pengjunjie/ragflow
1
0
Fork 1
mirror of https://github.com/infiniflow/ragflow.git synced 2026-02-03 00:55:10 +08:00
Code Issues Packages Projects Releases Wiki Activity

shortcut metadata_condition if there is none (#12835)

Browse Source 
### 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
This commit is contained in:
Mathias Panzenböck
2026-01-27 05:45:58 +01:00
committed by GitHub
parent c8338dec57
commit b36d9744ae
1 changed files with 12 additions and 8 deletions
Download Patch File Download Diff File Expand all files Collapse all files

20
api/apps/sdk/doc.py
View File

@ -1526,14 +1526,18 @@ async def retrieval_test(tenant_id):
if doc_id not in doc_ids_list: if doc_id not in doc_ids_list:
return get_error_data_result(f"The datasets don't own the document {doc_id}") return get_error_data_result(f"The datasets don't own the document {doc_id}")
if not doc_ids: if not doc_ids:
metadata_condition = req.get("metadata_condition", {}) or {} metadata_condition = req.get("metadata_condition")
metas = DocumentService.get_meta_by_kbs(kb_ids) if metadata_condition:
doc_ids = meta_filter(metas, convert_conditions(metadata_condition), metadata_condition.get("logic", "and")) metas = DocumentService.get_meta_by_kbs(kb_ids)
# If metadata_condition has conditions but no docs match, return empty result doc_ids = meta_filter(metas, convert_conditions(metadata_condition), metadata_condition.get("logic", "and"))
if not doc_ids and metadata_condition.get("conditions"): # If metadata_condition has conditions but no docs match, return empty result
return get_result(data={"total": 0, "chunks": [], "doc_aggs": {}}) if not doc_ids and metadata_condition.get("conditions"):
if metadata_condition and not doc_ids: return get_result(data={"total": 0, "chunks": [], "doc_aggs": {}})
doc_ids = ["-999"] if metadata_condition and not doc_ids:
doc_ids = ["-999"]
else:
# If doc_ids is None all documents of the datasets are used
doc_ids = None
similarity_threshold = float(req.get("similarity_threshold", 0.2)) similarity_threshold = float(req.get("similarity_threshold", 0.2))
vector_similarity_weight = float(req.get("vector_similarity_weight", 0.3)) vector_similarity_weight = float(req.get("vector_similarity_weight", 0.3))
top = int(req.get("top_k", 1024)) top = int(req.get("top_k", 1024))