refactor: strip external imports in workflow template transform

Co-authored-by: autofix-ci[bot] <114827586+autofix-ci[bot]@users.noreply.github.com>
Co-authored-by: Junyan Qin <rockchinq@gmail.com>
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>

.github/CODEOWNERS

@@ -24,6 +24,10 @@
 /api/services/tools/mcp_tools_manage_service.py @Nov1c444
 /api/controllers/mcp/ @Nov1c444
 /api/controllers/console/app/mcp_server.py @Nov1c444
+
+# Backend - Tests
+/api/tests/ @laipz8200 @QuantumGhost
+
 /api/tests/**/*mcp* @Nov1c444
 
 # Backend - Workflow - Engine (Core graph execution engine)
@@ -234,6 +238,9 @@
 # Frontend - Base Components
 /web/app/components/base/ @iamjoel @zxhlyh
 
+# Frontend - Base Components Tests
+/web/app/components/base/**/*.spec.tsx @hyoban @CodingOnStar
+
 # Frontend - Utils and Hooks
 /web/utils/classnames.ts @iamjoel @zxhlyh
 /web/utils/time.ts @iamjoel @zxhlyh

.github/workflows/autofix.yml

@@ -79,29 +79,6 @@ jobs:
           find . -name "*.py" -type f -exec sed -i.bak -E 's/"([^"]+)" \| None/Optional["\1"]/g; s/'"'"'([^'"'"']+)'"'"' \| None/Optional['"'"'\1'"'"']/g' {} \;
           find . -name "*.py.bak" -type f -delete
 
-      - name: Install pnpm
-        uses: pnpm/action-setup@v4
-        with:
-          package_json_file: web/package.json
-          run_install: false
-
-      - name: Setup Node.js
-        uses: actions/setup-node@v6
-        with:
-          node-version: 24
-          cache: pnpm
-          cache-dependency-path: ./web/pnpm-lock.yaml
-
-      - name: Install web dependencies
-        run: |
-          cd web
-          pnpm install --frozen-lockfile
-
-      - name: ESLint autofix
-        run: |
-          cd web
-          pnpm lint:fix || true
-
       # mdformat breaks YAML front matter in markdown files. Add --exclude for directories containing YAML front matter.
       - name: mdformat
         run: |

.github/workflows/web-tests.yml

@@ -39,7 +39,7 @@ jobs:
         run: pnpm install --frozen-lockfile
 
       - name: Run tests
-        run: pnpm test:coverage
+        run: pnpm test:ci
 
       - name: Coverage Summary
         if: always()

api/.importlinter

@@ -136,7 +136,6 @@ ignore_imports =
     core.workflow.nodes.llm.llm_utils -> models.provider
     core.workflow.nodes.llm.llm_utils -> services.credit_pool_service
     core.workflow.nodes.llm.node -> core.tools.signature
-    core.workflow.nodes.template_transform.template_transform_node -> configs
     core.workflow.nodes.tool.tool_node -> core.callback_handler.workflow_tool_callback_handler
     core.workflow.nodes.tool.tool_node -> core.tools.tool_engine
     core.workflow.nodes.tool.tool_node -> core.tools.tool_manager

api/controllers/console/app/app.py

@@ -1,3 +1,4 @@
+import logging
 import uuid
 from datetime import datetime
 from typing import Any, Literal, TypeAlias
@@ -54,6 +55,8 @@ ALLOW_CREATE_APP_MODES = ["chat", "agent-chat", "advanced-chat", "workflow", "co
 
 register_enum_models(console_ns, IconType)
 
+_logger = logging.getLogger(__name__)
+
 
 class AppListQuery(BaseModel):
     page: int = Field(default=1, ge=1, le=99999, description="Page number (1-99999)")
@@ -499,6 +502,7 @@ class AppListApi(Resource):
                     select(Workflow).where(
                         Workflow.version == Workflow.VERSION_DRAFT,
                         Workflow.app_id.in_(workflow_capable_app_ids),
+                        Workflow.tenant_id == current_tenant_id,
                     )
                 )
                 .scalars()
@@ -510,12 +514,14 @@ class AppListApi(Resource):
                 NodeType.TRIGGER_PLUGIN,
             }
             for workflow in draft_workflows:
+                node_id = None
                 try:
-                    for _, node_data in workflow.walk_nodes():
+                    for node_id, node_data in workflow.walk_nodes():
                         if node_data.get("type") in trigger_node_types:
                             draft_trigger_app_ids.add(str(workflow.app_id))
                             break
                 except Exception:
+                    _logger.exception("error while walking nodes, workflow_id=%s, node_id=%s", workflow.id, node_id)
                     continue
 
         for app in app_pagination.items:

api/controllers/console/explore/trial.py

@@ -1,15 +1,16 @@
 import logging
-from typing import Any, cast
+from typing import Any, Literal, cast
 
 from flask import request
-from flask_restx import Resource, fields, marshal, marshal_with, reqparse
+from flask_restx import Resource, fields, marshal, marshal_with
+from pydantic import BaseModel
 from werkzeug.exceptions import Forbidden, InternalServerError, NotFound
 
 import services
 from controllers.common.fields import Parameters as ParametersResponse
 from controllers.common.fields import Site as SiteResponse
 from controllers.common.schema import get_or_create_model
-from controllers.console import api
+from controllers.console import api, console_ns
 from controllers.console.app.error import (
     AppUnavailableError,
     AudioTooLargeError,
@@ -117,7 +118,56 @@ workflow_fields_copy["rag_pipeline_variables"] = fields.List(fields.Nested(pipel
 workflow_model = get_or_create_model("TrialWorkflow", workflow_fields_copy)
 
 
+# Pydantic models for request validation
+DEFAULT_REF_TEMPLATE_SWAGGER_2_0 = "#/definitions/{model}"
+
+
+class WorkflowRunRequest(BaseModel):
+    inputs: dict
+    files: list | None = None
+
+
+class ChatRequest(BaseModel):
+    inputs: dict
+    query: str
+    files: list | None = None
+    conversation_id: str | None = None
+    parent_message_id: str | None = None
+    retriever_from: str = "explore_app"
+
+
+class TextToSpeechRequest(BaseModel):
+    message_id: str | None = None
+    voice: str | None = None
+    text: str | None = None
+    streaming: bool | None = None
+
+
+class CompletionRequest(BaseModel):
+    inputs: dict
+    query: str = ""
+    files: list | None = None
+    response_mode: Literal["blocking", "streaming"] | None = None
+    retriever_from: str = "explore_app"
+
+
+# Register schemas for Swagger documentation
+console_ns.schema_model(
+    WorkflowRunRequest.__name__, WorkflowRunRequest.model_json_schema(ref_template=DEFAULT_REF_TEMPLATE_SWAGGER_2_0)
+)
+console_ns.schema_model(
+    ChatRequest.__name__, ChatRequest.model_json_schema(ref_template=DEFAULT_REF_TEMPLATE_SWAGGER_2_0)
+)
+console_ns.schema_model(
+    TextToSpeechRequest.__name__, TextToSpeechRequest.model_json_schema(ref_template=DEFAULT_REF_TEMPLATE_SWAGGER_2_0)
+)
+console_ns.schema_model(
+    CompletionRequest.__name__, CompletionRequest.model_json_schema(ref_template=DEFAULT_REF_TEMPLATE_SWAGGER_2_0)
+)
+
+
 class TrialAppWorkflowRunApi(TrialAppResource):
+    @console_ns.expect(console_ns.models[WorkflowRunRequest.__name__])
     def post(self, trial_app):
         """
         Run workflow
@@ -129,10 +179,8 @@ class TrialAppWorkflowRunApi(TrialAppResource):
         if app_mode != AppMode.WORKFLOW:
             raise NotWorkflowAppError()
 
-        parser = reqparse.RequestParser()
-        parser.add_argument("inputs", type=dict, required=True, nullable=False, location="json")
-        parser.add_argument("files", type=list, required=False, location="json")
-        args = parser.parse_args()
+        request_data = WorkflowRunRequest.model_validate(console_ns.payload)
+        args = request_data.model_dump()
         assert current_user is not None
         try:
             app_id = app_model.id
@@ -183,6 +231,7 @@ class TrialAppWorkflowTaskStopApi(TrialAppResource):
 
 
 class TrialChatApi(TrialAppResource):
+    @console_ns.expect(console_ns.models[ChatRequest.__name__])
     @trial_feature_enable
     def post(self, trial_app):
         app_model = trial_app
@@ -190,14 +239,14 @@ class TrialChatApi(TrialAppResource):
         if app_mode not in {AppMode.CHAT, AppMode.AGENT_CHAT, AppMode.ADVANCED_CHAT}:
             raise NotChatAppError()
 
-        parser = reqparse.RequestParser()
-        parser.add_argument("inputs", type=dict, required=True, location="json")
-        parser.add_argument("query", type=str, required=True, location="json")
-        parser.add_argument("files", type=list, required=False, location="json")
-        parser.add_argument("conversation_id", type=uuid_value, location="json")
-        parser.add_argument("parent_message_id", type=uuid_value, required=False, location="json")
-        parser.add_argument("retriever_from", type=str, required=False, default="explore_app", location="json")
-        args = parser.parse_args()
+        request_data = ChatRequest.model_validate(console_ns.payload)
+        args = request_data.model_dump()
+
+        # Validate UUID values if provided
+        if args.get("conversation_id"):
+            args["conversation_id"] = uuid_value(args["conversation_id"])
+        if args.get("parent_message_id"):
+            args["parent_message_id"] = uuid_value(args["parent_message_id"])
 
         args["auto_generate_name"] = False
 
@@ -320,20 +369,16 @@ class TrialChatAudioApi(TrialAppResource):
 
 
 class TrialChatTextApi(TrialAppResource):
+    @console_ns.expect(console_ns.models[TextToSpeechRequest.__name__])
     @trial_feature_enable
     def post(self, trial_app):
         app_model = trial_app
         try:
-            parser = reqparse.RequestParser()
-            parser.add_argument("message_id", type=str, required=False, location="json")
-            parser.add_argument("voice", type=str, location="json")
-            parser.add_argument("text", type=str, location="json")
-            parser.add_argument("streaming", type=bool, location="json")
-            args = parser.parse_args()
+            request_data = TextToSpeechRequest.model_validate(console_ns.payload)
 
-            message_id = args.get("message_id", None)
-            text = args.get("text", None)
-            voice = args.get("voice", None)
+            message_id = request_data.message_id
+            text = request_data.text
+            voice = request_data.voice
             if not isinstance(current_user, Account):
                 raise ValueError("current_user must be an Account instance")
 
@@ -371,19 +416,15 @@ class TrialChatTextApi(TrialAppResource):
 
 
 class TrialCompletionApi(TrialAppResource):
+    @console_ns.expect(console_ns.models[CompletionRequest.__name__])
     @trial_feature_enable
     def post(self, trial_app):
         app_model = trial_app
         if app_model.mode != "completion":
             raise NotCompletionAppError()
 
-        parser = reqparse.RequestParser()
-        parser.add_argument("inputs", type=dict, required=True, location="json")
-        parser.add_argument("query", type=str, location="json", default="")
-        parser.add_argument("files", type=list, required=False, location="json")
-        parser.add_argument("response_mode", type=str, choices=["blocking", "streaming"], location="json")
-        parser.add_argument("retriever_from", type=str, required=False, default="explore_app", location="json")
-        args = parser.parse_args()
+        request_data = CompletionRequest.model_validate(console_ns.payload)
+        args = request_data.model_dump()
 
         streaming = args["response_mode"] == "streaming"
         args["auto_generate_name"] = False

api/core/app/workflow/node_factory.py

@@ -47,6 +47,7 @@ class DifyNodeFactory(NodeFactory):
         code_providers: Sequence[type[CodeNodeProvider]] | None = None,
         code_limits: CodeNodeLimits | None = None,
         template_renderer: Jinja2TemplateRenderer | None = None,
+        template_transform_max_output_length: int | None = None,
         http_request_http_client: HttpClientProtocol | None = None,
         http_request_tool_file_manager_factory: Callable[[], ToolFileManager] = ToolFileManager,
         http_request_file_manager: FileManagerProtocol | None = None,
@@ -68,6 +69,9 @@ class DifyNodeFactory(NodeFactory):
             max_object_array_length=dify_config.CODE_MAX_OBJECT_ARRAY_LENGTH,
         )
         self._template_renderer = template_renderer or CodeExecutorJinja2TemplateRenderer()
+        self._template_transform_max_output_length = (
+            template_transform_max_output_length or dify_config.TEMPLATE_TRANSFORM_MAX_LENGTH
+        )
         self._http_request_http_client = http_request_http_client or ssrf_proxy
         self._http_request_tool_file_manager_factory = http_request_tool_file_manager_factory
         self._http_request_file_manager = http_request_file_manager or file_manager
@@ -122,6 +126,7 @@ class DifyNodeFactory(NodeFactory):
                 graph_init_params=self.graph_init_params,
                 graph_runtime_state=self.graph_runtime_state,
                 template_renderer=self._template_renderer,
+                max_output_length=self._template_transform_max_output_length,
             )
 
         if node_type == NodeType.HTTP_REQUEST:

api/core/helper/marketplace.py

@@ -6,7 +6,8 @@ from yarl import URL
 
 from configs import dify_config
 from core.helper.download import download_with_size_limit
-from core.plugin.entities.marketplace import MarketplacePluginDeclaration
+from core.plugin.entities.marketplace import MarketplacePluginDeclaration, MarketplacePluginSnapshot
+from extensions.ext_redis import redis_client
 
 marketplace_api_url = URL(str(dify_config.MARKETPLACE_API_URL))
 logger = logging.getLogger(__name__)
@@ -43,28 +44,37 @@ def batch_fetch_plugin_by_ids(plugin_ids: list[str]) -> list[dict]:
     return data.get("data", {}).get("plugins", [])
 
 
-def batch_fetch_plugin_manifests_ignore_deserialization_error(
-    plugin_ids: list[str],
-) -> Sequence[MarketplacePluginDeclaration]:
-    if len(plugin_ids) == 0:
-        return []
-
-    url = str(marketplace_api_url / "api/v1/plugins/batch")
-    response = httpx.post(url, json={"plugin_ids": plugin_ids}, headers={"X-Dify-Version": dify_config.project.version})
-    response.raise_for_status()
-    result: list[MarketplacePluginDeclaration] = []
-    for plugin in response.json()["data"]["plugins"]:
-        try:
-            result.append(MarketplacePluginDeclaration.model_validate(plugin))
-        except Exception:
-            logger.exception(
-                "Failed to deserialize marketplace plugin manifest for %s", plugin.get("plugin_id", "unknown")
-            )
-
-    return result
-
-
 def record_install_plugin_event(plugin_unique_identifier: str):
     url = str(marketplace_api_url / "api/v1/stats/plugins/install_count")
     response = httpx.post(url, json={"unique_identifier": plugin_unique_identifier})
     response.raise_for_status()
+
+
+def fetch_global_plugin_manifest(cache_key_prefix: str, cache_ttl: int) -> None:
+    """
+    Fetch all plugin manifests from marketplace and cache them in Redis.
+    This should be called once per check cycle to populate the instance-level cache.
+
+    Args:
+        cache_key_prefix: Redis key prefix for caching plugin manifests
+        cache_ttl: Cache TTL in seconds
+
+    Raises:
+        httpx.HTTPError: If the HTTP request fails
+        Exception: If any other error occurs during fetching or caching
+    """
+    url = str(marketplace_api_url / "api/v1/dist/plugins/manifest.json")
+    response = httpx.get(url, headers={"X-Dify-Version": dify_config.project.version}, timeout=30)
+    response.raise_for_status()
+
+    raw_json = response.json()
+    plugins_data = raw_json.get("plugins", [])
+
+    # Parse and cache all plugin snapshots
+    for plugin_data in plugins_data:
+        plugin_snapshot = MarketplacePluginSnapshot.model_validate(plugin_data)
+        redis_client.setex(
+            name=f"{cache_key_prefix}{plugin_snapshot.plugin_id}",
+            time=cache_ttl,
+            value=plugin_snapshot.model_dump_json(),
+        )

api/core/plugin/entities/marketplace.py

@@ -1,4 +1,4 @@
-from pydantic import BaseModel, Field, model_validator
+from pydantic import BaseModel, Field, computed_field, model_validator
 
 from core.model_runtime.entities.provider_entities import ProviderEntity
 from core.plugin.entities.endpoint import EndpointProviderDeclaration
@@ -48,3 +48,15 @@ class MarketplacePluginDeclaration(BaseModel):
         if "tool" in data and not data["tool"]:
             del data["tool"]
         return data
+
+
+class MarketplacePluginSnapshot(BaseModel):
+    org: str
+    name: str
+    latest_version: str
+    latest_package_identifier: str
+    latest_package_url: str
+
+    @computed_field
+    def plugin_id(self) -> str:
+        return f"{self.org}/{self.name}"

api/core/workflow/nodes/template_transform/template_transform_node.py

@@ -1,7 +1,6 @@
 from collections.abc import Mapping, Sequence
 from typing import TYPE_CHECKING, Any
 
-from configs import dify_config
 from core.workflow.enums import NodeType, WorkflowNodeExecutionStatus
 from core.workflow.node_events import NodeRunResult
 from core.workflow.nodes.base.node import Node
@@ -16,12 +15,13 @@ if TYPE_CHECKING:
     from core.workflow.entities import GraphInitParams
     from core.workflow.runtime import GraphRuntimeState
 
-MAX_TEMPLATE_TRANSFORM_OUTPUT_LENGTH = dify_config.TEMPLATE_TRANSFORM_MAX_LENGTH
+DEFAULT_TEMPLATE_TRANSFORM_MAX_OUTPUT_LENGTH = 400_000
 
 
 class TemplateTransformNode(Node[TemplateTransformNodeData]):
     node_type = NodeType.TEMPLATE_TRANSFORM
     _template_renderer: Jinja2TemplateRenderer
+    _max_output_length: int
 
     def __init__(
         self,
@@ -31,6 +31,7 @@ class TemplateTransformNode(Node[TemplateTransformNodeData]):
         graph_runtime_state: "GraphRuntimeState",
         *,
         template_renderer: Jinja2TemplateRenderer | None = None,
+        max_output_length: int | None = None,
     ) -> None:
         super().__init__(
             id=id,
@@ -40,6 +41,10 @@ class TemplateTransformNode(Node[TemplateTransformNodeData]):
         )
         self._template_renderer = template_renderer or CodeExecutorJinja2TemplateRenderer()
 
+        if max_output_length is not None and max_output_length <= 0:
+            raise ValueError("max_output_length must be a positive integer")
+        self._max_output_length = max_output_length or DEFAULT_TEMPLATE_TRANSFORM_MAX_OUTPUT_LENGTH
+
     @classmethod
     def get_default_config(cls, filters: Mapping[str, object] | None = None) -> Mapping[str, object]:
         """
@@ -69,11 +74,11 @@ class TemplateTransformNode(Node[TemplateTransformNodeData]):
         except TemplateRenderError as e:
             return NodeRunResult(inputs=variables, status=WorkflowNodeExecutionStatus.FAILED, error=str(e))
 
-        if len(rendered) > MAX_TEMPLATE_TRANSFORM_OUTPUT_LENGTH:
+        if len(rendered) > self._max_output_length:
             return NodeRunResult(
                 inputs=variables,
                 status=WorkflowNodeExecutionStatus.FAILED,
-                error=f"Output length exceeds {MAX_TEMPLATE_TRANSFORM_OUTPUT_LENGTH} characters",
+                error=f"Output length exceeds {self._max_output_length} characters",
             )
 
         return NodeRunResult(

api/migrations/versions/2025_12_25_1039-7df29de0f6be_add_credit_pool.py

@@ -10,6 +10,10 @@ import models as models
 import sqlalchemy as sa
 from sqlalchemy.dialects import postgresql
 
+
+def _is_pg(conn):
+    return conn.dialect.name == "postgresql"
+
 # revision identifiers, used by Alembic.
 revision = '7df29de0f6be'
 down_revision = '03ea244985ce'
@@ -19,16 +23,31 @@ depends_on = None
 
 def upgrade():
     # ### commands auto generated by Alembic - please adjust! ###
-    op.create_table('tenant_credit_pools',
-    sa.Column('id', models.types.StringUUID(), server_default=sa.text('uuid_generate_v4()'), nullable=False),
-    sa.Column('tenant_id', models.types.StringUUID(), nullable=False),
-    sa.Column('pool_type', sa.String(length=40), server_default='trial', nullable=False),
-    sa.Column('quota_limit', sa.BigInteger(), nullable=False),
-    sa.Column('quota_used', sa.BigInteger(), nullable=False),
-    sa.Column('created_at', sa.DateTime(), server_default=sa.text('CURRENT_TIMESTAMP'), nullable=False),
-    sa.Column('updated_at', sa.DateTime(), server_default=sa.text('CURRENT_TIMESTAMP'), nullable=False),
-    sa.PrimaryKeyConstraint('id', name='tenant_credit_pool_pkey')
-    )
+    conn = op.get_bind()
+
+    if _is_pg(conn):
+        op.create_table('tenant_credit_pools',
+        sa.Column('id', models.types.StringUUID(), server_default=sa.text('uuid_generate_v4()'), nullable=False),
+        sa.Column('tenant_id', models.types.StringUUID(), nullable=False),
+        sa.Column('pool_type', sa.String(length=40), server_default='trial', nullable=False),
+        sa.Column('quota_limit', sa.BigInteger(), nullable=False),
+        sa.Column('quota_used', sa.BigInteger(), nullable=False),
+        sa.Column('created_at', sa.DateTime(), server_default=sa.text('CURRENT_TIMESTAMP'), nullable=False),
+        sa.Column('updated_at', sa.DateTime(), server_default=sa.text('CURRENT_TIMESTAMP'), nullable=False),
+        sa.PrimaryKeyConstraint('id', name='tenant_credit_pool_pkey')
+        )
+    else:
+        # For MySQL and other databases, UUID should be generated at application level
+        op.create_table('tenant_credit_pools',
+        sa.Column('id', models.types.StringUUID(), nullable=False),
+        sa.Column('tenant_id', models.types.StringUUID(), nullable=False),
+        sa.Column('pool_type', sa.String(length=40), server_default='trial', nullable=False),
+        sa.Column('quota_limit', sa.BigInteger(), nullable=False),
+        sa.Column('quota_used', sa.BigInteger(), nullable=False),
+        sa.Column('created_at', sa.DateTime(), server_default=sa.func.current_timestamp(), nullable=False),
+        sa.Column('updated_at', sa.DateTime(), server_default=sa.func.current_timestamp(), nullable=False),
+        sa.PrimaryKeyConstraint('id', name='tenant_credit_pool_pkey')
+        )
     with op.batch_alter_table('tenant_credit_pools', schema=None) as batch_op:
         batch_op.create_index('tenant_credit_pool_pool_type_idx', ['pool_type'], unique=False)
         batch_op.create_index('tenant_credit_pool_tenant_id_idx', ['tenant_id'], unique=False)

api/models/model.py

@@ -2166,7 +2166,9 @@ class TenantCreditPool(TypeBase):
         sa.Index("tenant_credit_pool_pool_type_idx", "pool_type"),
     )
 
-    id: Mapped[str] = mapped_column(StringUUID, primary_key=True, server_default=text("uuid_generate_v4()"), init=False)
+    id: Mapped[str] = mapped_column(
+        StringUUID, insert_default=lambda: str(uuid4()), default_factory=lambda: str(uuid4()), init=False
+    )
     tenant_id: Mapped[str] = mapped_column(StringUUID, nullable=False)
     pool_type: Mapped[str] = mapped_column(String(40), nullable=False, default="trial", server_default="trial")
     quota_limit: Mapped[int] = mapped_column(BigInteger, nullable=False, default=0)

api/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "dify-api"
-version = "1.12.0"
+version = "1.12.1"
 requires-python = ">=3.11,<3.13"
 
 dependencies = [

api/schedule/check_upgradable_plugin_task.py

@@ -1,16 +1,24 @@
+import logging
 import math
 import time
 
 import click
 
 import app
+from core.helper.marketplace import fetch_global_plugin_manifest
 from extensions.ext_database import db
 from models.account import TenantPluginAutoUpgradeStrategy
 from tasks import process_tenant_plugin_autoupgrade_check_task as check_task
 
+logger = logging.getLogger(__name__)
+
 AUTO_UPGRADE_MINIMAL_CHECKING_INTERVAL = 15 * 60  # 15 minutes
 MAX_CONCURRENT_CHECK_TASKS = 20
 
+# Import cache constants from the task module
+CACHE_REDIS_KEY_PREFIX = check_task.CACHE_REDIS_KEY_PREFIX
+CACHE_REDIS_TTL = check_task.CACHE_REDIS_TTL
+
 
 @app.celery.task(queue="plugin")
 def check_upgradable_plugin_task():
@@ -40,6 +48,22 @@ def check_upgradable_plugin_task():
     )  # make sure all strategies are checked in this interval
     batch_interval_time = (AUTO_UPGRADE_MINIMAL_CHECKING_INTERVAL / batch_chunk_count) if batch_chunk_count > 0 else 0
 
+    if total_strategies == 0:
+        click.echo(click.style("no strategies to process, skipping plugin manifest fetch.", fg="green"))
+        return
+
+    # Fetch and cache all plugin manifests before processing tenants
+    # This reduces load on marketplace from 300k requests to 1 request per check cycle
+    logger.info("fetching global plugin manifest from marketplace")
+    try:
+        fetch_global_plugin_manifest(CACHE_REDIS_KEY_PREFIX, CACHE_REDIS_TTL)
+        logger.info("successfully fetched and cached global plugin manifest")
+    except Exception as e:
+        logger.exception("failed to fetch global plugin manifest")
+        click.echo(click.style(f"failed to fetch global plugin manifest: {e}", fg="red"))
+        click.echo(click.style("skipping plugin upgrade check for this cycle", fg="yellow"))
+        return
+
     for i in range(0, total_strategies, MAX_CONCURRENT_CHECK_TASKS):
         batch_strategies = strategies[i : i + MAX_CONCURRENT_CHECK_TASKS]
         for strategy in batch_strategies:

api/services/account_service.py

@@ -327,6 +327,17 @@ class AccountService:
     @staticmethod
     def delete_account(account: Account):
         """Delete account. This method only adds a task to the queue for deletion."""
+        # Queue account deletion sync tasks for all workspaces BEFORE account deletion (enterprise only)
+        from services.enterprise.account_deletion_sync import sync_account_deletion
+
+        sync_success = sync_account_deletion(account_id=account.id, source="account_deleted")
+        if not sync_success:
+            logger.warning(
+                "Enterprise account deletion sync failed for account %s; proceeding with local deletion.",
+                account.id,
+            )
+
+        # Now proceed with async account deletion
         delete_account_task.delay(account.id)
 
     @staticmethod
@@ -1230,6 +1241,19 @@ class TenantService:
         if dify_config.BILLING_ENABLED:
             BillingService.clean_billing_info_cache(tenant.id)
 
+        # Queue account deletion sync task for enterprise backend to reassign resources (enterprise only)
+        from services.enterprise.account_deletion_sync import sync_workspace_member_removal
+
+        sync_success = sync_workspace_member_removal(
+            workspace_id=tenant.id, member_id=account.id, source="workspace_member_removed"
+        )
+        if not sync_success:
+            logger.warning(
+                "Enterprise workspace member removal sync failed: workspace_id=%s, member_id=%s",
+                tenant.id,
+                account.id,
+            )
+
     @staticmethod
     def update_member_role(tenant: Tenant, member: Account, new_role: str, operator: Account):
         """Update member role"""

api/services/enterprise/account_deletion_sync.py

@@ -0,0 +1,115 @@
+import json
+import logging
+import uuid
+from datetime import UTC, datetime
+
+from redis import RedisError
+
+from configs import dify_config
+from extensions.ext_database import db
+from extensions.ext_redis import redis_client
+from models.account import TenantAccountJoin
+
+logger = logging.getLogger(__name__)
+
+ACCOUNT_DELETION_SYNC_QUEUE = "enterprise:member:sync:queue"
+ACCOUNT_DELETION_SYNC_TASK_TYPE = "sync_member_deletion_from_workspace"
+
+
+def _queue_task(workspace_id: str, member_id: str, *, source: str) -> bool:
+    """
+    Queue an account deletion sync task to Redis.
+
+    Internal helper function. Do not call directly - use the public functions instead.
+
+    Args:
+        workspace_id: The workspace/tenant ID to sync
+        member_id: The member/account ID that was removed
+        source: Source of the sync request (for debugging/tracking)
+
+    Returns:
+        bool: True if task was queued successfully, False otherwise
+    """
+    try:
+        task = {
+            "task_id": str(uuid.uuid4()),
+            "workspace_id": workspace_id,
+            "member_id": member_id,
+            "retry_count": 0,
+            "created_at": datetime.now(UTC).isoformat(),
+            "source": source,
+            "type": ACCOUNT_DELETION_SYNC_TASK_TYPE,
+        }
+
+        # Push to Redis list (queue) - LPUSH adds to the head, worker consumes from tail with RPOP
+        redis_client.lpush(ACCOUNT_DELETION_SYNC_QUEUE, json.dumps(task))
+
+        logger.info(
+            "Queued account deletion sync task for workspace %s, member %s, task_id: %s, source: %s",
+            workspace_id,
+            member_id,
+            task["task_id"],
+            source,
+        )
+        return True
+
+    except (RedisError, TypeError) as e:
+        logger.error(
+            "Failed to queue account deletion sync for workspace %s, member %s: %s",
+            workspace_id,
+            member_id,
+            str(e),
+            exc_info=True,
+        )
+        # Don't raise - we don't want to fail member deletion if queueing fails
+        return False
+
+
+def sync_workspace_member_removal(workspace_id: str, member_id: str, *, source: str) -> bool:
+    """
+    Sync a single workspace member removal (enterprise only).
+
+    Queues a task for the enterprise backend to reassign resources from the removed member.
+    Handles enterprise edition check internally. Safe to call in community edition (no-op).
+
+    Args:
+        workspace_id: The workspace/tenant ID
+        member_id: The member/account ID that was removed
+        source: Source of the sync request (e.g., "workspace_member_removed")
+
+    Returns:
+        bool: True if task was queued (or skipped in community), False if queueing failed
+    """
+    if not dify_config.ENTERPRISE_ENABLED:
+        return True
+
+    return _queue_task(workspace_id=workspace_id, member_id=member_id, source=source)
+
+
+def sync_account_deletion(account_id: str, *, source: str) -> bool:
+    """
+    Sync full account deletion across all workspaces (enterprise only).
+
+    Fetches all workspace memberships for the account and queues a sync task for each.
+    Handles enterprise edition check internally. Safe to call in community edition (no-op).
+
+    Args:
+        account_id: The account ID being deleted
+        source: Source of the sync request (e.g., "account_deleted")
+
+    Returns:
+        bool: True if all tasks were queued (or skipped in community), False if any queueing failed
+    """
+    if not dify_config.ENTERPRISE_ENABLED:
+        return True
+
+    # Fetch all workspaces the account belongs to
+    workspace_joins = db.session.query(TenantAccountJoin).filter_by(account_id=account_id).all()
+
+    # Queue sync task for each workspace
+    success = True
+    for join in workspace_joins:
+        if not _queue_task(workspace_id=join.tenant_id, member_id=account_id, source=source):
+            success = False
+
+    return success

api/tasks/document_indexing_update_task.py

@@ -8,7 +8,6 @@ from sqlalchemy import delete, select
 from core.db.session_factory import session_factory
 from core.indexing_runner import DocumentIsPausedError, IndexingRunner
 from core.rag.index_processor.index_processor_factory import IndexProcessorFactory
-from extensions.ext_database import db
 from libs.datetime_utils import naive_utc_now
 from models.dataset import Dataset, Document, DocumentSegment
 
@@ -27,7 +26,7 @@ def document_indexing_update_task(dataset_id: str, document_id: str):
     logger.info(click.style(f"Start update document: {document_id}", fg="green"))
     start_at = time.perf_counter()
 
-    with session_factory.create_session() as session:
+    with session_factory.create_session() as session, session.begin():
         document = session.query(Document).where(Document.id == document_id, Document.dataset_id == dataset_id).first()
 
         if not document:
@@ -36,7 +35,6 @@ def document_indexing_update_task(dataset_id: str, document_id: str):
 
         document.indexing_status = "parsing"
         document.processing_started_at = naive_utc_now()
-        session.commit()
 
         # delete all document segment and index
         try:
@@ -56,7 +54,7 @@ def document_indexing_update_task(dataset_id: str, document_id: str):
                 segment_ids = [segment.id for segment in segments]
                 segment_delete_stmt = delete(DocumentSegment).where(DocumentSegment.id.in_(segment_ids))
                 session.execute(segment_delete_stmt)
-                db.session.commit()
+
             end_at = time.perf_counter()
             logger.info(
                 click.style(

api/tasks/process_tenant_plugin_autoupgrade_check_task.py

@@ -6,8 +6,8 @@ import typing
 import click
 from celery import shared_task
 
-from core.helper import marketplace
-from core.helper.marketplace import MarketplacePluginDeclaration
+from core.helper.marketplace import record_install_plugin_event
+from core.plugin.entities.marketplace import MarketplacePluginSnapshot
 from core.plugin.entities.plugin import PluginInstallationSource
 from core.plugin.impl.plugin import PluginInstaller
 from extensions.ext_redis import redis_client
@@ -16,7 +16,7 @@ from models.account import TenantPluginAutoUpgradeStrategy
 logger = logging.getLogger(__name__)
 
 RETRY_TIMES_OF_ONE_PLUGIN_IN_ONE_TENANT = 3
-CACHE_REDIS_KEY_PREFIX = "plugin_autoupgrade_check_task:cached_plugin_manifests:"
+CACHE_REDIS_KEY_PREFIX = "plugin_autoupgrade_check_task:cached_plugin_snapshot:"
 CACHE_REDIS_TTL = 60 * 60  # 1 hour
 
 
@@ -25,11 +25,11 @@ def _get_redis_cache_key(plugin_id: str) -> str:
     return f"{CACHE_REDIS_KEY_PREFIX}{plugin_id}"
 
 
-def _get_cached_manifest(plugin_id: str) -> typing.Union[MarketplacePluginDeclaration, None, bool]:
+def _get_cached_manifest(plugin_id: str) -> typing.Union[MarketplacePluginSnapshot, None, bool]:
     """
     Get cached plugin manifest from Redis.
     Returns:
-        - MarketplacePluginDeclaration: if found in cache
+        - MarketplacePluginSnapshot: if found in cache
         - None: if cached as not found (marketplace returned no result)
         - False: if not in cache at all
     """
@@ -43,76 +43,31 @@ def _get_cached_manifest(plugin_id: str) -> typing.Union[MarketplacePluginDeclar
         if cached_json is None:
             return None
 
-        return MarketplacePluginDeclaration.model_validate(cached_json)
+        return MarketplacePluginSnapshot.model_validate(cached_json)
     except Exception:
         logger.exception("Failed to get cached manifest for plugin %s", plugin_id)
         return False
 
 
-def _set_cached_manifest(plugin_id: str, manifest: typing.Union[MarketplacePluginDeclaration, None]) -> None:
-    """
-    Cache plugin manifest in Redis.
-    Args:
-        plugin_id: The plugin ID
-        manifest: The manifest to cache, or None if not found in marketplace
-    """
-    try:
-        key = _get_redis_cache_key(plugin_id)
-        if manifest is None:
-            # Cache the fact that this plugin was not found
-            redis_client.setex(key, CACHE_REDIS_TTL, json.dumps(None))
-        else:
-            # Cache the manifest data
-            redis_client.setex(key, CACHE_REDIS_TTL, manifest.model_dump_json())
-    except Exception:
-        # If Redis fails, continue without caching
-        # traceback.print_exc()
-        logger.exception("Failed to set cached manifest for plugin %s", plugin_id)
-
-
 def marketplace_batch_fetch_plugin_manifests(
     plugin_ids_plain_list: list[str],
-) -> list[MarketplacePluginDeclaration]:
-    """Fetch plugin manifests with Redis caching support."""
-    cached_manifests: dict[str, typing.Union[MarketplacePluginDeclaration, None]] = {}
-    not_cached_plugin_ids: list[str] = []
+) -> list[MarketplacePluginSnapshot]:
+    """
+    Fetch plugin manifests from Redis cache only.
+    This function assumes fetch_global_plugin_manifest() has been called
+    to pre-populate the cache with all marketplace plugins.
+    """
+    result: list[MarketplacePluginSnapshot] = []
 
     # Check Redis cache for each plugin
     for plugin_id in plugin_ids_plain_list:
         cached_result = _get_cached_manifest(plugin_id)
-        if cached_result is False:
-            # Not in cache, need to fetch
-            not_cached_plugin_ids.append(plugin_id)
-        else:
-            # Either found manifest or cached as None (not found in marketplace)
-            # At this point, cached_result is either MarketplacePluginDeclaration or None
-            if isinstance(cached_result, bool):
-                # This should never happen due to the if condition above, but for type safety
-                continue
-            cached_manifests[plugin_id] = cached_result
+        if not isinstance(cached_result, MarketplacePluginSnapshot):
+            # cached_result is False (not in cache) or None (cached as not found)
+            logger.warning("plugin %s not found in cache, skipping", plugin_id)
+            continue
 
-    # Fetch uncached plugins from marketplace
-    if not_cached_plugin_ids:
-        manifests = marketplace.batch_fetch_plugin_manifests_ignore_deserialization_error(not_cached_plugin_ids)
-
-        # Cache the fetched manifests
-        for manifest in manifests:
-            cached_manifests[manifest.plugin_id] = manifest
-            _set_cached_manifest(manifest.plugin_id, manifest)
-
-        # Cache plugins that were not found in marketplace
-        fetched_plugin_ids = {manifest.plugin_id for manifest in manifests}
-        for plugin_id in not_cached_plugin_ids:
-            if plugin_id not in fetched_plugin_ids:
-                cached_manifests[plugin_id] = None
-                _set_cached_manifest(plugin_id, None)
-
-    # Build result list from cached manifests
-    result: list[MarketplacePluginDeclaration] = []
-    for plugin_id in plugin_ids_plain_list:
-        cached_manifest: typing.Union[MarketplacePluginDeclaration, None] = cached_manifests.get(plugin_id)
-        if cached_manifest is not None:
-            result.append(cached_manifest)
+        result.append(cached_result)
 
     return result
 
@@ -211,7 +166,7 @@ def process_tenant_plugin_autoupgrade_check_task(
                         # execute upgrade
                         new_unique_identifier = manifest.latest_package_identifier
 
-                        marketplace.record_install_plugin_event(new_unique_identifier)
+                        record_install_plugin_event(new_unique_identifier)
                         click.echo(
                             click.style(
                                 f"Upgrade plugin: {original_unique_identifier} -> {new_unique_identifier}",

api/tasks/remove_app_and_related_data_task.py

@@ -259,8 +259,8 @@ def _delete_app_workflow_app_logs(tenant_id: str, app_id: str):
 
 
 def _delete_app_workflow_archive_logs(tenant_id: str, app_id: str):
-    def del_workflow_archive_log(workflow_archive_log_id: str):
-        db.session.query(WorkflowArchiveLog).where(WorkflowArchiveLog.id == workflow_archive_log_id).delete(
+    def del_workflow_archive_log(session, workflow_archive_log_id: str):
+        session.query(WorkflowArchiveLog).where(WorkflowArchiveLog.id == workflow_archive_log_id).delete(
             synchronize_session=False
         )
 
@@ -420,7 +420,7 @@ def delete_draft_variables_batch(app_id: str, batch_size: int = 1000) -> int:
     total_files_deleted = 0
 
     while True:
-        with session_factory.create_session() as session:
+        with session_factory.create_session() as session, session.begin():
             # Get a batch of draft variable IDs along with their file_ids
             query_sql = """
                 SELECT id, file_id FROM workflow_draft_variables

api/tests/integration_tests/tasks/test_remove_app_and_related_data_task.py

@@ -10,7 +10,10 @@ from models import Tenant
 from models.enums import CreatorUserRole
 from models.model import App, UploadFile
 from models.workflow import WorkflowDraftVariable, WorkflowDraftVariableFile
-from tasks.remove_app_and_related_data_task import _delete_draft_variables, delete_draft_variables_batch
+from tasks.remove_app_and_related_data_task import (
+    _delete_draft_variables,
+    delete_draft_variables_batch,
+)
 
 
 @pytest.fixture
@@ -297,12 +300,18 @@ class TestDeleteDraftVariablesWithOffloadIntegration:
     def test_delete_draft_variables_with_offload_data(self, mock_storage, setup_offload_test_data):
         data = setup_offload_test_data
         app_id = data["app"].id
+        upload_file_ids = [uf.id for uf in data["upload_files"]]
+        variable_file_ids = [vf.id for vf in data["variable_files"]]
         mock_storage.delete.return_value = None
 
         with session_factory.create_session() as session:
             draft_vars_before = session.query(WorkflowDraftVariable).filter_by(app_id=app_id).count()
-            var_files_before = session.query(WorkflowDraftVariableFile).count()
-            upload_files_before = session.query(UploadFile).count()
+            var_files_before = (
+                session.query(WorkflowDraftVariableFile)
+                .where(WorkflowDraftVariableFile.id.in_(variable_file_ids))
+                .count()
+            )
+            upload_files_before = session.query(UploadFile).where(UploadFile.id.in_(upload_file_ids)).count()
         assert draft_vars_before == 3
         assert var_files_before == 2
         assert upload_files_before == 2
@@ -315,8 +324,12 @@ class TestDeleteDraftVariablesWithOffloadIntegration:
         assert draft_vars_after == 0
 
         with session_factory.create_session() as session:
-            var_files_after = session.query(WorkflowDraftVariableFile).count()
-            upload_files_after = session.query(UploadFile).count()
+            var_files_after = (
+                session.query(WorkflowDraftVariableFile)
+                .where(WorkflowDraftVariableFile.id.in_(variable_file_ids))
+                .count()
+            )
+            upload_files_after = session.query(UploadFile).where(UploadFile.id.in_(upload_file_ids)).count()
         assert var_files_after == 0
         assert upload_files_after == 0
 
@@ -329,6 +342,8 @@ class TestDeleteDraftVariablesWithOffloadIntegration:
     def test_delete_draft_variables_storage_failure_continues_cleanup(self, mock_storage, setup_offload_test_data):
         data = setup_offload_test_data
         app_id = data["app"].id
+        upload_file_ids = [uf.id for uf in data["upload_files"]]
+        variable_file_ids = [vf.id for vf in data["variable_files"]]
         mock_storage.delete.side_effect = [Exception("Storage error"), None]
 
         deleted_count = delete_draft_variables_batch(app_id, batch_size=10)
@@ -339,8 +354,12 @@ class TestDeleteDraftVariablesWithOffloadIntegration:
         assert draft_vars_after == 0
 
         with session_factory.create_session() as session:
-            var_files_after = session.query(WorkflowDraftVariableFile).count()
-            upload_files_after = session.query(UploadFile).count()
+            var_files_after = (
+                session.query(WorkflowDraftVariableFile)
+                .where(WorkflowDraftVariableFile.id.in_(variable_file_ids))
+                .count()
+            )
+            upload_files_after = session.query(UploadFile).where(UploadFile.id.in_(upload_file_ids)).count()
         assert var_files_after == 0
         assert upload_files_after == 0
 
@@ -395,3 +414,275 @@ class TestDeleteDraftVariablesWithOffloadIntegration:
                 if app2_obj:
                     session.delete(app2_obj)
                 session.commit()
+
+
+class TestDeleteDraftVariablesSessionCommit:
+    """Test suite to verify session commit behavior in delete_draft_variables_batch."""
+
+    @pytest.fixture
+    def setup_offload_test_data(self, app_and_tenant):
+        """Create test data with offload files for session commit tests."""
+        from core.variables.types import SegmentType
+        from libs.datetime_utils import naive_utc_now
+
+        tenant, app = app_and_tenant
+
+        with session_factory.create_session() as session:
+            upload_file1 = UploadFile(
+                tenant_id=tenant.id,
+                storage_type="local",
+                key="test/file1.json",
+                name="file1.json",
+                size=1024,
+                extension="json",
+                mime_type="application/json",
+                created_by_role=CreatorUserRole.ACCOUNT,
+                created_by=str(uuid.uuid4()),
+                created_at=naive_utc_now(),
+                used=False,
+            )
+            upload_file2 = UploadFile(
+                tenant_id=tenant.id,
+                storage_type="local",
+                key="test/file2.json",
+                name="file2.json",
+                size=2048,
+                extension="json",
+                mime_type="application/json",
+                created_by_role=CreatorUserRole.ACCOUNT,
+                created_by=str(uuid.uuid4()),
+                created_at=naive_utc_now(),
+                used=False,
+            )
+            session.add(upload_file1)
+            session.add(upload_file2)
+            session.flush()
+
+            var_file1 = WorkflowDraftVariableFile(
+                tenant_id=tenant.id,
+                app_id=app.id,
+                user_id=str(uuid.uuid4()),
+                upload_file_id=upload_file1.id,
+                size=1024,
+                length=10,
+                value_type=SegmentType.STRING,
+            )
+            var_file2 = WorkflowDraftVariableFile(
+                tenant_id=tenant.id,
+                app_id=app.id,
+                user_id=str(uuid.uuid4()),
+                upload_file_id=upload_file2.id,
+                size=2048,
+                length=20,
+                value_type=SegmentType.OBJECT,
+            )
+            session.add(var_file1)
+            session.add(var_file2)
+            session.flush()
+
+            draft_var1 = WorkflowDraftVariable.new_node_variable(
+                app_id=app.id,
+                node_id="node_1",
+                name="large_var_1",
+                value=StringSegment(value="truncated..."),
+                node_execution_id=str(uuid.uuid4()),
+                file_id=var_file1.id,
+            )
+            draft_var2 = WorkflowDraftVariable.new_node_variable(
+                app_id=app.id,
+                node_id="node_2",
+                name="large_var_2",
+                value=StringSegment(value="truncated..."),
+                node_execution_id=str(uuid.uuid4()),
+                file_id=var_file2.id,
+            )
+            draft_var3 = WorkflowDraftVariable.new_node_variable(
+                app_id=app.id,
+                node_id="node_3",
+                name="regular_var",
+                value=StringSegment(value="regular_value"),
+                node_execution_id=str(uuid.uuid4()),
+            )
+            session.add(draft_var1)
+            session.add(draft_var2)
+            session.add(draft_var3)
+            session.commit()
+
+            data = {
+                "app": app,
+                "tenant": tenant,
+                "upload_files": [upload_file1, upload_file2],
+                "variable_files": [var_file1, var_file2],
+                "draft_variables": [draft_var1, draft_var2, draft_var3],
+            }
+
+        yield data
+
+        with session_factory.create_session() as session:
+            for table, ids in [
+                (WorkflowDraftVariable, [v.id for v in data["draft_variables"]]),
+                (WorkflowDraftVariableFile, [vf.id for vf in data["variable_files"]]),
+                (UploadFile, [uf.id for uf in data["upload_files"]]),
+            ]:
+                cleanup_query = delete(table).where(table.id.in_(ids)).execution_options(synchronize_session=False)
+                session.execute(cleanup_query)
+            session.commit()
+
+    @pytest.fixture
+    def setup_commit_test_data(self, app_and_tenant):
+        """Create test data for session commit tests."""
+        tenant, app = app_and_tenant
+        variable_ids: list[str] = []
+
+        with session_factory.create_session() as session:
+            variables = []
+            for i in range(10):
+                var = WorkflowDraftVariable.new_node_variable(
+                    app_id=app.id,
+                    node_id=f"node_{i}",
+                    name=f"var_{i}",
+                    value=StringSegment(value="test_value"),
+                    node_execution_id=str(uuid.uuid4()),
+                )
+                session.add(var)
+                variables.append(var)
+            session.commit()
+            variable_ids = [v.id for v in variables]
+
+        yield {
+            "app": app,
+            "tenant": tenant,
+            "variable_ids": variable_ids,
+        }
+
+        with session_factory.create_session() as session:
+            cleanup_query = (
+                delete(WorkflowDraftVariable)
+                .where(WorkflowDraftVariable.id.in_(variable_ids))
+                .execution_options(synchronize_session=False)
+            )
+            session.execute(cleanup_query)
+            session.commit()
+
+    def test_session_commit_is_called_after_each_batch(self, setup_commit_test_data):
+        """Test that session.begin() is used for automatic transaction management."""
+        data = setup_commit_test_data
+        app_id = data["app"].id
+
+        # Since session.begin() is used, the transaction is automatically committed
+        # when the with block exits successfully. We verify this by checking that
+        # data is actually persisted.
+        deleted_count = delete_draft_variables_batch(app_id, batch_size=3)
+
+        # Verify all data was deleted (proves transaction was committed)
+        with session_factory.create_session() as session:
+            remaining_count = session.query(WorkflowDraftVariable).filter_by(app_id=app_id).count()
+
+        assert deleted_count == 10
+        assert remaining_count == 0
+
+    def test_data_persisted_after_batch_deletion(self, setup_commit_test_data):
+        """Test that data is actually persisted to database after batch deletion with commits."""
+        data = setup_commit_test_data
+        app_id = data["app"].id
+        variable_ids = data["variable_ids"]
+
+        # Verify initial state
+        with session_factory.create_session() as session:
+            initial_count = session.query(WorkflowDraftVariable).filter_by(app_id=app_id).count()
+        assert initial_count == 10
+
+        # Perform deletion with small batch size to force multiple commits
+        deleted_count = delete_draft_variables_batch(app_id, batch_size=3)
+
+        assert deleted_count == 10
+
+        # Verify all data is deleted in a new session (proves commits worked)
+        with session_factory.create_session() as session:
+            final_count = session.query(WorkflowDraftVariable).filter_by(app_id=app_id).count()
+        assert final_count == 0
+
+        # Verify specific IDs are deleted
+        with session_factory.create_session() as session:
+            remaining_vars = (
+                session.query(WorkflowDraftVariable).where(WorkflowDraftVariable.id.in_(variable_ids)).count()
+            )
+        assert remaining_vars == 0
+
+    def test_session_commit_with_empty_dataset(self, setup_commit_test_data):
+        """Test session behavior when deleting from an empty dataset."""
+        nonexistent_app_id = str(uuid.uuid4())
+
+        # Should not raise any errors and should return 0
+        deleted_count = delete_draft_variables_batch(nonexistent_app_id, batch_size=10)
+        assert deleted_count == 0
+
+    def test_session_commit_with_single_batch(self, setup_commit_test_data):
+        """Test that commit happens correctly when all data fits in a single batch."""
+        data = setup_commit_test_data
+        app_id = data["app"].id
+
+        with session_factory.create_session() as session:
+            initial_count = session.query(WorkflowDraftVariable).filter_by(app_id=app_id).count()
+        assert initial_count == 10
+
+        # Delete all in a single batch
+        deleted_count = delete_draft_variables_batch(app_id, batch_size=100)
+        assert deleted_count == 10
+
+        # Verify data is persisted
+        with session_factory.create_session() as session:
+            final_count = session.query(WorkflowDraftVariable).filter_by(app_id=app_id).count()
+        assert final_count == 0
+
+    def test_invalid_batch_size_raises_error(self, setup_commit_test_data):
+        """Test that invalid batch size raises ValueError."""
+        data = setup_commit_test_data
+        app_id = data["app"].id
+
+        with pytest.raises(ValueError, match="batch_size must be positive"):
+            delete_draft_variables_batch(app_id, batch_size=0)
+
+        with pytest.raises(ValueError, match="batch_size must be positive"):
+            delete_draft_variables_batch(app_id, batch_size=-1)
+
+    @patch("extensions.ext_storage.storage")
+    def test_session_commit_with_offload_data_cleanup(self, mock_storage, setup_offload_test_data):
+        """Test that session commits correctly when cleaning up offload data."""
+        data = setup_offload_test_data
+        app_id = data["app"].id
+        upload_file_ids = [uf.id for uf in data["upload_files"]]
+        mock_storage.delete.return_value = None
+
+        # Verify initial state
+        with session_factory.create_session() as session:
+            draft_vars_before = session.query(WorkflowDraftVariable).filter_by(app_id=app_id).count()
+            var_files_before = (
+                session.query(WorkflowDraftVariableFile)
+                .where(WorkflowDraftVariableFile.id.in_([vf.id for vf in data["variable_files"]]))
+                .count()
+            )
+            upload_files_before = session.query(UploadFile).where(UploadFile.id.in_(upload_file_ids)).count()
+        assert draft_vars_before == 3
+        assert var_files_before == 2
+        assert upload_files_before == 2
+
+        # Delete variables with offload data
+        deleted_count = delete_draft_variables_batch(app_id, batch_size=10)
+        assert deleted_count == 3
+
+        # Verify all data is persisted (deleted) in new session
+        with session_factory.create_session() as session:
+            draft_vars_after = session.query(WorkflowDraftVariable).filter_by(app_id=app_id).count()
+            var_files_after = (
+                session.query(WorkflowDraftVariableFile)
+                .where(WorkflowDraftVariableFile.id.in_([vf.id for vf in data["variable_files"]]))
+                .count()
+            )
+            upload_files_after = session.query(UploadFile).where(UploadFile.id.in_(upload_file_ids)).count()
+        assert draft_vars_after == 0
+        assert var_files_after == 0
+        assert upload_files_after == 0
+
+        # Verify storage cleanup was called
+        assert mock_storage.delete.call_count == 2

api/tests/test_containers_integration_tests/services/test_account_service.py

@@ -1016,7 +1016,7 @@ class TestAccountService:
 
     def test_delete_account(self, db_session_with_containers, mock_external_service_dependencies):
         """
-        Test account deletion (should add task to queue).
+        Test account deletion (should add task to queue and sync to enterprise).
         """
         fake = Faker()
         email = fake.email()
@@ -1034,10 +1034,18 @@ class TestAccountService:
             password=password,
         )
 
-        with patch("services.account_service.delete_account_task") as mock_delete_task:
+        with (
+            patch("services.account_service.delete_account_task") as mock_delete_task,
+            patch("services.enterprise.account_deletion_sync.sync_account_deletion") as mock_sync,
+        ):
+            mock_sync.return_value = True
+
             # Delete account
             AccountService.delete_account(account)
 
+            # Verify sync was called
+            mock_sync.assert_called_once_with(account_id=account.id, source="account_deleted")
+
             # Verify task was added to queue
             mock_delete_task.delay.assert_called_once_with(account.id)
 
@@ -1716,7 +1724,7 @@ class TestTenantService:
 
     def test_remove_member_from_tenant_success(self, db_session_with_containers, mock_external_service_dependencies):
         """
-        Test successful member removal from tenant.
+        Test successful member removal from tenant (should sync to enterprise).
         """
         fake = Faker()
         tenant_name = fake.company()
@@ -1751,7 +1759,15 @@ class TestTenantService:
         TenantService.create_tenant_member(tenant, member_account, role="normal")
 
         # Remove member
-        TenantService.remove_member_from_tenant(tenant, member_account, owner_account)
+        with patch("services.enterprise.account_deletion_sync.sync_workspace_member_removal") as mock_sync:
+            mock_sync.return_value = True
+
+            TenantService.remove_member_from_tenant(tenant, member_account, owner_account)
+
+            # Verify sync was called
+            mock_sync.assert_called_once_with(
+                workspace_id=tenant.id, member_id=member_account.id, source="workspace_member_removed"
+            )
 
         # Verify member was removed
         from extensions.ext_database import db

api/tests/test_containers_integration_tests/tasks/test_document_indexing_update_task.py

@@ -0,0 +1,182 @@
+from unittest.mock import MagicMock, patch
+
+import pytest
+from faker import Faker
+
+from models import Account, Tenant, TenantAccountJoin, TenantAccountRole
+from models.dataset import Dataset, Document, DocumentSegment
+from tasks.document_indexing_update_task import document_indexing_update_task
+
+
+class TestDocumentIndexingUpdateTask:
+    @pytest.fixture
+    def mock_external_dependencies(self):
+        """Patch external collaborators used by the update task.
+        - IndexProcessorFactory.init_index_processor().clean(...)
+        - IndexingRunner.run([...])
+        """
+        with (
+            patch("tasks.document_indexing_update_task.IndexProcessorFactory") as mock_factory,
+            patch("tasks.document_indexing_update_task.IndexingRunner") as mock_runner,
+        ):
+            processor_instance = MagicMock()
+            mock_factory.return_value.init_index_processor.return_value = processor_instance
+
+            runner_instance = MagicMock()
+            mock_runner.return_value = runner_instance
+
+            yield {
+                "factory": mock_factory,
+                "processor": processor_instance,
+                "runner": mock_runner,
+                "runner_instance": runner_instance,
+            }
+
+    def _create_dataset_document_with_segments(self, db_session_with_containers, *, segment_count: int = 2):
+        fake = Faker()
+
+        # Account and tenant
+        account = Account(
+            email=fake.email(),
+            name=fake.name(),
+            interface_language="en-US",
+            status="active",
+        )
+        db_session_with_containers.add(account)
+        db_session_with_containers.commit()
+
+        tenant = Tenant(name=fake.company(), status="normal")
+        db_session_with_containers.add(tenant)
+        db_session_with_containers.commit()
+
+        join = TenantAccountJoin(
+            tenant_id=tenant.id,
+            account_id=account.id,
+            role=TenantAccountRole.OWNER,
+            current=True,
+        )
+        db_session_with_containers.add(join)
+        db_session_with_containers.commit()
+
+        # Dataset and document
+        dataset = Dataset(
+            tenant_id=tenant.id,
+            name=fake.company(),
+            description=fake.text(max_nb_chars=64),
+            data_source_type="upload_file",
+            indexing_technique="high_quality",
+            created_by=account.id,
+        )
+        db_session_with_containers.add(dataset)
+        db_session_with_containers.commit()
+
+        document = Document(
+            tenant_id=tenant.id,
+            dataset_id=dataset.id,
+            position=0,
+            data_source_type="upload_file",
+            batch="test_batch",
+            name=fake.file_name(),
+            created_from="upload_file",
+            created_by=account.id,
+            indexing_status="waiting",
+            enabled=True,
+            doc_form="text_model",
+        )
+        db_session_with_containers.add(document)
+        db_session_with_containers.commit()
+
+        # Segments
+        node_ids = []
+        for i in range(segment_count):
+            node_id = f"node-{i + 1}"
+            seg = DocumentSegment(
+                tenant_id=tenant.id,
+                dataset_id=dataset.id,
+                document_id=document.id,
+                position=i,
+                content=fake.text(max_nb_chars=32),
+                answer=None,
+                word_count=10,
+                tokens=5,
+                index_node_id=node_id,
+                status="completed",
+                created_by=account.id,
+            )
+            db_session_with_containers.add(seg)
+            node_ids.append(node_id)
+        db_session_with_containers.commit()
+
+        # Refresh to ensure ORM state
+        db_session_with_containers.refresh(dataset)
+        db_session_with_containers.refresh(document)
+
+        return dataset, document, node_ids
+
+    def test_cleans_segments_and_reindexes(self, db_session_with_containers, mock_external_dependencies):
+        dataset, document, node_ids = self._create_dataset_document_with_segments(db_session_with_containers)
+
+        # Act
+        document_indexing_update_task(dataset.id, document.id)
+
+        # Ensure we see committed changes from another session
+        db_session_with_containers.expire_all()
+
+        # Assert document status updated before reindex
+        updated = db_session_with_containers.query(Document).where(Document.id == document.id).first()
+        assert updated.indexing_status == "parsing"
+        assert updated.processing_started_at is not None
+
+        # Segments should be deleted
+        remaining = (
+            db_session_with_containers.query(DocumentSegment).where(DocumentSegment.document_id == document.id).count()
+        )
+        assert remaining == 0
+
+        # Assert index processor clean was called with expected args
+        clean_call = mock_external_dependencies["processor"].clean.call_args
+        assert clean_call is not None
+        args, kwargs = clean_call
+        # args[0] is a Dataset instance (from another session) — validate by id
+        assert getattr(args[0], "id", None) == dataset.id
+        # args[1] should contain our node_ids
+        assert set(args[1]) == set(node_ids)
+        assert kwargs.get("with_keywords") is True
+        assert kwargs.get("delete_child_chunks") is True
+
+        # Assert indexing runner invoked with the updated document
+        run_call = mock_external_dependencies["runner_instance"].run.call_args
+        assert run_call is not None
+        run_docs = run_call[0][0]
+        assert len(run_docs) == 1
+        first = run_docs[0]
+        assert getattr(first, "id", None) == document.id
+
+    def test_clean_error_is_logged_and_indexing_continues(self, db_session_with_containers, mock_external_dependencies):
+        dataset, document, node_ids = self._create_dataset_document_with_segments(db_session_with_containers)
+
+        # Force clean to raise; task should continue to indexing
+        mock_external_dependencies["processor"].clean.side_effect = Exception("boom")
+
+        document_indexing_update_task(dataset.id, document.id)
+
+        # Ensure we see committed changes from another session
+        db_session_with_containers.expire_all()
+
+        # Indexing should still be triggered
+        mock_external_dependencies["runner_instance"].run.assert_called_once()
+
+        # Segments should remain (since clean failed before DB delete)
+        remaining = (
+            db_session_with_containers.query(DocumentSegment).where(DocumentSegment.document_id == document.id).count()
+        )
+        assert remaining > 0
+
+    def test_document_not_found_noop(self, db_session_with_containers, mock_external_dependencies):
+        fake = Faker()
+        # Act with non-existent document id
+        document_indexing_update_task(dataset_id=fake.uuid4(), document_id=fake.uuid4())
+
+        # Neither processor nor runner should be called
+        mock_external_dependencies["processor"].clean.assert_not_called()
+        mock_external_dependencies["runner_instance"].run.assert_not_called()

api/tests/unit_tests/core/workflow/nodes/template_transform/template_transform_node_spec.py

@@ -217,7 +217,6 @@ class TestTemplateTransformNode:
     @patch(
         "core.workflow.nodes.template_transform.template_transform_node.CodeExecutorJinja2TemplateRenderer.render_template"
     )
-    @patch("core.workflow.nodes.template_transform.template_transform_node.MAX_TEMPLATE_TRANSFORM_OUTPUT_LENGTH", 10)
     def test_run_output_length_exceeds_limit(
         self, mock_execute, basic_node_data, mock_graph, mock_graph_runtime_state, graph_init_params
     ):
@@ -231,6 +230,7 @@ class TestTemplateTransformNode:
             graph_init_params=graph_init_params,
             graph=mock_graph,
             graph_runtime_state=mock_graph_runtime_state,
+            max_output_length=10,
         )
 
         result = node._run()

api/tests/unit_tests/services/enterprise/test_account_deletion_sync.py

@@ -0,0 +1,276 @@
+"""Unit tests for account deletion synchronization.
+
+This test module verifies the enterprise account deletion sync functionality,
+including Redis queuing, error handling, and community vs enterprise behavior.
+"""
+
+from unittest.mock import MagicMock, patch
+
+import pytest
+from redis import RedisError
+
+from services.enterprise.account_deletion_sync import (
+    _queue_task,
+    sync_account_deletion,
+    sync_workspace_member_removal,
+)
+
+
+class TestQueueTask:
+    """Unit tests for the _queue_task helper function."""
+
+    @pytest.fixture
+    def mock_redis_client(self):
+        """Mock redis_client for testing."""
+        with patch("services.enterprise.account_deletion_sync.redis_client") as mock_redis:
+            yield mock_redis
+
+    @pytest.fixture
+    def mock_uuid(self):
+        """Mock UUID generation for predictable task IDs."""
+        with patch("services.enterprise.account_deletion_sync.uuid.uuid4") as mock_uuid_gen:
+            mock_uuid_gen.return_value = MagicMock(hex="test-task-id-1234")
+            yield mock_uuid_gen
+
+    def test_queue_task_success(self, mock_redis_client, mock_uuid):
+        """Test successful task queueing to Redis."""
+        # Arrange
+        workspace_id = "ws-123"
+        member_id = "member-456"
+        source = "test_source"
+
+        # Act
+        result = _queue_task(workspace_id=workspace_id, member_id=member_id, source=source)
+
+        # Assert
+        assert result is True
+        mock_redis_client.lpush.assert_called_once()
+
+        # Verify the task payload structure
+        call_args = mock_redis_client.lpush.call_args[0]
+        assert call_args[0] == "enterprise:member:sync:queue"
+
+        import json
+
+        task_data = json.loads(call_args[1])
+        assert task_data["workspace_id"] == workspace_id
+        assert task_data["member_id"] == member_id
+        assert task_data["source"] == source
+        assert task_data["type"] == "sync_member_deletion_from_workspace"
+        assert task_data["retry_count"] == 0
+        assert "task_id" in task_data
+        assert "created_at" in task_data
+
+    def test_queue_task_redis_error(self, mock_redis_client, caplog):
+        """Test handling of Redis connection errors."""
+        # Arrange
+        mock_redis_client.lpush.side_effect = RedisError("Connection failed")
+
+        # Act
+        result = _queue_task(workspace_id="ws-123", member_id="member-456", source="test_source")
+
+        # Assert
+        assert result is False
+        assert "Failed to queue account deletion sync" in caplog.text
+
+    def test_queue_task_type_error(self, mock_redis_client, caplog):
+        """Test handling of JSON serialization errors."""
+        # Arrange
+        mock_redis_client.lpush.side_effect = TypeError("Cannot serialize")
+
+        # Act
+        result = _queue_task(workspace_id="ws-123", member_id="member-456", source="test_source")
+
+        # Assert
+        assert result is False
+        assert "Failed to queue account deletion sync" in caplog.text
+
+
+class TestSyncWorkspaceMemberRemoval:
+    """Unit tests for sync_workspace_member_removal function."""
+
+    @pytest.fixture
+    def mock_queue_task(self):
+        """Mock _queue_task for testing."""
+        with patch("services.enterprise.account_deletion_sync._queue_task") as mock_queue:
+            mock_queue.return_value = True
+            yield mock_queue
+
+    def test_sync_workspace_member_removal_enterprise_enabled(self, mock_queue_task):
+        """Test sync when ENTERPRISE_ENABLED is True."""
+        # Arrange
+        workspace_id = "ws-123"
+        member_id = "member-456"
+        source = "workspace_member_removed"
+
+        with patch("services.enterprise.account_deletion_sync.dify_config") as mock_config:
+            mock_config.ENTERPRISE_ENABLED = True
+
+            # Act
+            result = sync_workspace_member_removal(workspace_id=workspace_id, member_id=member_id, source=source)
+
+            # Assert
+            assert result is True
+            mock_queue_task.assert_called_once_with(workspace_id=workspace_id, member_id=member_id, source=source)
+
+    def test_sync_workspace_member_removal_enterprise_disabled(self, mock_queue_task):
+        """Test sync when ENTERPRISE_ENABLED is False (community edition)."""
+        # Arrange
+        with patch("services.enterprise.account_deletion_sync.dify_config") as mock_config:
+            mock_config.ENTERPRISE_ENABLED = False
+
+            # Act
+            result = sync_workspace_member_removal(workspace_id="ws-123", member_id="member-456", source="test_source")
+
+            # Assert
+            assert result is True
+            mock_queue_task.assert_not_called()
+
+    def test_sync_workspace_member_removal_queue_failure(self, mock_queue_task):
+        """Test handling of queue task failures."""
+        # Arrange
+        mock_queue_task.return_value = False
+
+        with patch("services.enterprise.account_deletion_sync.dify_config") as mock_config:
+            mock_config.ENTERPRISE_ENABLED = True
+
+            # Act
+            result = sync_workspace_member_removal(workspace_id="ws-123", member_id="member-456", source="test_source")
+
+            # Assert
+            assert result is False
+
+
+class TestSyncAccountDeletion:
+    """Unit tests for sync_account_deletion function."""
+
+    @pytest.fixture
+    def mock_db_session(self):
+        """Mock database session for testing."""
+        with patch("services.enterprise.account_deletion_sync.db.session") as mock_session:
+            yield mock_session
+
+    @pytest.fixture
+    def mock_queue_task(self):
+        """Mock _queue_task for testing."""
+        with patch("services.enterprise.account_deletion_sync._queue_task") as mock_queue:
+            mock_queue.return_value = True
+            yield mock_queue
+
+    def test_sync_account_deletion_enterprise_disabled(self, mock_db_session, mock_queue_task):
+        """Test sync when ENTERPRISE_ENABLED is False (community edition)."""
+        # Arrange
+        with patch("services.enterprise.account_deletion_sync.dify_config") as mock_config:
+            mock_config.ENTERPRISE_ENABLED = False
+
+            # Act
+            result = sync_account_deletion(account_id="acc-123", source="account_deleted")
+
+            # Assert
+            assert result is True
+            mock_db_session.query.assert_not_called()
+            mock_queue_task.assert_not_called()
+
+    def test_sync_account_deletion_multiple_workspaces(self, mock_db_session, mock_queue_task):
+        """Test sync for account with multiple workspace memberships."""
+        # Arrange
+        account_id = "acc-123"
+
+        # Mock workspace joins
+        mock_join1 = MagicMock()
+        mock_join1.tenant_id = "tenant-1"
+        mock_join2 = MagicMock()
+        mock_join2.tenant_id = "tenant-2"
+        mock_join3 = MagicMock()
+        mock_join3.tenant_id = "tenant-3"
+
+        mock_query = MagicMock()
+        mock_query.filter_by.return_value.all.return_value = [mock_join1, mock_join2, mock_join3]
+        mock_db_session.query.return_value = mock_query
+
+        with patch("services.enterprise.account_deletion_sync.dify_config") as mock_config:
+            mock_config.ENTERPRISE_ENABLED = True
+
+            # Act
+            result = sync_account_deletion(account_id=account_id, source="account_deleted")
+
+            # Assert
+            assert result is True
+            assert mock_queue_task.call_count == 3
+
+            # Verify each workspace was queued
+            mock_queue_task.assert_any_call(workspace_id="tenant-1", member_id=account_id, source="account_deleted")
+            mock_queue_task.assert_any_call(workspace_id="tenant-2", member_id=account_id, source="account_deleted")
+            mock_queue_task.assert_any_call(workspace_id="tenant-3", member_id=account_id, source="account_deleted")
+
+    def test_sync_account_deletion_no_workspaces(self, mock_db_session, mock_queue_task):
+        """Test sync for account with no workspace memberships."""
+        # Arrange
+        mock_query = MagicMock()
+        mock_query.filter_by.return_value.all.return_value = []
+        mock_db_session.query.return_value = mock_query
+
+        with patch("services.enterprise.account_deletion_sync.dify_config") as mock_config:
+            mock_config.ENTERPRISE_ENABLED = True
+
+            # Act
+            result = sync_account_deletion(account_id="acc-123", source="account_deleted")
+
+            # Assert
+            assert result is True
+            mock_queue_task.assert_not_called()
+
+    def test_sync_account_deletion_partial_failure(self, mock_db_session, mock_queue_task):
+        """Test sync when some tasks fail to queue."""
+        # Arrange
+        account_id = "acc-123"
+
+        # Mock workspace joins
+        mock_join1 = MagicMock()
+        mock_join1.tenant_id = "tenant-1"
+        mock_join2 = MagicMock()
+        mock_join2.tenant_id = "tenant-2"
+        mock_join3 = MagicMock()
+        mock_join3.tenant_id = "tenant-3"
+
+        mock_query = MagicMock()
+        mock_query.filter_by.return_value.all.return_value = [mock_join1, mock_join2, mock_join3]
+        mock_db_session.query.return_value = mock_query
+
+        # Mock queue_task to fail for second workspace
+        def queue_side_effect(workspace_id, member_id, source):
+            return workspace_id != "tenant-2"
+
+        mock_queue_task.side_effect = queue_side_effect
+
+        with patch("services.enterprise.account_deletion_sync.dify_config") as mock_config:
+            mock_config.ENTERPRISE_ENABLED = True
+
+            # Act
+            result = sync_account_deletion(account_id=account_id, source="account_deleted")
+
+            # Assert
+            assert result is False  # Should return False if any task fails
+            assert mock_queue_task.call_count == 3
+
+    def test_sync_account_deletion_all_failures(self, mock_db_session, mock_queue_task):
+        """Test sync when all tasks fail to queue."""
+        # Arrange
+        mock_join = MagicMock()
+        mock_join.tenant_id = "tenant-1"
+
+        mock_query = MagicMock()
+        mock_query.filter_by.return_value.all.return_value = [mock_join]
+        mock_db_session.query.return_value = mock_query
+
+        mock_queue_task.return_value = False
+
+        with patch("services.enterprise.account_deletion_sync.dify_config") as mock_config:
+            mock_config.ENTERPRISE_ENABLED = True
+
+            # Act
+            result = sync_account_deletion(account_id="acc-123", source="account_deleted")
+
+            # Assert
+            assert result is False
+            mock_queue_task.assert_called_once()

api/tests/unit_tests/tasks/test_remove_app_and_related_data_task.py

@@ -350,7 +350,7 @@ class TestDeleteWorkflowArchiveLogs:
         mock_query.where.return_value = mock_delete_query
         mock_db.session.query.return_value = mock_query
 
-        delete_func("log-1")
+        delete_func(mock_db.session, "log-1")
 
         mock_db.session.query.assert_called_once_with(WorkflowArchiveLog)
         mock_query.where.assert_called_once()

api/uv.lock

@@ -1368,7 +1368,7 @@ wheels = [
 
 [[package]]
 name = "dify-api"
-version = "1.12.0"
+version = "1.12.1"
 source = { virtual = "." }
 dependencies = [
     { name = "aliyun-log-python-sdk" },

docker/docker-compose-template.yaml

@@ -21,7 +21,7 @@ services:
 
   # API service
   api:
-    image: langgenius/dify-api:1.12.0
+    image: langgenius/dify-api:1.12.1
     restart: always
     environment:
       # Use the shared environment variables.
@@ -63,7 +63,7 @@ services:
   # worker service
   # The Celery worker for processing all queues (dataset, workflow, mail, etc.)
   worker:
-    image: langgenius/dify-api:1.12.0
+    image: langgenius/dify-api:1.12.1
     restart: always
     environment:
       # Use the shared environment variables.
@@ -102,7 +102,7 @@ services:
   # worker_beat service
   # Celery beat for scheduling periodic tasks.
   worker_beat:
-    image: langgenius/dify-api:1.12.0
+    image: langgenius/dify-api:1.12.1
     restart: always
     environment:
       # Use the shared environment variables.
@@ -132,7 +132,7 @@ services:
 
   # Frontend web application.
   web:
-    image: langgenius/dify-web:1.12.0
+    image: langgenius/dify-web:1.12.1
     restart: always
     environment:
       CONSOLE_API_URL: ${CONSOLE_API_URL:-}

docker/docker-compose.yaml

@@ -707,7 +707,7 @@ services:
 
   # API service
   api:
-    image: langgenius/dify-api:1.12.0
+    image: langgenius/dify-api:1.12.1
     restart: always
     environment:
       # Use the shared environment variables.
@@ -749,7 +749,7 @@ services:
   # worker service
   # The Celery worker for processing all queues (dataset, workflow, mail, etc.)
   worker:
-    image: langgenius/dify-api:1.12.0
+    image: langgenius/dify-api:1.12.1
     restart: always
     environment:
       # Use the shared environment variables.
@@ -788,7 +788,7 @@ services:
   # worker_beat service
   # Celery beat for scheduling periodic tasks.
   worker_beat:
-    image: langgenius/dify-api:1.12.0
+    image: langgenius/dify-api:1.12.1
     restart: always
     environment:
       # Use the shared environment variables.
@@ -818,7 +818,7 @@ services:
 
   # Frontend web application.
   web:
-    image: langgenius/dify-web:1.12.0
+    image: langgenius/dify-web:1.12.1
     restart: always
     environment:
       CONSOLE_API_URL: ${CONSOLE_API_URL:-}

web/app/components/app/create-app-modal/index.spec.tsx

@@ -1,3 +1,4 @@
+import type { App } from '@/types/app'
 import { fireEvent, render, screen, waitFor } from '@testing-library/react'
 import { useRouter } from 'next/navigation'
 import { afterAll, beforeEach, describe, expect, it, vi } from 'vitest'
@@ -13,8 +14,8 @@ import { getRedirection } from '@/utils/app-redirection'
 import CreateAppModal from './index'
 
 vi.mock('ahooks', () => ({
-  useDebounceFn: (fn: (...args: any[]) => any) => {
-    const run = (...args: any[]) => fn(...args)
+  useDebounceFn: <T extends (...args: unknown[]) => unknown>(fn: T) => {
+    const run = (...args: Parameters<T>) => fn(...args)
     const cancel = vi.fn()
     const flush = vi.fn()
     return { run, cancel, flush }
@@ -83,7 +84,7 @@ describe('CreateAppModal', () => {
 
   beforeEach(() => {
     vi.clearAllMocks()
-    mockUseRouter.mockReturnValue({ push: mockPush } as any)
+    mockUseRouter.mockReturnValue({ push: mockPush } as unknown as ReturnType<typeof useRouter>)
     mockUseProviderContext.mockReturnValue({
       plan: {
         type: AppModeEnum.ADVANCED_CHAT,
@@ -92,10 +93,10 @@ describe('CreateAppModal', () => {
         reset: {},
       },
       enableBilling: true,
-    } as any)
+    } as unknown as ReturnType<typeof useProviderContext>)
     mockUseAppContext.mockReturnValue({
       isCurrentWorkspaceEditor: true,
-    } as any)
+    } as unknown as ReturnType<typeof useAppContext>)
     mockSetItem.mockClear()
     Object.defineProperty(window, 'localStorage', {
       value: {
@@ -118,8 +119,8 @@ describe('CreateAppModal', () => {
   })
 
   it('creates an app, notifies success, and fires callbacks', async () => {
-    const mockApp = { id: 'app-1', mode: AppModeEnum.ADVANCED_CHAT }
-    mockCreateApp.mockResolvedValue(mockApp as any)
+    const mockApp: Partial<App> = { id: 'app-1', mode: AppModeEnum.ADVANCED_CHAT }
+    mockCreateApp.mockResolvedValue(mockApp as App)
     const { onClose, onSuccess } = renderModal()
 
     const nameInput = screen.getByPlaceholderText('app.newApp.appNamePlaceholder')

web/app/components/base/effect/index.stories.tsx

@@ -1,4 +1,3 @@
-/* eslint-disable tailwindcss/classnames-order */
 import type { Meta, StoryObj } from '@storybook/nextjs-vite'
 import Effect from '.'
 
