Spaces:
Paused
Paused
Upload persian_legal_scraper.py
Browse files- app/persian_legal_scraper.py +1720 -0
app/persian_legal_scraper.py
ADDED
|
@@ -0,0 +1,1720 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
# app.py - سیستم پیشرفته استخراج و تحلیل اسناد حقوقی فارسی
|
| 2 |
+
import os
|
| 3 |
+
import gc
|
| 4 |
+
import sys
|
| 5 |
+
import time
|
| 6 |
+
import json
|
| 7 |
+
import logging
|
| 8 |
+
import resource
|
| 9 |
+
import requests
|
| 10 |
+
import threading
|
| 11 |
+
import re
|
| 12 |
+
import random
|
| 13 |
+
from pathlib import Path
|
| 14 |
+
from datetime import datetime
|
| 15 |
+
from typing import List, Dict, Any, Optional, Tuple, Union
|
| 16 |
+
from dataclasses import dataclass
|
| 17 |
+
from concurrent.futures import ThreadPoolExecutor, as_completed
|
| 18 |
+
from urllib.parse import urljoin, urlparse
|
| 19 |
+
import hashlib
|
| 20 |
+
|
| 21 |
+
import gradio as gr
|
| 22 |
+
import pandas as pd
|
| 23 |
+
import torch
|
| 24 |
+
from bs4 import BeautifulSoup
|
| 25 |
+
from transformers import (
|
| 26 |
+
AutoTokenizer,
|
| 27 |
+
AutoModelForSequenceClassification,
|
| 28 |
+
pipeline,
|
| 29 |
+
logging as transformers_logging
|
| 30 |
+
)
|
| 31 |
+
import warnings
|
| 32 |
+
|
| 33 |
+
# تنظیمات اولیه
|
| 34 |
+
warnings.filterwarnings('ignore')
|
| 35 |
+
transformers_logging.set_verbosity_error()
|
| 36 |
+
|
| 37 |
+
# محدودیت حافظه برای HF Spaces
|
| 38 |
+
try:
|
| 39 |
+
resource.setrlimit(resource.RLIMIT_AS, (2*1024*1024*1024, 2*1024*1024*1024))
|
| 40 |
+
except:
|
| 41 |
+
pass
|
| 42 |
+
|
| 43 |
+
# تنظیمات محیط
|
| 44 |
+
os.environ['TRANSFORMERS_CACHE'] = '/tmp/hf_cache'
|
| 45 |
+
os.environ['HF_HOME'] = '/tmp/hf_cache'
|
| 46 |
+
os.environ['TORCH_HOME'] = '/tmp/torch_cache'
|
| 47 |
+
os.environ['TOKENIZERS_PARALLELISM'] = 'false'
|
| 48 |
+
|
| 49 |
+
# تنظیم لاگ
|
| 50 |
+
logging.basicConfig(level=logging.INFO, format='%(asctime)s - %(levelname)s - %(message)s')
|
| 51 |
+
logger = logging.getLogger(__name__)
|
| 52 |
+
|
| 53 |
+
# SVG Icons
|
| 54 |
+
SVG_ICONS = {
|
| 55 |
+
'search': '🔍',
|
| 56 |
+
'document': '📄',
|
| 57 |
+
'analyze': '🤖',
|
| 58 |
+
'export': '📊',
|
| 59 |
+
'settings': '⚙️',
|
| 60 |
+
'preview': '👁️',
|
| 61 |
+
'link': '🔗',
|
| 62 |
+
'success': '✅',
|
| 63 |
+
'error': '❌',
|
| 64 |
+
'warning': '⚠️'
|
| 65 |
+
}
|
| 66 |
+
|
| 67 |
+
# منابع حقوقی معتبر ایران
|
| 68 |
+
LEGAL_SOURCES_CONFIG = {
|
| 69 |
+
"مجلس شورای اسلامی": {
|
| 70 |
+
"base_url": "https://rc.majlis.ir",
|
| 71 |
+
"patterns": ["/fa/law/", "/fa/report/", "/fa/news/"],
|
| 72 |
+
"selectors": [".main-content", ".article-body", ".content", "article"],
|
| 73 |
+
"delay_range": (2, 5),
|
| 74 |
+
"max_depth": 2
|
| 75 |
+
},
|
| 76 |
+
"پورتال ملی قوانین": {
|
| 77 |
+
"base_url": "https://www.dotic.ir",
|
| 78 |
+
"patterns": ["/portal/", "/law/", "/regulation/"],
|
| 79 |
+
"selectors": [".content-area", ".law-content", ".main-text"],
|
| 80 |
+
"delay_range": (1, 4),
|
| 81 |
+
"max_depth": 2
|
| 82 |
+
},
|
| 83 |
+
"قوه قضاییه": {
|
| 84 |
+
"base_url": "https://www.judiciary.ir",
|
| 85 |
+
"patterns": ["/fa/news/", "/fa/verdict/", "/fa/law/"],
|
| 86 |
+
"selectors": [".news-content", ".verdict-text", ".main-content"],
|
| 87 |
+
"delay_range": (3, 6),
|
| 88 |
+
"max_depth": 2
|
| 89 |
+
},
|
| 90 |
+
"وزارت دادگستری": {
|
| 91 |
+
"base_url": "https://www.moj.ir",
|
| 92 |
+
"patterns": ["/fa/news/", "/fa/law/", "/fa/regulation/"],
|
| 93 |
+
"selectors": [".news-body", ".law-text", ".content"],
|
| 94 |
+
"delay_range": (2, 4),
|
| 95 |
+
"max_depth": 2
|
| 96 |
+
},
|
| 97 |
+
"دیوان عدالت اداری": {
|
| 98 |
+
"base_url": "https://www.adcourt.ir",
|
| 99 |
+
"patterns": ["/fa/verdict/", "/fa/news/"],
|
| 100 |
+
"selectors": [".verdict-content", ".news-text"],
|
| 101 |
+
"delay_range": (1, 3),
|
| 102 |
+
"max_depth": 2
|
| 103 |
+
}
|
| 104 |
+
}
|
| 105 |
+
|
| 106 |
+
# واژگان حقوقی فارسی پیشرفته
|
| 107 |
+
PERSIAN_LEGAL_DICTIONARY = {
|
| 108 |
+
"قوانین_اساسی": [
|
| 109 |
+
"قانون اساسی", "اصول قانون اساسی", "مجلس شورای اسلامی", "شورای نگهبان",
|
| 110 |
+
"رهبری", "جمهوری اسلامی", "حاکمیت ملی", "ولایت فقیه"
|
| 111 |
+
],
|
| 112 |
+
"قوانین_عادی": [
|
| 113 |
+
"ماده", "تبصره", "اصل", "فصل", "باب", "قسمت", "بخش", "کتاب",
|
| 114 |
+
"قانون مدنی", "قانون جزا", "قانون آیین دادرسی", "قانون تجارت"
|
| 115 |
+
],
|
| 116 |
+
"مقررات_اجرایی": [
|
| 117 |
+
"آییننامه", "مقرره", "دستورالعمل", "شیوهنامه", "بند", "جزء", "فقره",
|
| 118 |
+
"ضابطه", "رهنمود", "دستور", "اعلامیه", "ابلاغیه"
|
| 119 |
+
],
|
| 120 |
+
"اصطلاحات_حقوقی": [
|
| 121 |
+
"شخص حقیقی", "شخص حقوقی", "حق", "تکلیف", "مسئولیت", "جرم", "مجازات",
|
| 122 |
+
"دعوا", "طرف دعوا", "خواهان", "خوانده", "شاکی", "متهم", "مجنیعلیه"
|
| 123 |
+
],
|
| 124 |
+
"نهادهای_قضایی": [
|
| 125 |
+
"دادگاه", "قاضی", "دادرس", "مدعیالعموم", "وکیل", "کارشناس", "مترجم",
|
| 126 |
+
"رای", "حکم", "قرار", "اجرائیه", "کیفرخواست", "لایحه دفاعیه"
|
| 127 |
+
],
|
| 128 |
+
"اصطلاحات_اداری": [
|
| 129 |
+
"وزارت", "اداره", "سازمان", "مدیر", "مقام", "مسئول", "کارمند", "کارگزار",
|
| 130 |
+
"بخشنامه", "تصویبنامه", "مصوبه", "تصمیم", "نظریه", "استعلام"
|
| 131 |
+
],
|
| 132 |
+
"مفاهیم_مالی": [
|
| 133 |
+
"مالیات", "عوارض", "پرداخت", "وجه", "ریال", "درهم", "خسارت", "دیه",
|
| 134 |
+
"تأمین", "ضمانت", "وثیقه", "سپرده", "جریمه", "جزا�� نقدی"
|
| 135 |
+
]
|
| 136 |
+
}
|
| 137 |
+
|
| 138 |
+
@dataclass
|
| 139 |
+
class ProcessingProgress:
|
| 140 |
+
current_step: str = ""
|
| 141 |
+
progress: float = 0.0
|
| 142 |
+
total_documents: int = 0
|
| 143 |
+
processed_documents: int = 0
|
| 144 |
+
status: str = "آماده"
|
| 145 |
+
error: Optional[str] = None
|
| 146 |
+
|
| 147 |
+
class MemoryManager:
|
| 148 |
+
"""مدیریت هوشمند حافظه برای HF Spaces"""
|
| 149 |
+
|
| 150 |
+
@staticmethod
|
| 151 |
+
def get_memory_usage() -> float:
|
| 152 |
+
try:
|
| 153 |
+
with open('/proc/self/status') as f:
|
| 154 |
+
for line in f:
|
| 155 |
+
if line.startswith('VmRSS:'):
|
| 156 |
+
return float(line.split()[1]) / 1024
|
| 157 |
+
return 0.0
|
| 158 |
+
except:
|
| 159 |
+
return 0.0
|
| 160 |
+
|
| 161 |
+
@staticmethod
|
| 162 |
+
def check_memory_available(required_mb: float) -> bool:
|
| 163 |
+
try:
|
| 164 |
+
current_usage = MemoryManager.get_memory_usage()
|
| 165 |
+
return current_usage + required_mb < 1800
|
| 166 |
+
except:
|
| 167 |
+
return True
|
| 168 |
+
|
| 169 |
+
@staticmethod
|
| 170 |
+
def cleanup_memory():
|
| 171 |
+
gc.collect()
|
| 172 |
+
if torch.cuda.is_available():
|
| 173 |
+
torch.cuda.empty_cache()
|
| 174 |
+
|
| 175 |
+
class SmartTextProcessor:
|
| 176 |
+
"""پردازشگر هوشمند متن با قابلیتهای پیشرفته"""
|
| 177 |
+
|
| 178 |
+
def __init__(self):
|
| 179 |
+
self.sentence_endings = ['۔', '.', '!', '؟', '?', ';', '؛']
|
| 180 |
+
self.paragraph_indicators = ['ماده', 'تبصره', 'بند', 'الف', 'ب', 'ج', 'د', 'ه', 'و']
|
| 181 |
+
|
| 182 |
+
# الگوهای شناسایی ارجاعات حقوقی
|
| 183 |
+
self.citation_patterns = [
|
| 184 |
+
r'ماده\s*(\d+)',
|
| 185 |
+
r'تبصره\s*(\d+)',
|
| 186 |
+
r'بند\s*([الف-ی]|\d+)',
|
| 187 |
+
r'فصل\s*(\d+)',
|
| 188 |
+
r'باب\s*(\d+)',
|
| 189 |
+
r'قسمت\s*(\d+)',
|
| 190 |
+
r'اصل\s*(\d+)'
|
| 191 |
+
]
|
| 192 |
+
|
| 193 |
+
# الگوهای پاکسازی متن
|
| 194 |
+
self.cleanup_patterns = [
|
| 195 |
+
(r'\s+', ' '),
|
| 196 |
+
(r'([۰-۹])\s+([۰-۹])', r'\1\2'),
|
| 197 |
+
(r'([a-zA-Z])\s+([a-zA-Z])', r'\1\2'),
|
| 198 |
+
(r'([ا-ی])\s+(ها|های|ان|ات|ین)', r'\1\2'),
|
| 199 |
+
(r'(می|نمی|خواهد)\s+(شود|گردد|باشد)', r'\1\2'),
|
| 200 |
+
]
|
| 201 |
+
|
| 202 |
+
def normalize_persian_text(self, text: str) -> str:
|
| 203 |
+
"""نرمالسازی پیشرفته متن فارسی"""
|
| 204 |
+
if not text:
|
| 205 |
+
return ""
|
| 206 |
+
|
| 207 |
+
# نرمالسازی کاراکترهای فارسی
|
| 208 |
+
persian_normalization = {
|
| 209 |
+
'ي': 'ی', 'ك': 'ک', 'ة': 'ه', 'ؤ': 'و', 'إ': 'ا', 'أ': 'ا',
|
| 210 |
+
'ء': '', 'ئ': 'ی', '٠': '۰', '١': '۱', '٢': '۲', '٣': '۳', '٤': '۴',
|
| 211 |
