pengjunjie/ragflow
1
0
Fork 1
mirror of https://github.com/infiniflow/ragflow.git synced 2025-12-08 20:42:30 +08:00
Code Issues Packages Projects Releases Wiki Activity

Feat: add extractor component. (#10271)

Browse Source 
### What problem does this PR solve?


### Type of change

- [x] New Feature (non-breaking change which adds functionality)
This commit is contained in:
Kevin Hu
2025-09-25 11:34:47 +08:00
committed by GitHub
parent 840b2b5809
commit 1b19d302c5
16 changed files with 379 additions and 127 deletions
Download Patch File Download Diff File Expand all files Collapse all files

18
api/db/services/document_service.py
View File

@ -121,12 +121,20 @@ class DocumentService(CommonService):
orderby, desc, keywords, run_status, types, suffix):
fields = cls.get_cls_model_fields()
if keywords:
docs = cls.model.select(*fields).join(File2Document, on=(File2Document.document_id == cls.model.id)).join(File, on=(File.id == File2Document.file_id)).where(
(cls.model.kb_id == kb_id),
(fn.LOWER(cls.model.name).contains(keywords.lower()))
)
docs = cls.model.select(*[*fields, UserCanvas.title])\
.join(File2Document, on=(File2Document.document_id == cls.model.id))\
.join(File, on=(File.id == File2Document.file_id))\
.join(UserCanvas, on=(cls.model.pipeline_id == UserCanvas.id), join_type=JOIN.LEFT_OUTER)\
.where(
(cls.model.kb_id == kb_id),
(fn.LOWER(cls.model.name).contains(keywords.lower()))
)
else:
docs = cls.model.select(*fields).join(File2Document, on=(File2Document.document_id == cls.model.id)).join(File, on=(File.id == File2Document.file_id)).where(cls.model.kb_id == kb_id)
docs = cls.model.select(*[*fields, UserCanvas.title])\
.join(File2Document, on=(File2Document.document_id == cls.model.id))\
.join(UserCanvas, on=(cls.model.pipeline_id == UserCanvas.id), join_type=JOIN.LEFT_OUTER)\
.join(File, on=(File.id == File2Document.file_id))\
.where(cls.model.kb_id == kb_id)
if run_status:
docs = docs.where(cls.model.run.in_(run_status))