### What problem does this PR solve?
test_doc_sdk_routes_unit had two flaky/incorrect branch assumptions:
1. parse/stop_parsing production logic gates on doc.run, but tests used
progress, causing branch mismatch and unintended fallthrough into
mutation/DB paths.
2. stop_parsing invalid-state test asserted an outdated message
fragment, making the contract brittle.
### Type of change
- [x] Bug Fix (non-breaking change which fixes an issue)
### What problem does this PR solve?
When the original code terminates the parsing task halfway, the progress
may not be 0 or 1, which will result in the inability to call the
interface to parse again
-Change the document parsing progress check to task status check, and
use TaskStatus.RUNNING.value to judge
-Update the condition judgment for stopping parsing documents, and check
whether the task is running instead
### Type of change
- [x] Bug Fix (non-breaking change which fixes an issue)
### What problem does this PR solve?
Fix stored XSS via HTML file upload and inline rendering in
/v1/file/get/<id>
### Type of change
- [x] Bug Fix (non-breaking change which fixes an issue)
### What problem does this PR solve?
When using a chat assistant that has a hardcoded `empty_response`, that
response was not returned correctly in streaming mode when no
information is found in the knowledge base. In this case only one
response with `"content": null` was yielded. If `"references": true`,
then the `empty_response` is still put into the `final_content` so there
is technically some content returned, but when `"references": false` no
content at all is returned.
I update the OpenAI chat completion endpoint to yield an additional
response with the `empty_response` in the content.
### Type of change
- [x] Bug Fix (non-breaking change which fixes an issue)
### What problem does this PR solve?
Decouple the memory API into a gateway layer (for routing/param parse)
and a service layer (for business logic).
### Type of change
- [x] Refactoring
### What problem does this PR solve?
Adjust highlight parsing, add row-count SQL override, tweak retrieval
thresholding, and update tests with engine-aware skips/utilities.
### Type of change
- [x] Bug Fix (non-breaking change which fixes an issue)
### What problem does this PR solve?
This PR adds an opt‑in way to include document‑level metadata in
OpenAI‑compatible reference chunks. Until now, metadata could be used
for filtering but wasn’t returned in responses. The change enables
clients to show richer citations (author/year/source, etc.) while
keeping payload size and privacy under control via an explicit request
flag and optional field allowlist.
### Type of change
- [ ] Bug Fix (non-breaking change which fixes an issue)
- [x] New Feature (non-breaking change which adds functionality)
- [x] Documentation Update
- [ ] Refactoring
- [ ] Performance Improvement
- [ ] Other (please describe):
Contribution during my time at RAGcon GmbH.
### What problem does this PR solve?
Put document metadata in ES/Infinity.
Index name of meta data: ragflow_doc_meta_{tenant_id}
### Type of change
- [x] Refactoring
## Description
This PR fixes issue #12805 by adding validation to handle
whitespace-only questions in the `/retrieval` endpoint.
## Problem
Sending a single space `" "` as the `question` parameter to `/retrieval`
crashes the request with an `AssertionError`. This happens because:
1. The endpoint doesn't trim or validate the question parameter
2. A whitespace-only string is treated as valid input
3. The retrieval logic only checks for empty strings (which are falsy),
but `" "` is truthy
4. Invalid match expressions are constructed, causing an assertion
failure in the Elasticsearch layer
## Solution
- Trim whitespace from the question parameter before processing
- Return an empty result for whitespace-only or empty questions
- Prevents the AssertionError and provides expected behavior
## Changes
- Added whitespace trimming and validation in `api/apps/sdk/doc.py`
- Returns empty result early if question is empty after trimming
## Testing
- Tested with single space input - now returns empty result instead of
crashing
- Tested with empty string - returns empty result
- Tested with normal questions - works as expected
Fixes#12805
Co-authored-by: Daniel <daniel@example.com>
### What problem does this PR solve?
If no `metadata_condition` parameter is given then don't load the
metadata of all documents into memory. Instead just pass `doc_ids` as
`None` to the `retrieval()` method, which means to use all documents of
the given datasets.
This is relevant if you have *a lot* of documents!
### Type of change
- [x] Performance Improvement
### What problem does this PR solve?
The OpenAI-compatible chat endpoint
(`/chats_openai/<chat_id>/chat/completions`) was not returning accurate
token
usage in streaming responses. The token counts were either missing or
inaccurate because the underlying LLM API
responses weren't being properly parsed for usage data.