+
'٥': '۵', '٦': '۶', '٧': '۷', '٨': '۸', '٩': '۹'
|
| 212 |
+
}
|
| 213 |
+
|
| 214 |
+
for old, new in persian_normalization.items():
|
| 215 |
+
text = text.replace(old, new)
|
| 216 |
+
|
| 217 |
+
# اعمال الگوهای پاکسازی
|
| 218 |
+
for pattern, replacement in self.cleanup_patterns:
|
| 219 |
+
text = re.sub(pattern, replacement, text)
|
| 220 |
+
|
| 221 |
+
# حذف کاراکترهای غیرضروری
|
| 222 |
+
text = re.sub(r'[^\u0600-\u06FF\u200C\u200D\s\w\d.,;:!؟()«»\-]', '', text)
|
| 223 |
+
|
| 224 |
+
return text.strip()
|
| 225 |
+
|
| 226 |
+
def detect_long_sentences(self, text: str) -> List[Dict[str, Any]]:
|
| 227 |
+
"""تشخیص جملات طولانی و پیچیده"""
|
| 228 |
+
sentences = self._split_into_sentences(text)
|
| 229 |
+
long_sentences = []
|
| 230 |
+
|
| 231 |
+
for i, sentence in enumerate(sentences):
|
| 232 |
+
sentence_info = {
|
| 233 |
+
'index': i,
|
| 234 |
+
'text': sentence,
|
| 235 |
+
'word_count': len(sentence.split()),
|
| 236 |
+
'is_long': False,
|
| 237 |
+
'suggestions': []
|
| 238 |
+
}
|
| 239 |
+
|
| 240 |
+
# بررسی طول جمله
|
| 241 |
+
if sentence_info['word_count'] > 30:
|
| 242 |
+
sentence_info['is_long'] = True
|
| 243 |
+
sentence_info['suggestions'].append('جمله بسیار طولانی - تقسیم توصیه میشود')
|
| 244 |
+
|
| 245 |
+
# بررسی پیچیدگی
|
| 246 |
+
complexity_score = self._calculate_complexity(sentence)
|
| 247 |
+
if complexity_score > 5:
|
| 248 |
+
sentence_info['suggestions'].append('جمله پیچیده - سادهسازی توصیه میشود')
|
| 249 |
+
|
| 250 |
+
if sentence_info['is_long'] or sentence_info['suggestions']:
|
| 251 |
+
long_sentences.append(sentence_info)
|
| 252 |
+
|
| 253 |
+
return long_sentences
|
| 254 |
+
|
| 255 |
+
def _split_into_sentences(self, text: str) -> List[str]:
|
| 256 |
+
"""تقسیم متن به جملات"""
|
| 257 |
+
boundaries = self.detect_sentence_boundaries(text)
|
| 258 |
+
sentences = []
|
| 259 |
+
|
| 260 |
+
start = 0
|
| 261 |
+
for boundary in boundaries:
|
| 262 |
+
sentence = text[start:boundary].strip()
|
| 263 |
+
if sentence and len(sentence) > 5:
|
| 264 |
+
sentences.append(sentence)
|
| 265 |
+
start = boundary
|
| 266 |
+
|
| 267 |
+
# جمله آخر
|
| 268 |
+
if start < len(text):
|
| 269 |
+
last_sentence = text[start:].strip()
|
| 270 |
+
if last_sentence and len(last_sentence) > 5:
|
| 271 |
+
sentences.append(last_sentence)
|
| 272 |
+
|
| 273 |
+
return sentences
|
| 274 |
+
|
| 275 |
+
def _calculate_complexity(self, sentence: str) -> float:
|
| 276 |
+
"""محاسبه پیچیدگی جمله"""
|
| 277 |
+
complexity = 0
|
| 278 |
+
|
| 279 |
+
# تعداد کلمات ربط
|
| 280 |
+
conjunctions = ['که', 'اگر', 'چون', 'زیرا', 'ولی', 'اما', 'درحالیکه', 'درصورتیکه']
|
| 281 |
+
complexity += sum(sentence.count(conj) for conj in conjunctions) * 0.5
|
| 282 |
+
|
| 283 |
+
# تعداد ویرگول
|
| 284 |
+
complexity += sentence.count('،') * 0.3
|
| 285 |
+
|
| 286 |
+
# تعداد کلمات
|
| 287 |
+
complexity += len(sentence.split()) * 0.02
|
| 288 |
+
|
| 289 |
+
# جملات تودرتو
|
| 290 |
+
if sentence.count('(') > 0:
|
| 291 |
+
complexity += sentence.count('(') * 0.5
|
| 292 |
+
|
| 293 |
+
return complexity
|
| 294 |
+
|
| 295 |
+
def detect_sentence_boundaries(self, text: str) -> List[int]:
|
| 296 |
+
"""تشخیص هوشمند مرزهای جمله در متون حقوقی فارسی"""
|
| 297 |
+
boundaries = []
|
| 298 |
+
|
| 299 |
+
for i, char in enumerate(text):
|
| 300 |
+
if char in self.sentence_endings:
|
| 301 |
+
is_real_ending = True
|
| 302 |
+
|
| 303 |
+
# بررسی برای اعداد و اختصارات
|
| 304 |
+
if i > 0 and text[i-1].isdigit() and char == '.':
|
| 305 |
+
is_real_ending = False
|
| 306 |
+
|
| 307 |
+
# بررسی برای اختصارات رایج
|
| 308 |
+
if i > 2:
|
| 309 |
+
prev_text = text[max(0, i-10):i].strip()
|
| 310 |
+
if any(abbr in prev_text for abbr in ['ماده', 'بند', 'ج.ا.ا', 'ق.م', 'ق.ج']):
|
| 311 |
+
if char == '.' and i < len(text) - 1 and not text[i+1].isspace():
|
| 312 |
+
is_real_ending = False
|
| 313 |
+
|
| 314 |
+
if is_real_ending:
|
| 315 |
+
if i < len(text) - 1:
|
| 316 |
+
next_char = text[i + 1]
|
| 317 |
+
if next_char.isspace() or next_char in '«"\'':
|
| 318 |
+
boundaries.append(i + 1)
|
| 319 |
+
else:
|
| 320 |
+
boundaries.append(i + 1)
|
| 321 |
+
|
| 322 |
+
return boundaries
|
| 323 |
+
|
| 324 |
+
def reconstruct_legal_text(self, content_fragments: List[str]) -> str:
|
| 325 |
+
"""بازسازی هوشمند متن حقوقی از قطعات پراکنده"""
|
| 326 |
+
if not content_fragments:
|
| 327 |
+
return ""
|
| 328 |
+
|
| 329 |
+
# مرحله 1: نرمالسازی تمام قطعات
|
| 330 |
+
normalized_fragments = []
|
| 331 |
+
for fragment in content_fragments:
|
| 332 |
+
normalized = self.normalize_persian_text(fragment)
|
| 333 |
+
if normalized and len(normalized.strip()) > 10:
|
| 334 |
+
normalized_fragments.append(normalized)
|
| 335 |
+
|
| 336 |
+
if not normalized_fragments:
|
| 337 |
+
return ""
|
| 338 |
+
|
| 339 |
+
# مرحله 2: ادغام هوشمند قطعات
|
| 340 |
+
combined_text = self._smart_join_fragments(normalized_fragments)
|
| 341 |
+
|
| 342 |
+
# مرحله 3: اعمال قالببندی حقوقی
|
| 343 |
+
formatted = self._apply_legal_formatting(combined_text)
|
| 344 |
+
|
| 345 |
+
return formatted
|
| 346 |
+
|
| 347 |
+
def _smart_join_fragments(self, fragments: List[str]) -> str:
|
| 348 |
+
"""ادغام هوشمند قطعات با در نظر گیری زمینه"""
|
| 349 |
+
if len(fragments) == 1:
|
| 350 |
+
return fragments[0]
|
| 351 |
+
|
| 352 |
+
result = [fragments[0]]
|
| 353 |
+
|
| 354 |
+
for i in range(1, len(fragments)):
|
| 355 |
+
current_fragment = fragments[i]
|
| 356 |
+
prev_fragment = result[-1]
|
| 357 |
+
|
| 358 |
+
# بررسی ادامه جمله
|
| 359 |
+
if self._should_continue_sentence(prev_fragment, current_fragment):
|
| 360 |
+
result[-1] += ' ' + current_fragment
|
| 361 |
+
# بررسی ادغام بدون فاصله (نیمفاصله)
|
| 362 |
+
elif self._should_join_without_space(prev_fragment, current_fragment):
|
| 363 |
+
result[-1] += current_fragment
|
| 364 |
+
# شروع پاراگراف جدید
|
| 365 |
+
elif self._is_new_paragraph(current_fragment):
|
| 366 |
+
result.append('\n\n' + current_fragment)
|
| 367 |
+
else:
|
| 368 |
+
result.append(' ' + current_fragment)
|
| 369 |
+
|
| 370 |
+
return ''.join(result)
|
| 371 |
+
|
| 372 |
+
def _should_continue_sentence(self, prev: str, current: str) -> bool:
|
| 373 |
+
"""تشخیص ادامه جمله"""
|
| 374 |
+
# کلمات ادامهدهنده
|
| 375 |
+
continuation_words = ['که', 'تا', 'اگر', 'چون', 'زیرا', 'ولی', 'اما', 'و', 'یا']
|
| 376 |
+
|
| 377 |
+
# اگر جمله قبلی ناتمام باشد
|
| 378 |
+
if not any(prev.endswith(end) for end in self.sentence_endings):
|
| 379 |
+
return True
|
| 380 |
+
|
| 381 |
+
# اگر جمله فعلی با کلمه ادامه شروع شود
|
| 382 |
+
if any(current.strip().startswith(word) for word in continuation_words):
|
| 383 |
+
return True
|
| 384 |
+
|
| 385 |
+
return False
|
| 386 |
+
|
| 387 |
+
def _should_join_without_space(self, prev: str, current: str) -> bool:
|
| 388 |
+
"""تشخیص ادغام بدون فاصله"""
|
| 389 |
+
# پسوندها و پیشوندها
|
| 390 |
+
suffixes = ['��ا', 'های', 'ان', 'ات', 'ین', 'تان', 'شان', 'تون', 'شون']
|
| 391 |
+
prefixes = ['می', 'نمی', 'برمی', 'درمی']
|
| 392 |
+
|
| 393 |
+
current_stripped = current.strip()
|
| 394 |
+
|
| 395 |
+
# بررسی پسوندها
|
| 396 |
+
if any(current_stripped.startswith(suffix) for suffix in suffixes):
|
| 397 |
+
return True
|
| 398 |
+
|
| 399 |
+
# بررسی ادامه فعل
|
| 400 |
+
if prev.endswith(('می', 'نمی', 'خواهد', 'است')):
|
| 401 |
+
verb_continuations = ['شود', 'گردد', 'باشد', 'کرد', 'کند']
|
| 402 |
+
if any(current_stripped.startswith(cont) for cont in verb_continuations):
|
| 403 |
+
return True
|
| 404 |
+
|
| 405 |
+
return False
|
| 406 |
+
|
| 407 |
+
def _is_new_paragraph(self, text: str) -> bool:
|
| 408 |
+
"""تشخیص شروع پاراگراف جدید"""
|
| 409 |
+
text_stripped = text.strip()
|
| 410 |
+
|
| 411 |
+
# شاخصهای پاراگراف در متون حقوقی
|
| 412 |
+
paragraph_starters = [
|
| 413 |
+
'ماده', 'تبصره', 'بند', 'فصل', 'باب', 'قسمت', 'کتاب',
|
| 414 |
+
'الف)', 'ب)', 'ج)', 'د)', 'ه)', 'و)', 'ز)', 'ح)', 'ط)',
|
| 415 |
+