@@ -29,8 +28,8 @@ type Story = StoryObj<typeof meta>
 export const Playground: Story = {
   render: () => (
     <div className="relative h-40 w-72 overflow-hidden rounded-2xl border border-divider-subtle bg-background-default-subtle">
-      <Effect className="top-6 left-8" />
-      <Effect className="top-14 right-10 bg-util-colors-purple-brand-purple-brand-500" />
+      <Effect className="left-8 top-6" />
+      <Effect className="bg-util-colors-purple-brand-purple-brand-500 right-10 top-14" />
       <div className="absolute inset-x-0 bottom-4 flex justify-center text-xs text-text-secondary">
         Accent glow
       </div>

web/app/components/base/tooltip/index.tsx

@@ -4,7 +4,7 @@ import type { FC } from 'react'
 import { RiQuestionLine } from '@remixicon/react'
 import { useBoolean } from 'ahooks'
 import * as React from 'react'
-import { useEffect, useRef, useState } from 'react'
+import { useCallback, useEffect, useRef, useState } from 'react'
 import { PortalToFollowElem, PortalToFollowElemContent, PortalToFollowElemTrigger } from '@/app/components/base/portal-to-follow-elem'
 import { cn } from '@/utils/classnames'
 import { tooltipManager } from './TooltipManager'
@@ -61,6 +61,20 @@ const Tooltip: FC<TooltipProps> = ({
     isHoverTriggerRef.current = isHoverTrigger
   }, [isHoverTrigger])
 
+  const closeTimeoutRef = useRef<ReturnType<typeof setTimeout> | null>(null)
+  const clearCloseTimeout = useCallback(() => {
+    if (closeTimeoutRef.current) {
+      clearTimeout(closeTimeoutRef.current)
+      closeTimeoutRef.current = null
+    }
+  }, [])
+
+  useEffect(() => {
+    return () => {
+      clearCloseTimeout()
+    }
+  }, [clearCloseTimeout])
+
   const close = () => setOpen(false)
 
   const handleLeave = (isTrigger: boolean) => {
@@ -71,7 +85,9 @@ const Tooltip: FC<TooltipProps> = ({
 
     // give time to move to the popup
     if (needsDelay) {
-      setTimeout(() => {
+      clearCloseTimeout()
+      closeTimeoutRef.current = setTimeout(() => {
+        closeTimeoutRef.current = null
         if (!isHoverPopupRef.current && !isHoverTriggerRef.current) {
           setOpen(false)
           tooltipManager.clear(close)
@@ -79,6 +95,7 @@ const Tooltip: FC<TooltipProps> = ({
       }, 300)
     }
     else {
+      clearCloseTimeout()
       setOpen(false)
       tooltipManager.clear(close)
     }
@@ -95,6 +112,7 @@ const Tooltip: FC<TooltipProps> = ({
         onClick={() => triggerMethod === 'click' && setOpen(v => !v)}
         onMouseEnter={() => {
           if (triggerMethod === 'hover') {
+            clearCloseTimeout()
             setHoverTrigger()
             tooltipManager.register(close)
             setOpen(true)
@@ -115,7 +133,12 @@ const Tooltip: FC<TooltipProps> = ({
               !noDecoration && 'system-xs-regular relative max-w-[300px] break-words rounded-md bg-components-panel-bg px-3 py-2 text-left text-text-tertiary shadow-lg',
               popupClassName,
             )}
-            onMouseEnter={() => triggerMethod === 'hover' && setHoverPopup()}
+            onMouseEnter={() => {
+              if (triggerMethod === 'hover') {
+                clearCloseTimeout()
+                setHoverPopup()
+              }
+            }}
             onMouseLeave={() => triggerMethod === 'hover' && handleLeave(false)}
           >
             {popupContent}

web/app/components/datasets/common/image-uploader/utils.spec.ts

@@ -216,13 +216,22 @@ describe('image-uploader utils', () => {
     type FileCallback = (file: MockFile) => void
     type EntriesCallback = (entries: FileSystemEntry[]) => void
 
+    // Helper to create mock FileSystemEntry with required properties
+    const createMockEntry = (props: {
+      isFile: boolean
+      isDirectory: boolean
+      name?: string
+      file?: (callback: FileCallback) => void
+      createReader?: () => { readEntries: (callback: EntriesCallback) => void }
+    }): FileSystemEntry => props as unknown as FileSystemEntry
+
     it('should resolve with file array for file entry', async () => {
       const mockFile: MockFile = { name: 'test.png' }
-      const mockEntry = {
+      const mockEntry = createMockEntry({
         isFile: true,
         isDirectory: false,
         file: (callback: FileCallback) => callback(mockFile),
-      }
+      })
 
       const result = await traverseFileEntry(mockEntry)
       expect(result).toHaveLength(1)
@@ -232,11 +241,11 @@ describe('image-uploader utils', () => {
 
     it('should resolve with file array with prefix for nested file', async () => {
       const mockFile: MockFile = { name: 'test.png' }
-      const mockEntry = {
+      const mockEntry = createMockEntry({
         isFile: true,
         isDirectory: false,
         file: (callback: FileCallback) => callback(mockFile),
-      }
+      })
 
       const result = await traverseFileEntry(mockEntry, 'folder/')
       expect(result).toHaveLength(1)
@@ -244,24 +253,24 @@ describe('image-uploader utils', () => {
     })
 
     it('should resolve empty array for unknown entry type', async () => {
-      const mockEntry = {
+      const mockEntry = createMockEntry({
         isFile: false,
         isDirectory: false,
-      }
+      })
 
       const result = await traverseFileEntry(mockEntry)
       expect(result).toEqual([])
     })
 
     it('should handle directory with no files', async () => {
-      const mockEntry = {
+      const mockEntry = createMockEntry({
         isFile: false,
         isDirectory: true,
         name: 'empty-folder',
         createReader: () => ({
           readEntries: (callback: EntriesCallback) => callback([]),
         }),
-      }
+      })
 
       const result = await traverseFileEntry(mockEntry)
       expect(result).toEqual([])
@@ -271,20 +280,20 @@ describe('image-uploader utils', () => {
       const mockFile1: MockFile = { name: 'file1.png' }
       const mockFile2: MockFile = { name: 'file2.png' }
 
-      const mockFileEntry1 = {
+      const mockFileEntry1 = createMockEntry({
         isFile: true,
         isDirectory: false,
         file: (callback: FileCallback) => callback(mockFile1),
-      }
+      })
 
-      const mockFileEntry2 = {
+      const mockFileEntry2 = createMockEntry({
         isFile: true,
         isDirectory: false,
         file: (callback: FileCallback) => callback(mockFile2),
-      }
+      })
 
       let readCount = 0
-      const mockEntry = {
+      const mockEntry = createMockEntry({
         isFile: false,
         isDirectory: true,
         name: 'folder',
@@ -292,14 +301,14 @@ describe('image-uploader utils', () => {
           readEntries: (callback: EntriesCallback) => {
             if (readCount === 0) {
               readCount++
-              callback([mockFileEntry1, mockFileEntry2] as unknown as FileSystemEntry[])
+              callback([mockFileEntry1, mockFileEntry2])
             }
             else {
               callback([])
             }
           },
         }),
-      }
+      })
 
       const result = await traverseFileEntry(mockEntry)
       expect(result).toHaveLength(2)

web/app/components/datasets/common/image-uploader/utils.ts

@@ -18,17 +18,17 @@ type FileWithPath = {
   relativePath?: string
 } & File
 
-export const traverseFileEntry = (entry: any, prefix = ''): Promise<FileWithPath[]> => {
+export const traverseFileEntry = (entry: FileSystemEntry, prefix = ''): Promise<FileWithPath[]> => {
   return new Promise((resolve) => {
     if (entry.isFile) {
-      entry.file((file: FileWithPath) => {
+      (entry as FileSystemFileEntry).file((file: FileWithPath) => {
         file.relativePath = `${prefix}${file.name}`
         resolve([file])
       })
     }
     else if (entry.isDirectory) {
-      const reader = entry.createReader()
-      const entries: any[] = []
+      const reader = (entry as FileSystemDirectoryEntry).createReader()
+      const entries: FileSystemEntry[] = []
       const read = () => {
         reader.readEntries(async (results: FileSystemEntry[]) => {
           if (!results.length) {

web/app/components/datasets/create-from-pipeline/create-options/create-from-dsl-modal/hooks/use-dsl-import.ts

@@ -0,0 +1,218 @@
+'use client'
+import { useDebounceFn } from 'ahooks'
+import { useRouter } from 'next/navigation'
+import { useCallback, useMemo, useRef, useState } from 'react'
+import { useTranslation } from 'react-i18next'
+import { useContext } from 'use-context-selector'
+import { ToastContext } from '@/app/components/base/toast'
+import { usePluginDependencies } from '@/app/components/workflow/plugin-dependency/hooks'
+import {
+  DSLImportMode,
+  DSLImportStatus,
+} from '@/models/app'
+import { useImportPipelineDSL, useImportPipelineDSLConfirm } from '@/service/use-pipeline'
+
+export enum CreateFromDSLModalTab {
+  FROM_FILE = 'from-file',
+  FROM_URL = 'from-url',
+}
+
+export type UseDSLImportOptions = {
+  activeTab?: CreateFromDSLModalTab
+  dslUrl?: string
+  onSuccess?: () => void
+  onClose?: () => void
+}
+
+export type DSLVersions = {
+  importedVersion: string
+  systemVersion: string
+}
+
+export const useDSLImport = ({
+  activeTab = CreateFromDSLModalTab.FROM_FILE,
+  dslUrl = '',
+  onSuccess,
+  onClose,
+}: UseDSLImportOptions) => {
+  const { push } = useRouter()
+  const { t } = useTranslation()
+  const { notify } = useContext(ToastContext)
+
+  const [currentFile, setDSLFile] = useState<File>()
+  const [fileContent, setFileContent] = useState<string>()
+  const [currentTab, setCurrentTab] = useState(activeTab)
+  const [dslUrlValue, setDslUrlValue] = useState(dslUrl)
+  const [showConfirmModal, setShowConfirmModal] = useState(false)
+  const [versions, setVersions] = useState<DSLVersions>()
+  const [importId, setImportId] = useState<string>()
+  const [isConfirming, setIsConfirming] = useState(false)
+
+  const { handleCheckPluginDependencies } = usePluginDependencies()
+  const isCreatingRef = useRef(false)
+
+  const { mutateAsync: importDSL } = useImportPipelineDSL()
+  const { mutateAsync: importDSLConfirm } = useImportPipelineDSLConfirm()
+
+  const readFile = useCallback((file: File) => {
+    const reader = new FileReader()
+    reader.onload = (event) => {
+      const content = event.target?.result
+      setFileContent(content as string)
+    }
+    reader.readAsText(file)
+  }, [])
+
+  const handleFile = useCallback((file?: File) => {
+    setDSLFile(file)
+    if (file)
+      readFile(file)
+    if (!file)
+      setFileContent('')
+  }, [readFile])
+
+  const onCreate = useCallback(async () => {
+    if (currentTab === CreateFromDSLModalTab.FROM_FILE && !currentFile)
+      return
+    if (currentTab === CreateFromDSLModalTab.FROM_URL && !dslUrlValue)
+      return
+    if (isCreatingRef.current)
+      return
+
+    isCreatingRef.current = true
+
+    let response
+    if (currentTab === CreateFromDSLModalTab.FROM_FILE) {
+      response = await importDSL({
+        mode: DSLImportMode.YAML_CONTENT,
+        yaml_content: fileContent || '',
+      })
+    }
+    if (currentTab === CreateFromDSLModalTab.FROM_URL) {
+      response = await importDSL({
+        mode: DSLImportMode.YAML_URL,
+        yaml_url: dslUrlValue || '',
+      })
+    }
+
+    if (!response) {
+      notify({ type: 'error', message: t('creation.errorTip', { ns: 'datasetPipeline' }) })
+      isCreatingRef.current = false
+      return
+    }
+
+    const { id, status, pipeline_id, dataset_id, imported_dsl_version, current_dsl_version } = response
+
+    if (status === DSLImportStatus.COMPLETED || status === DSLImportStatus.COMPLETED_WITH_WARNINGS) {
+      onSuccess?.()
+      onClose?.()
+
+      notify({
+        type: status === DSLImportStatus.COMPLETED ? 'success' : 'warning',
+        message: t(status === DSLImportStatus.COMPLETED ? 'creation.successTip' : 'creation.caution', { ns: 'datasetPipeline' }),
+        children: status === DSLImportStatus.COMPLETED_WITH_WARNINGS && t('newApp.appCreateDSLWarning', { ns: 'app' }),
+      })
+
+      if (pipeline_id)
+        await handleCheckPluginDependencies(pipeline_id, true)
+
+      push(`/datasets/${dataset_id}/pipeline`)
+      isCreatingRef.current = false
+    }
+    else if (status === DSLImportStatus.PENDING) {
+      setVersions({
+        importedVersion: imported_dsl_version ?? '',
+        systemVersion: current_dsl_version ?? '',
+      })
+      onClose?.()
+      setTimeout(() => {
+        setShowConfirmModal(true)
+      }, 300)
+      setImportId(id)
+      isCreatingRef.current = false
+    }
+    else {
+      notify({ type: 'error', message: t('creation.errorTip', { ns: 'datasetPipeline' }) })
+      isCreatingRef.current = false
+    }
+  }, [
+    currentTab,
+    currentFile,
+    dslUrlValue,
+    fileContent,
+    importDSL,
+    notify,
+    t,
+    onSuccess,
+    onClose,
+    handleCheckPluginDependencies,
+    push,
+  ])
+
+  const { run: handleCreateApp } = useDebounceFn(onCreate, { wait: 300 })
+
+  const onDSLConfirm = useCallback(async () => {
+    if (!importId)
+      return
+
+    setIsConfirming(true)
+    const response = await importDSLConfirm(importId)
+    setIsConfirming(false)
+
+    if (!response) {
+      notify({ type: 'error', message: t('creation.errorTip', { ns: 'datasetPipeline' }) })
+      return
+    }
+
+    const { status, pipeline_id, dataset_id } = response
+
+    if (status === DSLImportStatus.COMPLETED) {
+      onSuccess?.()
+      setShowConfirmModal(false)
+
+      notify({
+        type: 'success',
+        message: t('creation.successTip', { ns: 'datasetPipeline' }),
+      })
+
+      if (pipeline_id)
+        await handleCheckPluginDependencies(pipeline_id, true)
+
+      push(`/datasets/${dataset_id}/pipeline`)
+    }
+    else if (status === DSLImportStatus.FAILED) {
+      notify({ type: 'error', message: t('creation.errorTip', { ns: 'datasetPipeline' }) })
+    }
+  }, [importId, importDSLConfirm, notify, t, onSuccess, handleCheckPluginDependencies, push])
+
+  const handleCancelConfirm = useCallback(() => {
+    setShowConfirmModal(false)
+  }, [])
+
+  const buttonDisabled = useMemo(() => {
+    if (currentTab === CreateFromDSLModalTab.FROM_FILE)
+      return !currentFile
+    if (currentTab === CreateFromDSLModalTab.FROM_URL)
+      return !dslUrlValue
+    return false
+  }, [currentTab, currentFile, dslUrlValue])
+
+  return {
+    // State
+    currentFile,
+    currentTab,
+    dslUrlValue,
+    showConfirmModal,
+    versions,
+    buttonDisabled,
+    isConfirming,
+
+    // Actions
+    setCurrentTab,
+    setDslUrlValue,
+    handleFile,
+    handleCreateApp,
+    onDSLConfirm,
+    handleCancelConfirm,
+  }
+}

web/app/components/datasets/create-from-pipeline/create-options/create-from-dsl-modal/index.tsx

@@ -1,24 +1,18 @@
 'use client'
-import { useDebounceFn, useKeyPress } from 'ahooks'
+import { useKeyPress } from 'ahooks'
 import { noop } from 'es-toolkit/function'
-import { useRouter } from 'next/navigation'
-import { useMemo, useRef, useState } from 'react'
 import { useTranslation } from 'react-i18next'
-import { useContext } from 'use-context-selector'
 import Button from '@/app/components/base/button'
 import Input from '@/app/components/base/input'
 import Modal from '@/app/components/base/modal'
-import { ToastContext } from '@/app/components/base/toast'
-import { usePluginDependencies } from '@/app/components/workflow/plugin-dependency/hooks'
-import {
-  DSLImportMode,
-  DSLImportStatus,
-} from '@/models/app'
-import { useImportPipelineDSL, useImportPipelineDSLConfirm } from '@/service/use-pipeline'
+import DSLConfirmModal from './dsl-confirm-modal'
 import Header from './header'
+import { CreateFromDSLModalTab, useDSLImport } from './hooks/use-dsl-import'
 import Tab from './tab'
 import Uploader from './uploader'
 
+export { CreateFromDSLModalTab }
+
 type CreateFromDSLModalProps = {
   show: boolean
   onSuccess?: () => void
@@ -27,11 +21,6 @@ type CreateFromDSLModalProps = {
   dslUrl?: string
 }
 
-export enum CreateFromDSLModalTab {
-  FROM_FILE = 'from-file',
-  FROM_URL = 'from-url',
-}
-
 const CreateFromDSLModal = ({
   show,
   onSuccess,
@@ -39,149 +28,33 @@ const CreateFromDSLModal = ({
   activeTab = CreateFromDSLModalTab.FROM_FILE,
   dslUrl = '',
 }: CreateFromDSLModalProps) => {
-  const { push } = useRouter()
   const { t } = useTranslation()
-  const { notify } = useContext(ToastContext)
-  const [currentFile, setDSLFile] = useState<File>()
-  const [fileContent, setFileContent] = useState<string>()
-  const [currentTab, setCurrentTab] = useState(activeTab)
-  const [dslUrlValue, setDslUrlValue] = useState(dslUrl)
-  const [showErrorModal, setShowErrorModal] = useState(false)
-  const [versions, setVersions] = useState<{ importedVersion: string, systemVersion: string }>()
-  const [importId, setImportId] = useState<string>()
-  const { handleCheckPluginDependencies } = usePluginDependencies()
 
-  const readFile = (file: File) => {
-    const reader = new FileReader()
-    reader.onload = function (event) {
-      const content = event.target?.result
-      setFileContent(content as string)
-    }
-    reader.readAsText(file)
-  }
-
-  const handleFile = (file?: File) => {
-    setDSLFile(file)
-    if (file)
-      readFile(file)
-    if (!file)
-      setFileContent('')
-  }
-
-  const isCreatingRef = useRef(false)
-
-  const { mutateAsync: importDSL } = useImportPipelineDSL()
-
-  const onCreate = async () => {
-    if (currentTab === CreateFromDSLModalTab.FROM_FILE && !currentFile)
-      return
-    if (currentTab === CreateFromDSLModalTab.FROM_URL && !dslUrlValue)
-      return
-    if (isCreatingRef.current)
-      return
-    isCreatingRef.current = true
-    let response
-    if (currentTab === CreateFromDSLModalTab.FROM_FILE) {
-      response = await importDSL({
-        mode: DSLImportMode.YAML_CONTENT,
-        yaml_content: fileContent || '',
-      })
-    }
-    if (currentTab === CreateFromDSLModalTab.FROM_URL) {
-      response = await importDSL({
-        mode: DSLImportMode.YAML_URL,
-        yaml_url: dslUrlValue || '',
-      })
-    }
-
-    if (!response) {
-      notify({ type: 'error', message: t('creation.errorTip', { ns: 'datasetPipeline' }) })
-      isCreatingRef.current = false
-      return
-    }
-    const { id, status, pipeline_id, dataset_id, imported_dsl_version, current_dsl_version } = response
-    if (status === DSLImportStatus.COMPLETED || status === DSLImportStatus.COMPLETED_WITH_WARNINGS) {
-      if (onSuccess)
-        onSuccess()
-      if (onClose)
-        onClose()
-
-      notify({
-        type: status === DSLImportStatus.COMPLETED ? 'success' : 'warning',
-        message: t(status === DSLImportStatus.COMPLETED ? 'creation.successTip' : 'creation.caution', { ns: 'datasetPipeline' }),
-        children: status === DSLImportStatus.COMPLETED_WITH_WARNINGS && t('newApp.appCreateDSLWarning', { ns: 'app' }),
-      })
-      if (pipeline_id)
-        await handleCheckPluginDependencies(pipeline_id, true)
-      push(`/datasets/${dataset_id}/pipeline`)
-      isCreatingRef.current = false
-    }
-    else if (status === DSLImportStatus.PENDING) {
-      setVersions({
-        importedVersion: imported_dsl_version ?? '',
-        systemVersion: current_dsl_version ?? '',
-      })
-      if (onClose)
-        onClose()
-      setTimeout(() => {
-        setShowErrorModal(true)
-      }, 300)
-      setImportId(id)
-      isCreatingRef.current = false
-    }
-    else {
-      notify({ type: 'error', message: t('creation.errorTip', { ns: 'datasetPipeline' }) })
-      isCreatingRef.current = false
-    }
-  }
-
-  const { run: handleCreateApp } = useDebounceFn(onCreate, { wait: 300 })
-
-  useKeyPress('esc', () => {
-    if (show && !showErrorModal)
-      onClose()
+  const {
+    currentFile,
+    currentTab,
+    dslUrlValue,
+    showConfirmModal,
+    versions,
+    buttonDisabled,
+    isConfirming,
+    setCurrentTab,
+    setDslUrlValue,
+    handleFile,
+    handleCreateApp,
+    onDSLConfirm,
+    handleCancelConfirm,
+  } = useDSLImport({
+    activeTab,
+    dslUrl,
+    onSuccess,
+    onClose,
   })
 
-  const { mutateAsync: importDSLConfirm } = useImportPipelineDSLConfirm()
-
-  const onDSLConfirm = async () => {
-    if (!importId)
-      return
-    const response = await importDSLConfirm(importId)
-
-    if (!response) {
-      notify({ type: 'error', message: t('creation.errorTip', { ns: 'datasetPipeline' }) })
-      return
-    }
-
-    const { status, pipeline_id, dataset_id } = response
-
-    if (status === DSLImportStatus.COMPLETED) {
-      if (onSuccess)
-        onSuccess()
-      if (onClose)
-        onClose()
-
-      notify({
-        type: 'success',
-        message: t('creation.successTip', { ns: 'datasetPipeline' }),
-      })
-      if (pipeline_id)
-        await handleCheckPluginDependencies(pipeline_id, true)
-      push(`datasets/${dataset_id}/pipeline`)
-    }
-    else if (status === DSLImportStatus.FAILED) {
-      notify({ type: 'error', message: t('creation.errorTip', { ns: 'datasetPipeline' }) })
-    }
-  }
-
-  const buttonDisabled = useMemo(() => {
-    if (currentTab === CreateFromDSLModalTab.FROM_FILE)
-      return !currentFile
-    if (currentTab === CreateFromDSLModalTab.FROM_URL)
-      return !dslUrlValue
-    return false
-  }, [currentTab, currentFile, dslUrlValue])
+  useKeyPress('esc', () => {
+    if (show && !showConfirmModal)
+      onClose()
+  })
 
   return (
     <>
@@ -196,29 +69,25 @@ const CreateFromDSLModal = ({
           setCurrentTab={setCurrentTab}
         />
         <div className="px-6 py-4">
-          {
-            currentTab === CreateFromDSLModalTab.FROM_FILE && (
-              <Uploader
-                className="mt-0"
-                file={currentFile}
-                updateFile={handleFile}
-              />
-            )
-          }
-          {
-            currentTab === CreateFromDSLModalTab.FROM_URL && (
-              <div>
-                <div className="system-md-semibold leading6 mb-1 text-text-secondary">
-                  DSL URL
-                </div>
-                <Input
-                  placeholder={t('importFromDSLUrlPlaceholder', { ns: 'app' }) || ''}
-                  value={dslUrlValue}
-                  onChange={e => setDslUrlValue(e.target.value)}
-                />
+          {currentTab === CreateFromDSLModalTab.FROM_FILE && (
+            <Uploader
+              className="mt-0"
+              file={currentFile}
+              updateFile={handleFile}
+            />
+          )}
+          {currentTab === CreateFromDSLModalTab.FROM_URL && (
+            <div>
+              <div className="system-md-semibold leading6 mb-1 text-text-secondary">
+                DSL URL
               </div>
-            )
-          }
+              <Input
+                placeholder={t('importFromDSLUrlPlaceholder', { ns: 'app' }) || ''}
+                value={dslUrlValue}
+                onChange={e => setDslUrlValue(e.target.value)}
+              />
+            </div>
+          )}
         </div>
         <div className="flex justify-end gap-x-2 p-6 pt-5">
           <Button onClick={onClose}>
@@ -234,32 +103,14 @@ const CreateFromDSLModal = ({
           </Button>
         </div>
       </Modal>
-      <Modal
-        isShow={showErrorModal}
-        onClose={() => setShowErrorModal(false)}
-        className="w-[480px]"
-      >
-        <div className="flex flex-col items-start gap-2 self-stretch pb-4">
-          <div className="title-2xl-semi-bold text-text-primary">{t('newApp.appCreateDSLErrorTitle', { ns: 'app' })}</div>
-          <div className="system-md-regular flex grow flex-col text-text-secondary">
-            <div>{t('newApp.appCreateDSLErrorPart1', { ns: 'app' })}</div>
-            <div>{t('newApp.appCreateDSLErrorPart2', { ns: 'app' })}</div>
-            <br />
-            <div>
-              {t('newApp.appCreateDSLErrorPart3', { ns: 'app' })}
-              <span className="system-md-medium">{versions?.importedVersion}</span>
-            </div>
-            <div>
-              {t('newApp.appCreateDSLErrorPart4', { ns: 'app' })}
-              <span className="system-md-medium">{versions?.systemVersion}</span>
-            </div>
-          </div>
-        </div>
-        <div className="flex items-start justify-end gap-2 self-stretch pt-6">
-          <Button variant="secondary" onClick={() => setShowErrorModal(false)}>{t('newApp.Cancel', { ns: 'app' })}</Button>
-          <Button variant="primary" destructive onClick={onDSLConfirm}>{t('newApp.Confirm', { ns: 'app' })}</Button>
-        </div>
-      </Modal>
+      {showConfirmModal && (
+        <DSLConfirmModal
+          versions={versions}
+          onCancel={handleCancelConfirm}
+          onConfirm={onDSLConfirm}
+          confirmDisabled={isConfirming}
+        />
+      )}
     </>
   )
 }

web/app/components/datasets/create/file-uploader/components/file-list-item.spec.tsx

@@ -0,0 +1,334 @@
+import type { FileListItemProps } from './file-list-item'
+import type { CustomFile as File, FileItem } from '@/models/datasets'
+import { fireEvent, render, screen } from '@testing-library/react'
+import { beforeEach, describe, expect, it, vi } from 'vitest'
+import { PROGRESS_COMPLETE, PROGRESS_ERROR, PROGRESS_NOT_STARTED } from '../constants'
+import FileListItem from './file-list-item'
+
+// Mock theme hook - can be changed per test
+let mockTheme = 'light'
+vi.mock('@/hooks/use-theme', () => ({
+  default: () => ({ theme: mockTheme }),
+}))
+
+// Mock theme types
+vi.mock('@/types/app', () => ({
+  Theme: { dark: 'dark', light: 'light' },
+}))
+
+// Mock SimplePieChart with dynamic import handling
+vi.mock('next/dynamic', () => ({
+  default: () => {
+    const DynamicComponent = ({ percentage, stroke, fill }: { percentage: number, stroke: string, fill: string }) => (
+      <div data-testid="pie-chart" data-percentage={percentage} data-stroke={stroke} data-fill={fill}>
+        Pie Chart:
+        {' '}
+        {percentage}
+        %
+      </div>
+    )
+    DynamicComponent.displayName = 'SimplePieChart'
+    return DynamicComponent
+  },
+}))
+
+// Mock DocumentFileIcon
+vi.mock('@/app/components/datasets/common/document-file-icon', () => ({
+  default: ({ name, extension, size }: { name: string, extension: string, size: string }) => (
+    <div data-testid="document-icon" data-name={name} data-extension={extension} data-size={size}>
+      Document Icon
+    </div>
+  ),
+}))
+
+describe('FileListItem', () => {
+  const createMockFile = (overrides: Partial<File> = {}): File => ({
+    name: 'test-document.pdf',
+    size: 1024 * 100, // 100KB
+    type: 'application/pdf',
+    lastModified: Date.now(),
+    ...overrides,
+  } as File)
+
+  const createMockFileItem = (overrides: Partial<FileItem> = {}): FileItem => ({
+    fileID: 'file-123',
+    file: createMockFile(overrides.file as Partial<File>),
+    progress: PROGRESS_NOT_STARTED,
+    ...overrides,
+  })
+
+  const defaultProps: FileListItemProps = {
+    fileItem: createMockFileItem(),
+    onPreview: vi.fn(),
+    onRemove: vi.fn(),
+  }
+
+  beforeEach(() => {
+    vi.clearAllMocks()
+    mockTheme = 'light'
+  })
+
+  describe('rendering', () => {
+    it('should render the file item container', () => {
+      const { container } = render(<FileListItem {...defaultProps} />)
+      const item = container.firstChild as HTMLElement
+      expect(item).toHaveClass('flex', 'h-12', 'items-center', 'rounded-lg')
+    })
+
+    it('should render document icon with correct props', () => {
+      render(<FileListItem {...defaultProps} />)
+      const icon = screen.getByTestId('document-icon')
+      expect(icon).toBeInTheDocument()
+      expect(icon).toHaveAttribute('data-name', 'test-document.pdf')
+      expect(icon).toHaveAttribute('data-extension', 'pdf')
+      expect(icon).toHaveAttribute('data-size', 'xl')
+    })
+
+    it('should render file name', () => {
+      render(<FileListItem {...defaultProps} />)
+      expect(screen.getByText('test-document.pdf')).toBeInTheDocument()
+    })
+
+    it('should render file extension in uppercase via CSS class', () => {
+      render(<FileListItem {...defaultProps} />)
+      const extensionSpan = screen.getByText('pdf')
+      expect(extensionSpan).toBeInTheDocument()
+      expect(extensionSpan).toHaveClass('uppercase')
+    })
+
+    it('should render file size', () => {
+      render(<FileListItem {...defaultProps} />)
+      // Default mock file is 100KB (1024 * 100 bytes)
+      expect(screen.getByText('100.00 KB')).toBeInTheDocument()
+    })
+
+    it('should render delete button', () => {
+      const { container } = render(<FileListItem {...defaultProps} />)
+      const deleteButton = container.querySelector('.cursor-pointer')
+      expect(deleteButton).toBeInTheDocument()
+    })
+  })
+
+  describe('progress states', () => {
+    it('should show progress chart when uploading (0-99)', () => {
+      const fileItem = createMockFileItem({ progress: 50 })
+      render(<FileListItem {...defaultProps} fileItem={fileItem} />)
+
+      const pieChart = screen.getByTestId('pie-chart')
+      expect(pieChart).toBeInTheDocument()
+      expect(pieChart).toHaveAttribute('data-percentage', '50')
+    })
+
+    it('should show progress chart at 0%', () => {
+      const fileItem = createMockFileItem({ progress: 0 })
+      render(<FileListItem {...defaultProps} fileItem={fileItem} />)
+
+      const pieChart = screen.getByTestId('pie-chart')
+      expect(pieChart).toHaveAttribute('data-percentage', '0')
+    })
+
+    it('should not show progress chart when complete (100)', () => {
+      const fileItem = createMockFileItem({ progress: PROGRESS_COMPLETE })
+      render(<FileListItem {...defaultProps} fileItem={fileItem} />)
+
+      expect(screen.queryByTestId('pie-chart')).not.toBeInTheDocument()
+    })
+
+    it('should not show progress chart when not started (-1)', () => {
+      const fileItem = createMockFileItem({ progress: PROGRESS_NOT_STARTED })
+      render(<FileListItem {...defaultProps} fileItem={fileItem} />)
+
+      expect(screen.queryByTestId('pie-chart')).not.toBeInTheDocument()
+    })
+  })
+
+  describe('error state', () => {
+    it('should show error indicator when progress is PROGRESS_ERROR', () => {
+      const fileItem = createMockFileItem({ progress: PROGRESS_ERROR })
+      const { container } = render(<FileListItem {...defaultProps} fileItem={fileItem} />)
+
+      const errorIndicator = container.querySelector('.text-text-destructive')
+      expect(errorIndicator).toBeInTheDocument()
+    })
+
+    it('should not show error indicator when not in error state', () => {
+      const { container } = render(<FileListItem {...defaultProps} />)
+      const errorIndicator = container.querySelector('.text-text-destructive')
+      expect(errorIndicator).not.toBeInTheDocument()
+    })
+  })
+
+  describe('theme handling', () => {
+    it('should use correct chart color for light theme', () => {
+      mockTheme = 'light'
+      const fileItem = createMockFileItem({ progress: 50 })
+      render(<FileListItem {...defaultProps} fileItem={fileItem} />)
+
+      const pieChart = screen.getByTestId('pie-chart')
+      expect(pieChart).toHaveAttribute('data-stroke', '#296dff')
+      expect(pieChart).toHaveAttribute('data-fill', '#296dff')
+    })
+
+    it('should use correct chart color for dark theme', () => {
+      mockTheme = 'dark'
+      const fileItem = createMockFileItem({ progress: 50 })
+      render(<FileListItem {...defaultProps} fileItem={fileItem} />)
+
+      const pieChart = screen.getByTestId('pie-chart')
+      expect(pieChart).toHaveAttribute('data-stroke', '#5289ff')
+      expect(pieChart).toHaveAttribute('data-fill', '#5289ff')
+    })
+  })
+
+  describe('event handlers', () => {
+    it('should call onPreview when item is clicked with file id', () => {
+      const onPreview = vi.fn()
+      const fileItem = createMockFileItem({
+        file: createMockFile({ id: 'uploaded-id' } as Partial<File>),
+      })
+      render(<FileListItem {...defaultProps} fileItem={fileItem} onPreview={onPreview} />)
+
+      const item = screen.getByText('test-document.pdf').closest('[class*="flex h-12"]')!
+      fireEvent.click(item)
+
+      expect(onPreview).toHaveBeenCalledTimes(1)
+      expect(onPreview).toHaveBeenCalledWith(fileItem.file)
+    })
+
+    it('should not call onPreview when file has no id', () => {
+      const onPreview = vi.fn()
+      const fileItem = createMockFileItem()
+      render(<FileListItem {...defaultProps} fileItem={fileItem} onPreview={onPreview} />)
+
+      const item = screen.getByText('test-document.pdf').closest('[class*="flex h-12"]')!
+      fireEvent.click(item)
+
+      expect(onPreview).not.toHaveBeenCalled()
+    })
+
+    it('should call onRemove when delete button is clicked', () => {
+      const onRemove = vi.fn()
+      const fileItem = createMockFileItem()
+      const { container } = render(<FileListItem {...defaultProps} fileItem={fileItem} onRemove={onRemove} />)
+
+      const deleteButton = container.querySelector('.cursor-pointer')!
+      fireEvent.click(deleteButton)
+
+      expect(onRemove).toHaveBeenCalledTimes(1)
+      expect(onRemove).toHaveBeenCalledWith('file-123')
+    })
+
+    it('should stop propagation when delete button is clicked', () => {
+      const onPreview = vi.fn()
+      const onRemove = vi.fn()
+      const fileItem = createMockFileItem({
+        file: createMockFile({ id: 'uploaded-id' } as Partial<File>),
+      })
+      const { container } = render(<FileListItem {...defaultProps} fileItem={fileItem} onPreview={onPreview} onRemove={onRemove} />)
+
+      const deleteButton = container.querySelector('.cursor-pointer')!
+      fireEvent.click(deleteButton)
+
+      expect(onRemove).toHaveBeenCalledTimes(1)
+      expect(onPreview).not.toHaveBeenCalled()
+    })
+  })
+
+  describe('file type handling', () => {
+    it('should handle files with multiple dots in name', () => {
+      const fileItem = createMockFileItem({
+        file: createMockFile({ name: 'my.document.file.docx' }),
+      })
+      render(<FileListItem {...defaultProps} fileItem={fileItem} />)
+
+      expect(screen.getByText('my.document.file.docx')).toBeInTheDocument()
+      expect(screen.getByText('docx')).toBeInTheDocument()
+    })
+
+    it('should handle files without extension', () => {
+      const fileItem = createMockFileItem({
+        file: createMockFile({ name: 'README' }),
+      })
+      render(<FileListItem {...defaultProps} fileItem={fileItem} />)
+
+      // File name appears once, and extension area shows empty string
+      expect(screen.getByText('README')).toBeInTheDocument()
+    })
+
+    it('should handle various file extensions', () => {
+      const extensions = ['txt', 'md', 'json', 'csv', 'xlsx']
+
+      extensions.forEach((ext) => {
+        const fileItem = createMockFileItem({
+          file: createMockFile({ name: `file.${ext}` }),
+        })
+        const { unmount } = render(<FileListItem {...defaultProps} fileItem={fileItem} />)
+        expect(screen.getByText(ext)).toBeInTheDocument()
+        unmount()
+      })
+    })
+  })
+
+  describe('file size display', () => {
+    it('should display size in KB for small files', () => {
+      const fileItem = createMockFileItem({
+        file: createMockFile({ size: 5 * 1024 }),
+      })
+      render(<FileListItem {...defaultProps} fileItem={fileItem} />)
+      expect(screen.getByText('5.00 KB')).toBeInTheDocument()
+    })
+
+    it('should display size in MB for larger files', () => {
+      const fileItem = createMockFileItem({
+        file: createMockFile({ size: 5 * 1024 * 1024 }),
+      })
+      render(<FileListItem {...defaultProps} fileItem={fileItem} />)
+      expect(screen.getByText('5.00 MB')).toBeInTheDocument()
+    })
+  })
+
+  describe('upload progress values', () => {
+    it('should show chart at progress 1', () => {
+      const fileItem = createMockFileItem({ progress: 1 })
+      render(<FileListItem {...defaultProps} fileItem={fileItem} />)
+      expect(screen.getByTestId('pie-chart')).toBeInTheDocument()
+    })
+
+    it('should show chart at progress 99', () => {
+      const fileItem = createMockFileItem({ progress: 99 })
+      render(<FileListItem {...defaultProps} fileItem={fileItem} />)
+      expect(screen.getByTestId('pie-chart')).toHaveAttribute('data-percentage', '99')
+    })
+
+    it('should not show chart at progress 100', () => {
+      const fileItem = createMockFileItem({ progress: 100 })
+      render(<FileListItem {...defaultProps} fileItem={fileItem} />)
+      expect(screen.queryByTestId('pie-chart')).not.toBeInTheDocument()
+    })
+  })
+
+  describe('styling', () => {
+    it('should have proper shadow styling', () => {
+      const { container } = render(<FileListItem {...defaultProps} />)
+      const item = container.firstChild as HTMLElement
+      expect(item).toHaveClass('shadow-xs')
+    })
+
+    it('should have proper border styling', () => {
+      const { container } = render(<FileListItem {...defaultProps} />)
+      const item = container.firstChild as HTMLElement
+      expect(item).toHaveClass('border', 'border-components-panel-border')
+    })
+
+    it('should truncate long file names', () => {
+      const longFileName = 'this-is-a-very-long-file-name-that-should-be-truncated.pdf'
+      const fileItem = createMockFileItem({
+        file: createMockFile({ name: longFileName }),
+      })
+      render(<FileListItem {...defaultProps} fileItem={fileItem} />)
+
+      const nameElement = screen.getByText(longFileName)
+      expect(nameElement).toHaveClass('truncate')
+    })
+  })
+})

web/app/components/datasets/create/file-uploader/components/file-list-item.tsx

@@ -0,0 +1,89 @@
+'use client'
+import type { CustomFile as File, FileItem } from '@/models/datasets'
+import { RiDeleteBinLine, RiErrorWarningFill } from '@remixicon/react'
+import dynamic from 'next/dynamic'
+import { useMemo } from 'react'
+import DocumentFileIcon from '@/app/components/datasets/common/document-file-icon'
+import useTheme from '@/hooks/use-theme'
+import { Theme } from '@/types/app'
+import { formatFileSize, getFileExtension } from '@/utils/format'
+import { PROGRESS_COMPLETE, PROGRESS_ERROR } from '../constants'
+
+const SimplePieChart = dynamic(() => import('@/app/components/base/simple-pie-chart'), { ssr: false })
+
+export type FileListItemProps = {
+  fileItem: FileItem
+  onPreview: (file: File) => void
+  onRemove: (fileID: string) => void
+}
+
+const FileListItem = ({
+  fileItem,
+  onPreview,
+  onRemove,
+}: FileListItemProps) => {
+  const { theme } = useTheme()
+  const chartColor = useMemo(() => theme === Theme.dark ? '#5289ff' : '#296dff', [theme])
+
+  const isUploading = fileItem.progress >= 0 && fileItem.progress < PROGRESS_COMPLETE
+  const isError = fileItem.progress === PROGRESS_ERROR
+
+  const handleClick = () => {
+    if (fileItem.file?.id)
+      onPreview(fileItem.file)
+  }
+
+  const handleRemove = (e: React.MouseEvent) => {
+    e.stopPropagation()
+    onRemove(fileItem.fileID)
+  }
+
+  return (
+    <div
+      onClick={handleClick}
+      className="flex h-12 max-w-[640px] items-center rounded-lg border border-components-panel-border bg-components-panel-on-panel-item-bg text-xs leading-3 text-text-tertiary shadow-xs"
+    >
+      <div className="flex w-12 shrink-0 items-center justify-center">
+        <DocumentFileIcon
+          size="xl"
+          className="shrink-0"
+          name={fileItem.file.name}
+          extension={getFileExtension(fileItem.file.name)}
+        />
+      </div>
+      <div className="flex shrink grow flex-col gap-0.5">
+        <div className="flex w-full">
+          <div className="w-0 grow truncate text-sm leading-4 text-text-secondary">
+            {fileItem.file.name}
+          </div>
+        </div>
+        <div className="w-full truncate leading-3 text-text-tertiary">
+          <span className="uppercase">{getFileExtension(fileItem.file.name)}</span>
+          <span className="px-1 text-text-quaternary">·</span>
+          <span>{formatFileSize(fileItem.file.size)}</span>
+        </div>
+      </div>
+      <div className="flex w-16 shrink-0 items-center justify-end gap-1 pr-3">
+        {isUploading && (
+          <SimplePieChart
+            percentage={fileItem.progress}
+            stroke={chartColor}
+            fill={chartColor}
+            animationDuration={0}
+          />
+        )}
+        {isError && (
+          <RiErrorWarningFill className="size-4 text-text-destructive" />
+        )}
+        <span
+          className="flex h-6 w-6 cursor-pointer items-center justify-center"
+          onClick={handleRemove}
+        >
+          <RiDeleteBinLine className="size-4 text-text-tertiary" />
+        </span>
+      </div>
+    </div>
+  )
+}
+
+export default FileListItem

web/app/components/datasets/create/file-uploader/components/upload-dropzone.spec.tsx

@@ -0,0 +1,210 @@
+import type { RefObject } from 'react'
+import type { UploadDropzoneProps } from './upload-dropzone'
+import { fireEvent, render, screen } from '@testing-library/react'
+import { beforeEach, describe, expect, it, vi } from 'vitest'
+import UploadDropzone from './upload-dropzone'
+
+// Helper to create mock ref objects for testing
+const createMockRef = <T,>(value: T | null = null): RefObject<T | null> => ({ current: value })
+
+// Mock react-i18next
+vi.mock('react-i18next', () => ({
+  useTranslation: () => ({
+    t: (key: string, options?: Record<string, unknown>) => {
+      const translations: Record<string, string> = {
+        'stepOne.uploader.button': 'Drag and drop files, or',
+        'stepOne.uploader.buttonSingleFile': 'Drag and drop file, or',
+        'stepOne.uploader.browse': 'Browse',
+        'stepOne.uploader.tip': 'Supports {{supportTypes}}, Max {{size}}MB each, up to {{batchCount}} files at a time, {{totalCount}} files total',
+      }
+      let result = translations[key] || key
+      if (options && typeof options === 'object') {
+        Object.entries(options).forEach(([k, v]) => {
+          result = result.replace(`{{${k}}}`, String(v))
+        })
+      }
+      return result
+    },
+  }),
+}))
+
+describe('UploadDropzone', () => {
+  const defaultProps: UploadDropzoneProps = {
+    dropRef: createMockRef<HTMLDivElement>() as RefObject<HTMLDivElement | null>,
+    dragRef: createMockRef<HTMLDivElement>() as RefObject<HTMLDivElement | null>,
+    fileUploaderRef: createMockRef<HTMLInputElement>() as RefObject<HTMLInputElement | null>,
+    dragging: false,
+    supportBatchUpload: true,
+    supportTypesShowNames: 'PDF, DOCX, TXT',
+    fileUploadConfig: {
+      file_size_limit: 15,
+      batch_count_limit: 5,
+      file_upload_limit: 10,
+    },
+    acceptTypes: ['.pdf', '.docx', '.txt'],
+    onSelectFile: vi.fn(),
+    onFileChange: vi.fn(),
+  }
+
+  beforeEach(() => {
+    vi.clearAllMocks()
+  })
+
+  describe('rendering', () => {
+    it('should render the dropzone container', () => {
+      const { container } = render(<UploadDropzone {...defaultProps} />)
+      const dropzone = container.querySelector('[class*="border-dashed"]')
+      expect(dropzone).toBeInTheDocument()
+    })
+
+    it('should render hidden file input', () => {
+      render(<UploadDropzone {...defaultProps} />)
+      const input = document.getElementById('fileUploader') as HTMLInputElement
+      expect(input).toBeInTheDocument()
+      expect(input).toHaveClass('hidden')
+      expect(input).toHaveAttribute('type', 'file')
+    })
+
+    it('should render upload icon', () => {
+      render(<UploadDropzone {...defaultProps} />)
+      const icon = document.querySelector('svg')
+      expect(icon).toBeInTheDocument()
+    })
+
+    it('should render browse label when extensions are allowed', () => {
+      render(<UploadDropzone {...defaultProps} />)
+      expect(screen.getByText('Browse')).toBeInTheDocument()
+    })
+
+    it('should not render browse label when no extensions allowed', () => {
+      render(<UploadDropzone {...defaultProps} acceptTypes={[]} />)
+      expect(screen.queryByText('Browse')).not.toBeInTheDocument()
+    })
+
+    it('should render file size and count limits', () => {
+      render(<UploadDropzone {...defaultProps} />)
+      const tipText = screen.getByText(/Supports.*Max.*15MB/i)
+      expect(tipText).toBeInTheDocument()
+    })
+  })
+
+  describe('file input configuration', () => {
+    it('should allow multiple files when supportBatchUpload is true', () => {
+      render(<UploadDropzone {...defaultProps} supportBatchUpload={true} />)
+      const input = document.getElementById('fileUploader') as HTMLInputElement
+      expect(input).toHaveAttribute('multiple')
+    })
+
+    it('should not allow multiple files when supportBatchUpload is false', () => {
+      render(<UploadDropzone {...defaultProps} supportBatchUpload={false} />)
+      const input = document.getElementById('fileUploader') as HTMLInputElement
+      expect(input).not.toHaveAttribute('multiple')
+    })
+
+    it('should set accept attribute with correct types', () => {
+      render(<UploadDropzone {...defaultProps} acceptTypes={['.pdf', '.docx']} />)
+      const input = document.getElementById('fileUploader') as HTMLInputElement
+      expect(input).toHaveAttribute('accept', '.pdf,.docx')
+    })
+  })
+
+  describe('text content', () => {
+    it('should show batch upload text when supportBatchUpload is true', () => {
+      render(<UploadDropzone {...defaultProps} supportBatchUpload={true} />)
+      expect(screen.getByText(/Drag and drop files/i)).toBeInTheDocument()
+    })
+
+    it('should show single file text when supportBatchUpload is false', () => {
+      render(<UploadDropzone {...defaultProps} supportBatchUpload={false} />)
+      expect(screen.getByText(/Drag and drop file/i)).toBeInTheDocument()
+    })
+  })
+
+  describe('dragging state', () => {
+    it('should apply dragging styles when dragging is true', () => {
+      const { container } = render(<UploadDropzone {...defaultProps} dragging={true} />)
+      const dropzone = container.querySelector('[class*="border-components-dropzone-border-accent"]')
+      expect(dropzone).toBeInTheDocument()
+    })
+
+    it('should render drag overlay when dragging', () => {
+      const dragRef = createMockRef<HTMLDivElement>()
+      render(<UploadDropzone {...defaultProps} dragging={true} dragRef={dragRef as RefObject<HTMLDivElement | null>} />)
+      const overlay = document.querySelector('.absolute.left-0.top-0')
+      expect(overlay).toBeInTheDocument()
+    })
+
+    it('should not render drag overlay when not dragging', () => {
+      render(<UploadDropzone {...defaultProps} dragging={false} />)
+      const overlay = document.querySelector('.absolute.left-0.top-0')
+      expect(overlay).not.toBeInTheDocument()
+    })
+  })
+
+  describe('event handlers', () => {
+    it('should call onSelectFile when browse label is clicked', () => {
+      const onSelectFile = vi.fn()
+      render(<UploadDropzone {...defaultProps} onSelectFile={onSelectFile} />)
+
+      const browseLabel = screen.getByText('Browse')
+      fireEvent.click(browseLabel)
+
+      expect(onSelectFile).toHaveBeenCalledTimes(1)
+    })
+
+    it('should call onFileChange when files are selected', () => {
+      const onFileChange = vi.fn()
+      render(<UploadDropzone {...defaultProps} onFileChange={onFileChange} />)
+
+      const input = document.getElementById('fileUploader') as HTMLInputElement
+      const file = new File(['content'], 'test.pdf', { type: 'application/pdf' })
+
+      fireEvent.change(input, { target: { files: [file] } })
+
+      expect(onFileChange).toHaveBeenCalledTimes(1)
+    })
+  })
+
+  describe('refs', () => {
+    it('should attach dropRef to drop container', () => {
+      const dropRef = createMockRef<HTMLDivElement>()
+      render(<UploadDropzone {...defaultProps} dropRef={dropRef as RefObject<HTMLDivElement | null>} />)
+      expect(dropRef.current).toBeInstanceOf(HTMLDivElement)
+    })
+
+    it('should attach fileUploaderRef to input element', () => {
+      const fileUploaderRef = createMockRef<HTMLInputElement>()
+      render(<UploadDropzone {...defaultProps} fileUploaderRef={fileUploaderRef as RefObject<HTMLInputElement | null>} />)
+      expect(fileUploaderRef.current).toBeInstanceOf(HTMLInputElement)
+    })
+
+    it('should attach dragRef to overlay when dragging', () => {
+      const dragRef = createMockRef<HTMLDivElement>()
+      render(<UploadDropzone {...defaultProps} dragging={true} dragRef={dragRef as RefObject<HTMLDivElement | null>} />)
+      expect(dragRef.current).toBeInstanceOf(HTMLDivElement)
+    })
+  })
+
+  describe('styling', () => {
+    it('should have base dropzone styling', () => {
+      const { container } = render(<UploadDropzone {...defaultProps} />)
+      const dropzone = container.querySelector('[class*="border-dashed"]')
+      expect(dropzone).toBeInTheDocument()
+      expect(dropzone).toHaveClass('rounded-xl')
+    })
+
+    it('should have cursor-pointer on browse label', () => {
+      render(<UploadDropzone {...defaultProps} />)
+      const browseLabel = screen.getByText('Browse')
+      expect(browseLabel).toHaveClass('cursor-pointer')
+    })
+  })
+
+  describe('accessibility', () => {
+    it('should have an accessible file input', () => {
+      render(<UploadDropzone {...defaultProps} />)
+      const input = document.getElementById('fileUploader') as HTMLInputElement
+      expect(input).toHaveAttribute('id', 'fileUploader')
+    })
+  })
+})

web/app/components/datasets/create/file-uploader/components/upload-dropzone.tsx

@@ -0,0 +1,84 @@
+'use client'
+import type { RefObject } from 'react'
+import type { FileUploadConfig } from '../hooks/use-file-upload'
+import { RiUploadCloud2Line } from '@remixicon/react'
+import { useTranslation } from 'react-i18next'
+import { cn } from '@/utils/classnames'
+
+export type UploadDropzoneProps = {
+  dropRef: RefObject<HTMLDivElement | null>
+  dragRef: RefObject<HTMLDivElement | null>
+  fileUploaderRef: RefObject<HTMLInputElement | null>
+  dragging: boolean
+  supportBatchUpload: boolean
+  supportTypesShowNames: string
+  fileUploadConfig: FileUploadConfig
+  acceptTypes: string[]
+  onSelectFile: () => void
+  onFileChange: (e: React.ChangeEvent<HTMLInputElement>) => void
+}
+
+const UploadDropzone = ({
+  dropRef,
+  dragRef,
+  fileUploaderRef,
+  dragging,
+  supportBatchUpload,
+  supportTypesShowNames,
+  fileUploadConfig,
+  acceptTypes,
+  onSelectFile,
+  onFileChange,
+}: UploadDropzoneProps) => {
+  const { t } = useTranslation()
+
+  return (
+    <>
+      <input
+        ref={fileUploaderRef}
+        id="fileUploader"
+        className="hidden"
+        type="file"
+        multiple={supportBatchUpload}
+        accept={acceptTypes.join(',')}
+        onChange={onFileChange}
+      />
+      <div
+        ref={dropRef}
+        className={cn(
+          'relative mb-2 box-border flex min-h-20 max-w-[640px] flex-col items-center justify-center gap-1 rounded-xl border border-dashed border-components-dropzone-border bg-components-dropzone-bg px-4 py-3 text-xs leading-4 text-text-tertiary',
+          dragging && 'border-components-dropzone-border-accent bg-components-dropzone-bg-accent',
+        )}
+      >
+        <div className="flex min-h-5 items-center justify-center text-sm leading-4 text-text-secondary">
+          <RiUploadCloud2Line className="mr-2 size-5" />
+          <span>
+            {supportBatchUpload
+              ? t('stepOne.uploader.button', { ns: 'datasetCreation' })
+              : t('stepOne.uploader.buttonSingleFile', { ns: 'datasetCreation' })}
+            {acceptTypes.length > 0 && (
+              <label
+                className="ml-1 cursor-pointer text-text-accent"
+                onClick={onSelectFile}
+              >
+                {t('stepOne.uploader.browse', { ns: 'datasetCreation' })}
+              </label>
+            )}
+          </span>
+        </div>
+        <div>
+          {t('stepOne.uploader.tip', {
+            ns: 'datasetCreation',
+            size: fileUploadConfig.file_size_limit,
+            supportTypes: supportTypesShowNames,
+            batchCount: fileUploadConfig.batch_count_limit,
+            totalCount: fileUploadConfig.file_upload_limit,
+          })}
+        </div>
+        {dragging && <div ref={dragRef} className="absolute left-0 top-0 h-full w-full" />}
+      </div>
+    </>
+  )
+}
+
+export default UploadDropzone

web/app/components/datasets/create/file-uploader/constants.ts

@@ -0,0 +1,3 @@
+export const PROGRESS_NOT_STARTED = -1
+export const PROGRESS_ERROR = -2
+export const PROGRESS_COMPLETE = 100