This PR adds proper token counting using tiktoken (cl100k_base encoding)
as a fallback when the LLM API doesn't provide usage data in streaming
chunks. This ensures clients always receive token usage information in
the
response, which is essential for billing and quota management.
**Changes:**
- Add tiktoken-based token counting for streaming responses in
OpenAI-compatible endpoint
- Ensure `usage` field is always populated in the final streaming chunk
- Add unit tests for token usage calculation
Fixes#7850
### Type of change
- [x] Bug Fix (non-breaking change which fixes an issue)
### What problem does this PR solve?
Feat: Add a web search button to the chat box on the chat page.
### Type of change
- [x] New Feature (non-breaking change which adds functionality)
### What problem does this PR solve?
**Backend**
\rag\nlp\search.py
*Before the fix*
The top_k parameter was not applied to limit the total number of chunks,
and the rerank model also uses the exact whole valid_idx rather than
assigning valid_idx = valid_idx[:top] firstly.
*After the fix*
The top_k limit is applied to the total results before pagination, using
a default value of top = 1024 if top_k is not modified.
session.py
*Before the fix:*
When the frontend calls the retrieval API with `search_id`, the backend
only reads `meta_data_filter` from the saved `search_config`. The
`rerank_id`, `top_k`, `similarity_threshold`, and
`vector_similarity_weight` parameters are only taken from the direct
request body. Since the frontend doesn't pass these parameters
explicitly (it only passes `search_id`), they always fall back to
default values:
- `similarity_threshold` = 0.0
- `vector_similarity_weight` = 0.3
- `top_k` = 1024
- `rerank_id` = "" (no rerank)
This means user settings saved in the Search Settings page have no
effect on actual search results.
*After the fix:*
When a `search_id` is provided, the backend now reads all relevant
configuration from the saved `search_config`, including `rerank_id`,
`top_k`, `similarity_threshold`, and `vector_similarity_weight`. Request
parameters can still override these values if explicitly provided,
allowing flexibility. The rerank model is now properly instantiated
using the configured `rerank_id`, making the rerank feature actually
work.
**Frontend**
\web\src\pages\next-search\search-setting.tsx
*Before the fix*
search-setting.tsx file, the top_k input box is only displayed when
rerank is enabled (wrapped in the rerankModelDisabled condition). If the
rerank switch is turned off, the top_k input field will be hidden, but
the form value will remain unchanged. In other words: - When rerank is
enabled, users can modify top_k (default 1024). - When rerank is
disabled, top_k retains the previous value, but it's not visible on the
interface. Therefore, the backend will always receive the top_k
parameter; it's just that the frontend UI binds this configuration item
to the rerank switch. When rerank is turned off, top_k will not
automatically reset to 1024, but will retain its original value.
*After the fix*
On the contrary, if we switch off the button rerank model, the value
top-k will be reset to 1024. By the way, If we use top-k in an
individual method, rather than put it into the method retrieval, we can
control it separately
Now all methods valid
Using rerank
<img width="2378" height="1565" alt="Screenshot 2026-01-21 190206"
src="https://github.com/user-attachments/assets/fa2b0df0-1334-4ca3-b169-da6c5fd59935"
/>
Not using rerank
<img width="2596" height="1559" alt="Screenshot 2026-01-21 190229"
src="https://github.com/user-attachments/assets/c5a80522-a0e1-40e7-b349-42fe86df3138"
/>
Before fixing they are the same
### Type of change
- Bug Fix (non-breaking change which fixes an issue)
### What problem does this PR solve?
API adds audio to text and text to speech functions
### Type of change
- [x] New Feature (non-breaking change which adds functionality)
### What problem does this PR solve?
1) Create dataset using table parser for infinity
2) Answer questions in chat using SQL
### Type of change
- [x] New Feature (non-breaking change which adds functionality)
### What problem does this PR solve?
Fixes web API behavior mismatches that caused test failures by
normalizing error responses, tightening validations, correcting error
messages, and closing upload file handles.
### Type of change
- [x] Bug Fix (non-breaking change which fixes an issue)
### What problem does this PR solve?
Updates pre-existing HTTP API and SDK tests to align with current
backend behavior (validation errors, 404s, and schema defaults). This
ensures p3 regression coverage is accurate without changing production
code.
