lgcosta/paperless-ngx
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Support more characters known to be in the month names of other languages

Browse Source
This commit is contained in:
Fabien Dubuy 2024-06-25 21:41:08 +02:00 committed by GitHub
parent 759bd11fea
commit 405acc7f7b
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 8 additions and 7 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

15
src/documents/parsers.py
View File

@ -37,13 +37,14 @@ from documents.utils import run_subprocess
# TODO: isn't there a date parsing library for this?
DATE_REGEX = re.compile(
r"(\b|(?!=([_-])))([0-9]{1,2})[\.\/-]([0-9]{1,2})[\.\/-]([0-9]{4}|[0-9]{2})(\b|(?=([_-])))|"
r"(\b|(?!=([_-])))([0-9]{4}|[0-9]{2})[\.\/-]([0-9]{1,2})[\.\/-]([0-9]{1,2})(\b|(?=([_-])))|"
r"(\b|(?!=([_-])))([0-9]{1,2}[\. ]+[a-zA-ZÀ-ÖØ-öø-ÿ]{3,9} [0-9]{4}|[a-zA-ZÀ-ÖØ-öø-ÿ]{3,9} [0-9]{1,2}, [0-9]{4})(\b|(?=([_-])))|"
r"(\b|(?!=([_-])))([^\W\d_]{3,9} [0-9]{1,2}, ([0-9]{4}))(\b|(?=([_-])))|"
r"(\b|(?!=([_-])))([^\W\d_]{3,9} [0-9]{4})(\b|(?=([_-])))|"
r"(\b|(?!=([_-])))([0-9]{1,2}[^ ]{2}[\. ]+[^ ]{3,9}[ \.\/-][0-9]{4})(\b|(?=([_-])))|"
r"(\b|(?!=([_-])))(\b[0-9]{1,2}[ \.\/-][a-zA-ZÀ-ÖØ-öø-ÿ]{3}[ \.\/-][0-9]{4})(\b|(?=([_-])))",
r"(\b|(?!=([_-])))(\d{1,2})[\.\/-](\d{1,2})[\.\/-](\d{4}|\d{2})(\b|(?=([_-])))|"
r"(\b|(?!=([_-])))(\d{4}|\d{2})[\.\/-](\d{1,2})[\.\/-](\d{1,2})(\b|(?=([_-])))|"
r"(\b|(?!=([_-])))(\d{1,2}[\. ]+[a-zéûäëčžúřěáíóńźçŞığü]{3,9} \d{4}|[a-zéûäëčžúřěáíóńźçŞığü]{3,9} \d{1,2}, \d{4})(\b|(?=([_-])))|"
r"(\b|(?!=([_-])))([^\W\d_]{3,9} \d{1,2}, (\d{4}))(\b|(?=([_-])))|"
r"(\b|(?!=([_-])))([^\W\d_]{3,9} \d{4})(\b|(?=([_-])))|"
r"(\b|(?!=([_-])))(\d{1,2}[^ ]{2}[\. ]+[^ ]{3,9}[ \.\/-]\d{4})(\b|(?=([_-])))|"
r"(\b|(?!=([_-])))(\b\d{1,2}[ \.\/-][a-zéûäëčžúřěáíóńźçŞığü]{3}[ \.\/-]\d{4})(\b|(?=([_-])))",
re.IGNORECASE,
)