Feat: add meta data filter. (#9405)

### What problem does this PR solve?

#8531 
#7417 
#6761 
#6573
#6477

### Type of change

- [x] New Feature (non-breaking change which adds functionality)

api/apps/dialog_app.py

@@ -51,6 +51,7 @@ def set_dialog():
     similarity_threshold = req.get("similarity_threshold", 0.1)
     vector_similarity_weight = req.get("vector_similarity_weight", 0.3)
     llm_setting = req.get("llm_setting", {})
+    meta_data_filter = req.get("meta_data_filter", {})
     prompt_config = req["prompt_config"]
 
     if not is_create:
@@ -85,6 +86,7 @@ def set_dialog():
                 "llm_id": llm_id,
                 "llm_setting": llm_setting,
                 "prompt_config": prompt_config,
+                "meta_data_filter": meta_data_filter,
                 "top_n": top_n,
                 "top_k": top_k,
                 "rerank_id": rerank_id,

api/apps/document_app.py

@@ -681,6 +681,11 @@ def set_meta():
         return get_json_result(data=False, message="No authorization.", code=settings.RetCode.AUTHENTICATION_ERROR)
     try:
         meta = json.loads(req["meta"])
+        if not isinstance(meta, dict):
+            return get_json_result(data=False, message="Only dictionary type supported.", code=settings.RetCode.ARGUMENT_ERROR)
+        for k,v in meta.items():
+            if not isinstance(v, str) and not isinstance(v, int) and not isinstance(v, float):
+                return get_json_result(data=False, message=f"The type is not supported: {v}", code=settings.RetCode.ARGUMENT_ERROR)
     except Exception as e:
         return get_json_result(data=False, message=f"Json syntax error: {e}", code=settings.RetCode.ARGUMENT_ERROR)
     if not isinstance(meta, dict):

api/apps/kb_app.py

@@ -351,6 +351,7 @@ def knowledge_graph(kb_id):
             obj["graph"]["edges"] = sorted(filtered_edges, key=lambda x: x.get("weight", 0), reverse=True)[:128]
     return get_json_result(data=obj)
 
+
 @manager.route('/<kb_id>/knowledge_graph', methods=['DELETE'])  # noqa: F821
 @login_required
 def delete_knowledge_graph(kb_id):
@@ -364,3 +365,17 @@ def delete_knowledge_graph(kb_id):
     settings.docStoreConn.delete({"knowledge_graph_kwd": ["graph", "subgraph", "entity", "relation"]}, search.index_name(kb.tenant_id), kb_id)
 
     return get_json_result(data=True)
+
+
+@manager.route("/get_meta", methods=["GET"])  # noqa: F821
+@login_required
+def get_meta():
+    kb_ids = request.args.get("kb_ids", "").split(",")
+    for kb_id in kb_ids:
+        if not KnowledgebaseService.accessible(kb_id, current_user.id):
+            return get_json_result(
+                data=False,
+                message='No authorization.',
+                code=settings.RetCode.AUTHENTICATION_ERROR
+            )
+    return get_json_result(data=DocumentService.get_meta_by_kbs(kb_ids))

api/db/db_models.py

@@ -744,6 +744,7 @@ class Dialog(DataBaseModel):
         null=False,
         default={"system": "", "prologue": "Hi! I'm your assistant, what can I do for you?", "parameters": [], "empty_response": "Sorry! No relevant content was found in the knowledge base!"},
     )
+    meta_data_filter = JSONField(null=True, default={})
 
     similarity_threshold = FloatField(default=0.2)
     vector_similarity_weight = FloatField(default=0.3)
@@ -1015,4 +1016,8 @@ def migrate_db():
         migrate(migrator.add_column("api_4_conversation", "errors", TextField(null=True, help_text="errors")))
     except Exception:
         pass
+    try:
+        migrate(migrator.add_column("dialog", "meta_data_filter", JSONField(null=True, default={})))
+    except Exception:
+        pass
     logging.disable(logging.NOTSET)

api/db/services/dialog_service.py

@@ -30,6 +30,7 @@ from api import settings
 from api.db import LLMType, ParserType, StatusEnum
 from api.db.db_models import DB, Dialog
 from api.db.services.common_service import CommonService
+from api.db.services.document_service import DocumentService
 from api.db.services.knowledgebase_service import KnowledgebaseService
 from api.db.services.langfuse_service import TenantLangfuseService
 from api.db.services.llm_service import LLMBundle, TenantLLMService
@@ -38,6 +39,7 @@ from rag.app.resume import forbidden_select_fields4resume
 from rag.app.tag import label_question
 from rag.nlp.search import index_name
 from rag.prompts import chunks_format, citation_prompt, cross_languages, full_question, kb_prompt, keyword_extraction, message_fit_in
+from rag.prompts.prompts import gen_meta_filter
 from rag.utils import num_tokens_from_string, rmSpace
 from rag.utils.tavily_conn import Tavily
 
