pengjunjie/ragflow
1
0
Fork 1
mirror of https://github.com/infiniflow/ragflow.git synced 2025-12-25 08:06:48 +08:00
Code Issues Packages Projects Releases Wiki Activity

Fix: table tag on chunks. (#12126)

Browse Source 
### Type of change

- [x] Bug Fix (non-breaking change which fixes an issue)
This commit is contained in:
Kevin Hu
2025-12-24 09:32:19 +08:00
committed by GitHub
parent 17b8bb62b6
commit c33134ea2c
5 changed files with 26 additions and 17 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
rag/nlp/__init__.py
View File

@ -348,7 +348,8 @@ def tokenize_table(tbls, doc, eng, batch_size=10):
d["doc_type_kwd"] = "table"
if img:
d["image"] = img
d["doc_type_kwd"] = "image"
if d["content_with_weight"].find("<tr>") < 0:
d["doc_type_kwd"] = "image"
if poss:
add_positions(d, poss)
res.append(d)
@ -361,7 +362,8 @@ def tokenize_table(tbls, doc, eng, batch_size=10):
d["doc_type_kwd"] = "table"
if img:
d["image"] = img
d["doc_type_kwd"] = "image"
if d["content_with_weight"].find("<tr>") < 0:
d["doc_type_kwd"] = "image"
add_positions(d, poss)
res.append(d)
return res

4
rag/svr/task_executor.py
View File

@ -395,8 +395,8 @@ async def build_chunks(task, progress_callback):
await asyncio.gather(*tasks, return_exceptions=True)
raise
metadata = {}
for ck in cks:
metadata = update_metadata_to(metadata, ck["metadata_obj"])
for doc in docs:
metadata = update_metadata_to(metadata, doc["metadata_obj"])
del ck["metadata_obj"]
if metadata:
e, doc = DocumentService.get_by_id(task["doc_id"])