'۱-', '۲-', '۳-', '۴-', '۵-', '۶-', '۷-', '۸-', '۹-', '۱۰-'
|
| 416 |
+
]
|
| 417 |
+
|
| 418 |
+
return any(text_stripped.startswith(starter) for starter in paragraph_starters)
|
| 419 |
+
|
| 420 |
+
def _apply_legal_formatting(self, text: str) -> str:
|
| 421 |
+
"""اعمال قالببندی مخصوص اسناد حقوقی"""
|
| 422 |
+
lines = text.split('\n')
|
| 423 |
+
formatted_lines = []
|
| 424 |
+
|
| 425 |
+
for line in lines:
|
| 426 |
+
line = line.strip()
|
| 427 |
+
if not line:
|
| 428 |
+
continue
|
| 429 |
+
|
| 430 |
+
# قالببندی مواد و تبصرهها
|
| 431 |
+
if any(line.startswith(indicator) for indicator in ['ماده', 'تبصره']):
|
| 432 |
+
formatted_lines.append(f"\n{line}")
|
| 433 |
+
# قالببندی بندها
|
| 434 |
+
elif any(line.startswith(indicator) for indicator in ['الف)', 'ب)', 'ج)']):
|
| 435 |
+
formatted_lines.append(f" {line}")
|
| 436 |
+
# قالببندی فصول و ابواب
|
| 437 |
+
elif any(line.startswith(indicator) for indicator in ['فصل', 'باب', 'کتاب']):
|
| 438 |
+
formatted_lines.append(f"\n\n{line.upper()}\n")
|
| 439 |
+
else:
|
| 440 |
+
formatted_lines.append(line)
|
| 441 |
+
|
| 442 |
+
return '\n'.join(formatted_lines)
|
| 443 |
+
|
| 444 |
+
def extract_legal_entities(self, text: str) -> Dict[str, List[str]]:
|
| 445 |
+
"""استخراج موجودیتهای حقوقی از متن"""
|
| 446 |
+
entities = {
|
| 447 |
+
'articles': [],
|
| 448 |
+
'citations': [],
|
| 449 |
+
'legal_terms': [],
|
| 450 |
+
'organizations': [],
|
| 451 |
+
'laws': []
|
| 452 |
+
}
|
| 453 |
+
|
| 454 |
+
# استخراج مواد و تبصرهها
|
| 455 |
+
for pattern in self.citation_patterns:
|
| 456 |
+
matches = re.findall(pattern, text)
|
| 457 |
+
if matches:
|
| 458 |
+
entities['citations'].extend(matches)
|
| 459 |
+
|
| 460 |
+
# استخراج اصطلاحات حقوقی
|
| 461 |
+
for category, terms in PERSIAN_LEGAL_DICTIONARY.items():
|
| 462 |
+
found_terms = [term for term in terms if term in text]
|
| 463 |
+
entities['legal_terms'].extend(found_terms)
|
| 464 |
+
|
| 465 |
+
# استخراج نام قوانین
|
| 466 |
+
law_patterns = [
|
| 467 |
+
r'قانون\s+([^۔\.\n]{5,50})',
|
| 468 |
+
r'آییننامه\s+([^۔\.\n]{5,50})',
|
| 469 |
+
r'مقرره\s+([^۔\.\n]{5,50})'
|
| 470 |
+
]
|
| 471 |
+
|
| 472 |
+
for pattern in law_patterns:
|
| 473 |
+
matches = re.findall(pattern, text)
|
| 474 |
+
entities['laws'].extend(matches)
|
| 475 |
+
|
| 476 |
+
return entities
|
| 477 |
+
|
| 478 |
+
class AntiDDoSManager:
|
| 479 |
+
"""مدیریت ضد حملات DDoS و تنوع درخواستها"""
|
| 480 |
+
|
| 481 |
+
def __init__(self):
|
| 482 |
+
self.request_history = {}
|
| 483 |
+
self.user_agents = [
|
| 484 |
+
'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/91.0.4472.124 Safari/537.36',
|
| 485 |
+
'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/91.0.4472.124 Safari/537.36',
|
| 486 |
+
'Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/91.0.4472.124 Safari/537.36',
|
| 487 |
+
'Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:89.0) Gecko/20100101 Firefox/89.0',
|
| 488 |
+
'Mozilla/5.0 (Macintosh; Intel Mac OS X 10.15; rv:89.0) Gecko/20100101 Firefox/89.0'
|
| 489 |
+
]
|
| 490 |
+
|
| 491 |
+
def get_request_delay(self, domain: str) -> float:
|
| 492 |
+
"""محاسبه تأخیر مناسب برای هر منبع"""
|
| 493 |
+
source_info = self._identify_source(domain)
|
| 494 |
+
|
| 495 |
+
if source_info and 'delay_range' in source_info:
|
| 496 |
+
min_delay, max_delay = source_info['delay_range']
|
| 497 |
+
return random.uniform(min_delay, max_delay)
|
| 498 |
+
|
| 499 |
+
# تأخیر پیشفرض
|
| 500 |
+
return random.uniform(1, 3)
|
| 501 |
+
|
| 502 |
+
def get_random_headers(self) -> Dict[str, str]:
|
| 503 |
+
"""تولید هدرهای تصادفی"""
|
| 504 |
+
return {
|
| 505 |
+
'User-Agent': random.choice(self.user_agents),
|
| 506 |
+
'Accept': 'text/html,application/xhtml+xml,application/xml;q=0.9,image/webp,*/*;q=0.8',
|
| 507 |
+
'Accept-Language': random.choice(['fa,en;q=0.9', 'fa-IR,fa;q=0.9,en;q=0.8', 'fa;q=0.9,en;q=0.8']),
|
| 508 |
+
'Accept-Encoding': 'gzip, deflate',
|
| 509 |
+
'Connection': 'keep-alive',
|
| 510 |
+
'Upgrade-Insecure-Requests': '1',
|
| 511 |
+
'Cache-Control': random.choice(['no-cache', 'max-age=0', 'no-store'])
|
| 512 |
+
}
|
| 513 |
+
|
| 514 |
+
def _identify_source(self, domain: str) -> Optional[Dict]:
|
| 515 |
+
"""شناسایی منبع بر اساس دامنه"""
|
| 516 |
+
for source_name, config in LEGAL_SOURCES_CONFIG.items():
|
| 517 |
+
base_domain = config['base_url'].replace('https://', '').replace('http://', '')
|
| 518 |
+
if base_domain in domain:
|
| 519 |
+
return config
|
| 520 |
+
return None
|
| 521 |
+
|
| 522 |
+
def should_allow_request(self, domain: str) -> bool:
|
| 523 |
+
"""بررسی اینکه آیا درخواست مجاز است"""
|
| 524 |
+
current_time = time.time()
|
| 525 |
+
|
| 526 |
+
if domain not in self.request_history:
|
| 527 |
+
self.request_history[domain] = []
|
| 528 |
+
|
| 529 |
+
# حذف درخواستهای قدیمی (بیش از 1 ساعت)
|
| 530 |
+
self.request_history[domain] = [
|
| 531 |
+
req_time for req_time in self.request_history[domain]
|
| 532 |
+
if current_time - req_time < 3600
|
| 533 |
+
]
|
| 534 |
+
|
| 535 |
+
# بررسی تعداد درخواستها در ساعت گذشته
|
| 536 |
+
if len(self.request_history[domain]) >= 50: # حداکثر 50 درخواست در ساعت
|
| 537 |
+
return False
|
| 538 |
+
|
| 539 |
+
# ثبت درخواست جدید
|
| 540 |
+
self.request_history[domain].append(current_time)
|
| 541 |
+
return True
|
| 542 |
+
|
| 543 |
+
class ModelManager:
|
| 544 |
+
"""مدیریت پیشرفته مدلهای هوش مصنوعی"""
|
| 545 |
+
|
| 546 |
+
def __init__(self):
|
| 547 |
+
self.models = {}
|
| 548 |
+
self.model_status = {}
|
| 549 |
+
|
| 550 |
+
def load_models_progressively(self, progress_callback=None) -> Dict[str, Any]:
|
| 551 |
+
"""بارگذاری تدریجی مدلها با مدیریت حافظه"""
|
| 552 |
+
logger.info("🚀 شروع بارگذاری مدلهای هوش مصنوعی...")
|
| 553 |
+
|
| 554 |
+
if progress_callback:
|
| 555 |
+
progress_callback("آمادهسازی محیط...", 0.05)
|
| 556 |
+
|
| 557 |
+
# ایجاد دایرکتوری کش
|
| 558 |
+
cache_dir = "/tmp/hf_cache"
|
| 559 |
+
os.makedirs(cache_dir, exist_ok=True)
|
| 560 |
+
|
| 561 |
+
# فاز 1: Tokenizer
|
| 562 |
+
try:
|
| 563 |
+
if MemoryManager.check_memory_available(100):
|
| 564 |
+
logger.info("📝 بارگذاری Tokenizer...")
|
| 565 |
+
if progress_callback:
|
| 566 |
+
progress_callback("بارگذاری Tokenizer...", 0.2)
|
| 567 |
+
|
| 568 |
+
self.models['tokenizer'] = AutoTokenizer.from_pretrained(
|
| 569 |
+
"HooshvareLab/bert-fa-base-uncased",
|
| 570 |
+
cache_dir=cache_dir,
|
| 571 |
+
local_files_only=False
|
| 572 |
+
)
|
| 573 |
+
self.model_status['tokenizer'] = 'loaded'
|
| 574 |
+
logger.info("✅ Tokenizer بارگذاری شد")
|
| 575 |
+
else:
|
| 576 |
+
self.model_status['tokenizer'] = 'memory_insufficient'
|
| 577 |
+
except Exception as e:
|
| 578 |
+
logger.error(f"❌ خطا در بارگذاری Tokenizer: {e}")
|
| 579 |
+
self.model_status['tokenizer'] = 'failed'
|
| 580 |
+
|
| 581 |
+
# فاز 2: مدل طبقهبندی متن
|
| 582 |
+
try:
|
| 583 |
+
if MemoryManager.check_memory_available(400):
|
| 584 |
+
logger.info("🏷️ بارگذاری مدل طبقهبندی...")
|
| 585 |
+
if progress_callback:
|
| 586 |
+
progress_callback("بارگذاری مدل طبقهبندی...", 0.5)
|
| 587 |
+
|
| 588 |
+
self.models['classifier'] = pipeline(
|
| 589 |
+
"text-classification",
|
| 590 |
+
model="HooshvareLab/bert-fa-base-uncased-clf-persiannews",
|
| 591 |
+
tokenizer="HooshvareLab/bert-fa-base-uncased-clf-persiannews",
|
| 592 |
+
device=-1,
|
| 593 |
+
return_all_scores=True,
|
| 594 |
+
model_kwargs={"cache_dir": cache_dir}
|
| 595 |
+
)
|
| 596 |
+
self.model_status['classifier'] = 'loaded'
|
| 597 |
+
logger.info("✅ مدل طبقهبندی بارگذاری شد")
|
| 598 |
+
else:
|
| 599 |
+
self.model_status['classifier'] = 'memory_insufficient'
|
| 600 |
+
except Exception as e:
|
| 601 |
+
logger.error(f"❌ خطا در بارگذاری مدل طبقهبندی: {e}")
|
| 602 |
+
self.model_status['classifier'] = 'failed'
|
| 603 |
+
|
| 604 |
+
# فاز 3: مدل تشخیص موجودیت
|
| 605 |
+
try:
|
| 606 |
+
if MemoryManager.check_memory_available(500):
|
| 607 |
+
logger.info("👤 بارگذاری مدل NER...")