web/app/components/datasets/create/file-uploader/hooks/use-file-upload.spec.tsx

@@ -0,0 +1,921 @@
+import type { ReactNode } from 'react'
+import type { CustomFile, FileItem } from '@/models/datasets'
+import { act, render, renderHook, waitFor } from '@testing-library/react'
+import { beforeEach, describe, expect, it, vi } from 'vitest'
+import { ToastContext } from '@/app/components/base/toast'
+
+import { PROGRESS_COMPLETE, PROGRESS_ERROR, PROGRESS_NOT_STARTED } from '../constants'
+// Import after mocks
+import { useFileUpload } from './use-file-upload'
+
+// Mock notify function
+const mockNotify = vi.fn()
+const mockClose = vi.fn()
+
+// Mock ToastContext
+vi.mock('use-context-selector', async () => {
+  const actual = await vi.importActual<typeof import('use-context-selector')>('use-context-selector')
+  return {
+    ...actual,
+    useContext: vi.fn(() => ({ notify: mockNotify, close: mockClose })),
+  }
+})
+
+// Mock upload service
+const mockUpload = vi.fn()
+vi.mock('@/service/base', () => ({
+  upload: (...args: unknown[]) => mockUpload(...args),
+}))
+
+// Mock file upload config
+const mockFileUploadConfig = {
+  file_size_limit: 15,
+  batch_count_limit: 5,
+  file_upload_limit: 10,
+}
+
+const mockSupportTypes = {
+  allowed_extensions: ['pdf', 'docx', 'txt', 'md'],
+}
+
+vi.mock('@/service/use-common', () => ({
+  useFileUploadConfig: () => ({ data: mockFileUploadConfig }),
+  useFileSupportTypes: () => ({ data: mockSupportTypes }),
+}))
+
+// Mock i18n
+vi.mock('react-i18next', () => ({
+  useTranslation: () => ({
+    t: (key: string) => key,
+  }),
+}))
+
+// Mock locale
+vi.mock('@/context/i18n', () => ({
+  useLocale: () => 'en-US',
+}))
+
+vi.mock('@/i18n-config/language', () => ({
+  LanguagesSupported: ['en-US', 'zh-Hans'],
+}))
+
+// Mock config
+vi.mock('@/config', () => ({
+  IS_CE_EDITION: false,
+}))
+
+// Mock file upload error message
+vi.mock('@/app/components/base/file-uploader/utils', () => ({
+  getFileUploadErrorMessage: (_e: unknown, defaultMsg: string) => defaultMsg,
+}))
+
+const createWrapper = () => {
+  return ({ children }: { children: ReactNode }) => (
+    <ToastContext.Provider value={{ notify: mockNotify, close: mockClose }}>
+      {children}
+    </ToastContext.Provider>
+  )
+}
+
+describe('useFileUpload', () => {
+  const defaultOptions = {
+    fileList: [] as FileItem[],
+    prepareFileList: vi.fn(),
+    onFileUpdate: vi.fn(),
+    onFileListUpdate: vi.fn(),
+    onPreview: vi.fn(),
+    supportBatchUpload: true,
+  }
+
+  beforeEach(() => {
+    vi.clearAllMocks()
+    mockUpload.mockReset()
+    // Default mock to return a resolved promise to avoid unhandled rejections
+    mockUpload.mockResolvedValue({ id: 'default-id' })
+    mockNotify.mockReset()
+  })
+
+  describe('initialization', () => {
+    it('should initialize with default values', () => {
+      const { result } = renderHook(
+        () => useFileUpload(defaultOptions),
+        { wrapper: createWrapper() },
+      )
+
+      expect(result.current.dragging).toBe(false)
+      expect(result.current.hideUpload).toBe(false)
+      expect(result.current.dropRef.current).toBeNull()
+      expect(result.current.dragRef.current).toBeNull()
+      expect(result.current.fileUploaderRef.current).toBeNull()
+    })
+
+    it('should set hideUpload true when not batch upload and has files', () => {
+      const { result } = renderHook(
+        () => useFileUpload({
+          ...defaultOptions,
+          supportBatchUpload: false,
+          fileList: [{ fileID: 'file-1', file: {} as CustomFile, progress: 100 }],
+        }),
+        { wrapper: createWrapper() },
+      )
+
+      expect(result.current.hideUpload).toBe(true)
+    })
+
+    it('should compute acceptTypes correctly', () => {
+      const { result } = renderHook(
+        () => useFileUpload(defaultOptions),
+        { wrapper: createWrapper() },
+      )
+
+      expect(result.current.acceptTypes).toEqual(['.pdf', '.docx', '.txt', '.md'])
+    })
+
+    it('should compute supportTypesShowNames correctly', () => {
+      const { result } = renderHook(
+        () => useFileUpload(defaultOptions),
+        { wrapper: createWrapper() },
+      )
+
+      expect(result.current.supportTypesShowNames).toContain('PDF')
+      expect(result.current.supportTypesShowNames).toContain('DOCX')
+      expect(result.current.supportTypesShowNames).toContain('TXT')
+      // 'md' is mapped to 'markdown' in the extensionMap
+      expect(result.current.supportTypesShowNames).toContain('MARKDOWN')
+    })
+
+    it('should set batch limit to 1 when not batch upload', () => {
+      const { result } = renderHook(
+        () => useFileUpload({
+          ...defaultOptions,
+          supportBatchUpload: false,
+        }),
+        { wrapper: createWrapper() },
+      )
+
+      expect(result.current.fileUploadConfig.batch_count_limit).toBe(1)
+      expect(result.current.fileUploadConfig.file_upload_limit).toBe(1)
+    })
+  })
+
+  describe('selectHandle', () => {
+    it('should trigger click on file input', () => {
+      const { result } = renderHook(
+        () => useFileUpload(defaultOptions),
+        { wrapper: createWrapper() },
+      )
+
+      const mockClick = vi.fn()
+      const mockInput = { click: mockClick } as unknown as HTMLInputElement
+      Object.defineProperty(result.current.fileUploaderRef, 'current', {
+        value: mockInput,
+        writable: true,
+      })
+
+      act(() => {
+        result.current.selectHandle()
+      })
+
+      expect(mockClick).toHaveBeenCalled()
+    })
+
+    it('should do nothing when file input ref is null', () => {
+      const { result } = renderHook(
+        () => useFileUpload(defaultOptions),
+        { wrapper: createWrapper() },
+      )
+
+      expect(() => {
+        act(() => {
+          result.current.selectHandle()
+        })
+      }).not.toThrow()
+    })
+  })
+
+  describe('handlePreview', () => {
+    it('should call onPreview when file has id', () => {
+      const onPreview = vi.fn()
+      const { result } = renderHook(
+        () => useFileUpload({ ...defaultOptions, onPreview }),
+        { wrapper: createWrapper() },
+      )
+
+      const mockFile = { id: 'file-123', name: 'test.pdf', size: 1024 } as CustomFile
+
+      act(() => {
+        result.current.handlePreview(mockFile)
+      })
+
+      expect(onPreview).toHaveBeenCalledWith(mockFile)
+    })
+
+    it('should not call onPreview when file has no id', () => {
+      const onPreview = vi.fn()
+      const { result } = renderHook(
+        () => useFileUpload({ ...defaultOptions, onPreview }),
+        { wrapper: createWrapper() },
+      )
+
+      const mockFile = { name: 'test.pdf', size: 1024 } as CustomFile
+
+      act(() => {
+        result.current.handlePreview(mockFile)
+      })
+
+      expect(onPreview).not.toHaveBeenCalled()
+    })
+  })
+
+  describe('removeFile', () => {
+    it('should call onFileListUpdate with filtered list', () => {
+      const onFileListUpdate = vi.fn()
+      const { result } = renderHook(
+        () => useFileUpload({ ...defaultOptions, onFileListUpdate }),
+        { wrapper: createWrapper() },
+      )
+
+      act(() => {
+        result.current.removeFile('file-to-remove')
+      })
+
+      expect(onFileListUpdate).toHaveBeenCalled()
+    })
+
+    it('should clear file input value', () => {
+      const { result } = renderHook(
+        () => useFileUpload(defaultOptions),
+        { wrapper: createWrapper() },
+      )
+
+      const mockInput = { value: 'some-file' } as HTMLInputElement
+      Object.defineProperty(result.current.fileUploaderRef, 'current', {
+        value: mockInput,
+        writable: true,
+      })
+
+      act(() => {
+        result.current.removeFile('file-123')
+      })
+
+      expect(mockInput.value).toBe('')
+    })
+  })
+
+  describe('fileChangeHandle', () => {
+    it('should handle valid files', async () => {
+      mockUpload.mockResolvedValue({ id: 'uploaded-id' })
+
+      const prepareFileList = vi.fn()
+      const { result } = renderHook(
+        () => useFileUpload({ ...defaultOptions, prepareFileList }),
+        { wrapper: createWrapper() },
+      )
+
+      const mockFile = new File(['content'], 'test.pdf', { type: 'application/pdf' })
+      const event = {
+        target: { files: [mockFile] },
+      } as unknown as React.ChangeEvent<HTMLInputElement>
+
+      act(() => {
+        result.current.fileChangeHandle(event)
+      })
+
+      await waitFor(() => {
+        expect(prepareFileList).toHaveBeenCalled()
+      })
+    })
+
+    it('should limit files to batch count', () => {
+      const prepareFileList = vi.fn()
+      const { result } = renderHook(
+        () => useFileUpload({ ...defaultOptions, prepareFileList }),
+        { wrapper: createWrapper() },
+      )
+
+      const files = Array.from({ length: 10 }, (_, i) =>
+        new File(['content'], `file${i}.pdf`, { type: 'application/pdf' }))
+
+      const event = {
+        target: { files },
+      } as unknown as React.ChangeEvent<HTMLInputElement>
+
+      act(() => {
+        result.current.fileChangeHandle(event)
+      })
+
+      // Should be called with at most batch_count_limit files
+      if (prepareFileList.mock.calls.length > 0) {
+        const calledFiles = prepareFileList.mock.calls[0][0]
+        expect(calledFiles.length).toBeLessThanOrEqual(mockFileUploadConfig.batch_count_limit)
+      }
+    })
+
+    it('should reject invalid file types', () => {
+      const { result } = renderHook(
+        () => useFileUpload(defaultOptions),
+        { wrapper: createWrapper() },
+      )
+
+      const mockFile = new File(['content'], 'test.exe', { type: 'application/x-msdownload' })
+      const event = {
+        target: { files: [mockFile] },
+      } as unknown as React.ChangeEvent<HTMLInputElement>
+
+      act(() => {
+        result.current.fileChangeHandle(event)
+      })
+
+      expect(mockNotify).toHaveBeenCalledWith(
+        expect.objectContaining({ type: 'error' }),
+      )
+    })
+
+    it('should reject files exceeding size limit', () => {
+      const { result } = renderHook(
+        () => useFileUpload(defaultOptions),
+        { wrapper: createWrapper() },
+      )
+
+      // Create a file larger than the limit (15MB)
+      const largeFile = new File([new ArrayBuffer(20 * 1024 * 1024)], 'large.pdf', { type: 'application/pdf' })
+
+      const event = {
+        target: { files: [largeFile] },
+      } as unknown as React.ChangeEvent<HTMLInputElement>
+
+      act(() => {
+        result.current.fileChangeHandle(event)
+      })
+
+      expect(mockNotify).toHaveBeenCalledWith(
+        expect.objectContaining({ type: 'error' }),
+      )
+    })
+
+    it('should handle null files', () => {
+      const prepareFileList = vi.fn()
+      const { result } = renderHook(
+        () => useFileUpload({ ...defaultOptions, prepareFileList }),
+        { wrapper: createWrapper() },
+      )
+
+      const event = {
+        target: { files: null },
+      } as unknown as React.ChangeEvent<HTMLInputElement>
+
+      act(() => {
+        result.current.fileChangeHandle(event)
+      })
+
+      expect(prepareFileList).not.toHaveBeenCalled()
+    })
+  })
+
+  describe('drag and drop handlers', () => {
+    const TestDropzone = ({ options }: { options: typeof defaultOptions }) => {
+      const {
+        dropRef,
+        dragRef,
+        dragging,
+      } = useFileUpload(options)
+
+      return (
+        <div>
+          <div ref={dropRef} data-testid="dropzone">
+            {dragging && <div ref={dragRef} data-testid="drag-overlay" />}
+          </div>
+          <span data-testid="dragging">{String(dragging)}</span>
+        </div>
+      )
+    }
+
+    it('should set dragging true on dragenter', async () => {
+      const { getByTestId } = await act(async () =>
+        render(
+          <ToastContext.Provider value={{ notify: mockNotify, close: mockClose }}>
+            <TestDropzone options={defaultOptions} />
+          </ToastContext.Provider>,
+        ),
+      )
+
+      const dropzone = getByTestId('dropzone')
+
+      await act(async () => {
+        const dragEnterEvent = new Event('dragenter', { bubbles: true, cancelable: true })
+        dropzone.dispatchEvent(dragEnterEvent)
+      })
+
+      expect(getByTestId('dragging').textContent).toBe('true')
+    })
+
+    it('should handle dragover event', async () => {
+      const { getByTestId } = await act(async () =>
+        render(
+          <ToastContext.Provider value={{ notify: mockNotify, close: mockClose }}>
+            <TestDropzone options={defaultOptions} />
+          </ToastContext.Provider>,
+        ),
+      )
+
+      const dropzone = getByTestId('dropzone')
+
+      await act(async () => {
+        const dragOverEvent = new Event('dragover', { bubbles: true, cancelable: true })
+        dropzone.dispatchEvent(dragOverEvent)
+      })
+
+      expect(dropzone).toBeInTheDocument()
+    })
+
+    it('should set dragging false on dragleave from drag overlay', async () => {
+      const { getByTestId, queryByTestId } = await act(async () =>
+        render(
+          <ToastContext.Provider value={{ notify: mockNotify, close: mockClose }}>
+            <TestDropzone options={defaultOptions} />
+          </ToastContext.Provider>,
+        ),
+      )
+
+      const dropzone = getByTestId('dropzone')
+
+      await act(async () => {
+        const dragEnterEvent = new Event('dragenter', { bubbles: true, cancelable: true })
+        dropzone.dispatchEvent(dragEnterEvent)
+      })
+
+      expect(getByTestId('dragging').textContent).toBe('true')
+
+      const dragOverlay = queryByTestId('drag-overlay')
+      if (dragOverlay) {
+        await act(async () => {
+          const dragLeaveEvent = new Event('dragleave', { bubbles: true, cancelable: true })
+          Object.defineProperty(dragLeaveEvent, 'target', { value: dragOverlay })
+          dropzone.dispatchEvent(dragLeaveEvent)
+        })
+      }
+    })
+
+    it('should handle drop with files', async () => {
+      mockUpload.mockResolvedValue({ id: 'uploaded-id' })
+      const prepareFileList = vi.fn()
+
+      const { getByTestId } = await act(async () =>
+        render(
+          <ToastContext.Provider value={{ notify: mockNotify, close: mockClose }}>
+            <TestDropzone options={{ ...defaultOptions, prepareFileList }} />
+          </ToastContext.Provider>,
+        ),
+      )
+
+      const dropzone = getByTestId('dropzone')
+      const mockFile = new File(['content'], 'test.pdf', { type: 'application/pdf' })
+
+      await act(async () => {
+        const dropEvent = new Event('drop', { bubbles: true, cancelable: true }) as Event & { dataTransfer: DataTransfer | null }
+        Object.defineProperty(dropEvent, 'dataTransfer', {
+          value: {
+            items: [{
+              getAsFile: () => mockFile,
+              webkitGetAsEntry: () => null,
+            }],
+          },
+        })
+        dropzone.dispatchEvent(dropEvent)
+      })
+
+      await waitFor(() => {
+        expect(prepareFileList).toHaveBeenCalled()
+      })
+    })
+
+    it('should handle drop without dataTransfer', async () => {
+      const prepareFileList = vi.fn()
+
+      const { getByTestId } = await act(async () =>
+        render(
+          <ToastContext.Provider value={{ notify: mockNotify, close: mockClose }}>
+            <TestDropzone options={{ ...defaultOptions, prepareFileList }} />
+          </ToastContext.Provider>,
+        ),
+      )
+
+      const dropzone = getByTestId('dropzone')
+
+      await act(async () => {
+        const dropEvent = new Event('drop', { bubbles: true, cancelable: true }) as Event & { dataTransfer: DataTransfer | null }
+        Object.defineProperty(dropEvent, 'dataTransfer', { value: null })
+        dropzone.dispatchEvent(dropEvent)
+      })
+
+      expect(prepareFileList).not.toHaveBeenCalled()
+    })
+
+    it('should limit to single file on drop when supportBatchUpload is false', async () => {
+      mockUpload.mockResolvedValue({ id: 'uploaded-id' })
+      const prepareFileList = vi.fn()
+
+      const { getByTestId } = await act(async () =>
+        render(
+          <ToastContext.Provider value={{ notify: mockNotify, close: mockClose }}>
+            <TestDropzone options={{ ...defaultOptions, supportBatchUpload: false, prepareFileList }} />
+          </ToastContext.Provider>,
+        ),
+      )
+
+      const dropzone = getByTestId('dropzone')
+      const files = [
+        new File(['content1'], 'test1.pdf', { type: 'application/pdf' }),
+        new File(['content2'], 'test2.pdf', { type: 'application/pdf' }),
+      ]
+
+      await act(async () => {
+        const dropEvent = new Event('drop', { bubbles: true, cancelable: true }) as Event & { dataTransfer: DataTransfer | null }
+        Object.defineProperty(dropEvent, 'dataTransfer', {
+          value: {
+            items: files.map(f => ({
+              getAsFile: () => f,
+              webkitGetAsEntry: () => null,
+            })),
+          },
+        })
+        dropzone.dispatchEvent(dropEvent)
+      })
+
+      await waitFor(() => {
+        if (prepareFileList.mock.calls.length > 0) {
+          const calledFiles = prepareFileList.mock.calls[0][0]
+          expect(calledFiles.length).toBe(1)
+        }
+      })
+    })
+
+    it('should handle drop with FileSystemFileEntry', async () => {
+      mockUpload.mockResolvedValue({ id: 'uploaded-id' })
+      const prepareFileList = vi.fn()
+      const mockFile = new File(['content'], 'test.pdf', { type: 'application/pdf' })
+
+      const { getByTestId } = await act(async () =>
+        render(
+          <ToastContext.Provider value={{ notify: mockNotify, close: mockClose }}>
+            <TestDropzone options={{ ...defaultOptions, prepareFileList }} />
+          </ToastContext.Provider>,
+        ),
+      )
+
+      const dropzone = getByTestId('dropzone')
+
+      await act(async () => {
+        const dropEvent = new Event('drop', { bubbles: true, cancelable: true }) as Event & { dataTransfer: DataTransfer | null }
+        Object.defineProperty(dropEvent, 'dataTransfer', {
+          value: {
+            items: [{
+              getAsFile: () => mockFile,
+              webkitGetAsEntry: () => ({
+                isFile: true,
+                isDirectory: false,
+                file: (callback: (file: File) => void) => callback(mockFile),
+              }),
+            }],
+          },
+        })
+        dropzone.dispatchEvent(dropEvent)
+      })
+
+      await waitFor(() => {
+        expect(prepareFileList).toHaveBeenCalled()
+      })
+    })
+
+    it('should handle drop with FileSystemDirectoryEntry', async () => {
+      mockUpload.mockResolvedValue({ id: 'uploaded-id' })
+      const prepareFileList = vi.fn()
+      const mockFile = new File(['content'], 'nested.pdf', { type: 'application/pdf' })
+
+      const { getByTestId } = await act(async () =>
+        render(
+          <ToastContext.Provider value={{ notify: mockNotify, close: mockClose }}>
+            <TestDropzone options={{ ...defaultOptions, prepareFileList }} />
+          </ToastContext.Provider>,
+        ),
+      )
+
+      const dropzone = getByTestId('dropzone')
+
+      await act(async () => {
+        let callCount = 0
+        const dropEvent = new Event('drop', { bubbles: true, cancelable: true }) as Event & { dataTransfer: DataTransfer | null }
+        Object.defineProperty(dropEvent, 'dataTransfer', {
+          value: {
+            items: [{
+              getAsFile: () => null,
+              webkitGetAsEntry: () => ({
+                isFile: false,
+                isDirectory: true,
+                name: 'folder',
+                createReader: () => ({
+                  readEntries: (callback: (entries: Array<{ isFile: boolean, isDirectory: boolean, name?: string, file?: (cb: (f: File) => void) => void }>) => void) => {
+                    // First call returns file entry, second call returns empty (signals end)
+                    if (callCount === 0) {
+                      callCount++
+                      callback([{
+                        isFile: true,
+                        isDirectory: false,
+                        name: 'nested.pdf',
+                        file: (cb: (f: File) => void) => cb(mockFile),
+                      }])
+                    }
+                    else {
+                      callback([])
+                    }
+                  },
+                }),
+              }),
+            }],
+          },
+        })
+        dropzone.dispatchEvent(dropEvent)
+      })
+
+      await waitFor(() => {
+        expect(prepareFileList).toHaveBeenCalled()
+      })
+    })
+
+    it('should handle drop with empty directory', async () => {
+      const prepareFileList = vi.fn()
+
+      const { getByTestId } = await act(async () =>
+        render(
+          <ToastContext.Provider value={{ notify: mockNotify, close: mockClose }}>
+            <TestDropzone options={{ ...defaultOptions, prepareFileList }} />
+          </ToastContext.Provider>,
+        ),
+      )
+
+      const dropzone = getByTestId('dropzone')
+
+      await act(async () => {
+        const dropEvent = new Event('drop', { bubbles: true, cancelable: true }) as Event & { dataTransfer: DataTransfer | null }
+        Object.defineProperty(dropEvent, 'dataTransfer', {
+          value: {
+            items: [{
+              getAsFile: () => null,
+              webkitGetAsEntry: () => ({
+                isFile: false,
+                isDirectory: true,
+                name: 'empty-folder',
+                createReader: () => ({
+                  readEntries: (callback: (entries: never[]) => void) => {
+                    callback([])
+                  },
+                }),
+              }),
+            }],
+          },
+        })
+        dropzone.dispatchEvent(dropEvent)
+      })
+
+      // Should not prepare file list if no valid files
+      await new Promise(resolve => setTimeout(resolve, 100))
+    })
+
+    it('should handle entry that is neither file nor directory', async () => {
+      const prepareFileList = vi.fn()
+
+      const { getByTestId } = await act(async () =>
+        render(
+          <ToastContext.Provider value={{ notify: mockNotify, close: mockClose }}>
+            <TestDropzone options={{ ...defaultOptions, prepareFileList }} />
+          </ToastContext.Provider>,
+        ),
+      )
+
+      const dropzone = getByTestId('dropzone')
+
+      await act(async () => {
+        const dropEvent = new Event('drop', { bubbles: true, cancelable: true }) as Event & { dataTransfer: DataTransfer | null }
+        Object.defineProperty(dropEvent, 'dataTransfer', {
+          value: {
+            items: [{
+              getAsFile: () => null,
+              webkitGetAsEntry: () => ({
+                isFile: false,
+                isDirectory: false,
+              }),
+            }],
+          },
+        })
+        dropzone.dispatchEvent(dropEvent)
+      })
+
+      // Should not throw and should handle gracefully
+      await new Promise(resolve => setTimeout(resolve, 100))
+    })
+  })
+
+  describe('file upload', () => {
+    it('should call upload with correct parameters', async () => {
+      mockUpload.mockResolvedValue({ id: 'uploaded-id', name: 'test.pdf' })
+      const onFileUpdate = vi.fn()
+
+      const { result } = renderHook(
+        () => useFileUpload({ ...defaultOptions, onFileUpdate }),
+        { wrapper: createWrapper() },
+      )
+
+      const mockFile = new File(['content'], 'test.pdf', { type: 'application/pdf' })
+      const event = {
+        target: { files: [mockFile] },
+      } as unknown as React.ChangeEvent<HTMLInputElement>
+
+      act(() => {
+        result.current.fileChangeHandle(event)
+      })
+
+      await waitFor(() => {
+        expect(mockUpload).toHaveBeenCalled()
+      })
+    })
+
+    it('should update progress during upload', async () => {
+      let progressCallback: ((e: ProgressEvent) => void) | undefined
+
+      mockUpload.mockImplementation(async (options: { onprogress: (e: ProgressEvent) => void }) => {
+        progressCallback = options.onprogress
+        return { id: 'uploaded-id' }
+      })
+
+      const onFileUpdate = vi.fn()
+
+      const { result } = renderHook(
+        () => useFileUpload({ ...defaultOptions, onFileUpdate }),
+        { wrapper: createWrapper() },
+      )
+
+      const mockFile = new File(['content'], 'test.pdf', { type: 'application/pdf' })
+      const event = {
+        target: { files: [mockFile] },
+      } as unknown as React.ChangeEvent<HTMLInputElement>
+
+      act(() => {
+        result.current.fileChangeHandle(event)
+      })
+
+      await waitFor(() => {
+        expect(mockUpload).toHaveBeenCalled()
+      })
+
+      if (progressCallback) {
+        act(() => {
+          progressCallback!({
+            lengthComputable: true,
+            loaded: 50,
+            total: 100,
+          } as ProgressEvent)
+        })
+
+        expect(onFileUpdate).toHaveBeenCalled()
+      }
+    })
+
+    it('should handle upload error', async () => {
+      mockUpload.mockRejectedValue(new Error('Upload failed'))
+      const onFileUpdate = vi.fn()
+
+      const { result } = renderHook(
+        () => useFileUpload({ ...defaultOptions, onFileUpdate }),
+        { wrapper: createWrapper() },
+      )
+
+      const mockFile = new File(['content'], 'test.pdf', { type: 'application/pdf' })
+      const event = {
+        target: { files: [mockFile] },
+      } as unknown as React.ChangeEvent<HTMLInputElement>
+
+      act(() => {
+        result.current.fileChangeHandle(event)
+      })
+
+      await waitFor(() => {
+        expect(mockNotify).toHaveBeenCalledWith(
+          expect.objectContaining({ type: 'error' }),
+        )
+      })
+    })
+
+    it('should update file with PROGRESS_COMPLETE on success', async () => {
+      mockUpload.mockResolvedValue({ id: 'uploaded-id', name: 'test.pdf' })
+      const onFileUpdate = vi.fn()
+
+      const { result } = renderHook(
+        () => useFileUpload({ ...defaultOptions, onFileUpdate }),
+        { wrapper: createWrapper() },
+      )
+
+      const mockFile = new File(['content'], 'test.pdf', { type: 'application/pdf' })
+      const event = {
+        target: { files: [mockFile] },
+      } as unknown as React.ChangeEvent<HTMLInputElement>
+
+      act(() => {
+        result.current.fileChangeHandle(event)
+      })
+
+      await waitFor(() => {
+        const completeCalls = onFileUpdate.mock.calls.filter(
+          ([, progress]) => progress === PROGRESS_COMPLETE,
+        )
+        expect(completeCalls.length).toBeGreaterThan(0)
+      })
+    })
+
+    it('should update file with PROGRESS_ERROR on failure', async () => {
+      mockUpload.mockRejectedValue(new Error('Upload failed'))
+      const onFileUpdate = vi.fn()
+
+      const { result } = renderHook(
+        () => useFileUpload({ ...defaultOptions, onFileUpdate }),
+        { wrapper: createWrapper() },
+      )
+
+      const mockFile = new File(['content'], 'test.pdf', { type: 'application/pdf' })
+      const event = {
+        target: { files: [mockFile] },
+      } as unknown as React.ChangeEvent<HTMLInputElement>
+
+      act(() => {
+        result.current.fileChangeHandle(event)
+      })
+
+      await waitFor(() => {
+        const errorCalls = onFileUpdate.mock.calls.filter(
+          ([, progress]) => progress === PROGRESS_ERROR,
+        )
+        expect(errorCalls.length).toBeGreaterThan(0)
+      })
+    })
+  })
+
+  describe('file count validation', () => {
+    it('should reject when total files exceed limit', () => {
+      const existingFiles: FileItem[] = Array.from({ length: 8 }, (_, i) => ({
+        fileID: `existing-${i}`,
+        file: { name: `existing-${i}.pdf`, size: 1024 } as CustomFile,
+        progress: 100,
+      }))
+
+      const { result } = renderHook(
+        () => useFileUpload({
+          ...defaultOptions,
+          fileList: existingFiles,
+        }),
+        { wrapper: createWrapper() },
+      )
+
+      const files = Array.from({ length: 5 }, (_, i) =>
+        new File(['content'], `new-${i}.pdf`, { type: 'application/pdf' }))
+
+      const event = {
+        target: { files },
+      } as unknown as React.ChangeEvent<HTMLInputElement>
+
+      act(() => {
+        result.current.fileChangeHandle(event)
+      })
+
+      expect(mockNotify).toHaveBeenCalledWith(
+        expect.objectContaining({ type: 'error' }),
+      )
+    })
+  })
+
+  describe('progress constants', () => {
+    it('should use PROGRESS_NOT_STARTED for new files', async () => {
+      mockUpload.mockResolvedValue({ id: 'file-id' })
+
+      const prepareFileList = vi.fn()
+      const { result } = renderHook(
+        () => useFileUpload({ ...defaultOptions, prepareFileList }),
+        { wrapper: createWrapper() },
+      )
+
+      const mockFile = new File(['content'], 'test.pdf', { type: 'application/pdf' })
+      const event = {
+        target: { files: [mockFile] },
+      } as unknown as React.ChangeEvent<HTMLInputElement>
+
+      act(() => {
+        result.current.fileChangeHandle(event)
+      })
+
+      await waitFor(() => {
+        if (prepareFileList.mock.calls.length > 0) {
+          const files = prepareFileList.mock.calls[0][0]
+          expect(files[0].progress).toBe(PROGRESS_NOT_STARTED)
+        }
+      })
+    })
+  })
+})

web/app/components/datasets/create/file-uploader/hooks/use-file-upload.ts

@@ -0,0 +1,351 @@
+'use client'
+import type { RefObject } from 'react'
+import type { CustomFile as File, FileItem } from '@/models/datasets'
+import { useCallback, useEffect, useMemo, useRef, useState } from 'react'
+import { useTranslation } from 'react-i18next'
+import { useContext } from 'use-context-selector'
+import { getFileUploadErrorMessage } from '@/app/components/base/file-uploader/utils'
+import { ToastContext } from '@/app/components/base/toast'
+import { IS_CE_EDITION } from '@/config'
+import { useLocale } from '@/context/i18n'
+import { LanguagesSupported } from '@/i18n-config/language'
+import { upload } from '@/service/base'
+import { useFileSupportTypes, useFileUploadConfig } from '@/service/use-common'
+import { getFileExtension } from '@/utils/format'
+import { PROGRESS_COMPLETE, PROGRESS_ERROR, PROGRESS_NOT_STARTED } from '../constants'
+
+export type FileUploadConfig = {
+  file_size_limit: number
+  batch_count_limit: number
+  file_upload_limit: number
+}
+
+export type UseFileUploadOptions = {
+  fileList: FileItem[]
+  prepareFileList: (files: FileItem[]) => void
+  onFileUpdate: (fileItem: FileItem, progress: number, list: FileItem[]) => void
+  onFileListUpdate?: (files: FileItem[]) => void
+  onPreview: (file: File) => void
+  supportBatchUpload?: boolean
+  /**
+   * Optional list of allowed file extensions. If not provided, fetches from API.
+   * Pass this when you need custom extension filtering instead of using the global config.
+   */
+  allowedExtensions?: string[]
+}
+
+export type UseFileUploadReturn = {
+  // Refs
+  dropRef: RefObject<HTMLDivElement | null>
+  dragRef: RefObject<HTMLDivElement | null>
+  fileUploaderRef: RefObject<HTMLInputElement | null>
+
+  // State
+  dragging: boolean
+
+  // Config
+  fileUploadConfig: FileUploadConfig
+  acceptTypes: string[]
+  supportTypesShowNames: string
+  hideUpload: boolean
+
+  // Handlers
+  selectHandle: () => void
+  fileChangeHandle: (e: React.ChangeEvent<HTMLInputElement>) => void
+  removeFile: (fileID: string) => void
+  handlePreview: (file: File) => void
+}
+
+type FileWithPath = {
+  relativePath?: string
+} & File
+
+export const useFileUpload = ({
+  fileList,
+  prepareFileList,
+  onFileUpdate,
+  onFileListUpdate,
+  onPreview,
+  supportBatchUpload = false,
+  allowedExtensions,
+}: UseFileUploadOptions): UseFileUploadReturn => {
+  const { t } = useTranslation()
+  const { notify } = useContext(ToastContext)
+  const locale = useLocale()
+
+  const [dragging, setDragging] = useState(false)
+  const dropRef = useRef<HTMLDivElement>(null)
+  const dragRef = useRef<HTMLDivElement>(null)
+  const fileUploaderRef = useRef<HTMLInputElement>(null)
+  const fileListRef = useRef<FileItem[]>([])
+
+  const hideUpload = !supportBatchUpload && fileList.length > 0
+
+  const { data: fileUploadConfigResponse } = useFileUploadConfig()
+  const { data: supportFileTypesResponse } = useFileSupportTypes()
+  // Use provided allowedExtensions or fetch from API
+  const supportTypes = useMemo(
+    () => allowedExtensions ?? supportFileTypesResponse?.allowed_extensions ?? [],
+    [allowedExtensions, supportFileTypesResponse?.allowed_extensions],
+  )
+
+  const supportTypesShowNames = useMemo(() => {
+    const extensionMap: { [key: string]: string } = {
+      md: 'markdown',
+      pptx: 'pptx',
+      htm: 'html',
+      xlsx: 'xlsx',
+      docx: 'docx',
+    }
+
+    return [...supportTypes]
+      .map(item => extensionMap[item] || item)
+      .map(item => item.toLowerCase())
+      .filter((item, index, self) => self.indexOf(item) === index)
+      .map(item => item.toUpperCase())
+      .join(locale !== LanguagesSupported[1] ? ', ' : '、 ')
+  }, [supportTypes, locale])
+
+  const acceptTypes = useMemo(() => supportTypes.map((ext: string) => `.${ext}`), [supportTypes])
+
+  const fileUploadConfig = useMemo(() => ({
+    file_size_limit: fileUploadConfigResponse?.file_size_limit ?? 15,
+    batch_count_limit: supportBatchUpload ? (fileUploadConfigResponse?.batch_count_limit ?? 5) : 1,
+    file_upload_limit: supportBatchUpload ? (fileUploadConfigResponse?.file_upload_limit ?? 5) : 1,
+  }), [fileUploadConfigResponse, supportBatchUpload])
+
+  const isValid = useCallback((file: File) => {
+    const { size } = file
+    const ext = `.${getFileExtension(file.name)}`
+    const isValidType = acceptTypes.includes(ext.toLowerCase())
+    if (!isValidType)
+      notify({ type: 'error', message: t('stepOne.uploader.validation.typeError', { ns: 'datasetCreation' }) })
+
+    const isValidSize = size <= fileUploadConfig.file_size_limit * 1024 * 1024
+    if (!isValidSize)
+      notify({ type: 'error', message: t('stepOne.uploader.validation.size', { ns: 'datasetCreation', size: fileUploadConfig.file_size_limit }) })
+
+    return isValidType && isValidSize
+  }, [fileUploadConfig, notify, t, acceptTypes])
+
+  const fileUpload = useCallback(async (fileItem: FileItem): Promise<FileItem> => {
+    const formData = new FormData()
+    formData.append('file', fileItem.file)
+    const onProgress = (e: ProgressEvent) => {
+      if (e.lengthComputable) {
+        const percent = Math.floor(e.loaded / e.total * 100)
+        onFileUpdate(fileItem, percent, fileListRef.current)
+      }
+    }
+
+    return upload({
+      xhr: new XMLHttpRequest(),
+      data: formData,
+      onprogress: onProgress,
+    }, false, undefined, '?source=datasets')
+      .then((res) => {
+        const completeFile = {
+          fileID: fileItem.fileID,
+          file: res as unknown as File,
+          progress: PROGRESS_NOT_STARTED,
+        }
+        const index = fileListRef.current.findIndex(item => item.fileID === fileItem.fileID)
+        fileListRef.current[index] = completeFile
+        onFileUpdate(completeFile, PROGRESS_COMPLETE, fileListRef.current)
+        return Promise.resolve({ ...completeFile })
+      })
+      .catch((e) => {
+        const errorMessage = getFileUploadErrorMessage(e, t('stepOne.uploader.failed', { ns: 'datasetCreation' }), t)
+        notify({ type: 'error', message: errorMessage })
+        onFileUpdate(fileItem, PROGRESS_ERROR, fileListRef.current)
+        return Promise.resolve({ ...fileItem })
+      })
+      .finally()
+  }, [notify, onFileUpdate, t])
+
+  const uploadBatchFiles = useCallback((bFiles: FileItem[]) => {
+    bFiles.forEach(bf => (bf.progress = 0))
+    return Promise.all(bFiles.map(fileUpload))
+  }, [fileUpload])
+
+  const uploadMultipleFiles = useCallback(async (files: FileItem[]) => {
+    const batchCountLimit = fileUploadConfig.batch_count_limit
+    const length = files.length
+    let start = 0
+    let end = 0
+
+    while (start < length) {
+      if (start + batchCountLimit > length)
+        end = length
+      else
+        end = start + batchCountLimit
+      const bFiles = files.slice(start, end)
+      await uploadBatchFiles(bFiles)
+      start = end
+    }
+  }, [fileUploadConfig, uploadBatchFiles])
+
+  const initialUpload = useCallback((files: File[]) => {
+    const filesCountLimit = fileUploadConfig.file_upload_limit
+    if (!files.length)
+      return false
+
+    if (files.length + fileList.length > filesCountLimit && !IS_CE_EDITION) {
+      notify({ type: 'error', message: t('stepOne.uploader.validation.filesNumber', { ns: 'datasetCreation', filesNumber: filesCountLimit }) })
+      return false
+    }
+
+    const preparedFiles = files.map((file, index) => ({
+      fileID: `file${index}-${Date.now()}`,
+      file,
+      progress: PROGRESS_NOT_STARTED,
+    }))
+    const newFiles = [...fileListRef.current, ...preparedFiles]
+    prepareFileList(newFiles)
+    fileListRef.current = newFiles
+    uploadMultipleFiles(preparedFiles)
+  }, [prepareFileList, uploadMultipleFiles, notify, t, fileList, fileUploadConfig])
+
+  const traverseFileEntry = useCallback(
+    (entry: FileSystemEntry, prefix = ''): Promise<FileWithPath[]> => {
+      return new Promise((resolve) => {
+        if (entry.isFile) {
+          (entry as FileSystemFileEntry).file((file: FileWithPath) => {
+            file.relativePath = `${prefix}${file.name}`
+            resolve([file])
+          })
+        }
+        else if (entry.isDirectory) {
+          const reader = (entry as FileSystemDirectoryEntry).createReader()
+          const entries: FileSystemEntry[] = []
+          const read = () => {
+            reader.readEntries(async (results: FileSystemEntry[]) => {
+              if (!results.length) {
+                const files = await Promise.all(
+                  entries.map(ent =>
+                    traverseFileEntry(ent, `${prefix}${entry.name}/`),
+                  ),
+                )
+                resolve(files.flat())
+              }
+              else {
+                entries.push(...results)
+                read()
+              }
+            })
+          }
+          read()
+        }
+        else {
+          resolve([])
+        }
+      })
+    },
+    [],
+  )
+
+  const handleDragEnter = useCallback((e: DragEvent) => {
+    e.preventDefault()
+    e.stopPropagation()
+    if (e.target !== dragRef.current)
+      setDragging(true)
+  }, [])
+
+  const handleDragOver = useCallback((e: DragEvent) => {
+    e.preventDefault()
+    e.stopPropagation()
+  }, [])
+
+  const handleDragLeave = useCallback((e: DragEvent) => {
+    e.preventDefault()
+    e.stopPropagation()
+    if (e.target === dragRef.current)
+      setDragging(false)
+  }, [])
+
+  const handleDrop = useCallback(
+    async (e: DragEvent) => {
+      e.preventDefault()
+      e.stopPropagation()
+      setDragging(false)
+      if (!e.dataTransfer)
+        return
+      const nested = await Promise.all(
+        Array.from(e.dataTransfer.items).map((it) => {
+          const entry = (it as DataTransferItem & { webkitGetAsEntry?: () => FileSystemEntry | null }).webkitGetAsEntry?.()
+          if (entry)
+            return traverseFileEntry(entry)
+          const f = it.getAsFile?.()
+          return f ? Promise.resolve([f as FileWithPath]) : Promise.resolve([])
+        }),
+      )
+      let files = nested.flat()
+      if (!supportBatchUpload)
+        files = files.slice(0, 1)
+      files = files.slice(0, fileUploadConfig.batch_count_limit)
+      const valid = files.filter(isValid)
+      initialUpload(valid)
+    },
+    [initialUpload, isValid, supportBatchUpload, traverseFileEntry, fileUploadConfig],
+  )
+
+  const selectHandle = useCallback(() => {
+    if (fileUploaderRef.current)
+      fileUploaderRef.current.click()
+  }, [])
+
+  const removeFile = useCallback((fileID: string) => {
+    if (fileUploaderRef.current)
+      fileUploaderRef.current.value = ''
+
+    fileListRef.current = fileListRef.current.filter(item => item.fileID !== fileID)
+    onFileListUpdate?.([...fileListRef.current])
+  }, [onFileListUpdate])
+
+  const fileChangeHandle = useCallback((e: React.ChangeEvent<HTMLInputElement>) => {
+    let files = Array.from(e.target.files ?? []) as File[]
+    files = files.slice(0, fileUploadConfig.batch_count_limit)
+    initialUpload(files.filter(isValid))
+  }, [isValid, initialUpload, fileUploadConfig])
+
+  const handlePreview = useCallback((file: File) => {
+    if (file?.id)
+      onPreview(file)
+  }, [onPreview])
+
+  useEffect(() => {
+    const dropArea = dropRef.current
+    dropArea?.addEventListener('dragenter', handleDragEnter)
+    dropArea?.addEventListener('dragover', handleDragOver)
+    dropArea?.addEventListener('dragleave', handleDragLeave)
+    dropArea?.addEventListener('drop', handleDrop)
+    return () => {
+      dropArea?.removeEventListener('dragenter', handleDragEnter)
+      dropArea?.removeEventListener('dragover', handleDragOver)
+      dropArea?.removeEventListener('dragleave', handleDragLeave)
+      dropArea?.removeEventListener('drop', handleDrop)
+    }
+  }, [handleDragEnter, handleDragOver, handleDragLeave, handleDrop])
+
+  return {
+    // Refs
+    dropRef,
+    dragRef,
+    fileUploaderRef,
+
+    // State
+    dragging,
+
+    // Config
+    fileUploadConfig,
+    acceptTypes,
+    supportTypesShowNames,
+    hideUpload,
+
+    // Handlers
+    selectHandle,
+    fileChangeHandle,
+    removeFile,
+    handlePreview,
+  }
+}

web/app/components/datasets/create/file-uploader/index.spec.tsx

@@ -0,0 +1,278 @@
+import type { CustomFile as File, FileItem } from '@/models/datasets'
+import { fireEvent, render, screen } from '@testing-library/react'
+import { beforeEach, describe, expect, it, vi } from 'vitest'
+import { PROGRESS_NOT_STARTED } from './constants'
+import FileUploader from './index'
+
+// Mock react-i18next
+vi.mock('react-i18next', () => ({
+  useTranslation: () => ({
+    t: (key: string) => {
+      const translations: Record<string, string> = {
+        'stepOne.uploader.title': 'Upload Files',
+        'stepOne.uploader.button': 'Drag and drop files, or',
+        'stepOne.uploader.buttonSingleFile': 'Drag and drop file, or',
+        'stepOne.uploader.browse': 'Browse',
+        'stepOne.uploader.tip': 'Supports various file types',
+      }
+      return translations[key] || key
+    },
+  }),
+}))
+
+// Mock ToastContext
+const mockNotify = vi.fn()
+vi.mock('use-context-selector', async () => {
+  const actual = await vi.importActual<typeof import('use-context-selector')>('use-context-selector')
+  return {
+    ...actual,
+    useContext: vi.fn(() => ({ notify: mockNotify })),
+  }
+})
+
+// Mock services
+vi.mock('@/service/base', () => ({
+  upload: vi.fn().mockResolvedValue({ id: 'uploaded-id' }),
+}))
+
+vi.mock('@/service/use-common', () => ({
+  useFileUploadConfig: () => ({
+    data: { file_size_limit: 15, batch_count_limit: 5, file_upload_limit: 10 },
+  }),
+  useFileSupportTypes: () => ({
+    data: { allowed_extensions: ['pdf', 'docx', 'txt'] },
+  }),
+}))
+
+vi.mock('@/context/i18n', () => ({
+  useLocale: () => 'en-US',
+}))
+
+vi.mock('@/i18n-config/language', () => ({
+  LanguagesSupported: ['en-US', 'zh-Hans'],
+}))
+
+vi.mock('@/config', () => ({
+  IS_CE_EDITION: false,
+}))
+
+vi.mock('@/app/components/base/file-uploader/utils', () => ({
+  getFileUploadErrorMessage: () => 'Upload error',
+}))
+
+// Mock theme
+vi.mock('@/hooks/use-theme', () => ({
+  default: () => ({ theme: 'light' }),
+}))
+
+vi.mock('@/types/app', () => ({
+  Theme: { dark: 'dark', light: 'light' },
+}))
+
+// Mock DocumentFileIcon - uses relative path from file-list-item.tsx
+vi.mock('@/app/components/datasets/common/document-file-icon', () => ({
+  default: ({ extension }: { extension: string }) => <div data-testid="document-icon">{extension}</div>,
+}))
+
+// Mock SimplePieChart
+vi.mock('next/dynamic', () => ({
+  default: () => {
+    const Component = ({ percentage }: { percentage: number }) => (
+      <div data-testid="pie-chart">
+        {percentage}
+        %
+      </div>
+    )
+    return Component
+  },
+}))
+
+describe('FileUploader', () => {
+  const createMockFile = (overrides: Partial<File> = {}): File => ({
+    name: 'test.pdf',
+    size: 1024,
+    type: 'application/pdf',
+    ...overrides,
+  } as File)
+
+  const createMockFileItem = (overrides: Partial<FileItem> = {}): FileItem => ({
+    fileID: `file-${Date.now()}`,
+    file: createMockFile(overrides.file as Partial<File>),
+    progress: PROGRESS_NOT_STARTED,
+    ...overrides,
+  })
+
+  const defaultProps = {
+    fileList: [] as FileItem[],
+    prepareFileList: vi.fn(),
+    onFileUpdate: vi.fn(),
+    onFileListUpdate: vi.fn(),
+    onPreview: vi.fn(),
+    supportBatchUpload: true,
+  }
+
+  beforeEach(() => {
+    vi.clearAllMocks()
+  })
+
+  describe('rendering', () => {
+    it('should render the component', () => {
+      render(<FileUploader {...defaultProps} />)
+      expect(screen.getByText('Upload Files')).toBeInTheDocument()
+    })
+
+    it('should render dropzone when no files', () => {
+      render(<FileUploader {...defaultProps} />)
+      expect(screen.getByText(/Drag and drop files/i)).toBeInTheDocument()
+    })
+
+    it('should render browse button', () => {
+      render(<FileUploader {...defaultProps} />)
+      expect(screen.getByText('Browse')).toBeInTheDocument()
+    })
+
+    it('should apply custom title className', () => {
+      render(<FileUploader {...defaultProps} titleClassName="custom-class" />)
+      const title = screen.getByText('Upload Files')
+      expect(title).toHaveClass('custom-class')
+    })
+  })
+
+  describe('file list rendering', () => {
+    it('should render file items when fileList has items', () => {
+      const fileList = [
+        createMockFileItem({ file: createMockFile({ name: 'file1.pdf' }) }),
+        createMockFileItem({ file: createMockFile({ name: 'file2.pdf' }) }),
+      ]
+
+      render(<FileUploader {...defaultProps} fileList={fileList} />)
+
+      expect(screen.getByText('file1.pdf')).toBeInTheDocument()
+      expect(screen.getByText('file2.pdf')).toBeInTheDocument()
+    })
+
+    it('should render document icons for files', () => {
+      const fileList = [createMockFileItem()]
+      render(<FileUploader {...defaultProps} fileList={fileList} />)
+
+      expect(screen.getByTestId('document-icon')).toBeInTheDocument()
+    })
+  })
+
+  describe('batch upload mode', () => {
+    it('should show dropzone with batch upload enabled', () => {
+      render(<FileUploader {...defaultProps} supportBatchUpload={true} />)
+      expect(screen.getByText(/Drag and drop files/i)).toBeInTheDocument()
+    })
+
+    it('should show single file text when batch upload disabled', () => {
+      render(<FileUploader {...defaultProps} supportBatchUpload={false} />)
+      expect(screen.getByText(/Drag and drop file/i)).toBeInTheDocument()
+    })
+
+    it('should hide dropzone when not batch upload and has files', () => {
+      const fileList = [createMockFileItem()]
+      render(<FileUploader {...defaultProps} supportBatchUpload={false} fileList={fileList} />)
+
+      expect(screen.queryByText(/Drag and drop/i)).not.toBeInTheDocument()
+    })
+  })
+
+  describe('event handlers', () => {
+    it('should handle file preview click', () => {
+      const onPreview = vi.fn()
+      const fileItem = createMockFileItem({
+        file: createMockFile({ id: 'file-id' } as Partial<File>),
+      })
+
+      const { container } = render(<FileUploader {...defaultProps} fileList={[fileItem]} onPreview={onPreview} />)
+
+      // Find the file list item container by its class pattern
+      const fileElement = container.querySelector('[class*="flex h-12"]')
+      if (fileElement)
+        fireEvent.click(fileElement)
+
+      expect(onPreview).toHaveBeenCalledWith(fileItem.file)
+    })
+
+    it('should handle file remove click', () => {
+      const onFileListUpdate = vi.fn()
+      const fileItem = createMockFileItem()
+
+      const { container } = render(
+        <FileUploader {...defaultProps} fileList={[fileItem]} onFileListUpdate={onFileListUpdate} />,
+      )
+
+      // Find the delete button (the span with cursor-pointer containing the icon)
+      const deleteButtons = container.querySelectorAll('[class*="cursor-pointer"]')
+      // Get the last one which should be the delete button (not the browse label)
+      const deleteButton = deleteButtons[deleteButtons.length - 1]
+      if (deleteButton)
+        fireEvent.click(deleteButton)
+
+      expect(onFileListUpdate).toHaveBeenCalled()
+    })
+
+    it('should handle browse button click', () => {
+      render(<FileUploader {...defaultProps} />)
+
+      // The browse label should trigger file input click
+      const browseLabel = screen.getByText('Browse')
+      expect(browseLabel).toHaveClass('cursor-pointer')
+    })
+  })
+
+  describe('upload progress', () => {
+    it('should show progress chart for uploading files', () => {
+      const fileItem = createMockFileItem({ progress: 50 })
+      render(<FileUploader {...defaultProps} fileList={[fileItem]} />)
+
+      expect(screen.getByTestId('pie-chart')).toBeInTheDocument()
+      expect(screen.getByText('50%')).toBeInTheDocument()
+    })
+
+    it('should not show progress chart for completed files', () => {
+      const fileItem = createMockFileItem({ progress: 100 })
+      render(<FileUploader {...defaultProps} fileList={[fileItem]} />)
+
+      expect(screen.queryByTestId('pie-chart')).not.toBeInTheDocument()
+    })
+
+    it('should not show progress chart for not started files', () => {
+      const fileItem = createMockFileItem({ progress: PROGRESS_NOT_STARTED })
+      render(<FileUploader {...defaultProps} fileList={[fileItem]} />)
+
+      expect(screen.queryByTestId('pie-chart')).not.toBeInTheDocument()
+    })
+  })
+
+  describe('multiple files', () => {
+    it('should render all files in the list', () => {
+      const fileList = [
+        createMockFileItem({ fileID: 'f1', file: createMockFile({ name: 'doc1.pdf' }) }),
+        createMockFileItem({ fileID: 'f2', file: createMockFile({ name: 'doc2.docx' }) }),
+        createMockFileItem({ fileID: 'f3', file: createMockFile({ name: 'doc3.txt' }) }),
+      ]
+
+      render(<FileUploader {...defaultProps} fileList={fileList} />)
+
+      expect(screen.getByText('doc1.pdf')).toBeInTheDocument()
+      expect(screen.getByText('doc2.docx')).toBeInTheDocument()
+      expect(screen.getByText('doc3.txt')).toBeInTheDocument()
+    })
+  })
+
+  describe('styling', () => {
+    it('should have correct container width', () => {
+      const { container } = render(<FileUploader {...defaultProps} />)
+      const wrapper = container.firstChild as HTMLElement
+      expect(wrapper).toHaveClass('w-[640px]')
+    })
+
+    it('should have proper spacing', () => {
+      const { container } = render(<FileUploader {...defaultProps} />)
+      const wrapper = container.firstChild as HTMLElement
+      expect(wrapper).toHaveClass('mb-5')
+    })
+  })
+})

web/app/components/datasets/create/file-uploader/index.tsx

@@ -1,23 +1,10 @@
 'use client'
 import type { CustomFile as File, FileItem } from '@/models/datasets'
-import { RiDeleteBinLine, RiUploadCloud2Line } from '@remixicon/react'
-import * as React from 'react'
-import { useCallback, useEffect, useMemo, useRef, useState } from 'react'
 import { useTranslation } from 'react-i18next'
-import { useContext } from 'use-context-selector'
-import { getFileUploadErrorMessage } from '@/app/components/base/file-uploader/utils'
-import SimplePieChart from '@/app/components/base/simple-pie-chart'
-import { ToastContext } from '@/app/components/base/toast'
-import { IS_CE_EDITION } from '@/config'
-
-import { useLocale } from '@/context/i18n'
-import useTheme from '@/hooks/use-theme'
-import { LanguagesSupported } from '@/i18n-config/language'
-import { upload } from '@/service/base'
-import { useFileSupportTypes, useFileUploadConfig } from '@/service/use-common'
-import { Theme } from '@/types/app'
 import { cn } from '@/utils/classnames'
-import DocumentFileIcon from '../../common/document-file-icon'
+import FileListItem from './components/file-list-item'
+import UploadDropzone from './components/upload-dropzone'
+import { useFileUpload } from './hooks/use-file-upload'
 
 type IFileUploaderProps = {
   fileList: FileItem[]
@@ -39,358 +26,62 @@ const FileUploader = ({
   supportBatchUpload = false,
 }: IFileUploaderProps) => {
   const { t } = useTranslation()
-  const { notify } = useContext(ToastContext)
-  const locale = useLocale()
-  const [dragging, setDragging] = useState(false)
-  const dropRef = useRef<HTMLDivElement>(null)
-  const dragRef = useRef<HTMLDivElement>(null)
-  const fileUploader = useRef<HTMLInputElement>(null)
-  const hideUpload = !supportBatchUpload && fileList.length > 0
 
-  const { data: fileUploadConfigResponse } = useFileUploadConfig()
-  const { data: supportFileTypesResponse } = useFileSupportTypes()
-  const supportTypes = supportFileTypesResponse?.allowed_extensions || []
-  const supportTypesShowNames = (() => {
-    const extensionMap: { [key: string]: string } = {
-      md: 'markdown',
-      pptx: 'pptx',
-      htm: 'html',
-      xlsx: 'xlsx',
-      docx: 'docx',
-    }
-
-    return [...supportTypes]
-      .map(item => extensionMap[item] || item) // map to standardized extension
-      .map(item => item.toLowerCase()) // convert to lower case
-      .filter((item, index, self) => self.indexOf(item) === index) // remove duplicates
-      .map(item => item.toUpperCase()) // convert to upper case
-      .join(locale !== LanguagesSupported[1] ? ', ' : '、 ')
-  })()
-  const ACCEPTS = supportTypes.map((ext: string) => `.${ext}`)
-  const fileUploadConfig = useMemo(() => ({
-    file_size_limit: fileUploadConfigResponse?.file_size_limit ?? 15,
-    batch_count_limit: supportBatchUpload ? (fileUploadConfigResponse?.batch_count_limit ?? 5) : 1,
-    file_upload_limit: supportBatchUpload ? (fileUploadConfigResponse?.file_upload_limit ?? 5) : 1,
-  }), [fileUploadConfigResponse, supportBatchUpload])
-
-  const fileListRef = useRef<FileItem[]>([])
-
-  // utils
-  const getFileType = (currentFile: File) => {
-    if (!currentFile)
-      return ''
-
-    const arr = currentFile.name.split('.')
-    return arr[arr.length - 1]
-  }
-
-  const getFileSize = (size: number) => {
-    if (size / 1024 < 10)
-      return `${(size / 1024).toFixed(2)}KB`
-
-    return `${(size / 1024 / 1024).toFixed(2)}MB`
-  }
-
-  const isValid = useCallback((file: File) => {
-    const { size } = file
-    const ext = `.${getFileType(file)}`
-    const isValidType = ACCEPTS.includes(ext.toLowerCase())
-    if (!isValidType)
-      notify({ type: 'error', message: t('stepOne.uploader.validation.typeError', { ns: 'datasetCreation' }) })
-
-    const isValidSize = size <= fileUploadConfig.file_size_limit * 1024 * 1024
-    if (!isValidSize)
-      notify({ type: 'error', message: t('stepOne.uploader.validation.size', { ns: 'datasetCreation', size: fileUploadConfig.file_size_limit }) })
-
-    return isValidType && isValidSize
-  }, [fileUploadConfig, notify, t, ACCEPTS])
-
-  const fileUpload = useCallback(async (fileItem: FileItem): Promise<FileItem> => {
-    const formData = new FormData()
-    formData.append('file', fileItem.file)
-    const onProgress = (e: ProgressEvent) => {
-      if (e.lengthComputable) {
-        const percent = Math.floor(e.loaded / e.total * 100)
-        onFileUpdate(fileItem, percent, fileListRef.current)
-      }
-    }
-
-    return upload({
-      xhr: new XMLHttpRequest(),
-      data: formData,
-      onprogress: onProgress,
-    }, false, undefined, '?source=datasets')
-      .then((res) => {
-        const completeFile = {
-          fileID: fileItem.fileID,
-          file: res as unknown as File,
-          progress: -1,
-        }
-        const index = fileListRef.current.findIndex(item => item.fileID === fileItem.fileID)
-        fileListRef.current[index] = completeFile
-        onFileUpdate(completeFile, 100, fileListRef.current)
-        return Promise.resolve({ ...completeFile })
-      })
-      .catch((e) => {
-        const errorMessage = getFileUploadErrorMessage(e, t('stepOne.uploader.failed', { ns: 'datasetCreation' }), t)
-        notify({ type: 'error', message: errorMessage })
-        onFileUpdate(fileItem, -2, fileListRef.current)
-        return Promise.resolve({ ...fileItem })
-      })
-      .finally()
-  }, [fileListRef, notify, onFileUpdate, t])
-
-  const uploadBatchFiles = useCallback((bFiles: FileItem[]) => {
-    bFiles.forEach(bf => (bf.progress = 0))
-    return Promise.all(bFiles.map(fileUpload))
-  }, [fileUpload])
-
-  const uploadMultipleFiles = useCallback(async (files: FileItem[]) => {
-    const batchCountLimit = fileUploadConfig.batch_count_limit
-    const length = files.length
-    let start = 0
-    let end = 0
-
-    while (start < length) {
-      if (start + batchCountLimit > length)
-        end = length
-      else
-        end = start + batchCountLimit
-      const bFiles = files.slice(start, end)
-      await uploadBatchFiles(bFiles)
-      start = end
-    }
-  }, [fileUploadConfig, uploadBatchFiles])
-
-  const initialUpload = useCallback((files: File[]) => {
-    const filesCountLimit = fileUploadConfig.file_upload_limit
-    if (!files.length)
-      return false
-
-    if (files.length + fileList.length > filesCountLimit && !IS_CE_EDITION) {
-      notify({ type: 'error', message: t('stepOne.uploader.validation.filesNumber', { ns: 'datasetCreation', filesNumber: filesCountLimit }) })
-      return false
-    }
-
-    const preparedFiles = files.map((file, index) => ({
-      fileID: `file${index}-${Date.now()}`,
-      file,
-      progress: -1,
-    }))
-    const newFiles = [...fileListRef.current, ...preparedFiles]
-    prepareFileList(newFiles)
-    fileListRef.current = newFiles
-    uploadMultipleFiles(preparedFiles)
-  }, [prepareFileList, uploadMultipleFiles, notify, t, fileList, fileUploadConfig])
-
-  const handleDragEnter = (e: DragEvent) => {
-    e.preventDefault()
-    e.stopPropagation()
-    if (e.target !== dragRef.current)
-      setDragging(true)
-  }
-  const handleDragOver = (e: DragEvent) => {
-    e.preventDefault()
-    e.stopPropagation()
-  }
-  const handleDragLeave = (e: DragEvent) => {
-    e.preventDefault()
-    e.stopPropagation()
-    if (e.target === dragRef.current)
-      setDragging(false)
-  }
-  type FileWithPath = {
-    relativePath?: string
-  } & File
-  const traverseFileEntry = useCallback(
-    (entry: any, prefix = ''): Promise<FileWithPath[]> => {
-      return new Promise((resolve) => {
-        if (entry.isFile) {
-          entry.file((file: FileWithPath) => {
-            file.relativePath = `${prefix}${file.name}`
-            resolve([file])
-          })
-        }
-        else if (entry.isDirectory) {
-          const reader = entry.createReader()
-          const entries: any[] = []
-          const read = () => {
-            reader.readEntries(async (results: FileSystemEntry[]) => {
-              if (!results.length) {
-                const files = await Promise.all(
-                  entries.map(ent =>
-                    traverseFileEntry(ent, `${prefix}${entry.name}/`),
-                  ),
-                )
-                resolve(files.flat())
-              }
-              else {
-                entries.push(...results)
-                read()
-              }
-            })
-          }
-          read()
-        }
-        else {
-          resolve([])
-        }
-      })
-    },
-    [],
-  )
-
-  const handleDrop = useCallback(
-    async (e: DragEvent) => {
-      e.preventDefault()
-      e.stopPropagation()
-      setDragging(false)
-      if (!e.dataTransfer)
-        return
-      const nested = await Promise.all(
-        Array.from(e.dataTransfer.items).map((it) => {
-          const entry = (it as any).webkitGetAsEntry?.()
-          if (entry)
-            return traverseFileEntry(entry)
-          const f = it.getAsFile?.()
-          return f ? Promise.resolve([f]) : Promise.resolve([])
-        }),
-      )
-      let files = nested.flat()
-      if (!supportBatchUpload)
-        files = files.slice(0, 1)
-      files = files.slice(0, fileUploadConfig.batch_count_limit)
-      const valid = files.filter(isValid)
-      initialUpload(valid)
-    },
-    [initialUpload, isValid, supportBatchUpload, traverseFileEntry, fileUploadConfig],
-  )
-  const selectHandle = () => {
-    if (fileUploader.current)
-      fileUploader.current.click()
-  }
-
-  const removeFile = (fileID: string) => {
-    if (fileUploader.current)
-      fileUploader.current.value = ''
-
-    fileListRef.current = fileListRef.current.filter(item => item.fileID !== fileID)
-    onFileListUpdate?.([...fileListRef.current])
-  }
-  const fileChangeHandle = useCallback((e: React.ChangeEvent<HTMLInputElement>) => {
-    let files = Array.from(e.target.files ?? []) as File[]
-    files = files.slice(0, fileUploadConfig.batch_count_limit)
-    initialUpload(files.filter(isValid))
-  }, [isValid, initialUpload, fileUploadConfig])
-
-  const { theme } = useTheme()
-  const chartColor = useMemo(() => theme === Theme.dark ? '#5289ff' : '#296dff', [theme])
-
-  useEffect(() => {
-    dropRef.current?.addEventListener('dragenter', handleDragEnter)
-    dropRef.current?.addEventListener('dragover', handleDragOver)
-    dropRef.current?.addEventListener('dragleave', handleDragLeave)
-    dropRef.current?.addEventListener('drop', handleDrop)
-    return () => {
-      dropRef.current?.removeEventListener('dragenter', handleDragEnter)
-      dropRef.current?.removeEventListener('dragover', handleDragOver)
-      dropRef.current?.removeEventListener('dragleave', handleDragLeave)
-      dropRef.current?.removeEventListener('drop', handleDrop)
-    }
-  }, [handleDrop])
+  const {
+    dropRef,
+    dragRef,
+    fileUploaderRef,
+    dragging,
+    fileUploadConfig,
+    acceptTypes,
+    supportTypesShowNames,
+    hideUpload,
+    selectHandle,
+    fileChangeHandle,
+    removeFile,
+    handlePreview,
+  } = useFileUpload({
+    fileList,
+    prepareFileList,
+    onFileUpdate,
+    onFileListUpdate,
+    onPreview,
+    supportBatchUpload,
+  })
 
   return (
     <div className="mb-5 w-[640px]">
+      <div className={cn('mb-1 text-sm font-semibold leading-6 text-text-secondary', titleClassName)}>
+        {t('stepOne.uploader.title', { ns: 'datasetCreation' })}
+      </div>
+
       {!hideUpload && (
-        <input
-          ref={fileUploader}
-          id="fileUploader"
-          className="hidden"
-          type="file"
-          multiple={supportBatchUpload}
-          accept={ACCEPTS.join(',')}
-          onChange={fileChangeHandle}
+        <UploadDropzone
+          dropRef={dropRef}
+          dragRef={dragRef}
+          fileUploaderRef={fileUploaderRef}
+          dragging={dragging}
+          supportBatchUpload={supportBatchUpload}
+          supportTypesShowNames={supportTypesShowNames}
+          fileUploadConfig={fileUploadConfig}
+          acceptTypes={acceptTypes}
+          onSelectFile={selectHandle}
+          onFileChange={fileChangeHandle}
         />
       )}
 
-      <div className={cn('mb-1 text-sm font-semibold leading-6 text-text-secondary', titleClassName)}>{t('stepOne.uploader.title', { ns: 'datasetCreation' })}</div>
-
-      {!hideUpload && (
-        <div ref={dropRef} className={cn('relative mb-2 box-border flex min-h-20 max-w-[640px] flex-col items-center justify-center gap-1 rounded-xl border border-dashed border-components-dropzone-border bg-components-dropzone-bg px-4 py-3 text-xs leading-4 text-text-tertiary', dragging && 'border-components-dropzone-border-accent bg-components-dropzone-bg-accent')}>
-          <div className="flex min-h-5 items-center justify-center text-sm leading-4 text-text-secondary">
-            <RiUploadCloud2Line className="mr-2 size-5" />
-
-            <span>
-              {supportBatchUpload ? t('stepOne.uploader.button', { ns: 'datasetCreation' }) : t('stepOne.uploader.buttonSingleFile', { ns: 'datasetCreation' })}
-              {supportTypes.length > 0 && (
-                <label className="ml-1 cursor-pointer text-text-accent" onClick={selectHandle}>{t('stepOne.uploader.browse', { ns: 'datasetCreation' })}</label>
-              )}
-            </span>
-          </div>
-          <div>
-            {t('stepOne.uploader.tip', {
-              ns: 'datasetCreation',
-              size: fileUploadConfig.file_size_limit,
-              supportTypes: supportTypesShowNames,
-              batchCount: fileUploadConfig.batch_count_limit,
-              totalCount: fileUploadConfig.file_upload_limit,
-            })}
-          </div>
-          {dragging && <div ref={dragRef} className="absolute left-0 top-0 h-full w-full" />}
+      {fileList.length > 0 && (
+        <div className="max-w-[640px] cursor-default space-y-1">
+          {fileList.map(fileItem => (
+            <FileListItem
+              key={fileItem.fileID}
+              fileItem={fileItem}
+              onPreview={handlePreview}
+              onRemove={removeFile}
+            />
+          ))}
         </div>
       )}
-      <div className="max-w-[640px] cursor-default space-y-1">
-
-        {fileList.map((fileItem, index) => (
-          <div
-            key={`${fileItem.fileID}-${index}`}
-            onClick={() => fileItem.file?.id && onPreview(fileItem.file)}
-            className={cn(
-              'flex h-12 max-w-[640px] items-center rounded-lg border border-components-panel-border bg-components-panel-on-panel-item-bg text-xs leading-3 text-text-tertiary shadow-xs',
-              // 'border-state-destructive-border bg-state-destructive-hover',
-            )}
-          >
-            <div className="flex w-12 shrink-0 items-center justify-center">
-              <DocumentFileIcon
-                size="xl"
-                className="shrink-0"
-                name={fileItem.file.name}
-                extension={getFileType(fileItem.file)}
-              />
-            </div>
-            <div className="flex shrink grow flex-col gap-0.5">
-              <div className="flex w-full">
-                <div className="w-0 grow truncate text-sm leading-4 text-text-secondary">{fileItem.file.name}</div>
-              </div>
-              <div className="w-full truncate leading-3 text-text-tertiary">
-                <span className="uppercase">{getFileType(fileItem.file)}</span>
-                <span className="px-1 text-text-quaternary">·</span>
-                <span>{getFileSize(fileItem.file.size)}</span>
-                {/* <span className='px-1 text-text-quaternary'>·</span>
-                  <span>10k characters</span> */}
-              </div>
-            </div>
-            <div className="flex w-16 shrink-0 items-center justify-end gap-1 pr-3">
-              {/* <span className="flex justify-center items-center w-6 h-6 cursor-pointer">
-                  <RiErrorWarningFill className='size-4 text-text-warning' />
-                </span> */}
-              {(fileItem.progress < 100 && fileItem.progress >= 0) && (
-                // <div className={s.percent}>{`${fileItem.progress}%`}</div>
-                <SimplePieChart percentage={fileItem.progress} stroke={chartColor} fill={chartColor} animationDuration={0} />
-              )}
-              <span
-                className="flex h-6 w-6 cursor-pointer items-center justify-center"
-                onClick={(e) => {
-                  e.stopPropagation()
-                  removeFile(fileItem.fileID)
-                }}
-              >
-                <RiDeleteBinLine className="size-4 text-text-tertiary" />
-              </span>
-            </div>
-          </div>
-        ))}
-      </div>
     </div>
   )
 }

web/app/components/datasets/create/step-two/components/general-chunking-options.tsx

@@ -154,7 +154,7 @@ export const GeneralChunkingOptions: FC<GeneralChunkingOptionsProps> = ({
               </div>
             ))}
             {
-              showSummaryIndexSetting && (
+              showSummaryIndexSetting && IS_CE_EDITION && (
                 <div className="mt-3">
                   <SummaryIndexSetting
                     entry="create-document"

web/app/components/datasets/create/step-two/components/parent-child-options.tsx

@@ -12,6 +12,7 @@ import Divider from '@/app/components/base/divider'
 import { ParentChildChunk } from '@/app/components/base/icons/src/vender/knowledge'
 import RadioCard from '@/app/components/base/radio-card'
 import SummaryIndexSetting from '@/app/components/datasets/settings/summary-index-setting'
+import { IS_CE_EDITION } from '@/config'
 import { ChunkingMode } from '@/models/datasets'
 import FileList from '../../assets/file-list-3-fill.svg'
 import Note from '../../assets/note-mod.svg'
@@ -191,7 +192,7 @@ export const ParentChildOptions: FC<ParentChildOptionsProps> = ({
               </div>
             ))}
             {
-              showSummaryIndexSetting && (
+              showSummaryIndexSetting && IS_CE_EDITION && (
                 <div className="mt-3">
                   <SummaryIndexSetting
                     entry="create-document"

web/app/components/datasets/documents/components/document-list/components/document-source-icon.spec.tsx

@@ -0,0 +1,262 @@
+import type { SimpleDocumentDetail } from '@/models/datasets'
+import { render } from '@testing-library/react'
+import { describe, expect, it } from 'vitest'
+import { DataSourceType } from '@/models/datasets'
+import { DatasourceType } from '@/models/pipeline'
+import DocumentSourceIcon from './document-source-icon'
+
+const createMockDoc = (overrides: Record<string, unknown> = {}): SimpleDocumentDetail => ({
+  id: 'doc-1',
+  position: 1,
+  data_source_type: DataSourceType.FILE,
+  data_source_info: {},
+  data_source_detail_dict: {},
+  dataset_process_rule_id: 'rule-1',
+  dataset_id: 'dataset-1',
+  batch: 'batch-1',
+  name: 'test-document.txt',
+  created_from: 'web',
+  created_by: 'user-1',
+  created_at: Date.now(),
+  tokens: 100,
+  indexing_status: 'completed',
+  error: null,
+  enabled: true,
+  disabled_at: null,
+  disabled_by: null,
+  archived: false,
+  archived_reason: null,
+  archived_by: null,
+  archived_at: null,
+  updated_at: Date.now(),
+  doc_type: null,
+  doc_metadata: undefined,
+  doc_language: 'en',
+  display_status: 'available',
+  word_count: 100,
+  hit_count: 10,
+  doc_form: 'text_model',
+  ...overrides,
+}) as unknown as SimpleDocumentDetail
+
+describe('DocumentSourceIcon', () => {
+  describe('Rendering', () => {
+    it('should render without crashing', () => {
+      const doc = createMockDoc()
+      const { container } = render(<DocumentSourceIcon doc={doc} />)
+      expect(container.firstChild).toBeInTheDocument()
+    })
+  })
+
+  describe('Local File Icon', () => {
+    it('should render FileTypeIcon for FILE data source type', () => {
+      const doc = createMockDoc({
+        data_source_type: DataSourceType.FILE,
+        data_source_info: {
+          upload_file: { extension: 'pdf' },
+        },
+      })
+
+      const { container } = render(<DocumentSourceIcon doc={doc} fileType="pdf" />)
+      const icon = container.querySelector('svg, img')
+      expect(icon).toBeInTheDocument()
+    })
+
+    it('should render FileTypeIcon for localFile data source type', () => {
+      const doc = createMockDoc({
+        data_source_type: DatasourceType.localFile,
+        created_from: 'rag-pipeline',
+        data_source_info: {
+          extension: 'docx',
+        },
+      })
+
+      const { container } = render(<DocumentSourceIcon doc={doc} />)
+      const icon = container.querySelector('svg, img')
+      expect(icon).toBeInTheDocument()
+    })
+
+    it('should use extension from upload_file for legacy data source', () => {
+      const doc = createMockDoc({
+        data_source_type: DataSourceType.FILE,
+        created_from: 'web',
+        data_source_info: {
+          upload_file: { extension: 'txt' },
+        },
+      })
+
+      const { container } = render(<DocumentSourceIcon doc={doc} />)
+      expect(container.firstChild).toBeInTheDocument()
+    })
+
+    it('should use fileType prop as fallback for extension', () => {
+      const doc = createMockDoc({
+        data_source_type: DataSourceType.FILE,
+        created_from: 'web',
+        data_source_info: {},
+      })
+
+      const { container } = render(<DocumentSourceIcon doc={doc} fileType="csv" />)
+      expect(container.firstChild).toBeInTheDocument()
+    })
+  })
+
+  describe('Notion Icon', () => {
+    it('should render NotionIcon for NOTION data source type', () => {
+      const doc = createMockDoc({
+        data_source_type: DataSourceType.NOTION,
+        created_from: 'web',
+        data_source_info: {
+          notion_page_icon: 'https://notion.so/icon.png',
+        },
+      })
+
+      const { container } = render(<DocumentSourceIcon doc={doc} />)
+      expect(container.firstChild).toBeInTheDocument()
+    })
+
+    it('should render NotionIcon for onlineDocument data source type', () => {
+      const doc = createMockDoc({
+        data_source_type: DatasourceType.onlineDocument,
+        created_from: 'rag-pipeline',
+        data_source_info: {
+          page: { page_icon: 'https://notion.so/icon.png' },
+        },
+      })
+
+      const { container } = render(<DocumentSourceIcon doc={doc} />)
+      expect(container.firstChild).toBeInTheDocument()
+    })
+
+    it('should use page_icon for rag-pipeline created documents', () => {
+      const doc = createMockDoc({
+        data_source_type: DataSourceType.NOTION,
+        created_from: 'rag-pipeline',
+        data_source_info: {
+          page: { page_icon: 'https://notion.so/custom-icon.png' },
+        },
+      })
+
+      const { container } = render(<DocumentSourceIcon doc={doc} />)
+      expect(container.firstChild).toBeInTheDocument()
+    })
+  })
+
+  describe('Web Crawl Icon', () => {
+    it('should render globe icon for WEB data source type', () => {
+      const doc = createMockDoc({
+        data_source_type: DataSourceType.WEB,
+      })
+
+      const { container } = render(<DocumentSourceIcon doc={doc} />)
+      const icon = container.querySelector('svg')
+      expect(icon).toBeInTheDocument()
+      expect(icon).toHaveClass('mr-1.5')
+      expect(icon).toHaveClass('size-4')
+    })
+
+    it('should render globe icon for websiteCrawl data source type', () => {
+      const doc = createMockDoc({
+        data_source_type: DatasourceType.websiteCrawl,
+      })
+
+      const { container } = render(<DocumentSourceIcon doc={doc} />)
+      const icon = container.querySelector('svg')
+      expect(icon).toBeInTheDocument()
+    })
+  })
+
+  describe('Online Drive Icon', () => {
+    it('should render FileTypeIcon for onlineDrive data source type', () => {
+      const doc = createMockDoc({
+        data_source_type: DatasourceType.onlineDrive,
+        data_source_info: {
+          name: 'document.xlsx',
+        },
+      })
+
+      const { container } = render(<DocumentSourceIcon doc={doc} />)
+      expect(container.firstChild).toBeInTheDocument()
+    })
+
+    it('should extract extension from file name', () => {
+      const doc = createMockDoc({
+        data_source_type: DatasourceType.onlineDrive,
+        data_source_info: {
+          name: 'spreadsheet.xlsx',
+        },
+      })
+
+      const { container } = render(<DocumentSourceIcon doc={doc} />)
+      expect(container.firstChild).toBeInTheDocument()
+    })
+
+    it('should handle file name without extension', () => {
+      const doc = createMockDoc({
+        data_source_type: DatasourceType.onlineDrive,
+        data_source_info: {
+          name: 'noextension',
+        },
+      })
+
+      const { container } = render(<DocumentSourceIcon doc={doc} />)
+      expect(container.firstChild).toBeInTheDocument()
+    })
+
+    it('should handle empty file name', () => {
+      const doc = createMockDoc({
+        data_source_type: DatasourceType.onlineDrive,
+        data_source_info: {
+          name: '',
+        },
+      })
+
+      const { container } = render(<DocumentSourceIcon doc={doc} />)
+      expect(container.firstChild).toBeInTheDocument()
+    })
+
+    it('should handle hidden files (starting with dot)', () => {
+      const doc = createMockDoc({
+        data_source_type: DatasourceType.onlineDrive,
+        data_source_info: {
+          name: '.gitignore',
+        },
+      })
+
+      const { container } = render(<DocumentSourceIcon doc={doc} />)
+      expect(container.firstChild).toBeInTheDocument()
+    })
+  })
+
+  describe('Unknown Data Source Type', () => {
+    it('should return null for unknown data source type', () => {
+      const doc = createMockDoc({
+        data_source_type: 'unknown',
+      })
+
+      const { container } = render(<DocumentSourceIcon doc={doc} />)
+      expect(container.firstChild).toBeNull()
+    })
+  })
+
+  describe('Edge Cases', () => {
+    it('should handle undefined data_source_info', () => {
+      const doc = createMockDoc({
+        data_source_type: DataSourceType.FILE,
+        data_source_info: undefined,
+      })
+
+      const { container } = render(<DocumentSourceIcon doc={doc} />)
+      expect(container.firstChild).toBeInTheDocument()
+    })
+
+    it('should memoize the component', () => {
+      const doc = createMockDoc()
+      const { rerender, container } = render(<DocumentSourceIcon doc={doc} />)
+
+      const firstRender = container.innerHTML
+      rerender(<DocumentSourceIcon doc={doc} />)
+      expect(container.innerHTML).toBe(firstRender)
+    })
+  })
+})

web/app/components/datasets/documents/components/document-list/components/document-source-icon.tsx

@@ -0,0 +1,100 @@
+import type { FC } from 'react'
+import type { LegacyDataSourceInfo, LocalFileInfo, OnlineDocumentInfo, OnlineDriveInfo, SimpleDocumentDetail } from '@/models/datasets'
+import { RiGlobalLine } from '@remixicon/react'
+import * as React from 'react'
+import FileTypeIcon from '@/app/components/base/file-uploader/file-type-icon'
+import NotionIcon from '@/app/components/base/notion-icon'
+import { extensionToFileType } from '@/app/components/datasets/hit-testing/utils/extension-to-file-type'
+import { DataSourceType } from '@/models/datasets'
+import { DatasourceType } from '@/models/pipeline'
+
+type DocumentSourceIconProps = {
+  doc: SimpleDocumentDetail
+  fileType?: string
+}
+
+const isLocalFile = (dataSourceType: DataSourceType | DatasourceType) => {
+  return dataSourceType === DatasourceType.localFile || dataSourceType === DataSourceType.FILE
+}
+
+const isOnlineDocument = (dataSourceType: DataSourceType | DatasourceType) => {
+  return dataSourceType === DatasourceType.onlineDocument || dataSourceType === DataSourceType.NOTION
+}
+
+const isWebsiteCrawl = (dataSourceType: DataSourceType | DatasourceType) => {
+  return dataSourceType === DatasourceType.websiteCrawl || dataSourceType === DataSourceType.WEB
+}
+
+const isOnlineDrive = (dataSourceType: DataSourceType | DatasourceType) => {
+  return dataSourceType === DatasourceType.onlineDrive
+}
+
+const isCreateFromRAGPipeline = (createdFrom: string) => {
+  return createdFrom === 'rag-pipeline'
+}
+
+const getFileExtension = (fileName: string): string => {
+  if (!fileName)
+    return ''
+  const parts = fileName.split('.')
+  if (parts.length <= 1 || (parts[0] === '' && parts.length === 2))
+    return ''
+  return parts[parts.length - 1].toLowerCase()
+}
+
+const DocumentSourceIcon: FC<DocumentSourceIconProps> = React.memo(({
+  doc,
+  fileType,
+}) => {
+  if (isOnlineDocument(doc.data_source_type)) {
+    return (
+      <NotionIcon
+        className="mr-1.5"
+        type="page"
+        src={
+          isCreateFromRAGPipeline(doc.created_from)
+            ? (doc.data_source_info as OnlineDocumentInfo).page.page_icon
+            : (doc.data_source_info as LegacyDataSourceInfo).notion_page_icon
+        }
+      />
+    )
+  }
+
+  if (isLocalFile(doc.data_source_type)) {
+    return (
+      <FileTypeIcon
+        type={
+          extensionToFileType(
+            isCreateFromRAGPipeline(doc.created_from)
+              ? (doc?.data_source_info as LocalFileInfo)?.extension
+              : ((doc?.data_source_info as LegacyDataSourceInfo)?.upload_file?.extension ?? fileType),
+          )
+        }
+        className="mr-1.5"
+      />
+    )
+  }
+
+  if (isOnlineDrive(doc.data_source_type)) {
+    return (
+      <FileTypeIcon
+        type={
+          extensionToFileType(
+            getFileExtension((doc?.data_source_info as unknown as OnlineDriveInfo)?.name),
+          )
+        }
+        className="mr-1.5"
+      />
+    )
+  }
+
+  if (isWebsiteCrawl(doc.data_source_type)) {
+    return <RiGlobalLine className="mr-1.5 size-4" />
+  }
+
+  return null
+})
+
+DocumentSourceIcon.displayName = 'DocumentSourceIcon'
+
+export default DocumentSourceIcon

web/app/components/datasets/documents/components/document-list/components/document-table-row.spec.tsx

