mirror of
https://github.com/infiniflow/ragflow.git
synced 2025-12-20 04:39:00 +08:00
Refa: refactor metadata filter (#11907)
### 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>
This commit is contained in:
@ -21,10 +21,10 @@ import re
|
||||
import xxhash
|
||||
from quart import request
|
||||
|
||||
from api.db.services.dialog_service import meta_filter
|
||||
from api.db.services.document_service import DocumentService
|
||||
from api.db.services.knowledgebase_service import KnowledgebaseService
|
||||
from api.db.services.llm_service import LLMBundle
|
||||
from common.metadata_utils import apply_meta_data_filter
|
||||
from api.db.services.search_service import SearchService
|
||||
from api.db.services.user_service import UserTenantService
|
||||
from api.utils.api_utils import get_data_error_result, get_json_result, server_error_response, validate_request, \
|
||||
@ -32,7 +32,7 @@ from api.utils.api_utils import get_data_error_result, get_json_result, server_e
|
||||
from rag.app.qa import beAdoc, rmPrefix
|
||||
from rag.app.tag import label_question
|
||||
from rag.nlp import rag_tokenizer, search
|
||||
from rag.prompts.generator import gen_meta_filter, cross_languages, keyword_extraction
|
||||
from rag.prompts.generator import cross_languages, keyword_extraction
|
||||
from common.string_utils import remove_redundant_spaces
|
||||
from common.constants import RetCode, LLMType, ParserType, PAGERANK_FLD
|
||||
from common import settings
|
||||
@ -317,54 +317,21 @@ async def retrieval_test():
|
||||
local_doc_ids = list(doc_ids) if doc_ids else []
|
||||
tenant_ids = []
|
||||
|
||||
meta_data_filter = {}
|
||||
chat_mdl = None
|
||||
if req.get("search_id", ""):
|
||||
search_config = SearchService.get_detail(req.get("search_id", "")).get("search_config", {})
|
||||
meta_data_filter = search_config.get("meta_data_filter", {})
|
||||
metas = DocumentService.get_meta_by_kbs(kb_ids)
|
||||
if meta_data_filter.get("method") == "auto":
|
||||
if meta_data_filter.get("method") in ["auto", "semi_auto"]:
|
||||
chat_mdl = LLMBundle(user_id, LLMType.CHAT, llm_name=search_config.get("chat_id", ""))
|
||||
filters: dict = await gen_meta_filter(chat_mdl, metas, question)
|
||||
local_doc_ids.extend(meta_filter(metas, filters["conditions"], filters.get("logic", "and")))
|
||||
if not local_doc_ids:
|
||||
local_doc_ids = None
|
||||
elif meta_data_filter.get("method") == "semi_auto":
|
||||
selected_keys = meta_data_filter.get("semi_auto", [])
|
||||
if selected_keys:
|
||||
filtered_metas = {key: metas[key] for key in selected_keys if key in metas}
|
||||
if filtered_metas:
|
||||
chat_mdl = LLMBundle(user_id, LLMType.CHAT, llm_name=search_config.get("chat_id", ""))
|
||||
filters: dict = await gen_meta_filter(chat_mdl, filtered_metas, question)
|
||||
local_doc_ids.extend(meta_filter(metas, filters["conditions"], filters.get("logic", "and")))
|
||||
if not local_doc_ids:
|
||||
local_doc_ids = None
|
||||
elif meta_data_filter.get("method") == "manual":
|
||||
local_doc_ids.extend(meta_filter(metas, meta_data_filter["manual"], meta_data_filter.get("logic", "and")))
|
||||
if meta_data_filter["manual"] and not local_doc_ids:
|
||||
local_doc_ids = ["-999"]
|
||||
else:
|
||||
meta_data_filter = req.get("meta_data_filter")
|
||||
if meta_data_filter:
|
||||
metas = DocumentService.get_meta_by_kbs(kb_ids)
|
||||
if meta_data_filter.get("method") == "auto":
|
||||
chat_mdl = LLMBundle(user_id, LLMType.CHAT)
|
||||
filters: dict = await gen_meta_filter(chat_mdl, metas, question)
|
||||
local_doc_ids.extend(meta_filter(metas, filters["conditions"], filters.get("logic", "and")))
|
||||
if not local_doc_ids:
|
||||
local_doc_ids = None
|
||||
elif meta_data_filter.get("method") == "semi_auto":
|
||||
selected_keys = meta_data_filter.get("semi_auto", [])
|
||||
if selected_keys:
|
||||
filtered_metas = {key: metas[key] for key in selected_keys if key in metas}
|
||||
if filtered_metas:
|
||||
chat_mdl = LLMBundle(user_id, LLMType.CHAT)
|
||||
filters: dict = await gen_meta_filter(chat_mdl, filtered_metas, question)
|
||||
local_doc_ids.extend(meta_filter(metas, filters["conditions"], filters.get("logic", "and")))
|
||||
if not local_doc_ids:
|
||||
local_doc_ids = None
|
||||
elif meta_data_filter.get("method") == "manual":
|
||||
local_doc_ids.extend(meta_filter(metas, meta_data_filter["manual"], meta_data_filter.get("logic", "and")))
|
||||
if meta_data_filter["manual"] and not local_doc_ids:
|
||||
local_doc_ids = ["-999"]
|
||||
meta_data_filter = req.get("meta_data_filter") or {}
|
||||
if meta_data_filter.get("method") in ["auto", "semi_auto"]:
|
||||
chat_mdl = LLMBundle(user_id, LLMType.CHAT)
|
||||
|
||||
if meta_data_filter:
|
||||
metas = DocumentService.get_meta_by_kbs(kb_ids)
|
||||
local_doc_ids = await apply_meta_data_filter(meta_data_filter, metas, question, chat_mdl, local_doc_ids)
|
||||
|
||||
tenants = UserTenantService.query(user_id=user_id)
|
||||
for kb_id in kb_ids:
|
||||
|
||||
Reference in New Issue
Block a user