|
| 608 |
+
if progress_callback:
|
| 609 |
+
progress_callback("بارگذاری مدل تشخیص موجودیت...", 0.8)
|
| 610 |
+
|
| 611 |
+
self.models['ner'] = pipeline(
|
| 612 |
+
"ner",
|
| 613 |
+
model="HooshvareLab/bert-fa-base-uncased-ner",
|
| 614 |
+
tokenizer="HooshvareLab/bert-fa-base-uncased-ner",
|
| 615 |
+
device=-1,
|
| 616 |
+
aggregation_strategy="simple",
|
| 617 |
+
model_kwargs={"cache_dir": cache_dir}
|
| 618 |
+
)
|
| 619 |
+
self.model_status['ner'] = 'loaded'
|
| 620 |
+
logger.info("✅ مدل NER بارگذاری شد")
|
| 621 |
+
else:
|
| 622 |
+
self.model_status['ner'] = 'memory_insufficient'
|
| 623 |
+
except Exception as e:
|
| 624 |
+
logger.error(f"❌ خطا در بارگذاری مدل NER: {e}")
|
| 625 |
+
self.model_status['ner'] = 'failed'
|
| 626 |
+
|
| 627 |
+
if progress_callback:
|
| 628 |
+
progress_callback("تکمیل بارگذاری مدلها", 1.0)
|
| 629 |
+
|
| 630 |
+
# پاکسازی حافظه
|
| 631 |
+
MemoryManager.cleanup_memory()
|
| 632 |
+
|
| 633 |
+
loaded_count = sum(1 for status in self.model_status.values() if status == 'loaded')
|
| 634 |
+
logger.info(f"🎯 {loaded_count} مدل با موفقیت بارگذاری شد")
|
| 635 |
+
|
| 636 |
+
return self.models
|
| 637 |
+
|
| 638 |
+
def get_model_status(self) -> str:
|
| 639 |
+
"""دریافت وضعیت مدلها"""
|
| 640 |
+
status_lines = []
|
| 641 |
+
status_icons = {
|
| 642 |
+
'loaded': '✅',
|
| 643 |
+
'failed': '❌',
|
| 644 |
+
'memory_insufficient': '⚠️',
|
| 645 |
+
'not_loaded': '⏳'
|
| 646 |
+
}
|
| 647 |
+
|
| 648 |
+
for model_name, status in self.model_status.items():
|
| 649 |
+
icon = status_icons.get(status, '❓')
|
| 650 |
+
status_persian = {
|
| 651 |
+
'loaded': 'بارگذاری شده',
|
| 652 |
+
'failed': 'خطا در بارگذاری',
|
| 653 |
+
'memory_insufficient': 'حافظه ناکافی',
|
| 654 |
+
'not_loaded': 'بارگذاری نشده'
|
| 655 |
+
}.get(status, 'نامشخص')
|
| 656 |
+
|
| 657 |
+
status_lines.append(f"{icon} {model_name}: {status_persian}")
|
| 658 |
+
|
| 659 |
+
memory_usage = MemoryManager.get_memory_usage()
|
| 660 |
+
status_lines.append(f"\n💾 مصرف حافظه: {memory_usage:.1f} MB")
|
| 661 |
+
|
| 662 |
+
return '\n'.join(status_lines)
|
| 663 |
+
|
| 664 |
+
class LegalDocumentScraper:
|
| 665 |
+
"""استخراجکننده پیشرفته اسناد حقوقی"""
|
| 666 |
+
|
| 667 |
+
def __init__(self, model_manager: ModelManager, text_processor: SmartTextProcessor):
|
| 668 |
+
self.model_manager = model_manager
|
| 669 |
+
self.text_processor = text_processor
|
| 670 |
+
self.anti_ddos = AntiDDoSManager()
|
| 671 |
+
self.session = self._create_session()
|
| 672 |
+
|
| 673 |
+
def _create_session(self) -> requests.Session:
|
| 674 |
+
"""ایجاد session با تنظیمات بهینه"""
|
| 675 |
+
session = requests.Session()
|
| 676 |
+
|
| 677 |
+
# تنظیم adapter برای retry
|
| 678 |
+
from requests.adapters import HTTPAdapter
|
| 679 |
+
from urllib3.util.retry import Retry
|
| 680 |
+
|
| 681 |
+
retry_strategy = Retry(
|
| 682 |
+
total=3,
|
| 683 |
+
backoff_factor=1,
|
| 684 |
+
status_forcelist=[429, 500, 502, 503, 504]
|
| 685 |
+
)
|
| 686 |
+
adapter = HTTPAdapter(max_retries=retry_strategy)
|
| 687 |
+
session.mount("http://", adapter)
|
| 688 |
+
session.mount("https://", adapter)
|
| 689 |
+
|
| 690 |
+
return session
|
| 691 |
+
|
| 692 |
+
def scrape_legal_source(self, url: str, progress_callback=None) -> Dict[str, Any]:
|
| 693 |
+
"""استخراج هوشمند سند حقوقی"""
|
| 694 |
+
try:
|
| 695 |
+
domain = urlparse(url).netloc
|
| 696 |
+
|
| 697 |
+
# بررسی مجوز
|
| 698 |
+
if not self.anti_ddos.should_allow_request(domain):
|
| 699 |
+
raise Exception("تعداد درخواستها از حد مجاز تجاوز کرده است")
|
| 700 |
+
|
| 701 |
+
if progress_callback:
|
| 702 |
+
progress_callback(f"🔗 اتصال به {domain}...", 0.1)
|
| 703 |
+
|
| 704 |
+
# اعمال تأخیر ضد DDoS
|
| 705 |
+
delay = self.anti_ddos.get_request_delay(domain)
|
| 706 |
+
time.sleep(delay)
|
| 707 |
+
|
| 708 |
+
# درخواست اصلی
|
| 709 |
+
headers = self.anti_ddos.get_random_headers()
|
| 710 |
+
response = self.session.get(url, headers=headers, timeout=30, allow_redirects=True)
|
| 711 |
+
response.raise_for_status()
|
| 712 |
+
response.encoding = 'utf-8'
|
| 713 |
+
|
| 714 |
+
if progress_callback:
|
| 715 |
+
progress_callback("📄 تجزیه محتوای HTML...", 0.3)
|
| 716 |
+
|
| 717 |
+
# تجزیه HTML
|
| 718 |
+
soup = BeautifulSoup(response.content, 'html.parser')
|
| 719 |
+
self._clean_soup(soup)
|
| 720 |
+
|
| 721 |
+
# شناسایی منبع
|
| 722 |
+
source_info = self._identify_legal_source(url)
|
| 723 |
+
|
| 724 |
+
if progress_callback:
|
| 725 |
+
progress_callback("🎯 استخراج محتوای هدفمند...", 0.5)
|
| 726 |
+
|
| 727 |
+
# استخراج محتوا
|
| 728 |
+
content_fragments = self._extract_content_intelligently(soup, source_info)
|
| 729 |
+
|
| 730 |
+
# بازسازی متن
|
| 731 |
+
reconstructed_text = self.text_processor.reconstruct_legal_text(content_fragments)
|
| 732 |
+
|
| 733 |
+
if not reconstructed_text or len(reconstructed_text.strip()) < 50:
|
| 734 |
+
raise Exception("محتوای کافی استخراج نشد")
|
| 735 |
+
|
| 736 |
+
if progress_callback:
|
| 737 |
+
progress_callback("🔧 تحلیل و پردازش متن...", 0.7)
|
| 738 |
+
|
| 739 |
+
# تحلیلهای پیشرفته
|
| 740 |
+
quality_assessment = self._assess_content_quality(reconstructed_text)
|
| 741 |
+
legal_entities = self.text_processor.extract_legal_entities(reconstructed_text)
|
| 742 |
+
long_sentences = self.text_processor.detect_long_sentences(reconstructed_text)
|
| 743 |
+
ai_analysis = self._apply_ai_analysis(reconstructed_text)
|
| 744 |
+
|
| 745 |
+
# نتیجه نهایی
|
| 746 |
+
result = {
|
| 747 |
+
'url': url,
|
| 748 |
+
'source_info': source_info,
|
| 749 |
+
'title': self._extract_title(soup),
|
| 750 |
+
'content': reconstructed_text,
|
| 751 |
+
'word_count': len(reconstructed_text.split()),
|
| 752 |
+
'character_count': len(reconstructed_text),
|
| 753 |
+
'quality_assessment': quality_assessment,
|
| 754 |
+
'legal_entities': legal_entities,
|
| 755 |
+
'long_sentences': long_sentences,
|
| 756 |
+
'ai_analysis': ai_analysis,
|
| 757 |
+
'extraction_metadata': {
|
| 758 |
+
'method': 'advanced_extraction',
|
| 759 |
+
'fragments_count': len(content_fragments),
|
| 760 |
+
'response_size': len(response.content),
|
| 761 |
+
'encoding': response.encoding,
|
| 762 |
+
'anti_ddos_delay': delay
|
| 763 |
+
},
|
| 764 |
+
'timestamp': datetime.now().isoformat(),
|
| 765 |
+
'status': 'موفق'
|
| 766 |
+
}
|
| 767 |
+
|
| 768 |
+
if progress_callback:
|
| 769 |
+
progress_callback("✅ استخراج با موفقیت تکمیل شد", 1.0)
|
| 770 |
+
|
| 771 |
+
return result
|
| 772 |
+
|
| 773 |
+
except requests.RequestException as e:
|
| 774 |
+
return self._create_error_result(url, f"خطای شبکه: {str(e)}")
|
| 775 |
+
except Exception as e:
|
| 776 |
+
return self._create_error_result(url, f"خطای پردازش: {str(e)}")
|
| 777 |
+
|
| 778 |
+
def _identify_legal_source(self, url: str) -> Dict[str, Any]:
|
| 779 |
+
"""شناسایی منبع حقوقی"""
|
| 780 |
+
domain = urlparse(url).netloc
|
| 781 |
+
|
| 782 |
+
for source_name, config in LEGAL_SOURCES_CONFIG.items():
|
| 783 |
+
if config['base_url'].replace('https://', '').replace('http://', '') in domain:
|
| 784 |
+
return {
|
| 785 |
+
'name': source_name,
|
| 786 |
+
'type': 'official',
|
| 787 |
+
'credibility': 'high',
|
| 788 |
+
'config': config
|
| 789 |
+
}
|
| 790 |
+
|
| 791 |
+
# بررسی الگوهای عمومی سایتهای حقوقی
|
| 792 |
+
legal_indicators = ['law', 'legal', 'court', 'judiciary', 'قانون', 'حقوق', 'دادگاه']
|
| 793 |
+
if any(indicator in domain.lower() for indicator in legal_indicators):
|
| 794 |
+
return {
|
| 795 |
+
'name': 'منبع حقوقی شناخته نشده',
|
| 796 |
+
'type': 'legal_related',
|
| 797 |
+
'credibility': 'medium',
|
| 798 |
+
'config': {'max_depth': 1, 'delay_range': (2, 4)}
|
| 799 |
+
}
|
| 800 |
+
|
| 801 |
+
return {
|
| 802 |
+
'name': 'منبع عمومی',
|
| 803 |
+
'type': 'general',
|
| 804 |
+
'credibility': 'low',
|
| 805 |
+
'config': {'max_depth': 0, 'delay_range': (3, 6)}
|
| 806 |
+
}
|
| 807 |
+
|
| 808 |
+
def _clean_soup(self, soup: BeautifulSoup) -> None:
|
| 809 |
+
"""پاکسازی پیشرفته HTML"""
|
| 810 |
+
# حذف عناصر غیرضروری
|
| 811 |
+
unwanted_tags = [
|
| 812 |
+
'script', 'style', 'nav', 'footer', 'header', 'aside',
|
| 813 |
+
'advertisement', 'ads', 'sidebar', 'menu', 'breadcrumb',
|
| 814 |
+
'social', 'share', 'comment', 'popup', 'modal'
|
| 815 |
+
]
|
| 816 |
+
|
| 817 |
+
for tag in unwanted_tags:
|
| 818 |
+
for element in soup.find_all(tag):
|
| 819 |
+
element.decompose()
|
| 820 |
+
|
| 821 |
+
# حذف عناصر با کلاسهای مشخص
|
| 822 |
+
unwanted_classes = [
|
| 823 |
+
'ad', 'ads', 'advertisement', 'sidebar', 'menu', 'nav',
|
| 824 |
+
'footer', 'header', 'social', 'share', 'comment', 'popup'
|
| 825 |
+
]
|
| 826 |
+
|
| 827 |
+
for class_name in unwanted_classes:
|
| 828 |
+
for element in soup.find_all(class_=lambda x: x and class_name in ' '.join(x).lower()):
|
| 829 |
+
element.decompose()
|
| 830 |
+
|
| 831 |
+
def _extract_content_intelligently(self, soup: BeautifulSoup, source_info: Dict) -> List[str]:
|
| 832 |
+
"""استخراج هوشمند محتوا"""
|
| 833 |
+
fragments = []
|
| 834 |
+
|
| 835 |
+
# استراتژی 1: استفاده از تنظیمات منبع شناخته شده
|
| 836 |
+
if source_info.get('config') and source_info['config'].get('selectors'):
|
| 837 |
+
for selector in source_info['config']['selectors']:
|
| 838 |
+
elements = soup.select(selector)
|
| 839 |
+
for element in elements:
|
| 840 |
+
text = self._extract_clean_text(element)
|
| 841 |
+
if self._is_valid_content(text):
|
| 842 |
+
fragments.append(text)
|
| 843 |
+
|
| 844 |
+
# استراتژی 2: جستجوی محتوای اصلی
|
| 845 |
+
if not fragments:
|
| 846 |
+
main_selectors = [
|
| 847 |
+