@@ -0,0 +1,342 @@
+import type { ReactNode } from 'react'
+import type { SimpleDocumentDetail } from '@/models/datasets'
+import { QueryClient, QueryClientProvider } from '@tanstack/react-query'
+import { fireEvent, render, screen } from '@testing-library/react'
+import { beforeEach, describe, expect, it, vi } from 'vitest'
+import { DataSourceType } from '@/models/datasets'
+import DocumentTableRow from './document-table-row'
+
+const mockPush = vi.fn()
+
+vi.mock('next/navigation', () => ({
+  useRouter: () => ({
+    push: mockPush,
+  }),
+}))
+
+const createTestQueryClient = () => new QueryClient({
+  defaultOptions: {
+    queries: { retry: false, gcTime: 0 },
+    mutations: { retry: false },
+  },
+})
+
+const createWrapper = () => {
+  const queryClient = createTestQueryClient()
+  return ({ children }: { children: ReactNode }) => (
+    <QueryClientProvider client={queryClient}>
+      <table>
+        <tbody>
+          {children}
+        </tbody>
+      </table>
+    </QueryClientProvider>
+  )
+}
+
+type LocalDoc = SimpleDocumentDetail & { percent?: number }
+
+const createMockDoc = (overrides: Record<string, unknown> = {}): LocalDoc => ({
+  id: 'doc-1',
+  position: 1,
+  data_source_type: DataSourceType.FILE,
+  data_source_info: {},
+  data_source_detail_dict: {
+    upload_file: { name: 'test.txt', extension: 'txt' },
+  },
+  dataset_process_rule_id: 'rule-1',
+  dataset_id: 'dataset-1',
+  batch: 'batch-1',
+  name: 'test-document.txt',
+  created_from: 'web',
+  created_by: 'user-1',
+  created_at: Date.now(),
+  tokens: 100,
+  indexing_status: 'completed',
+  error: null,
+  enabled: true,
+  disabled_at: null,
+  disabled_by: null,
+  archived: false,
+  archived_reason: null,
+  archived_by: null,
+  archived_at: null,
+  updated_at: Date.now(),
+  doc_type: null,
+  doc_metadata: undefined,
+  doc_language: 'en',
+  display_status: 'available',
+  word_count: 500,
+  hit_count: 10,
+  doc_form: 'text_model',
+  ...overrides,
+}) as unknown as LocalDoc
+
+// Helper to find the custom checkbox div (Checkbox component renders as a div, not a native checkbox)
+const findCheckbox = (container: HTMLElement): HTMLElement | null => {
+  return container.querySelector('[class*="shadow-xs"]')
+}
+
+describe('DocumentTableRow', () => {
+  const defaultProps = {
+    doc: createMockDoc(),
+    index: 0,
+    datasetId: 'dataset-1',
+    isSelected: false,
+    isGeneralMode: true,
+    isQAMode: false,
+    embeddingAvailable: true,
+    selectedIds: [],
+    onSelectOne: vi.fn(),
+    onSelectedIdChange: vi.fn(),
+    onShowRenameModal: vi.fn(),
+    onUpdate: vi.fn(),
+  }
+
+  beforeEach(() => {
+    vi.clearAllMocks()
+  })
+
+  describe('Rendering', () => {
+    it('should render without crashing', () => {
+      render(<DocumentTableRow {...defaultProps} />, { wrapper: createWrapper() })
+      expect(screen.getByText('test-document.txt')).toBeInTheDocument()
+    })
+
+    it('should render index number correctly', () => {
+      render(<DocumentTableRow {...defaultProps} index={5} />, { wrapper: createWrapper() })
+      expect(screen.getByText('6')).toBeInTheDocument()
+    })
+
+    it('should render document name with tooltip', () => {
+      render(<DocumentTableRow {...defaultProps} />, { wrapper: createWrapper() })
+      expect(screen.getByText('test-document.txt')).toBeInTheDocument()
+    })
+
+    it('should render checkbox element', () => {
+      const { container } = render(<DocumentTableRow {...defaultProps} />, { wrapper: createWrapper() })
+      const checkbox = findCheckbox(container)
+      expect(checkbox).toBeInTheDocument()
+    })
+  })
+
+  describe('Selection', () => {
+    it('should show check icon when isSelected is true', () => {
+      const { container } = render(<DocumentTableRow {...defaultProps} isSelected />, { wrapper: createWrapper() })
+      // When selected, the checkbox should have a check icon (RiCheckLine svg)
+      const checkbox = findCheckbox(container)
+      expect(checkbox).toBeInTheDocument()
+      const checkIcon = checkbox?.querySelector('svg')
+      expect(checkIcon).toBeInTheDocument()
+    })
+
+    it('should not show check icon when isSelected is false', () => {
+      const { container } = render(<DocumentTableRow {...defaultProps} isSelected={false} />, { wrapper: createWrapper() })
+      const checkbox = findCheckbox(container)
+      expect(checkbox).toBeInTheDocument()
+      // When not selected, there should be no check icon inside the checkbox
+      const checkIcon = checkbox?.querySelector('svg')
+      expect(checkIcon).not.toBeInTheDocument()
+    })
+
+    it('should call onSelectOne when checkbox is clicked', () => {
+      const onSelectOne = vi.fn()
+      const { container } = render(<DocumentTableRow {...defaultProps} onSelectOne={onSelectOne} />, { wrapper: createWrapper() })
+
+      const checkbox = findCheckbox(container)
+      if (checkbox) {
+        fireEvent.click(checkbox)
+        expect(onSelectOne).toHaveBeenCalledWith('doc-1')
+      }
+    })
+
+    it('should stop propagation when checkbox container is clicked', () => {
+      const { container } = render(<DocumentTableRow {...defaultProps} />, { wrapper: createWrapper() })
+
+      // Click the div containing the checkbox (which has stopPropagation)
+      const checkboxContainer = container.querySelector('td')?.querySelector('div')
+      if (checkboxContainer) {
+        fireEvent.click(checkboxContainer)
+        expect(mockPush).not.toHaveBeenCalled()
+      }
+    })
+  })
+
+  describe('Row Navigation', () => {
+    it('should navigate to document detail on row click', () => {
+      render(<DocumentTableRow {...defaultProps} />, { wrapper: createWrapper() })
+
+      const row = screen.getByRole('row')
+      fireEvent.click(row)
+
+      expect(mockPush).toHaveBeenCalledWith('/datasets/dataset-1/documents/doc-1')
+    })
+
+    it('should navigate with correct datasetId and documentId', () => {
+      render(
+        <DocumentTableRow
+          {...defaultProps}
+          datasetId="custom-dataset"
+          doc={createMockDoc({ id: 'custom-doc' })}
+        />,
+        { wrapper: createWrapper() },
+      )
+
+      const row = screen.getByRole('row')
+      fireEvent.click(row)
+
+      expect(mockPush).toHaveBeenCalledWith('/datasets/custom-dataset/documents/custom-doc')
+    })
+  })
+
+  describe('Word Count Display', () => {
+    it('should display word count less than 1000 as is', () => {
+      const doc = createMockDoc({ word_count: 500 })
+      render(<DocumentTableRow {...defaultProps} doc={doc} />, { wrapper: createWrapper() })
+      expect(screen.getByText('500')).toBeInTheDocument()
+    })
+
+    it('should display word count 1000 or more in k format', () => {
+      const doc = createMockDoc({ word_count: 1500 })
+      render(<DocumentTableRow {...defaultProps} doc={doc} />, { wrapper: createWrapper() })
+      expect(screen.getByText('1.5k')).toBeInTheDocument()
+    })
+
+    it('should display 0 with empty style when word_count is 0', () => {
+      const doc = createMockDoc({ word_count: 0 })
+      const { container } = render(<DocumentTableRow {...defaultProps} doc={doc} />, { wrapper: createWrapper() })
+      const zeroCells = container.querySelectorAll('.text-text-tertiary')
+      expect(zeroCells.length).toBeGreaterThan(0)
+    })
+
+    it('should handle undefined word_count', () => {
+      const doc = createMockDoc({ word_count: undefined as unknown as number })
+      const { container } = render(<DocumentTableRow {...defaultProps} doc={doc} />, { wrapper: createWrapper() })
+      expect(container).toBeInTheDocument()
+    })
+  })
+
+  describe('Hit Count Display', () => {
+    it('should display hit count less than 1000 as is', () => {
+      const doc = createMockDoc({ hit_count: 100 })
+      render(<DocumentTableRow {...defaultProps} doc={doc} />, { wrapper: createWrapper() })
+      expect(screen.getByText('100')).toBeInTheDocument()
+    })
+
+    it('should display hit count 1000 or more in k format', () => {
+      const doc = createMockDoc({ hit_count: 2500 })
+      render(<DocumentTableRow {...defaultProps} doc={doc} />, { wrapper: createWrapper() })
+      expect(screen.getByText('2.5k')).toBeInTheDocument()
+    })
+
+    it('should display 0 with empty style when hit_count is 0', () => {
+      const doc = createMockDoc({ hit_count: 0 })
+      const { container } = render(<DocumentTableRow {...defaultProps} doc={doc} />, { wrapper: createWrapper() })
+      const zeroCells = container.querySelectorAll('.text-text-tertiary')
+      expect(zeroCells.length).toBeGreaterThan(0)
+    })
+  })
+
+  describe('Chunking Mode', () => {
+    it('should render ChunkingModeLabel with general mode', () => {
+      render(<DocumentTableRow {...defaultProps} isGeneralMode isQAMode={false} />, { wrapper: createWrapper() })
+      // ChunkingModeLabel should be rendered
+      expect(screen.getByRole('row')).toBeInTheDocument()
+    })
+
+    it('should render ChunkingModeLabel with QA mode', () => {
+      render(<DocumentTableRow {...defaultProps} isGeneralMode={false} isQAMode />, { wrapper: createWrapper() })
+      expect(screen.getByRole('row')).toBeInTheDocument()
+    })
+  })
+
+  describe('Summary Status', () => {
+    it('should render SummaryStatus when summary_index_status is present', () => {
+      const doc = createMockDoc({ summary_index_status: 'completed' })
+      render(<DocumentTableRow {...defaultProps} doc={doc} />, { wrapper: createWrapper() })
+      expect(screen.getByRole('row')).toBeInTheDocument()
+    })
+
+    it('should not render SummaryStatus when summary_index_status is absent', () => {
+      const doc = createMockDoc({ summary_index_status: undefined })
+      render(<DocumentTableRow {...defaultProps} doc={doc} />, { wrapper: createWrapper() })
+      expect(screen.getByRole('row')).toBeInTheDocument()
+    })
+  })
+
+  describe('Rename Action', () => {
+    it('should call onShowRenameModal when rename button is clicked', () => {
+      const onShowRenameModal = vi.fn()
+      const { container } = render(
+        <DocumentTableRow {...defaultProps} onShowRenameModal={onShowRenameModal} />,
+        { wrapper: createWrapper() },
+      )
+
+      // Find the rename button by finding the RiEditLine icon's parent
+      const renameButtons = container.querySelectorAll('.cursor-pointer.rounded-md')
+      if (renameButtons.length > 0) {
+        fireEvent.click(renameButtons[0])
+        expect(onShowRenameModal).toHaveBeenCalledWith(defaultProps.doc)
+        expect(mockPush).not.toHaveBeenCalled()
+      }
+    })
+  })
+
+  describe('Operations', () => {
+    it('should pass selectedIds to Operations component', () => {
+      render(<DocumentTableRow {...defaultProps} selectedIds={['doc-1', 'doc-2']} />, { wrapper: createWrapper() })
+      expect(screen.getByRole('row')).toBeInTheDocument()
+    })
+
+    it('should pass onSelectedIdChange to Operations component', () => {
+      const onSelectedIdChange = vi.fn()
+      render(<DocumentTableRow {...defaultProps} onSelectedIdChange={onSelectedIdChange} />, { wrapper: createWrapper() })
+      expect(screen.getByRole('row')).toBeInTheDocument()
+    })
+  })
+
+  describe('Document Source Icon', () => {
+    it('should render with FILE data source type', () => {
+      const doc = createMockDoc({ data_source_type: DataSourceType.FILE })
+      render(<DocumentTableRow {...defaultProps} doc={doc} />, { wrapper: createWrapper() })
+      expect(screen.getByRole('row')).toBeInTheDocument()
+    })
+
+    it('should render with NOTION data source type', () => {
+      const doc = createMockDoc({
+        data_source_type: DataSourceType.NOTION,
+        data_source_info: { notion_page_icon: 'icon.png' },
+      })
+      render(<DocumentTableRow {...defaultProps} doc={doc} />, { wrapper: createWrapper() })
+      expect(screen.getByRole('row')).toBeInTheDocument()
+    })
+
+    it('should render with WEB data source type', () => {
+      const doc = createMockDoc({ data_source_type: DataSourceType.WEB })
+      render(<DocumentTableRow {...defaultProps} doc={doc} />, { wrapper: createWrapper() })
+      expect(screen.getByRole('row')).toBeInTheDocument()
+    })
+  })
+
+  describe('Edge Cases', () => {
+    it('should handle document with very long name', () => {
+      const doc = createMockDoc({ name: `${'a'.repeat(500)}.txt` })
+      render(<DocumentTableRow {...defaultProps} doc={doc} />, { wrapper: createWrapper() })
+      expect(screen.getByRole('row')).toBeInTheDocument()
+    })
+
+    it('should handle document with special characters in name', () => {
+      const doc = createMockDoc({ name: '<script>test</script>.txt' })
+      render(<DocumentTableRow {...defaultProps} doc={doc} />, { wrapper: createWrapper() })
+      expect(screen.getByText('<script>test</script>.txt')).toBeInTheDocument()
+    })
+
+    it('should memoize the component', () => {
+      const wrapper = createWrapper()
+      const { rerender } = render(<DocumentTableRow {...defaultProps} />, { wrapper })
+
+      rerender(<DocumentTableRow {...defaultProps} />)
+      expect(screen.getByRole('row')).toBeInTheDocument()
+    })
+  })
+})

web/app/components/datasets/documents/components/document-list/components/document-table-row.tsx

@@ -0,0 +1,152 @@
+import type { FC } from 'react'
+import type { SimpleDocumentDetail } from '@/models/datasets'
+import { RiEditLine } from '@remixicon/react'
+import { pick } from 'es-toolkit/object'
+import { useRouter } from 'next/navigation'
+import * as React from 'react'
+import { useCallback } from 'react'
+import { useTranslation } from 'react-i18next'
+import Checkbox from '@/app/components/base/checkbox'
+import Tooltip from '@/app/components/base/tooltip'
+import ChunkingModeLabel from '@/app/components/datasets/common/chunking-mode-label'
+import Operations from '@/app/components/datasets/documents/components/operations'
+import SummaryStatus from '@/app/components/datasets/documents/detail/completed/common/summary-status'
+import StatusItem from '@/app/components/datasets/documents/status-item'
+import useTimestamp from '@/hooks/use-timestamp'
+import { DataSourceType } from '@/models/datasets'
+import { formatNumber } from '@/utils/format'
+import DocumentSourceIcon from './document-source-icon'
+import { renderTdValue } from './utils'
+
+type LocalDoc = SimpleDocumentDetail & { percent?: number }
+
+type DocumentTableRowProps = {
+  doc: LocalDoc
+  index: number
+  datasetId: string
+  isSelected: boolean
+  isGeneralMode: boolean
+  isQAMode: boolean
+  embeddingAvailable: boolean
+  selectedIds: string[]
+  onSelectOne: (docId: string) => void
+  onSelectedIdChange: (ids: string[]) => void
+  onShowRenameModal: (doc: LocalDoc) => void
+  onUpdate: () => void
+}
+
+const renderCount = (count: number | undefined) => {
+  if (!count)
+    return renderTdValue(0, true)
+
+  if (count < 1000)
+    return count
+
+  return `${formatNumber((count / 1000).toFixed(1))}k`
+}
+
+const DocumentTableRow: FC<DocumentTableRowProps> = React.memo(({
+  doc,
+  index,
+  datasetId,
+  isSelected,
+  isGeneralMode,
+  isQAMode,
+  embeddingAvailable,
+  selectedIds,
+  onSelectOne,
+  onSelectedIdChange,
+  onShowRenameModal,
+  onUpdate,
+}) => {
+  const { t } = useTranslation()
+  const { formatTime } = useTimestamp()
+  const router = useRouter()
+
+  const isFile = doc.data_source_type === DataSourceType.FILE
+  const fileType = isFile ? doc.data_source_detail_dict?.upload_file?.extension : ''
+
+  const handleRowClick = useCallback(() => {
+    router.push(`/datasets/${datasetId}/documents/${doc.id}`)
+  }, [router, datasetId, doc.id])
+
+  const handleCheckboxClick = useCallback((e: React.MouseEvent) => {
+    e.stopPropagation()
+  }, [])
+
+  const handleRenameClick = useCallback((e: React.MouseEvent) => {
+    e.stopPropagation()
+    onShowRenameModal(doc)
+  }, [doc, onShowRenameModal])
+
+  return (
+    <tr
+      className="h-8 cursor-pointer border-b border-divider-subtle hover:bg-background-default-hover"
+      onClick={handleRowClick}
+    >
+      <td className="text-left align-middle text-xs text-text-tertiary">
+        <div className="flex items-center" onClick={handleCheckboxClick}>
+          <Checkbox
+            className="mr-2 shrink-0"
+            checked={isSelected}
+            onCheck={() => onSelectOne(doc.id)}
+          />
+          {index + 1}
+        </div>
+      </td>
+      <td>
+        <div className="group mr-6 flex max-w-[460px] items-center hover:mr-0">
+          <div className="flex shrink-0 items-center">
+            <DocumentSourceIcon doc={doc} fileType={fileType} />
+          </div>
+          <Tooltip popupContent={doc.name}>
+            <span className="grow-1 truncate text-sm">{doc.name}</span>
+          </Tooltip>
+          {doc.summary_index_status && (
+            <div className="ml-1 hidden shrink-0 group-hover:flex">
+              <SummaryStatus status={doc.summary_index_status} />
+            </div>
+          )}
+          <div className="hidden shrink-0 group-hover:ml-auto group-hover:flex">
+            <Tooltip popupContent={t('list.table.rename', { ns: 'datasetDocuments' })}>
+              <div
+                className="cursor-pointer rounded-md p-1 hover:bg-state-base-hover"
+                onClick={handleRenameClick}
+              >
+                <RiEditLine className="h-4 w-4 text-text-tertiary" />
+              </div>
+            </Tooltip>
+          </div>
+        </div>
+      </td>
+      <td>
+        <ChunkingModeLabel
+          isGeneralMode={isGeneralMode}
+          isQAMode={isQAMode}
+        />
+      </td>
+      <td>{renderCount(doc.word_count)}</td>
+      <td>{renderCount(doc.hit_count)}</td>
+      <td className="text-[13px] text-text-secondary">
+        {formatTime(doc.created_at, t('dateTimeFormat', { ns: 'datasetHitTesting' }) as string)}
+      </td>
+      <td>
+        <StatusItem status={doc.display_status} />
+      </td>
+      <td>
+        <Operations
+          selectedIds={selectedIds}
+          onSelectedIdChange={onSelectedIdChange}
+          embeddingAvailable={embeddingAvailable}
+          datasetId={datasetId}
+          detail={pick(doc, ['name', 'enabled', 'archived', 'id', 'data_source_type', 'doc_form', 'display_status'])}
+          onUpdate={onUpdate}
+        />
+      </td>
+    </tr>
+  )
+})
+
+DocumentTableRow.displayName = 'DocumentTableRow'
+
+export default DocumentTableRow

web/app/components/datasets/documents/components/document-list/components/index.ts

@@ -0,0 +1,4 @@
+export { default as DocumentSourceIcon } from './document-source-icon'
+export { default as DocumentTableRow } from './document-table-row'
+export { default as SortHeader } from './sort-header'
+export { renderTdValue } from './utils'

web/app/components/datasets/documents/components/document-list/components/sort-header.spec.tsx

@@ -0,0 +1,124 @@
+import { fireEvent, render, screen } from '@testing-library/react'
+import { describe, expect, it, vi } from 'vitest'
+import SortHeader from './sort-header'
+
+describe('SortHeader', () => {
+  const defaultProps = {
+    field: 'name' as const,
+    label: 'File Name',
+    currentSortField: null,
+    sortOrder: 'desc' as const,
+    onSort: vi.fn(),
+  }
+
+  describe('rendering', () => {
+    it('should render the label', () => {
+      render(<SortHeader {...defaultProps} />)
+      expect(screen.getByText('File Name')).toBeInTheDocument()
+    })
+
+    it('should render the sort icon', () => {
+      const { container } = render(<SortHeader {...defaultProps} />)
+      const icon = container.querySelector('svg')
+      expect(icon).toBeInTheDocument()
+    })
+  })
+
+  describe('inactive state', () => {
+    it('should have disabled text color when not active', () => {
+      const { container } = render(<SortHeader {...defaultProps} />)
+      const icon = container.querySelector('svg')
+      expect(icon).toHaveClass('text-text-disabled')
+    })
+
+    it('should not be rotated when not active', () => {
+      const { container } = render(<SortHeader {...defaultProps} />)
+      const icon = container.querySelector('svg')
+      expect(icon).not.toHaveClass('rotate-180')
+    })
+  })
+
+  describe('active state', () => {
+    it('should have tertiary text color when active', () => {
+      const { container } = render(
+        <SortHeader {...defaultProps} currentSortField="name" />,
+      )
+      const icon = container.querySelector('svg')
+      expect(icon).toHaveClass('text-text-tertiary')
+    })
+
+    it('should not be rotated when active and desc', () => {
+      const { container } = render(
+        <SortHeader {...defaultProps} currentSortField="name" sortOrder="desc" />,
+      )
+      const icon = container.querySelector('svg')
+      expect(icon).not.toHaveClass('rotate-180')
+    })
+
+    it('should be rotated when active and asc', () => {
+      const { container } = render(
+        <SortHeader {...defaultProps} currentSortField="name" sortOrder="asc" />,
+      )
+      const icon = container.querySelector('svg')
+      expect(icon).toHaveClass('rotate-180')
+    })
+  })
+
+  describe('interaction', () => {
+    it('should call onSort when clicked', () => {
+      const onSort = vi.fn()
+      render(<SortHeader {...defaultProps} onSort={onSort} />)
+
+      fireEvent.click(screen.getByText('File Name'))
+
+      expect(onSort).toHaveBeenCalledWith('name')
+    })
+
+    it('should call onSort with correct field', () => {
+      const onSort = vi.fn()
+      render(<SortHeader {...defaultProps} field="word_count" onSort={onSort} />)
+
+      fireEvent.click(screen.getByText('File Name'))
+
+      expect(onSort).toHaveBeenCalledWith('word_count')
+    })
+  })
+
+  describe('different fields', () => {
+    it('should work with word_count field', () => {
+      render(
+        <SortHeader
+          {...defaultProps}
+          field="word_count"
+          label="Words"
+          currentSortField="word_count"
+        />,
+      )
+      expect(screen.getByText('Words')).toBeInTheDocument()
+    })
+
+    it('should work with hit_count field', () => {
+      render(
+        <SortHeader
+          {...defaultProps}
+          field="hit_count"
+          label="Hit Count"
+          currentSortField="hit_count"
+        />,
+      )
+      expect(screen.getByText('Hit Count')).toBeInTheDocument()
+    })
+
+    it('should work with created_at field', () => {
+      render(
+        <SortHeader
+          {...defaultProps}
+          field="created_at"
+          label="Upload Time"
+          currentSortField="created_at"
+        />,
+      )
+      expect(screen.getByText('Upload Time')).toBeInTheDocument()
+    })
+  })
+})

web/app/components/datasets/documents/components/document-list/components/sort-header.tsx

@@ -0,0 +1,44 @@
+import type { FC } from 'react'
+import type { SortField, SortOrder } from '../hooks'
+import { RiArrowDownLine } from '@remixicon/react'
+import * as React from 'react'
+import { cn } from '@/utils/classnames'
+
+type SortHeaderProps = {
+  field: Exclude<SortField, null>
+  label: string
+  currentSortField: SortField
+  sortOrder: SortOrder
+  onSort: (field: SortField) => void
+}
+
+const SortHeader: FC<SortHeaderProps> = React.memo(({
+  field,
+  label,
+  currentSortField,
+  sortOrder,
+  onSort,
+}) => {
+  const isActive = currentSortField === field
+  const isDesc = isActive && sortOrder === 'desc'
+
+  return (
+    <div
+      className="flex cursor-pointer items-center hover:text-text-secondary"
+      onClick={() => onSort(field)}
+    >
+      {label}
+      <RiArrowDownLine
+        className={cn(
+          'ml-0.5 h-3 w-3 transition-all',
+          isActive ? 'text-text-tertiary' : 'text-text-disabled',
+          isActive && !isDesc ? 'rotate-180' : '',
+        )}
+      />
+    </div>
+  )
+})
+
+SortHeader.displayName = 'SortHeader'
+
+export default SortHeader

web/app/components/datasets/documents/components/document-list/components/utils.spec.tsx

@@ -0,0 +1,90 @@
+import { render, screen } from '@testing-library/react'
+import { describe, expect, it } from 'vitest'
+import { renderTdValue } from './utils'
+
+describe('renderTdValue', () => {
+  describe('Rendering', () => {
+    it('should render string value correctly', () => {
+      const { container } = render(<>{renderTdValue('test value')}</>)
+      expect(screen.getByText('test value')).toBeInTheDocument()
+      expect(container.querySelector('div')).toHaveClass('text-text-secondary')
+    })
+
+    it('should render number value correctly', () => {
+      const { container } = render(<>{renderTdValue(42)}</>)
+      expect(screen.getByText('42')).toBeInTheDocument()
+      expect(container.querySelector('div')).toHaveClass('text-text-secondary')
+    })
+
+    it('should render zero correctly', () => {
+      const { container } = render(<>{renderTdValue(0)}</>)
+      expect(screen.getByText('0')).toBeInTheDocument()
+      expect(container.querySelector('div')).toHaveClass('text-text-secondary')
+    })
+  })
+
+  describe('Null and undefined handling', () => {
+    it('should render dash for null value', () => {
+      render(<>{renderTdValue(null)}</>)
+      expect(screen.getByText('-')).toBeInTheDocument()
+    })
+
+    it('should render dash for null value with empty style', () => {
+      const { container } = render(<>{renderTdValue(null, true)}</>)
+      expect(screen.getByText('-')).toBeInTheDocument()
+      expect(container.querySelector('div')).toHaveClass('text-text-tertiary')
+    })
+  })
+
+  describe('Empty style', () => {
+    it('should apply text-text-tertiary class when isEmptyStyle is true', () => {
+      const { container } = render(<>{renderTdValue('value', true)}</>)
+      expect(container.querySelector('div')).toHaveClass('text-text-tertiary')
+    })
+
+    it('should apply text-text-secondary class when isEmptyStyle is false', () => {
+      const { container } = render(<>{renderTdValue('value', false)}</>)
+      expect(container.querySelector('div')).toHaveClass('text-text-secondary')
+    })
+
+    it('should apply text-text-secondary class when isEmptyStyle is not provided', () => {
+      const { container } = render(<>{renderTdValue('value')}</>)
+      expect(container.querySelector('div')).toHaveClass('text-text-secondary')
+    })
+  })
+
+  describe('Edge Cases', () => {
+    it('should handle empty string', () => {
+      render(<>{renderTdValue('')}</>)
+      // Empty string should still render but with no visible text
+      const div = document.querySelector('div')
+      expect(div).toBeInTheDocument()
+    })
+
+    it('should handle large numbers', () => {
+      render(<>{renderTdValue(1234567890)}</>)
+      expect(screen.getByText('1234567890')).toBeInTheDocument()
+    })
+
+    it('should handle negative numbers', () => {
+      render(<>{renderTdValue(-42)}</>)
+      expect(screen.getByText('-42')).toBeInTheDocument()
+    })
+
+    it('should handle special characters in string', () => {
+      render(<>{renderTdValue('<script>alert("xss")</script>')}</>)
+      expect(screen.getByText('<script>alert("xss")</script>')).toBeInTheDocument()
+    })
+
+    it('should handle unicode characters', () => {
+      render(<>{renderTdValue('Test Unicode: \u4E2D\u6587')}</>)
+      expect(screen.getByText('Test Unicode: \u4E2D\u6587')).toBeInTheDocument()
+    })
+
+    it('should handle very long strings', () => {
+      const longString = 'a'.repeat(1000)
+      render(<>{renderTdValue(longString)}</>)
+      expect(screen.getByText(longString)).toBeInTheDocument()
+    })
+  })
+})

web/app/components/datasets/documents/components/document-list/components/utils.tsx

@@ -0,0 +1,16 @@
+import type { ReactNode } from 'react'
+import { cn } from '@/utils/classnames'
+import s from '../../../style.module.css'
+
+export const renderTdValue = (value: string | number | null, isEmptyStyle = false): ReactNode => {
+  const className = cn(
+    isEmptyStyle ? 'text-text-tertiary' : 'text-text-secondary',
+    s.tdValue,
+  )
+
+  return (
+    <div className={className}>
+      {value ?? '-'}
+    </div>
+  )
+}

web/app/components/datasets/documents/components/document-list/hooks/index.ts

@@ -0,0 +1,4 @@
+export { useDocumentActions } from './use-document-actions'
+export { useDocumentSelection } from './use-document-selection'
+export { useDocumentSort } from './use-document-sort'
+export type { SortField, SortOrder } from './use-document-sort'

web/app/components/datasets/documents/components/document-list/hooks/use-document-actions.spec.tsx

@@ -0,0 +1,438 @@
+import type { ReactNode } from 'react'
+import { QueryClient, QueryClientProvider } from '@tanstack/react-query'
+import { act, renderHook, waitFor } from '@testing-library/react'
+import { beforeEach, describe, expect, it, vi } from 'vitest'
+import { DocumentActionType } from '@/models/datasets'
+import * as useDocument from '@/service/knowledge/use-document'
+import { useDocumentActions } from './use-document-actions'
+
+vi.mock('@/service/knowledge/use-document')
+
+const mockUseDocumentArchive = vi.mocked(useDocument.useDocumentArchive)
+const mockUseDocumentSummary = vi.mocked(useDocument.useDocumentSummary)
+const mockUseDocumentEnable = vi.mocked(useDocument.useDocumentEnable)
+const mockUseDocumentDisable = vi.mocked(useDocument.useDocumentDisable)
+const mockUseDocumentDelete = vi.mocked(useDocument.useDocumentDelete)
+const mockUseDocumentBatchRetryIndex = vi.mocked(useDocument.useDocumentBatchRetryIndex)
+const mockUseDocumentDownloadZip = vi.mocked(useDocument.useDocumentDownloadZip)
+
+const createTestQueryClient = () => new QueryClient({
+  defaultOptions: {
+    queries: { retry: false },
+    mutations: { retry: false },
+  },
+})
+
+const createWrapper = () => {
+  const queryClient = createTestQueryClient()
+  return ({ children }: { children: ReactNode }) => (
+    <QueryClientProvider client={queryClient}>
+      {children}
+    </QueryClientProvider>
+  )
+}
+
+describe('useDocumentActions', () => {
+  const mockMutateAsync = vi.fn()
+
+  beforeEach(() => {
+    vi.clearAllMocks()
+
+    // Setup all mocks with default values
+    const createMockMutation = () => ({
+      mutateAsync: mockMutateAsync,
+      isPending: false,
+      isError: false,
+      isSuccess: false,
+      isIdle: true,
+      data: undefined,
+      error: null,
+      mutate: vi.fn(),
+      reset: vi.fn(),
+      status: 'idle' as const,
+      variables: undefined,
+      context: undefined,
+      failureCount: 0,
+      failureReason: null,
+      submittedAt: 0,
+    })
+
+    mockUseDocumentArchive.mockReturnValue(createMockMutation() as unknown as ReturnType<typeof useDocument.useDocumentArchive>)
+    mockUseDocumentSummary.mockReturnValue(createMockMutation() as unknown as ReturnType<typeof useDocument.useDocumentSummary>)
+    mockUseDocumentEnable.mockReturnValue(createMockMutation() as unknown as ReturnType<typeof useDocument.useDocumentEnable>)
+    mockUseDocumentDisable.mockReturnValue(createMockMutation() as unknown as ReturnType<typeof useDocument.useDocumentDisable>)
+    mockUseDocumentDelete.mockReturnValue(createMockMutation() as unknown as ReturnType<typeof useDocument.useDocumentDelete>)
+    mockUseDocumentBatchRetryIndex.mockReturnValue(createMockMutation() as unknown as ReturnType<typeof useDocument.useDocumentBatchRetryIndex>)
+    mockUseDocumentDownloadZip.mockReturnValue({
+      ...createMockMutation(),
+      isPending: false,
+    } as unknown as ReturnType<typeof useDocument.useDocumentDownloadZip>)
+  })
+
+  describe('handleAction', () => {
+    it('should call archive mutation when archive action is triggered', async () => {
+      mockMutateAsync.mockResolvedValue({ result: 'success' })
+      const onUpdate = vi.fn()
+      const onClearSelection = vi.fn()
+
+      const { result } = renderHook(
+        () => useDocumentActions({
+          datasetId: 'ds1',
+          selectedIds: ['doc1'],
+          downloadableSelectedIds: [],
+          onUpdate,
+          onClearSelection,
+        }),
+        { wrapper: createWrapper() },
+      )
+
+      await act(async () => {
+        await result.current.handleAction(DocumentActionType.archive)()
+      })
+
+      expect(mockMutateAsync).toHaveBeenCalledWith({
+        datasetId: 'ds1',
+        documentIds: ['doc1'],
+      })
+    })
+
+    it('should call onUpdate on successful action', async () => {
+      mockMutateAsync.mockResolvedValue({ result: 'success' })
+      const onUpdate = vi.fn()
+      const onClearSelection = vi.fn()
+
+      const { result } = renderHook(
+        () => useDocumentActions({
+          datasetId: 'ds1',
+          selectedIds: ['doc1'],
+          downloadableSelectedIds: [],
+          onUpdate,
+          onClearSelection,
+        }),
+        { wrapper: createWrapper() },
+      )
+
+      await act(async () => {
+        await result.current.handleAction(DocumentActionType.enable)()
+      })
+
+      await waitFor(() => {
+        expect(onUpdate).toHaveBeenCalled()
+      })
+    })
+
+    it('should call onClearSelection on delete action', async () => {
+      mockMutateAsync.mockResolvedValue({ result: 'success' })
+      const onUpdate = vi.fn()
+      const onClearSelection = vi.fn()
+
+      const { result } = renderHook(
+        () => useDocumentActions({
+          datasetId: 'ds1',
+          selectedIds: ['doc1'],
+          downloadableSelectedIds: [],
+          onUpdate,
+          onClearSelection,
+        }),
+        { wrapper: createWrapper() },
+      )
+
+      await act(async () => {
+        await result.current.handleAction(DocumentActionType.delete)()
+      })
+
+      await waitFor(() => {
+        expect(onClearSelection).toHaveBeenCalled()
+      })
+    })
+  })
+
+  describe('handleBatchReIndex', () => {
+    it('should call retry index mutation', async () => {
+      mockMutateAsync.mockResolvedValue({ result: 'success' })
+      const onUpdate = vi.fn()
+      const onClearSelection = vi.fn()
+
+      const { result } = renderHook(
+        () => useDocumentActions({
+          datasetId: 'ds1',
+          selectedIds: ['doc1', 'doc2'],
+          downloadableSelectedIds: [],
+          onUpdate,
+          onClearSelection,
+        }),
+        { wrapper: createWrapper() },
+      )
+
+      await act(async () => {
+        await result.current.handleBatchReIndex()
+      })
+
+      expect(mockMutateAsync).toHaveBeenCalledWith({
+        datasetId: 'ds1',
+        documentIds: ['doc1', 'doc2'],
+      })
+    })
+
+    it('should call onClearSelection on success', async () => {
+      mockMutateAsync.mockResolvedValue({ result: 'success' })
+      const onUpdate = vi.fn()
+      const onClearSelection = vi.fn()
+
+      const { result } = renderHook(
+        () => useDocumentActions({
+          datasetId: 'ds1',
+          selectedIds: ['doc1'],
+          downloadableSelectedIds: [],
+          onUpdate,
+          onClearSelection,
+        }),
+        { wrapper: createWrapper() },
+      )
+
+      await act(async () => {
+        await result.current.handleBatchReIndex()
+      })
+
+      await waitFor(() => {
+        expect(onClearSelection).toHaveBeenCalled()
+        expect(onUpdate).toHaveBeenCalled()
+      })
+    })
+  })
+
+  describe('handleBatchDownload', () => {
+    it('should not proceed when already downloading', async () => {
+      mockUseDocumentDownloadZip.mockReturnValue({
+        mutateAsync: mockMutateAsync,
+        isPending: true,
+      } as unknown as ReturnType<typeof useDocument.useDocumentDownloadZip>)
+
+      const { result } = renderHook(
+        () => useDocumentActions({
+          datasetId: 'ds1',
+          selectedIds: ['doc1'],
+          downloadableSelectedIds: ['doc1'],
+          onUpdate: vi.fn(),
+          onClearSelection: vi.fn(),
+        }),
+        { wrapper: createWrapper() },
+      )
+
+      await act(async () => {
+        await result.current.handleBatchDownload()
+      })
+
+      expect(mockMutateAsync).not.toHaveBeenCalled()
+    })
+
+    it('should call download mutation with downloadable ids', async () => {
+      const mockBlob = new Blob(['test'])
+      mockMutateAsync.mockResolvedValue(mockBlob)
+
+      mockUseDocumentDownloadZip.mockReturnValue({
+        mutateAsync: mockMutateAsync,
+        isPending: false,
+      } as unknown as ReturnType<typeof useDocument.useDocumentDownloadZip>)
+
+      const { result } = renderHook(
+        () => useDocumentActions({
+          datasetId: 'ds1',
+          selectedIds: ['doc1', 'doc2'],
+          downloadableSelectedIds: ['doc1'],
+          onUpdate: vi.fn(),
+          onClearSelection: vi.fn(),
+        }),
+        { wrapper: createWrapper() },
+      )
+
+      await act(async () => {
+        await result.current.handleBatchDownload()
+      })
+
+      expect(mockMutateAsync).toHaveBeenCalledWith({
+        datasetId: 'ds1',
+        documentIds: ['doc1'],
+      })
+    })
+  })
+
+  describe('isDownloadingZip', () => {
+    it('should reflect isPending state from mutation', () => {
+      mockUseDocumentDownloadZip.mockReturnValue({
+        mutateAsync: mockMutateAsync,
+        isPending: true,
+      } as unknown as ReturnType<typeof useDocument.useDocumentDownloadZip>)
+
+      const { result } = renderHook(
+        () => useDocumentActions({
+          datasetId: 'ds1',
+          selectedIds: [],
+          downloadableSelectedIds: [],
+          onUpdate: vi.fn(),
+          onClearSelection: vi.fn(),
+        }),
+        { wrapper: createWrapper() },
+      )
+
+      expect(result.current.isDownloadingZip).toBe(true)
+    })
+  })
+
+  describe('error handling', () => {
+    it('should show error toast when handleAction fails', async () => {
+      mockMutateAsync.mockRejectedValue(new Error('Action failed'))
+      const onUpdate = vi.fn()
+      const onClearSelection = vi.fn()
+
+      const { result } = renderHook(
+        () => useDocumentActions({
+          datasetId: 'ds1',
+          selectedIds: ['doc1'],
+          downloadableSelectedIds: [],
+          onUpdate,
+          onClearSelection,
+        }),
+        { wrapper: createWrapper() },
+      )
+
+      await act(async () => {
+        await result.current.handleAction(DocumentActionType.archive)()
+      })
+
+      // onUpdate should not be called on error
+      expect(onUpdate).not.toHaveBeenCalled()
+    })
+
+    it('should show error toast when handleBatchReIndex fails', async () => {
+      mockMutateAsync.mockRejectedValue(new Error('Re-index failed'))
+      const onUpdate = vi.fn()
+      const onClearSelection = vi.fn()
+
+      const { result } = renderHook(
+        () => useDocumentActions({
+          datasetId: 'ds1',
+          selectedIds: ['doc1'],
+          downloadableSelectedIds: [],
+          onUpdate,
+          onClearSelection,
+        }),
+        { wrapper: createWrapper() },
+      )
+
+      await act(async () => {
+        await result.current.handleBatchReIndex()
+      })
+
+      // onUpdate and onClearSelection should not be called on error
+      expect(onUpdate).not.toHaveBeenCalled()
+      expect(onClearSelection).not.toHaveBeenCalled()
+    })
+
+    it('should show error toast when handleBatchDownload fails', async () => {
+      mockMutateAsync.mockRejectedValue(new Error('Download failed'))
+
+      mockUseDocumentDownloadZip.mockReturnValue({
+        mutateAsync: mockMutateAsync,
+        isPending: false,
+      } as unknown as ReturnType<typeof useDocument.useDocumentDownloadZip>)
+
+      const { result } = renderHook(
+        () => useDocumentActions({
+          datasetId: 'ds1',
+          selectedIds: ['doc1'],
+          downloadableSelectedIds: ['doc1'],
+          onUpdate: vi.fn(),
+          onClearSelection: vi.fn(),
+        }),
+        { wrapper: createWrapper() },
+      )
+
+      await act(async () => {
+        await result.current.handleBatchDownload()
+      })
+
+      // Mutation was called but failed
+      expect(mockMutateAsync).toHaveBeenCalled()
+    })
+
+    it('should show error toast when handleBatchDownload returns null blob', async () => {
+      mockMutateAsync.mockResolvedValue(null)
+
+      mockUseDocumentDownloadZip.mockReturnValue({
+        mutateAsync: mockMutateAsync,
+        isPending: false,
+      } as unknown as ReturnType<typeof useDocument.useDocumentDownloadZip>)
+
+      const { result } = renderHook(
+        () => useDocumentActions({
+          datasetId: 'ds1',
+          selectedIds: ['doc1'],
+          downloadableSelectedIds: ['doc1'],
+          onUpdate: vi.fn(),
+          onClearSelection: vi.fn(),
+        }),
+        { wrapper: createWrapper() },
+      )
+
+      await act(async () => {
+        await result.current.handleBatchDownload()
+      })
+
+      // Mutation was called but returned null
+      expect(mockMutateAsync).toHaveBeenCalled()
+    })
+  })
+
+  describe('all action types', () => {
+    it('should handle summary action', async () => {
+      mockMutateAsync.mockResolvedValue({ result: 'success' })
+      const onUpdate = vi.fn()
+
+      const { result } = renderHook(
+        () => useDocumentActions({
+          datasetId: 'ds1',
+          selectedIds: ['doc1'],
+          downloadableSelectedIds: [],
+          onUpdate,
+          onClearSelection: vi.fn(),
+        }),
+        { wrapper: createWrapper() },
+      )
+
+      await act(async () => {
+        await result.current.handleAction(DocumentActionType.summary)()
+      })
+
+      expect(mockMutateAsync).toHaveBeenCalled()
+      await waitFor(() => {
+        expect(onUpdate).toHaveBeenCalled()
+      })
+    })
+
+    it('should handle disable action', async () => {
+      mockMutateAsync.mockResolvedValue({ result: 'success' })
+      const onUpdate = vi.fn()
+
+      const { result } = renderHook(
+        () => useDocumentActions({
+          datasetId: 'ds1',
+          selectedIds: ['doc1'],
+          downloadableSelectedIds: [],
+          onUpdate,
+          onClearSelection: vi.fn(),
+        }),
+        { wrapper: createWrapper() },
+      )
+
+      await act(async () => {
+        await result.current.handleAction(DocumentActionType.disable)()
+      })
+
+      expect(mockMutateAsync).toHaveBeenCalled()
+      await waitFor(() => {
+        expect(onUpdate).toHaveBeenCalled()
+      })
+    })
+  })
+})

web/app/components/datasets/documents/components/document-list/hooks/use-document-actions.ts

@@ -0,0 +1,126 @@
+import type { CommonResponse } from '@/models/common'
+import { useCallback, useMemo } from 'react'
+import { useTranslation } from 'react-i18next'
+import Toast from '@/app/components/base/toast'
+import { DocumentActionType } from '@/models/datasets'
+import {
+  useDocumentArchive,
+  useDocumentBatchRetryIndex,
+  useDocumentDelete,
+  useDocumentDisable,
+  useDocumentDownloadZip,
+  useDocumentEnable,
+  useDocumentSummary,
+} from '@/service/knowledge/use-document'
+import { asyncRunSafe } from '@/utils'
+import { downloadBlob } from '@/utils/download'
+
+type UseDocumentActionsOptions = {
+  datasetId: string
+  selectedIds: string[]
+  downloadableSelectedIds: string[]
+  onUpdate: () => void
+  onClearSelection: () => void
+}
+
+/**
+ * Generate a random ZIP filename for bulk document downloads.
+ * We intentionally avoid leaking dataset info in the exported archive name.
+ */
+const generateDocsZipFileName = (): string => {
+  const randomPart = (typeof crypto !== 'undefined' && typeof crypto.randomUUID === 'function')
+    ? crypto.randomUUID()
+    : `${Date.now().toString(36)}${Math.random().toString(36).slice(2, 10)}`
+  return `${randomPart}-docs.zip`
+}
+
+export const useDocumentActions = ({
+  datasetId,
+  selectedIds,
+  downloadableSelectedIds,
+  onUpdate,
+  onClearSelection,
+}: UseDocumentActionsOptions) => {
+  const { t } = useTranslation()
+
+  const { mutateAsync: archiveDocument } = useDocumentArchive()
+  const { mutateAsync: generateSummary } = useDocumentSummary()
+  const { mutateAsync: enableDocument } = useDocumentEnable()
+  const { mutateAsync: disableDocument } = useDocumentDisable()
+  const { mutateAsync: deleteDocument } = useDocumentDelete()
+  const { mutateAsync: retryIndexDocument } = useDocumentBatchRetryIndex()
+  const { mutateAsync: requestDocumentsZip, isPending: isDownloadingZip } = useDocumentDownloadZip()
+
+  type SupportedActionType
+    = | typeof DocumentActionType.archive
+      | typeof DocumentActionType.summary
+      | typeof DocumentActionType.enable
+      | typeof DocumentActionType.disable
+      | typeof DocumentActionType.delete
+
+  const actionMutationMap = useMemo(() => ({
+    [DocumentActionType.archive]: archiveDocument,
+    [DocumentActionType.summary]: generateSummary,
+    [DocumentActionType.enable]: enableDocument,
+    [DocumentActionType.disable]: disableDocument,
+    [DocumentActionType.delete]: deleteDocument,
+  } as const), [archiveDocument, generateSummary, enableDocument, disableDocument, deleteDocument])
+
+  const handleAction = useCallback((actionName: SupportedActionType) => {
+    return async () => {
+      const opApi = actionMutationMap[actionName]
+      if (!opApi)
+        return
+
+      const [e] = await asyncRunSafe<CommonResponse>(
+        opApi({ datasetId, documentIds: selectedIds }),
+      )
+
+      if (!e) {
+        if (actionName === DocumentActionType.delete)
+          onClearSelection()
+        Toast.notify({ type: 'success', message: t('actionMsg.modifiedSuccessfully', { ns: 'common' }) })
+        onUpdate()
+      }
+      else {
+        Toast.notify({ type: 'error', message: t('actionMsg.modifiedUnsuccessfully', { ns: 'common' }) })
+      }
+    }
+  }, [actionMutationMap, datasetId, selectedIds, onClearSelection, onUpdate, t])
+
+  const handleBatchReIndex = useCallback(async () => {
+    const [e] = await asyncRunSafe<CommonResponse>(
+      retryIndexDocument({ datasetId, documentIds: selectedIds }),
+    )
+    if (!e) {
+      onClearSelection()
+      Toast.notify({ type: 'success', message: t('actionMsg.modifiedSuccessfully', { ns: 'common' }) })
+      onUpdate()
+    }
+    else {
+      Toast.notify({ type: 'error', message: t('actionMsg.modifiedUnsuccessfully', { ns: 'common' }) })
+    }
+  }, [retryIndexDocument, datasetId, selectedIds, onClearSelection, onUpdate, t])
+
+  const handleBatchDownload = useCallback(async () => {
+    if (isDownloadingZip)
+      return
+
+    const [e, blob] = await asyncRunSafe(
+      requestDocumentsZip({ datasetId, documentIds: downloadableSelectedIds }),
+    )
+    if (e || !blob) {
+      Toast.notify({ type: 'error', message: t('actionMsg.downloadUnsuccessfully', { ns: 'common' }) })
+      return
+    }
+
+    downloadBlob({ data: blob, fileName: generateDocsZipFileName() })
+  }, [datasetId, downloadableSelectedIds, isDownloadingZip, requestDocumentsZip, t])
+
+  return {
+    handleAction,
+    handleBatchReIndex,
+    handleBatchDownload,
+    isDownloadingZip,
+  }
+}

web/app/components/datasets/documents/components/document-list/hooks/use-document-selection.spec.ts

@@ -0,0 +1,317 @@
+import type { SimpleDocumentDetail } from '@/models/datasets'
+import { act, renderHook } from '@testing-library/react'
+import { describe, expect, it, vi } from 'vitest'
+import { DataSourceType } from '@/models/datasets'
+import { useDocumentSelection } from './use-document-selection'
+
+type LocalDoc = SimpleDocumentDetail & { percent?: number }
+
+const createMockDocument = (overrides: Partial<LocalDoc> = {}): LocalDoc => ({
+  id: 'doc1',
+  name: 'Test Document',
+  data_source_type: DataSourceType.FILE,
+  data_source_info: {},
+  data_source_detail_dict: {},
+  word_count: 100,
+  hit_count: 10,
+  created_at: 1000000,
+  position: 1,
+  doc_form: 'text_model',
+  enabled: true,
+  archived: false,
+  display_status: 'available',
+  created_from: 'api',
+  ...overrides,
+} as LocalDoc)
+
+describe('useDocumentSelection', () => {
+  describe('isAllSelected', () => {
+    it('should return false when documents is empty', () => {
+      const onSelectedIdChange = vi.fn()
+      const { result } = renderHook(() =>
+        useDocumentSelection({
+          documents: [],
+          selectedIds: [],
+          onSelectedIdChange,
+        }),
+      )
+
+      expect(result.current.isAllSelected).toBe(false)
+    })
+
+    it('should return true when all documents are selected', () => {
+      const docs = [
+        createMockDocument({ id: 'doc1' }),
+        createMockDocument({ id: 'doc2' }),
+      ]
+      const onSelectedIdChange = vi.fn()
+
+      const { result } = renderHook(() =>
+        useDocumentSelection({
+          documents: docs,
+          selectedIds: ['doc1', 'doc2'],
+          onSelectedIdChange,
+        }),
+      )
+
+      expect(result.current.isAllSelected).toBe(true)
+    })
+
+    it('should return false when not all documents are selected', () => {
+      const docs = [
+        createMockDocument({ id: 'doc1' }),
+        createMockDocument({ id: 'doc2' }),
+      ]
+      const onSelectedIdChange = vi.fn()
+
+      const { result } = renderHook(() =>
+        useDocumentSelection({
+          documents: docs,
+          selectedIds: ['doc1'],
+          onSelectedIdChange,
+        }),
+      )
+
+      expect(result.current.isAllSelected).toBe(false)
+    })
+  })
+
+  describe('isSomeSelected', () => {
+    it('should return false when no documents are selected', () => {
+      const docs = [createMockDocument({ id: 'doc1' })]
+      const onSelectedIdChange = vi.fn()
+
+      const { result } = renderHook(() =>
+        useDocumentSelection({
+          documents: docs,
+          selectedIds: [],
+          onSelectedIdChange,
+        }),
+      )
+
+      expect(result.current.isSomeSelected).toBe(false)
+    })
+
+    it('should return true when some documents are selected', () => {
+      const docs = [
+        createMockDocument({ id: 'doc1' }),
+        createMockDocument({ id: 'doc2' }),
+      ]
+      const onSelectedIdChange = vi.fn()
+
+      const { result } = renderHook(() =>
+        useDocumentSelection({
+          documents: docs,
+          selectedIds: ['doc1'],
+          onSelectedIdChange,
+        }),
+      )
+
+      expect(result.current.isSomeSelected).toBe(true)
+    })
+  })
+
+  describe('onSelectAll', () => {
+    it('should select all documents when none are selected', () => {
+      const docs = [
+        createMockDocument({ id: 'doc1' }),
+        createMockDocument({ id: 'doc2' }),
+      ]
+      const onSelectedIdChange = vi.fn()
+
+      const { result } = renderHook(() =>
+        useDocumentSelection({
+          documents: docs,
+          selectedIds: [],
+          onSelectedIdChange,
+        }),
+      )
+
+      act(() => {
+        result.current.onSelectAll()
+      })
+
+      expect(onSelectedIdChange).toHaveBeenCalledWith(['doc1', 'doc2'])
+    })
+
+    it('should deselect all when all are selected', () => {
+      const docs = [
+        createMockDocument({ id: 'doc1' }),
+        createMockDocument({ id: 'doc2' }),
+      ]
+      const onSelectedIdChange = vi.fn()
+
+      const { result } = renderHook(() =>
+        useDocumentSelection({
+          documents: docs,
+          selectedIds: ['doc1', 'doc2'],
+          onSelectedIdChange,
+        }),
+      )
+
+      act(() => {
+        result.current.onSelectAll()
+      })
+
+      expect(onSelectedIdChange).toHaveBeenCalledWith([])
+    })
+
+    it('should add to existing selection when some are selected', () => {
+      const docs = [
+        createMockDocument({ id: 'doc1' }),
+        createMockDocument({ id: 'doc2' }),
+        createMockDocument({ id: 'doc3' }),
+      ]
+      const onSelectedIdChange = vi.fn()
+
+      const { result } = renderHook(() =>
+        useDocumentSelection({
+          documents: docs,
+          selectedIds: ['doc1'],
+          onSelectedIdChange,
+        }),
+      )
+
+      act(() => {
+        result.current.onSelectAll()
+      })
+
+      expect(onSelectedIdChange).toHaveBeenCalledWith(['doc1', 'doc2', 'doc3'])
+    })
+  })
+
+  describe('onSelectOne', () => {
+    it('should add document to selection when not selected', () => {
+      const onSelectedIdChange = vi.fn()
+
+      const { result } = renderHook(() =>
+        useDocumentSelection({
+          documents: [],
+          selectedIds: [],
+          onSelectedIdChange,
+        }),
+      )
+
+      act(() => {
+        result.current.onSelectOne('doc1')
+      })
+
+      expect(onSelectedIdChange).toHaveBeenCalledWith(['doc1'])
+    })
+
+    it('should remove document from selection when already selected', () => {
+      const onSelectedIdChange = vi.fn()
+
+      const { result } = renderHook(() =>
+        useDocumentSelection({
+          documents: [],
+          selectedIds: ['doc1', 'doc2'],
+          onSelectedIdChange,
+        }),
+      )
+
+      act(() => {
+        result.current.onSelectOne('doc1')
+      })
+
+      expect(onSelectedIdChange).toHaveBeenCalledWith(['doc2'])
+    })
+  })
+
+  describe('hasErrorDocumentsSelected', () => {
+    it('should return false when no error documents are selected', () => {
+      const docs = [
+        createMockDocument({ id: 'doc1', display_status: 'available' }),
+        createMockDocument({ id: 'doc2', display_status: 'error' }),
+      ]
+      const onSelectedIdChange = vi.fn()
+
+      const { result } = renderHook(() =>
+        useDocumentSelection({
+          documents: docs,
+          selectedIds: ['doc1'],
+          onSelectedIdChange,
+        }),
+      )
+
+      expect(result.current.hasErrorDocumentsSelected).toBe(false)
+    })
+
+    it('should return true when an error document is selected', () => {
+      const docs = [
+        createMockDocument({ id: 'doc1', display_status: 'available' }),
+        createMockDocument({ id: 'doc2', display_status: 'error' }),
+      ]
+      const onSelectedIdChange = vi.fn()
+
+      const { result } = renderHook(() =>
+        useDocumentSelection({
+          documents: docs,
+          selectedIds: ['doc2'],
+          onSelectedIdChange,
+        }),
+      )
+
+      expect(result.current.hasErrorDocumentsSelected).toBe(true)
+    })
+  })
+
+  describe('downloadableSelectedIds', () => {
+    it('should return only FILE type documents from selection', () => {
+      const docs = [
+        createMockDocument({ id: 'doc1', data_source_type: DataSourceType.FILE }),
+        createMockDocument({ id: 'doc2', data_source_type: DataSourceType.NOTION }),
+        createMockDocument({ id: 'doc3', data_source_type: DataSourceType.FILE }),
+      ]
+      const onSelectedIdChange = vi.fn()
+
+      const { result } = renderHook(() =>
+        useDocumentSelection({
+          documents: docs,
+          selectedIds: ['doc1', 'doc2', 'doc3'],
+          onSelectedIdChange,
+        }),
+      )
+
+      expect(result.current.downloadableSelectedIds).toEqual(['doc1', 'doc3'])
+    })
+
+    it('should return empty array when no FILE documents selected', () => {
+      const docs = [
+        createMockDocument({ id: 'doc1', data_source_type: DataSourceType.NOTION }),
+        createMockDocument({ id: 'doc2', data_source_type: DataSourceType.WEB }),
+      ]
+      const onSelectedIdChange = vi.fn()
+
+      const { result } = renderHook(() =>
+        useDocumentSelection({
+          documents: docs,
+          selectedIds: ['doc1', 'doc2'],
+          onSelectedIdChange,
+        }),
+      )
+
+      expect(result.current.downloadableSelectedIds).toEqual([])
+    })
+  })
+
+  describe('clearSelection', () => {
+    it('should call onSelectedIdChange with empty array', () => {
+      const onSelectedIdChange = vi.fn()
+
+      const { result } = renderHook(() =>
+        useDocumentSelection({
+          documents: [],
+          selectedIds: ['doc1', 'doc2'],
+          onSelectedIdChange,
+        }),
+      )
+
+      act(() => {
+        result.current.clearSelection()
+      })
+
+      expect(onSelectedIdChange).toHaveBeenCalledWith([])
+    })
+  })
+})

web/app/components/datasets/documents/components/document-list/hooks/use-document-selection.ts

@@ -0,0 +1,66 @@
+import type { SimpleDocumentDetail } from '@/models/datasets'
+import { uniq } from 'es-toolkit/array'
+import { useCallback, useMemo } from 'react'
+import { DataSourceType } from '@/models/datasets'
+
+type LocalDoc = SimpleDocumentDetail & { percent?: number }
+
+type UseDocumentSelectionOptions = {
+  documents: LocalDoc[]
+  selectedIds: string[]
+  onSelectedIdChange: (selectedIds: string[]) => void
+}
+
+export const useDocumentSelection = ({
+  documents,
+  selectedIds,
+  onSelectedIdChange,
+}: UseDocumentSelectionOptions) => {
+  const isAllSelected = useMemo(() => {
+    return documents.length > 0 && documents.every(doc => selectedIds.includes(doc.id))
+  }, [documents, selectedIds])
+
+  const isSomeSelected = useMemo(() => {
+    return documents.some(doc => selectedIds.includes(doc.id))
+  }, [documents, selectedIds])
+
+  const onSelectAll = useCallback(() => {
+    if (isAllSelected)
+      onSelectedIdChange([])
+    else
+      onSelectedIdChange(uniq([...selectedIds, ...documents.map(doc => doc.id)]))
+  }, [isAllSelected, documents, onSelectedIdChange, selectedIds])
+
+  const onSelectOne = useCallback((docId: string) => {
+    onSelectedIdChange(
+      selectedIds.includes(docId)
+        ? selectedIds.filter(id => id !== docId)
+        : [...selectedIds, docId],
+    )
+  }, [selectedIds, onSelectedIdChange])
+
+  const hasErrorDocumentsSelected = useMemo(() => {
+    return documents.some(doc => selectedIds.includes(doc.id) && doc.display_status === 'error')
+  }, [documents, selectedIds])
+
+  const downloadableSelectedIds = useMemo(() => {
+    const selectedSet = new Set(selectedIds)
+    return documents
+      .filter(doc => selectedSet.has(doc.id) && doc.data_source_type === DataSourceType.FILE)
+      .map(doc => doc.id)
+  }, [documents, selectedIds])
+
+  const clearSelection = useCallback(() => {
+    onSelectedIdChange([])
+  }, [onSelectedIdChange])
+
+  return {
+    isAllSelected,
+    isSomeSelected,
+    onSelectAll,
+    onSelectOne,
+    hasErrorDocumentsSelected,
+    downloadableSelectedIds,
+    clearSelection,
+  }
+}

web/app/components/datasets/documents/components/document-list/hooks/use-document-sort.spec.ts

@@ -0,0 +1,340 @@
+import type { SimpleDocumentDetail } from '@/models/datasets'
+import { act, renderHook } from '@testing-library/react'
+import { describe, expect, it } from 'vitest'
+import { useDocumentSort } from './use-document-sort'
+
+type LocalDoc = SimpleDocumentDetail & { percent?: number }
+
+const createMockDocument = (overrides: Partial<LocalDoc> = {}): LocalDoc => ({
+  id: 'doc1',
+  name: 'Test Document',
+  data_source_type: 'upload_file',
+  data_source_info: {},
+  data_source_detail_dict: {},
+  word_count: 100,
+  hit_count: 10,
+  created_at: 1000000,
+  position: 1,
+  doc_form: 'text_model',
+  enabled: true,
+  archived: false,
+  display_status: 'available',
+  created_from: 'api',
+  ...overrides,
+} as LocalDoc)
+
+describe('useDocumentSort', () => {
+  describe('initial state', () => {
+    it('should return null sortField initially', () => {
+      const { result } = renderHook(() =>
+        useDocumentSort({
+          documents: [],
+          statusFilterValue: '',
+          remoteSortValue: '',
+        }),
+      )
+
+      expect(result.current.sortField).toBeNull()
+      expect(result.current.sortOrder).toBe('desc')
+    })
+
+    it('should return documents unchanged when no sort is applied', () => {
+      const docs = [
+        createMockDocument({ id: 'doc1', name: 'B' }),
+        createMockDocument({ id: 'doc2', name: 'A' }),
+      ]
+
+      const { result } = renderHook(() =>
+        useDocumentSort({
+          documents: docs,
+          statusFilterValue: '',
+          remoteSortValue: '',
+        }),
+      )
+
+      expect(result.current.sortedDocuments).toEqual(docs)
+    })
+  })
+
+  describe('handleSort', () => {
+    it('should set sort field when called', () => {
+      const { result } = renderHook(() =>
+        useDocumentSort({
+          documents: [],
+          statusFilterValue: '',
+          remoteSortValue: '',
+        }),
+      )
+
+      act(() => {
+        result.current.handleSort('name')
+      })
+
+      expect(result.current.sortField).toBe('name')
+      expect(result.current.sortOrder).toBe('desc')
+    })
+
+    it('should toggle sort order when same field is clicked twice', () => {
+      const { result } = renderHook(() =>
+        useDocumentSort({
+          documents: [],
+          statusFilterValue: '',
+          remoteSortValue: '',
+        }),
+      )
+
+      act(() => {
+        result.current.handleSort('name')
+      })
+      expect(result.current.sortOrder).toBe('desc')
+
+      act(() => {
+        result.current.handleSort('name')
+      })
+      expect(result.current.sortOrder).toBe('asc')
+
+      act(() => {
+        result.current.handleSort('name')
+      })
+      expect(result.current.sortOrder).toBe('desc')
+    })
+
+    it('should reset to desc when different field is selected', () => {
+      const { result } = renderHook(() =>
+        useDocumentSort({
+          documents: [],
+          statusFilterValue: '',
+          remoteSortValue: '',
+        }),
+      )
+
+      act(() => {
+        result.current.handleSort('name')
+      })
+      act(() => {
+        result.current.handleSort('name')
+      })
+      expect(result.current.sortOrder).toBe('asc')
+
+      act(() => {
+        result.current.handleSort('word_count')
+      })
+      expect(result.current.sortField).toBe('word_count')
+      expect(result.current.sortOrder).toBe('desc')
+    })
+
+    it('should not change state when null is passed', () => {
+      const { result } = renderHook(() =>
+        useDocumentSort({
+          documents: [],
+          statusFilterValue: '',
+          remoteSortValue: '',
+        }),
+      )
+
+      act(() => {
+        result.current.handleSort(null)
+      })
+
+      expect(result.current.sortField).toBeNull()
+    })
+  })
+
+  describe('sorting documents', () => {
+    const docs = [
+      createMockDocument({ id: 'doc1', name: 'Banana', word_count: 200, hit_count: 5, created_at: 3000 }),
+      createMockDocument({ id: 'doc2', name: 'Apple', word_count: 100, hit_count: 10, created_at: 1000 }),
+      createMockDocument({ id: 'doc3', name: 'Cherry', word_count: 300, hit_count: 1, created_at: 2000 }),
+    ]
+
+    it('should sort by name descending', () => {
+      const { result } = renderHook(() =>
+        useDocumentSort({
+          documents: docs,
+          statusFilterValue: '',
+          remoteSortValue: '',
+        }),
+      )
+
+      act(() => {
+        result.current.handleSort('name')
+      })
+
+      const names = result.current.sortedDocuments.map(d => d.name)
+      expect(names).toEqual(['Cherry', 'Banana', 'Apple'])
+    })
+
+    it('should sort by name ascending', () => {
+      const { result } = renderHook(() =>
+        useDocumentSort({
+          documents: docs,
+          statusFilterValue: '',
+          remoteSortValue: '',
+        }),
+      )
+
+      act(() => {
+        result.current.handleSort('name')
+      })
+      act(() => {
+        result.current.handleSort('name')
+      })
+
+      const names = result.current.sortedDocuments.map(d => d.name)
+      expect(names).toEqual(['Apple', 'Banana', 'Cherry'])
+    })
+
+    it('should sort by word_count descending', () => {
+      const { result } = renderHook(() =>
+        useDocumentSort({
+          documents: docs,
+          statusFilterValue: '',
+          remoteSortValue: '',
+        }),
+      )
+
+      act(() => {
+        result.current.handleSort('word_count')
+      })
+
+      const counts = result.current.sortedDocuments.map(d => d.word_count)
+      expect(counts).toEqual([300, 200, 100])
+    })
+
+    it('should sort by hit_count ascending', () => {
+      const { result } = renderHook(() =>
+        useDocumentSort({
+          documents: docs,
+          statusFilterValue: '',
+          remoteSortValue: '',
+        }),
+      )
+
+      act(() => {
+        result.current.handleSort('hit_count')
+      })
+      act(() => {
+        result.current.handleSort('hit_count')
+      })
+
+      const counts = result.current.sortedDocuments.map(d => d.hit_count)
+      expect(counts).toEqual([1, 5, 10])
+    })
+
+    it('should sort by created_at descending', () => {
+      const { result } = renderHook(() =>
+        useDocumentSort({
+          documents: docs,
+          statusFilterValue: '',
+          remoteSortValue: '',
+        }),
+      )
+
+      act(() => {
+        result.current.handleSort('created_at')
+      })
+
+      const times = result.current.sortedDocuments.map(d => d.created_at)
+      expect(times).toEqual([3000, 2000, 1000])
+    })
+  })
+
+  describe('status filtering', () => {
+    const docs = [
+      createMockDocument({ id: 'doc1', display_status: 'available' }),
+      createMockDocument({ id: 'doc2', display_status: 'error' }),
+      createMockDocument({ id: 'doc3', display_status: 'available' }),
+    ]
+
+    it('should not filter when statusFilterValue is empty', () => {
+      const { result } = renderHook(() =>
+        useDocumentSort({
+          documents: docs,
+          statusFilterValue: '',
+          remoteSortValue: '',
+        }),
+      )
+
+      expect(result.current.sortedDocuments.length).toBe(3)
+    })
+
+    it('should not filter when statusFilterValue is all', () => {
+      const { result } = renderHook(() =>
+        useDocumentSort({
+          documents: docs,
+          statusFilterValue: 'all',
+          remoteSortValue: '',
+        }),
+      )
+
+      expect(result.current.sortedDocuments.length).toBe(3)
+    })
+  })
+
+  describe('remoteSortValue reset', () => {
+    it('should reset sort state when remoteSortValue changes', () => {
+      const { result, rerender } = renderHook(
+        ({ remoteSortValue }) =>
+          useDocumentSort({
+            documents: [],
+            statusFilterValue: '',
+            remoteSortValue,
+          }),
+        { initialProps: { remoteSortValue: 'initial' } },
+      )
+
+      act(() => {
+        result.current.handleSort('name')
+      })
+      act(() => {
+        result.current.handleSort('name')
+      })
+      expect(result.current.sortField).toBe('name')
+      expect(result.current.sortOrder).toBe('asc')
+
+      rerender({ remoteSortValue: 'changed' })
+
+      expect(result.current.sortField).toBeNull()
+      expect(result.current.sortOrder).toBe('desc')
+    })
+  })
+
+  describe('edge cases', () => {
+    it('should handle documents with missing values', () => {
+      const docs = [
+        createMockDocument({ id: 'doc1', name: undefined as unknown as string, word_count: undefined }),
+        createMockDocument({ id: 'doc2', name: 'Test', word_count: 100 }),
+      ]
+
+      const { result } = renderHook(() =>
+        useDocumentSort({
+          documents: docs,
+          statusFilterValue: '',
+          remoteSortValue: '',
+        }),
+      )
+
+      act(() => {
+        result.current.handleSort('name')
+      })
+
+      expect(result.current.sortedDocuments.length).toBe(2)
+    })
+
+    it('should handle empty documents array', () => {
+      const { result } = renderHook(() =>
+        useDocumentSort({
+          documents: [],
+          statusFilterValue: '',
+          remoteSortValue: '',
+        }),
+      )
+
+      act(() => {
+        result.current.handleSort('name')
+      })
+
+      expect(result.current.sortedDocuments).toEqual([])
+    })
+  })
+})

web/app/components/datasets/documents/components/document-list/hooks/use-document-sort.ts

