Fix: Infinity keyword round-trip, highlight fallback, and KB update guards (#12660)

### What problem does this PR solve?

Fixes Infinity-specific API regressions: preserves ```important_kwd```
round‑trip for ```[""]```, restores required highlight key in retrieval
responses, and enforces Infinity guards for unsupported
```parser_id=tag``` and pagerank in ```/v1/kb/update```. Also removes a
slow/buggy pandas row-wise apply that was throwing ```ValueError``` and
causing flakiness.

### Type of change

- [x] Bug Fix (non-breaking change which fixes an issue)

api/apps/kb_app.py

@@ -82,6 +82,20 @@ async def update():
         return get_data_error_result(
             message=f"Dataset name length is {len(req['name'])} which is large than {DATASET_NAME_LIMIT}")
     req["name"] = req["name"].strip()
+    if settings.DOC_ENGINE_INFINITY:
+        parser_id = req.get("parser_id")
+        if isinstance(parser_id, str) and parser_id.lower() == "tag":
+            return get_json_result(
+                code=RetCode.OPERATING_ERROR,
+                message="The chunking method Tag has not been supported by Infinity yet.",
+                data=False,
+            )
+        if "pagerank" in req:
+            return get_json_result(
+                code=RetCode.DATA_ERROR,
+                message="'pagerank' can only be set when doc_engine is elasticsearch",
+                data=False,
+            )
 
     if not KnowledgebaseService.accessible4deletion(req["kb_id"], current_user.id):
         return get_json_result(

common/doc_store/infinity_conn_base.py

@@ -367,6 +367,9 @@ class InfinityConnectionBase(DocStoreConnection):
         num_rows = len(res)
         column_id = res["id"]
         if field_name not in res:
+            if field_name == "content_with_weight" and "content" in res:
+                field_name = "content"
+            else:
                 return {}
         for i in range(num_rows):
             id = column_id[i]

conf/infinity_mapping.json

@@ -9,6 +9,7 @@
 	"docnm": {"type": "varchar", "default": "", "analyzer": ["rag-coarse", "rag-fine"], "comment": "docnm_kwd, title_tks, title_sm_tks"},
 	"name_kwd": {"type": "varchar", "default": "", "analyzer": "whitespace-#"},
 	"tag_kwd": {"type": "varchar", "default": "", "analyzer": "whitespace-#"},
+	"important_kwd_empty_count": {"type": "integer", "default": 0},
 	"important_keywords": {"type": "varchar", "default": "", "analyzer": ["rag-coarse", "rag-fine"], "comment": "important_kwd, important_tks"},
 	"questions": {"type": "varchar", "default": "", "analyzer": ["rag-coarse", "rag-fine"], "comment": "question_kwd, question_tks"},
 	"content": {"type": "varchar", "default": "", "analyzer": ["rag-coarse", "rag-fine"], "comment": "content_with_weight, content_ltks, content_sm_ltks"},

rag/utils/infinity_conn.py

@@ -42,6 +42,7 @@ class InfinityConnection(InfinityConnectionBase):
         return False
 
     def convert_select_fields(self, output_fields: list[str]) -> list[str]:
+        need_empty_count = "important_kwd" in output_fields
         for i, field in enumerate(output_fields):
             if field in ["docnm_kwd", "title_tks", "title_sm_tks"]:
                 output_fields[i] = "docnm"
@@ -53,6 +54,8 @@ class InfinityConnection(InfinityConnectionBase):
                 output_fields[i] = "content"
             elif field in ["authors_tks", "authors_sm_tks"]:
                 output_fields[i] = "authors"
+        if need_empty_count and "important_kwd_empty_count" not in output_fields:
+            output_fields.append("important_kwd_empty_count")
         return list(set(output_fields))
 
     @staticmethod
@@ -340,6 +343,12 @@ class InfinityConnection(InfinityConnectionBase):
                     if not d.get("docnm_kwd"):
                         d["docnm"] = self.list2str(v)
                 elif k == "important_kwd":
+                    if isinstance(v, list):
+                        empty_count = sum(1 for kw in v if kw == "")
+                        tokens = [kw for kw in v if kw != ""]
+                        d["important_keywords"] = self.list2str(tokens, ",")
+                        d["important_kwd_empty_count"] = empty_count
+                    else:
                         d["important_keywords"] = self.list2str(v, ",")
                 elif k == "important_tks":
                     if not d.get("important_kwd"):
@@ -429,6 +438,12 @@ class InfinityConnection(InfinityConnectionBase):
                 if not new_value.get("docnm_kwd"):
                     new_value["docnm"] = v
             elif k == "important_kwd":
+                if isinstance(v, list):
+                    empty_count = sum(1 for kw in v if kw == "")
+                    tokens = [kw for kw in v if kw != ""]
+                    new_value["important_keywords"] = self.list2str(tokens, ",")
+                    new_value["important_kwd_empty_count"] = empty_count
+                else:
                     new_value["important_keywords"] = self.list2str(v, ",")
             elif k == "important_tks":
                 if not new_value.get("important_kwd"):
@@ -532,6 +547,14 @@ class InfinityConnection(InfinityConnectionBase):
                     res[field] = res["docnm"]
         if "important_keywords" in res.columns:
             if "important_kwd" in fields_all:
+                if "important_kwd_empty_count" in res.columns:
+                    base = res["important_keywords"].apply(lambda raw: raw.split(",") if raw else [])
+                    counts = res["important_kwd_empty_count"].fillna(0).astype(int)
+                    res["important_kwd"] = [
+                        tokens + [""] * empty_count
+                        for tokens, empty_count in zip(base.tolist(), counts.tolist())
+                    ]
+                else:
                     res["important_kwd"] = res["important_keywords"].apply(lambda v: v.split(",") if v else [])
             if "important_tks" in fields_all:
                 res["important_tks"] = res["important_keywords"]