'main', 'article', '.main-content', '.content', '.post-content',
|
| 848 |
+
'.article-content', '.news-content', '.law-content', '.legal-text',
|
| 849 |
+
'#main', '#content', '#article', '.document-content'
|
| 850 |
+
]
|
| 851 |
+
|
| 852 |
+
for selector in main_selectors:
|
| 853 |
+
elements = soup.select(selector)
|
| 854 |
+
for element in elements:
|
| 855 |
+
text = self._extract_clean_text(element)
|
| 856 |
+
if self._is_valid_content(text):
|
| 857 |
+
fragments.append(text)
|
| 858 |
+
if fragments:
|
| 859 |
+
break
|
| 860 |
+
|
| 861 |
+
# استراتژی 3: استخراج از پاراگرافها
|
| 862 |
+
if not fragments:
|
| 863 |
+
for tag in ['p', 'div', 'section', 'article']:
|
| 864 |
+
elements = soup.find_all(tag)
|
| 865 |
+
for element in elements:
|
| 866 |
+
text = self._extract_clean_text(element)
|
| 867 |
+
if self._is_valid_content(text, min_length=30):
|
| 868 |
+
fragments.append(text)
|
| 869 |
+
|
| 870 |
+
# استراتژی 4: fallback به body
|
| 871 |
+
if not fragments:
|
| 872 |
+
body = soup.find('body')
|
| 873 |
+
if body:
|
| 874 |
+
text = self._extract_clean_text(body)
|
| 875 |
+
if text:
|
| 876 |
+
paragraphs = [p.strip() for p in text.split('\n') if p.strip()]
|
| 877 |
+
fragments.extend(p for p in paragraphs if self._is_valid_content(p, min_length=20))
|
| 878 |
+
|
| 879 |
+
return fragments
|
| 880 |
+
|
| 881 |
+
def _extract_clean_text(self, element) -> str:
|
| 882 |
+
"""استخراج متن تمیز"""
|
| 883 |
+
if not element:
|
| 884 |
+
return ""
|
| 885 |
+
|
| 886 |
+
# حذف عناصر تودرتو غیرضروری
|
| 887 |
+
for unwanted in element.find_all(['script', 'style', 'nav', 'aside']):
|
| 888 |
+
unwanted.decompose()
|
| 889 |
+
|
| 890 |
+
text = element.get_text(separator=' ', strip=True)
|
| 891 |
+
text = re.sub(r'\s+', ' ', text)
|
| 892 |
+
|
| 893 |
+
return text.strip()
|
| 894 |
+
|
| 895 |
+
def _is_valid_content(self, text: str, min_length: int = 50) -> bool:
|
| 896 |
+
"""بررسی اعتبار محتوا"""
|
| 897 |
+
if not text or len(text) < min_length:
|
| 898 |
+
return False
|
| 899 |
+
|
| 900 |
+
# بررسی نسبت کاراکترهای فارسی
|
| 901 |
+
persian_chars = sum(1 for c in text if '\u0600' <= c <= '\u06FF')
|
| 902 |
+
persian_ratio = persian_chars / len(text) if text else 0
|
| 903 |
+
|
| 904 |
+
if persian_ratio < 0.2:
|
| 905 |
+
return False
|
| 906 |
+
|
| 907 |
+
# بررسی کلمات کلیدی حقوقی
|
| 908 |
+
legal_keywords = ['قانون', 'ماده', 'تبصره', 'مقرره', 'آییننامه', 'دادگاه', 'حکم', 'رای']
|
| 909 |
+
has_legal_content = any(keyword in text for keyword in legal_keywords)
|
| 910 |
+
|
| 911 |
+
return has_legal_content or persian_ratio > 0.5
|
| 912 |
+
|
| 913 |
+
def _extract_title(self, soup: BeautifulSoup) -> str:
|
| 914 |
+
"""استخراج عنوان"""
|
| 915 |
+
title_selectors = [
|
| 916 |
+
'h1', 'h2', 'title', '.page-title', '.article-title',
|
| 917 |
+
'.post-title', '.document-title', '.news-title', '.law-title'
|
| 918 |
+
]
|
| 919 |
+
|
| 920 |
+
for selector in title_selectors:
|
| 921 |
+
element = soup.select_one(selector)
|
| 922 |
+
if element:
|
| 923 |
+
title = element.get_text(strip=True)
|
| 924 |
+
if title and 5 < len(title) < 200:
|
| 925 |
+
return re.sub(r'\s+', ' ', title)
|
| 926 |
+
|
| 927 |
+
return "بدون عنوان"
|
| 928 |
+
|
| 929 |
+
def _assess_content_quality(self, text: str) -> Dict[str, Any]:
|
| 930 |
+
"""ارزیابی کیفیت محتوا"""
|
| 931 |
+
assessment = {
|
| 932 |
+
'overall_score': 0,
|
| 933 |
+
'factors': {},
|
| 934 |
+
'issues': [],
|
| 935 |
+
'strengths': []
|
| 936 |
+
}
|
| 937 |
+
|
| 938 |
+
if not text:
|
| 939 |
+
assessment['issues'].append('متن خالی')
|
| 940 |
+
return assessment
|
| 941 |
+
|
| 942 |
+
factors = {}
|
| 943 |
+
|
| 944 |
+
# طول متن
|
| 945 |
+
word_count = len(text.split())
|
| 946 |
+
if word_count >= 100:
|
| 947 |
+
factors['length'] = min(100, word_count / 10)
|
| 948 |
+
assessment['strengths'].append('طول مناسب متن')
|
| 949 |
+
else:
|
| 950 |
+
factors['length'] = word_count
|
| 951 |
+
assessment['issues'].append('متن کوتاه')
|
| 952 |
+
|
| 953 |
+
# نسبت فارسی
|
| 954 |
+
persian_chars = sum(1 for c in text if '\u0600' <= c <= '\u06FF')
|
| 955 |
+
persian_ratio = persian_chars / len(text) if text else 0
|
| 956 |
+
factors['persian_content'] = persian_ratio * 100
|
| 957 |
+
|
| 958 |
+
if persian_ratio >= 0.7:
|
| 959 |
+
assessment['strengths'].append('محتوای فارسی غنی')
|
| 960 |
+
elif persian_ratio < 0.3:
|
| 961 |
+
assessment['issues'].append('محتوای فارسی کم')
|
| 962 |
+
|
| 963 |
+
# محتوای حقوقی
|
| 964 |
+
legal_terms_count = 0
|
| 965 |
+
for category, terms in PERSIAN_LEGAL_DICTIONARY.items():
|
| 966 |
+
legal_terms_count += sum(1 for term in terms if term in text)
|
| 967 |
+
|
| 968 |
+
factors['legal_content'] = min(100, legal_terms_count * 5)
|
| 969 |
+
|
| 970 |
+
if legal_terms_count >= 10:
|
| 971 |
+
assessment['strengths'].append('غنی از اصطلاحات حقوقی')
|
| 972 |
+
elif legal_terms_count < 3:
|
| 973 |
+
assessment['issues'].append('فقر اصطلاحات حقوقی')
|
| 974 |
+
|
| 975 |
+
# ساختار متن
|
| 976 |
+
structure_score = 0
|
| 977 |
+
if 'ماده' in text:
|
| 978 |
+
structure_score += 20
|
| 979 |
+
if any(indicator in text for indicator in ['تبصره', 'بند', 'فصل']):
|
| 980 |
+
structure_score += 15
|
| 981 |
+
if re.search(r'[۰-۹]+', text):
|
| 982 |
+
structure_score += 10
|
| 983 |
+
|
| 984 |
+
factors['structure'] = structure_score
|
| 985 |
+
|
| 986 |
+
if structure_score >= 30:
|
| 987 |
+
assessment['strengths'].append('ساختار منظم حقوقی')
|
| 988 |
+
|
| 989 |
+
# محاسبه امتیاز کلی
|
| 990 |
+
assessment['factors'] = factors
|
| 991 |
+
assessment['overall_score'] = sum(factors.values()) / len(factors) if factors else 0
|
| 992 |
+
assessment['overall_score'] = min(100, max(0, assessment['overall_score']))
|
| 993 |
+
|
| 994 |
+
return assessment
|
| 995 |
+
|
| 996 |
+
def _apply_ai_analysis(self, text: str) -> Dict[str, Any]:
|
| 997 |
+
"""اعمال تحلیل هوش مصنوعی"""
|
| 998 |
+
analysis = {
|
| 999 |
+
'classification': None,
|
| 1000 |
+
'entities': [],
|
| 1001 |
+
'confidence_scores': {},
|
| 1002 |
+
'model_performance': {}
|
| 1003 |
+
}
|
| 1004 |
+
|
| 1005 |
+
if not text or len(text.split()) < 10:
|
| 1006 |
+
return analysis
|
| 1007 |
+
|
| 1008 |
+
# آمادهسازی متن
|
| 1009 |
+
text_sample = ' '.join(text.split()[:300])
|
| 1010 |
+
|
| 1011 |
+
# طبقهبندی متن
|
| 1012 |
+
if 'classifier' in self.model_manager.models:
|
| 1013 |
+
try:
|
| 1014 |
+
start_time = time.time()
|
| 1015 |
+
classification_result = self.model_manager.models['classifier'](text_sample)
|
| 1016 |
+
|
| 1017 |
+
if classification_result:
|
| 1018 |
+
analysis['classification'] = classification_result[:3]
|
| 1019 |
+
analysis['confidence_scores']['classification'] = classification_result[0]['score']
|
| 1020 |
+
|
| 1021 |
+
analysis['model_performance']['classification_time'] = time.time() - start_time
|
| 1022 |
+
|
| 1023 |
+
except Exception as e:
|
| 1024 |
+
logger.error(f"خطا در طبقهبندی: {e}")
|
| 1025 |
+
|
| 1026 |
+
# تشخیص موجودیت
|
| 1027 |
+
if 'ner' in self.model_manager.models:
|
| 1028 |
+
try:
|
| 1029 |
+
start_time = time.time()
|
| 1030 |
+
ner_result = self.model_manager.models['ner'](text_sample[:1000])
|
| 1031 |
+
|
| 1032 |
+
if ner_result:
|
| 1033 |
+
high_confidence_entities = [
|
| 1034 |
+
entity for entity in ner_result
|
| 1035 |
+
if entity.get('score', 0) > 0.5
|
| 1036 |
+
]
|
| 1037 |
+
analysis['entities'] = high_confidence_entities[:15]
|
| 1038 |
+
|
| 1039 |
+
analysis['model_performance']['ner_time'] = time.time() - start_time
|
| 1040 |
+
|
| 1041 |
+
except Exception as e:
|
| 1042 |
+
logger.error(f"خطا در تشخیص موجودیت: {e}")
|
| 1043 |
+
|
| 1044 |
+
return analysis
|
| 1045 |
+
|
| 1046 |
+
def _create_error_result(self, url: str, error_message: str) -> Dict[str, Any]:
|
| 1047 |
+
"""ایجاد نتیجه خطا"""
|
| 1048 |
+
return {
|
| 1049 |
+
'url': url,
|
| 1050 |
+
'error': error_message,
|
| 1051 |
+
'status': 'ناموفق',
|
| 1052 |
+
'timestamp': datetime.now().isoformat(),
|
| 1053 |
+
'content': '',
|
| 1054 |
+
'word_count': 0,
|
| 1055 |
+
'quality_assessment': {'overall_score': 0, 'issues': [error_message]}
|
| 1056 |
+
}
|
| 1057 |
+
|
| 1058 |
+
class PersianLegalScraperApp:
|
| 1059 |
+
"""اپلیکیشن اصلی با رابط کاربری پیشرفته"""
|
| 1060 |
+
|
| 1061 |
+
def __init__(self):
|
| 1062 |
+
self.text_processor = SmartTextProcessor()
|
| 1063 |
+
self.model_manager = ModelManager()
|
| 1064 |
+
self.scraper = None
|
| 1065 |
+
self.results = []
|
| 1066 |
+
self.processing_stats = {
|
| 1067 |
+
'total_processed': 0,
|
| 1068 |
+
'successful': 0,
|
| 1069 |
+
'failed': 0,
|
| 1070 |
+
'total_words': 0
|
| 1071 |
+
}
|
| 1072 |
+
|
| 1073 |
+
# مقداردهی اولیه
|
| 1074 |
+
self._initialize_system()
|
| 1075 |
+
|
| 1076 |
+
def _initialize_system(self):
|
| 1077 |
+
"""مقداردهی سیستم"""
|
| 1078 |
+
try:
|
| 1079 |
+
logger.info("🚀 مقداردهی سیستم...")
|
| 1080 |
+
self.model_manager.load_models_progressively()
|
| 1081 |
+
self.scraper = LegalDocumentScraper(self.model_manager, self.text_processor)
|
| 1082 |
+
logger.info("✅ سیستم آماده است")
|
| 1083 |
+
except Exception as e:
|
| 1084 |
+
logger.error(f"❌ خطا در مقداردهی: {e}")
|
| 1085 |
+
|
| 1086 |
+
def process_single_url(self, url: str, progress=gr.Progress()) -> Tuple[str, str, str]:
|
| 1087 |
+
"""پردازش یک URL"""
|
| 1088 |
+
if not url or not url.strip():
|
| 1089 |
+
return "❌ خطا: آدرس معتبری وارد کنید", "", ""
|
| 1090 |
+
|
| 1091 |
+
url = url.strip()
|
| 1092 |
+
if not url.startswith(('http://', 'https://')):
|
| 1093 |
+
url = 'https://' + url
|
| 1094 |
+
|
| 1095 |
+
try:
|
| 1096 |
+
progress(0.0, desc="شروع پردازش...")
|
| 1097 |
+
|
| 1098 |
+
def progress_callback(message: str, value: float):
|
| 1099 |
+
progress(value, desc=message)
|
| 1100 |
+
|
| 1101 |
+
result = self.scraper.scrape_legal_source(url, progress_callback)
|
| 1102 |
+
|
| 1103 |
+
if result.get('status') == 'موفق':
|
| 1104 |
+
# بهروزرسانی آمار
|
| 1105 |
+
self.processing_stats['total_processed'] += 1
|
| 1106 |
+
self.processing_stats['successful'] += 1
|
| 1107 |
+
self.processing_stats['total_words'] += result.get('word_count', 0)
|
| 1108 |
+
|
| 1109 |
+
# ذخیره نتیجه
|
| 1110 |
+
self.results.append(result)
|
| 1111 |
+
|
| 1112 |
+
# تنظیم خروجیها
|
| 1113 |
+
status_text = self._format_single_result(result)
|
| 1114 |
+
analysis_text = self._format_analysis_result(result)
|
| 1115 |
+
content_text = result.get('content', '')
|
| 1116 |
+
|
| 1117 |
+
return status_text, content_text, analysis_text
|
| 1118 |
+
else:
|
| 1119 |
+
self.processing_stats['total_processed'] += 1
|
| 1120 |
+
self.processing_stats['failed'] += 1
|
| 1121 |
+
error_msg = result.get('error', 'خطای نامشخص')
|
| 1122 |
+
return f"❌ خطا: {error_msg}", "", ""
|
| 1123 |
+
|
| 1124 |
+
except Exception as e:
|
| 1125 |
+
self.processing_stats['total_processed'] += 1
|
| 1126 |
+
self.processing_stats['failed'] += 1
|
| 1127 |
+
logger.error(f"خطا در پردازش: {e}")
|
| 1128 |
+
return f"❌ خطای سیستمی: {str(e)}", "", ""
|
| 1129 |
+
|
| 1130 |
+
def _format_single_result(self, result: Dict[str, Any]) -> str:
|
| 1131 |
+
"""قالببندی نتیجه با آیکونهای SVG"""
|
| 1132 |
+
quality = result.get('quality_assessment', {})
|
| 1133 |
+
source_info = result.get('source_info', {})
|
| 1134 |
+
legal_entities = result.get('legal_entities', {})
|
| 1135 |
+
long_sentences = result.get('long_sentences', [])
|
| 1136 |
+
|
| 1137 |
+
lines = [
|
| 1138 |
+
f"{SVG_ICONS['success']} **وضعیت**: {result.get('status')}",
|
| 1139 |
+
f"{SVG_ICONS['document']} **عنوان**: {result.get('title', 'بدون عنوان')}",
|
| 1140 |
+
f"🏛️ **منبع**: {source_info.get('name', 'نامشخص')} ({source_info.get('credibility', 'نامشخص')})",
|
| 1141 |
+
f"📊 **آمار محتوا**: {result.get('word_count', 0):,} کلمه، {result.get('character_count', 0):,} کاراکتر",
|
| 1142 |
+
f"🎯 **کیفیت کلی**: {quality.get('overall_score', 0):.1f}/100",
|
| 1143 |
+
f"📈 **محتوای فارسی**: {quality.get('factors', {}).get('persian_content', 0):.1f}%",
|
| 1144 |
+
f"⚖️ **محتوای حقوقی**: {quality.get('factors', {}).get('legal_content', 0):.1f}/100",
|
| 1145 |
+
f"📚 **ارجاعات حقوقی**: {len(legal_entities.get('citations', []))} مورد",
|
| 1146 |
+
f"🏷️ **اصطلاحات حقوقی**: {len(legal_entities.get('legal_terms', []))} مورد"
|
| 1147 |
+
]
|
| 1148 |
+
|
| 1149 |
+
# جملات طولانی
|
| 1150 |
+
if long_sentences:
|
| 1151 |
+
lines.append(f"{SVG_ICONS['warning']} **جملات طولانی**: {len(long_sentences)} مورد")
|
| 1152 |
+
|
| 1153 |
+
# نقاط قوت
|
| 1154 |
+
strengths = quality.get('strengths', [])
|
| 1155 |
+
if strengths:
|
| 1156 |
+
lines.append(f"\n✨ **نقاط قوت**: {' | '.join(strengths)}")
|
| 1157 |
+
|
| 1158 |
+
# مشکلات
|
| 1159 |
+
issues = quality.get('issues', [])
|
| 1160 |
+
if issues:
|
| 1161 |
+
lines.append(f"{SVG_ICONS['warning']} **نکات**: {' | '.join(issues)}")
|
| 1162 |
+
|
| 1163 |
+
lines.append(f"🕐 **زمان**: {result.get('timestamp', '')[:19]}")
|
| 1164 |
+
|
| 1165 |
+
return '\n'.join(lines)
|
| 1166 |
+
|
| 1167 |
+
def _format_analysis_result(self, result: Dict[str, Any]) -> str:
|
| 1168 |
+
"""قالببندی تحلیل هوش مصنوعی"""
|
| 1169 |
+
ai_analysis = result.get('ai_analysis', {})
|
| 1170 |
+
legal_entities = result.get('legal_entities', {})
|
| 1171 |
+
long_sentences = result.get('long_sentences', [])
|
| 1172 |
+
|
| 1173 |
+
lines = [
|
| 1174 |
+
f"{SVG_ICONS['analyze']} **تحلیل هوش مصنوعی**\n",
|
| 1175 |
+
f"📊 **وضعیت مدلها**: {self.model_manager.get_model_status()}\n"
|
| 1176 |
+
]
|
| 1177 |
+
|
| 1178 |
+
# طبقهبندی
|
| 1179 |
+
classification = ai_analysis.get('classification')
|
| 1180 |
+
if classification:
|
| 1181 |
+
lines.append("🏷️ **طبقهبندی محتوا**:")
|
| 1182 |
+
for i, item in enumerate(classification[:3], 1):
|
| 1183 |
+
label = item.get('label', 'نامشخص')
|
| 1184 |
+
score = item.get('score', 0)
|
| 1185 |
+
lines.append(f"{i}. {label}: {score:.1%}")
|
| 1186 |
+
|
| 1187 |
+
# موجودیتهای شناسایی شده
|
| 1188 |
+
entities = ai_analysis.get('entities', [])
|
| 1189 |
+
if entities:
|
| 1190 |
+
lines.append("\n👥 **موجودیتهای شناسایی شده**:")
|
| 1191 |
+
for entity in entities[:8]: # 8 مورد اول
|
| 1192 |
+
word = entity.get('word', '')
|
| 1193 |
+
label = entity.get('entity_group', '')
|
| 1194 |
+
score = entity.get('score', 0)
|
| 1195 |
+
lines.append(f"• {word} ({label}): {score:.1%}")
|
| 1196 |
+
|
| 1197 |
+
# ارجاعات حقوقی
|
| 1198 |
+
citations = legal_entities.get('citations', [])
|
| 1199 |
+
if citations:
|
| 1200 |
+
lines.append(f"\n📚 **ارجاعات حقوقی**: {len(citations)} مورد")
|
| 1201 |
+
unique_citations = list(set(citations))[:10]
|
| 1202 |
+
lines.append(f"نمونه: {', '.join(unique_citations)}")
|
| 1203 |
+
|
| 1204 |
+
# قوانین شناسایی شده
|
| 1205 |
+
laws = legal_entities.get('laws', [])
|
| 1206 |
+
if laws:
|
| 1207 |
+
lines.append(f"\n⚖️ **قوانین شناسایی شده**: {len(laws)} مورد")
|
| 1208 |
+
for law in laws[:3]:
|
| 1209 |
+
lines.append(f"• {law}")
|
| 1210 |
+
|
| 1211 |
+
# جملات طولانی
|
| 1212 |
+
if long_sentences:
|
| 1213 |
+
lines.append(f"\n{SVG_ICONS['warning']} **جملات طولانی شناسایی شده**:")
|
| 1214 |
+
for i, sentence_info in enumerate(long_sentences[:3], 1):
|
| 1215 |
+
word_count = sentence_info.get('word_count', 0)
|
| 1216 |
+
suggestions = sentence_info.get('suggestions', [])
|
| 1217 |
+
lines.append(f"{i}. {word_count} کلمه - {', '.join(suggestions[:2])}")
|
| 1218 |
+
|
| 1219 |
+
# عملکرد مدل
|
| 1220 |
+
performance = ai_analysis.get('model_performance', {})
|
| 1221 |
+
if performance:
|
| 1222 |
+
lines.append(f"\n⏱️ **عملکرد**: ")
|
| 1223 |
+
if 'classification_time' in performance:
|
| 1224 |
+
lines.append(f"طبقهبندی: {performance['classification_time']:.2f}s")
|
| 1225 |
+
if 'ner_time' in performance:
|
| 1226 |
+
lines.append(f"تشخیص موجودیت: {performance['ner_time']:.2f}s")
|
| 1227 |
+
|
| 1228 |
+
return '\n'.join(lines)
|
| 1229 |
+
|
| 1230 |
+
def process_multiple_urls(self, urls_text: str, progress=gr.Progress()) -> Tuple[str, str]:
|
| 1231 |
+
"""پردازش چندین URL"""
|
| 1232 |
+
if not urls_text or not urls_text.strip():
|
| 1233 |
+
return "❌ لطفا لیست آدرسها را وارد کنید", ""
|
| 1234 |
+
|
| 1235 |
+
urls = [url.strip() for url in urls_text.split('\n') if url.strip()]
|
| 1236 |
+
if not urls:
|
| 1237 |
+
return "❌ آدرس معتبری یافت نشد", ""
|
| 1238 |
+
|
| 1239 |
+
# محدودیت تعداد برای HF Spaces
|
| 1240 |
+
if len(urls) > 10:
|
| 1241 |
+
urls = urls[:10]
|
| 1242 |
+
warning_msg = f"{SVG_ICONS['warning']} به دلیل محدودیتها، تنها 10 آدرس اول پردازش میشود.\n\n"
|
| 1243 |
+
else:
|
| 1244 |
+
warning_msg = ""
|
| 1245 |
+
|
| 1246 |
+
results = []
|
| 1247 |
+
total_urls = len(urls)
|
| 1248 |
+
|
| 1249 |
+
try:
|
| 1250 |
+
progress(0.0, desc=f"شروع پردازش {total_urls} آدرس...")
|
| 1251 |
+
|
| 1252 |
+
for i, url in enumerate(urls):
|
| 1253 |
+
if not url.startswith(('http://', 'https://')):
|
| 1254 |
+
url = 'https://' + url
|
| 1255 |
+
|
| 1256 |
+
progress_value = i / total_urls
|
| 1257 |
+
progress(progress_value, desc=f"پردازش {i+1} از {total_urls}: {url[:50]}...")
|
| 1258 |
+
|
| 1259 |
+
def progress_callback(message: str, value: float):
|
| 1260 |
+
overall_progress = progress_value + (value * (1/total_urls))
|
| 1261 |
+
progress(overall_progress, desc=f"{message} ({i+1}/{total_urls})")
|
| 1262 |
+
|
| 1263 |
+
result = self.scraper.scrape_legal_source(url, progress_callback)
|
| 1264 |
+
results.append(result)
|
| 1265 |
+
|
| 1266 |
+
# بهروزرسانی آمار
|
| 1267 |
+
self.processing_stats['total_processed'] += 1
|
| 1268 |
+
if result.get('status') == 'موفق':
|
| 1269 |
+
self.processing_stats['successful'] += 1
|
| 1270 |
+
self.processing_stats['total_words'] += result.get('word_count', 0)
|
| 1271 |
+
else:
|
| 1272 |
+
self.processing_stats['failed'] += 1
|
| 1273 |
+
|
| 1274 |
+
# پاکسازی حافظه هر 3 درخواست
|
| 1275 |
+
if (i + 1) % 3 == 0:
|
| 1276 |
+
MemoryManager.cleanup_memory()
|
| 1277 |
+
time.sleep(1) # کمی استراحت
|
| 1278 |
+
|
| 1279 |
+
progress(1.0, desc="تکمیل پردازش")
|
| 1280 |
+
|
| 1281 |
+
# ذخیره نتایج
|
| 1282 |
+
self.results.extend(results)
|
| 1283 |
+
|
| 1284 |
+
# تنظیم خروجیها
|
| 1285 |
+
summary_text = warning_msg + self._format_batch_summary(results)
|
| 1286 |
+
detailed_results = self._format_batch_details(results)
|
| 1287 |
+
|
| 1288 |
+
return summary_text, detailed_results
|
| 1289 |
+
|
| 1290 |
+
except Exception as e:
|
| 1291 |
+
logger.error(f"خطا در پردازش دستهای: {e}")
|
| 1292 |
+
return f"❌ خطای سیستمی: {str(e)}", ""
|
| 1293 |
+
|
| 1294 |
+
def _format_batch_summary(self, results: List[Dict[str, Any]]) -> str:
|
| 1295 |
+
"""خلاصه پردازش دستهای با آیکونها"""
|
| 1296 |
+
total = len(results)
|
| 1297 |
+
successful = sum(1 for r in results if r.get('status') == 'موفق')
|
| 1298 |
+
failed = total - successful
|
| 1299 |
+
|
| 1300 |
+
if successful > 0:
|
| 1301 |
+
total_words = sum(r.get('word_count', 0) for r in results if r.get('status') == 'موفق')
|
| 1302 |
+
avg_quality = sum(r.get('quality_assessment', {}).get('overall_score', 0)
|
| 1303 |
+
for r in results if r.get('status') == 'موفق') / successful
|
| 1304 |
+
else:
|
| 1305 |
+
total_words = 0
|
| 1306 |
+
avg_quality = 0
|
| 1307 |
+
|
| 1308 |
+
lines = [
|
| 1309 |
+
f"{SVG_ICONS['analyze']} **خلاصه پردازش دستهای**",
|
| 1310 |
+
f"📈 **کل آدرسها**: {total}",
|
| 1311 |
+
f"{SVG_ICONS['success']} **موفق**: {successful} ({successful/total*100:.1f}%)",
|
| 1312 |
+
f"{SVG_ICONS['error']} **ناموفق**: {failed} ({failed/total*100:.1f}%)",
|
| 1313 |
+
f"📝 **کل کلمات**: {total_words:,}",
|
| 1314 |
+
f"🎯 **میانگین کیفیت**: {avg_quality:.1f}/100",
|
| 1315 |
+
f"💾 **حافظه**: {MemoryManager.get_memory_usage():.1f} MB",
|
| 1316 |
+
f"🕐 **زمان**: {datetime.now().strftime('%H:%M:%S')}"
|
| 1317 |
+
]
|
| 1318 |
+
|
| 1319 |
+
return '\n'.join(lines)
|
| 1320 |
+
|
| 1321 |
+
def _format_batch_details(self, results: List[Dict[str, Any]]) -> str:
|
| 1322 |
+
"""جزئیات نتایج دستهای"""
|
| 1323 |
+
lines = []
|
| 1324 |
+
|
| 1325 |
+
for i, result in enumerate(results, 1):
|
| 1326 |
+
url = result.get('url', '')
|
| 1327 |
+
status = result.get('status', '')
|
| 1328 |
+
|
| 1329 |
+
if status == 'موفق':
|
| 1330 |
+
title = result.get('title', 'بدون عنوان')
|
| 1331 |
+
word_count = result.get('word_count', 0)
|
| 1332 |
+
quality = result.get('quality_assessment', {}).get('overall_score', 0)
|
| 1333 |
+
source = result.get('source_info', {}).get('name', 'نامشخص')
|
| 1334 |
+
|
| 1335 |
+
lines.extend([
|
| 1336 |
+
f"\n**{i}. {SVG_ICONS['success']} {title}**",
|
| 1337 |
+
f"{SVG_ICONS['link']} {url[:70]}{'...' if len(url) > 70 else ''}",
|
| 1338 |
+
f"🏛️ منبع: {source}",
|
| 1339 |
+
f"📊 {word_count:,} کلمه | کیفیت: {quality:.1f}/100"
|
| 1340 |
+
])
|
| 1341 |
+
else:
|
| 1342 |
+
error = result.get('error', 'خطای نامشخص')
|
| 1343 |
+
lines.extend([
|
| 1344 |
+
f"\n**{i}. {SVG_ICONS['error']} ناموفق**",
|
| 1345 |
+
f"{SVG_ICONS['link']} {url[:70]}{'...' if len(url) > 70 else ''}",
|
| 1346 |
+
f"❗ {error[:80]}{'...' if len(error) > 80 else ''}"
|
| 1347 |
+
])
|
| 1348 |
+
|
| 1349 |
+
return '\n'.join(lines)
|
| 1350 |
+
|
| 1351 |
+
def export_results(self) -> Tuple[str, Optional[str]]:
|
| 1352 |
+
"""صادرات نتایج"""
|
| 1353 |
+
if not self.results:
|
| 1354 |
+
return f"{SVG_ICONS['error']} نتیجهای برای صادرات وجود ندارد", None
|
| 1355 |
+
|
| 1356 |
+
try:
|
| 1357 |
+
successful_results = [r for r in self.results if r.get('status') == 'موفق']
|
| 1358 |
+
|
| 1359 |
+
if not successful_results:
|
| 1360 |
+
return f"{SVG_ICONS['error']} نتیجه موفقی برای صادرات وجود ندارد", None
|
| 1361 |
+
|
| 1362 |
+
export_data = []
|
| 1363 |
+
for result in successful_results:
|
| 1364 |
+
quality = result.get('quality_assessment', {})
|
| 1365 |
+
source_info = result.get('source_info', {})
|
| 1366 |
+
ai_analysis = result.get('ai_analysis', {})
|
| 1367 |
+
|
| 1368 |
+
# استخراج اطلاعات طبقهبندی
|
| 1369 |
+
classification = ai_analysis.get('classification', [])
|
| 1370 |
+
top_class = classification[0].get('label', '') if classification else ''
|
| 1371 |
+
|
| 1372 |
+
export_data.append({
|
| 1373 |
+
'آدرس': result.get('url', ''),
|
| 1374 |
+
'عنوان': result.get('title', ''),
|
| 1375 |
+
'منبع': source_info.get('name', ''),
|
| 1376 |
+
'اعتبار منبع': source_info.get('credibility', ''),
|
| 1377 |
+
'تعداد کلمات': result.get('word_count', 0),
|
| 1378 |
+
'کیفیت کلی': round(quality.get('overall_score', 0), 1),
|
| 1379 |
+
'محتوای فارسی (%)': round(quality.get('factors', {}).get('persian_content', 0), 1),
|
| 1380 |
+
'محتوای حقوقی': round(quality.get('factors', {}).get('legal_content', 0), 1),
|
| 1381 |
+
'طبقهبندی AI': top_class,
|
| 1382 |
+
'زمان استخراج': result.get('timestamp', ''),
|
| 1383 |
+
'محتوا': result.get('content', '')[:2000] + '...' if len(result.get('content', '')) > 2000 else result.get('content', '')
|
| 1384 |
+
})
|
| 1385 |
+
|
| 1386 |
+
df = pd.DataFrame(export_data)
|
| 1387 |
+
|
| 1388 |
+
# ذخیره فایل
|
| 1389 |
+
timestamp = datetime.now().strftime('%Y%m%d_%H%M%S')
|
| 1390 |
+
csv_path = f"/tmp/legal_scraping_results_{timestamp}.csv"
|
| 1391 |
+
df.to_csv(csv_path, index=False, encoding='utf-8-sig')
|
| 1392 |
+
|
| 1393 |
+
summary = f"{SVG_ICONS['export']} {len(export_data)} سند با موفقیت صادر شد"
|
| 1394 |
+
return summary, csv_path
|
| 1395 |
+
|
| 1396 |
+
except Exception as e:
|
| 1397 |
+
logger.error(f"خطا در صادرات: {e}")
|
| 1398 |
+
return f"{SVG_ICONS['error']} خطا در صادرات: {str(e)}", None
|
| 1399 |
+
|
| 1400 |
+
def get_system_status(self) -> str:
|
| 1401 |
+
"""وضعیت سیستم"""
|
| 1402 |
+
model_status = self.model_manager.get_model_status()
|
| 1403 |
+
memory_usage = MemoryManager.get_memory_usage()
|
| 1404 |
+
|
| 1405 |
+
lines = [
|
| 1406 |
+
f"{SVG_ICONS['settings']} **وضعیت سیستم**\n",
|
| 1407 |
+
f"💾 **حافظه**: {memory_usage:.1f} MB",
|
| 1408 |
+
f"📊 **آمار کلی**:",
|
| 1409 |
+
f" • کل پردازش شده: {self.processing_stats['total_processed']}",
|
| 1410 |
+
f" • موفق: {self.processing_stats['successful']}",
|
| 1411 |
+
f" • ناموفق: {self.processing_stats['failed']}",
|
| 1412 |
+
f" • کل کلمات: {self.processing_stats['total_words']:,}",
|
| 1413 |
+
f"\n🤖 **مدلها**:",
|
| 1414 |
+
model_status,
|
| 1415 |
+
f"\n⏰ **آخرین بروزرسانی**: {datetime.now().strftime('%Y/%m/%d %H:%M:%S')}"
|
| 1416 |
+
]
|
| 1417 |
+
|
| 1418 |
+
return '\n'.join(lines)
|
| 1419 |
+
|
| 1420 |
+
def clear_results(self) -> str:
|
| 1421 |
+
"""پاکسازی نتایج و حافظه"""
|
| 1422 |
+
self.results.clear()
|
| 1423 |
+
self.processing_stats = {
|
| 1424 |
+
'total_processed': 0,
|
| 1425 |
+
'successful': 0,
|
| 1426 |
+
'failed': 0,
|
| 1427 |
+
'total_words': 0
|
| 1428 |
+
}
|
| 1429 |
+
MemoryManager.cleanup_memory()
|
| 1430 |
+
return f"{SVG_ICONS['success']} نتایج و حافظه پاکسازی شد"
|
| 1431 |
+
|
| 1432 |
+
def create_interface(self):
|
| 1433 |
+
"""ایجاد رابط کاربری Gradio"""
|
| 1434 |
+
|
| 1435 |
+
# CSS سفارشی برای RTL و فونت فارسی
|
| 1436 |
+
custom_css = """
|
| 1437 |
+
.rtl {
|
| 1438 |
+
direction: rtl;
|
| 1439 |
+
text-align: right;
|
| 1440 |
+
font-family: 'Vazirmatn', 'Tahoma', sans-serif;
|
| 1441 |
+
}
|
| 1442 |
+
|
| 1443 |
+
.persian-title {
|
| 1444 |
+
background: linear-gradient(135deg, #667eea 0%, #764ba2 100%);
|
| 1445 |
+
color: white;
|
| 1446 |
+
padding: 20px;
|
| 1447 |
+
border-radius: 10px;
|
| 1448 |
+
text-align: center;
|
| 1449 |
+
margin-bottom: 20px;
|
| 1450 |
+
font-family: 'Vazirmatn', 'Tahoma', sans-serif;
|
| 1451 |
+
}
|
| 1452 |
+
|
| 1453 |
+
.status-box {
|
| 1454 |
+
border: 1px solid #ddd;
|
| 1455 |
+
border-radius: 8px;
|
| 1456 |
+
padding: 15px;
|
| 1457 |