@@ -0,0 +1,102 @@
+import type { SimpleDocumentDetail } from '@/models/datasets'
+import { useCallback, useMemo, useRef, useState } from 'react'
+import { normalizeStatusForQuery } from '@/app/components/datasets/documents/status-filter'
+
+export type SortField = 'name' | 'word_count' | 'hit_count' | 'created_at' | null
+export type SortOrder = 'asc' | 'desc'
+
+type LocalDoc = SimpleDocumentDetail & { percent?: number }
+
+type UseDocumentSortOptions = {
+  documents: LocalDoc[]
+  statusFilterValue: string
+  remoteSortValue: string
+}
+
+export const useDocumentSort = ({
+  documents,
+  statusFilterValue,
+  remoteSortValue,
+}: UseDocumentSortOptions) => {
+  const [sortField, setSortField] = useState<SortField>(null)
+  const [sortOrder, setSortOrder] = useState<SortOrder>('desc')
+  const prevRemoteSortValueRef = useRef(remoteSortValue)
+
+  // Reset sort when remote sort changes
+  if (prevRemoteSortValueRef.current !== remoteSortValue) {
+    prevRemoteSortValueRef.current = remoteSortValue
+    setSortField(null)
+    setSortOrder('desc')
+  }
+
+  const handleSort = useCallback((field: SortField) => {
+    if (field === null)
+      return
+
+    if (sortField === field) {
+      setSortOrder(prev => prev === 'asc' ? 'desc' : 'asc')
+    }
+    else {
+      setSortField(field)
+      setSortOrder('desc')
+    }
+  }, [sortField])
+
+  const sortedDocuments = useMemo(() => {
+    let filteredDocs = documents
+
+    if (statusFilterValue && statusFilterValue !== 'all') {
+      filteredDocs = filteredDocs.filter(doc =>
+        typeof doc.display_status === 'string'
+        && normalizeStatusForQuery(doc.display_status) === statusFilterValue,
+      )
+    }
+
+    if (!sortField)
+      return filteredDocs
+
+    const sortedDocs = [...filteredDocs].sort((a, b) => {
+      let aValue: string | number
+      let bValue: string | number
+
+      switch (sortField) {
+        case 'name':
+          aValue = a.name?.toLowerCase() || ''
+          bValue = b.name?.toLowerCase() || ''
+          break
+        case 'word_count':
+          aValue = a.word_count || 0
+          bValue = b.word_count || 0
+          break
+        case 'hit_count':
+          aValue = a.hit_count || 0
+          bValue = b.hit_count || 0
+          break
+        case 'created_at':
+          aValue = a.created_at
+          bValue = b.created_at
+          break
+        default:
+          return 0
+      }
+
+      if (sortField === 'name') {
+        const result = (aValue as string).localeCompare(bValue as string)
+        return sortOrder === 'asc' ? result : -result
+      }
+      else {
+        const result = (aValue as number) - (bValue as number)
+        return sortOrder === 'asc' ? result : -result
+      }
+    })
+
+    return sortedDocs
+  }, [documents, sortField, sortOrder, statusFilterValue])
+
+  return {
+    sortField,
+    sortOrder,
+    handleSort,
+    sortedDocuments,
+  }
+}

web/app/components/datasets/documents/components/document-list/index.spec.tsx

@@ -0,0 +1,487 @@
+import type { ReactNode } from 'react'
+import type { Props as PaginationProps } from '@/app/components/base/pagination'
+import type { SimpleDocumentDetail } from '@/models/datasets'
+import { QueryClient, QueryClientProvider } from '@tanstack/react-query'
+import { fireEvent, render, screen } from '@testing-library/react'
+import { beforeEach, describe, expect, it, vi } from 'vitest'
+import { ChunkingMode, DataSourceType } from '@/models/datasets'
+import DocumentList from '../list'
+
+const mockPush = vi.fn()
+
+vi.mock('next/navigation', () => ({
+  useRouter: () => ({
+    push: mockPush,
+  }),
+}))
+
+vi.mock('@/context/dataset-detail', () => ({
+  useDatasetDetailContextWithSelector: (selector: (state: { dataset: { doc_form: string } }) => unknown) =>
+    selector({ dataset: { doc_form: ChunkingMode.text } }),
+}))
+
+const createTestQueryClient = () => new QueryClient({
+  defaultOptions: {
+    queries: { retry: false, gcTime: 0 },
+    mutations: { retry: false },
+  },
+})
+
+const createWrapper = () => {
+  const queryClient = createTestQueryClient()
+  return ({ children }: { children: ReactNode }) => (
+    <QueryClientProvider client={queryClient}>
+      {children}
+    </QueryClientProvider>
+  )
+}
+
+const createMockDoc = (overrides: Partial<SimpleDocumentDetail> = {}): SimpleDocumentDetail => ({
+  id: `doc-${Math.random().toString(36).substr(2, 9)}`,
+  position: 1,
+  data_source_type: DataSourceType.FILE,
+  data_source_info: {},
+  data_source_detail_dict: {
+    upload_file: { name: 'test.txt', extension: 'txt' },
+  },
+  dataset_process_rule_id: 'rule-1',
+  batch: 'batch-1',
+  name: 'test-document.txt',
+  created_from: 'web',
+  created_by: 'user-1',
+  created_at: Date.now(),
+  tokens: 100,
+  indexing_status: 'completed',
+  error: null,
+  enabled: true,
+  disabled_at: null,
+  disabled_by: null,
+  archived: false,
+  archived_reason: null,
+  archived_by: null,
+  archived_at: null,
+  updated_at: Date.now(),
+  doc_type: null,
+  doc_metadata: undefined,
+  display_status: 'available',
+  word_count: 500,
+  hit_count: 10,
+  doc_form: 'text_model',
+  ...overrides,
+} as SimpleDocumentDetail)
+
+const defaultPagination: PaginationProps = {
+  current: 1,
+  onChange: vi.fn(),
+  total: 100,
+}
+
+describe('DocumentList', () => {
+  const defaultProps = {
+    embeddingAvailable: true,
+    documents: [
+      createMockDoc({ id: 'doc-1', name: 'Document 1.txt', word_count: 100, hit_count: 5 }),
+      createMockDoc({ id: 'doc-2', name: 'Document 2.txt', word_count: 200, hit_count: 10 }),
+      createMockDoc({ id: 'doc-3', name: 'Document 3.txt', word_count: 300, hit_count: 15 }),
+    ],
+    selectedIds: [] as string[],
+    onSelectedIdChange: vi.fn(),
+    datasetId: 'dataset-1',
+    pagination: defaultPagination,
+    onUpdate: vi.fn(),
+    onManageMetadata: vi.fn(),
+    statusFilterValue: '',
+    remoteSortValue: '',
+  }
+
+  beforeEach(() => {
+    vi.clearAllMocks()
+  })
+
+  describe('Rendering', () => {
+    it('should render without crashing', () => {
+      render(<DocumentList {...defaultProps} />, { wrapper: createWrapper() })
+      expect(screen.getByRole('table')).toBeInTheDocument()
+    })
+
+    it('should render all documents', () => {
+      render(<DocumentList {...defaultProps} />, { wrapper: createWrapper() })
+      expect(screen.getByText('Document 1.txt')).toBeInTheDocument()
+      expect(screen.getByText('Document 2.txt')).toBeInTheDocument()
+      expect(screen.getByText('Document 3.txt')).toBeInTheDocument()
+    })
+
+    it('should render table headers', () => {
+      render(<DocumentList {...defaultProps} />, { wrapper: createWrapper() })
+      expect(screen.getByText('#')).toBeInTheDocument()
+    })
+
+    it('should render pagination when total is provided', () => {
+      render(<DocumentList {...defaultProps} />, { wrapper: createWrapper() })
+      // Pagination component should be present
+      expect(screen.getByRole('table')).toBeInTheDocument()
+    })
+
+    it('should not render pagination when total is 0', () => {
+      const props = {
+        ...defaultProps,
+        pagination: { ...defaultPagination, total: 0 },
+      }
+      render(<DocumentList {...props} />, { wrapper: createWrapper() })
+      expect(screen.getByRole('table')).toBeInTheDocument()
+    })
+
+    it('should render empty table when no documents', () => {
+      const props = { ...defaultProps, documents: [] }
+      render(<DocumentList {...props} />, { wrapper: createWrapper() })
+      expect(screen.getByRole('table')).toBeInTheDocument()
+    })
+  })
+
+  describe('Selection', () => {
+    // Helper to find checkboxes (custom div components, not native checkboxes)
+    const findCheckboxes = (container: HTMLElement): NodeListOf<Element> => {
+      return container.querySelectorAll('[class*="shadow-xs"]')
+    }
+
+    it('should render header checkbox when embeddingAvailable', () => {
+      const { container } = render(<DocumentList {...defaultProps} />, { wrapper: createWrapper() })
+      const checkboxes = findCheckboxes(container)
+      expect(checkboxes.length).toBeGreaterThan(0)
+    })
+
+    it('should not render header checkbox when embedding not available', () => {
+      const props = { ...defaultProps, embeddingAvailable: false }
+      render(<DocumentList {...props} />, { wrapper: createWrapper() })
+      // Row checkboxes should still be there, but header checkbox should be hidden
+      expect(screen.getByRole('table')).toBeInTheDocument()
+    })
+
+    it('should call onSelectedIdChange when select all is clicked', () => {
+      const onSelectedIdChange = vi.fn()
+      const props = { ...defaultProps, onSelectedIdChange }
+      const { container } = render(<DocumentList {...props} />, { wrapper: createWrapper() })
+
+      const checkboxes = findCheckboxes(container)
+      if (checkboxes.length > 0) {
+        fireEvent.click(checkboxes[0])
+        expect(onSelectedIdChange).toHaveBeenCalled()
+      }
+    })
+
+    it('should show all checkboxes as checked when all are selected', () => {
+      const props = {
+        ...defaultProps,
+        selectedIds: ['doc-1', 'doc-2', 'doc-3'],
+      }
+      const { container } = render(<DocumentList {...props} />, { wrapper: createWrapper() })
+
+      const checkboxes = findCheckboxes(container)
+      // When checked, checkbox should have a check icon (svg) inside
+      checkboxes.forEach((checkbox) => {
+        const checkIcon = checkbox.querySelector('svg')
+        expect(checkIcon).toBeInTheDocument()
+      })
+    })
+
+    it('should show indeterminate state when some are selected', () => {
+      const props = {
+        ...defaultProps,
+        selectedIds: ['doc-1'],
+      }
+      const { container } = render(<DocumentList {...props} />, { wrapper: createWrapper() })
+
+      // First checkbox is the header checkbox which should be indeterminate
+      const checkboxes = findCheckboxes(container)
+      expect(checkboxes.length).toBeGreaterThan(0)
+      // Header checkbox should show indeterminate icon, not check icon
+      // Just verify it's rendered
+      expect(checkboxes[0]).toBeInTheDocument()
+    })
+
+    it('should call onSelectedIdChange with single document when row checkbox is clicked', () => {
+      const onSelectedIdChange = vi.fn()
+      const props = { ...defaultProps, onSelectedIdChange }
+      const { container } = render(<DocumentList {...props} />, { wrapper: createWrapper() })
+
+      // Click the second checkbox (first row checkbox)
+      const checkboxes = findCheckboxes(container)
+      if (checkboxes.length > 1) {
+        fireEvent.click(checkboxes[1])
+        expect(onSelectedIdChange).toHaveBeenCalled()
+      }
+    })
+  })
+
+  describe('Sorting', () => {
+    it('should render sort headers for sortable columns', () => {
+      render(<DocumentList {...defaultProps} />, { wrapper: createWrapper() })
+      // Find svg icons which indicate sortable columns
+      const sortIcons = document.querySelectorAll('svg')
+      expect(sortIcons.length).toBeGreaterThan(0)
+    })
+
+    it('should update sort order when sort header is clicked', () => {
+      render(<DocumentList {...defaultProps} />, { wrapper: createWrapper() })
+
+      // Find and click a sort header by its parent div containing the label text
+      const sortableHeaders = document.querySelectorAll('[class*="cursor-pointer"]')
+      if (sortableHeaders.length > 0) {
+        fireEvent.click(sortableHeaders[0])
+      }
+
+      expect(screen.getByRole('table')).toBeInTheDocument()
+    })
+  })
+
+  describe('Batch Actions', () => {
+    it('should show batch action bar when documents are selected', () => {
+      const props = {
+        ...defaultProps,
+        selectedIds: ['doc-1', 'doc-2'],
+      }
+      render(<DocumentList {...props} />, { wrapper: createWrapper() })
+
+      // BatchAction component should be visible
+      expect(screen.getByRole('table')).toBeInTheDocument()
+    })
+
+    it('should not show batch action bar when no documents selected', () => {
+      render(<DocumentList {...defaultProps} />, { wrapper: createWrapper() })
+
+      // BatchAction should not be present
+      expect(screen.getByRole('table')).toBeInTheDocument()
+    })
+
+    it('should render batch action bar with archive option', () => {
+      const props = {
+        ...defaultProps,
+        selectedIds: ['doc-1'],
+      }
+      render(<DocumentList {...props} />, { wrapper: createWrapper() })
+
+      // BatchAction component should be visible when documents are selected
+      expect(screen.getByRole('table')).toBeInTheDocument()
+    })
+
+    it('should render batch action bar with enable option', () => {
+      const props = {
+        ...defaultProps,
+        selectedIds: ['doc-1'],
+      }
+      render(<DocumentList {...props} />, { wrapper: createWrapper() })
+
+      expect(screen.getByRole('table')).toBeInTheDocument()
+    })
+
+    it('should render batch action bar with disable option', () => {
+      const props = {
+        ...defaultProps,
+        selectedIds: ['doc-1'],
+      }
+      render(<DocumentList {...props} />, { wrapper: createWrapper() })
+
+      expect(screen.getByRole('table')).toBeInTheDocument()
+    })
+
+    it('should render batch action bar with delete option', () => {
+      const props = {
+        ...defaultProps,
+        selectedIds: ['doc-1'],
+      }
+      render(<DocumentList {...props} />, { wrapper: createWrapper() })
+
+      expect(screen.getByRole('table')).toBeInTheDocument()
+    })
+
+    it('should clear selection when cancel is clicked', () => {
+      const onSelectedIdChange = vi.fn()
+      const props = {
+        ...defaultProps,
+        selectedIds: ['doc-1'],
+        onSelectedIdChange,
+      }
+      render(<DocumentList {...props} />, { wrapper: createWrapper() })
+
+      const cancelButton = screen.queryByRole('button', { name: /cancel/i })
+      if (cancelButton) {
+        fireEvent.click(cancelButton)
+        expect(onSelectedIdChange).toHaveBeenCalledWith([])
+      }
+    })
+
+    it('should show download option for downloadable documents', () => {
+      const props = {
+        ...defaultProps,
+        selectedIds: ['doc-1'],
+        documents: [
+          createMockDoc({ id: 'doc-1', data_source_type: DataSourceType.FILE }),
+        ],
+      }
+      render(<DocumentList {...props} />, { wrapper: createWrapper() })
+
+      // BatchAction should be visible
+      expect(screen.getByRole('table')).toBeInTheDocument()
+    })
+
+    it('should show re-index option for error documents', () => {
+      const props = {
+        ...defaultProps,
+        selectedIds: ['doc-1'],
+        documents: [
+          createMockDoc({ id: 'doc-1', display_status: 'error' }),
+        ],
+      }
+      render(<DocumentList {...props} />, { wrapper: createWrapper() })
+
+      // BatchAction with re-index should be present for error documents
+      expect(screen.getByRole('table')).toBeInTheDocument()
+    })
+  })
+
+  describe('Row Click Navigation', () => {
+    it('should navigate to document detail when row is clicked', () => {
+      render(<DocumentList {...defaultProps} />, { wrapper: createWrapper() })
+
+      const rows = screen.getAllByRole('row')
+      // First row is header, second row is first document
+      if (rows.length > 1) {
+        fireEvent.click(rows[1])
+        expect(mockPush).toHaveBeenCalledWith('/datasets/dataset-1/documents/doc-1')
+      }
+    })
+  })
+
+  describe('Rename Modal', () => {
+    it('should not show rename modal initially', () => {
+      render(<DocumentList {...defaultProps} />, { wrapper: createWrapper() })
+
+      // RenameModal should not be visible initially
+      const modal = screen.queryByRole('dialog')
+      expect(modal).not.toBeInTheDocument()
+    })
+
+    it('should show rename modal when rename button is clicked', () => {
+      const { container } = render(<DocumentList {...defaultProps} />, { wrapper: createWrapper() })
+
+      // Find and click the rename button in the first row
+      const renameButtons = container.querySelectorAll('.cursor-pointer.rounded-md')
+      if (renameButtons.length > 0) {
+        fireEvent.click(renameButtons[0])
+      }
+
+      // After clicking rename, the modal should potentially be visible
+      expect(screen.getByRole('table')).toBeInTheDocument()
+    })
+
+    it('should call onUpdate when document is renamed', () => {
+      const onUpdate = vi.fn()
+      const props = { ...defaultProps, onUpdate }
+      render(<DocumentList {...props} />, { wrapper: createWrapper() })
+
+      // The handleRenamed callback wraps onUpdate
+      expect(screen.getByRole('table')).toBeInTheDocument()
+    })
+  })
+
+  describe('Edit Metadata Modal', () => {
+    it('should handle edit metadata action', () => {
+      const props = {
+        ...defaultProps,
+        selectedIds: ['doc-1'],
+      }
+      render(<DocumentList {...props} />, { wrapper: createWrapper() })
+
+      const editButton = screen.queryByRole('button', { name: /metadata/i })
+      if (editButton) {
+        fireEvent.click(editButton)
+      }
+
+      expect(screen.getByRole('table')).toBeInTheDocument()
+    })
+
+    it('should call onManageMetadata when manage metadata is triggered', () => {
+      const onManageMetadata = vi.fn()
+      const props = {
+        ...defaultProps,
+        selectedIds: ['doc-1'],
+        onManageMetadata,
+      }
+      render(<DocumentList {...props} />, { wrapper: createWrapper() })
+
+      // The onShowManage callback in EditMetadataBatchModal should call hideEditModal then onManageMetadata
+      expect(screen.getByRole('table')).toBeInTheDocument()
+    })
+  })
+
+  describe('Chunking Mode', () => {
+    it('should render with general mode', () => {
+      render(<DocumentList {...defaultProps} />, { wrapper: createWrapper() })
+      expect(screen.getByRole('table')).toBeInTheDocument()
+    })
+
+    it('should render with QA mode', () => {
+      // This test uses the default mock which returns ChunkingMode.text
+      // The component will compute isQAMode based on doc_form
+      render(<DocumentList {...defaultProps} />, { wrapper: createWrapper() })
+      expect(screen.getByRole('table')).toBeInTheDocument()
+    })
+
+    it('should render with parent-child mode', () => {
+      render(<DocumentList {...defaultProps} />, { wrapper: createWrapper() })
+      expect(screen.getByRole('table')).toBeInTheDocument()
+    })
+  })
+
+  describe('Edge Cases', () => {
+    it('should handle empty documents array', () => {
+      const props = { ...defaultProps, documents: [] }
+      render(<DocumentList {...props} />, { wrapper: createWrapper() })
+
+      expect(screen.getByRole('table')).toBeInTheDocument()
+    })
+
+    it('should handle documents with missing optional fields', () => {
+      const docWithMissingFields = createMockDoc({
+        word_count: undefined as unknown as number,
+        hit_count: undefined as unknown as number,
+      })
+      const props = {
+        ...defaultProps,
+        documents: [docWithMissingFields],
+      }
+      render(<DocumentList {...props} />, { wrapper: createWrapper() })
+
+      expect(screen.getByRole('table')).toBeInTheDocument()
+    })
+
+    it('should handle status filter value', () => {
+      const props = {
+        ...defaultProps,
+        statusFilterValue: 'completed',
+      }
+      render(<DocumentList {...props} />, { wrapper: createWrapper() })
+
+      expect(screen.getByRole('table')).toBeInTheDocument()
+    })
+
+    it('should handle remote sort value', () => {
+      const props = {
+        ...defaultProps,
+        remoteSortValue: 'created_at',
+      }
+      render(<DocumentList {...props} />, { wrapper: createWrapper() })
+
+      expect(screen.getByRole('table')).toBeInTheDocument()
+    })
+
+    it('should handle large number of documents', () => {
+      const manyDocs = Array.from({ length: 20 }, (_, i) =>
+        createMockDoc({ id: `doc-${i}`, name: `Document ${i}.txt` }))
+      const props = { ...defaultProps, documents: manyDocs }
+      render(<DocumentList {...props} />, { wrapper: createWrapper() })
+
+      expect(screen.getByRole('table')).toBeInTheDocument()
+    }, 10000)
+  })
+})

web/app/components/datasets/documents/components/document-list/index.tsx

@@ -0,0 +1,3 @@
+// Re-export from parent for backwards compatibility
+export { default } from '../list'
+export { renderTdValue } from './components'

web/app/components/datasets/documents/components/list.tsx

@@ -1,67 +1,26 @@
 'use client'
 import type { FC } from 'react'
 import type { Props as PaginationProps } from '@/app/components/base/pagination'
-import type { CommonResponse } from '@/models/common'
-import type { LegacyDataSourceInfo, LocalFileInfo, OnlineDocumentInfo, OnlineDriveInfo, SimpleDocumentDetail } from '@/models/datasets'
-import {
-  RiArrowDownLine,
-  RiEditLine,
-  RiGlobalLine,
-} from '@remixicon/react'
+import type { SimpleDocumentDetail } from '@/models/datasets'
 import { useBoolean } from 'ahooks'
-import { uniq } from 'es-toolkit/array'
-import { pick } from 'es-toolkit/object'
-import { useRouter } from 'next/navigation'
 import * as React from 'react'
-import { useCallback, useEffect, useMemo, useState } from 'react'
+import { useCallback, useState } from 'react'
 import { useTranslation } from 'react-i18next'
 import Checkbox from '@/app/components/base/checkbox'
-import FileTypeIcon from '@/app/components/base/file-uploader/file-type-icon'
-import NotionIcon from '@/app/components/base/notion-icon'
 import Pagination from '@/app/components/base/pagination'
-import Toast from '@/app/components/base/toast'
-import Tooltip from '@/app/components/base/tooltip'
-import ChunkingModeLabel from '@/app/components/datasets/common/chunking-mode-label'
-import { normalizeStatusForQuery } from '@/app/components/datasets/documents/status-filter'
-import { extensionToFileType } from '@/app/components/datasets/hit-testing/utils/extension-to-file-type'
 import EditMetadataBatchModal from '@/app/components/datasets/metadata/edit-metadata-batch/modal'
 import useBatchEditDocumentMetadata from '@/app/components/datasets/metadata/hooks/use-batch-edit-document-metadata'
 import { useDatasetDetailContextWithSelector as useDatasetDetailContext } from '@/context/dataset-detail'
-import useTimestamp from '@/hooks/use-timestamp'
-import { ChunkingMode, DataSourceType, DocumentActionType } from '@/models/datasets'
-import { DatasourceType } from '@/models/pipeline'
-import { useDocumentArchive, useDocumentBatchRetryIndex, useDocumentDelete, useDocumentDisable, useDocumentDownloadZip, useDocumentEnable, useDocumentSummary } from '@/service/knowledge/use-document'
-import { asyncRunSafe } from '@/utils'
-import { cn } from '@/utils/classnames'
-import { downloadBlob } from '@/utils/download'
-import { formatNumber } from '@/utils/format'
+import { ChunkingMode, DocumentActionType } from '@/models/datasets'
 import BatchAction from '../detail/completed/common/batch-action'
-import SummaryStatus from '../detail/completed/common/summary-status'
-import StatusItem from '../status-item'
 import s from '../style.module.css'
-import Operations from './operations'
+import { DocumentTableRow, renderTdValue, SortHeader } from './document-list/components'
+import { useDocumentActions, useDocumentSelection, useDocumentSort } from './document-list/hooks'
 import RenameModal from './rename-modal'
 
-export const renderTdValue = (value: string | number | null, isEmptyStyle = false) => {
-  return (
-    <div className={cn(isEmptyStyle ? 'text-text-tertiary' : 'text-text-secondary', s.tdValue)}>
-      {value ?? '-'}
-    </div>
-  )
-}
-
-const renderCount = (count: number | undefined) => {
-  if (!count)
-    return renderTdValue(0, true)
-
-  if (count < 1000)
-    return count
-
-  return `${formatNumber((count / 1000).toFixed(1))}k`
-}
-
 type LocalDoc = SimpleDocumentDetail & { percent?: number }
-type IDocumentListProps = {
+
+type DocumentListProps = {
   embeddingAvailable: boolean
   documents: LocalDoc[]
   selectedIds: string[]
@@ -77,7 +36,7 @@ type IDocumentListProps = {
 /**
  * Document list component including basic information
  */
-const DocumentList: FC<IDocumentListProps> = ({
+const DocumentList: FC<DocumentListProps> = ({
   embeddingAvailable,
   documents = [],
   selectedIds,
@@ -90,20 +49,43 @@ const DocumentList: FC<IDocumentListProps> = ({
   remoteSortValue,
 }) => {
   const { t } = useTranslation()
-  const { formatTime } = useTimestamp()
-  const router = useRouter()
   const datasetConfig = useDatasetDetailContext(s => s.dataset)
   const chunkingMode = datasetConfig?.doc_form
   const isGeneralMode = chunkingMode !== ChunkingMode.parentChild
   const isQAMode = chunkingMode === ChunkingMode.qa
-  const [sortField, setSortField] = useState<'name' | 'word_count' | 'hit_count' | 'created_at' | null>(null)
-  const [sortOrder, setSortOrder] = useState<'asc' | 'desc'>('desc')
 
-  useEffect(() => {
-    setSortField(null)
-    setSortOrder('desc')
-  }, [remoteSortValue])
+  // Sorting
+  const { sortField, sortOrder, handleSort, sortedDocuments } = useDocumentSort({
+    documents,
+    statusFilterValue,
+    remoteSortValue,
+  })
 
+  // Selection
+  const {
+    isAllSelected,
+    isSomeSelected,
+    onSelectAll,
+    onSelectOne,
+    hasErrorDocumentsSelected,
+    downloadableSelectedIds,
+    clearSelection,
+  } = useDocumentSelection({
+    documents: sortedDocuments,
+    selectedIds,
+    onSelectedIdChange,
+  })
+
+  // Actions
+  const { handleAction, handleBatchReIndex, handleBatchDownload } = useDocumentActions({
+    datasetId,
+    selectedIds,
+    downloadableSelectedIds,
+    onUpdate,
+    onClearSelection: clearSelection,
+  })
+
+  // Batch edit metadata
   const {
     isShowEditModal,
     showEditModal,
@@ -113,233 +95,26 @@ const DocumentList: FC<IDocumentListProps> = ({
   } = useBatchEditDocumentMetadata({
     datasetId,
     docList: documents.filter(doc => selectedIds.includes(doc.id)),
-    selectedDocumentIds: selectedIds, // Pass all selected IDs separately
+    selectedDocumentIds: selectedIds,
     onUpdate,
   })
 
-  const localDocs = useMemo(() => {
-    let filteredDocs = documents
-
-    if (statusFilterValue && statusFilterValue !== 'all') {
-      filteredDocs = filteredDocs.filter(doc =>
-        typeof doc.display_status === 'string'
-        && normalizeStatusForQuery(doc.display_status) === statusFilterValue,
-      )
-    }
-
-    if (!sortField)
-      return filteredDocs
-
-    const sortedDocs = [...filteredDocs].sort((a, b) => {
-      let aValue: any
-      let bValue: any
-
-      switch (sortField) {
-        case 'name':
-          aValue = a.name?.toLowerCase() || ''
-          bValue = b.name?.toLowerCase() || ''
-          break
-        case 'word_count':
-          aValue = a.word_count || 0
-          bValue = b.word_count || 0
-          break
-        case 'hit_count':
-          aValue = a.hit_count || 0
-          bValue = b.hit_count || 0
-          break
-        case 'created_at':
-          aValue = a.created_at
-          bValue = b.created_at
-          break
-        default:
-          return 0
-      }
-
-      if (sortField === 'name') {
-        const result = aValue.localeCompare(bValue)
-        return sortOrder === 'asc' ? result : -result
-      }
-      else {
-        const result = aValue - bValue
-        return sortOrder === 'asc' ? result : -result
-      }
-    })
-
-    return sortedDocs
-  }, [documents, sortField, sortOrder, statusFilterValue])
-
-  const handleSort = (field: 'name' | 'word_count' | 'hit_count' | 'created_at') => {
-    if (sortField === field) {
-      setSortOrder(sortOrder === 'asc' ? 'desc' : 'asc')
-    }
-    else {
-      setSortField(field)
-      setSortOrder('desc')
-    }
-  }
-
-  const renderSortHeader = (field: 'name' | 'word_count' | 'hit_count' | 'created_at', label: string) => {
-    const isActive = sortField === field
-    const isDesc = isActive && sortOrder === 'desc'
-
-    return (
-      <div className="flex cursor-pointer items-center hover:text-text-secondary" onClick={() => handleSort(field)}>
-        {label}
-        <RiArrowDownLine
-          className={cn('ml-0.5 h-3 w-3 transition-all', isActive ? 'text-text-tertiary' : 'text-text-disabled', isActive && !isDesc ? 'rotate-180' : '')}
-        />
-      </div>
-    )
-  }
-
+  // Rename modal
   const [currDocument, setCurrDocument] = useState<LocalDoc | null>(null)
   const [isShowRenameModal, {
     setTrue: setShowRenameModalTrue,
     setFalse: setShowRenameModalFalse,
   }] = useBoolean(false)
+
   const handleShowRenameModal = useCallback((doc: LocalDoc) => {
     setCurrDocument(doc)
     setShowRenameModalTrue()
   }, [setShowRenameModalTrue])
+
   const handleRenamed = useCallback(() => {
     onUpdate()
   }, [onUpdate])
 
-  const isAllSelected = useMemo(() => {
-    return localDocs.length > 0 && localDocs.every(doc => selectedIds.includes(doc.id))
-  }, [localDocs, selectedIds])
-
-  const isSomeSelected = useMemo(() => {
-    return localDocs.some(doc => selectedIds.includes(doc.id))
-  }, [localDocs, selectedIds])
-
-  const onSelectedAll = useCallback(() => {
-    if (isAllSelected)
-      onSelectedIdChange([])
-    else
-      onSelectedIdChange(uniq([...selectedIds, ...localDocs.map(doc => doc.id)]))
-  }, [isAllSelected, localDocs, onSelectedIdChange, selectedIds])
-  const { mutateAsync: archiveDocument } = useDocumentArchive()
-  const { mutateAsync: generateSummary } = useDocumentSummary()
-  const { mutateAsync: enableDocument } = useDocumentEnable()
-  const { mutateAsync: disableDocument } = useDocumentDisable()
-  const { mutateAsync: deleteDocument } = useDocumentDelete()
-  const { mutateAsync: retryIndexDocument } = useDocumentBatchRetryIndex()
-  const { mutateAsync: requestDocumentsZip, isPending: isDownloadingZip } = useDocumentDownloadZip()
-
-  const handleAction = (actionName: DocumentActionType) => {
-    return async () => {
-      let opApi
-      switch (actionName) {
-        case DocumentActionType.archive:
-          opApi = archiveDocument
-          break
-        case DocumentActionType.summary:
-          opApi = generateSummary
-          break
-        case DocumentActionType.enable:
-          opApi = enableDocument
-          break
-        case DocumentActionType.disable:
-          opApi = disableDocument
-          break
-        default:
-          opApi = deleteDocument
-          break
-      }
-      const [e] = await asyncRunSafe<CommonResponse>(opApi({ datasetId, documentIds: selectedIds }) as Promise<CommonResponse>)
-
-      if (!e) {
-        if (actionName === DocumentActionType.delete)
-          onSelectedIdChange([])
-        Toast.notify({ type: 'success', message: t('actionMsg.modifiedSuccessfully', { ns: 'common' }) })
-        onUpdate()
-      }
-      else { Toast.notify({ type: 'error', message: t('actionMsg.modifiedUnsuccessfully', { ns: 'common' }) }) }
-    }
-  }
-
-  const handleBatchReIndex = async () => {
-    const [e] = await asyncRunSafe<CommonResponse>(retryIndexDocument({ datasetId, documentIds: selectedIds }))
-    if (!e) {
-      onSelectedIdChange([])
-      Toast.notify({ type: 'success', message: t('actionMsg.modifiedSuccessfully', { ns: 'common' }) })
-      onUpdate()
-    }
-    else {
-      Toast.notify({ type: 'error', message: t('actionMsg.modifiedUnsuccessfully', { ns: 'common' }) })
-    }
-  }
-
-  const hasErrorDocumentsSelected = useMemo(() => {
-    return localDocs.some(doc => selectedIds.includes(doc.id) && doc.display_status === 'error')
-  }, [localDocs, selectedIds])
-
-  const getFileExtension = useCallback((fileName: string): string => {
-    if (!fileName)
-      return ''
-    const parts = fileName.split('.')
-    if (parts.length <= 1 || (parts[0] === '' && parts.length === 2))
-      return ''
-
-    return parts[parts.length - 1].toLowerCase()
-  }, [])
-
-  const isCreateFromRAGPipeline = useCallback((createdFrom: string) => {
-    return createdFrom === 'rag-pipeline'
-  }, [])
-
-  /**
-   * Calculate the data source type
-   * DataSourceType: FILE, NOTION, WEB (legacy)
-   * DatasourceType: localFile, onlineDocument, websiteCrawl, onlineDrive (new)
-   */
-  const isLocalFile = useCallback((dataSourceType: DataSourceType | DatasourceType) => {
-    return dataSourceType === DatasourceType.localFile || dataSourceType === DataSourceType.FILE
-  }, [])
-  const isOnlineDocument = useCallback((dataSourceType: DataSourceType | DatasourceType) => {
-    return dataSourceType === DatasourceType.onlineDocument || dataSourceType === DataSourceType.NOTION
-  }, [])
-  const isWebsiteCrawl = useCallback((dataSourceType: DataSourceType | DatasourceType) => {
-    return dataSourceType === DatasourceType.websiteCrawl || dataSourceType === DataSourceType.WEB
-  }, [])
-  const isOnlineDrive = useCallback((dataSourceType: DataSourceType | DatasourceType) => {
-    return dataSourceType === DatasourceType.onlineDrive
-  }, [])
-
-  const downloadableSelectedIds = useMemo(() => {
-    const selectedSet = new Set(selectedIds)
-    return localDocs
-      .filter(doc => selectedSet.has(doc.id) && doc.data_source_type === DataSourceType.FILE)
-      .map(doc => doc.id)
-  }, [localDocs, selectedIds])
-
-  /**
-   * Generate a random ZIP filename for bulk document downloads.
-   * We intentionally avoid leaking dataset info in the exported archive name.
-   */
-  const generateDocsZipFileName = useCallback((): string => {
-    // Prefer UUID for uniqueness; fall back to time+random when unavailable.
-    const randomPart = (typeof crypto !== 'undefined' && typeof crypto.randomUUID === 'function')
-      ? crypto.randomUUID()
-      : `${Date.now().toString(36)}${Math.random().toString(36).slice(2, 10)}`
-    return `${randomPart}-docs.zip`
-  }, [])
-
-  const handleBatchDownload = useCallback(async () => {
-    if (isDownloadingZip)
-      return
-
-    // Download as a single ZIP to avoid browser caps on multiple automatic downloads.
-    const [e, blob] = await asyncRunSafe(requestDocumentsZip({ datasetId, documentIds: downloadableSelectedIds }))
-    if (e || !blob) {
-      Toast.notify({ type: 'error', message: t('actionMsg.downloadUnsuccessfully', { ns: 'common' }) })
-      return
-    }
-
-    downloadBlob({ data: blob, fileName: generateDocsZipFileName() })
-  }, [datasetId, downloadableSelectedIds, generateDocsZipFileName, isDownloadingZip, requestDocumentsZip, t])
-
   return (
     <div className="relative mt-3 flex h-full w-full flex-col">
       <div className="relative h-0 grow overflow-x-auto">
@@ -353,157 +128,76 @@ const DocumentList: FC<IDocumentListProps> = ({
                       className="mr-2 shrink-0"
                       checked={isAllSelected}
                       indeterminate={!isAllSelected && isSomeSelected}
-                      onCheck={onSelectedAll}
+                      onCheck={onSelectAll}
                     />
                   )}
                   #
                 </div>
               </td>
               <td>
-                {renderSortHeader('name', t('list.table.header.fileName', { ns: 'datasetDocuments' }))}
+                <SortHeader
+                  field="name"
+                  label={t('list.table.header.fileName', { ns: 'datasetDocuments' })}
+                  currentSortField={sortField}
+                  sortOrder={sortOrder}
+                  onSort={handleSort}
+                />
               </td>
               <td className="w-[130px]">{t('list.table.header.chunkingMode', { ns: 'datasetDocuments' })}</td>
               <td className="w-24">
-                {renderSortHeader('word_count', t('list.table.header.words', { ns: 'datasetDocuments' }))}
+                <SortHeader
+                  field="word_count"
+                  label={t('list.table.header.words', { ns: 'datasetDocuments' })}
+                  currentSortField={sortField}
+                  sortOrder={sortOrder}
+                  onSort={handleSort}
+                />
               </td>
               <td className="w-44">
-                {renderSortHeader('hit_count', t('list.table.header.hitCount', { ns: 'datasetDocuments' }))}
+                <SortHeader
+                  field="hit_count"
+                  label={t('list.table.header.hitCount', { ns: 'datasetDocuments' })}
+                  currentSortField={sortField}
+                  sortOrder={sortOrder}
+                  onSort={handleSort}
+                />
               </td>
               <td className="w-44">
-                {renderSortHeader('created_at', t('list.table.header.uploadTime', { ns: 'datasetDocuments' }))}
+                <SortHeader
+                  field="created_at"
+                  label={t('list.table.header.uploadTime', { ns: 'datasetDocuments' })}
+                  currentSortField={sortField}
+                  sortOrder={sortOrder}
+                  onSort={handleSort}
+                />
               </td>
               <td className="w-40">{t('list.table.header.status', { ns: 'datasetDocuments' })}</td>
               <td className="w-20">{t('list.table.header.action', { ns: 'datasetDocuments' })}</td>
             </tr>
           </thead>
           <tbody className="text-text-secondary">
-            {localDocs.map((doc, index) => {
-              const isFile = isLocalFile(doc.data_source_type)
-              const fileType = isFile ? doc.data_source_detail_dict?.upload_file?.extension : ''
-              return (
-                <tr
-                  key={doc.id}
-                  className="h-8 cursor-pointer border-b border-divider-subtle hover:bg-background-default-hover"
-                  onClick={() => {
-                    router.push(`/datasets/${datasetId}/documents/${doc.id}`)
-                  }}
-                >
-                  <td className="text-left align-middle text-xs text-text-tertiary">
-                    <div className="flex items-center" onClick={e => e.stopPropagation()}>
-                      <Checkbox
-                        className="mr-2 shrink-0"
-                        checked={selectedIds.includes(doc.id)}
-                        onCheck={() => {
-                          onSelectedIdChange(
-                            selectedIds.includes(doc.id)
-                              ? selectedIds.filter(id => id !== doc.id)
-                              : [...selectedIds, doc.id],
-                          )
-                        }}
-                      />
-                      {index + 1}
-                    </div>
-                  </td>
-                  <td>
-                    <div className="group mr-6 flex max-w-[460px] items-center hover:mr-0">
-                      <div className="flex shrink-0 items-center">
-                        {isOnlineDocument(doc.data_source_type) && (
-                          <NotionIcon
-                            className="mr-1.5"
-                            type="page"
-                            src={
-                              isCreateFromRAGPipeline(doc.created_from)
-                                ? (doc.data_source_info as OnlineDocumentInfo).page.page_icon
-                                : (doc.data_source_info as LegacyDataSourceInfo).notion_page_icon
-                            }
-                          />
-                        )}
-                        {isLocalFile(doc.data_source_type) && (
-                          <FileTypeIcon
-                            type={
-                              extensionToFileType(
-                                isCreateFromRAGPipeline(doc.created_from)
-                                  ? (doc?.data_source_info as LocalFileInfo)?.extension
-                                  : ((doc?.data_source_info as LegacyDataSourceInfo)?.upload_file?.extension ?? fileType),
-                              )
-                            }
-                            className="mr-1.5"
-                          />
-                        )}
-                        {isOnlineDrive(doc.data_source_type) && (
-                          <FileTypeIcon
-                            type={
-                              extensionToFileType(
-                                getFileExtension((doc?.data_source_info as unknown as OnlineDriveInfo)?.name),
-                              )
-                            }
-                            className="mr-1.5"
-                          />
-                        )}
-                        {isWebsiteCrawl(doc.data_source_type) && (
-                          <RiGlobalLine className="mr-1.5 size-4" />
-                        )}
-                      </div>
-                      <Tooltip
-                        popupContent={doc.name}
-                      >
-                        <span className="grow-1 truncate text-sm">{doc.name}</span>
-                      </Tooltip>
-                      {
-                        doc.summary_index_status && (
-                          <div className="ml-1 hidden shrink-0 group-hover:flex">
-                            <SummaryStatus status={doc.summary_index_status} />
-                          </div>
-                        )
-                      }
-                      <div className="hidden shrink-0 group-hover:ml-auto group-hover:flex">
-                        <Tooltip
-                          popupContent={t('list.table.rename', { ns: 'datasetDocuments' })}
-                        >
-                          <div
-                            className="cursor-pointer rounded-md p-1 hover:bg-state-base-hover"
-                            onClick={(e) => {
-                              e.stopPropagation()
-                              handleShowRenameModal(doc)
-                            }}
-                          >
-                            <RiEditLine className="h-4 w-4 text-text-tertiary" />
-                          </div>
-                        </Tooltip>
-                      </div>
-                    </div>
-                  </td>
-                  <td>
-                    <ChunkingModeLabel
-                      isGeneralMode={isGeneralMode}
-                      isQAMode={isQAMode}
-                    />
-                  </td>
-                  <td>{renderCount(doc.word_count)}</td>
-                  <td>{renderCount(doc.hit_count)}</td>
-                  <td className="text-[13px] text-text-secondary">
-                    {formatTime(doc.created_at, t('dateTimeFormat', { ns: 'datasetHitTesting' }) as string)}
-                  </td>
-                  <td>
-                    <StatusItem status={doc.display_status} />
-                  </td>
-                  <td>
-                    <Operations
-                      selectedIds={selectedIds}
-                      onSelectedIdChange={onSelectedIdChange}
-                      embeddingAvailable={embeddingAvailable}
-                      datasetId={datasetId}
-                      detail={pick(doc, ['name', 'enabled', 'archived', 'id', 'data_source_type', 'doc_form', 'display_status'])}
-                      onUpdate={onUpdate}
-                    />
-                  </td>
-                </tr>
-              )
-            })}
+            {sortedDocuments.map((doc, index) => (
+              <DocumentTableRow
+                key={doc.id}
+                doc={doc}
+                index={index}
+                datasetId={datasetId}
+                isSelected={selectedIds.includes(doc.id)}
+                isGeneralMode={isGeneralMode}
+                isQAMode={isQAMode}
+                embeddingAvailable={embeddingAvailable}
+                selectedIds={selectedIds}
+                onSelectOne={onSelectOne}
+                onSelectedIdChange={onSelectedIdChange}
+                onShowRenameModal={handleShowRenameModal}
+                onUpdate={onUpdate}
+              />
+            ))}
           </tbody>
         </table>
       </div>
-      {(selectedIds.length > 0) && (
+
+      {selectedIds.length > 0 && (
         <BatchAction
           className="absolute bottom-16 left-0 z-20"
           selectedIds={selectedIds}
@@ -515,12 +209,10 @@ const DocumentList: FC<IDocumentListProps> = ({
           onBatchDelete={handleAction(DocumentActionType.delete)}
           onEditMetadata={showEditModal}
           onBatchReIndex={hasErrorDocumentsSelected ? handleBatchReIndex : undefined}
-          onCancel={() => {
-            onSelectedIdChange([])
-          }}
+          onCancel={clearSelection}
         />
       )}
-      {/* Show Pagination only if the total is more than the limit */}
+
       {!!pagination.total && (
         <Pagination
           {...pagination}
@@ -556,3 +248,5 @@ const DocumentList: FC<IDocumentListProps> = ({
 }
 
 export default DocumentList
+
+export { renderTdValue }

web/app/components/datasets/documents/components/operations.tsx

@@ -26,6 +26,7 @@ import CustomPopover from '@/app/components/base/popover'
 import Switch from '@/app/components/base/switch'
 import { ToastContext } from '@/app/components/base/toast'
 import Tooltip from '@/app/components/base/tooltip'
+import { IS_CE_EDITION } from '@/config'
 import { DataSourceType, DocumentActionType } from '@/models/datasets'
 import {
   useDocumentArchive,
@@ -263,10 +264,14 @@ const Operations = ({
                         <span className={s.actionName}>{t('list.action.sync', { ns: 'datasetDocuments' })}</span>
                       </div>
                     )}
-                    <div className={s.actionItem} onClick={() => onOperate('summary')}>
-                      <SearchLinesSparkle className="h-4 w-4 text-text-tertiary" />
-                      <span className={s.actionName}>{t('list.action.summary', { ns: 'datasetDocuments' })}</span>
-                    </div>
+                    {
+                      IS_CE_EDITION && (
+                        <div className={s.actionItem} onClick={() => onOperate('summary')}>
+                          <SearchLinesSparkle className="h-4 w-4 text-text-tertiary" />
+                          <span className={s.actionName}>{t('list.action.summary', { ns: 'datasetDocuments' })}</span>
+                        </div>
+                      )
+                    }
                     <Divider className="my-1" />
                   </>
                 )}

web/app/components/datasets/documents/create-from-pipeline/data-source/local-file/components/file-list-item.spec.tsx

@@ -0,0 +1,351 @@
+import type { FileListItemProps } from './file-list-item'
+import type { CustomFile as File, FileItem } from '@/models/datasets'
+import { fireEvent, render, screen } from '@testing-library/react'
+import { beforeEach, describe, expect, it, vi } from 'vitest'
+import { PROGRESS_ERROR, PROGRESS_NOT_STARTED } from '../constants'
+import FileListItem from './file-list-item'
+
+// Mock theme hook - can be changed per test
+let mockTheme = 'light'
+vi.mock('@/hooks/use-theme', () => ({
+  default: () => ({ theme: mockTheme }),
+}))
+
+// Mock theme types
+vi.mock('@/types/app', () => ({
+  Theme: { dark: 'dark', light: 'light' },
+}))
+
+// Mock SimplePieChart with dynamic import handling
+vi.mock('next/dynamic', () => ({
+  default: () => {
+    const DynamicComponent = ({ percentage, stroke, fill }: { percentage: number, stroke: string, fill: string }) => (
+      <div data-testid="pie-chart" data-percentage={percentage} data-stroke={stroke} data-fill={fill}>
+        Pie Chart:
+        {' '}
+        {percentage}
+        %
+      </div>
+    )
+    DynamicComponent.displayName = 'SimplePieChart'
+    return DynamicComponent
+  },
+}))
+
+// Mock DocumentFileIcon
+vi.mock('@/app/components/datasets/common/document-file-icon', () => ({
+  default: ({ name, extension, size }: { name: string, extension: string, size: string }) => (
+    <div data-testid="document-icon" data-name={name} data-extension={extension} data-size={size}>
+      Document Icon
+    </div>
+  ),
+}))
+
+describe('FileListItem', () => {
+  const createMockFile = (overrides: Partial<File> = {}): File => ({
+    name: 'test-document.pdf',
+    size: 1024 * 100, // 100KB
+    type: 'application/pdf',
+    lastModified: Date.now(),
+    ...overrides,
+  } as File)
+
+  const createMockFileItem = (overrides: Partial<FileItem> = {}): FileItem => ({
+    fileID: 'file-123',
+    file: createMockFile(overrides.file as Partial<File>),
+    progress: PROGRESS_NOT_STARTED,
+    ...overrides,
+  })
+
+  const defaultProps: FileListItemProps = {
+    fileItem: createMockFileItem(),
+    onPreview: vi.fn(),
+    onRemove: vi.fn(),
+  }
+
+  beforeEach(() => {
+    vi.clearAllMocks()
+  })
+
+  describe('rendering', () => {
+    it('should render the file item container', () => {
+      const { container } = render(<FileListItem {...defaultProps} />)
+
+      const item = container.firstChild as HTMLElement
+      expect(item).toHaveClass('flex', 'h-12', 'items-center', 'rounded-lg')
+    })
+
+    it('should render document icon with correct props', () => {
+      render(<FileListItem {...defaultProps} />)
+
+      const icon = screen.getByTestId('document-icon')
+      expect(icon).toBeInTheDocument()
+      expect(icon).toHaveAttribute('data-name', 'test-document.pdf')
+      expect(icon).toHaveAttribute('data-extension', 'pdf')
+      expect(icon).toHaveAttribute('data-size', 'lg')
+    })
+
+    it('should render file name', () => {
+      render(<FileListItem {...defaultProps} />)
+
+      expect(screen.getByText('test-document.pdf')).toBeInTheDocument()
+    })
+
+    it('should render file extension in uppercase via CSS class', () => {
+      render(<FileListItem {...defaultProps} />)
+
+      // Extension is rendered in lowercase but styled with uppercase CSS
+      const extensionSpan = screen.getByText('pdf')
+      expect(extensionSpan).toBeInTheDocument()
+      expect(extensionSpan).toHaveClass('uppercase')
+    })
+
+    it('should render file size', () => {
+      render(<FileListItem {...defaultProps} />)
+
+      // 100KB (102400 bytes) formatted with formatFileSize
+      expect(screen.getByText('100.00 KB')).toBeInTheDocument()
+    })
+
+    it('should render delete button', () => {
+      const { container } = render(<FileListItem {...defaultProps} />)
+
+      const deleteButton = container.querySelector('.cursor-pointer')
+      expect(deleteButton).toBeInTheDocument()
+    })
+  })
+
+  describe('progress states', () => {
+    it('should show progress chart when uploading (0-99)', () => {
+      const fileItem = createMockFileItem({ progress: 50 })
+      render(<FileListItem {...defaultProps} fileItem={fileItem} />)
+
+      const pieChart = screen.getByTestId('pie-chart')
+      expect(pieChart).toBeInTheDocument()
+      expect(pieChart).toHaveAttribute('data-percentage', '50')
+    })
+
+    it('should show progress chart at 0%', () => {
+      const fileItem = createMockFileItem({ progress: 0 })
+      render(<FileListItem {...defaultProps} fileItem={fileItem} />)
+
+      const pieChart = screen.getByTestId('pie-chart')
+      expect(pieChart).toHaveAttribute('data-percentage', '0')
+    })
+
+    it('should not show progress chart when complete (100)', () => {
+      const fileItem = createMockFileItem({ progress: 100 })
+      render(<FileListItem {...defaultProps} fileItem={fileItem} />)
+
+      expect(screen.queryByTestId('pie-chart')).not.toBeInTheDocument()
+    })
+
+    it('should not show progress chart when not started (-1)', () => {
+      const fileItem = createMockFileItem({ progress: PROGRESS_NOT_STARTED })
+      render(<FileListItem {...defaultProps} fileItem={fileItem} />)
+
+      expect(screen.queryByTestId('pie-chart')).not.toBeInTheDocument()
+    })
+  })
+
+  describe('error state', () => {
+    it('should show error icon when progress is PROGRESS_ERROR', () => {
+      const fileItem = createMockFileItem({ progress: PROGRESS_ERROR })
+      const { container } = render(<FileListItem {...defaultProps} fileItem={fileItem} />)
+
+      const errorIcon = container.querySelector('.text-text-destructive')
+      expect(errorIcon).toBeInTheDocument()
+    })
+
+    it('should apply error styling to container', () => {
+      const fileItem = createMockFileItem({ progress: PROGRESS_ERROR })
+      const { container } = render(<FileListItem {...defaultProps} fileItem={fileItem} />)
+
+      const item = container.firstChild as HTMLElement
+      expect(item).toHaveClass('border-state-destructive-border', 'bg-state-destructive-hover')
+    })
+
+    it('should not show error styling when not in error state', () => {
+      const { container } = render(<FileListItem {...defaultProps} />)
+
+      const item = container.firstChild as HTMLElement
+      expect(item).not.toHaveClass('border-state-destructive-border')
+    })
+  })
+
+  describe('theme handling', () => {
+    it('should use correct chart color for light theme', () => {
+      mockTheme = 'light'
+      const fileItem = createMockFileItem({ progress: 50 })
+      render(<FileListItem {...defaultProps} fileItem={fileItem} />)
+
+      const pieChart = screen.getByTestId('pie-chart')
+      expect(pieChart).toHaveAttribute('data-stroke', '#296dff')
+      expect(pieChart).toHaveAttribute('data-fill', '#296dff')
+    })
+
+    it('should use correct chart color for dark theme', () => {
+      mockTheme = 'dark'
+      const fileItem = createMockFileItem({ progress: 50 })
+      render(<FileListItem {...defaultProps} fileItem={fileItem} />)
+
+      const pieChart = screen.getByTestId('pie-chart')
+      expect(pieChart).toHaveAttribute('data-stroke', '#5289ff')
+      expect(pieChart).toHaveAttribute('data-fill', '#5289ff')
+    })
+  })
+
+  describe('event handlers', () => {
+    it('should call onPreview when item is clicked', () => {
+      const onPreview = vi.fn()
+      const fileItem = createMockFileItem()
+      render(<FileListItem {...defaultProps} fileItem={fileItem} onPreview={onPreview} />)
+
+      const item = screen.getByText('test-document.pdf').closest('[class*="flex h-12"]')!
+      fireEvent.click(item)
+
+      expect(onPreview).toHaveBeenCalledTimes(1)
+      expect(onPreview).toHaveBeenCalledWith(fileItem.file)
+    })
+
+    it('should call onRemove when delete button is clicked', () => {
+      const onRemove = vi.fn()
+      const fileItem = createMockFileItem()
+      const { container } = render(<FileListItem {...defaultProps} fileItem={fileItem} onRemove={onRemove} />)
+
+      const deleteButton = container.querySelector('.cursor-pointer')!
+      fireEvent.click(deleteButton)
+
+      expect(onRemove).toHaveBeenCalledTimes(1)
+      expect(onRemove).toHaveBeenCalledWith('file-123')
+    })
+
+    it('should stop propagation when delete button is clicked', () => {
+      const onPreview = vi.fn()
+      const onRemove = vi.fn()
+      const { container } = render(<FileListItem {...defaultProps} onPreview={onPreview} onRemove={onRemove} />)
+
+      const deleteButton = container.querySelector('.cursor-pointer')!
+      fireEvent.click(deleteButton)
+
+      expect(onRemove).toHaveBeenCalledTimes(1)
+      expect(onPreview).not.toHaveBeenCalled()
+    })
+  })
+
+  describe('file type handling', () => {
+    it('should handle files with multiple dots in name', () => {
+      const fileItem = createMockFileItem({
+        file: createMockFile({ name: 'my.document.file.docx' }),
+      })
+      render(<FileListItem {...defaultProps} fileItem={fileItem} />)
+
+      expect(screen.getByText('my.document.file.docx')).toBeInTheDocument()
+      // Extension is lowercase with uppercase CSS class
+      expect(screen.getByText('docx')).toBeInTheDocument()
+    })
+
+    it('should handle files without extension', () => {
+      const fileItem = createMockFileItem({
+        file: createMockFile({ name: 'README' }),
+      })
+      render(<FileListItem {...defaultProps} fileItem={fileItem} />)
+
+      // getFileType returns 'README' when there's no extension (last part after split)
+      expect(screen.getAllByText('README')).toHaveLength(2) // filename and extension
+    })
+
+    it('should handle various file extensions', () => {
+      const extensions = ['txt', 'md', 'json', 'csv', 'xlsx']
+
+      extensions.forEach((ext) => {
+        const fileItem = createMockFileItem({
+          file: createMockFile({ name: `file.${ext}` }),
+        })
+        const { unmount } = render(<FileListItem {...defaultProps} fileItem={fileItem} />)
+        // Extension is rendered in lowercase with uppercase CSS class
+        expect(screen.getByText(ext)).toBeInTheDocument()
+        unmount()
+      })
+    })
+  })
+
+  describe('file size display', () => {
+    it('should display size in KB for small files', () => {
+      const fileItem = createMockFileItem({
+        file: createMockFile({ size: 5 * 1024 }), // 5KB
+      })
+      render(<FileListItem {...defaultProps} fileItem={fileItem} />)
+
+      expect(screen.getByText('5.00 KB')).toBeInTheDocument()
+    })
+
+    it('should display size in MB for larger files', () => {
+      const fileItem = createMockFileItem({
+        file: createMockFile({ size: 5 * 1024 * 1024 }), // 5MB
+      })
+      render(<FileListItem {...defaultProps} fileItem={fileItem} />)
+
+      expect(screen.getByText('5.00 MB')).toBeInTheDocument()
+    })
+
+    it('should display size at threshold (10KB)', () => {
+      const fileItem = createMockFileItem({
+        file: createMockFile({ size: 10 * 1024 }), // 10KB
+      })
+      render(<FileListItem {...defaultProps} fileItem={fileItem} />)
+
+      expect(screen.getByText('10.00 KB')).toBeInTheDocument()
+    })
+  })
+
+  describe('upload progress values', () => {
+    it('should show chart at progress 1', () => {
+      const fileItem = createMockFileItem({ progress: 1 })
+      render(<FileListItem {...defaultProps} fileItem={fileItem} />)
+
+      expect(screen.getByTestId('pie-chart')).toBeInTheDocument()
+    })
+
+    it('should show chart at progress 99', () => {
+      const fileItem = createMockFileItem({ progress: 99 })
+      render(<FileListItem {...defaultProps} fileItem={fileItem} />)
+
+      expect(screen.getByTestId('pie-chart')).toHaveAttribute('data-percentage', '99')
+    })
+
+    it('should not show chart at progress 100', () => {
+      const fileItem = createMockFileItem({ progress: 100 })
+      render(<FileListItem {...defaultProps} fileItem={fileItem} />)
+
+      expect(screen.queryByTestId('pie-chart')).not.toBeInTheDocument()
+    })
+  })
+
+  describe('styling', () => {
+    it('should have proper shadow styling', () => {
+      const { container } = render(<FileListItem {...defaultProps} />)
+
+      const item = container.firstChild as HTMLElement
+      expect(item).toHaveClass('shadow-xs')
+    })
+
+    it('should have proper border styling', () => {
+      const { container } = render(<FileListItem {...defaultProps} />)
+
+      const item = container.firstChild as HTMLElement
+      expect(item).toHaveClass('border', 'border-components-panel-border')
+    })
+
+    it('should truncate long file names', () => {
+      const longFileName = 'this-is-a-very-long-file-name-that-should-be-truncated.pdf'
+      const fileItem = createMockFileItem({
+        file: createMockFile({ name: longFileName }),
+      })
+      render(<FileListItem {...defaultProps} fileItem={fileItem} />)
+
+      const nameElement = screen.getByText(longFileName)
+      expect(nameElement).toHaveClass('truncate')
+    })
+  })
+})

web/app/components/datasets/documents/create-from-pipeline/data-source/local-file/components/file-list-item.tsx

@@ -0,0 +1,85 @@
+import type { CustomFile as File, FileItem } from '@/models/datasets'
+import { RiDeleteBinLine, RiErrorWarningFill } from '@remixicon/react'
+import dynamic from 'next/dynamic'
+import { useMemo } from 'react'
+import DocumentFileIcon from '@/app/components/datasets/common/document-file-icon'
+import { getFileType } from '@/app/components/datasets/common/image-uploader/utils'
+import useTheme from '@/hooks/use-theme'
+import { Theme } from '@/types/app'
+import { cn } from '@/utils/classnames'
+import { formatFileSize } from '@/utils/format'
+import { PROGRESS_ERROR } from '../constants'
+
+const SimplePieChart = dynamic(() => import('@/app/components/base/simple-pie-chart'), { ssr: false })
+
+export type FileListItemProps = {
+  fileItem: FileItem
+  onPreview: (file: File) => void
+  onRemove: (fileID: string) => void
+}
+
+const FileListItem = ({
+  fileItem,
+  onPreview,
+  onRemove,
+}: FileListItemProps) => {
+  const { theme } = useTheme()
+  const chartColor = useMemo(() => theme === Theme.dark ? '#5289ff' : '#296dff', [theme])
+
+  const isUploading = fileItem.progress >= 0 && fileItem.progress < 100
+  const isError = fileItem.progress === PROGRESS_ERROR
+
+  const handleClick = () => {
+    onPreview(fileItem.file)
+  }
+
+  const handleRemove = (e: React.MouseEvent) => {
+    e.stopPropagation()
+    onRemove(fileItem.fileID)
+  }
+
+  return (
+    <div
+      onClick={handleClick}
+      className={cn(
+        'flex h-12 items-center rounded-lg border border-components-panel-border bg-components-panel-on-panel-item-bg shadow-xs shadow-shadow-shadow-4',
+        isError && 'border-state-destructive-border bg-state-destructive-hover',
+      )}
+    >
+      <div className="flex w-12 shrink-0 items-center justify-center">
+        <DocumentFileIcon
+          size="lg"
+          className="shrink-0"
+          name={fileItem.file.name}
+          extension={getFileType(fileItem.file)}
+        />
+      </div>
+      <div className="flex shrink grow flex-col gap-0.5">
+        <div className="flex w-full">
+          <div className="w-0 grow truncate text-xs text-text-secondary">{fileItem.file.name}</div>
+        </div>
+        <div className="w-full truncate text-2xs leading-3 text-text-tertiary">
+          <span className="uppercase">{getFileType(fileItem.file)}</span>
+          <span className="px-1 text-text-quaternary">·</span>
+          <span>{formatFileSize(fileItem.file.size)}</span>
+        </div>
+      </div>
+      <div className="flex w-16 shrink-0 items-center justify-end gap-1 pr-3">
+        {isUploading && (
+          <SimplePieChart percentage={fileItem.progress} stroke={chartColor} fill={chartColor} animationDuration={0} />
+        )}
+        {isError && (
+          <RiErrorWarningFill className="size-4 text-text-destructive" />
+        )}
+        <span
+          className="flex h-6 w-6 cursor-pointer items-center justify-center"
+          onClick={handleRemove}
+        >
+          <RiDeleteBinLine className="size-4 text-text-tertiary" />
+        </span>
+      </div>
+    </div>
+  )
+}
+
+export default FileListItem

web/app/components/datasets/documents/create-from-pipeline/data-source/local-file/components/upload-dropzone.spec.tsx

@@ -0,0 +1,231 @@
+import type { RefObject } from 'react'
+import type { UploadDropzoneProps } from './upload-dropzone'
+import { fireEvent, render, screen } from '@testing-library/react'
+import { beforeEach, describe, expect, it, vi } from 'vitest'
+import UploadDropzone from './upload-dropzone'
+
+// Helper to create mock ref objects for testing
+const createMockRef = <T,>(value: T | null = null): RefObject<T | null> => ({ current: value })
+
+// Mock react-i18next
+vi.mock('react-i18next', () => ({
+  useTranslation: () => ({
+    t: (key: string, options?: { ns?: string }) => {
+      const translations: Record<string, string> = {
+        'stepOne.uploader.button': 'Drag and drop files, or',
+        'stepOne.uploader.buttonSingleFile': 'Drag and drop file, or',
+        'stepOne.uploader.browse': 'Browse',
+        'stepOne.uploader.tip': 'Supports {{supportTypes}}, Max {{size}}MB each, up to {{batchCount}} files at a time, {{totalCount}} files total',
+      }
+      let result = translations[key] || key
+      if (options && typeof options === 'object') {
+        Object.entries(options).forEach(([k, v]) => {
+          result = result.replace(`{{${k}}}`, String(v))
+        })
+      }
+      return result
+    },
+  }),
+}))
+
+describe('UploadDropzone', () => {
+  const defaultProps: UploadDropzoneProps = {
+    dropRef: createMockRef<HTMLDivElement>() as RefObject<HTMLDivElement | null>,
+    dragRef: createMockRef<HTMLDivElement>() as RefObject<HTMLDivElement | null>,
+    fileUploaderRef: createMockRef<HTMLInputElement>() as RefObject<HTMLInputElement | null>,
+    dragging: false,
+    supportBatchUpload: true,
+    supportTypesShowNames: 'PDF, DOCX, TXT',
+    fileUploadConfig: {
+      file_size_limit: 15,
+      batch_count_limit: 5,
+      file_upload_limit: 10,
+    },
+    acceptTypes: ['.pdf', '.docx', '.txt'],
+    onSelectFile: vi.fn(),
+    onFileChange: vi.fn(),
+    allowedExtensions: ['pdf', 'docx', 'txt'],
+  }
+
+  beforeEach(() => {
+    vi.clearAllMocks()
+  })
+
+  describe('rendering', () => {
+    it('should render the dropzone container', () => {
+      const { container } = render(<UploadDropzone {...defaultProps} />)
+
+      const dropzone = container.querySelector('[class*="border-dashed"]')
+      expect(dropzone).toBeInTheDocument()
+    })
+
+    it('should render hidden file input', () => {
+      render(<UploadDropzone {...defaultProps} />)
+
+      const input = document.getElementById('fileUploader') as HTMLInputElement
+      expect(input).toBeInTheDocument()
+      expect(input).toHaveClass('hidden')
+      expect(input).toHaveAttribute('type', 'file')
+    })
+
+    it('should render upload icon', () => {
+      render(<UploadDropzone {...defaultProps} />)
+
+      const icon = document.querySelector('svg')
+      expect(icon).toBeInTheDocument()
+    })
+
+    it('should render browse label when extensions are allowed', () => {
+      render(<UploadDropzone {...defaultProps} />)
+
+      expect(screen.getByText('Browse')).toBeInTheDocument()
+    })
+
+    it('should not render browse label when no extensions allowed', () => {
+      render(<UploadDropzone {...defaultProps} allowedExtensions={[]} />)
+
+      expect(screen.queryByText('Browse')).not.toBeInTheDocument()
+    })
+
+    it('should render file size and count limits', () => {
+      render(<UploadDropzone {...defaultProps} />)
+
+      const tipText = screen.getByText(/Supports.*Max.*15MB/i)
+      expect(tipText).toBeInTheDocument()
+    })
+  })
+
+  describe('file input configuration', () => {
+    it('should allow multiple files when supportBatchUpload is true', () => {
+      render(<UploadDropzone {...defaultProps} supportBatchUpload={true} />)
+
+      const input = document.getElementById('fileUploader') as HTMLInputElement
+      expect(input).toHaveAttribute('multiple')
+    })
+
+    it('should not allow multiple files when supportBatchUpload is false', () => {
+      render(<UploadDropzone {...defaultProps} supportBatchUpload={false} />)
+
+      const input = document.getElementById('fileUploader') as HTMLInputElement
+      expect(input).not.toHaveAttribute('multiple')
+    })
+
+    it('should set accept attribute with correct types', () => {
+      render(<UploadDropzone {...defaultProps} acceptTypes={['.pdf', '.docx']} />)
+
+      const input = document.getElementById('fileUploader') as HTMLInputElement
+      expect(input).toHaveAttribute('accept', '.pdf,.docx')
+    })
+  })
+
+  describe('text content', () => {
+    it('should show batch upload text when supportBatchUpload is true', () => {
+      render(<UploadDropzone {...defaultProps} supportBatchUpload={true} />)
+
+      expect(screen.getByText(/Drag and drop files/i)).toBeInTheDocument()
+    })
+
+    it('should show single file text when supportBatchUpload is false', () => {
+      render(<UploadDropzone {...defaultProps} supportBatchUpload={false} />)
+
+      expect(screen.getByText(/Drag and drop file/i)).toBeInTheDocument()
+    })
+  })
+
+  describe('dragging state', () => {
+    it('should apply dragging styles when dragging is true', () => {
+      const { container } = render(<UploadDropzone {...defaultProps} dragging={true} />)
+
+      const dropzone = container.querySelector('[class*="border-components-dropzone-border-accent"]')
+      expect(dropzone).toBeInTheDocument()
+    })
+
+    it('should render drag overlay when dragging', () => {
+      const dragRef = createMockRef<HTMLDivElement>()
+      render(<UploadDropzone {...defaultProps} dragging={true} dragRef={dragRef as RefObject<HTMLDivElement | null>} />)
+
+      const overlay = document.querySelector('.absolute.left-0.top-0')
+      expect(overlay).toBeInTheDocument()
+    })
+
+    it('should not render drag overlay when not dragging', () => {
+      render(<UploadDropzone {...defaultProps} dragging={false} />)
+
+      const overlay = document.querySelector('.absolute.left-0.top-0')
+      expect(overlay).not.toBeInTheDocument()
+    })
+  })
+
+  describe('event handlers', () => {
+    it('should call onSelectFile when browse label is clicked', () => {
+      const onSelectFile = vi.fn()
+      render(<UploadDropzone {...defaultProps} onSelectFile={onSelectFile} />)
+
+      const browseLabel = screen.getByText('Browse')
+      fireEvent.click(browseLabel)
+
+      expect(onSelectFile).toHaveBeenCalledTimes(1)
+    })
+
+    it('should call onFileChange when files are selected', () => {
+      const onFileChange = vi.fn()
+      render(<UploadDropzone {...defaultProps} onFileChange={onFileChange} />)
+
+      const input = document.getElementById('fileUploader') as HTMLInputElement
+      const file = new File(['content'], 'test.pdf', { type: 'application/pdf' })
+
+      fireEvent.change(input, { target: { files: [file] } })
+
+      expect(onFileChange).toHaveBeenCalledTimes(1)
+    })
+  })
+
+  describe('refs', () => {
+    it('should attach dropRef to drop container', () => {
+      const dropRef = createMockRef<HTMLDivElement>()
+      render(<UploadDropzone {...defaultProps} dropRef={dropRef as RefObject<HTMLDivElement | null>} />)
+
+      expect(dropRef.current).toBeInstanceOf(HTMLDivElement)
+    })
+
+    it('should attach fileUploaderRef to input element', () => {
+      const fileUploaderRef = createMockRef<HTMLInputElement>()
+      render(<UploadDropzone {...defaultProps} fileUploaderRef={fileUploaderRef as RefObject<HTMLInputElement | null>} />)
+
+      expect(fileUploaderRef.current).toBeInstanceOf(HTMLInputElement)
+    })
+
+    it('should attach dragRef to overlay when dragging', () => {
+      const dragRef = createMockRef<HTMLDivElement>()
+      render(<UploadDropzone {...defaultProps} dragging={true} dragRef={dragRef as RefObject<HTMLDivElement | null>} />)
+
+      expect(dragRef.current).toBeInstanceOf(HTMLDivElement)
+    })
+  })
+
+  describe('styling', () => {
+    it('should have base dropzone styling', () => {
+      const { container } = render(<UploadDropzone {...defaultProps} />)
+
+      const dropzone = container.querySelector('[class*="border-dashed"]')
+      expect(dropzone).toBeInTheDocument()
+      expect(dropzone).toHaveClass('rounded-xl')
+    })
+
+    it('should have cursor-pointer on browse label', () => {
+      render(<UploadDropzone {...defaultProps} />)
+
+      const browseLabel = screen.getByText('Browse')
+      expect(browseLabel).toHaveClass('cursor-pointer')
+    })
+  })
+
+  describe('accessibility', () => {
+    it('should have an accessible file input', () => {
+      render(<UploadDropzone {...defaultProps} />)
+
+      const input = document.getElementById('fileUploader') as HTMLInputElement
+      expect(input).toHaveAttribute('id', 'fileUploader')
+    })
+  })
+})

web/app/components/datasets/documents/create-from-pipeline/data-source/local-file/components/upload-dropzone.tsx

@@ -0,0 +1,83 @@
+import type { ChangeEvent, RefObject } from 'react'
+import { RiUploadCloud2Line } from '@remixicon/react'
+import { useTranslation } from 'react-i18next'
+import { cn } from '@/utils/classnames'
+
+type FileUploadConfig = {
+  file_size_limit: number
+  batch_count_limit: number
+  file_upload_limit: number
+}
+
+export type UploadDropzoneProps = {
+  dropRef: RefObject<HTMLDivElement | null>
+  dragRef: RefObject<HTMLDivElement | null>
+  fileUploaderRef: RefObject<HTMLInputElement | null>
+  dragging: boolean
+  supportBatchUpload: boolean
+  supportTypesShowNames: string
+  fileUploadConfig: FileUploadConfig
+  acceptTypes: string[]
+  onSelectFile: () => void
+  onFileChange: (e: ChangeEvent<HTMLInputElement>) => void
+  allowedExtensions: string[]
+}
+
+const UploadDropzone = ({
+  dropRef,
+  dragRef,
+  fileUploaderRef,
+  dragging,
+  supportBatchUpload,
+  supportTypesShowNames,
+  fileUploadConfig,
+  acceptTypes,
+  onSelectFile,
+  onFileChange,
+  allowedExtensions,
+}: UploadDropzoneProps) => {
+  const { t } = useTranslation()
+
+  return (
+    <>
+      <input
+        ref={fileUploaderRef}
+        id="fileUploader"
+        className="hidden"
+        type="file"
+        multiple={supportBatchUpload}
+        accept={acceptTypes.join(',')}
+        onChange={onFileChange}
+      />
+      <div
+        ref={dropRef}
+        className={cn(
+          'relative box-border flex min-h-20 flex-col items-center justify-center gap-1 rounded-xl border border-dashed border-components-dropzone-border bg-components-dropzone-bg px-4 py-3 text-xs leading-4 text-text-tertiary',
+          dragging && 'border-components-dropzone-border-accent bg-components-dropzone-bg-accent',
+        )}
+      >
+        <div className="flex min-h-5 items-center justify-center text-sm leading-4 text-text-secondary">
+          <RiUploadCloud2Line className="mr-2 size-5" />
+          <span>
+            {supportBatchUpload ? t('stepOne.uploader.button', { ns: 'datasetCreation' }) : t('stepOne.uploader.buttonSingleFile', { ns: 'datasetCreation' })}
+            {allowedExtensions.length > 0 && (
+              <label className="ml-1 cursor-pointer text-text-accent" onClick={onSelectFile}>{t('stepOne.uploader.browse', { ns: 'datasetCreation' })}</label>
+            )}
+          </span>
+        </div>
+        <div>
+          {t('stepOne.uploader.tip', {
+            ns: 'datasetCreation',
+            size: fileUploadConfig.file_size_limit,
+            supportTypes: supportTypesShowNames,
+            batchCount: fileUploadConfig.batch_count_limit,
+            totalCount: fileUploadConfig.file_upload_limit,
+          })}
+        </div>
+        {dragging && <div ref={dragRef} className="absolute left-0 top-0 h-full w-full" />}
+      </div>
+    </>
+  )
+}
+
+export default UploadDropzone

web/app/components/datasets/documents/create-from-pipeline/data-source/local-file/constants.ts

@@ -0,0 +1,3 @@
+export const PROGRESS_NOT_STARTED = -1
+export const PROGRESS_ERROR = -2
+export const PROGRESS_COMPLETE = 100

web/app/components/datasets/documents/create-from-pipeline/data-source/local-file/hooks/use-local-file-upload.spec.tsx

@@ -0,0 +1,911 @@
+import type { ReactNode } from 'react'
+import type { CustomFile, FileItem } from '@/models/datasets'
+import { act, render, renderHook, waitFor } from '@testing-library/react'
+import { beforeEach, describe, expect, it, vi } from 'vitest'
+import { PROGRESS_ERROR, PROGRESS_NOT_STARTED } from '../constants'
+
+// Mock notify function - defined before mocks
+const mockNotify = vi.fn()
+const mockClose = vi.fn()
+
+// Mock ToastContext with factory function
+vi.mock('@/app/components/base/toast', async () => {
+  const { createContext, useContext } = await import('use-context-selector')
+  const context = createContext({ notify: mockNotify, close: mockClose })
+  return {
+    ToastContext: context,
+    useToastContext: () => useContext(context),
+  }
+})
+
+// Mock file uploader utils
+vi.mock('@/app/components/base/file-uploader/utils', () => ({
+  getFileUploadErrorMessage: (e: Error, defaultMsg: string) => e.message || defaultMsg,
+}))
+
+// Mock format utils used by the shared hook
+vi.mock('@/utils/format', () => ({
+  getFileExtension: (filename: string) => {
+    const parts = filename.split('.')
+    return parts[parts.length - 1] || ''
+  },
+}))
+
+// Mock react-i18next
+vi.mock('react-i18next', () => ({
+  useTranslation: () => ({
+    t: (key: string) => key,
+  }),
+}))
+
+// Mock locale context
+vi.mock('@/context/i18n', () => ({
+  useLocale: () => 'en-US',
+}))
+
+// Mock i18n config
+vi.mock('@/i18n-config/language', () => ({
+  LanguagesSupported: ['en-US', 'zh-Hans'],
+}))
+
+// Mock config
+vi.mock('@/config', () => ({
+  IS_CE_EDITION: false,
+}))
+
+// Mock store functions
+const mockSetLocalFileList = vi.fn()
+const mockSetCurrentLocalFile = vi.fn()
+const mockGetState = vi.fn(() => ({
+  setLocalFileList: mockSetLocalFileList,
+  setCurrentLocalFile: mockSetCurrentLocalFile,
+}))
+const mockStore = { getState: mockGetState }
+
+vi.mock('../../store', () => ({
+  useDataSourceStoreWithSelector: vi.fn((selector: (state: { localFileList: FileItem[] }) => FileItem[]) =>
+    selector({ localFileList: [] }),
+  ),
+  useDataSourceStore: vi.fn(() => mockStore),
+}))
+
+// Mock file upload config
+vi.mock('@/service/use-common', () => ({
+  useFileUploadConfig: vi.fn(() => ({
+    data: {
+      file_size_limit: 15,
+      batch_count_limit: 5,
+      file_upload_limit: 10,
+    },
+  })),
+  // Required by the shared useFileUpload hook
+  useFileSupportTypes: vi.fn(() => ({
+    data: {
+      allowed_extensions: ['pdf', 'docx', 'txt'],
+    },
+  })),
+}))
+
+// Mock upload service
+const mockUpload = vi.fn()
+vi.mock('@/service/base', () => ({
+  upload: (...args: unknown[]) => mockUpload(...args),
+}))
+
+// Import after all mocks are set up
+const { useLocalFileUpload } = await import('./use-local-file-upload')
+const { ToastContext } = await import('@/app/components/base/toast')
+
+const createWrapper = () => {
+  return ({ children }: { children: ReactNode }) => (
+    <ToastContext.Provider value={{ notify: mockNotify, close: mockClose }}>
+      {children}
+    </ToastContext.Provider>
+  )
+}
+
+describe('useLocalFileUpload', () => {
+  beforeEach(() => {
+    vi.clearAllMocks()
+    mockUpload.mockReset()
+  })
+
+  describe('initialization', () => {
+    it('should initialize with default values', () => {
+      const { result } = renderHook(
+        () => useLocalFileUpload({ allowedExtensions: ['pdf', 'docx'] }),
+        { wrapper: createWrapper() },
+      )
+
+      expect(result.current.dragging).toBe(false)
+      expect(result.current.localFileList).toEqual([])
+      expect(result.current.hideUpload).toBe(false)
+    })
+
+    it('should create refs for dropzone, drag area, and file uploader', () => {
+      const { result } = renderHook(
+        () => useLocalFileUpload({ allowedExtensions: ['pdf'] }),
+        { wrapper: createWrapper() },
+      )
+
+      expect(result.current.dropRef).toBeDefined()
+      expect(result.current.dragRef).toBeDefined()
+      expect(result.current.fileUploaderRef).toBeDefined()
+    })
+
+    it('should compute acceptTypes from allowedExtensions', () => {
+      const { result } = renderHook(
+        () => useLocalFileUpload({ allowedExtensions: ['pdf', 'docx', 'txt'] }),
+        { wrapper: createWrapper() },
+      )
+
+      expect(result.current.acceptTypes).toEqual(['.pdf', '.docx', '.txt'])
+    })
+
+    it('should compute supportTypesShowNames correctly', () => {
+      const { result } = renderHook(
+        () => useLocalFileUpload({ allowedExtensions: ['pdf', 'docx', 'md'] }),
+        { wrapper: createWrapper() },
+      )
+
+      expect(result.current.supportTypesShowNames).toContain('PDF')
+      expect(result.current.supportTypesShowNames).toContain('DOCX')
+      expect(result.current.supportTypesShowNames).toContain('MARKDOWN')
+    })
+
+    it('should provide file upload config with defaults', () => {
+      const { result } = renderHook(
+        () => useLocalFileUpload({ allowedExtensions: ['pdf'] }),
+        { wrapper: createWrapper() },
+      )
+
+      expect(result.current.fileUploadConfig.file_size_limit).toBe(15)
+      expect(result.current.fileUploadConfig.batch_count_limit).toBe(5)
+      expect(result.current.fileUploadConfig.file_upload_limit).toBe(10)
+    })
+  })
+
+  describe('supportBatchUpload option', () => {
+    it('should use batch limits when supportBatchUpload is true', () => {
+      const { result } = renderHook(
+        () => useLocalFileUpload({ allowedExtensions: ['pdf'], supportBatchUpload: true }),
+        { wrapper: createWrapper() },
+      )
+
+      expect(result.current.fileUploadConfig.batch_count_limit).toBe(5)
+      expect(result.current.fileUploadConfig.file_upload_limit).toBe(10)
+    })
+
+    it('should use single file limits when supportBatchUpload is false', () => {
+      const { result } = renderHook(
+        () => useLocalFileUpload({ allowedExtensions: ['pdf'], supportBatchUpload: false }),
+        { wrapper: createWrapper() },
+      )
+
+      expect(result.current.fileUploadConfig.batch_count_limit).toBe(1)
+      expect(result.current.fileUploadConfig.file_upload_limit).toBe(1)
+    })
+  })
+
+  describe('selectHandle', () => {
+    it('should trigger file input click', () => {
+      const { result } = renderHook(
+        () => useLocalFileUpload({ allowedExtensions: ['pdf'] }),
+        { wrapper: createWrapper() },
+      )
+
+      const mockClick = vi.fn()
+      const mockInput = { click: mockClick } as unknown as HTMLInputElement
+      Object.defineProperty(result.current.fileUploaderRef, 'current', {
+        value: mockInput,
+        writable: true,
+      })
+
+      act(() => {
+        result.current.selectHandle()
+      })
+
+      expect(mockClick).toHaveBeenCalled()
+    })
+
+    it('should handle null fileUploaderRef gracefully', () => {
+      const { result } = renderHook(
+        () => useLocalFileUpload({ allowedExtensions: ['pdf'] }),
+        { wrapper: createWrapper() },
+      )
+
+      expect(() => {
+        act(() => {
+          result.current.selectHandle()
+        })
+      }).not.toThrow()
+    })
+  })
+
+  describe('removeFile', () => {
+    it('should remove file from list', () => {
+      const { result } = renderHook(
+        () => useLocalFileUpload({ allowedExtensions: ['pdf'] }),
+        { wrapper: createWrapper() },
+      )
+
+      act(() => {
+        result.current.removeFile('file-id-123')
+      })
+
+      expect(mockSetLocalFileList).toHaveBeenCalled()
+    })
+
+    it('should clear file input value when removing', () => {
+      const { result } = renderHook(
+        () => useLocalFileUpload({ allowedExtensions: ['pdf'] }),
+        { wrapper: createWrapper() },
+      )
+
+      const mockInput = { value: 'some-file.pdf' } as HTMLInputElement
+      Object.defineProperty(result.current.fileUploaderRef, 'current', {
+        value: mockInput,
+        writable: true,
+      })
+
+      act(() => {
+        result.current.removeFile('file-id')
+      })
+
+      expect(mockInput.value).toBe('')
+    })
+  })
+
+  describe('handlePreview', () => {
+    it('should set current local file when file has id', () => {
+      const { result } = renderHook(
+        () => useLocalFileUpload({ allowedExtensions: ['pdf'] }),
+        { wrapper: createWrapper() },
+      )
+
+      const mockFile = { id: 'file-123', name: 'test.pdf', size: 1024 }
+
+      act(() => {
+        result.current.handlePreview(mockFile as unknown as CustomFile)
+      })
+
+      expect(mockSetCurrentLocalFile).toHaveBeenCalledWith(mockFile)
+    })
+
+    it('should not set current file when file has no id', () => {
+      const { result } = renderHook(
+        () => useLocalFileUpload({ allowedExtensions: ['pdf'] }),
+        { wrapper: createWrapper() },
+      )
+
+      const mockFile = { name: 'test.pdf', size: 1024 }
+
+      act(() => {
+        result.current.handlePreview(mockFile as unknown as CustomFile)
+      })
+
+      expect(mockSetCurrentLocalFile).not.toHaveBeenCalled()
+    })
+  })
+
+  describe('fileChangeHandle', () => {
+    it('should handle valid files', async () => {
+      mockUpload.mockResolvedValue({ id: 'uploaded-id' })
+
+      const { result } = renderHook(
+        () => useLocalFileUpload({ allowedExtensions: ['pdf'] }),
+        { wrapper: createWrapper() },
+      )
+
+      const mockFile = new File(['content'], 'test.pdf', { type: 'application/pdf' })
+      const event = {
+        target: {
+          files: [mockFile],
+        },
+      } as unknown as React.ChangeEvent<HTMLInputElement>
+
+      act(() => {
+        result.current.fileChangeHandle(event)
+      })
+
+      await waitFor(() => {
+        expect(mockSetLocalFileList).toHaveBeenCalled()
+      })
+    })
+
+    it('should handle empty file list', () => {
+      const { result } = renderHook(
+        () => useLocalFileUpload({ allowedExtensions: ['pdf'] }),
+        { wrapper: createWrapper() },
+      )
+
+      const event = {
+        target: {
+          files: null,
+        },
+      } as unknown as React.ChangeEvent<HTMLInputElement>
+
+      act(() => {
+        result.current.fileChangeHandle(event)
+      })
+
+      expect(mockSetLocalFileList).not.toHaveBeenCalled()
+    })
+
+    it('should reject files with invalid type', () => {
+      const { result } = renderHook(
+        () => useLocalFileUpload({ allowedExtensions: ['pdf'] }),
+        { wrapper: createWrapper() },
+      )
+
+      const mockFile = new File(['content'], 'test.exe', { type: 'application/exe' })
+      const event = {
+        target: {
+          files: [mockFile],
+        },
+      } as unknown as React.ChangeEvent<HTMLInputElement>
+
+      act(() => {
+        result.current.fileChangeHandle(event)
+      })
+
+      expect(mockNotify).toHaveBeenCalledWith(
+        expect.objectContaining({ type: 'error' }),
+      )
+    })
+
+    it('should reject files exceeding size limit', () => {
+      const { result } = renderHook(
+        () => useLocalFileUpload({ allowedExtensions: ['pdf'] }),
+        { wrapper: createWrapper() },
+      )
+
+      // Create a mock file larger than 15MB
+      const largeSize = 20 * 1024 * 1024
+      const mockFile = new File([''], 'large.pdf', { type: 'application/pdf' })
+      Object.defineProperty(mockFile, 'size', { value: largeSize })
+
+      const event = {
+        target: {
+          files: [mockFile],
+        },
+      } as unknown as React.ChangeEvent<HTMLInputElement>
+
+      act(() => {
+        result.current.fileChangeHandle(event)
+      })
+
+      expect(mockNotify).toHaveBeenCalledWith(
+        expect.objectContaining({ type: 'error' }),
+      )
+    })
+
+    it('should limit files to batch count limit', async () => {
+      mockUpload.mockResolvedValue({ id: 'uploaded-id' })
+
+      const { result } = renderHook(
+        () => useLocalFileUpload({ allowedExtensions: ['pdf'] }),
+        { wrapper: createWrapper() },
+      )
+
+      // Create 10 files but batch limit is 5
+      const files = Array.from({ length: 10 }, (_, i) =>
+        new File(['content'], `file${i}.pdf`, { type: 'application/pdf' }))
+
+      const event = {
+        target: {
+          files,
+        },
+      } as unknown as React.ChangeEvent<HTMLInputElement>
+
+      act(() => {
+        result.current.fileChangeHandle(event)
+      })
+
+      await waitFor(() => {
+        expect(mockSetLocalFileList).toHaveBeenCalled()
+      })
+
+      // Should only process first 5 files (batch_count_limit)
+      const firstCall = mockSetLocalFileList.mock.calls[0]
+      expect(firstCall[0].length).toBeLessThanOrEqual(5)
+    })
+  })
+
+  describe('upload handling', () => {
+    it('should handle successful upload', async () => {
+      const uploadedResponse = { id: 'server-file-id' }
+      mockUpload.mockResolvedValue(uploadedResponse)
+
+      const { result } = renderHook(
+        () => useLocalFileUpload({ allowedExtensions: ['pdf'] }),
+        { wrapper: createWrapper() },
+      )
+
+      const mockFile = new File(['content'], 'test.pdf', { type: 'application/pdf' })
+      const event = {
+        target: {
+          files: [mockFile],
+        },
+      } as unknown as React.ChangeEvent<HTMLInputElement>
+
+      act(() => {
+        result.current.fileChangeHandle(event)
+      })
+
+      await waitFor(() => {
+        expect(mockUpload).toHaveBeenCalled()
+      })
+    })
+
+    it('should handle upload error', async () => {
+      mockUpload.mockRejectedValue(new Error('Upload failed'))
+
+      const { result } = renderHook(
+        () => useLocalFileUpload({ allowedExtensions: ['pdf'] }),
+        { wrapper: createWrapper() },
+      )
+
+      const mockFile = new File(['content'], 'test.pdf', { type: 'application/pdf' })
+      const event = {
+        target: {
+          files: [mockFile],
+        },
+      } as unknown as React.ChangeEvent<HTMLInputElement>
+
+      act(() => {
+        result.current.fileChangeHandle(event)
+      })
+
+      await waitFor(() => {
+        expect(mockNotify).toHaveBeenCalledWith(
+          expect.objectContaining({ type: 'error' }),
+        )
+      })
+    })
+
+    it('should call upload with correct parameters', async () => {
+      mockUpload.mockResolvedValue({ id: 'file-id' })
+
+      const { result } = renderHook(
+        () => useLocalFileUpload({ allowedExtensions: ['pdf'] }),
+        { wrapper: createWrapper() },
+      )
+
+      const mockFile = new File(['content'], 'test.pdf', { type: 'application/pdf' })
+      const event = {
+        target: {
+          files: [mockFile],
+        },
+      } as unknown as React.ChangeEvent<HTMLInputElement>
+
+      act(() => {
+        result.current.fileChangeHandle(event)
+      })
+
+      await waitFor(() => {
+        expect(mockUpload).toHaveBeenCalledWith(
+          expect.objectContaining({
+            xhr: expect.any(XMLHttpRequest),
+            data: expect.any(FormData),
+          }),
+          false,
+          undefined,
+          '?source=datasets',
+        )
+      })
+    })
+  })
+
+  describe('extension mapping', () => {
+    it('should map md to markdown', () => {
+      const { result } = renderHook(
+        () => useLocalFileUpload({ allowedExtensions: ['md'] }),
+        { wrapper: createWrapper() },
+      )
+
+      expect(result.current.supportTypesShowNames).toContain('MARKDOWN')
+    })
+
+    it('should map htm to html', () => {
+      const { result } = renderHook(
+        () => useLocalFileUpload({ allowedExtensions: ['htm'] }),
+        { wrapper: createWrapper() },
+      )
+
+      expect(result.current.supportTypesShowNames).toContain('HTML')
+    })
+
+    it('should preserve unmapped extensions', () => {
+      const { result } = renderHook(
+        () => useLocalFileUpload({ allowedExtensions: ['pdf', 'txt'] }),
+        { wrapper: createWrapper() },
+      )
+
+      expect(result.current.supportTypesShowNames).toContain('PDF')
+      expect(result.current.supportTypesShowNames).toContain('TXT')
+    })
+
+    it('should remove duplicate extensions', () => {
+      const { result } = renderHook(
+        () => useLocalFileUpload({ allowedExtensions: ['pdf', 'pdf', 'PDF'] }),
+        { wrapper: createWrapper() },
+      )
+
+      const count = (result.current.supportTypesShowNames.match(/PDF/g) || []).length
+      expect(count).toBe(1)
+    })
+  })
+
+  describe('drag and drop handlers', () => {
+    // Helper component that renders with the hook and connects refs
+    const TestDropzone = ({ allowedExtensions, supportBatchUpload = true }: {
+      allowedExtensions: string[]
+      supportBatchUpload?: boolean
+    }) => {
+      const {
+        dropRef,
+        dragRef,
+        dragging,
+      } = useLocalFileUpload({ allowedExtensions, supportBatchUpload })
+
+      return (
+        <div>
+          <div ref={dropRef} data-testid="dropzone">
+            {dragging && <div ref={dragRef} data-testid="drag-overlay" />}
+          </div>
+          <span data-testid="dragging">{String(dragging)}</span>
+        </div>
+      )
+    }
+
+    it('should set dragging true on dragenter', async () => {
+      const { getByTestId } = await act(async () =>
+        render(
+          <ToastContext.Provider value={{ notify: mockNotify, close: mockClose }}>
+            <TestDropzone allowedExtensions={['pdf']} />
+          </ToastContext.Provider>,
+        ),
+      )
+
+      const dropzone = getByTestId('dropzone')
+
+      await act(async () => {
+        const dragEnterEvent = new Event('dragenter', { bubbles: true, cancelable: true })
+        dropzone.dispatchEvent(dragEnterEvent)
+      })
+
+      expect(getByTestId('dragging').textContent).toBe('true')
+    })
+
+    it('should handle dragover event', async () => {
+      const { getByTestId } = await act(async () =>
+        render(
+          <ToastContext.Provider value={{ notify: mockNotify, close: mockClose }}>
+            <TestDropzone allowedExtensions={['pdf']} />
+          </ToastContext.Provider>,
+        ),
+      )
+
+      const dropzone = getByTestId('dropzone')
+
+      await act(async () => {
+        const dragOverEvent = new Event('dragover', { bubbles: true, cancelable: true })
+        dropzone.dispatchEvent(dragOverEvent)
+      })
+
+      // dragover should not throw
+      expect(dropzone).toBeInTheDocument()
+    })
+
+    it('should set dragging false on dragleave from drag overlay', async () => {
+      const { getByTestId, queryByTestId } = await act(async () =>
+        render(
+          <ToastContext.Provider value={{ notify: mockNotify, close: mockClose }}>
+            <TestDropzone allowedExtensions={['pdf']} />
+          </ToastContext.Provider>,
+        ),
+      )
+
+      const dropzone = getByTestId('dropzone')
+
+      // First trigger dragenter to set dragging true
+      await act(async () => {
+        const dragEnterEvent = new Event('dragenter', { bubbles: true, cancelable: true })
+        dropzone.dispatchEvent(dragEnterEvent)
+      })
+
+      expect(getByTestId('dragging').textContent).toBe('true')
+
+      // Now the drag overlay should be rendered
+      const dragOverlay = queryByTestId('drag-overlay')
+      if (dragOverlay) {
+        await act(async () => {
+          const dragLeaveEvent = new Event('dragleave', { bubbles: true, cancelable: true })
+          Object.defineProperty(dragLeaveEvent, 'target', { value: dragOverlay })
+          dropzone.dispatchEvent(dragLeaveEvent)
+        })
+      }
+    })
+
+    it('should handle drop with files', async () => {
+      mockUpload.mockResolvedValue({ id: 'uploaded-id' })
+
+      const { getByTestId } = await act(async () =>
+        render(
+          <ToastContext.Provider value={{ notify: mockNotify, close: mockClose }}>
+            <TestDropzone allowedExtensions={['pdf']} />
+          </ToastContext.Provider>,
+        ),
+      )
+
+      const dropzone = getByTestId('dropzone')
+      const mockFile = new File(['content'], 'test.pdf', { type: 'application/pdf' })
+
+      await act(async () => {
+        const dropEvent = new Event('drop', { bubbles: true, cancelable: true }) as Event & {
+          dataTransfer: { items: DataTransferItem[], files: File[] } | null
+        }
+        // Mock dataTransfer with items array (used by the shared hook for directory traversal)
+        dropEvent.dataTransfer = {
+          items: [{
+            kind: 'file',
+            getAsFile: () => mockFile,
+          }] as unknown as DataTransferItem[],
+          files: [mockFile],
+        }
+        dropzone.dispatchEvent(dropEvent)
+      })
+
+      await waitFor(() => {
+        expect(mockSetLocalFileList).toHaveBeenCalled()
+      })
+    })
+
+    it('should handle drop without dataTransfer', async () => {
+      const { getByTestId } = await act(async () =>
+        render(
+          <ToastContext.Provider value={{ notify: mockNotify, close: mockClose }}>
+            <TestDropzone allowedExtensions={['pdf']} />
+          </ToastContext.Provider>,
+        ),
+      )
+
+      const dropzone = getByTestId('dropzone')
+      mockSetLocalFileList.mockClear()
+
+      await act(async () => {
+        const dropEvent = new Event('drop', { bubbles: true, cancelable: true }) as Event & { dataTransfer: { files: File[] } | null }
+        dropEvent.dataTransfer = null
+        dropzone.dispatchEvent(dropEvent)
+      })
+
+      // Should not upload when no dataTransfer
+      expect(mockSetLocalFileList).not.toHaveBeenCalled()
+    })
+
+    it('should limit to single file on drop when supportBatchUpload is false', async () => {
+      mockUpload.mockResolvedValue({ id: 'uploaded-id' })
+
+      const { getByTestId } = await act(async () =>
+        render(
+          <ToastContext.Provider value={{ notify: mockNotify, close: mockClose }}>
+            <TestDropzone allowedExtensions={['pdf']} supportBatchUpload={false} />
+          </ToastContext.Provider>,
+        ),
+      )
+
+      const dropzone = getByTestId('dropzone')
+      const files = [
+        new File(['content1'], 'test1.pdf', { type: 'application/pdf' }),
+        new File(['content2'], 'test2.pdf', { type: 'application/pdf' }),
+      ]
+
+      await act(async () => {
+        const dropEvent = new Event('drop', { bubbles: true, cancelable: true }) as Event & {
+          dataTransfer: { items: DataTransferItem[], files: File[] } | null
+        }
+        // Mock dataTransfer with items array (used by the shared hook for directory traversal)
+        dropEvent.dataTransfer = {
+          items: files.map(f => ({
+            kind: 'file',
+            getAsFile: () => f,
+          })) as unknown as DataTransferItem[],
+          files,
+        }
+        dropzone.dispatchEvent(dropEvent)
+      })
+
+      await waitFor(() => {
+        expect(mockSetLocalFileList).toHaveBeenCalled()
+        // Should only have 1 file (limited by supportBatchUpload: false)
+        const callArgs = mockSetLocalFileList.mock.calls[0][0]
+        expect(callArgs.length).toBe(1)
+      })
+    })
+  })
+
+  describe('file upload limit', () => {
+    it('should reject files exceeding total file upload limit', async () => {
+      // Mock store to return existing files
+      const { useDataSourceStoreWithSelector } = vi.mocked(await import('../../store'))
+      const existingFiles: FileItem[] = Array.from({ length: 8 }, (_, i) => ({
+        fileID: `existing-${i}`,
+        file: { name: `existing-${i}.pdf`, size: 1024 } as CustomFile,
+        progress: 100,
+      }))
+      vi.mocked(useDataSourceStoreWithSelector).mockImplementation(selector =>
+        selector({ localFileList: existingFiles } as Parameters<typeof selector>[0]),
+      )
+
+      const { result } = renderHook(
+        () => useLocalFileUpload({ allowedExtensions: ['pdf'] }),
+        { wrapper: createWrapper() },
+      )
+
+      // Try to add 5 more files when limit is 10 and we already have 8
+      const files = Array.from({ length: 5 }, (_, i) =>
+        new File(['content'], `new-${i}.pdf`, { type: 'application/pdf' }))
+
+      const event = {
+        target: { files },
+      } as unknown as React.ChangeEvent<HTMLInputElement>
+
+      act(() => {
+        result.current.fileChangeHandle(event)
+      })
+
+      // Should show error about files number limit
+      expect(mockNotify).toHaveBeenCalledWith(
+        expect.objectContaining({ type: 'error' }),
+      )
+
+      // Reset mock for other tests
+      vi.mocked(useDataSourceStoreWithSelector).mockImplementation(selector =>
+        selector({ localFileList: [] as FileItem[] } as Parameters<typeof selector>[0]),
+      )
+    })
+  })
+
+  describe('upload progress tracking', () => {
+    it('should track upload progress', async () => {
+      let progressCallback: ((e: ProgressEvent) => void) | undefined
+
+      mockUpload.mockImplementation(async (options: { onprogress: (e: ProgressEvent) => void }) => {
+        progressCallback = options.onprogress
+        return { id: 'uploaded-id' }
+      })
+
+      const { result } = renderHook(
+        () => useLocalFileUpload({ allowedExtensions: ['pdf'] }),
+        { wrapper: createWrapper() },
+      )
+
+      const mockFile = new File(['content'], 'test.pdf', { type: 'application/pdf' })
+      const event = {
+        target: { files: [mockFile] },
+      } as unknown as React.ChangeEvent<HTMLInputElement>
+
+      act(() => {
+        result.current.fileChangeHandle(event)
+      })
+
+      await waitFor(() => {
+        expect(mockUpload).toHaveBeenCalled()
+      })
+
+      // Simulate progress event
+      if (progressCallback) {
+        act(() => {
+          progressCallback!({
+            lengthComputable: true,
+            loaded: 50,
+            total: 100,
+          } as ProgressEvent)
+        })
+
+        expect(mockSetLocalFileList).toHaveBeenCalled()
+      }
+    })
+
+    it('should not update progress when not lengthComputable', async () => {
+      let progressCallback: ((e: ProgressEvent) => void) | undefined
+      const uploadCallCount = { value: 0 }
+
+      mockUpload.mockImplementation(async (options: { onprogress: (e: ProgressEvent) => void }) => {
+        progressCallback = options.onprogress
+        uploadCallCount.value++
+        return { id: 'uploaded-id' }
+      })
+
+      const { result } = renderHook(
+        () => useLocalFileUpload({ allowedExtensions: ['pdf'] }),
+        { wrapper: createWrapper() },
+      )
+
+      const mockFile = new File(['content'], 'test.pdf', { type: 'application/pdf' })
+      const event = {
+        target: { files: [mockFile] },
+      } as unknown as React.ChangeEvent<HTMLInputElement>
+
+      mockSetLocalFileList.mockClear()
+
+      act(() => {
+        result.current.fileChangeHandle(event)
+      })
+
+      await waitFor(() => {
+        expect(mockUpload).toHaveBeenCalled()
+      })
+
+      const callsBeforeProgress = mockSetLocalFileList.mock.calls.length
+
+      // Simulate progress event without lengthComputable
+      if (progressCallback) {
+        act(() => {
+          progressCallback!({
+            lengthComputable: false,
+            loaded: 50,
+            total: 100,
+          } as ProgressEvent)
+        })
+
+        // Should not have additional calls
+        expect(mockSetLocalFileList.mock.calls.length).toBe(callsBeforeProgress)
+      }
+    })
+  })
+
+  describe('file progress constants', () => {
+    it('should use PROGRESS_NOT_STARTED for new files', async () => {
+      mockUpload.mockResolvedValue({ id: 'file-id' })
+
+      const { result } = renderHook(
+        () => useLocalFileUpload({ allowedExtensions: ['pdf'] }),
+        { wrapper: createWrapper() },
+      )
+
+      const mockFile = new File(['content'], 'test.pdf', { type: 'application/pdf' })
+      const event = {
+        target: {
+          files: [mockFile],
+        },
+      } as unknown as React.ChangeEvent<HTMLInputElement>
+
+      act(() => {
+        result.current.fileChangeHandle(event)
+      })
+
+      await waitFor(() => {
+        const callArgs = mockSetLocalFileList.mock.calls[0][0]
+        expect(callArgs[0].progress).toBe(PROGRESS_NOT_STARTED)
+      })
+    })
+
+    it('should set PROGRESS_ERROR on upload failure', async () => {
+      mockUpload.mockRejectedValue(new Error('Upload failed'))
+
+      const { result } = renderHook(
+        () => useLocalFileUpload({ allowedExtensions: ['pdf'] }),
+        { wrapper: createWrapper() },
+      )
+
+      const mockFile = new File(['content'], 'test.pdf', { type: 'application/pdf' })
+      const event = {
+        target: {
+          files: [mockFile],
+        },
+      } as unknown as React.ChangeEvent<HTMLInputElement>
+
+      act(() => {
+        result.current.fileChangeHandle(event)
+      })
+
+      await waitFor(() => {
+        const calls = mockSetLocalFileList.mock.calls
+        const lastCall = calls[calls.length - 1][0]
+        expect(lastCall.some((f: FileItem) => f.progress === PROGRESS_ERROR)).toBe(true)
+      })
+    })
+  })
+})

web/app/components/datasets/documents/create-from-pipeline/data-source/local-file/hooks/use-local-file-upload.ts

@@ -0,0 +1,105 @@
+import type { CustomFile as File, FileItem } from '@/models/datasets'
+import { produce } from 'immer'
+import { useCallback, useRef } from 'react'
+import { useFileUpload } from '@/app/components/datasets/create/file-uploader/hooks/use-file-upload'
+import { useDataSourceStore, useDataSourceStoreWithSelector } from '../../store'
+
+export type UseLocalFileUploadOptions = {
+  allowedExtensions: string[]
+  supportBatchUpload?: boolean
+}
+
+/**
+ * Hook for handling local file uploads in the create-from-pipeline flow.
+ * This is a thin wrapper around the generic useFileUpload hook that provides
+ * Zustand store integration for state management.
+ */
+export const useLocalFileUpload = ({
+  allowedExtensions,
+  supportBatchUpload = true,
+}: UseLocalFileUploadOptions) => {
+  const localFileList = useDataSourceStoreWithSelector(state => state.localFileList)
+  const dataSourceStore = useDataSourceStore()
+  const fileListRef = useRef<FileItem[]>([])
+
+  // Sync fileListRef with localFileList for internal tracking
+  fileListRef.current = localFileList
+
+  const prepareFileList = useCallback((files: FileItem[]) => {
+    const { setLocalFileList } = dataSourceStore.getState()
+    setLocalFileList(files)
+    fileListRef.current = files
+  }, [dataSourceStore])
+
+  const onFileUpdate = useCallback((fileItem: FileItem, progress: number, list: FileItem[]) => {
+    const { setLocalFileList } = dataSourceStore.getState()
+    const newList = produce(list, (draft) => {
+      const targetIndex = draft.findIndex(file => file.fileID === fileItem.fileID)
+      if (targetIndex !== -1) {
+        draft[targetIndex] = {
+          ...draft[targetIndex],
+          ...fileItem,
+          progress,
+        }
+      }
+    })
+    setLocalFileList(newList)
+  }, [dataSourceStore])
+
+  const onFileListUpdate = useCallback((files: FileItem[]) => {
+    const { setLocalFileList } = dataSourceStore.getState()
+    setLocalFileList(files)
+    fileListRef.current = files
+  }, [dataSourceStore])
+
+  const onPreview = useCallback((file: File) => {
+    const { setCurrentLocalFile } = dataSourceStore.getState()
+    setCurrentLocalFile(file)
+  }, [dataSourceStore])
+
+  const {
+    dropRef,
+    dragRef,
+    fileUploaderRef,
+    dragging,
+    fileUploadConfig,
+    acceptTypes,
+    supportTypesShowNames,
+    hideUpload,
+    selectHandle,
+    fileChangeHandle,
+    removeFile,
+    handlePreview,
+  } = useFileUpload({
+    fileList: localFileList,
+    prepareFileList,
+    onFileUpdate,
+    onFileListUpdate,
+    onPreview,
+    supportBatchUpload,
+    allowedExtensions,
+  })
+
+  return {
+    // Refs
+    dropRef,
+    dragRef,
+    fileUploaderRef,
+
+    // State
+    dragging,
+    localFileList,
+
+    // Config
+    fileUploadConfig,
+    acceptTypes,
+    supportTypesShowNames,
+    hideUpload,
+
+    // Handlers
+    selectHandle,
+    fileChangeHandle,
+    removeFile,
+    handlePreview,
+  }
+}

web/app/components/datasets/documents/create-from-pipeline/data-source/local-file/index.spec.tsx

@@ -0,0 +1,398 @@
+import type { FileItem } from '@/models/datasets'
+import { render, screen } from '@testing-library/react'
+import { beforeEach, describe, expect, it, vi } from 'vitest'
+import LocalFile from './index'
+
+// Mock the hook
+const mockUseLocalFileUpload = vi.fn()
+vi.mock('./hooks/use-local-file-upload', () => ({
+  useLocalFileUpload: (...args: unknown[]) => mockUseLocalFileUpload(...args),
+}))
+
+// Mock react-i18next for sub-components
+vi.mock('react-i18next', () => ({
+  useTranslation: () => ({
+    t: (key: string) => key,
+  }),
+}))
+
+// Mock theme hook for sub-components
+vi.mock('@/hooks/use-theme', () => ({
+  default: () => ({ theme: 'light' }),
+}))
+
+// Mock theme types
+vi.mock('@/types/app', () => ({
+  Theme: { dark: 'dark', light: 'light' },
+}))
+
+// Mock DocumentFileIcon
+vi.mock('@/app/components/datasets/common/document-file-icon', () => ({
+  default: ({ name }: { name: string }) => <div data-testid="document-icon">{name}</div>,
+}))
+
+// Mock SimplePieChart
+vi.mock('next/dynamic', () => ({
+  default: () => {
+    const Component = ({ percentage }: { percentage: number }) => (
+      <div data-testid="pie-chart">
+        {percentage}
+        %
+      </div>
+    )
+    return Component
+  },
+}))
+
+describe('LocalFile', () => {
+  const mockDropRef = { current: null }
+  const mockDragRef = { current: null }
+  const mockFileUploaderRef = { current: null }
+
+  const defaultHookReturn = {
+    dropRef: mockDropRef,
+    dragRef: mockDragRef,
+    fileUploaderRef: mockFileUploaderRef,
+    dragging: false,
+    localFileList: [] as FileItem[],
+    fileUploadConfig: {
+      file_size_limit: 15,
+      batch_count_limit: 5,
+      file_upload_limit: 10,
+    },
+    acceptTypes: ['.pdf', '.docx'],
+    supportTypesShowNames: 'PDF, DOCX',
+    hideUpload: false,
+    selectHandle: vi.fn(),
+    fileChangeHandle: vi.fn(),
+    removeFile: vi.fn(),
+    handlePreview: vi.fn(),
+  }
+
+  beforeEach(() => {
+    vi.clearAllMocks()
+    mockUseLocalFileUpload.mockReturnValue(defaultHookReturn)
+  })
+
+  describe('rendering', () => {
+    it('should render the component container', () => {
+      const { container } = render(
+        <LocalFile allowedExtensions={['pdf', 'docx']} />,
+      )
+
+      expect(container.firstChild).toHaveClass('flex', 'flex-col')
+    })
+
+    it('should render UploadDropzone when hideUpload is false', () => {
+      render(<LocalFile allowedExtensions={['pdf']} />)
+
+      const fileInput = document.getElementById('fileUploader')
+      expect(fileInput).toBeInTheDocument()
+    })
+
+    it('should not render UploadDropzone when hideUpload is true', () => {
+      mockUseLocalFileUpload.mockReturnValue({
+        ...defaultHookReturn,
+        hideUpload: true,
+      })
+
+      render(<LocalFile allowedExtensions={['pdf']} />)
+
+      const fileInput = document.getElementById('fileUploader')
+      expect(fileInput).not.toBeInTheDocument()
+    })
+  })
+
+  describe('file list rendering', () => {
+    it('should not render file list when empty', () => {
+      render(<LocalFile allowedExtensions={['pdf']} />)
+
+      expect(screen.queryByTestId('document-icon')).not.toBeInTheDocument()
+    })
+
+    it('should render file list when files exist', () => {
+      const mockFile = {
+        name: 'test.pdf',
+        size: 1024,
+        type: 'application/pdf',
+        lastModified: Date.now(),
+      } as File
+
+      mockUseLocalFileUpload.mockReturnValue({
+        ...defaultHookReturn,
+        localFileList: [
+          {
+            fileID: 'file-1',
+            file: mockFile,
+            progress: -1,
+          },
+        ],
+      })
+
+      render(<LocalFile allowedExtensions={['pdf']} />)
+
+      expect(screen.getByTestId('document-icon')).toBeInTheDocument()
+    })
+
+    it('should render multiple file items', () => {
+      const createMockFile = (name: string) => ({
+        name,
+        size: 1024,
+        type: 'application/pdf',
+        lastModified: Date.now(),
+      }) as File
+
+      mockUseLocalFileUpload.mockReturnValue({
+        ...defaultHookReturn,
+        localFileList: [
+          { fileID: 'file-1', file: createMockFile('doc1.pdf'), progress: -1 },
+          { fileID: 'file-2', file: createMockFile('doc2.pdf'), progress: -1 },
+          { fileID: 'file-3', file: createMockFile('doc3.pdf'), progress: -1 },
+        ],
+      })
+
+      render(<LocalFile allowedExtensions={['pdf']} />)
+
+      const icons = screen.getAllByTestId('document-icon')
+      expect(icons).toHaveLength(3)
+    })
+
+    it('should use correct key for file items', () => {
+      const mockFile = {
+        name: 'test.pdf',
+        size: 1024,
+        type: 'application/pdf',
+        lastModified: Date.now(),
+      } as File
+
+      mockUseLocalFileUpload.mockReturnValue({
+        ...defaultHookReturn,
+        localFileList: [
+          { fileID: 'unique-id-123', file: mockFile, progress: -1 },
+        ],
+      })
+
+      render(<LocalFile allowedExtensions={['pdf']} />)
+
+      // The component should render without errors (key is used internally)
+      expect(screen.getByTestId('document-icon')).toBeInTheDocument()
+    })
+  })
+
+  describe('hook integration', () => {
+    it('should pass allowedExtensions to hook', () => {
+      render(<LocalFile allowedExtensions={['pdf', 'docx', 'txt']} />)
+
+      expect(mockUseLocalFileUpload).toHaveBeenCalledWith({
+        allowedExtensions: ['pdf', 'docx', 'txt'],
+        supportBatchUpload: true,
+      })
+    })
+
+    it('should pass supportBatchUpload true by default', () => {
+      render(<LocalFile allowedExtensions={['pdf']} />)
+
+      expect(mockUseLocalFileUpload).toHaveBeenCalledWith(
+        expect.objectContaining({ supportBatchUpload: true }),
+      )
+    })
+
+    it('should pass supportBatchUpload false when specified', () => {
+      render(<LocalFile allowedExtensions={['pdf']} supportBatchUpload={false} />)
+
+      expect(mockUseLocalFileUpload).toHaveBeenCalledWith(
+        expect.objectContaining({ supportBatchUpload: false }),
+      )
+    })
+  })
+
+  describe('props passed to UploadDropzone', () => {
+    it('should pass all required props to UploadDropzone', () => {
+      const selectHandle = vi.fn()
+      const fileChangeHandle = vi.fn()
+
+      mockUseLocalFileUpload.mockReturnValue({
+        ...defaultHookReturn,
+        selectHandle,
+        fileChangeHandle,
+        supportTypesShowNames: 'PDF, DOCX',
+        acceptTypes: ['.pdf', '.docx'],
+        fileUploadConfig: {
+          file_size_limit: 20,
+          batch_count_limit: 10,
+          file_upload_limit: 50,
+        },
+      })
+
+      render(<LocalFile allowedExtensions={['pdf', 'docx']} supportBatchUpload={true} />)
+
+      // Verify the dropzone is rendered with correct configuration
+      const fileInput = document.getElementById('fileUploader')
+      expect(fileInput).toBeInTheDocument()
+      expect(fileInput).toHaveAttribute('accept', '.pdf,.docx')
+      expect(fileInput).toHaveAttribute('multiple')
+    })
+  })
+
+  describe('props passed to FileListItem', () => {
+    it('should pass correct props to file items', () => {
+      const handlePreview = vi.fn()
+      const removeFile = vi.fn()
+      const mockFile = {
+        name: 'document.pdf',
+        size: 2048,
+        type: 'application/pdf',
+        lastModified: Date.now(),
+      } as File
+
+      mockUseLocalFileUpload.mockReturnValue({
+        ...defaultHookReturn,
+        handlePreview,
+        removeFile,
+        localFileList: [
+          { fileID: 'test-id', file: mockFile, progress: 50 },
+        ],
+      })
+
+      render(<LocalFile allowedExtensions={['pdf']} />)
+
+      expect(screen.getByTestId('document-icon')).toHaveTextContent('document.pdf')
+    })
+  })
+
+  describe('conditional rendering', () => {
+    it('should show both dropzone and file list when files exist and hideUpload is false', () => {
+      const mockFile = {
+        name: 'test.pdf',
+        size: 1024,
+        type: 'application/pdf',
+        lastModified: Date.now(),
+      } as File
+
+      mockUseLocalFileUpload.mockReturnValue({
+        ...defaultHookReturn,
+        hideUpload: false,
+        localFileList: [
+          { fileID: 'file-1', file: mockFile, progress: -1 },
+        ],
+      })
+
+      render(<LocalFile allowedExtensions={['pdf']} />)
+
+      expect(document.getElementById('fileUploader')).toBeInTheDocument()
+      expect(screen.getByTestId('document-icon')).toBeInTheDocument()
+    })
+
+    it('should show only file list when hideUpload is true', () => {
+      const mockFile = {
+        name: 'test.pdf',
+        size: 1024,
+        type: 'application/pdf',
+        lastModified: Date.now(),
+      } as File
+
+      mockUseLocalFileUpload.mockReturnValue({
+        ...defaultHookReturn,
+        hideUpload: true,
+        localFileList: [
+          { fileID: 'file-1', file: mockFile, progress: -1 },
+        ],
+      })
+
+      render(<LocalFile allowedExtensions={['pdf']} />)
+
+      expect(document.getElementById('fileUploader')).not.toBeInTheDocument()
+      expect(screen.getByTestId('document-icon')).toBeInTheDocument()
+    })
+  })
+
+  describe('file list container styling', () => {
+    it('should apply correct container classes for file list', () => {
+      const mockFile = {
+        name: 'test.pdf',
+        size: 1024,
+        type: 'application/pdf',
+        lastModified: Date.now(),
+      } as File
+
+      mockUseLocalFileUpload.mockReturnValue({
+        ...defaultHookReturn,
+        localFileList: [
+          { fileID: 'file-1', file: mockFile, progress: -1 },
+        ],
+      })
+
+      const { container } = render(<LocalFile allowedExtensions={['pdf']} />)
+
+      const fileListContainer = container.querySelector('.mt-1.flex.flex-col.gap-y-1')
+      expect(fileListContainer).toBeInTheDocument()
+    })
+  })
+
+  describe('edge cases', () => {
+    it('should handle empty allowedExtensions', () => {
+      render(<LocalFile allowedExtensions={[]} />)
+
+      expect(mockUseLocalFileUpload).toHaveBeenCalledWith({
+        allowedExtensions: [],
+        supportBatchUpload: true,
+      })
+    })
+
+    it('should handle files with same fileID but different index', () => {
+      const mockFile = {
+        name: 'test.pdf',
+        size: 1024,
+        type: 'application/pdf',
+        lastModified: Date.now(),
+      } as File
+
+      mockUseLocalFileUpload.mockReturnValue({
+        ...defaultHookReturn,
+        localFileList: [
+          { fileID: 'same-id', file: { ...mockFile, name: 'doc1.pdf' } as File, progress: -1 },
+          { fileID: 'same-id', file: { ...mockFile, name: 'doc2.pdf' } as File, progress: -1 },
+        ],
+      })
+
+      // Should render without key collision errors due to index in key
+      render(<LocalFile allowedExtensions={['pdf']} />)
+
+      const icons = screen.getAllByTestId('document-icon')
+      expect(icons).toHaveLength(2)
+    })
+  })
+
+  describe('component integration', () => {
+    it('should render complete component tree', () => {
+      const mockFile = {
+        name: 'complete-test.pdf',
+        size: 5 * 1024,
+        type: 'application/pdf',
+        lastModified: Date.now(),
+      } as File
+
+      mockUseLocalFileUpload.mockReturnValue({
+        ...defaultHookReturn,
+        hideUpload: false,
+        localFileList: [
+          { fileID: 'file-1', file: mockFile, progress: 50 },
+        ],
+        dragging: false,
+      })
+
+      const { container } = render(
+        <LocalFile allowedExtensions={['pdf', 'docx']} supportBatchUpload={true} />,
+      )
+
+      // Main container
+      expect(container.firstChild).toHaveClass('flex', 'flex-col')
+
+      // Dropzone exists
+      expect(document.getElementById('fileUploader')).toBeInTheDocument()
+
+      // File list exists
+      expect(screen.getByTestId('document-icon')).toBeInTheDocument()
+    })
+  })
+})

web/app/components/datasets/documents/create-from-pipeline/data-source/local-file/index.tsx

@@ -1,26 +1,7 @@
 'use client'
-import type { CustomFile as File, FileItem } from '@/models/datasets'
-import { RiDeleteBinLine, RiErrorWarningFill, RiUploadCloud2Line } from '@remixicon/react'
-import { produce } from 'immer'
-import dynamic from 'next/dynamic'
-import * as React from 'react'
-import { useCallback, useEffect, useMemo, useRef, useState } from 'react'
-import { useTranslation } from 'react-i18next'
-import { useContext } from 'use-context-selector'
-import { getFileUploadErrorMessage } from '@/app/components/base/file-uploader/utils'
-import { ToastContext } from '@/app/components/base/toast'
-import DocumentFileIcon from '@/app/components/datasets/common/document-file-icon'
-import { IS_CE_EDITION } from '@/config'
-import { useLocale } from '@/context/i18n'
-import useTheme from '@/hooks/use-theme'
-import { LanguagesSupported } from '@/i18n-config/language'
-import { upload } from '@/service/base'
-import { useFileUploadConfig } from '@/service/use-common'
-import { Theme } from '@/types/app'
-import { cn } from '@/utils/classnames'
-import { useDataSourceStore, useDataSourceStoreWithSelector } from '../store'
-
-const SimplePieChart = dynamic(() => import('@/app/components/base/simple-pie-chart'), { ssr: false })
+import FileListItem from './components/file-list-item'
+import UploadDropzone from './components/upload-dropzone'
+import { useLocalFileUpload } from './hooks/use-local-file-upload'
 
 export type LocalFileProps = {
   allowedExtensions: string[]
@@ -31,345 +12,49 @@ const LocalFile = ({
   allowedExtensions,
   supportBatchUpload = true,
 }: LocalFileProps) => {
-  const { t } = useTranslation()
-  const { notify } = useContext(ToastContext)
-  const locale = useLocale()
-  const localFileList = useDataSourceStoreWithSelector(state => state.localFileList)
-  const dataSourceStore = useDataSourceStore()
-  const [dragging, setDragging] = useState(false)
-
-  const dropRef = useRef<HTMLDivElement>(null)
-  const dragRef = useRef<HTMLDivElement>(null)
-  const fileUploader = useRef<HTMLInputElement>(null)
-  const fileListRef = useRef<FileItem[]>([])
-
-  const hideUpload = !supportBatchUpload && localFileList.length > 0
-
-  const { data: fileUploadConfigResponse } = useFileUploadConfig()
-  const supportTypesShowNames = useMemo(() => {
-    const extensionMap: { [key: string]: string } = {
-      md: 'markdown',
-      pptx: 'pptx',
-      htm: 'html',
-      xlsx: 'xlsx',
-      docx: 'docx',
-    }
-
-    return allowedExtensions
-      .map(item => extensionMap[item] || item) // map to standardized extension
-      .map(item => item.toLowerCase()) // convert to lower case
-      .filter((item, index, self) => self.indexOf(item) === index) // remove duplicates
-      .map(item => item.toUpperCase()) // convert to upper case
-      .join(locale !== LanguagesSupported[1] ? ', ' : '、 ')
-  }, [locale, allowedExtensions])
-  const ACCEPTS = allowedExtensions.map((ext: string) => `.${ext}`)
-  const fileUploadConfig = useMemo(() => ({
-    file_size_limit: fileUploadConfigResponse?.file_size_limit ?? 15,
-    batch_count_limit: supportBatchUpload ? (fileUploadConfigResponse?.batch_count_limit ?? 5) : 1,
-    file_upload_limit: supportBatchUpload ? (fileUploadConfigResponse?.file_upload_limit ?? 5) : 1,
-  }), [fileUploadConfigResponse, supportBatchUpload])
-
-  const updateFile = useCallback((fileItem: FileItem, progress: number, list: FileItem[]) => {
-    const { setLocalFileList } = dataSourceStore.getState()
-    const newList = produce(list, (draft) => {
-      const targetIndex = draft.findIndex(file => file.fileID === fileItem.fileID)
-      draft[targetIndex] = {
-        ...draft[targetIndex],
-        progress,
-      }
-    })
-    setLocalFileList(newList)
-  }, [dataSourceStore])
-
-  const updateFileList = useCallback((preparedFiles: FileItem[]) => {
-    const { setLocalFileList } = dataSourceStore.getState()
-    setLocalFileList(preparedFiles)
-  }, [dataSourceStore])
-
-  const handlePreview = useCallback((file: File) => {
-    const { setCurrentLocalFile } = dataSourceStore.getState()
-    if (file.id)
-      setCurrentLocalFile(file)
-  }, [dataSourceStore])
-
-  // utils
-  const getFileType = (currentFile: File) => {
-    if (!currentFile)
-      return ''
-
-    const arr = currentFile.name.split('.')
-    return arr[arr.length - 1]
-  }
-
-  const getFileSize = (size: number) => {
-    if (size / 1024 < 10)
-      return `${(size / 1024).toFixed(2)}KB`
-
-    return `${(size / 1024 / 1024).toFixed(2)}MB`
-  }
-
-  const isValid = useCallback((file: File) => {
-    const { size } = file
-    const ext = `.${getFileType(file)}`
-    const isValidType = ACCEPTS.includes(ext.toLowerCase())
-    if (!isValidType)
-      notify({ type: 'error', message: t('stepOne.uploader.validation.typeError', { ns: 'datasetCreation' }) })
-
-    const isValidSize = size <= fileUploadConfig.file_size_limit * 1024 * 1024
-    if (!isValidSize)
-      notify({ type: 'error', message: t('stepOne.uploader.validation.size', { ns: 'datasetCreation', size: fileUploadConfig.file_size_limit }) })
-
-    return isValidType && isValidSize
-  }, [notify, t, ACCEPTS, fileUploadConfig.file_size_limit])
-
-  type UploadResult = Awaited<ReturnType<typeof upload>>
-
-  const fileUpload = useCallback(async (fileItem: FileItem): Promise<FileItem> => {
-    const formData = new FormData()
-    formData.append('file', fileItem.file)
-    const onProgress = (e: ProgressEvent) => {
-      if (e.lengthComputable) {
-        const percent = Math.floor(e.loaded / e.total * 100)
-        updateFile(fileItem, percent, fileListRef.current)
-      }
-    }
-
-    return upload({
-      xhr: new XMLHttpRequest(),
-      data: formData,
-      onprogress: onProgress,
-    }, false, undefined, '?source=datasets')
-      .then((res: UploadResult) => {
-        const updatedFile = Object.assign({}, fileItem.file, {
-          id: res.id,
-          ...(res as Partial<File>),
-        }) as File
-        const completeFile: FileItem = {
-          fileID: fileItem.fileID,
-          file: updatedFile,
-          progress: -1,
-        }
-        const index = fileListRef.current.findIndex(item => item.fileID === fileItem.fileID)
-        fileListRef.current[index] = completeFile
-        updateFile(completeFile, 100, fileListRef.current)
-        return Promise.resolve({ ...completeFile })
-      })
-      .catch((e) => {
-        const errorMessage = getFileUploadErrorMessage(e, t('stepOne.uploader.failed', { ns: 'datasetCreation' }), t)
-        notify({ type: 'error', message: errorMessage })
-        updateFile(fileItem, -2, fileListRef.current)
-        return Promise.resolve({ ...fileItem })
-      })
-      .finally()
-  }, [fileListRef, notify, updateFile, t])
-
-  const uploadBatchFiles = useCallback((bFiles: FileItem[]) => {
-    bFiles.forEach(bf => (bf.progress = 0))
-    return Promise.all(bFiles.map(fileUpload))
-  }, [fileUpload])
-
-  const uploadMultipleFiles = useCallback(async (files: FileItem[]) => {
-    const batchCountLimit = fileUploadConfig.batch_count_limit
-    const length = files.length
-    let start = 0
-    let end = 0
-
-    while (start < length) {
-      if (start + batchCountLimit > length)
-        end = length
-      else
-        end = start + batchCountLimit
-      const bFiles = files.slice(start, end)
-      await uploadBatchFiles(bFiles)
-      start = end
-    }
-  }, [fileUploadConfig, uploadBatchFiles])
-
-  const initialUpload = useCallback((files: File[]) => {
-    const filesCountLimit = fileUploadConfig.file_upload_limit
-    if (!files.length)
-      return false
-
-    if (files.length + localFileList.length > filesCountLimit && !IS_CE_EDITION) {
-      notify({ type: 'error', message: t('stepOne.uploader.validation.filesNumber', { ns: 'datasetCreation', filesNumber: filesCountLimit }) })
-      return false
-    }
-
-    const preparedFiles = files.map((file, index) => ({
-      fileID: `file${index}-${Date.now()}`,
-      file,
-      progress: -1,
-    }))
-    const newFiles = [...fileListRef.current, ...preparedFiles]
-    updateFileList(newFiles)
-    fileListRef.current = newFiles
-    uploadMultipleFiles(preparedFiles)
-  }, [fileUploadConfig.file_upload_limit, localFileList.length, updateFileList, uploadMultipleFiles, notify, t])
-
-  const handleDragEnter = (e: DragEvent) => {
-    e.preventDefault()
-    e.stopPropagation()
-    if (e.target !== dragRef.current)
-      setDragging(true)
-  }
-  const handleDragOver = (e: DragEvent) => {
-    e.preventDefault()
-    e.stopPropagation()
-  }
-  const handleDragLeave = (e: DragEvent) => {
-    e.preventDefault()
-    e.stopPropagation()
-    if (e.target === dragRef.current)
-      setDragging(false)
-  }
-
-  const handleDrop = useCallback((e: DragEvent) => {
-    e.preventDefault()
-    e.stopPropagation()
-    setDragging(false)
-    if (!e.dataTransfer)
-      return
-
-    let files = Array.from(e.dataTransfer.files) as File[]
-    if (!supportBatchUpload)
-      files = files.slice(0, 1)
-
-    const validFiles = files.filter(isValid)
-    initialUpload(validFiles)
-  }, [initialUpload, isValid, supportBatchUpload])
-
-  const selectHandle = useCallback(() => {
-    if (fileUploader.current)
-      fileUploader.current.click()
-  }, [])
-
-  const removeFile = (fileID: string) => {
-    if (fileUploader.current)
-      fileUploader.current.value = ''
-
-    fileListRef.current = fileListRef.current.filter(item => item.fileID !== fileID)
-    updateFileList([...fileListRef.current])
-  }
-  const fileChangeHandle = useCallback((e: React.ChangeEvent<HTMLInputElement>) => {
-    let files = Array.from(e.target.files ?? []) as File[]
-    files = files.slice(0, fileUploadConfig.batch_count_limit)
-    initialUpload(files.filter(isValid))
-  }, [isValid, initialUpload, fileUploadConfig.batch_count_limit])
-
-  const { theme } = useTheme()
-  const chartColor = useMemo(() => theme === Theme.dark ? '#5289ff' : '#296dff', [theme])
-
-  useEffect(() => {
-    const dropElement = dropRef.current
-    dropElement?.addEventListener('dragenter', handleDragEnter)
-    dropElement?.addEventListener('dragover', handleDragOver)
-    dropElement?.addEventListener('dragleave', handleDragLeave)
-    dropElement?.addEventListener('drop', handleDrop)
-    return () => {
-      dropElement?.removeEventListener('dragenter', handleDragEnter)
-      dropElement?.removeEventListener('dragover', handleDragOver)
-      dropElement?.removeEventListener('dragleave', handleDragLeave)
-      dropElement?.removeEventListener('drop', handleDrop)
-    }
-  }, [handleDrop])
+  const {
+    dropRef,
+    dragRef,
+    fileUploaderRef,
+    dragging,
+    localFileList,
+    fileUploadConfig,
+    acceptTypes,
+    supportTypesShowNames,
+    hideUpload,
+    selectHandle,
+    fileChangeHandle,
+    removeFile,
+    handlePreview,
+  } = useLocalFileUpload({ allowedExtensions, supportBatchUpload })
 
   return (
     <div className="flex flex-col">
       {!hideUpload && (
-        <input
-          ref={fileUploader}
-          id="fileUploader"
-          className="hidden"
-          type="file"
-          multiple={supportBatchUpload}
-          accept={ACCEPTS.join(',')}
-          onChange={fileChangeHandle}
+        <UploadDropzone
+          dropRef={dropRef}
+          dragRef={dragRef}
+          fileUploaderRef={fileUploaderRef}
+          dragging={dragging}
+          supportBatchUpload={supportBatchUpload}
+          supportTypesShowNames={supportTypesShowNames}
+          fileUploadConfig={fileUploadConfig}
+          acceptTypes={acceptTypes}
+          onSelectFile={selectHandle}
+          onFileChange={fileChangeHandle}
+          allowedExtensions={allowedExtensions}
         />
       )}
-      {!hideUpload && (
-        <div
-          ref={dropRef}
-          className={cn(
-            'relative box-border flex min-h-20 flex-col items-center justify-center gap-1 rounded-xl border border-dashed border-components-dropzone-border bg-components-dropzone-bg px-4 py-3 text-xs leading-4 text-text-tertiary',
-            dragging && 'border-components-dropzone-border-accent bg-components-dropzone-bg-accent',
-          )}
-        >
-          <div className="flex min-h-5 items-center justify-center text-sm leading-4 text-text-secondary">
-            <RiUploadCloud2Line className="mr-2 size-5" />
-
-            <span>
-              {supportBatchUpload ? t('stepOne.uploader.button', { ns: 'datasetCreation' }) : t('stepOne.uploader.buttonSingleFile', { ns: 'datasetCreation' })}
-              {allowedExtensions.length > 0 && (
-                <label className="ml-1 cursor-pointer text-text-accent" onClick={selectHandle}>{t('stepOne.uploader.browse', { ns: 'datasetCreation' })}</label>
-              )}
-            </span>
-          </div>
-          <div>
-            {t('stepOne.uploader.tip', {
-              ns: 'datasetCreation',
-              size: fileUploadConfig.file_size_limit,
-              supportTypes: supportTypesShowNames,
-              batchCount: fileUploadConfig.batch_count_limit,
-              totalCount: fileUploadConfig.file_upload_limit,
-            })}
-          </div>
-          {dragging && <div ref={dragRef} className="absolute left-0 top-0 h-full w-full" />}
-        </div>
-      )}
       {localFileList.length > 0 && (
         <div className="mt-1 flex flex-col gap-y-1">
-          {localFileList.map((fileItem, index) => {
-            const isUploading = fileItem.progress >= 0 && fileItem.progress < 100
-            const isError = fileItem.progress === -2
-            return (
-              <div
-                key={`${fileItem.fileID}-${index}`}
-                onClick={handlePreview.bind(null, fileItem.file)}
-                className={cn(
-                  'flex h-12 items-center rounded-lg border border-components-panel-border bg-components-panel-on-panel-item-bg shadow-xs shadow-shadow-shadow-4',
-                  isError && 'border-state-destructive-border bg-state-destructive-hover',
-                )}
-              >
-                <div className="flex w-12 shrink-0 items-center justify-center">
-                  <DocumentFileIcon
-                    size="lg"
-                    className="shrink-0"
-                    name={fileItem.file.name}
-                    extension={getFileType(fileItem.file)}
-                  />
-                </div>
-                <div className="flex shrink grow flex-col gap-0.5">
-                  <div className="flex w-full">
-                    <div className="w-0 grow truncate text-xs text-text-secondary">{fileItem.file.name}</div>
-                  </div>
-                  <div className="w-full truncate text-2xs leading-3 text-text-tertiary">
-                    <span className="uppercase">{getFileType(fileItem.file)}</span>
-                    <span className="px-1 text-text-quaternary">·</span>
-                    <span>{getFileSize(fileItem.file.size)}</span>
-                  </div>
-                </div>
-                <div className="flex w-16 shrink-0 items-center justify-end gap-1 pr-3">
-                  {isUploading && (
-                    <SimplePieChart percentage={fileItem.progress} stroke={chartColor} fill={chartColor} animationDuration={0} />
-                  )}
-                  {
-                    isError && (
-                      <RiErrorWarningFill className="size-4 text-text-destructive" />
-                    )
-                  }
-                  <span
-                    className="flex h-6 w-6 cursor-pointer items-center justify-center"
-                    onClick={(e) => {
-                      e.stopPropagation()
-                      removeFile(fileItem.fileID)
-                    }}
-                  >
-                    <RiDeleteBinLine className="size-4 text-text-tertiary" />
-                  </span>
-                </div>
-              </div>
-            )
-          })}
+          {localFileList.map((fileItem, index) => (
+            <FileListItem
+              key={`${fileItem.fileID}-${index}`}
+              fileItem={fileItem}
+              onPreview={handlePreview}
+              onRemove={removeFile}
+            />
+          ))}
         </div>
       )}
     </div>

web/app/components/datasets/documents/create-from-pipeline/data-source/website-crawl/base/error-message.tsx

@@ -14,7 +14,6 @@ const ErrorMessage = ({
   errorMsg,
 }: ErrorMessageProps) => {
   return (
-    // eslint-disable-next-line tailwindcss/migration-from-tailwind-2
     <div className={cn(
       'flex gap-x-0.5 rounded-xl border-[0.5px] border-components-panel-border bg-opacity-40 bg-toast-error-bg p-2 shadow-xs shadow-shadow-shadow-3',
       className,

web/app/components/datasets/documents/detail/completed/common/batch-action.tsx

@@ -7,6 +7,7 @@ import Button from '@/app/components/base/button'
 import Confirm from '@/app/components/base/confirm'
 import Divider from '@/app/components/base/divider'
 import { SearchLinesSparkle } from '@/app/components/base/icons/src/vender/knowledge'
+import { IS_CE_EDITION } from '@/config'
 import { cn } from '@/utils/classnames'
 
 const i18nPrefix = 'batchAction'
@@ -87,7 +88,7 @@ const BatchAction: FC<IBatchActionProps> = ({
             <span className="px-0.5">{t('metadata.metadata', { ns: 'dataset' })}</span>
           </Button>
         )}
-        {onBatchSummary && (
+        {onBatchSummary && IS_CE_EDITION && (
           <Button
             variant="ghost"
             className="gap-x-0.5 px-3"

web/app/components/datasets/documents/detail/embedding/components/index.ts

@@ -0,0 +1,4 @@
+export { default as ProgressBar } from './progress-bar'
+export { default as RuleDetail } from './rule-detail'
+export { default as SegmentProgress } from './segment-progress'
+export { default as StatusHeader } from './status-header'

web/app/components/datasets/documents/detail/embedding/components/progress-bar.spec.tsx

@@ -0,0 +1,159 @@
+import { render } from '@testing-library/react'
+import { describe, expect, it } from 'vitest'
+import ProgressBar from './progress-bar'
+
+describe('ProgressBar', () => {
+  const defaultProps = {
+    percent: 50,
+    isEmbedding: false,
+    isCompleted: false,
+    isPaused: false,
+    isError: false,
+  }
+
+  const getProgressElements = (container: HTMLElement) => {
+    const wrapper = container.firstChild as HTMLElement
+    const progressBar = wrapper.firstChild as HTMLElement
+    return { wrapper, progressBar }
+  }
+
+  describe('Rendering', () => {
+    it('should render without crashing', () => {
+      const { container } = render(<ProgressBar {...defaultProps} />)
+      const { wrapper, progressBar } = getProgressElements(container)
+      expect(wrapper).toBeInTheDocument()
+      expect(progressBar).toBeInTheDocument()
+    })
+
+    it('should render progress bar container with correct classes', () => {
+      const { container } = render(<ProgressBar {...defaultProps} />)
+      const { wrapper } = getProgressElements(container)
+      expect(wrapper).toHaveClass('flex', 'h-2', 'w-full', 'items-center', 'overflow-hidden', 'rounded-md')
+    })
+
+    it('should render inner progress bar with transition classes', () => {
+      const { container } = render(<ProgressBar {...defaultProps} />)
+      const { progressBar } = getProgressElements(container)
+      expect(progressBar).toHaveClass('h-full', 'transition-all', 'duration-300')
+    })
+  })
+
+  describe('Progress Width', () => {
+    it('should set progress width to 0%', () => {
+      const { container } = render(<ProgressBar {...defaultProps} percent={0} />)
+      const { progressBar } = getProgressElements(container)
+      expect(progressBar).toHaveStyle({ width: '0%' })
+    })
+
+    it('should set progress width to 50%', () => {
+      const { container } = render(<ProgressBar {...defaultProps} percent={50} />)
+      const { progressBar } = getProgressElements(container)
+      expect(progressBar).toHaveStyle({ width: '50%' })
+    })
+
+    it('should set progress width to 100%', () => {
+      const { container } = render(<ProgressBar {...defaultProps} percent={100} />)
+      const { progressBar } = getProgressElements(container)
+      expect(progressBar).toHaveStyle({ width: '100%' })
+    })
+
+    it('should set progress width to 75%', () => {
+      const { container } = render(<ProgressBar {...defaultProps} percent={75} />)
+      const { progressBar } = getProgressElements(container)
+      expect(progressBar).toHaveStyle({ width: '75%' })
+    })
+  })
+
+  describe('Container Background States', () => {
+    it('should apply semi-transparent background when isEmbedding is true', () => {
+      const { container } = render(<ProgressBar {...defaultProps} isEmbedding />)
+      const { wrapper } = getProgressElements(container)
+      expect(wrapper).toHaveClass('bg-components-progress-bar-bg/50')
+    })
+
+    it('should apply default background when isEmbedding is false', () => {
+      const { container } = render(<ProgressBar {...defaultProps} isEmbedding={false} />)
+      const { wrapper } = getProgressElements(container)
+      expect(wrapper).toHaveClass('bg-components-progress-bar-bg')
+      expect(wrapper).not.toHaveClass('bg-components-progress-bar-bg/50')
+    })
+  })
+
+  describe('Progress Bar Fill States', () => {
+    it('should apply solid progress style when isEmbedding is true', () => {
+      const { container } = render(<ProgressBar {...defaultProps} isEmbedding />)
+      const { progressBar } = getProgressElements(container)
+      expect(progressBar).toHaveClass('bg-components-progress-bar-progress-solid')
+    })
+
+    it('should apply solid progress style when isCompleted is true', () => {
+      const { container } = render(<ProgressBar {...defaultProps} isCompleted />)
+      const { progressBar } = getProgressElements(container)
+      expect(progressBar).toHaveClass('bg-components-progress-bar-progress-solid')
+    })
+
+    it('should apply highlight style when isPaused is true', () => {
+      const { container } = render(<ProgressBar {...defaultProps} isPaused />)
+      const { progressBar } = getProgressElements(container)
+      expect(progressBar).toHaveClass('bg-components-progress-bar-progress-highlight')
+    })
+
+    it('should apply highlight style when isError is true', () => {
+      const { container } = render(<ProgressBar {...defaultProps} isError />)
+      const { progressBar } = getProgressElements(container)
+      expect(progressBar).toHaveClass('bg-components-progress-bar-progress-highlight')
+    })
+
+    it('should not apply fill styles when no status flags are set', () => {
+      const { container } = render(<ProgressBar {...defaultProps} />)
+      const { progressBar } = getProgressElements(container)
+      expect(progressBar).not.toHaveClass('bg-components-progress-bar-progress-solid')
+      expect(progressBar).not.toHaveClass('bg-components-progress-bar-progress-highlight')
+    })
+  })
+
+  describe('Combined States', () => {
+    it('should apply highlight when isEmbedding and isPaused', () => {
+      const { container } = render(<ProgressBar {...defaultProps} isEmbedding isPaused />)
+      const { progressBar } = getProgressElements(container)
+      // highlight takes precedence since isPaused condition is separate
+      expect(progressBar).toHaveClass('bg-components-progress-bar-progress-highlight')
+    })
+
+    it('should apply highlight when isCompleted and isError', () => {
+      const { container } = render(<ProgressBar {...defaultProps} isCompleted isError />)
+      const { progressBar } = getProgressElements(container)
+      // highlight takes precedence since isError condition is separate
+      expect(progressBar).toHaveClass('bg-components-progress-bar-progress-highlight')
+    })
+
+    it('should apply semi-transparent bg for embedding and highlight for paused', () => {
+      const { container } = render(<ProgressBar {...defaultProps} isEmbedding isPaused />)
+      const { wrapper } = getProgressElements(container)
+      expect(wrapper).toHaveClass('bg-components-progress-bar-bg/50')
+    })
+  })
+
+  describe('Edge Cases', () => {
+    it('should handle all props set to false', () => {
+      const { container } = render(
+        <ProgressBar
+          percent={0}
+          isEmbedding={false}
+          isCompleted={false}
+          isPaused={false}
+          isError={false}
+        />,
+      )
+      const { wrapper, progressBar } = getProgressElements(container)
+      expect(wrapper).toBeInTheDocument()
+      expect(progressBar).toHaveStyle({ width: '0%' })
+    })
+
+    it('should handle decimal percent values', () => {
+      const { container } = render(<ProgressBar {...defaultProps} percent={33.33} />)
+      const { progressBar } = getProgressElements(container)
+      expect(progressBar).toHaveStyle({ width: '33.33%' })
+    })
+  })
+})

web/app/components/datasets/documents/detail/embedding/components/progress-bar.tsx

@@ -0,0 +1,44 @@
+import type { FC } from 'react'
+import * as React from 'react'
+import { cn } from '@/utils/classnames'
+
+type ProgressBarProps = {
+  percent: number
+  isEmbedding: boolean
+  isCompleted: boolean
+  isPaused: boolean
+  isError: boolean
+}
+
+const ProgressBar: FC<ProgressBarProps> = React.memo(({
+  percent,
+  isEmbedding,
+  isCompleted,
+  isPaused,
+  isError,
+}) => {
+  const isActive = isEmbedding || isCompleted
+  const isHighlighted = isPaused || isError
+
+  return (
+    <div
+      className={cn(
+        'flex h-2 w-full items-center overflow-hidden rounded-md border border-components-progress-bar-border',
+        isEmbedding ? 'bg-components-progress-bar-bg/50' : 'bg-components-progress-bar-bg',
+      )}
+    >
+      <div
+        className={cn(
+          'h-full transition-all duration-300',
+          isActive && 'bg-components-progress-bar-progress-solid',
+          isHighlighted && 'bg-components-progress-bar-progress-highlight',
+        )}
+        style={{ width: `${percent}%` }}
+      />
+    </div>
+  )
+})
+
+ProgressBar.displayName = 'ProgressBar'
+
+export default ProgressBar

web/app/components/datasets/documents/detail/embedding/components/rule-detail.spec.tsx

