admin-valentin/open-notebook
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

fix pdf encoding

Browse source
This commit is contained in:
LUIS NOVO 2024-10-24 13:24:39 -03:00
parent 356b2a5434
commit 10ec265a44
1 changed files with 3 additions and 1 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

4
open_notebook/graphs/content_process.py
View file

@ -1,4 +1,5 @@
import re import re
import unicodedata
import fitz # type: ignore import fitz # type: ignore
import magic import magic
@ -68,7 +69,8 @@ def _extract_text_from_pdf(pdf_path):
for page in doc: for page in doc:
text += page.get_text() text += page.get_text()
doc.close() doc.close()
return text normalized_content = unicodedata.normalize("NFKD", text)
return normalized_content
def extract_pdf(state: SourceState): def extract_pdf(state: SourceState):