+
background-color: #f9f9f9;
|
| 1458 |
+
direction: rtl;
|
| 1459 |
+
font-family: 'Vazirmatn', 'Tahoma', sans-serif;
|
| 1460 |
+
}
|
| 1461 |
+
|
| 1462 |
+
.gradio-container {
|
| 1463 |
+
font-family: 'Vazirmatn', 'Tahoma', sans-serif !important;
|
| 1464 |
+
}
|
| 1465 |
+
"""
|
| 1466 |
+
|
| 1467 |
+
with gr.Blocks(
|
| 1468 |
+
title="سیستم پیشرفته استخراج و تحلیل اسناد حقوقی فارسی",
|
| 1469 |
+
css=custom_css,
|
| 1470 |
+
theme=gr.themes.Soft()
|
| 1471 |
+
) as interface:
|
| 1472 |
+
|
| 1473 |
+
# عنوان اصلی
|
| 1474 |
+
gr.HTML(f"""
|
| 1475 |
+
<div class="persian-title">
|
| 1476 |
+
<h1>{SVG_ICONS['document']} سیستم پیشرفته استخراج و تحلیل اسناد حقوقی فارسی</h1>
|
| 1477 |
+
<p>{SVG_ICONS['analyze']} مجهز به مدلهای BERT فارسی | 📊 تحلیل هوشمند محتوا | ⚡ بهینهسازی شده برای Hugging Face Spaces</p>
|
| 1478 |
+
<p>🎯 منابع معتبر: مجلس، قوه قضاییه، وزارت دادگستری، دیوان عدالت اداری</p>
|
| 1479 |
+
</div>
|
| 1480 |
+
""")
|
| 1481 |
+
|
| 1482 |
+
with gr.Tabs():
|
| 1483 |
+
|
| 1484 |
+
# تب پردازش تک URL
|
| 1485 |
+
with gr.Tab(f"{SVG_ICONS['search']} پردازش تک آدرس"):
|
| 1486 |
+
with gr.Row():
|
| 1487 |
+
with gr.Column(scale=2):
|
| 1488 |
+
single_url = gr.Textbox(
|
| 1489 |
+
label=f"{SVG_ICONS['link']} آدرس سند حقوقی",
|
| 1490 |
+
placeholder="https://rc.majlis.ir/fa/law/show/12345",
|
| 1491 |
+
lines=2,
|
| 1492 |
+
elem_classes=["rtl"]
|
| 1493 |
+
)
|
| 1494 |
+
|
| 1495 |
+
with gr.Row():
|
| 1496 |
+
single_btn = gr.Button(
|
| 1497 |
+
f"{SVG_ICONS['analyze']} شروع استخراج و تحلیل",
|
| 1498 |
+
variant="primary",
|
| 1499 |
+
size="lg"
|
| 1500 |
+
)
|
| 1501 |
+
clear_single_btn = gr.Button(
|
| 1502 |
+
"🧹 پاک کردن",
|
| 1503 |
+
variant="secondary"
|
| 1504 |
+
)
|
| 1505 |
+
|
| 1506 |
+
with gr.Column(scale=1):
|
| 1507 |
+
system_status = gr.Textbox(
|
| 1508 |
+
label=f"{SVG_ICONS['settings']} وضعیت سیستم",
|
| 1509 |
+
interactive=False,
|
| 1510 |
+
lines=12,
|
| 1511 |
+
elem_classes=["rtl", "status-box"]
|
| 1512 |
+
)
|
| 1513 |
+
|
| 1514 |
+
with gr.Row():
|
| 1515 |
+
with gr.Column():
|
| 1516 |
+
single_status = gr.Textbox(
|
| 1517 |
+
label=f"{SVG_ICONS['analyze']} خلاصه نتایج",
|
| 1518 |
+
interactive=False,
|
| 1519 |
+
lines=12,
|
| 1520 |
+
elem_classes=["rtl"]
|
| 1521 |
+
)
|
| 1522 |
+
|
| 1523 |
+
with gr.Column():
|
| 1524 |
+
single_analysis = gr.Textbox(
|
| 1525 |
+
label=f"{SVG_ICONS['analyze']} تحلیل هوش مصنوعی",
|
| 1526 |
+
interactive=False,
|
| 1527 |
+
lines=12,
|
| 1528 |
+
elem_classes=["rtl"]
|
| 1529 |
+
)
|
| 1530 |
+
|
| 1531 |
+
single_content = gr.Textbox(
|
| 1532 |
+
label=f"{SVG_ICONS['document']} محتوای استخراج شده",
|
| 1533 |
+
interactive=False,
|
| 1534 |
+
lines=15,
|
| 1535 |
+
elem_classes=["rtl"]
|
| 1536 |
+
)
|
| 1537 |
+
|
| 1538 |
+
# تب پردازش چندتایی
|
| 1539 |
+
with gr.Tab(f"{SVG_ICONS['document']} پردازش دستهای"):
|
| 1540 |
+
gr.Markdown("""
|
| 1541 |
+
### 📝 راهنمای استفاده:
|
| 1542 |
+
- هر آدرس را در خط جداگانهای قرار دهید
|
| 1543 |
+
- حداکثر 10 آدرس به دلیل محدودیتهای سیستم
|
| 1544 |
+
- از منابع معتبر حقوقی استفاده کنید
|
| 1545 |
+
""", elem_classes=["rtl"])
|
| 1546 |
+
|
| 1547 |
+
multi_urls = gr.Textbox(
|
| 1548 |
+
label=f"{SVG_ICONS['document']} فهرست آدرسها (هر آدرس در خط جداگانه)",
|
| 1549 |
+
placeholder="""https://rc.majlis.ir/fa/law/show/12345
|
| 1550 |
+
https://www.judiciary.ir/fa/news/67890
|
| 1551 |
+
https://www.dotic.ir/portal/law/54321""",
|
| 1552 |
+
lines=8,
|
| 1553 |
+
elem_classes=["rtl"]
|
| 1554 |
+
)
|
| 1555 |
+
|
| 1556 |
+
with gr.Row():
|
| 1557 |
+
multi_btn = gr.Button(
|
| 1558 |
+
f"{SVG_ICONS['analyze']} شروع پردازش دستهای",
|
| 1559 |
+
variant="primary",
|
| 1560 |
+
size="lg"
|
| 1561 |
+
)
|
| 1562 |
+
clear_multi_btn = gr.Button(
|
| 1563 |
+
"🧹 پاک کردن",
|
| 1564 |
+
variant="secondary"
|
| 1565 |
+
)
|
| 1566 |
+
|
| 1567 |
+
with gr.Row():
|
| 1568 |
+
with gr.Column():
|
| 1569 |
+
batch_summary = gr.Textbox(
|
| 1570 |
+
label=f"{SVG_ICONS['analyze']} خلاصه پردازش",
|
| 1571 |
+
interactive=False,
|
| 1572 |
+
lines=10,
|
| 1573 |
+
elem_classes=["rtl"]
|
| 1574 |
+
)
|
| 1575 |
+
|
| 1576 |
+
with gr.Column():
|
| 1577 |
+
batch_details = gr.Textbox(
|
| 1578 |
+
label=f"{SVG_ICONS['document']} جزئیات نتایج",
|
| 1579 |
+
interactive=False,
|
| 1580 |
+
lines=10,
|
| 1581 |
+
elem_classes=["rtl"]
|
| 1582 |
+
)
|
| 1583 |
+
|
| 1584 |
+
# تب صادرات و مدیریت
|
| 1585 |
+
with gr.Tab(f"{SVG_ICONS['export']} مدیریت و صادرات"):
|
| 1586 |
+
with gr.Row():
|
| 1587 |
+
with gr.Column():
|
| 1588 |
+
gr.Markdown(f"### {SVG_ICONS['export']} صادرات نتایج", elem_classes=["rtl"])
|
| 1589 |
+
|
| 1590 |
+
export_btn = gr.Button(
|
| 1591 |
+
f"{SVG_ICONS['export']} صادرات به CSV",
|
| 1592 |
+
variant="primary"
|
| 1593 |
+
)
|
| 1594 |
+
|
| 1595 |
+
export_status = gr.Textbox(
|
| 1596 |
+
label="وضعیت صادرات",
|
| 1597 |
+
interactive=False,
|
| 1598 |
+
lines=3,
|
| 1599 |
+
elem_classes=["rtl"]
|
| 1600 |
+
)
|
| 1601 |
+
|
| 1602 |
+
export_file = gr.File(
|
| 1603 |
+
label="📁 فایل صادر شده",
|
| 1604 |
+
interactive=False
|
| 1605 |
+
)
|
| 1606 |
+
|
| 1607 |
+
with gr.Column():
|
| 1608 |
+
gr.Markdown(f"### {SVG_ICONS['settings']} مدیریت سیستم", elem_classes=["rtl"])
|
| 1609 |
+
|
| 1610 |
+
with gr.Row():
|
| 1611 |
+
refresh_btn = gr.Button(
|
| 1612 |
+
"🔄 بروزرسانی وضعیت",
|
| 1613 |
+
variant="secondary"
|
| 1614 |
+
)
|
| 1615 |
+
cleanup_btn = gr.Button(
|
| 1616 |
+
"🧹 پاکسازی حافظه",
|
| 1617 |
+
variant="secondary"
|
| 1618 |
+
)
|
| 1619 |
+
|
| 1620 |
+
clear_results_btn = gr.Button(
|
| 1621 |
+
"🗑️ پاک کردن تمام نتایج",
|
| 1622 |
+
variant="stop"
|
| 1623 |
+
)
|
| 1624 |
+
|
| 1625 |
+
management_status = gr.Textbox(
|
| 1626 |
+
label="وضعیت عملیات",
|
| 1627 |
+
interactive=False,
|
| 1628 |
+
lines=5,
|
| 1629 |
+
elem_classes=["rtl"]
|
| 1630 |
+
)
|
| 1631 |
+
|
| 1632 |
+
# اتصال event handlerها
|
| 1633 |
+
|
| 1634 |
+
# تک URL
|
| 1635 |
+
single_btn.click(
|
| 1636 |
+
fn=self.process_single_url,
|
| 1637 |
+
inputs=[single_url],
|
| 1638 |
+
outputs=[single_status, single_content, single_analysis],
|
| 1639 |
+
show_progress=True
|
| 1640 |
+
)
|
| 1641 |
+
|
| 1642 |
+
clear_single_btn.click(
|
| 1643 |
+
lambda: ("", "", "", ""),
|
| 1644 |
+
outputs=[single_url, single_status, single_content, single_analysis]
|
| 1645 |
+
)
|
| 1646 |
+
|
| 1647 |
+
# چندتایی
|
| 1648 |
+
multi_btn.click(
|
| 1649 |
+
fn=self.process_multiple_urls,
|
| 1650 |
+
inputs=[multi_urls],
|
| 1651 |
+
outputs=[batch_summary, batch_details],
|
| 1652 |
+
show_progress=True
|
| 1653 |
+
)
|
| 1654 |
+
|
| 1655 |
+
clear_multi_btn.click(
|
| 1656 |
+
lambda: ("", "", ""),
|
| 1657 |
+
outputs=[multi_urls, batch_summary, batch_details]
|
| 1658 |
+
)
|
| 1659 |
+
|
| 1660 |
+
# صادرات
|
| 1661 |
+
export_btn.click(
|
| 1662 |
+
fn=self.export_results,
|
| 1663 |
+
outputs=[export_status, export_file]
|
| 1664 |
+
)
|
| 1665 |
+
|
| 1666 |
+
# مدیریت
|
| 1667 |
+
refresh_btn.click(
|
| 1668 |
+
fn=self.get_system_status,
|
| 1669 |
+
outputs=[system_status]
|
| 1670 |
+
)
|
| 1671 |
+
|
| 1672 |
+
cleanup_btn.click(
|
| 1673 |
+
fn=MemoryManager.cleanup_memory,
|
| 1674 |
+
outputs=[management_status]
|
| 1675 |
+
).then(
|
| 1676 |
+
lambda: "✅ حافظه پاکسازی شد",
|
| 1677 |
+
outputs=[management_status]
|
| 1678 |
+
)
|
| 1679 |
+
|
| 1680 |
+
clear_results_btn.click(
|
| 1681 |
+
fn=self.clear_results,
|
| 1682 |
+
outputs=[management_status]
|
| 1683 |
+
)
|
| 1684 |
+
|
| 1685 |
+
# بارگذاری اولیه وضعیت سیستم
|
| 1686 |
+
interface.load(
|
| 1687 |
+
fn=self.get_system_status,
|
| 1688 |
+
outputs=[system_status]
|
| 1689 |
+
)
|
| 1690 |
+
|
| 1691 |
+
return interface
|
| 1692 |
+
|
| 1693 |
+
def main():
|
| 1694 |
+
"""تابع اصلی برای اجرای برنامه"""
|
| 1695 |
+
logger.info("🚀 راهاندازی سیستم استخراج اسناد حقوقی فارسی...")
|
| 1696 |
+
|
| 1697 |
+
try:
|
| 1698 |
+
# ایجاد نمونه برنامه
|
| 1699 |
+
app = PersianLegalScraperApp()
|
| 1700 |
+
|
| 1701 |
+
# ایجاد و راهاندازی رابط
|
| 1702 |
+
interface = app.create_interface()
|
| 1703 |
+
|
| 1704 |
+
# راهاندازی با پیکربندی Hugging Face Spaces
|
| 1705 |
+
interface.launch(
|
| 1706 |
+
server_name="0.0.0.0",
|
| 1707 |
+
server_port=7860,
|
| 1708 |
+
share=False,
|
| 1709 |
+
show_error=True,
|
| 1710 |
+
show_tips=True,
|
| 1711 |
+
enable_queue=True,
|
| 1712 |
+
max_threads=2
|
| 1713 |
+
)
|
| 1714 |
+
|
| 1715 |
+
except Exception as e:
|
| 1716 |
+
logger.error(f"خطا در راهاندازی برنامه: {e}")
|
| 1717 |
+
raise
|
| 1718 |
+
|
| 1719 |
+
if __name__ == "__main__":
|
| 1720 |
+
main()
|