github/langchain
1
0
Fork 0
mirror of https://github.com/hwchase17/langchain.git synced 2025-06-24 07:35:18 +00:00
Code Issues Packages Projects Releases Wiki Activity

community[patch]: Fix empty producer in PDF Parsers (#30620)

Browse Source 
Fix an issue where if a pdf file doesn't have a “producer” in metadata, it generates an exception.
This commit is contained in:
Philippe PRADOS 2025-04-04 21:53:49 +02:00 committed by GitHub
parent f0159c7125
commit d8e3b7667f
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 12 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

15
libs/community/langchain_community/document_loaders/parsers/pdf.py
View File

@ -723,7 +723,8 @@ class PDFMinerParser(BaseBlobParser):
pages = PDFPage.get_pages(pdf_file_obj, password=self.password or "")
rsrcmgr = PDFResourceManager()
doc_metadata = _purge_metadata(
self._get_metadata(pdf_file_obj, password=self.password or "")
{"producer": "PDFMiner", "creator": "PDFMiner", "creationdate": ""}
| self._get_metadata(pdf_file_obj, password=self.password or "")
)
doc_metadata["source"] = blob.source
@ -996,7 +997,11 @@ class PyMuPDFParser(BaseBlobParser):
doc = pymupdf.open(stream=file_path, filetype="pdf")
if doc.is_encrypted:
doc.authenticate(self.password)
doc_metadata = self._extract_metadata(doc, blob)
doc_metadata = {
"producer": "PyMuPDF",
"creator": "PyMuPDF",
"creationdate": "",
} | self._extract_metadata(doc, blob)
full_content = []
for page in doc:
all_text = self._get_page_content(doc, page, text_kwargs).strip()
@ -1302,7 +1307,11 @@ class PyPDFium2Parser(BaseBlobParser):
)
full_content = []
doc_metadata = _purge_metadata(pdf_reader.get_metadata_dict())
doc_metadata = {
"producer": "PyPDFium2",
"creator": "PyPDFium2",
"creationdate": "",
} | _purge_metadata(pdf_reader.get_metadata_dict())
doc_metadata["source"] = blob.source
doc_metadata["total_pages"] = len(pdf_reader)