### Type of change
- [x] Other (please describe): align p3 HTTP/SDK tests with current
backend behavior
---------
Co-authored-by: Liu An <asiro@qq.com>
### What problem does this PR solve?
Move memory and message apis to /api, and add sdk support.
### Type of change
- [x] New Feature (non-breaking change which adds functionality)
### What problem does this PR solve?
Fix: add multimodel models in chat api #11986
### Type of change
- [x] Bug Fix (non-breaking change which fixes an issue)
---------
Co-authored-by: Kevin Hu <kevinhu.sh@gmail.com>
### What problem does this PR solve?
change:
Enhance delta streaming in chat functions for improved reasoning and
content handling
### Type of change
- [x] Refactoring
### What problem does this PR solve?
when a kb contains many documents, say 50000, and the retrieval is only
made against some kb without specifying any doc ids, the query for all
docs from the db is not necessary, and can be omitted to improve
performance.
### Type of change
- [x] Performance Improvement
### What problem does this PR solve?
change:
initialize webhook configuration in webhook function
remove debug print statement from airtable_connector
remove redundant uuid import in imap_connector
### Type of change
- [x] Bug Fix (non-breaking change which fixes an issue)
## Summary
Fixes#12311
Changes the `reference` field initialization from `[{}]` to `[]` in
session creation.
### Problem
When creating a session via the SDK API, the `reference` field was
incorrectly initialized as `[{}]`. This caused:
- First dialogue round: Empty reference
- Second dialogue round: Reference pointing to first round's data
- Overall misalignment between dialogue rounds and their references
### Solution
Changed the initialization to `[]` (empty list), which:
- Matches the `Conversation` model's expected default
- Ensures references grow correctly one-to-one with assistant responses
- Aligns with the service layer's expectations
### Testing
After applying this fix:
1. Create a session via `POST /api/v1/chats/{conversation_id}/sessions`
2. Send multiple questions via `POST
/api/v1/chats/{conversation_id}/completions`
3. View the conversation on web - references should now align correctly
with each dialogue round
### What problem does this PR solve?
Fixes a bug in the `rm_chunk` SDK interface where an `UnboundLocalError`
could
occur if `chunk_ids` is not provided in the request.
- `unique_chunk_ids` and `duplicate_messages` are now always initialized
in the `else` branch when `chunk_ids` is missing.
- API behavior remains unchanged when `chunk_ids` is present.
### Type of change
- [x] Bug Fix (non-breaking change which fixes an issue)
### What problem does this PR solve?
Manage message and use in agent.
Issue #4213
### Type of change
- [x] New Feature (non-breaking change which adds functionality)
### What problem does this PR solve?
change：
enhance webhook response to include status and success fields and
simplify ReAct agent
### Type of change
- [x] New Feature (non-breaking change which adds functionality)
### What problem does this PR solve?
Chats completions API supports metadata filtering.
### Type of change
- [x] New Feature (non-breaking change which adds functionality)
### What problem does this PR solve?
issue:
https://github.com/infiniflow/ragflow/issues/10427https://github.com/infiniflow/ragflow/issues/8115
change:
- Support for Multiple HTTP Methods (POST / GET / PUT / PATCH / DELETE /
HEAD)
- Security Validation
1. max_body_size
2. IP whitelist
3. rate limit
4. token / basic / jwt authentication
- File Upload Support
- Unified Content-Type Handling
- Full Schema-Based Extraction & Type Validation
- Two Execution Modes: Immediately / Streaming
### Type of change
- [x] New Feature (non-breaking change which adds functionality)
### What problem does this PR solve?
Trace information can be returned by the agent completion API.
### Type of change
- [x] New Feature (non-breaking change which adds functionality)
…tant, but model is available via UI
Fix: add multimodel models in chat api
Fixes#8549
### What problem does this PR solve?
Add a parameter model_type in chat api.
### Type of change
- [x] Bug Fix (non-breaking change which fixes an issue)
---------
Co-authored-by: writinwaters <93570324+writinwaters@users.noreply.github.com>
### What problem does this PR solve?
Refactor metadata filter.
### Type of change
- [x] Bug Fix (non-breaking change which fixes an issue)
- [x] Refactoring
---------
Co-authored-by: Kevin Hu <kevinhu.sh@gmail.com>
### What problem does this PR solve?
Add DeepseekV3.2 of Tongyi-Qianwen and remove unused code
### Type of change
- [x] New Feature (non-breaking change which adds functionality)