@@ -0,0 +1,203 @@
+import type { ProcessRuleResponse } from '@/models/datasets'
+import { render, screen } from '@testing-library/react'
+import { describe, expect, it } from 'vitest'
+import { ProcessMode } from '@/models/datasets'
+import { RETRIEVE_METHOD } from '@/types/app'
+import { IndexingType } from '../../../../create/step-two'
+import RuleDetail from './rule-detail'
+
+describe('RuleDetail', () => {
+  const defaultProps = {
+    indexingType: IndexingType.QUALIFIED,
+    retrievalMethod: RETRIEVE_METHOD.semantic,
+  }
+
+  const createSourceData = (overrides: Partial<ProcessRuleResponse> = {}): ProcessRuleResponse => ({
+    mode: ProcessMode.general,
+    rules: {
+      segmentation: {
+        separator: '\n',
+        max_tokens: 500,
+        chunk_overlap: 50,
+      },
+      pre_processing_rules: [
+        { id: 'remove_extra_spaces', enabled: true },
+        { id: 'remove_urls_emails', enabled: false },
+      ],
+      parent_mode: 'full-doc',
+      subchunk_segmentation: {
+        separator: '\n',
+        max_tokens: 200,
+        chunk_overlap: 20,
+      },
+    },
+    limits: { indexing_max_segmentation_tokens_length: 4000 },
+    ...overrides,
+  })
+
+  describe('Rendering', () => {
+    it('should render without crashing', () => {
+      render(<RuleDetail {...defaultProps} />)
+      expect(screen.getByText(/stepTwo\.indexMode/i)).toBeInTheDocument()
+    })
+
+    it('should render with sourceData', () => {
+      const sourceData = createSourceData()
+      render(<RuleDetail {...defaultProps} sourceData={sourceData} />)
+      expect(screen.getByText(/embedding\.mode/i)).toBeInTheDocument()
+    })
+
+    it('should render all segmentation rule fields', () => {
+      const sourceData = createSourceData()
+      render(<RuleDetail {...defaultProps} sourceData={sourceData} />)
+      expect(screen.getByText(/embedding\.mode/i)).toBeInTheDocument()
+      expect(screen.getByText(/embedding\.segmentLength/i)).toBeInTheDocument()
+      expect(screen.getByText(/embedding\.textCleaning/i)).toBeInTheDocument()
+    })
+  })
+
+  describe('Mode Display', () => {
+    it('should display custom mode for general process mode', () => {
+      const sourceData = createSourceData({ mode: ProcessMode.general })
+      render(<RuleDetail {...defaultProps} sourceData={sourceData} />)
+      expect(screen.getByText(/embedding\.custom/i)).toBeInTheDocument()
+    })
+
+    it('should display mode label field', () => {
+      const sourceData = createSourceData()
+      render(<RuleDetail {...defaultProps} sourceData={sourceData} />)
+      expect(screen.getByText(/embedding\.mode/i)).toBeInTheDocument()
+    })
+  })
+
+  describe('Segment Length Display', () => {
+    it('should display max tokens for general mode', () => {
+      const sourceData = createSourceData({
+        mode: ProcessMode.general,
+        rules: {
+          segmentation: { separator: '\n', max_tokens: 500, chunk_overlap: 50 },
+          pre_processing_rules: [],
+          parent_mode: 'full-doc',
+          subchunk_segmentation: { separator: '\n', max_tokens: 200, chunk_overlap: 20 },
+        },
+      })
+      render(<RuleDetail {...defaultProps} sourceData={sourceData} />)
+      expect(screen.getByText('500')).toBeInTheDocument()
+    })
+
+    it('should display segment length label', () => {
+      const sourceData = createSourceData()
+      render(<RuleDetail {...defaultProps} sourceData={sourceData} />)
+      expect(screen.getByText(/embedding\.segmentLength/i)).toBeInTheDocument()
+    })
+  })
+
+  describe('Text Cleaning Display', () => {
+    it('should display enabled pre-processing rules', () => {
+      const sourceData = createSourceData({
+        rules: {
+          segmentation: { separator: '\n', max_tokens: 500, chunk_overlap: 50 },
+          pre_processing_rules: [
+            { id: 'remove_extra_spaces', enabled: true },
+            { id: 'remove_urls_emails', enabled: true },
+          ],
+          parent_mode: 'full-doc',
+          subchunk_segmentation: { separator: '\n', max_tokens: 200, chunk_overlap: 20 },
+        },
+      })
+      render(<RuleDetail {...defaultProps} sourceData={sourceData} />)
+      expect(screen.getByText(/removeExtraSpaces/i)).toBeInTheDocument()
+      expect(screen.getByText(/removeUrlEmails/i)).toBeInTheDocument()
+    })
+
+    it('should display text cleaning label', () => {
+      const sourceData = createSourceData()
+      render(<RuleDetail {...defaultProps} sourceData={sourceData} />)
+      expect(screen.getByText(/embedding\.textCleaning/i)).toBeInTheDocument()
+    })
+  })
+
+  describe('Index Mode Display', () => {
+    it('should display economical mode when indexingType is ECONOMICAL', () => {
+      render(<RuleDetail {...defaultProps} indexingType={IndexingType.ECONOMICAL} />)
+      expect(screen.getByText(/stepTwo\.economical/i)).toBeInTheDocument()
+    })
+
+    it('should display qualified mode when indexingType is QUALIFIED', () => {
+      render(<RuleDetail {...defaultProps} indexingType={IndexingType.QUALIFIED} />)
+      expect(screen.getByText(/stepTwo\.qualified/i)).toBeInTheDocument()
+    })
+  })
+
+  describe('Retrieval Method Display', () => {
+    it('should display keyword search for economical mode', () => {
+      render(<RuleDetail {...defaultProps} indexingType={IndexingType.ECONOMICAL} />)
+      expect(screen.getByText(/retrieval\.keyword_search\.title/i)).toBeInTheDocument()
+    })
+
+    it('should display semantic search as default for qualified mode', () => {
+      render(<RuleDetail {...defaultProps} indexingType={IndexingType.QUALIFIED} />)
+      expect(screen.getByText(/retrieval\.semantic_search\.title/i)).toBeInTheDocument()
+    })
+
+    it('should display full text search when retrievalMethod is fullText', () => {
+      render(<RuleDetail {...defaultProps} retrievalMethod={RETRIEVE_METHOD.fullText} />)
+      expect(screen.getByText(/retrieval\.full_text_search\.title/i)).toBeInTheDocument()
+    })
+
+    it('should display hybrid search when retrievalMethod is hybrid', () => {
+      render(<RuleDetail {...defaultProps} retrievalMethod={RETRIEVE_METHOD.hybrid} />)
+      expect(screen.getByText(/retrieval\.hybrid_search\.title/i)).toBeInTheDocument()
+    })
+  })
+
+  describe('Edge Cases', () => {
+    it('should display dash for missing sourceData', () => {
+      render(<RuleDetail {...defaultProps} />)
+      const dashes = screen.getAllByText('-')
+      expect(dashes.length).toBeGreaterThan(0)
+    })
+
+    it('should display dash when mode is undefined', () => {
+      const sourceData = { rules: {} } as ProcessRuleResponse
+      render(<RuleDetail {...defaultProps} sourceData={sourceData} />)
+      const dashes = screen.getAllByText('-')
+      expect(dashes.length).toBeGreaterThan(0)
+    })
+
+    it('should handle undefined retrievalMethod', () => {
+      render(<RuleDetail indexingType={IndexingType.QUALIFIED} />)
+      expect(screen.getByText(/retrieval\.semantic_search\.title/i)).toBeInTheDocument()
+    })
+
+    it('should handle empty pre_processing_rules array', () => {
+      const sourceData = createSourceData({
+        rules: {
+          segmentation: { separator: '\n', max_tokens: 500, chunk_overlap: 50 },
+          pre_processing_rules: [],
+          parent_mode: 'full-doc',
+          subchunk_segmentation: { separator: '\n', max_tokens: 200, chunk_overlap: 20 },
+        },
+      })
+      render(<RuleDetail {...defaultProps} sourceData={sourceData} />)
+      expect(screen.getByText(/embedding\.textCleaning/i)).toBeInTheDocument()
+    })
+
+    it('should render container with correct structure', () => {
+      const { container } = render(<RuleDetail {...defaultProps} />)
+      const wrapper = container.firstChild as HTMLElement
+      expect(wrapper).toHaveClass('py-3')
+    })
+
+    it('should handle undefined indexingType', () => {
+      render(<RuleDetail retrievalMethod={RETRIEVE_METHOD.semantic} />)
+      expect(screen.getByText(/stepTwo\.indexMode/i)).toBeInTheDocument()
+    })
+
+    it('should render divider between sections', () => {
+      const { container } = render(<RuleDetail {...defaultProps} />)
+      const dividers = container.querySelectorAll('.bg-divider-subtle')
+      expect(dividers.length).toBeGreaterThan(0)
+    })
+  })
+})

web/app/components/datasets/documents/detail/embedding/components/rule-detail.tsx

@@ -0,0 +1,128 @@
+import type { FC } from 'react'
+import type { ProcessRuleResponse } from '@/models/datasets'
+import type { RETRIEVE_METHOD } from '@/types/app'
+import Image from 'next/image'
+import * as React from 'react'
+import { useCallback } from 'react'
+import { useTranslation } from 'react-i18next'
+import Divider from '@/app/components/base/divider'
+import { ProcessMode } from '@/models/datasets'
+import { indexMethodIcon, retrievalIcon } from '../../../../create/icons'
+import { IndexingType } from '../../../../create/step-two'
+import { FieldInfo } from '../../metadata'
+
+type RuleDetailProps = {
+  sourceData?: ProcessRuleResponse
+  indexingType?: IndexingType
+  retrievalMethod?: RETRIEVE_METHOD
+}
+
+const getRetrievalIcon = (method?: RETRIEVE_METHOD) => {
+  if (method === 'full_text_search')
+    return retrievalIcon.fullText
+  if (method === 'hybrid_search')
+    return retrievalIcon.hybrid
+  return retrievalIcon.vector
+}
+
+const RuleDetail: FC<RuleDetailProps> = React.memo(({
+  sourceData,
+  indexingType,
+  retrievalMethod,
+}) => {
+  const { t } = useTranslation()
+
+  const segmentationRuleMap = {
+    mode: t('embedding.mode', { ns: 'datasetDocuments' }),
+    segmentLength: t('embedding.segmentLength', { ns: 'datasetDocuments' }),
+    textCleaning: t('embedding.textCleaning', { ns: 'datasetDocuments' }),
+  }
+
+  const getRuleName = useCallback((key: string) => {
+    const ruleNameMap: Record<string, string> = {
+      remove_extra_spaces: t('stepTwo.removeExtraSpaces', { ns: 'datasetCreation' }),
+      remove_urls_emails: t('stepTwo.removeUrlEmails', { ns: 'datasetCreation' }),
+      remove_stopwords: t('stepTwo.removeStopwords', { ns: 'datasetCreation' }),
+    }
+    return ruleNameMap[key]
+  }, [t])
+
+  const getValue = useCallback((field: string) => {
+    const defaultValue = '-'
+
+    if (!sourceData?.mode)
+      return defaultValue
+
+    const maxTokens = typeof sourceData?.rules?.segmentation?.max_tokens === 'number'
+      ? sourceData.rules.segmentation.max_tokens
+      : defaultValue
+
+    const childMaxTokens = typeof sourceData?.rules?.subchunk_segmentation?.max_tokens === 'number'
+      ? sourceData.rules.subchunk_segmentation.max_tokens
+      : defaultValue
+
+    const isGeneralMode = sourceData.mode === ProcessMode.general
+
+    const fieldValueMap: Record<string, string | number> = {
+      mode: isGeneralMode
+        ? t('embedding.custom', { ns: 'datasetDocuments' })
+        : `${t('embedding.hierarchical', { ns: 'datasetDocuments' })} · ${
+          sourceData?.rules?.parent_mode === 'paragraph'
+            ? t('parentMode.paragraph', { ns: 'dataset' })
+            : t('parentMode.fullDoc', { ns: 'dataset' })
+        }`,
+      segmentLength: isGeneralMode
+        ? maxTokens
+        : `${t('embedding.parentMaxTokens', { ns: 'datasetDocuments' })} ${maxTokens}; ${t('embedding.childMaxTokens', { ns: 'datasetDocuments' })} ${childMaxTokens}`,
+      textCleaning: sourceData?.rules?.pre_processing_rules
+        ?.filter(rule => rule.enabled)
+        .map(rule => getRuleName(rule.id))
+        .join(',') || defaultValue,
+    }
+
+    return fieldValueMap[field] ?? defaultValue
+  }, [sourceData, t, getRuleName])
+
+  const isEconomical = indexingType === IndexingType.ECONOMICAL
+
+  return (
+    <div className="py-3">
+      <div className="flex flex-col gap-y-1">
+        {Object.keys(segmentationRuleMap).map(field => (
+          <FieldInfo
+            key={field}
+            label={segmentationRuleMap[field as keyof typeof segmentationRuleMap]}
+            displayedValue={String(getValue(field))}
+          />
+        ))}
+      </div>
+      <Divider type="horizontal" className="bg-divider-subtle" />
+      <FieldInfo
+        label={t('stepTwo.indexMode', { ns: 'datasetCreation' })}
+        displayedValue={t(`stepTwo.${isEconomical ? 'economical' : 'qualified'}`, { ns: 'datasetCreation' }) as string}
+        valueIcon={(
+          <Image
+            className="size-4"
+            src={isEconomical ? indexMethodIcon.economical : indexMethodIcon.high_quality}
+            alt=""
+          />
+        )}
+      />
+      <FieldInfo
+        label={t('form.retrievalSetting.title', { ns: 'datasetSettings' })}
+        displayedValue={t(`retrieval.${isEconomical ? 'keyword_search' : retrievalMethod ?? 'semantic_search'}.title`, { ns: 'dataset' })}
+        valueIcon={(
+          <Image
+            className="size-4"
+            src={getRetrievalIcon(retrievalMethod)}
+            alt=""
+          />
+        )}
+      />
+    </div>
+  )
+})
+
+RuleDetail.displayName = 'RuleDetail'
+
+export default RuleDetail

web/app/components/datasets/documents/detail/embedding/components/segment-progress.spec.tsx

@@ -0,0 +1,81 @@
+import { render, screen } from '@testing-library/react'
+import { describe, expect, it } from 'vitest'
+import SegmentProgress from './segment-progress'
+
+describe('SegmentProgress', () => {
+  const defaultProps = {
+    completedSegments: 50,
+    totalSegments: 100,
+    percent: 50,
+  }
+
+  describe('Rendering', () => {
+    it('should render without crashing', () => {
+      render(<SegmentProgress {...defaultProps} />)
+      expect(screen.getByText(/segments/i)).toBeInTheDocument()
+    })
+
+    it('should render with correct CSS classes', () => {
+      const { container } = render(<SegmentProgress {...defaultProps} />)
+      const wrapper = container.firstChild as HTMLElement
+      expect(wrapper).toHaveClass('flex', 'w-full', 'items-center')
+    })
+
+    it('should render text with correct styling class', () => {
+      render(<SegmentProgress {...defaultProps} />)
+      const text = screen.getByText(/segments/i)
+      expect(text).toHaveClass('system-xs-medium', 'text-text-secondary')
+    })
+  })
+
+  describe('Progress Display', () => {
+    it('should display completed and total segments', () => {
+      render(<SegmentProgress completedSegments={50} totalSegments={100} percent={50} />)
+      expect(screen.getByText(/50\/100/)).toBeInTheDocument()
+    })
+
+    it('should display percent value', () => {
+      render(<SegmentProgress completedSegments={50} totalSegments={100} percent={50} />)
+      expect(screen.getByText(/50%/)).toBeInTheDocument()
+    })
+
+    it('should display 0/0 when segments are 0', () => {
+      render(<SegmentProgress completedSegments={0} totalSegments={0} percent={0} />)
+      expect(screen.getByText(/0\/0/)).toBeInTheDocument()
+      expect(screen.getByText(/0%/)).toBeInTheDocument()
+    })
+
+    it('should display 100% when completed', () => {
+      render(<SegmentProgress completedSegments={100} totalSegments={100} percent={100} />)
+      expect(screen.getByText(/100\/100/)).toBeInTheDocument()
+      expect(screen.getByText(/100%/)).toBeInTheDocument()
+    })
+  })
+
+  describe('Edge Cases', () => {
+    it('should display -- when completedSegments is undefined', () => {
+      render(<SegmentProgress totalSegments={100} percent={0} />)
+      expect(screen.getByText(/--\/100/)).toBeInTheDocument()
+    })
+
+    it('should display -- when totalSegments is undefined', () => {
+      render(<SegmentProgress completedSegments={50} percent={50} />)
+      expect(screen.getByText(/50\/--/)).toBeInTheDocument()
+    })
+
+    it('should display --/-- when both segments are undefined', () => {
+      render(<SegmentProgress percent={0} />)
+      expect(screen.getByText(/--\/--/)).toBeInTheDocument()
+    })
+
+    it('should handle large numbers', () => {
+      render(<SegmentProgress completedSegments={999999} totalSegments={1000000} percent={99} />)
+      expect(screen.getByText(/999999\/1000000/)).toBeInTheDocument()
+    })
+
+    it('should handle decimal percent', () => {
+      render(<SegmentProgress completedSegments={33} totalSegments={100} percent={33.33} />)
+      expect(screen.getByText(/33.33%/)).toBeInTheDocument()
+    })
+  })
+})

web/app/components/datasets/documents/detail/embedding/components/segment-progress.tsx

@@ -0,0 +1,32 @@
+import type { FC } from 'react'
+import * as React from 'react'
+import { useTranslation } from 'react-i18next'
+
+type SegmentProgressProps = {
+  completedSegments?: number
+  totalSegments?: number
+  percent: number
+}
+
+const SegmentProgress: FC<SegmentProgressProps> = React.memo(({
+  completedSegments,
+  totalSegments,
+  percent,
+}) => {
+  const { t } = useTranslation()
+
+  const completed = completedSegments ?? '--'
+  const total = totalSegments ?? '--'
+
+  return (
+    <div className="flex w-full items-center">
+      <span className="system-xs-medium text-text-secondary">
+        {`${t('embedding.segments', { ns: 'datasetDocuments' })} ${completed}/${total} · ${percent}%`}
+      </span>
+    </div>
+  )
+})
+
+SegmentProgress.displayName = 'SegmentProgress'
+
+export default SegmentProgress

web/app/components/datasets/documents/detail/embedding/components/status-header.spec.tsx

@@ -0,0 +1,155 @@
+import { fireEvent, render, screen } from '@testing-library/react'
+import { beforeEach, describe, expect, it, vi } from 'vitest'
+import StatusHeader from './status-header'
+
+describe('StatusHeader', () => {
+  const defaultProps = {
+    isEmbedding: false,
+    isCompleted: false,
+    isPaused: false,
+    isError: false,
+    onPause: vi.fn(),
+    onResume: vi.fn(),
+  }
+
+  beforeEach(() => {
+    vi.clearAllMocks()
+  })
+
+  describe('Rendering', () => {
+    it('should render without crashing', () => {
+      const { container } = render(<StatusHeader {...defaultProps} />)
+      expect(container.firstChild).toBeInTheDocument()
+    })
+
+    it('should render with correct container classes', () => {
+      const { container } = render(<StatusHeader {...defaultProps} />)
+      const wrapper = container.firstChild as HTMLElement
+      expect(wrapper).toHaveClass('flex', 'h-6', 'items-center', 'gap-x-1')
+    })
+  })
+
+  describe('Status Text', () => {
+    it('should display processing text when isEmbedding is true', () => {
+      render(<StatusHeader {...defaultProps} isEmbedding />)
+      expect(screen.getByText(/embedding\.processing/i)).toBeInTheDocument()
+    })
+
+    it('should display completed text when isCompleted is true', () => {
+      render(<StatusHeader {...defaultProps} isCompleted />)
+      expect(screen.getByText(/embedding\.completed/i)).toBeInTheDocument()
+    })
+
+    it('should display paused text when isPaused is true', () => {
+      render(<StatusHeader {...defaultProps} isPaused />)
+      expect(screen.getByText(/embedding\.paused/i)).toBeInTheDocument()
+    })
+
+    it('should display error text when isError is true', () => {
+      render(<StatusHeader {...defaultProps} isError />)
+      expect(screen.getByText(/embedding\.error/i)).toBeInTheDocument()
+    })
+
+    it('should display empty text when no status flags are set', () => {
+      render(<StatusHeader {...defaultProps} />)
+      const statusText = screen.getByText('', { selector: 'span.system-md-semibold-uppercase' })
+      expect(statusText).toBeInTheDocument()
+    })
+  })
+
+  describe('Loading Spinner', () => {
+    it('should show loading spinner when isEmbedding is true', () => {
+      const { container } = render(<StatusHeader {...defaultProps} isEmbedding />)
+      const spinner = container.querySelector('svg.animate-spin')
+      expect(spinner).toBeInTheDocument()
+    })
+
+    it('should not show loading spinner when isEmbedding is false', () => {
+      const { container } = render(<StatusHeader {...defaultProps} isEmbedding={false} />)
+      const spinner = container.querySelector('svg.animate-spin')
+      expect(spinner).not.toBeInTheDocument()
+    })
+  })
+
+  describe('Pause Button', () => {
+    it('should show pause button when isEmbedding is true', () => {
+      render(<StatusHeader {...defaultProps} isEmbedding />)
+      expect(screen.getByRole('button')).toBeInTheDocument()
+      expect(screen.getByText(/embedding\.pause/i)).toBeInTheDocument()
+    })
+
+    it('should not show pause button when isEmbedding is false', () => {
+      render(<StatusHeader {...defaultProps} isEmbedding={false} />)
+      expect(screen.queryByText(/embedding\.pause/i)).not.toBeInTheDocument()
+    })
+
+    it('should call onPause when pause button is clicked', () => {
+      const onPause = vi.fn()
+      render(<StatusHeader {...defaultProps} isEmbedding onPause={onPause} />)
+      fireEvent.click(screen.getByRole('button'))
+      expect(onPause).toHaveBeenCalledTimes(1)
+    })
+
+    it('should disable pause button when isPauseLoading is true', () => {
+      render(<StatusHeader {...defaultProps} isEmbedding isPauseLoading />)
+      expect(screen.getByRole('button')).toBeDisabled()
+    })
+  })
+
+  describe('Resume Button', () => {
+    it('should show resume button when isPaused is true', () => {
+      render(<StatusHeader {...defaultProps} isPaused />)
+      expect(screen.getByRole('button')).toBeInTheDocument()
+      expect(screen.getByText(/embedding\.resume/i)).toBeInTheDocument()
+    })
+
+    it('should not show resume button when isPaused is false', () => {
+      render(<StatusHeader {...defaultProps} isPaused={false} />)
+      expect(screen.queryByText(/embedding\.resume/i)).not.toBeInTheDocument()
+    })
+
+    it('should call onResume when resume button is clicked', () => {
+      const onResume = vi.fn()
+      render(<StatusHeader {...defaultProps} isPaused onResume={onResume} />)
+      fireEvent.click(screen.getByRole('button'))
+      expect(onResume).toHaveBeenCalledTimes(1)
+    })
+
+    it('should disable resume button when isResumeLoading is true', () => {
+      render(<StatusHeader {...defaultProps} isPaused isResumeLoading />)
+      expect(screen.getByRole('button')).toBeDisabled()
+    })
+  })
+
+  describe('Button Styles', () => {
+    it('should have correct button styles for pause button', () => {
+      render(<StatusHeader {...defaultProps} isEmbedding />)
+      const button = screen.getByRole('button')
+      expect(button).toHaveClass('flex', 'items-center', 'gap-x-1', 'rounded-md')
+    })
+
+    it('should have correct button styles for resume button', () => {
+      render(<StatusHeader {...defaultProps} isPaused />)
+      const button = screen.getByRole('button')
+      expect(button).toHaveClass('flex', 'items-center', 'gap-x-1', 'rounded-md')
+    })
+  })
+
+  describe('Edge Cases', () => {
+    it('should not show any buttons when isCompleted', () => {
+      render(<StatusHeader {...defaultProps} isCompleted />)
+      expect(screen.queryByRole('button')).not.toBeInTheDocument()
+    })
+
+    it('should not show any buttons when isError', () => {
+      render(<StatusHeader {...defaultProps} isError />)
+      expect(screen.queryByRole('button')).not.toBeInTheDocument()
+    })
+
+    it('should show both buttons when isEmbedding and isPaused are both true', () => {
+      render(<StatusHeader {...defaultProps} isEmbedding isPaused />)
+      const buttons = screen.getAllByRole('button')
+      expect(buttons.length).toBe(2)
+    })
+  })
+})

web/app/components/datasets/documents/detail/embedding/components/status-header.tsx

@@ -0,0 +1,84 @@
+import type { FC } from 'react'
+import { RiLoader2Line, RiPauseCircleLine, RiPlayCircleLine } from '@remixicon/react'
+import * as React from 'react'
+import { useTranslation } from 'react-i18next'
+
+type StatusHeaderProps = {
+  isEmbedding: boolean
+  isCompleted: boolean
+  isPaused: boolean
+  isError: boolean
+  onPause: () => void
+  onResume: () => void
+  isPauseLoading?: boolean
+  isResumeLoading?: boolean
+}
+
+const StatusHeader: FC<StatusHeaderProps> = React.memo(({
+  isEmbedding,
+  isCompleted,
+  isPaused,
+  isError,
+  onPause,
+  onResume,
+  isPauseLoading,
+  isResumeLoading,
+}) => {
+  const { t } = useTranslation()
+
+  const getStatusText = () => {
+    if (isEmbedding)
+      return t('embedding.processing', { ns: 'datasetDocuments' })
+    if (isCompleted)
+      return t('embedding.completed', { ns: 'datasetDocuments' })
+    if (isPaused)
+      return t('embedding.paused', { ns: 'datasetDocuments' })
+    if (isError)
+      return t('embedding.error', { ns: 'datasetDocuments' })
+    return ''
+  }
+
+  const buttonBaseClass = `flex items-center gap-x-1 rounded-md border-[0.5px]
+    border-components-button-secondary-border bg-components-button-secondary-bg
+    px-1.5 py-1 shadow-xs shadow-shadow-shadow-3 backdrop-blur-[5px]
+    disabled:cursor-not-allowed disabled:opacity-50`
+
+  return (
+    <div className="flex h-6 items-center gap-x-1">
+      {isEmbedding && <RiLoader2Line className="h-4 w-4 animate-spin text-text-secondary" />}
+      <span className="system-md-semibold-uppercase grow text-text-secondary">
+        {getStatusText()}
+      </span>
+      {isEmbedding && (
+        <button
+          type="button"
+          className={buttonBaseClass}
+          onClick={onPause}
+          disabled={isPauseLoading}
+        >
+          <RiPauseCircleLine className="h-3.5 w-3.5 text-components-button-secondary-text" />
+          <span className="system-xs-medium pr-[3px] text-components-button-secondary-text">
+            {t('embedding.pause', { ns: 'datasetDocuments' })}
+          </span>
+        </button>
+      )}
+      {isPaused && (
+        <button
+          type="button"
+          className={buttonBaseClass}
+          onClick={onResume}
+          disabled={isResumeLoading}
+        >
+          <RiPlayCircleLine className="h-3.5 w-3.5 text-components-button-secondary-text" />
+          <span className="system-xs-medium pr-[3px] text-components-button-secondary-text">
+            {t('embedding.resume', { ns: 'datasetDocuments' })}
+          </span>
+        </button>
+      )}
+    </div>
+  )
+})
+
+StatusHeader.displayName = 'StatusHeader'
+
+export default StatusHeader

web/app/components/datasets/documents/detail/embedding/hooks/index.ts

@@ -0,0 +1,10 @@
+export {
+  calculatePercent,
+  isEmbeddingStatus,
+  isTerminalStatus,
+  useEmbeddingStatus,
+  useInvalidateEmbeddingStatus,
+  usePauseIndexing,
+  useResumeIndexing,
+} from './use-embedding-status'
+export type { EmbeddingStatusType } from './use-embedding-status'

web/app/components/datasets/documents/detail/embedding/hooks/use-embedding-status.spec.tsx

@@ -0,0 +1,462 @@
+import type { ReactNode } from 'react'
+import type { IndexingStatusResponse } from '@/models/datasets'
+import { QueryClient, QueryClientProvider } from '@tanstack/react-query'
+import { act, renderHook, waitFor } from '@testing-library/react'
+import { beforeEach, describe, expect, it, vi } from 'vitest'
+import * as datasetsService from '@/service/datasets'
+import {
+  calculatePercent,
+  isEmbeddingStatus,
+  isTerminalStatus,
+  useEmbeddingStatus,
+  useInvalidateEmbeddingStatus,
+  usePauseIndexing,
+  useResumeIndexing,
+} from './use-embedding-status'
+
+vi.mock('@/service/datasets')
+
+const mockFetchIndexingStatus = vi.mocked(datasetsService.fetchIndexingStatus)
+const mockPauseDocIndexing = vi.mocked(datasetsService.pauseDocIndexing)
+const mockResumeDocIndexing = vi.mocked(datasetsService.resumeDocIndexing)
+
+const createTestQueryClient = () => new QueryClient({
+  defaultOptions: {
+    queries: { retry: false },
+    mutations: { retry: false },
+  },
+})
+
+const createWrapper = () => {
+  const queryClient = createTestQueryClient()
+  return ({ children }: { children: ReactNode }) => (
+    <QueryClientProvider client={queryClient}>
+      {children}
+    </QueryClientProvider>
+  )
+}
+
+const mockIndexingStatus = (overrides: Partial<IndexingStatusResponse> = {}): IndexingStatusResponse => ({
+  id: 'doc1',
+  indexing_status: 'indexing',
+  completed_segments: 50,
+  total_segments: 100,
+  processing_started_at: 0,
+  parsing_completed_at: 0,
+  cleaning_completed_at: 0,
+  splitting_completed_at: 0,
+  completed_at: null,
+  paused_at: null,
+  error: null,
+  stopped_at: null,
+  ...overrides,
+})
+
+describe('use-embedding-status', () => {
+  beforeEach(() => {
+    vi.clearAllMocks()
+  })
+
+  describe('isEmbeddingStatus', () => {
+    it('should return true for indexing status', () => {
+      expect(isEmbeddingStatus('indexing')).toBe(true)
+    })
+
+    it('should return true for splitting status', () => {
+      expect(isEmbeddingStatus('splitting')).toBe(true)
+    })
+
+    it('should return true for parsing status', () => {
+      expect(isEmbeddingStatus('parsing')).toBe(true)
+    })
+
+    it('should return true for cleaning status', () => {
+      expect(isEmbeddingStatus('cleaning')).toBe(true)
+    })
+
+    it('should return false for completed status', () => {
+      expect(isEmbeddingStatus('completed')).toBe(false)
+    })
+
+    it('should return false for paused status', () => {
+      expect(isEmbeddingStatus('paused')).toBe(false)
+    })
+
+    it('should return false for error status', () => {
+      expect(isEmbeddingStatus('error')).toBe(false)
+    })
+
+    it('should return false for undefined', () => {
+      expect(isEmbeddingStatus(undefined)).toBe(false)
+    })
+
+    it('should return false for empty string', () => {
+      expect(isEmbeddingStatus('')).toBe(false)
+    })
+  })
+
+  describe('isTerminalStatus', () => {
+    it('should return true for completed status', () => {
+      expect(isTerminalStatus('completed')).toBe(true)
+    })
+
+    it('should return true for error status', () => {
+      expect(isTerminalStatus('error')).toBe(true)
+    })
+
+    it('should return true for paused status', () => {
+      expect(isTerminalStatus('paused')).toBe(true)
+    })
+
+    it('should return false for indexing status', () => {
+      expect(isTerminalStatus('indexing')).toBe(false)
+    })
+
+    it('should return false for undefined', () => {
+      expect(isTerminalStatus(undefined)).toBe(false)
+    })
+  })
+
+  describe('calculatePercent', () => {
+    it('should calculate percent correctly', () => {
+      expect(calculatePercent(50, 100)).toBe(50)
+    })
+
+    it('should return 0 when total is 0', () => {
+      expect(calculatePercent(50, 0)).toBe(0)
+    })
+
+    it('should return 0 when total is undefined', () => {
+      expect(calculatePercent(50, undefined)).toBe(0)
+    })
+
+    it('should return 0 when completed is undefined', () => {
+      expect(calculatePercent(undefined, 100)).toBe(0)
+    })
+
+    it('should cap at 100 when percent exceeds 100', () => {
+      expect(calculatePercent(150, 100)).toBe(100)
+    })
+
+    it('should round to nearest integer', () => {
+      expect(calculatePercent(33, 100)).toBe(33)
+      expect(calculatePercent(1, 3)).toBe(33)
+    })
+  })
+
+  describe('useEmbeddingStatus', () => {
+    it('should return initial state when disabled', () => {
+      const { result } = renderHook(
+        () => useEmbeddingStatus({ datasetId: 'ds1', documentId: 'doc1', enabled: false }),
+        { wrapper: createWrapper() },
+      )
+
+      expect(result.current.isEmbedding).toBe(false)
+      expect(result.current.isCompleted).toBe(false)
+      expect(result.current.isPaused).toBe(false)
+      expect(result.current.isError).toBe(false)
+      expect(result.current.percent).toBe(0)
+    })
+
+    it('should not fetch when datasetId is missing', () => {
+      renderHook(
+        () => useEmbeddingStatus({ documentId: 'doc1' }),
+        { wrapper: createWrapper() },
+      )
+
+      expect(mockFetchIndexingStatus).not.toHaveBeenCalled()
+    })
+
+    it('should not fetch when documentId is missing', () => {
+      renderHook(
+        () => useEmbeddingStatus({ datasetId: 'ds1' }),
+        { wrapper: createWrapper() },
+      )
+
+      expect(mockFetchIndexingStatus).not.toHaveBeenCalled()
+    })
+
+    it('should fetch indexing status when enabled with valid ids', async () => {
+      mockFetchIndexingStatus.mockResolvedValue(mockIndexingStatus())
+
+      const { result } = renderHook(
+        () => useEmbeddingStatus({ datasetId: 'ds1', documentId: 'doc1' }),
+        { wrapper: createWrapper() },
+      )
+
+      await waitFor(() => {
+        expect(result.current.isEmbedding).toBe(true)
+      })
+
+      expect(mockFetchIndexingStatus).toHaveBeenCalledWith({
+        datasetId: 'ds1',
+        documentId: 'doc1',
+      })
+      expect(result.current.percent).toBe(50)
+    })
+
+    it('should set isCompleted when status is completed', async () => {
+      mockFetchIndexingStatus.mockResolvedValue(mockIndexingStatus({
+        indexing_status: 'completed',
+        completed_segments: 100,
+      }))
+
+      const { result } = renderHook(
+        () => useEmbeddingStatus({ datasetId: 'ds1', documentId: 'doc1' }),
+        { wrapper: createWrapper() },
+      )
+
+      await waitFor(() => {
+        expect(result.current.isCompleted).toBe(true)
+      })
+
+      expect(result.current.percent).toBe(100)
+    })
+
+    it('should set isPaused when status is paused', async () => {
+      mockFetchIndexingStatus.mockResolvedValue(mockIndexingStatus({
+        indexing_status: 'paused',
+      }))
+
+      const { result } = renderHook(
+        () => useEmbeddingStatus({ datasetId: 'ds1', documentId: 'doc1' }),
+        { wrapper: createWrapper() },
+      )
+
+      await waitFor(() => {
+        expect(result.current.isPaused).toBe(true)
+      })
+    })
+
+    it('should set isError when status is error', async () => {
+      mockFetchIndexingStatus.mockResolvedValue(mockIndexingStatus({
+        indexing_status: 'error',
+        completed_segments: 25,
+      }))
+
+      const { result } = renderHook(
+        () => useEmbeddingStatus({ datasetId: 'ds1', documentId: 'doc1' }),
+        { wrapper: createWrapper() },
+      )
+
+      await waitFor(() => {
+        expect(result.current.isError).toBe(true)
+      })
+    })
+
+    it('should provide invalidate function', async () => {
+      mockFetchIndexingStatus.mockResolvedValue(mockIndexingStatus())
+
+      const { result } = renderHook(
+        () => useEmbeddingStatus({ datasetId: 'ds1', documentId: 'doc1' }),
+        { wrapper: createWrapper() },
+      )
+
+      await waitFor(() => {
+        expect(result.current.isEmbedding).toBe(true)
+      })
+
+      expect(typeof result.current.invalidate).toBe('function')
+
+      // Call invalidate should not throw
+      await act(async () => {
+        result.current.invalidate()
+      })
+    })
+
+    it('should provide resetStatus function that clears data', async () => {
+      mockFetchIndexingStatus.mockResolvedValue(mockIndexingStatus())
+
+      const { result } = renderHook(
+        () => useEmbeddingStatus({ datasetId: 'ds1', documentId: 'doc1' }),
+        { wrapper: createWrapper() },
+      )
+
+      await waitFor(() => {
+        expect(result.current.data).toBeDefined()
+      })
+
+      // Reset status should clear the data
+      await act(async () => {
+        result.current.resetStatus()
+      })
+
+      await waitFor(() => {
+        expect(result.current.data).toBeNull()
+      })
+    })
+  })
+
+  describe('usePauseIndexing', () => {
+    it('should call pauseDocIndexing when mutate is called', async () => {
+      mockPauseDocIndexing.mockResolvedValue({ result: 'success' })
+
+      const { result } = renderHook(
+        () => usePauseIndexing({ datasetId: 'ds1', documentId: 'doc1' }),
+        { wrapper: createWrapper() },
+      )
+
+      await act(async () => {
+        result.current.mutate()
+      })
+
+      await waitFor(() => {
+        expect(mockPauseDocIndexing).toHaveBeenCalledWith({
+          datasetId: 'ds1',
+          documentId: 'doc1',
+        })
+      })
+    })
+
+    it('should call onSuccess callback on successful pause', async () => {
+      mockPauseDocIndexing.mockResolvedValue({ result: 'success' })
+      const onSuccess = vi.fn()
+
+      const { result } = renderHook(
+        () => usePauseIndexing({ datasetId: 'ds1', documentId: 'doc1', onSuccess }),
+        { wrapper: createWrapper() },
+      )
+
+      await act(async () => {
+        result.current.mutate()
+      })
+
+      await waitFor(() => {
+        expect(onSuccess).toHaveBeenCalled()
+      })
+    })
+
+    it('should call onError callback on failed pause', async () => {
+      const error = new Error('Network error')
+      mockPauseDocIndexing.mockRejectedValue(error)
+      const onError = vi.fn()
+
+      const { result } = renderHook(
+        () => usePauseIndexing({ datasetId: 'ds1', documentId: 'doc1', onError }),
+        { wrapper: createWrapper() },
+      )
+
+      await act(async () => {
+        result.current.mutate()
+      })
+
+      await waitFor(() => {
+        expect(onError).toHaveBeenCalled()
+        expect(onError.mock.calls[0][0]).toEqual(error)
+      })
+    })
+  })
+
+  describe('useResumeIndexing', () => {
+    it('should call resumeDocIndexing when mutate is called', async () => {
+      mockResumeDocIndexing.mockResolvedValue({ result: 'success' })
+
+      const { result } = renderHook(
+        () => useResumeIndexing({ datasetId: 'ds1', documentId: 'doc1' }),
+        { wrapper: createWrapper() },
+      )
+
+      await act(async () => {
+        result.current.mutate()
+      })
+
+      await waitFor(() => {
+        expect(mockResumeDocIndexing).toHaveBeenCalledWith({
+          datasetId: 'ds1',
+          documentId: 'doc1',
+        })
+      })
+    })
+
+    it('should call onSuccess callback on successful resume', async () => {
+      mockResumeDocIndexing.mockResolvedValue({ result: 'success' })
+      const onSuccess = vi.fn()
+
+      const { result } = renderHook(
+        () => useResumeIndexing({ datasetId: 'ds1', documentId: 'doc1', onSuccess }),
+        { wrapper: createWrapper() },
+      )
+
+      await act(async () => {
+        result.current.mutate()
+      })
+
+      await waitFor(() => {
+        expect(onSuccess).toHaveBeenCalled()
+      })
+    })
+  })
+
+  describe('useInvalidateEmbeddingStatus', () => {
+    it('should return a function', () => {
+      const { result } = renderHook(
+        () => useInvalidateEmbeddingStatus(),
+        { wrapper: createWrapper() },
+      )
+
+      expect(typeof result.current).toBe('function')
+    })
+
+    it('should invalidate specific query when datasetId and documentId are provided', async () => {
+      const queryClient = createTestQueryClient()
+      const wrapper = ({ children }: { children: ReactNode }) => (
+        <QueryClientProvider client={queryClient}>
+          {children}
+        </QueryClientProvider>
+      )
+
+      // Set some initial data in the cache
+      queryClient.setQueryData(['embedding', 'indexing-status', 'ds1', 'doc1'], {
+        id: 'doc1',
+        indexing_status: 'indexing',
+      })
+
+      const { result } = renderHook(
+        () => useInvalidateEmbeddingStatus(),
+        { wrapper },
+      )
+
+      await act(async () => {
+        result.current('ds1', 'doc1')
+      })
+
+      // The query should be invalidated (marked as stale)
+      const queryState = queryClient.getQueryState(['embedding', 'indexing-status', 'ds1', 'doc1'])
+      expect(queryState?.isInvalidated).toBe(true)
+    })
+
+    it('should invalidate all embedding status queries when ids are not provided', async () => {
+      const queryClient = createTestQueryClient()
+      const wrapper = ({ children }: { children: ReactNode }) => (
+        <QueryClientProvider client={queryClient}>
+          {children}
+        </QueryClientProvider>
+      )
+
+      // Set some initial data in the cache for multiple documents
+      queryClient.setQueryData(['embedding', 'indexing-status', 'ds1', 'doc1'], {
+        id: 'doc1',
+        indexing_status: 'indexing',
+      })
+      queryClient.setQueryData(['embedding', 'indexing-status', 'ds2', 'doc2'], {
+        id: 'doc2',
+        indexing_status: 'completed',
+      })
+
+      const { result } = renderHook(
+        () => useInvalidateEmbeddingStatus(),
+        { wrapper },
+      )
+
+      await act(async () => {
+        result.current()
+      })
+
+      // Both queries should be invalidated
+      const queryState1 = queryClient.getQueryState(['embedding', 'indexing-status', 'ds1', 'doc1'])
+      const queryState2 = queryClient.getQueryState(['embedding', 'indexing-status', 'ds2', 'doc2'])
+      expect(queryState1?.isInvalidated).toBe(true)
+      expect(queryState2?.isInvalidated).toBe(true)
+    })
+  })
+})

web/app/components/datasets/documents/detail/embedding/hooks/use-embedding-status.ts

@@ -0,0 +1,149 @@
+import type { CommonResponse } from '@/models/common'
+import type { IndexingStatusResponse } from '@/models/datasets'
+import { useMutation, useQuery, useQueryClient } from '@tanstack/react-query'
+import { useCallback, useEffect, useMemo, useRef } from 'react'
+import {
+  fetchIndexingStatus,
+  pauseDocIndexing,
+  resumeDocIndexing,
+} from '@/service/datasets'
+
+const NAME_SPACE = 'embedding'
+
+export type EmbeddingStatusType = 'indexing' | 'splitting' | 'parsing' | 'cleaning' | 'completed' | 'paused' | 'error' | 'waiting' | ''
+
+const EMBEDDING_STATUSES = ['indexing', 'splitting', 'parsing', 'cleaning'] as const
+const TERMINAL_STATUSES = ['completed', 'error', 'paused'] as const
+
+export const isEmbeddingStatus = (status?: string): boolean => {
+  return EMBEDDING_STATUSES.includes(status as typeof EMBEDDING_STATUSES[number])
+}
+
+export const isTerminalStatus = (status?: string): boolean => {
+  return TERMINAL_STATUSES.includes(status as typeof TERMINAL_STATUSES[number])
+}
+
+export const calculatePercent = (completed?: number, total?: number): number => {
+  if (!total || total === 0)
+    return 0
+  const percent = Math.round((completed || 0) * 100 / total)
+  return Math.min(percent, 100)
+}
+
+type UseEmbeddingStatusOptions = {
+  datasetId?: string
+  documentId?: string
+  enabled?: boolean
+  onComplete?: () => void
+}
+
+export const useEmbeddingStatus = ({
+  datasetId,
+  documentId,
+  enabled = true,
+  onComplete,
+}: UseEmbeddingStatusOptions) => {
+  const queryClient = useQueryClient()
+  const isPolling = useRef(false)
+  const onCompleteRef = useRef(onComplete)
+  onCompleteRef.current = onComplete
+
+  const queryKey = useMemo(
+    () => [NAME_SPACE, 'indexing-status', datasetId, documentId] as const,
+    [datasetId, documentId],
+  )
+
+  const query = useQuery<IndexingStatusResponse>({
+    queryKey,
+    queryFn: () => fetchIndexingStatus({ datasetId: datasetId!, documentId: documentId! }),
+    enabled: enabled && !!datasetId && !!documentId,
+    refetchInterval: (query) => {
+      const status = query.state.data?.indexing_status
+      if (isTerminalStatus(status)) {
+        return false
+      }
+      return 2500
+    },
+    refetchOnWindowFocus: false,
+  })
+
+  const status = query.data?.indexing_status || ''
+  const isEmbedding = isEmbeddingStatus(status)
+  const isCompleted = status === 'completed'
+  const isPaused = status === 'paused'
+  const isError = status === 'error'
+  const percent = calculatePercent(query.data?.completed_segments, query.data?.total_segments)
+
+  // Handle completion callback
+  useEffect(() => {
+    if (isTerminalStatus(status) && isPolling.current) {
+      isPolling.current = false
+      onCompleteRef.current?.()
+    }
+    if (isEmbedding) {
+      isPolling.current = true
+    }
+  }, [status, isEmbedding])
+
+  const invalidate = useCallback(() => {
+    queryClient.invalidateQueries({ queryKey })
+  }, [queryClient, queryKey])
+
+  const resetStatus = useCallback(() => {
+    queryClient.setQueryData(queryKey, null)
+  }, [queryClient, queryKey])
+
+  return {
+    data: query.data,
+    isLoading: query.isLoading,
+    isEmbedding,
+    isCompleted,
+    isPaused,
+    isError,
+    percent,
+    invalidate,
+    resetStatus,
+    refetch: query.refetch,
+  }
+}
+
+type UsePauseResumeOptions = {
+  datasetId?: string
+  documentId?: string
+  onSuccess?: () => void
+  onError?: (error: Error) => void
+}
+
+export const usePauseIndexing = ({ datasetId, documentId, onSuccess, onError }: UsePauseResumeOptions) => {
+  return useMutation<CommonResponse, Error>({
+    mutationKey: [NAME_SPACE, 'pause', datasetId, documentId],
+    mutationFn: () => pauseDocIndexing({ datasetId: datasetId!, documentId: documentId! }),
+    onSuccess,
+    onError,
+  })
+}
+
+export const useResumeIndexing = ({ datasetId, documentId, onSuccess, onError }: UsePauseResumeOptions) => {
+  return useMutation<CommonResponse, Error>({
+    mutationKey: [NAME_SPACE, 'resume', datasetId, documentId],
+    mutationFn: () => resumeDocIndexing({ datasetId: datasetId!, documentId: documentId! }),
+    onSuccess,
+    onError,
+  })
+}
+
+export const useInvalidateEmbeddingStatus = () => {
+  const queryClient = useQueryClient()
+  return useCallback((datasetId?: string, documentId?: string) => {
+    if (datasetId && documentId) {
+      queryClient.invalidateQueries({
+        queryKey: [NAME_SPACE, 'indexing-status', datasetId, documentId],
+      })
+    }
+    else {
+      queryClient.invalidateQueries({
+        queryKey: [NAME_SPACE, 'indexing-status'],
+      })
+    }
+  }, [queryClient])
+}

web/app/components/datasets/documents/detail/embedding/index.spec.tsx

@@ -0,0 +1,337 @@
+import type { ReactNode } from 'react'
+import type { DocumentContextValue } from '../context'
+import type { IndexingStatusResponse, ProcessRuleResponse } from '@/models/datasets'
+import { QueryClient, QueryClientProvider } from '@tanstack/react-query'
+import { render, screen, waitFor } from '@testing-library/react'
+import userEvent from '@testing-library/user-event'
+import { beforeEach, describe, expect, it, vi } from 'vitest'
+import { ProcessMode } from '@/models/datasets'
+import * as datasetsService from '@/service/datasets'
+import * as useDataset from '@/service/knowledge/use-dataset'
+import { RETRIEVE_METHOD } from '@/types/app'
+import { IndexingType } from '../../../create/step-two'
+import { DocumentContext } from '../context'
+import EmbeddingDetail from './index'
+
+vi.mock('@/service/datasets')
+vi.mock('@/service/knowledge/use-dataset')
+
+const mockFetchIndexingStatus = vi.mocked(datasetsService.fetchIndexingStatus)
+const mockPauseDocIndexing = vi.mocked(datasetsService.pauseDocIndexing)
+const mockResumeDocIndexing = vi.mocked(datasetsService.resumeDocIndexing)
+const mockUseProcessRule = vi.mocked(useDataset.useProcessRule)
+
+const createTestQueryClient = () => new QueryClient({
+  defaultOptions: {
+    queries: { retry: false, gcTime: 0 },
+    mutations: { retry: false },
+  },
+})
+
+const createWrapper = (contextValue: DocumentContextValue = { datasetId: 'ds1', documentId: 'doc1' }) => {
+  const queryClient = createTestQueryClient()
+  return ({ children }: { children: ReactNode }) => (
+    <QueryClientProvider client={queryClient}>
+      <DocumentContext.Provider value={contextValue}>
+        {children}
+      </DocumentContext.Provider>
+    </QueryClientProvider>
+  )
+}
+
+const mockIndexingStatus = (overrides: Partial<IndexingStatusResponse> = {}): IndexingStatusResponse => ({
+  id: 'doc1',
+  indexing_status: 'indexing',
+  completed_segments: 50,
+  total_segments: 100,
+  processing_started_at: Date.now(),
+  parsing_completed_at: 0,
+  cleaning_completed_at: 0,
+  splitting_completed_at: 0,
+  completed_at: null,
+  paused_at: null,
+  error: null,
+  stopped_at: null,
+  ...overrides,
+})
+
+const mockProcessRule = (overrides: Partial<ProcessRuleResponse> = {}): ProcessRuleResponse => ({
+  mode: ProcessMode.general,
+  rules: {
+    segmentation: { separator: '\n', max_tokens: 500, chunk_overlap: 50 },
+    pre_processing_rules: [{ id: 'remove_extra_spaces', enabled: true }],
+    parent_mode: 'full-doc',
+    subchunk_segmentation: { separator: '\n', max_tokens: 200, chunk_overlap: 20 },
+  },
+  limits: { indexing_max_segmentation_tokens_length: 4000 },
+  ...overrides,
+})
+
+describe('EmbeddingDetail', () => {
+  const defaultProps = {
+    detailUpdate: vi.fn(),
+    indexingType: IndexingType.QUALIFIED,
+    retrievalMethod: RETRIEVE_METHOD.semantic,
+  }
+
+  beforeEach(() => {
+    vi.clearAllMocks()
+
+    mockUseProcessRule.mockReturnValue({
+      data: mockProcessRule(),
+      isLoading: false,
+      error: null,
+    } as ReturnType<typeof useDataset.useProcessRule>)
+  })
+
+  describe('Rendering', () => {
+    it('should render without crashing', async () => {
+      mockFetchIndexingStatus.mockResolvedValue(mockIndexingStatus())
+
+      render(<EmbeddingDetail {...defaultProps} />, { wrapper: createWrapper() })
+
+      await waitFor(() => {
+        expect(screen.getByText(/embedding\.processing/i)).toBeInTheDocument()
+      })
+    })
+
+    it('should render with provided datasetId and documentId props', async () => {
+      mockFetchIndexingStatus.mockResolvedValue(mockIndexingStatus())
+
+      render(
+        <EmbeddingDetail {...defaultProps} datasetId="custom-ds" documentId="custom-doc" />,
+        { wrapper: createWrapper({ datasetId: '', documentId: '' }) },
+      )
+
+      await waitFor(() => {
+        expect(mockFetchIndexingStatus).toHaveBeenCalledWith({
+          datasetId: 'custom-ds',
+          documentId: 'custom-doc',
+        })
+      })
+    })
+
+    it('should fall back to context values when props are not provided', async () => {
+      mockFetchIndexingStatus.mockResolvedValue(mockIndexingStatus())
+
+      render(<EmbeddingDetail {...defaultProps} />, { wrapper: createWrapper() })
+
+      await waitFor(() => {
+        expect(mockFetchIndexingStatus).toHaveBeenCalledWith({
+          datasetId: 'ds1',
+          documentId: 'doc1',
+        })
+      })
+    })
+  })
+
+  describe('Status Display', () => {
+    it('should show processing status when indexing', async () => {
+      mockFetchIndexingStatus.mockResolvedValue(mockIndexingStatus({ indexing_status: 'indexing' }))
+
+      render(<EmbeddingDetail {...defaultProps} />, { wrapper: createWrapper() })
+
+      await waitFor(() => {
+        expect(screen.getByText(/embedding\.processing/i)).toBeInTheDocument()
+      })
+    })
+
+    it('should show completed status', async () => {
+      mockFetchIndexingStatus.mockResolvedValue(mockIndexingStatus({ indexing_status: 'completed' }))
+
+      render(<EmbeddingDetail {...defaultProps} />, { wrapper: createWrapper() })
+
+      await waitFor(() => {
+        expect(screen.getByText(/embedding\.completed/i)).toBeInTheDocument()
+      })
+    })
+
+    it('should show paused status', async () => {
+      mockFetchIndexingStatus.mockResolvedValue(mockIndexingStatus({ indexing_status: 'paused' }))
+
+      render(<EmbeddingDetail {...defaultProps} />, { wrapper: createWrapper() })
+
+      await waitFor(() => {
+        expect(screen.getByText(/embedding\.paused/i)).toBeInTheDocument()
+      })
+    })
+
+    it('should show error status', async () => {
+      mockFetchIndexingStatus.mockResolvedValue(mockIndexingStatus({ indexing_status: 'error' }))
+
+      render(<EmbeddingDetail {...defaultProps} />, { wrapper: createWrapper() })
+
+      await waitFor(() => {
+        expect(screen.getByText(/embedding\.error/i)).toBeInTheDocument()
+      })
+    })
+  })
+
+  describe('Progress Display', () => {
+    it('should display segment progress', async () => {
+      mockFetchIndexingStatus.mockResolvedValue(mockIndexingStatus({
+        completed_segments: 50,
+        total_segments: 100,
+      }))
+
+      render(<EmbeddingDetail {...defaultProps} />, { wrapper: createWrapper() })
+
+      await waitFor(() => {
+        expect(screen.getByText(/50\/100/)).toBeInTheDocument()
+        expect(screen.getByText(/50%/)).toBeInTheDocument()
+      })
+    })
+  })
+
+  describe('Pause/Resume Actions', () => {
+    it('should show pause button when embedding is in progress', async () => {
+      mockFetchIndexingStatus.mockResolvedValue(mockIndexingStatus({ indexing_status: 'indexing' }))
+
+      render(<EmbeddingDetail {...defaultProps} />, { wrapper: createWrapper() })
+
+      await waitFor(() => {
+        expect(screen.getByText(/embedding\.pause/i)).toBeInTheDocument()
+      })
+    })
+
+    it('should show resume button when paused', async () => {
+      mockFetchIndexingStatus.mockResolvedValue(mockIndexingStatus({ indexing_status: 'paused' }))
+
+      render(<EmbeddingDetail {...defaultProps} />, { wrapper: createWrapper() })
+
+      await waitFor(() => {
+        expect(screen.getByText(/embedding\.resume/i)).toBeInTheDocument()
+      })
+    })
+
+    it('should call pause API when pause button is clicked', async () => {
+      const user = userEvent.setup()
+      mockFetchIndexingStatus.mockResolvedValue(mockIndexingStatus({ indexing_status: 'indexing' }))
+      mockPauseDocIndexing.mockResolvedValue({ result: 'success' })
+
+      render(<EmbeddingDetail {...defaultProps} />, { wrapper: createWrapper() })
+
+      await waitFor(() => {
+        expect(screen.getByText(/embedding\.pause/i)).toBeInTheDocument()
+      })
+
+      await user.click(screen.getByRole('button', { name: /pause/i }))
+
+      await waitFor(() => {
+        expect(mockPauseDocIndexing).toHaveBeenCalledWith({
+          datasetId: 'ds1',
+          documentId: 'doc1',
+        })
+      })
+    })
+
+    it('should call resume API when resume button is clicked', async () => {
+      const user = userEvent.setup()
+      mockFetchIndexingStatus.mockResolvedValue(mockIndexingStatus({ indexing_status: 'paused' }))
+      mockResumeDocIndexing.mockResolvedValue({ result: 'success' })
+
+      render(<EmbeddingDetail {...defaultProps} />, { wrapper: createWrapper() })
+
+      await waitFor(() => {
+        expect(screen.getByText(/embedding\.resume/i)).toBeInTheDocument()
+      })
+
+      await user.click(screen.getByRole('button', { name: /resume/i }))
+
+      await waitFor(() => {
+        expect(mockResumeDocIndexing).toHaveBeenCalledWith({
+          datasetId: 'ds1',
+          documentId: 'doc1',
+        })
+      })
+    })
+  })
+
+  describe('Rule Detail', () => {
+    it('should display rule detail section', async () => {
+      mockFetchIndexingStatus.mockResolvedValue(mockIndexingStatus())
+
+      render(<EmbeddingDetail {...defaultProps} />, { wrapper: createWrapper() })
+
+      await waitFor(() => {
+        expect(screen.getByText(/stepTwo\.indexMode/i)).toBeInTheDocument()
+      })
+    })
+
+    it('should display qualified index mode', async () => {
+      mockFetchIndexingStatus.mockResolvedValue(mockIndexingStatus())
+
+      render(
+        <EmbeddingDetail {...defaultProps} indexingType={IndexingType.QUALIFIED} />,
+        { wrapper: createWrapper() },
+      )
+
+      await waitFor(() => {
+        expect(screen.getByText(/stepTwo\.qualified/i)).toBeInTheDocument()
+      })
+    })
+
+    it('should display economical index mode', async () => {
+      mockFetchIndexingStatus.mockResolvedValue(mockIndexingStatus())
+
+      render(
+        <EmbeddingDetail {...defaultProps} indexingType={IndexingType.ECONOMICAL} />,
+        { wrapper: createWrapper() },
+      )
+
+      await waitFor(() => {
+        expect(screen.getByText(/stepTwo\.economical/i)).toBeInTheDocument()
+      })
+    })
+  })
+
+  describe('detailUpdate Callback', () => {
+    it('should call detailUpdate when status becomes terminal', async () => {
+      const detailUpdate = vi.fn()
+      // First call returns indexing, subsequent call returns completed
+      mockFetchIndexingStatus
+        .mockResolvedValueOnce(mockIndexingStatus({ indexing_status: 'indexing' }))
+        .mockResolvedValueOnce(mockIndexingStatus({ indexing_status: 'completed' }))
+
+      render(
+        <EmbeddingDetail {...defaultProps} detailUpdate={detailUpdate} />,
+        { wrapper: createWrapper() },
+      )
+
+      // Wait for the terminal status to trigger detailUpdate
+      await waitFor(() => {
+        expect(mockFetchIndexingStatus).toHaveBeenCalled()
+      }, { timeout: 5000 })
+    })
+  })
+
+  describe('Edge Cases', () => {
+    it('should handle missing context values', async () => {
+      mockFetchIndexingStatus.mockResolvedValue(mockIndexingStatus())
+
+      render(
+        <EmbeddingDetail {...defaultProps} datasetId="explicit-ds" documentId="explicit-doc" />,
+        { wrapper: createWrapper({ datasetId: undefined, documentId: undefined }) },
+      )
+
+      await waitFor(() => {
+        expect(mockFetchIndexingStatus).toHaveBeenCalledWith({
+          datasetId: 'explicit-ds',
+          documentId: 'explicit-doc',
+        })
+      })
+    })
+
+    it('should render skeleton component', async () => {
+      mockFetchIndexingStatus.mockResolvedValue(mockIndexingStatus())
+
+      const { container } = render(<EmbeddingDetail {...defaultProps} />, { wrapper: createWrapper() })
+
+      // EmbeddingSkeleton should be rendered - check for the skeleton wrapper element
+      await waitFor(() => {
+        const skeletonWrapper = container.querySelector('.bg-dataset-chunk-list-mask-bg')
+        expect(skeletonWrapper).toBeInTheDocument()
+      })
+    })
+  })
+})

web/app/components/datasets/documents/detail/embedding/index.tsx

@@ -1,31 +1,18 @@
 import type { FC } from 'react'
-import type { CommonResponse } from '@/models/common'
-import type { IndexingStatusResponse, ProcessRuleResponse } from '@/models/datasets'
-import { RiLoader2Line, RiPauseCircleLine, RiPlayCircleLine } from '@remixicon/react'
-import Image from 'next/image'
+import type { IndexingType } from '../../../create/step-two'
+import type { RETRIEVE_METHOD } from '@/types/app'
 import * as React from 'react'
-import { useCallback, useEffect, useMemo, useRef, useState } from 'react'
+import { useCallback } from 'react'
 import { useTranslation } from 'react-i18next'
 import { useContext } from 'use-context-selector'
-import Divider from '@/app/components/base/divider'
 import { ToastContext } from '@/app/components/base/toast'
-import { ProcessMode } from '@/models/datasets'
-import {
-  fetchIndexingStatus as doFetchIndexingStatus,
-  pauseDocIndexing,
-  resumeDocIndexing,
-} from '@/service/datasets'
 import { useProcessRule } from '@/service/knowledge/use-dataset'
-import { RETRIEVE_METHOD } from '@/types/app'
-import { asyncRunSafe, sleep } from '@/utils'
-import { cn } from '@/utils/classnames'
-import { indexMethodIcon, retrievalIcon } from '../../../create/icons'
-import { IndexingType } from '../../../create/step-two'
 import { useDocumentContext } from '../context'
-import { FieldInfo } from '../metadata'
+import { ProgressBar, RuleDetail, SegmentProgress, StatusHeader } from './components'
+import { useEmbeddingStatus, usePauseIndexing, useResumeIndexing } from './hooks'
 import EmbeddingSkeleton from './skeleton'
 
-type IEmbeddingDetailProps = {
+type EmbeddingDetailProps = {
   datasetId?: string
   documentId?: string
   indexingType?: IndexingType
@@ -33,128 +20,7 @@ type IEmbeddingDetailProps = {
   detailUpdate: VoidFunction
 }
 
-type IRuleDetailProps = {
-  sourceData?: ProcessRuleResponse
-  indexingType?: IndexingType
-  retrievalMethod?: RETRIEVE_METHOD
-}
-
-const RuleDetail: FC<IRuleDetailProps> = React.memo(({
-  sourceData,
-  indexingType,
-  retrievalMethod,
-}) => {
-  const { t } = useTranslation()
-
-  const segmentationRuleMap = {
-    mode: t('embedding.mode', { ns: 'datasetDocuments' }),
-    segmentLength: t('embedding.segmentLength', { ns: 'datasetDocuments' }),
-    textCleaning: t('embedding.textCleaning', { ns: 'datasetDocuments' }),
-  }
-
-  const getRuleName = (key: string) => {
-    if (key === 'remove_extra_spaces')
-      return t('stepTwo.removeExtraSpaces', { ns: 'datasetCreation' })
-
-    if (key === 'remove_urls_emails')
-      return t('stepTwo.removeUrlEmails', { ns: 'datasetCreation' })
-
-    if (key === 'remove_stopwords')
-      return t('stepTwo.removeStopwords', { ns: 'datasetCreation' })
-  }
-
-  const isNumber = (value: unknown) => {
-    return typeof value === 'number'
-  }
-
-  const getValue = useCallback((field: string) => {
-    let value: string | number | undefined = '-'
-    const maxTokens = isNumber(sourceData?.rules?.segmentation?.max_tokens)
-      ? sourceData.rules.segmentation.max_tokens
-      : value
-    const childMaxTokens = isNumber(sourceData?.rules?.subchunk_segmentation?.max_tokens)
-      ? sourceData.rules.subchunk_segmentation.max_tokens
-      : value
-    switch (field) {
-      case 'mode':
-        value = !sourceData?.mode
-          ? value
-          : sourceData.mode === ProcessMode.general
-            ? (t('embedding.custom', { ns: 'datasetDocuments' }) as string)
-            : `${t('embedding.hierarchical', { ns: 'datasetDocuments' })} · ${sourceData?.rules?.parent_mode === 'paragraph'
-              ? t('parentMode.paragraph', { ns: 'dataset' })
-              : t('parentMode.fullDoc', { ns: 'dataset' })}`
-        break
-      case 'segmentLength':
-        value = !sourceData?.mode
-          ? value
-          : sourceData.mode === ProcessMode.general
-            ? maxTokens
-            : `${t('embedding.parentMaxTokens', { ns: 'datasetDocuments' })} ${maxTokens}; ${t('embedding.childMaxTokens', { ns: 'datasetDocuments' })} ${childMaxTokens}`
-        break
-      default:
-        value = !sourceData?.mode
-          ? value
-          : sourceData?.rules?.pre_processing_rules?.filter(rule =>
-              rule.enabled).map(rule => getRuleName(rule.id)).join(',')
-        break
-    }
-    return value
-  }, [sourceData])
-
-  return (
-    <div className="py-3">
-      <div className="flex flex-col gap-y-1">
-        {Object.keys(segmentationRuleMap).map((field) => {
-          return (
-            <FieldInfo
-              key={field}
-              label={segmentationRuleMap[field as keyof typeof segmentationRuleMap]}
-              displayedValue={String(getValue(field))}
-            />
-          )
-        })}
-      </div>
-      <Divider type="horizontal" className="bg-divider-subtle" />
-      <FieldInfo
-        label={t('stepTwo.indexMode', { ns: 'datasetCreation' })}
-        displayedValue={t(`stepTwo.${indexingType === IndexingType.ECONOMICAL ? 'economical' : 'qualified'}`, { ns: 'datasetCreation' }) as string}
-        valueIcon={(
-          <Image
-            className="size-4"
-            src={
-              indexingType === IndexingType.ECONOMICAL
-                ? indexMethodIcon.economical
-                : indexMethodIcon.high_quality
-            }
-            alt=""
-          />
-        )}
-      />
-      <FieldInfo
-        label={t('form.retrievalSetting.title', { ns: 'datasetSettings' })}
-        displayedValue={t(`retrieval.${indexingType === IndexingType.ECONOMICAL ? 'keyword_search' : retrievalMethod ?? 'semantic_search'}.title`, { ns: 'dataset' })}
-        valueIcon={(
-          <Image
-            className="size-4"
-            src={
-              retrievalMethod === RETRIEVE_METHOD.fullText
-                ? retrievalIcon.fullText
-                : retrievalMethod === RETRIEVE_METHOD.hybrid
-                  ? retrievalIcon.hybrid
-                  : retrievalIcon.vector
-            }
-            alt=""
-          />
-        )}
-      />
-    </div>
-  )
-})
-
-RuleDetail.displayName = 'RuleDetail'
-
-const EmbeddingDetail: FC<IEmbeddingDetailProps> = ({
+const EmbeddingDetail: FC<EmbeddingDetailProps> = ({
   datasetId: dstId,
   documentId: docId,
   detailUpdate,
@@ -164,144 +30,95 @@ const EmbeddingDetail: FC<IEmbeddingDetailProps> = ({
   const { t } = useTranslation()
   const { notify } = useContext(ToastContext)
 
-  const datasetId = useDocumentContext(s => s.datasetId)
-  const documentId = useDocumentContext(s => s.documentId)
-  const localDatasetId = dstId ?? datasetId
-  const localDocumentId = docId ?? documentId
+  const contextDatasetId = useDocumentContext(s => s.datasetId)
+  const contextDocumentId = useDocumentContext(s => s.documentId)
+  const datasetId = dstId ?? contextDatasetId
+  const documentId = docId ?? contextDocumentId
 
-  const [indexingStatusDetail, setIndexingStatusDetail] = useState<IndexingStatusResponse | null>(null)
-  const fetchIndexingStatus = async () => {
-    const status = await doFetchIndexingStatus({ datasetId: localDatasetId, documentId: localDocumentId })
-    setIndexingStatusDetail(status)
-    return status
-  }
+  const {
+    data: indexingStatus,
+    isEmbedding,
+    isCompleted,
+    isPaused,
+    isError,
+    percent,
+    resetStatus,
+    refetch,
+  } = useEmbeddingStatus({
+    datasetId,
+    documentId,
+    onComplete: detailUpdate,
+  })
 
-  const isStopQuery = useRef(false)
-  const stopQueryStatus = useCallback(() => {
-    isStopQuery.current = true
-  }, [])
+  const { data: ruleDetail } = useProcessRule(documentId)
 
-  const startQueryStatus = useCallback(async () => {
-    if (isStopQuery.current)
-      return
+  const handleSuccess = useCallback(() => {
+    notify({ type: 'success', message: t('actionMsg.modifiedSuccessfully', { ns: 'common' }) })
+  }, [notify, t])
 
-    try {
-      const indexingStatusDetail = await fetchIndexingStatus()
-      if (['completed', 'error', 'paused'].includes(indexingStatusDetail?.indexing_status)) {
-        stopQueryStatus()
-        detailUpdate()
-        return
-      }
+  const handleError = useCallback(() => {
+    notify({ type: 'error', message: t('actionMsg.modifiedUnsuccessfully', { ns: 'common' }) })
+  }, [notify, t])
 
-      await sleep(2500)
-      await startQueryStatus()
-    }
-    catch {
-      await sleep(2500)
-      await startQueryStatus()
-    }
-  }, [stopQueryStatus])
+  const pauseMutation = usePauseIndexing({
+    datasetId,
+    documentId,
+    onSuccess: () => {
+      handleSuccess()
+      resetStatus()
+    },
+    onError: handleError,
+  })
 
-  useEffect(() => {
-    isStopQuery.current = false
-    startQueryStatus()
-    return () => {
-      stopQueryStatus()
-    }
-  }, [startQueryStatus, stopQueryStatus])
+  const resumeMutation = useResumeIndexing({
+    datasetId,
+    documentId,
+    onSuccess: () => {
+      handleSuccess()
+      refetch()
+      detailUpdate()
+    },
+    onError: handleError,
+  })
 
-  const { data: ruleDetail } = useProcessRule(localDocumentId)
+  const handlePause = useCallback(() => {
+    pauseMutation.mutate()
+  }, [pauseMutation])
 
