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

fix 'KeyError: "There is no item named 'word/NULL' in the archive"' (#9455)

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

Issue referring to:
https://github.com/python-openxml/python-docx/issues/797
Fix referring to:
https://github.com/python-openxml/python-docx/issues/1105#issuecomment-1298075246

### Type of change

- [x] Bug Fix (non-breaking change which fixes an issue)
This commit is contained in:
Jay Xu
2025-08-14 12:14:03 +08:00
committed by GitHub
parent 6e862553cb
commit 6d1078b538
1 changed files with 22 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

23
rag/app/naive.py
View File

@ -22,6 +22,8 @@ from timeit import default_timer as timer
from docx import Document from docx import Document
from docx.image.exceptions import InvalidImageStreamError, UnexpectedEndOfFileError, UnrecognizedImageError from docx.image.exceptions import InvalidImageStreamError, UnexpectedEndOfFileError, UnrecognizedImageError
from docx.opc.pkgreader import _SerializedRelationships, _SerializedRelationship
from docx.opc.oxml import parse_xml
from markdown import markdown from markdown import markdown
from PIL import Image from PIL import Image
from tika import parser from tika import parser
@ -47,8 +49,8 @@ class Docx(DocxParser):
if not embed: if not embed:
return None return None
embed = embed[0] embed = embed[0]
related_part = document.part.related_parts[embed]
try: try:
related_part = document.part.related_parts[embed]
image_blob = related_part.image.blob image_blob = related_part.image.blob
except UnrecognizedImageError: except UnrecognizedImageError:
logging.info("Unrecognized image format. Skipping image.") logging.info("Unrecognized image format. Skipping image.")
@ -62,6 +64,9 @@ class Docx(DocxParser):
except UnicodeDecodeError: except UnicodeDecodeError:
logging.info("The recognized image stream appears to be corrupted. Skipping image.") logging.info("The recognized image stream appears to be corrupted. Skipping image.")
return None return None
except Exception:
logging.info("The recognized image stream appears to be corrupted. Skipping image.")
return None
try: try:
image = Image.open(BytesIO(image_blob)).convert('RGB') image = Image.open(BytesIO(image_blob)).convert('RGB')
return image return image
@ -360,6 +365,20 @@ class Markdown(MarkdownParser):
tbls.append(((None, markdown(table, extensions=['markdown.extensions.tables'])), "")) tbls.append(((None, markdown(table, extensions=['markdown.extensions.tables'])), ""))
return sections, tbls return sections, tbls
def load_from_xml_v2(baseURI, rels_item_xml):
"""
Return |_SerializedRelationships| instance loaded with the
relationships contained in *rels_item_xml*. Returns an empty
collection if *rels_item_xml* is |None|.
"""
srels = _SerializedRelationships()
if rels_item_xml is not None:
rels_elm = parse_xml(rels_item_xml)
for rel_elm in rels_elm.Relationship_lst:
if rel_elm.target_ref in ('../NULL', 'NULL'):
continue
srels._srels.append(_SerializedRelationship(baseURI, rel_elm))
return srels
def chunk(filename, binary=None, from_page=0, to_page=100000, def chunk(filename, binary=None, from_page=0, to_page=100000,
lang="Chinese", callback=None, **kwargs): lang="Chinese", callback=None, **kwargs):
@ -391,6 +410,8 @@ def chunk(filename, binary=None, from_page=0, to_page=100000,
except Exception: except Exception:
vision_model = None vision_model = None
# fix "There is no item named 'word/NULL' in the archive", referring to https://github.com/python-openxml/python-docx/issues/1105#issuecomment-1298075246
_SerializedRelationships.load_from_xml = load_from_xml_v2
sections, tables = Docx()(filename, binary) sections, tables = Docx()(filename, binary)
if vision_model: if vision_model: