AI/ragflow
1
0
Fork 0
mirror of https://git.mirrors.martin98.com/https://github.com/infiniflow/ragflow.git synced 2025-08-12 06:49:00 +08:00
Code Issues Packages Projects Releases Wiki Activity

fix ppt file from filemanager error (#2470)

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

#2467

### Type of change

- [x] Bug Fix (non-breaking change which fixes an issue)
This commit is contained in:
Kevin Hu 2024-09-18 09:22:14 +08:00 committed by GitHub
parent e41268efc6
commit e7dd487779
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
2 changed files with 14 additions and 9 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

1
api/apps/file2document_app.py
View File

@ -85,6 +85,7 @@ def convert():
"location": file.location,
"size": file.size
})
FileService.set_constant_parser(doc, file.name)
file2document = File2DocumentService.insert({
"id": get_uuid(),
"file_id": id,

22
api/db/services/file_service.py
View File

@ -366,14 +366,7 @@ class FileService(CommonService):
"size": len(blob),
"thumbnail": thumbnail(filename, blob)
}
if doc["type"] == FileType.VISUAL:
doc["parser_id"] = ParserType.PICTURE.value
if doc["type"] == FileType.AURAL:
doc["parser_id"] = ParserType.AUDIO.value
if re.search(r"\.(ppt|pptx|pages)$", filename):
doc["parser_id"] = ParserType.PRESENTATION.value
if re.search(r"\.(eml)$", filename):
doc["parser_id"] = ParserType.EMAIL.value
self.set_constant_parser(doc, filename)
DocumentService.insert(doc)
FileService.add_file_from_kb(doc, kb_folder["id"], kb.tenant_id)
@ -381,4 +374,15 @@ class FileService(CommonService):
except Exception as e:
err.append(file.filename + ": " + str(e))
return err, files
return err, files
@staticmethod
def set_constant_parser(doc, filename):
if doc["type"] == FileType.VISUAL:
doc["parser_id"] = ParserType.PICTURE.value
if doc["type"] == FileType.AURAL:
doc["parser_id"] = ParserType.AUDIO.value
if re.search(r"\.(ppt|pptx|pages)$", filename):
doc["parser_id"] = ParserType.PRESENTATION.value
if re.search(r"\.(eml)$", filename):
doc["parser_id"] = ParserType.EMAIL.value