-  const isEmbedding = useMemo(() => ['indexing', 'splitting', 'parsing', 'cleaning'].includes(indexingStatusDetail?.indexing_status || ''), [indexingStatusDetail])
-  const isEmbeddingCompleted = useMemo(() => ['completed'].includes(indexingStatusDetail?.indexing_status || ''), [indexingStatusDetail])
-  const isEmbeddingPaused = useMemo(() => ['paused'].includes(indexingStatusDetail?.indexing_status || ''), [indexingStatusDetail])
-  const isEmbeddingError = useMemo(() => ['error'].includes(indexingStatusDetail?.indexing_status || ''), [indexingStatusDetail])
-  const percent = useMemo(() => {
-    const completedCount = indexingStatusDetail?.completed_segments || 0
-    const totalCount = indexingStatusDetail?.total_segments || 0
-    if (totalCount === 0)
-      return 0
-    const percent = Math.round(completedCount * 100 / totalCount)
-    return percent > 100 ? 100 : percent
-  }, [indexingStatusDetail])
-
-  const handleSwitch = async () => {
-    const opApi = isEmbedding ? pauseDocIndexing : resumeDocIndexing
-    const [e] = await asyncRunSafe<CommonResponse>(opApi({ datasetId: localDatasetId, documentId: localDocumentId }) as Promise<CommonResponse>)
-    if (!e) {
-      notify({ type: 'success', message: t('actionMsg.modifiedSuccessfully', { ns: 'common' }) })
-      // if the embedding is resumed from paused, we need to start the query status
-      if (isEmbeddingPaused) {
-        isStopQuery.current = false
-        startQueryStatus()
-        detailUpdate()
-      }
-      setIndexingStatusDetail(null)
-    }
-    else {
-      notify({ type: 'error', message: t('actionMsg.modifiedUnsuccessfully', { ns: 'common' }) })
-    }
-  }
+  const handleResume = useCallback(() => {
+    resumeMutation.mutate()
+  }, [resumeMutation])
 
   return (
     <>
       <div className="flex flex-col gap-y-2 px-16 py-12">
-        <div className="flex h-6 items-center gap-x-1">
-          {isEmbedding && <RiLoader2Line className="h-4 w-4 animate-spin text-text-secondary" />}
-          <span className="system-md-semibold-uppercase grow text-text-secondary">
-            {isEmbedding && t('embedding.processing', { ns: 'datasetDocuments' })}
-            {isEmbeddingCompleted && t('embedding.completed', { ns: 'datasetDocuments' })}
-            {isEmbeddingPaused && t('embedding.paused', { ns: 'datasetDocuments' })}
-            {isEmbeddingError && t('embedding.error', { ns: 'datasetDocuments' })}
-          </span>
-          {isEmbedding && (
-            <button
-              type="button"
-              className={`flex items-center gap-x-1 rounded-md border-[0.5px]
-              border-components-button-secondary-border bg-components-button-secondary-bg px-1.5 py-1 shadow-xs shadow-shadow-shadow-3 backdrop-blur-[5px]`}
-              onClick={handleSwitch}
-            >
-              <RiPauseCircleLine className="h-3.5 w-3.5 text-components-button-secondary-text" />
-              <span className="system-xs-medium pr-[3px] text-components-button-secondary-text">
-                {t('embedding.pause', { ns: 'datasetDocuments' })}
-              </span>
-            </button>
-          )}
-          {isEmbeddingPaused && (
-            <button
-              type="button"
-              className={`flex items-center gap-x-1 rounded-md border-[0.5px]
-              border-components-button-secondary-border bg-components-button-secondary-bg px-1.5 py-1 shadow-xs shadow-shadow-shadow-3 backdrop-blur-[5px]`}
-              onClick={handleSwitch}
-            >
-              <RiPlayCircleLine className="h-3.5 w-3.5 text-components-button-secondary-text" />
-              <span className="system-xs-medium pr-[3px] text-components-button-secondary-text">
-                {t('embedding.resume', { ns: 'datasetDocuments' })}
-              </span>
-            </button>
-          )}
-        </div>
-        {/* progress bar */}
-        <div className={cn(
-          'flex h-2 w-full items-center overflow-hidden rounded-md border border-components-progress-bar-border',
-          isEmbedding ? 'bg-components-progress-bar-bg/50' : 'bg-components-progress-bar-bg',
-        )}
-        >
-          <div
-            className={cn(
-              'h-full',
-              (isEmbedding || isEmbeddingCompleted) && 'bg-components-progress-bar-progress-solid',
-              (isEmbeddingPaused || isEmbeddingError) && 'bg-components-progress-bar-progress-highlight',
-            )}
-            style={{ width: `${percent}%` }}
-          />
-        </div>
-        <div className="flex w-full items-center">
-          <span className="system-xs-medium text-text-secondary">
-            {`${t('embedding.segments', { ns: 'datasetDocuments' })} ${indexingStatusDetail?.completed_segments || '--'}/${indexingStatusDetail?.total_segments || '--'} · ${percent}%`}
-          </span>
-        </div>
-        <RuleDetail sourceData={ruleDetail} indexingType={indexingType} retrievalMethod={retrievalMethod} />
+        <StatusHeader
+          isEmbedding={isEmbedding}
+          isCompleted={isCompleted}
+          isPaused={isPaused}
+          isError={isError}
+          onPause={handlePause}
+          onResume={handleResume}
+          isPauseLoading={pauseMutation.isPending}
+          isResumeLoading={resumeMutation.isPending}
+        />
+        <ProgressBar
+          percent={percent}
+          isEmbedding={isEmbedding}
+          isCompleted={isCompleted}
+          isPaused={isPaused}
+          isError={isError}
+        />
+        <SegmentProgress
+          completedSegments={indexingStatus?.completed_segments}
+          totalSegments={indexingStatus?.total_segments}
+          percent={percent}
+        />
+        <RuleDetail
+          sourceData={ruleDetail}
+          indexingType={indexingType}
+          retrievalMethod={retrievalMethod}
+        />
       </div>
       <EmbeddingSkeleton />
     </>

web/app/components/datasets/list/dataset-card/components/dataset-card-header.spec.tsx

@@ -6,6 +6,13 @@ import { ChunkingMode, DatasetPermission, DataSourceType } from '@/models/datase
 import { RETRIEVE_METHOD } from '@/types/app'
 import DatasetCardHeader from './dataset-card-header'
 
+// Mock AppIcon component to avoid emoji-mart initialization issues
+vi.mock('@/app/components/base/app-icon', () => ({
+  default: ({ icon, className }: { icon?: string, className?: string }) => (
+    <div data-testid="app-icon" className={className}>{icon}</div>
+  ),
+}))
+
 // Mock useFormatTimeFromNow hook
 vi.mock('@/hooks/use-format-time-from-now', () => ({
   useFormatTimeFromNow: () => ({

web/app/components/datasets/list/dataset-card/components/dataset-card-modals.spec.tsx

@@ -19,6 +19,28 @@ vi.mock('../../../rename-modal', () => ({
   ),
 }))
 
+// Mock Confirm component since it uses createPortal which can cause issues in tests
+vi.mock('@/app/components/base/confirm', () => ({
+  default: ({ isShow, title, content, onConfirm, onCancel }: {
+    isShow: boolean
+    title: string
+    content?: React.ReactNode
+    onConfirm: () => void
+    onCancel: () => void
+  }) => (
+    isShow
+      ? (
+          <div data-testid="confirm-modal">
+            <div data-testid="confirm-title">{title}</div>
+            <div data-testid="confirm-content">{content}</div>
+            <button onClick={onCancel} role="button" aria-label="cancel">Cancel</button>
+            <button onClick={onConfirm} role="button" aria-label="confirm">Confirm</button>
+          </div>
+        )
+      : null
+  ),
+}))
+
 describe('DatasetCardModals', () => {
   const mockDataset: DataSet = {
     id: 'dataset-1',
@@ -172,11 +194,9 @@ describe('DatasetCardModals', () => {
         />,
       )
 
-      // Find and click the confirm button
-      const confirmButton = screen.getByRole('button', { name: /confirm|ok|delete/i })
-        || screen.getAllByRole('button').find(btn => btn.textContent?.toLowerCase().includes('confirm'))
-      if (confirmButton)
-        fireEvent.click(confirmButton)
+      // Find and click the confirm button using our mocked Confirm component
+      const confirmButton = screen.getByRole('button', { name: /confirm/i })
+      fireEvent.click(confirmButton)
 
       expect(onConfirmDelete).toHaveBeenCalledTimes(1)
     })

web/app/components/datasets/settings/form/components/basic-info-section.spec.tsx

@@ -0,0 +1,441 @@
+import type { Member } from '@/models/common'
+import type { DataSet, IconInfo } from '@/models/datasets'
+import type { RetrievalConfig } from '@/types/app'
+import { fireEvent, render, screen, waitFor } from '@testing-library/react'
+import { ChunkingMode, DatasetPermission, DataSourceType } from '@/models/datasets'
+import { RETRIEVE_METHOD } from '@/types/app'
+import { IndexingType } from '../../../create/step-two'
+import BasicInfoSection from './basic-info-section'
+
+// Mock app-context
+vi.mock('@/context/app-context', () => ({
+  useSelector: () => ({
+    id: 'user-1',
+    name: 'Current User',
+    email: 'current@example.com',
+    avatar_url: '',
+    role: 'owner',
+  }),
+}))
+
+// Mock image uploader hooks for AppIconPicker
+vi.mock('@/app/components/base/image-uploader/hooks', () => ({
+  useLocalFileUploader: () => ({
+    disabled: false,
+    handleLocalFileUpload: vi.fn(),
+  }),
+  useImageFiles: () => ({
+    files: [],
+    onUpload: vi.fn(),
+    onRemove: vi.fn(),
+    onReUpload: vi.fn(),
+    onImageLinkLoadError: vi.fn(),
+    onImageLinkLoadSuccess: vi.fn(),
+    onClear: vi.fn(),
+  }),
+}))
+
+describe('BasicInfoSection', () => {
+  const mockDataset: DataSet = {
+    id: 'dataset-1',
+    name: 'Test Dataset',
+    description: 'Test description',
+    permission: DatasetPermission.onlyMe,
+    icon_info: {
+      icon_type: 'emoji',
+      icon: '📚',
+      icon_background: '#FFFFFF',
+      icon_url: '',
+    },
+    indexing_technique: IndexingType.QUALIFIED,
+    indexing_status: 'completed',
+    data_source_type: DataSourceType.FILE,
+    doc_form: ChunkingMode.text,
+    embedding_model: 'text-embedding-ada-002',
+    embedding_model_provider: 'openai',
+    embedding_available: true,
+    app_count: 0,
+    document_count: 5,
+    total_document_count: 5,
+    word_count: 1000,
+    provider: 'vendor',
+    tags: [],
+    partial_member_list: [],
+    external_knowledge_info: {
+      external_knowledge_id: 'ext-1',
+      external_knowledge_api_id: 'api-1',
+      external_knowledge_api_name: 'External API',
+      external_knowledge_api_endpoint: 'https://api.example.com',
+    },
+    external_retrieval_model: {
+      top_k: 3,
+      score_threshold: 0.7,
+      score_threshold_enabled: true,
+    },
+    retrieval_model_dict: {
+      search_method: RETRIEVE_METHOD.semantic,
+      reranking_enable: false,
+      reranking_model: {
+        reranking_provider_name: '',
+        reranking_model_name: '',
+      },
+      top_k: 3,
+      score_threshold_enabled: false,
+      score_threshold: 0.5,
+    } as RetrievalConfig,
+    retrieval_model: {
+      search_method: RETRIEVE_METHOD.semantic,
+      reranking_enable: false,
+      reranking_model: {
+        reranking_provider_name: '',
+        reranking_model_name: '',
+      },
+      top_k: 3,
+      score_threshold_enabled: false,
+      score_threshold: 0.5,
+    } as RetrievalConfig,
+    built_in_field_enabled: false,
+    keyword_number: 10,
+    created_by: 'user-1',
+    updated_by: 'user-1',
+    updated_at: Date.now(),
+    runtime_mode: 'general',
+    enable_api: true,
+    is_multimodal: false,
+  }
+
+  const mockMemberList: Member[] = [
+    { id: 'user-1', name: 'User 1', email: 'user1@example.com', role: 'owner', avatar: '', avatar_url: '', last_login_at: '', created_at: '', status: 'active' },
+    { id: 'user-2', name: 'User 2', email: 'user2@example.com', role: 'admin', avatar: '', avatar_url: '', last_login_at: '', created_at: '', status: 'active' },
+  ]
+
+  const mockIconInfo: IconInfo = {
+    icon_type: 'emoji',
+    icon: '📚',
+    icon_background: '#FFFFFF',
+    icon_url: '',
+  }
+
+  const defaultProps = {
+    currentDataset: mockDataset,
+    isCurrentWorkspaceDatasetOperator: false,
+    name: 'Test Dataset',
+    setName: vi.fn(),
+    description: 'Test description',
+    setDescription: vi.fn(),
+    iconInfo: mockIconInfo,
+    showAppIconPicker: false,
+    handleOpenAppIconPicker: vi.fn(),
+    handleSelectAppIcon: vi.fn(),
+    handleCloseAppIconPicker: vi.fn(),
+    permission: DatasetPermission.onlyMe,
+    setPermission: vi.fn(),
+    selectedMemberIDs: ['user-1'],
+    setSelectedMemberIDs: vi.fn(),
+    memberList: mockMemberList,
+  }
+
+  beforeEach(() => {
+    vi.clearAllMocks()
+  })
+
+  describe('Rendering', () => {
+    it('should render without crashing', () => {
+      render(<BasicInfoSection {...defaultProps} />)
+      expect(screen.getByText(/form\.nameAndIcon/i)).toBeInTheDocument()
+    })
+
+    it('should render name and icon section', () => {
+      render(<BasicInfoSection {...defaultProps} />)
+      expect(screen.getByText(/form\.nameAndIcon/i)).toBeInTheDocument()
+    })
+
+    it('should render description section', () => {
+      render(<BasicInfoSection {...defaultProps} />)
+      expect(screen.getByText(/form\.desc/i)).toBeInTheDocument()
+    })
+
+    it('should render permissions section', () => {
+      render(<BasicInfoSection {...defaultProps} />)
+      // Use exact match to avoid matching "permissionsOnlyMe"
+      expect(screen.getByText('datasetSettings.form.permissions')).toBeInTheDocument()
+    })
+
+    it('should render name input with correct value', () => {
+      render(<BasicInfoSection {...defaultProps} />)
+      const nameInput = screen.getByDisplayValue('Test Dataset')
+      expect(nameInput).toBeInTheDocument()
+    })
+
+    it('should render description textarea with correct value', () => {
+      render(<BasicInfoSection {...defaultProps} />)
+      const descriptionTextarea = screen.getByDisplayValue('Test description')
+      expect(descriptionTextarea).toBeInTheDocument()
+    })
+
+    it('should render app icon with emoji', () => {
+      const { container } = render(<BasicInfoSection {...defaultProps} />)
+      // The icon section should be rendered (emoji may be in a span or SVG)
+      const iconSection = container.querySelector('[class*="cursor-pointer"]')
+      expect(iconSection).toBeInTheDocument()
+    })
+  })
+
+  describe('Name Input', () => {
+    it('should call setName when name input changes', () => {
+      const setName = vi.fn()
+      render(<BasicInfoSection {...defaultProps} setName={setName} />)
+
+      const nameInput = screen.getByDisplayValue('Test Dataset')
+      fireEvent.change(nameInput, { target: { value: 'New Name' } })
+
+      expect(setName).toHaveBeenCalledWith('New Name')
+    })
+
+    it('should disable name input when embedding is not available', () => {
+      const datasetWithoutEmbedding = { ...mockDataset, embedding_available: false }
+      render(<BasicInfoSection {...defaultProps} currentDataset={datasetWithoutEmbedding} />)
+
+      const nameInput = screen.getByDisplayValue('Test Dataset')
+      expect(nameInput).toBeDisabled()
+    })
+
+    it('should enable name input when embedding is available', () => {
+      render(<BasicInfoSection {...defaultProps} />)
+
+      const nameInput = screen.getByDisplayValue('Test Dataset')
+      expect(nameInput).not.toBeDisabled()
+    })
+
+    it('should display empty name', () => {
+      const { container } = render(<BasicInfoSection {...defaultProps} name="" />)
+
+      // Find the name input by its structure - may be type=text or just input
+      const nameInput = container.querySelector('input')
+      expect(nameInput).toHaveValue('')
+    })
+  })
+
+  describe('Description Textarea', () => {
+    it('should call setDescription when description changes', () => {
+      const setDescription = vi.fn()
+      render(<BasicInfoSection {...defaultProps} setDescription={setDescription} />)
+
+      const descriptionTextarea = screen.getByDisplayValue('Test description')
+      fireEvent.change(descriptionTextarea, { target: { value: 'New Description' } })
+
+      expect(setDescription).toHaveBeenCalledWith('New Description')
+    })
+
+    it('should disable description textarea when embedding is not available', () => {
+      const datasetWithoutEmbedding = { ...mockDataset, embedding_available: false }
+      render(<BasicInfoSection {...defaultProps} currentDataset={datasetWithoutEmbedding} />)
+
+      const descriptionTextarea = screen.getByDisplayValue('Test description')
+      expect(descriptionTextarea).toBeDisabled()
+    })
+
+    it('should render placeholder', () => {
+      render(<BasicInfoSection {...defaultProps} description="" />)
+
+      const descriptionTextarea = screen.getByPlaceholderText(/form\.descPlaceholder/i)
+      expect(descriptionTextarea).toBeInTheDocument()
+    })
+  })
+
+  describe('App Icon', () => {
+    it('should call handleOpenAppIconPicker when icon is clicked', () => {
+      const handleOpenAppIconPicker = vi.fn()
+      const { container } = render(<BasicInfoSection {...defaultProps} handleOpenAppIconPicker={handleOpenAppIconPicker} />)
+
+      // Find the clickable icon element - it's inside a wrapper that handles the click
+      const iconWrapper = container.querySelector('[class*="cursor-pointer"]')
+      if (iconWrapper) {
+        fireEvent.click(iconWrapper)
+        expect(handleOpenAppIconPicker).toHaveBeenCalled()
+      }
+    })
+
+    it('should render AppIconPicker when showAppIconPicker is true', () => {
+      const { baseElement } = render(<BasicInfoSection {...defaultProps} showAppIconPicker={true} />)
+
+      // AppIconPicker renders a modal with emoji tabs and options via portal
+      // We just verify the component renders without crashing when picker is shown
+      expect(baseElement).toBeInTheDocument()
+    })
+
+    it('should not render AppIconPicker when showAppIconPicker is false', () => {
+      const { container } = render(<BasicInfoSection {...defaultProps} showAppIconPicker={false} />)
+
+      // Check that AppIconPicker is not rendered
+      expect(container.querySelector('[data-testid="app-icon-picker"]')).not.toBeInTheDocument()
+    })
+
+    it('should render image icon when icon_type is image', () => {
+      const imageIconInfo: IconInfo = {
+        icon_type: 'image',
+        icon: 'file-123',
+        icon_background: undefined,
+        icon_url: 'https://example.com/icon.png',
+      }
+      render(<BasicInfoSection {...defaultProps} iconInfo={imageIconInfo} />)
+
+      // For image type, it renders an img element
+      const img = screen.queryByRole('img')
+      if (img) {
+        expect(img).toHaveAttribute('src', expect.stringContaining('icon.png'))
+      }
+    })
+  })
+
+  describe('Permission Selector', () => {
+    it('should render with correct permission value', () => {
+      render(<BasicInfoSection {...defaultProps} permission={DatasetPermission.onlyMe} />)
+
+      expect(screen.getByText(/form\.permissionsOnlyMe/i)).toBeInTheDocument()
+    })
+
+    it('should render all team members permission', () => {
+      render(<BasicInfoSection {...defaultProps} permission={DatasetPermission.allTeamMembers} />)
+
+      expect(screen.getByText(/form\.permissionsAllMember/i)).toBeInTheDocument()
+    })
+
+    it('should be disabled when embedding is not available', () => {
+      const datasetWithoutEmbedding = { ...mockDataset, embedding_available: false }
+      const { container } = render(
+        <BasicInfoSection {...defaultProps} currentDataset={datasetWithoutEmbedding} />,
+      )
+
+      // Check for disabled state via cursor-not-allowed class
+      const disabledElement = container.querySelector('[class*="cursor-not-allowed"]')
+      expect(disabledElement).toBeInTheDocument()
+    })
+
+    it('should be disabled when user is dataset operator', () => {
+      const { container } = render(
+        <BasicInfoSection {...defaultProps} isCurrentWorkspaceDatasetOperator={true} />,
+      )
+
+      const disabledElement = container.querySelector('[class*="cursor-not-allowed"]')
+      expect(disabledElement).toBeInTheDocument()
+    })
+
+    it('should call setPermission when permission changes', async () => {
+      const setPermission = vi.fn()
+      render(<BasicInfoSection {...defaultProps} setPermission={setPermission} />)
+
+      // Open dropdown
+      const trigger = screen.getByText(/form\.permissionsOnlyMe/i)
+      fireEvent.click(trigger)
+
+      await waitFor(() => {
+        // Click All Team Members option
+        const allMemberOptions = screen.getAllByText(/form\.permissionsAllMember/i)
+        fireEvent.click(allMemberOptions[0])
+      })
+
+      expect(setPermission).toHaveBeenCalledWith(DatasetPermission.allTeamMembers)
+    })
+
+    it('should call setSelectedMemberIDs when members are selected', async () => {
+      const setSelectedMemberIDs = vi.fn()
+      const { container } = render(
+        <BasicInfoSection
+          {...defaultProps}
+          permission={DatasetPermission.partialMembers}
+          setSelectedMemberIDs={setSelectedMemberIDs}
+        />,
+      )
+
+      // For partial members permission, the member selector should be visible
+      // The exact interaction depends on the MemberSelector component
+      // We verify the component renders without crashing
+      expect(container).toBeInTheDocument()
+    })
+  })
+
+  describe('Undefined Dataset', () => {
+    it('should handle undefined currentDataset gracefully', () => {
+      render(<BasicInfoSection {...defaultProps} currentDataset={undefined} />)
+
+      // Should still render but inputs might behave differently
+      expect(screen.getByText(/form\.nameAndIcon/i)).toBeInTheDocument()
+    })
+  })
+
+  describe('Props Validation', () => {
+    it('should update when name prop changes', () => {
+      const { rerender } = render(<BasicInfoSection {...defaultProps} name="Initial Name" />)
+
+      expect(screen.getByDisplayValue('Initial Name')).toBeInTheDocument()
+
+      rerender(<BasicInfoSection {...defaultProps} name="Updated Name" />)
+
+      expect(screen.getByDisplayValue('Updated Name')).toBeInTheDocument()
+    })
+
+    it('should update when description prop changes', () => {
+      const { rerender } = render(<BasicInfoSection {...defaultProps} description="Initial Description" />)
+
+      expect(screen.getByDisplayValue('Initial Description')).toBeInTheDocument()
+
+      rerender(<BasicInfoSection {...defaultProps} description="Updated Description" />)
+
+      expect(screen.getByDisplayValue('Updated Description')).toBeInTheDocument()
+    })
+
+    it('should update when permission prop changes', () => {
+      const { rerender } = render(<BasicInfoSection {...defaultProps} permission={DatasetPermission.onlyMe} />)
+
+      expect(screen.getByText(/form\.permissionsOnlyMe/i)).toBeInTheDocument()
+
+      rerender(<BasicInfoSection {...defaultProps} permission={DatasetPermission.allTeamMembers} />)
+
+      expect(screen.getByText(/form\.permissionsAllMember/i)).toBeInTheDocument()
+    })
+  })
+
+  describe('Member List', () => {
+    it('should pass member list to PermissionSelector', () => {
+      const { container } = render(
+        <BasicInfoSection
+          {...defaultProps}
+          permission={DatasetPermission.partialMembers}
+          memberList={mockMemberList}
+        />,
+      )
+
+      // For partial members, a member selector component should be rendered
+      // We verify it renders without crashing
+      expect(container).toBeInTheDocument()
+    })
+
+    it('should handle empty member list', () => {
+      render(
+        <BasicInfoSection
+          {...defaultProps}
+          memberList={[]}
+        />,
+      )
+
+      expect(screen.getByText(/form\.permissionsOnlyMe/i)).toBeInTheDocument()
+    })
+  })
+
+  describe('Accessibility', () => {
+    it('should have accessible name input', () => {
+      render(<BasicInfoSection {...defaultProps} />)
+
+      const nameInput = screen.getByDisplayValue('Test Dataset')
+      expect(nameInput.tagName.toLowerCase()).toBe('input')
+    })
+
+    it('should have accessible description textarea', () => {
+      render(<BasicInfoSection {...defaultProps} />)
+
+      const descriptionTextarea = screen.getByDisplayValue('Test description')
+      expect(descriptionTextarea.tagName.toLowerCase()).toBe('textarea')
+    })
+  })
+})

web/app/components/datasets/settings/form/components/basic-info-section.tsx

@@ -0,0 +1,124 @@
+'use client'
+import type { AppIconSelection } from '@/app/components/base/app-icon-picker'
+import type { Member } from '@/models/common'
+import type { DataSet, DatasetPermission, IconInfo } from '@/models/datasets'
+import type { AppIconType } from '@/types/app'
+import { useTranslation } from 'react-i18next'
+import AppIcon from '@/app/components/base/app-icon'
+import AppIconPicker from '@/app/components/base/app-icon-picker'
+import Input from '@/app/components/base/input'
+import Textarea from '@/app/components/base/textarea'
+import PermissionSelector from '../../permission-selector'
+
+const rowClass = 'flex gap-x-1'
+const labelClass = 'flex items-center shrink-0 w-[180px] h-7 pt-1'
+
+type BasicInfoSectionProps = {
+  currentDataset: DataSet | undefined
+  isCurrentWorkspaceDatasetOperator: boolean
+  name: string
+  setName: (value: string) => void
+  description: string
+  setDescription: (value: string) => void
+  iconInfo: IconInfo
+  showAppIconPicker: boolean
+  handleOpenAppIconPicker: () => void
+  handleSelectAppIcon: (icon: AppIconSelection) => void
+  handleCloseAppIconPicker: () => void
+  permission: DatasetPermission | undefined
+  setPermission: (value: DatasetPermission | undefined) => void
+  selectedMemberIDs: string[]
+  setSelectedMemberIDs: (value: string[]) => void
+  memberList: Member[]
+}
+
+const BasicInfoSection = ({
+  currentDataset,
+  isCurrentWorkspaceDatasetOperator,
+  name,
+  setName,
+  description,
+  setDescription,
+  iconInfo,
+  showAppIconPicker,
+  handleOpenAppIconPicker,
+  handleSelectAppIcon,
+  handleCloseAppIconPicker,
+  permission,
+  setPermission,
+  selectedMemberIDs,
+  setSelectedMemberIDs,
+  memberList,
+}: BasicInfoSectionProps) => {
+  const { t } = useTranslation()
+
+  return (
+    <>
+      {/* Dataset name and icon */}
+      <div className={rowClass}>
+        <div className={labelClass}>
+          <div className="system-sm-semibold text-text-secondary">{t('form.nameAndIcon', { ns: 'datasetSettings' })}</div>
+        </div>
+        <div className="flex grow items-center gap-x-2">
+          <AppIcon
+            size="small"
+            onClick={handleOpenAppIconPicker}
+            className="cursor-pointer"
+            iconType={iconInfo.icon_type as AppIconType}
+            icon={iconInfo.icon}
+            background={iconInfo.icon_background}
+            imageUrl={iconInfo.icon_url}
+            showEditIcon
+          />
+          <Input
+            disabled={!currentDataset?.embedding_available}
+            value={name}
+            onChange={e => setName(e.target.value)}
+          />
+        </div>
+      </div>
+
+      {/* Dataset description */}
+      <div className={rowClass}>
+        <div className={labelClass}>
+          <div className="system-sm-semibold text-text-secondary">{t('form.desc', { ns: 'datasetSettings' })}</div>
+        </div>
+        <div className="grow">
+          <Textarea
+            disabled={!currentDataset?.embedding_available}
+            className="resize-none"
+            placeholder={t('form.descPlaceholder', { ns: 'datasetSettings' }) || ''}
+            value={description}
+            onChange={e => setDescription(e.target.value)}
+          />
+        </div>
+      </div>
+
+      {/* Permissions */}
+      <div className={rowClass}>
+        <div className={labelClass}>
+          <div className="system-sm-semibold text-text-secondary">{t('form.permissions', { ns: 'datasetSettings' })}</div>
+        </div>
+        <div className="grow">
+          <PermissionSelector
+            disabled={!currentDataset?.embedding_available || isCurrentWorkspaceDatasetOperator}
+            permission={permission}
+            value={selectedMemberIDs}
+            onChange={v => setPermission(v)}
+            onMemberSelect={setSelectedMemberIDs}
+            memberList={memberList}
+          />
+        </div>
+      </div>
+
+      {showAppIconPicker && (
+        <AppIconPicker
+          onSelect={handleSelectAppIcon}
+          onClose={handleCloseAppIconPicker}
+        />
+      )}
+    </>
+  )
+}
+
+export default BasicInfoSection

web/app/components/datasets/settings/form/components/external-knowledge-section.spec.tsx

@@ -0,0 +1,362 @@
+import type { DataSet } from '@/models/datasets'
+import type { RetrievalConfig } from '@/types/app'
+import { render, screen } from '@testing-library/react'
+import { ChunkingMode, DatasetPermission, DataSourceType } from '@/models/datasets'
+import { RETRIEVE_METHOD } from '@/types/app'
+import { IndexingType } from '../../../create/step-two'
+import ExternalKnowledgeSection from './external-knowledge-section'
+
+describe('ExternalKnowledgeSection', () => {
+  const mockRetrievalConfig: RetrievalConfig = {
+    search_method: RETRIEVE_METHOD.semantic,
+    reranking_enable: false,
+    reranking_model: {
+      reranking_provider_name: '',
+      reranking_model_name: '',
+    },
+    top_k: 3,
+    score_threshold_enabled: false,
+    score_threshold: 0.5,
+  }
+
+  const mockDataset: DataSet = {
+    id: 'dataset-1',
+    name: 'External Dataset',
+    description: 'External dataset description',
+    permission: DatasetPermission.onlyMe,
+    icon_info: {
+      icon_type: 'emoji',
+      icon: '📚',
+      icon_background: '#FFFFFF',
+      icon_url: '',
+    },
+    indexing_technique: IndexingType.QUALIFIED,
+    indexing_status: 'completed',
+    data_source_type: DataSourceType.FILE,
+    doc_form: ChunkingMode.text,
+    embedding_model: 'text-embedding-ada-002',
+    embedding_model_provider: 'openai',
+    embedding_available: true,
+    app_count: 0,
+    document_count: 5,
+    total_document_count: 5,
+    word_count: 1000,
+    provider: 'external',
+    tags: [],
+    partial_member_list: [],
+    external_knowledge_info: {
+      external_knowledge_id: 'ext-knowledge-123',
+      external_knowledge_api_id: 'api-456',
+      external_knowledge_api_name: 'My External API',
+      external_knowledge_api_endpoint: 'https://api.external.example.com/v1',
+    },
+    external_retrieval_model: {
+      top_k: 5,
+      score_threshold: 0.8,
+      score_threshold_enabled: true,
+    },
+    retrieval_model_dict: mockRetrievalConfig,
+    retrieval_model: mockRetrievalConfig,
+    built_in_field_enabled: false,
+    keyword_number: 10,
+    created_by: 'user-1',
+    updated_by: 'user-1',
+    updated_at: Date.now(),
+    runtime_mode: 'general',
+    enable_api: true,
+    is_multimodal: false,
+  }
+
+  const defaultProps = {
+    currentDataset: mockDataset,
+    topK: 5,
+    scoreThreshold: 0.8,
+    scoreThresholdEnabled: true,
+    handleSettingsChange: vi.fn(),
+  }
+
+  beforeEach(() => {
+    vi.clearAllMocks()
+  })
+
+  describe('Rendering', () => {
+    it('should render without crashing', () => {
+      render(<ExternalKnowledgeSection {...defaultProps} />)
+      expect(screen.getByText(/form\.retrievalSetting\.title/i)).toBeInTheDocument()
+    })
+
+    it('should render retrieval settings section', () => {
+      render(<ExternalKnowledgeSection {...defaultProps} />)
+      expect(screen.getByText(/form\.retrievalSetting\.title/i)).toBeInTheDocument()
+    })
+
+    it('should render external knowledge API section', () => {
+      render(<ExternalKnowledgeSection {...defaultProps} />)
+      expect(screen.getByText(/form\.externalKnowledgeAPI/i)).toBeInTheDocument()
+    })
+
+    it('should render external knowledge ID section', () => {
+      render(<ExternalKnowledgeSection {...defaultProps} />)
+      expect(screen.getByText(/form\.externalKnowledgeID/i)).toBeInTheDocument()
+    })
+  })
+
+  describe('External Knowledge API Info', () => {
+    it('should display external API name', () => {
+      render(<ExternalKnowledgeSection {...defaultProps} />)
+      expect(screen.getByText('My External API')).toBeInTheDocument()
+    })
+
+    it('should display external API endpoint', () => {
+      render(<ExternalKnowledgeSection {...defaultProps} />)
+      expect(screen.getByText('https://api.external.example.com/v1')).toBeInTheDocument()
+    })
+
+    it('should render API connection icon', () => {
+      const { container } = render(<ExternalKnowledgeSection {...defaultProps} />)
+      // The ApiConnectionMod icon should be rendered
+      const icon = container.querySelector('svg')
+      expect(icon).toBeInTheDocument()
+    })
+
+    it('should display API name and endpoint in the same row', () => {
+      render(<ExternalKnowledgeSection {...defaultProps} />)
+
+      const apiName = screen.getByText('My External API')
+      const apiEndpoint = screen.getByText('https://api.external.example.com/v1')
+
+      // Both should be in the same container
+      expect(apiName.parentElement?.parentElement).toBe(apiEndpoint.parentElement?.parentElement)
+    })
+  })
+
+  describe('External Knowledge ID', () => {
+    it('should display external knowledge ID value', () => {
+      render(<ExternalKnowledgeSection {...defaultProps} />)
+      expect(screen.getByText('ext-knowledge-123')).toBeInTheDocument()
+    })
+
+    it('should render ID in a read-only display', () => {
+      render(<ExternalKnowledgeSection {...defaultProps} />)
+
+      const idElement = screen.getByText('ext-knowledge-123')
+      // The ID should be in a div with input-like styling, not an actual input
+      expect(idElement.tagName.toLowerCase()).toBe('div')
+    })
+  })
+
+  describe('Retrieval Settings', () => {
+    it('should pass topK to RetrievalSettings', () => {
+      render(<ExternalKnowledgeSection {...defaultProps} topK={10} />)
+
+      // RetrievalSettings should receive topK prop
+      // The exact rendering depends on RetrievalSettings component
+    })
+
+    it('should pass scoreThreshold to RetrievalSettings', () => {
+      render(<ExternalKnowledgeSection {...defaultProps} scoreThreshold={0.9} />)
+
+      // RetrievalSettings should receive scoreThreshold prop
+    })
+
+    it('should pass scoreThresholdEnabled to RetrievalSettings', () => {
+      render(<ExternalKnowledgeSection {...defaultProps} scoreThresholdEnabled={false} />)
+
+      // RetrievalSettings should receive scoreThresholdEnabled prop
+    })
+
+    it('should call handleSettingsChange when settings change', () => {
+      const handleSettingsChange = vi.fn()
+      render(<ExternalKnowledgeSection {...defaultProps} handleSettingsChange={handleSettingsChange} />)
+
+      // The handler should be properly passed to RetrievalSettings
+      // Actual interaction depends on RetrievalSettings implementation
+    })
+  })
+
+  describe('Dividers', () => {
+    it('should render dividers between sections', () => {
+      const { container } = render(<ExternalKnowledgeSection {...defaultProps} />)
+
+      const dividers = container.querySelectorAll('.bg-divider-subtle')
+      expect(dividers.length).toBeGreaterThanOrEqual(2)
+    })
+  })
+
+  describe('Props Updates', () => {
+    it('should update when currentDataset changes', () => {
+      const { rerender } = render(<ExternalKnowledgeSection {...defaultProps} />)
+
+      expect(screen.getByText('My External API')).toBeInTheDocument()
+
+      const updatedDataset = {
+        ...mockDataset,
+        external_knowledge_info: {
+          ...mockDataset.external_knowledge_info,
+          external_knowledge_api_name: 'Updated API Name',
+        },
+      }
+
+      rerender(<ExternalKnowledgeSection {...defaultProps} currentDataset={updatedDataset} />)
+
+      expect(screen.getByText('Updated API Name')).toBeInTheDocument()
+    })
+
+    it('should update when external knowledge ID changes', () => {
+      const { rerender } = render(<ExternalKnowledgeSection {...defaultProps} />)
+
+      expect(screen.getByText('ext-knowledge-123')).toBeInTheDocument()
+
+      const updatedDataset = {
+        ...mockDataset,
+        external_knowledge_info: {
+          ...mockDataset.external_knowledge_info,
+          external_knowledge_id: 'new-ext-id-789',
+        },
+      }
+
+      rerender(<ExternalKnowledgeSection {...defaultProps} currentDataset={updatedDataset} />)
+
+      expect(screen.getByText('new-ext-id-789')).toBeInTheDocument()
+    })
+
+    it('should update when API endpoint changes', () => {
+      const { rerender } = render(<ExternalKnowledgeSection {...defaultProps} />)
+
+      expect(screen.getByText('https://api.external.example.com/v1')).toBeInTheDocument()
+
+      const updatedDataset = {
+        ...mockDataset,
+        external_knowledge_info: {
+          ...mockDataset.external_knowledge_info,
+          external_knowledge_api_endpoint: 'https://new-api.example.com/v2',
+        },
+      }
+
+      rerender(<ExternalKnowledgeSection {...defaultProps} currentDataset={updatedDataset} />)
+
+      expect(screen.getByText('https://new-api.example.com/v2')).toBeInTheDocument()
+    })
+  })
+
+  describe('Layout', () => {
+    it('should have consistent row layout', () => {
+      const { container } = render(<ExternalKnowledgeSection {...defaultProps} />)
+
+      // Check for flex gap-x-1 class on rows
+      const rows = container.querySelectorAll('.flex.gap-x-1')
+      expect(rows.length).toBeGreaterThan(0)
+    })
+
+    it('should have consistent label width', () => {
+      const { container } = render(<ExternalKnowledgeSection {...defaultProps} />)
+
+      // Check for w-[180px] label containers
+      const labels = container.querySelectorAll('.w-\\[180px\\]')
+      expect(labels.length).toBeGreaterThan(0)
+    })
+  })
+
+  describe('Styling', () => {
+    it('should apply correct background to info displays', () => {
+      const { container } = render(<ExternalKnowledgeSection {...defaultProps} />)
+
+      // Info displays should have bg-components-input-bg-normal
+      const infoDisplays = container.querySelectorAll('.bg-components-input-bg-normal')
+      expect(infoDisplays.length).toBeGreaterThan(0)
+    })
+
+    it('should apply rounded corners to info displays', () => {
+      const { container } = render(<ExternalKnowledgeSection {...defaultProps} />)
+
+      const roundedElements = container.querySelectorAll('.rounded-lg')
+      expect(roundedElements.length).toBeGreaterThan(0)
+    })
+  })
+
+  describe('Different External Knowledge Info', () => {
+    it('should handle long API names', () => {
+      const longNameDataset = {
+        ...mockDataset,
+        external_knowledge_info: {
+          ...mockDataset.external_knowledge_info,
+          external_knowledge_api_name: 'This is a very long external knowledge API name that should be truncated',
+        },
+      }
+
+      render(<ExternalKnowledgeSection {...defaultProps} currentDataset={longNameDataset} />)
+
+      expect(screen.getByText(/This is a very long external knowledge API name/)).toBeInTheDocument()
+    })
+
+    it('should handle long API endpoints', () => {
+      const longEndpointDataset = {
+        ...mockDataset,
+        external_knowledge_info: {
+          ...mockDataset.external_knowledge_info,
+          external_knowledge_api_endpoint: 'https://api.very-long-domain-name.example.com/api/v1/external/knowledge',
+        },
+      }
+
+      render(<ExternalKnowledgeSection {...defaultProps} currentDataset={longEndpointDataset} />)
+
+      expect(screen.getByText(/https:\/\/api.very-long-domain-name.example.com/)).toBeInTheDocument()
+    })
+
+    it('should handle special characters in API name', () => {
+      const specialCharDataset = {
+        ...mockDataset,
+        external_knowledge_info: {
+          ...mockDataset.external_knowledge_info,
+          external_knowledge_api_name: 'API & Service <Test>',
+        },
+      }
+
+      render(<ExternalKnowledgeSection {...defaultProps} currentDataset={specialCharDataset} />)
+
+      expect(screen.getByText('API & Service <Test>')).toBeInTheDocument()
+    })
+  })
+
+  describe('RetrievalSettings Integration', () => {
+    it('should pass isInRetrievalSetting=true to RetrievalSettings', () => {
+      render(<ExternalKnowledgeSection {...defaultProps} />)
+
+      // The RetrievalSettings component should be rendered with isInRetrievalSetting=true
+      // This affects the component's layout/styling
+    })
+
+    it('should handle settings change for top_k', () => {
+      const handleSettingsChange = vi.fn()
+      render(<ExternalKnowledgeSection {...defaultProps} handleSettingsChange={handleSettingsChange} />)
+
+      // Find and interact with the top_k control in RetrievalSettings
+      // The exact interaction depends on RetrievalSettings implementation
+    })
+
+    it('should handle settings change for score_threshold', () => {
+      const handleSettingsChange = vi.fn()
+      render(<ExternalKnowledgeSection {...defaultProps} handleSettingsChange={handleSettingsChange} />)
+
+      // Find and interact with the score_threshold control in RetrievalSettings
+    })
+
+    it('should handle settings change for score_threshold_enabled', () => {
+      const handleSettingsChange = vi.fn()
+      render(<ExternalKnowledgeSection {...defaultProps} handleSettingsChange={handleSettingsChange} />)
+
+      // Find and interact with the score_threshold_enabled toggle in RetrievalSettings
+    })
+  })
+
+  describe('Accessibility', () => {
+    it('should have semantic structure', () => {
+      render(<ExternalKnowledgeSection {...defaultProps} />)
+
+      // Section labels should be present
+      expect(screen.getByText(/form\.retrievalSetting\.title/i)).toBeInTheDocument()
+      expect(screen.getByText(/form\.externalKnowledgeAPI/i)).toBeInTheDocument()
+      expect(screen.getByText(/form\.externalKnowledgeID/i)).toBeInTheDocument()
+    })
+  })
+})

web/app/components/datasets/settings/form/components/external-knowledge-section.tsx

@@ -0,0 +1,84 @@
+'use client'
+import type { DataSet } from '@/models/datasets'
+import { useTranslation } from 'react-i18next'
+import Divider from '@/app/components/base/divider'
+import { ApiConnectionMod } from '@/app/components/base/icons/src/vender/solid/development'
+import RetrievalSettings from '../../../external-knowledge-base/create/RetrievalSettings'
+
+const rowClass = 'flex gap-x-1'
+const labelClass = 'flex items-center shrink-0 w-[180px] h-7 pt-1'
+
+type ExternalKnowledgeSectionProps = {
+  currentDataset: DataSet
+  topK: number
+  scoreThreshold: number
+  scoreThresholdEnabled: boolean
+  handleSettingsChange: (data: { top_k?: number, score_threshold?: number, score_threshold_enabled?: boolean }) => void
+}
+
+const ExternalKnowledgeSection = ({
+  currentDataset,
+  topK,
+  scoreThreshold,
+  scoreThresholdEnabled,
+  handleSettingsChange,
+}: ExternalKnowledgeSectionProps) => {
+  const { t } = useTranslation()
+
+  return (
+    <>
+      <Divider type="horizontal" className="my-1 h-px bg-divider-subtle" />
+
+      {/* Retrieval Settings */}
+      <div className={rowClass}>
+        <div className={labelClass}>
+          <div className="system-sm-semibold text-text-secondary">{t('form.retrievalSetting.title', { ns: 'datasetSettings' })}</div>
+        </div>
+        <RetrievalSettings
+          topK={topK}
+          scoreThreshold={scoreThreshold}
+          scoreThresholdEnabled={scoreThresholdEnabled}
+          onChange={handleSettingsChange}
+          isInRetrievalSetting={true}
+        />
+      </div>
+
+      <Divider type="horizontal" className="my-1 h-px bg-divider-subtle" />
+
+      {/* External Knowledge API */}
+      <div className={rowClass}>
+        <div className={labelClass}>
+          <div className="system-sm-semibold text-text-secondary">{t('form.externalKnowledgeAPI', { ns: 'datasetSettings' })}</div>
+        </div>
+        <div className="w-full">
+          <div className="flex h-full items-center gap-1 rounded-lg bg-components-input-bg-normal px-3 py-2">
+            <ApiConnectionMod className="h-4 w-4 text-text-secondary" />
+            <div className="system-sm-medium overflow-hidden text-ellipsis text-text-secondary">
+              {currentDataset.external_knowledge_info.external_knowledge_api_name}
+            </div>
+            <div className="system-xs-regular text-text-tertiary">·</div>
+            <div className="system-xs-regular text-text-tertiary">
+              {currentDataset.external_knowledge_info.external_knowledge_api_endpoint}
+            </div>
+          </div>
+        </div>
+      </div>
+
+      {/* External Knowledge ID */}
+      <div className={rowClass}>
+        <div className={labelClass}>
+          <div className="system-sm-semibold text-text-secondary">{t('form.externalKnowledgeID', { ns: 'datasetSettings' })}</div>
+        </div>
+        <div className="w-full">
+          <div className="flex h-full items-center gap-1 rounded-lg bg-components-input-bg-normal px-3 py-2">
+            <div className="system-xs-regular text-text-tertiary">
+              {currentDataset.external_knowledge_info.external_knowledge_id}
+            </div>
+          </div>
+        </div>
+      </div>
+    </>
+  )
+}
+
+export default ExternalKnowledgeSection

web/app/components/datasets/settings/form/components/indexing-section.spec.tsx

@@ -0,0 +1,501 @@
+import type { DefaultModel, Model } from '@/app/components/header/account-setting/model-provider-page/declarations'
+import type { DataSet, SummaryIndexSetting } from '@/models/datasets'
+import type { RetrievalConfig } from '@/types/app'
+import { fireEvent, render, screen } from '@testing-library/react'
+import { ConfigurationMethodEnum, ModelStatusEnum, ModelTypeEnum } from '@/app/components/header/account-setting/model-provider-page/declarations'
+import { ChunkingMode, DatasetPermission, DataSourceType } from '@/models/datasets'
+import { RETRIEVE_METHOD } from '@/types/app'
+import { IndexingType } from '../../../create/step-two'
+import IndexingSection from './indexing-section'
+
+// Mock i18n doc link
+vi.mock('@/context/i18n', () => ({
+  useDocLink: () => (path: string) => `https://docs.dify.ai${path}`,
+}))
+
+// Mock app-context for child components
+vi.mock('@/context/app-context', () => ({
+  useSelector: (selector: (state: unknown) => unknown) => {
+    const state = {
+      isCurrentWorkspaceDatasetOperator: false,
+      userProfile: {
+        id: 'user-1',
+        name: 'Current User',
+        email: 'current@example.com',
+        avatar_url: '',
+        role: 'owner',
+      },
+    }
+    return selector(state)
+  },
+}))
+
+// Mock model-provider-page hooks
+vi.mock('@/app/components/header/account-setting/model-provider-page/hooks', () => ({
+  useModelList: () => ({ data: [], mutate: vi.fn(), isLoading: false }),
+  useCurrentProviderAndModel: () => ({ currentProvider: undefined, currentModel: undefined }),
+  useDefaultModel: () => ({ data: undefined, mutate: vi.fn(), isLoading: false }),
+  useModelListAndDefaultModel: () => ({ modelList: [], defaultModel: undefined }),
+  useModelListAndDefaultModelAndCurrentProviderAndModel: () => ({
+    modelList: [],
+    defaultModel: undefined,
+    currentProvider: undefined,
+    currentModel: undefined,
+  }),
+  useUpdateModelList: () => vi.fn(),
+  useUpdateModelProviders: () => vi.fn(),
+  useLanguage: () => 'en_US',
+  useSystemDefaultModelAndModelList: () => [undefined, vi.fn()],
+  useProviderCredentialsAndLoadBalancing: () => ({
+    credentials: undefined,
+    loadBalancing: undefined,
+    mutate: vi.fn(),
+    isLoading: false,
+  }),
+  useAnthropicBuyQuota: () => vi.fn(),
+  useMarketplaceAllPlugins: () => ({ plugins: [], isLoading: false }),
+  useRefreshModel: () => ({ handleRefreshModel: vi.fn() }),
+  useModelModalHandler: () => vi.fn(),
+}))
+
+// Mock provider-context
+vi.mock('@/context/provider-context', () => ({
+  useProviderContext: () => ({
+    textGenerationModelList: [],
+    embeddingsModelList: [],
+    rerankModelList: [],
+    agentThoughtModelList: [],
+    modelProviders: [],
+    textEmbeddingModelList: [],
+    speech2textModelList: [],
+    ttsModelList: [],
+    moderationModelList: [],
+    hasSettedApiKey: true,
+    plan: { type: 'free' },
+    enableBilling: false,
+    onPlanInfoChanged: vi.fn(),
+    isCurrentWorkspaceDatasetOperator: false,
+    supportRetrievalMethods: ['semantic_search', 'full_text_search', 'hybrid_search'],
+  }),
+}))
+
+describe('IndexingSection', () => {
+  const mockRetrievalConfig: RetrievalConfig = {
+    search_method: RETRIEVE_METHOD.semantic,
+    reranking_enable: false,
+    reranking_model: {
+      reranking_provider_name: '',
+      reranking_model_name: '',
+    },
+    top_k: 3,
+    score_threshold_enabled: false,
+    score_threshold: 0.5,
+  }
+
+  const mockDataset: DataSet = {
+    id: 'dataset-1',
+    name: 'Test Dataset',
+    description: 'Test description',
+    permission: DatasetPermission.onlyMe,
+    icon_info: {
+      icon_type: 'emoji',
+      icon: '📚',
+      icon_background: '#FFFFFF',
+      icon_url: '',
+    },
+    indexing_technique: IndexingType.QUALIFIED,
+    indexing_status: 'completed',
+    data_source_type: DataSourceType.FILE,
+    doc_form: ChunkingMode.text,
+    embedding_model: 'text-embedding-ada-002',
+    embedding_model_provider: 'openai',
+    embedding_available: true,
+    app_count: 0,
+    document_count: 5,
+    total_document_count: 5,
+    word_count: 1000,
+    provider: 'vendor',
+    tags: [],
+    partial_member_list: [],
+    external_knowledge_info: {
+      external_knowledge_id: 'ext-1',
+      external_knowledge_api_id: 'api-1',
+      external_knowledge_api_name: 'External API',
+      external_knowledge_api_endpoint: 'https://api.example.com',
+    },
+    external_retrieval_model: {
+      top_k: 3,
+      score_threshold: 0.7,
+      score_threshold_enabled: true,
+    },
+    retrieval_model_dict: mockRetrievalConfig,
+    retrieval_model: mockRetrievalConfig,
+    built_in_field_enabled: false,
+    keyword_number: 10,
+    created_by: 'user-1',
+    updated_by: 'user-1',
+    updated_at: Date.now(),
+    runtime_mode: 'general',
+    enable_api: true,
+    is_multimodal: false,
+  }
+
+  const mockEmbeddingModel: DefaultModel = {
+    provider: 'openai',
+    model: 'text-embedding-ada-002',
+  }
+
+  const mockEmbeddingModelList: Model[] = [
+    {
+      provider: 'openai',
+      label: { en_US: 'OpenAI', zh_Hans: 'OpenAI' },
+      icon_small: { en_US: '', zh_Hans: '' },
+      status: ModelStatusEnum.active,
+      models: [
+        {
+          model: 'text-embedding-ada-002',
+          label: { en_US: 'text-embedding-ada-002', zh_Hans: 'text-embedding-ada-002' },
+          model_type: ModelTypeEnum.textEmbedding,
+          features: [],
+          fetch_from: ConfigurationMethodEnum.predefinedModel,
+          model_properties: {},
+          deprecated: false,
+          status: ModelStatusEnum.active,
+          load_balancing_enabled: false,
+        },
+      ],
+    },
+  ]
+
+  const mockSummaryIndexSetting: SummaryIndexSetting = {
+    enable: false,
+  }
+
+  const defaultProps = {
+    currentDataset: mockDataset,
+    indexMethod: IndexingType.QUALIFIED,
+    setIndexMethod: vi.fn(),
+    keywordNumber: 10,
+    setKeywordNumber: vi.fn(),
+    embeddingModel: mockEmbeddingModel,
+    setEmbeddingModel: vi.fn(),
+    embeddingModelList: mockEmbeddingModelList,
+    retrievalConfig: mockRetrievalConfig,
+    setRetrievalConfig: vi.fn(),
+    summaryIndexSetting: mockSummaryIndexSetting,
+    handleSummaryIndexSettingChange: vi.fn(),
+    showMultiModalTip: false,
+  }
+
+  beforeEach(() => {
+    vi.clearAllMocks()
+  })
+
+  describe('Rendering', () => {
+    it('should render without crashing', () => {
+      render(<IndexingSection {...defaultProps} />)
+      expect(screen.getByText(/form\.chunkStructure\.title/i)).toBeInTheDocument()
+    })
+
+    it('should render chunk structure section when doc_form is set', () => {
+      render(<IndexingSection {...defaultProps} />)
+      expect(screen.getByText(/form\.chunkStructure\.title/i)).toBeInTheDocument()
+    })
+
+    it('should render index method section when conditions are met', () => {
+      render(<IndexingSection {...defaultProps} />)
+      // May match multiple elements (label and descriptions)
+      expect(screen.getAllByText(/form\.indexMethod/i).length).toBeGreaterThan(0)
+    })
+
+    it('should render embedding model section when indexMethod is high_quality', () => {
+      render(<IndexingSection {...defaultProps} indexMethod={IndexingType.QUALIFIED} />)
+      expect(screen.getByText(/form\.embeddingModel/i)).toBeInTheDocument()
+    })
+
+    it('should render retrieval settings section', () => {
+      render(<IndexingSection {...defaultProps} />)
+      expect(screen.getByText(/form\.retrievalSetting\.title/i)).toBeInTheDocument()
+    })
+  })
+
+  describe('Chunk Structure Section', () => {
+    it('should not render chunk structure when doc_form is not set', () => {
+      const datasetWithoutDocForm = { ...mockDataset, doc_form: undefined as unknown as ChunkingMode }
+      render(<IndexingSection {...defaultProps} currentDataset={datasetWithoutDocForm} />)
+
+      expect(screen.queryByText(/form\.chunkStructure\.title/i)).not.toBeInTheDocument()
+    })
+
+    it('should render learn more link for chunk structure', () => {
+      render(<IndexingSection {...defaultProps} />)
+
+      const learnMoreLink = screen.getByText(/form\.chunkStructure\.learnMore/i)
+      expect(learnMoreLink).toBeInTheDocument()
+      expect(learnMoreLink).toHaveAttribute('href', expect.stringContaining('chunking-and-cleaning-text'))
+    })
+
+    it('should render chunk structure description', () => {
+      render(<IndexingSection {...defaultProps} />)
+
+      expect(screen.getByText(/form\.chunkStructure\.description/i)).toBeInTheDocument()
+    })
+  })
+
+  describe('Index Method Section', () => {
+    it('should not render index method for parentChild chunking mode', () => {
+      const parentChildDataset = { ...mockDataset, doc_form: ChunkingMode.parentChild }
+      render(<IndexingSection {...defaultProps} currentDataset={parentChildDataset} />)
+
+      expect(screen.queryByText(/form\.indexMethod/i)).not.toBeInTheDocument()
+    })
+
+    it('should render high quality option', () => {
+      render(<IndexingSection {...defaultProps} />)
+
+      expect(screen.getByText(/stepTwo\.qualified/i)).toBeInTheDocument()
+    })
+
+    it('should render economy option', () => {
+      render(<IndexingSection {...defaultProps} />)
+
+      // May match multiple elements (title and tip)
+      expect(screen.getAllByText(/form\.indexMethodEconomy/i).length).toBeGreaterThan(0)
+    })
+
+    it('should call setIndexMethod when index method changes', () => {
+      const setIndexMethod = vi.fn()
+      const { container } = render(<IndexingSection {...defaultProps} setIndexMethod={setIndexMethod} />)
+
+      // Find the economy option card by looking for clickable elements containing the economy text
+      const economyOptions = screen.getAllByText(/form\.indexMethodEconomy/i)
+      if (economyOptions.length > 0) {
+        const economyCard = economyOptions[0].closest('[class*="cursor-pointer"]')
+        if (economyCard) {
+          fireEvent.click(economyCard)
+        }
+      }
+
+      // The handler should be properly passed - verify component renders without crashing
+      expect(container).toBeInTheDocument()
+    })
+
+    it('should show upgrade warning when switching from economy to high quality', () => {
+      const economyDataset = { ...mockDataset, indexing_technique: IndexingType.ECONOMICAL }
+      render(
+        <IndexingSection
+          {...defaultProps}
+          currentDataset={economyDataset}
+          indexMethod={IndexingType.QUALIFIED}
+        />,
+      )
+
+      expect(screen.getByText(/form\.upgradeHighQualityTip/i)).toBeInTheDocument()
+    })
+
+    it('should not show upgrade warning when already on high quality', () => {
+      render(
+        <IndexingSection
+          {...defaultProps}
+          indexMethod={IndexingType.QUALIFIED}
+        />,
+      )
+
+      expect(screen.queryByText(/form\.upgradeHighQualityTip/i)).not.toBeInTheDocument()
+    })
+
+    it('should disable index method when embedding is not available', () => {
+      const datasetWithoutEmbedding = { ...mockDataset, embedding_available: false }
+      render(<IndexingSection {...defaultProps} currentDataset={datasetWithoutEmbedding} />)
+
+      // Index method options should be disabled
+      // The exact implementation depends on the IndexMethod component
+    })
+  })
+
+  describe('Embedding Model Section', () => {
+    it('should render embedding model when indexMethod is high_quality', () => {
+      render(<IndexingSection {...defaultProps} indexMethod={IndexingType.QUALIFIED} />)
+
+      expect(screen.getByText(/form\.embeddingModel/i)).toBeInTheDocument()
+    })
+
+    it('should not render embedding model when indexMethod is economy', () => {
+      render(<IndexingSection {...defaultProps} indexMethod={IndexingType.ECONOMICAL} />)
+
+      expect(screen.queryByText(/form\.embeddingModel/i)).not.toBeInTheDocument()
+    })
+
+    it('should call setEmbeddingModel when model changes', () => {
+      const setEmbeddingModel = vi.fn()
+      render(
+        <IndexingSection
+          {...defaultProps}
+          setEmbeddingModel={setEmbeddingModel}
+          indexMethod={IndexingType.QUALIFIED}
+        />,
+      )
+
+      // The embedding model selector should be rendered
+      expect(screen.getByText(/form\.embeddingModel/i)).toBeInTheDocument()
+    })
+  })
+
+  describe('Summary Index Setting Section', () => {
+    it('should render summary index setting for high quality with text chunking', () => {
+      render(
+        <IndexingSection
+          {...defaultProps}
+          indexMethod={IndexingType.QUALIFIED}
+        />,
+      )
+
+      // Summary index setting should be rendered based on conditions
+      // The exact rendering depends on the SummaryIndexSetting component
+    })
+
+    it('should not render summary index setting for economy indexing', () => {
+      render(
+        <IndexingSection
+          {...defaultProps}
+          indexMethod={IndexingType.ECONOMICAL}
+        />,
+      )
+
+      // Summary index setting should not be rendered for economy
+    })
+
+    it('should call handleSummaryIndexSettingChange when setting changes', () => {
+      const handleSummaryIndexSettingChange = vi.fn()
+      render(
+        <IndexingSection
+          {...defaultProps}
+          handleSummaryIndexSettingChange={handleSummaryIndexSettingChange}
+          indexMethod={IndexingType.QUALIFIED}
+        />,
+      )
+
+      // The handler should be properly passed
+    })
+  })
+
+  describe('Retrieval Settings Section', () => {
+    it('should render retrieval settings', () => {
+      render(<IndexingSection {...defaultProps} />)
+
+      expect(screen.getByText(/form\.retrievalSetting\.title/i)).toBeInTheDocument()
+    })
+
+    it('should render learn more link for retrieval settings', () => {
+      render(<IndexingSection {...defaultProps} />)
+
+      const learnMoreLinks = screen.getAllByText(/learnMore/i)
+      const retrievalLearnMore = learnMoreLinks.find(link =>
+        link.closest('a')?.href?.includes('setting-indexing-methods'),
+      )
+      expect(retrievalLearnMore).toBeInTheDocument()
+    })
+
+    it('should render RetrievalMethodConfig for high quality indexing', () => {
+      render(<IndexingSection {...defaultProps} indexMethod={IndexingType.QUALIFIED} />)
+
+      // RetrievalMethodConfig should be rendered
+      expect(screen.getByText(/form\.retrievalSetting\.title/i)).toBeInTheDocument()
+    })
+
+    it('should render EconomicalRetrievalMethodConfig for economy indexing', () => {
+      render(<IndexingSection {...defaultProps} indexMethod={IndexingType.ECONOMICAL} />)
+
+      // EconomicalRetrievalMethodConfig should be rendered
+      expect(screen.getByText(/form\.retrievalSetting\.title/i)).toBeInTheDocument()
+    })
+
+    it('should call setRetrievalConfig when config changes', () => {
+      const setRetrievalConfig = vi.fn()
+      render(<IndexingSection {...defaultProps} setRetrievalConfig={setRetrievalConfig} />)
+
+      // The handler should be properly passed
+    })
+
+    it('should pass showMultiModalTip to RetrievalMethodConfig', () => {
+      render(<IndexingSection {...defaultProps} showMultiModalTip={true} />)
+
+      // The tip should be passed to the config component
+    })
+  })
+
+  describe('External Provider', () => {
+    it('should not render retrieval config for external provider', () => {
+      const externalDataset = { ...mockDataset, provider: 'external' }
+      render(<IndexingSection {...defaultProps} currentDataset={externalDataset} />)
+
+      // Retrieval config should not be rendered for external provider
+      // This is handled by the parent component, but we verify the condition
+    })
+  })
+
+  describe('Conditional Rendering', () => {
+    it('should show divider between sections', () => {
+      const { container } = render(<IndexingSection {...defaultProps} />)
+
+      // Dividers should be present
+      const dividers = container.querySelectorAll('.bg-divider-subtle')
+      expect(dividers.length).toBeGreaterThan(0)
+    })
+
+    it('should not render index method when indexing_technique is not set', () => {
+      const datasetWithoutTechnique = { ...mockDataset, indexing_technique: undefined as unknown as IndexingType }
+      render(<IndexingSection {...defaultProps} currentDataset={datasetWithoutTechnique} indexMethod={undefined} />)
+
+      expect(screen.queryByText(/form\.indexMethod/i)).not.toBeInTheDocument()
+    })
+  })
+
+  describe('Keyword Number', () => {
+    it('should pass keywordNumber to IndexMethod', () => {
+      render(<IndexingSection {...defaultProps} keywordNumber={15} />)
+
+      // The keyword number should be displayed in the economy option description
+      // The exact rendering depends on the IndexMethod component
+    })
+
+    it('should call setKeywordNumber when keyword number changes', () => {
+      const setKeywordNumber = vi.fn()
+      render(<IndexingSection {...defaultProps} setKeywordNumber={setKeywordNumber} />)
+
+      // The handler should be properly passed
+    })
+  })
+
+  describe('Props Updates', () => {
+    it('should update when indexMethod changes', () => {
+      const { rerender } = render(<IndexingSection {...defaultProps} indexMethod={IndexingType.QUALIFIED} />)
+
+      expect(screen.getByText(/form\.embeddingModel/i)).toBeInTheDocument()
+
+      rerender(<IndexingSection {...defaultProps} indexMethod={IndexingType.ECONOMICAL} />)
+
+      expect(screen.queryByText(/form\.embeddingModel/i)).not.toBeInTheDocument()
+    })
+
+    it('should update when currentDataset changes', () => {
+      const { rerender } = render(<IndexingSection {...defaultProps} />)
+
+      expect(screen.getByText(/form\.chunkStructure\.title/i)).toBeInTheDocument()
+
+      const datasetWithoutDocForm = { ...mockDataset, doc_form: undefined as unknown as ChunkingMode }
+      rerender(<IndexingSection {...defaultProps} currentDataset={datasetWithoutDocForm} />)
+
+      expect(screen.queryByText(/form\.chunkStructure\.title/i)).not.toBeInTheDocument()
+    })
+  })
+
+  describe('Undefined Dataset', () => {
+    it('should handle undefined currentDataset gracefully', () => {
+      render(<IndexingSection {...defaultProps} currentDataset={undefined} />)
+
+      // Should not crash and should handle undefined gracefully
+      // Most sections should not render without a dataset
+    })
+  })
+})

web/app/components/datasets/settings/form/components/indexing-section.tsx

@@ -0,0 +1,208 @@
+'use client'
+import type { DefaultModel, Model } from '@/app/components/header/account-setting/model-provider-page/declarations'
+import type { DataSet, SummaryIndexSetting as SummaryIndexSettingType } from '@/models/datasets'
+import type { RetrievalConfig } from '@/types/app'
+import { RiAlertFill } from '@remixicon/react'
+import { useTranslation } from 'react-i18next'
+import Divider from '@/app/components/base/divider'
+import EconomicalRetrievalMethodConfig from '@/app/components/datasets/common/economical-retrieval-method-config'
+import RetrievalMethodConfig from '@/app/components/datasets/common/retrieval-method-config'
+import ModelSelector from '@/app/components/header/account-setting/model-provider-page/model-selector'
+import { IS_CE_EDITION } from '@/config'
+import { useDocLink } from '@/context/i18n'
+import { ChunkingMode } from '@/models/datasets'
+import { IndexingType } from '../../../create/step-two'
+import ChunkStructure from '../../chunk-structure'
+import IndexMethod from '../../index-method'
+import SummaryIndexSetting from '../../summary-index-setting'
+
+const rowClass = 'flex gap-x-1'
+const labelClass = 'flex items-center shrink-0 w-[180px] h-7 pt-1'
+
+type IndexingSectionProps = {
+  currentDataset: DataSet | undefined
+  indexMethod: IndexingType | undefined
+  setIndexMethod: (value: IndexingType | undefined) => void
+  keywordNumber: number
+  setKeywordNumber: (value: number) => void
+  embeddingModel: DefaultModel
+  setEmbeddingModel: (value: DefaultModel) => void
+  embeddingModelList: Model[]
+  retrievalConfig: RetrievalConfig
+  setRetrievalConfig: (value: RetrievalConfig) => void
+  summaryIndexSetting: SummaryIndexSettingType | undefined
+  handleSummaryIndexSettingChange: (payload: SummaryIndexSettingType) => void
+  showMultiModalTip: boolean
+}
+
+const IndexingSection = ({
+  currentDataset,
+  indexMethod,
+  setIndexMethod,
+  keywordNumber,
+  setKeywordNumber,
+  embeddingModel,
+  setEmbeddingModel,
+  embeddingModelList,
+  retrievalConfig,
+  setRetrievalConfig,
+  summaryIndexSetting,
+  handleSummaryIndexSettingChange,
+  showMultiModalTip,
+}: IndexingSectionProps) => {
+  const { t } = useTranslation()
+  const docLink = useDocLink()
+
+  const isShowIndexMethod = currentDataset
+    && currentDataset.doc_form !== ChunkingMode.parentChild
+    && currentDataset.indexing_technique
+    && indexMethod
+
+  const showUpgradeWarning = currentDataset?.indexing_technique === IndexingType.ECONOMICAL
+    && indexMethod === IndexingType.QUALIFIED
+
+  const showSummaryIndexSetting = indexMethod === IndexingType.QUALIFIED
+    && [ChunkingMode.text, ChunkingMode.parentChild].includes(currentDataset?.doc_form as ChunkingMode)
+    && IS_CE_EDITION
+
+  return (
+    <>
+      {/* Chunk Structure */}
+      {!!currentDataset?.doc_form && (
+        <>
+          <Divider type="horizontal" className="my-1 h-px bg-divider-subtle" />
+          <div className={rowClass}>
+            <div className="flex w-[180px] shrink-0 flex-col">
+              <div className="system-sm-semibold flex h-8 items-center text-text-secondary">
+                {t('form.chunkStructure.title', { ns: 'datasetSettings' })}
+              </div>
+              <div className="body-xs-regular text-text-tertiary">
+                <a
+                  target="_blank"
+                  rel="noopener noreferrer"
+                  href={docLink('/use-dify/knowledge/create-knowledge/chunking-and-cleaning-text')}
+                  className="text-text-accent"
+                >
+                  {t('form.chunkStructure.learnMore', { ns: 'datasetSettings' })}
+                </a>
+                {t('form.chunkStructure.description', { ns: 'datasetSettings' })}
+              </div>
+            </div>
+            <div className="grow">
+              <ChunkStructure chunkStructure={currentDataset?.doc_form} />
+            </div>
+          </div>
+        </>
+      )}
+
+      {!!(isShowIndexMethod || indexMethod === 'high_quality') && (
+        <Divider type="horizontal" className="my-1 h-px bg-divider-subtle" />
+      )}
+
+      {/* Index Method */}
+      {!!isShowIndexMethod && (
+        <div className={rowClass}>
+          <div className={labelClass}>
+            <div className="system-sm-semibold text-text-secondary">{t('form.indexMethod', { ns: 'datasetSettings' })}</div>
+          </div>
+          <div className="grow">
+            <IndexMethod
+              value={indexMethod!}
+              disabled={!currentDataset?.embedding_available}
+              onChange={setIndexMethod}
+              currentValue={currentDataset.indexing_technique}
+              keywordNumber={keywordNumber}
+              onKeywordNumberChange={setKeywordNumber}
+            />
+            {showUpgradeWarning && (
+              <div className="relative mt-2 flex h-10 items-center gap-x-0.5 overflow-hidden rounded-xl border-[0.5px] border-components-panel-border bg-components-panel-bg-blur px-2 shadow-xs shadow-shadow-shadow-3">
+                <div className="absolute left-0 top-0 flex h-full w-full items-center bg-toast-warning-bg opacity-40" />
+                <div className="p-1">
+                  <RiAlertFill className="size-4 text-text-warning-secondary" />
+                </div>
+                <span className="system-xs-medium text-text-primary">
+                  {t('form.upgradeHighQualityTip', { ns: 'datasetSettings' })}
+                </span>
+              </div>
+            )}
+          </div>
+        </div>
+      )}
+
+      {/* Embedding Model */}
+      {indexMethod === IndexingType.QUALIFIED && (
+        <div className={rowClass}>
+          <div className={labelClass}>
+            <div className="system-sm-semibold text-text-secondary">
+              {t('form.embeddingModel', { ns: 'datasetSettings' })}
+            </div>
+          </div>
+          <div className="grow">
+            <ModelSelector
+              defaultModel={embeddingModel}
+              modelList={embeddingModelList}
+              onSelect={setEmbeddingModel}
+            />
+          </div>
+        </div>
+      )}
+
+      {/* Summary Index Setting */}
+      {showSummaryIndexSetting && (
+        <>
+          <Divider type="horizontal" className="my-1 h-px bg-divider-subtle" />
+          <SummaryIndexSetting
+            entry="dataset-settings"
+            summaryIndexSetting={summaryIndexSetting}
+            onSummaryIndexSettingChange={handleSummaryIndexSettingChange}
+          />
+        </>
+      )}
+
+      {/* Retrieval Method Config */}
+      {indexMethod && currentDataset?.provider !== 'external' && (
+        <>
+          <Divider type="horizontal" className="my-1 h-px bg-divider-subtle" />
+          <div className={rowClass}>
+            <div className={labelClass}>
+              <div className="flex w-[180px] shrink-0 flex-col">
+                <div className="system-sm-semibold flex h-7 items-center pt-1 text-text-secondary">
+                  {t('form.retrievalSetting.title', { ns: 'datasetSettings' })}
+                </div>
+                <div className="body-xs-regular text-text-tertiary">
+                  <a
+                    target="_blank"
+                    rel="noopener noreferrer"
+                    href={docLink('/use-dify/knowledge/create-knowledge/setting-indexing-methods')}
+                    className="text-text-accent"
+                  >
+                    {t('form.retrievalSetting.learnMore', { ns: 'datasetSettings' })}
+                  </a>
+                  {t('form.retrievalSetting.description', { ns: 'datasetSettings' })}
+                </div>
+              </div>
+            </div>
+            <div className="grow">
+              {indexMethod === IndexingType.QUALIFIED
+                ? (
+                    <RetrievalMethodConfig
+                      value={retrievalConfig}
+                      onChange={setRetrievalConfig}
+                      showMultiModalTip={showMultiModalTip}
+                    />
+                  )
+                : (
+                    <EconomicalRetrievalMethodConfig
+                      value={retrievalConfig}
+                      onChange={setRetrievalConfig}
+                    />
+                  )}
+            </div>
+          </div>
+        </>
+      )}
+    </>
+  )
+}
+
+export default IndexingSection