@@ -250,6 +252,46 @@ def repair_bad_citation_formats(answer: str, kbinfos: dict, idx: set):
     return answer, idx
 
 
+def meta_filter(metas: dict, filters: list[dict]):
+    doc_ids = []
+    def filter_out(v2docs, operator, value):
+        nonlocal doc_ids
+        for input,docids in v2docs.items():
+            try:
+                input = float(input)
+                value = float(value)
+            except Exception:
+                input = str(input)
+                value = str(value)
+
+            for conds in [
+                    (operator == "contains", str(value).lower() in str(input).lower()),
+                    (operator == "not contains", str(value).lower() not in str(input).lower()),
+                    (operator == "start with", str(input).lower().startswith(str(value).lower())),
+                    (operator == "end with", str(input).lower().endswith(str(value).lower())),
+                    (operator == "empty", not input),
+                    (operator == "not empty", input),
+                    (operator == "=", input == value),
+                    (operator == "≠", input != value),
+                    (operator == ">", input > value),
+                    (operator == "<", input < value),
+                    (operator == "≥", input >= value),
+                    (operator == "≤", input <= value),
+                ]:
+                try:
+                    if all(conds):
+                        doc_ids.extend(docids)
+                except Exception:
+                    pass
+
+    for k, v2docs in metas.items():
+        for f in filters:
+            if k != f["key"]:
+                continue
+            filter_out(v2docs, f["op"], f["value"])
+    return doc_ids
+
+
 def chat(dialog, messages, stream=True, **kwargs):
     assert messages[-1]["role"] == "user", "The last content of this conversation is not from user."
     if not dialog.kb_ids and not dialog.prompt_config.get("tavily_api_key"):
@@ -287,9 +329,10 @@ def chat(dialog, messages, stream=True, **kwargs):
 
     retriever = settings.retrievaler
     questions = [m["content"] for m in messages if m["role"] == "user"][-3:]
-    attachments = kwargs["doc_ids"].split(",") if "doc_ids" in kwargs else None
+    attachments = kwargs["doc_ids"].split(",") if "doc_ids" in kwargs else []
     if "doc_ids" in messages[-1]:
         attachments = messages[-1]["doc_ids"]
+
     prompt_config = dialog.prompt_config
     field_map = KnowledgebaseService.get_field_map(dialog.kb_ids)
     # try to use sql if field mapping is good to go
@@ -316,6 +359,14 @@ def chat(dialog, messages, stream=True, **kwargs):
     if prompt_config.get("cross_languages"):
         questions = [cross_languages(dialog.tenant_id, dialog.llm_id, questions[0], prompt_config["cross_languages"])]
 
+    if dialog.meta_data_filter:
+        metas = DocumentService.get_meta_by_kbs(dialog.kb_ids)
+        if dialog.meta_data_filter.get("method") == "auto":
+            filters = gen_meta_filter(chat_mdl, metas, questions[-1])
+            attachments.extend(meta_filter(metas, filters))
+        elif dialog.meta_data_filter.get("method") == "manual":
+            attachments.extend(meta_filter(metas, dialog.meta_data_filter["manual"]))
+
     if prompt_config.get("keyword", False):
         questions[-1] += keyword_extraction(chat_mdl, questions[-1])
 

api/db/services/document_service.py

@@ -574,6 +574,25 @@ class DocumentService(CommonService):
     def update_meta_fields(cls, doc_id, meta_fields):
         return cls.update_by_id(doc_id, {"meta_fields": meta_fields})
 
+    @classmethod
+    @DB.connection_context()
+    def get_meta_by_kbs(cls, kb_ids):
+        fields = [
+            cls.model.id,
+            cls.model.meta_fields,
+        ]
+        meta = {}
+        for r in cls.model.select(*fields).where(cls.model.kb_id.in_(kb_ids)):
+            doc_id = r.id
+            for k,v in r.meta_fields.items():
+                if k not in meta:
+                    meta[k] = {}
+                v = str(v)
+                if v not in meta[k]:
+                    meta[k][v] = []
+                meta[k][v].append(doc_id)
+        return meta
+
     @classmethod
     @DB.connection_context()
     def update_progress(cls):