884a8eed96
- Corrections mineures dans les pipelines - Optimisations de l'API complète - Améliorations de la documentation - Finalisation du système
66 lines
1.8 KiB
Python
66 lines
1.8 KiB
Python
"""
|
|
Pipeline d'extraction d'entités
|
|
"""
|
|
|
|
import os
|
|
import logging
|
|
import re
|
|
from typing import Dict, Any, List
|
|
|
|
logger = logging.getLogger(__name__)
|
|
|
|
def run(doc_id: str, ctx: Dict[str, Any]) -> None:
|
|
"""Pipeline d'extraction d'entités"""
|
|
logger.info(f"🔍 Extraction d'entités pour le document {doc_id}")
|
|
|
|
try:
|
|
ocr_text = ctx.get("ocr_text", "")
|
|
document_type = ctx.get("document_type", "autre")
|
|
|
|
# Extraction basique
|
|
entities = _extract_basic_entities(ocr_text, document_type)
|
|
|
|
ctx.update({
|
|
"extracted_entities": entities,
|
|
"entities_count": len(entities)
|
|
})
|
|
logger.info(f"✅ Extraction terminée pour {doc_id}: {len(entities)} entités")
|
|
except Exception as e:
|
|
logger.error(f"❌ Erreur extraction {doc_id}: {e}")
|
|
ctx["extraction_error"] = str(e)
|
|
|
|
def _extract_basic_entities(text: str, doc_type: str) -> List[Dict[str, Any]]:
|
|
"""Extraction basique d'entités"""
|
|
entities = []
|
|
|
|
# Emails
|
|
emails = re.findall(r'\b[A-Za-z0-9._%+-]+@[A-Za-z0-9.-]+\.[A-Z|a-z]{2,}\b', text)
|
|
for email in emails:
|
|
entities.append({
|
|
"type": "contact",
|
|
"subtype": "email",
|
|
"value": email,
|
|
"confidence": 0.95
|
|
})
|
|
|
|
# Téléphones
|
|
phones = re.findall(r'\b0[1-9](?:[.\-\s]?\d{2}){4}\b', text)
|
|
for phone in phones:
|
|
entities.append({
|
|
"type": "contact",
|
|
"subtype": "phone",
|
|
"value": phone,
|
|
"confidence": 0.9
|
|
})
|
|
|
|
# Dates
|
|
dates = re.findall(r'\b\d{1,2}[\/\-\.]\d{1,2}[\/\-\.]\d{4}\b', text)
|
|
for date in dates:
|
|
entities.append({
|
|
"type": "date",
|
|
"subtype": "generic",
|
|
"value": date,
|
|
"confidence": 0.8
|
|
})
|
|
|
|
return entities |