From 162f9f37b13dc812b328169d03361e87de960cd3 Mon Sep 17 00:00:00 2001
From: Nacim <nacim@biqoz.com>
Date: Mon, 23 Jun 2025 15:17:29 +0200
Subject: [PATCH] Update app.py

---
 app.py | 31 +++++++++++++++++--------------
 1 file changed, 17 insertions(+), 14 deletions(-)

diff --git a/app.py b/app.py
index c6523ff..d7b45da 100644
--- a/app.py
+++ b/app.py
@@ -3,13 +3,15 @@ import logging
 import yaml
 from flask import Flask, request, jsonify, make_response
 
+# Import des classes nécessaires de Presidio
 from presidio_analyzer import AnalyzerEngine, RecognizerRegistry, PatternRecognizer, Pattern
 from presidio_analyzer.nlp_engine import NlpEngineProvider
-from presidio_analyzer.deny_list_recognizer import DenyListRecognizer
 
+# Configuration du logging
 logging.basicConfig(level=logging.INFO, format='%(asctime)s - %(name)s - %(levelname)s - %(message)s')
 logger = logging.getLogger(__name__)
 
+# --- CHARGEMENT MANUEL ET EXPLICITE DE LA CONFIGURATION ---
 CONFIG_FILE_PATH = os.environ.get("PRESIDIO_ANALYZER_CONFIG_FILE", "conf/default.yaml")
 logger.info(f"Loading configuration from: {CONFIG_FILE_PATH}")
 
@@ -18,31 +20,31 @@ try:
     with open(CONFIG_FILE_PATH, 'r', encoding='utf-8') as f:
         config = yaml.safe_load(f)
     logger.info("Configuration file loaded successfully.")
-except FileNotFoundError:
-    logger.error(f"Configuration file not found at {CONFIG_FILE_PATH}. Using default empty config.")
-except yaml.YAMLError as e:
-    logger.error(f"Error parsing YAML configuration file: {e}. Using default empty config.")
+except Exception as e:
+    logger.exception(f"Could not load or parse configuration file at {CONFIG_FILE_PATH}")
+    # En cas d'échec, on continue avec une config vide pour ne pas planter, mais le service sera limité.
+    config = {}
 
+# On récupère les langues supportées depuis la config pour les utiliser partout
 supported_languages_from_config = config.get("supported_languages", ["en"])
 logger.info(f"Languages supported according to config: {supported_languages_from_config}")
 
+# Création du fournisseur de moteur NLP
 logger.info("Creating NLP engine provider...")
 nlp_engine_provider = NlpEngineProvider(nlp_configuration=config.get("nlp_engine_configuration"))
 nlp_engine = nlp_engine_provider.create_engine()
 logger.info(f"NLP engine created with models for: {nlp_engine.get_supported_languages()}")
 
-# --- CORRECTION ICI ---
-# On s'assure que le registry est initialisé avec les mêmes langues que le reste
+# Création du registre de recognizers
 logger.info("Creating and populating recognizer registry...")
 registry = RecognizerRegistry()
-registry.load_predefined_recognizers(languages=supported_languages_from_config) # On utilise la variable
-logger.info(f"Recognizer registry loaded for languages: {supported_languages_from_config}")
+# On initialise le registre avec TOUTES les langues supportées
+registry.load_predefined_recognizers(languages=supported_languages_from_config)
 
-# Ajout des recognizers personnalisés
+# Ajout des recognizers personnalisés définis dans le YAML
 custom_recognizers_conf = config.get("recognizers", [])
 for recognizer_conf in custom_recognizers_conf:
     patterns = [Pattern(name=p['name'], regex=p['regex'], score=p['score']) for p in recognizer_conf['patterns']]
-    
     custom_recognizer = PatternRecognizer(
         supported_entity=recognizer_conf['entity_name'],
         name=recognizer_conf['name'],
@@ -50,16 +52,16 @@ for recognizer_conf in custom_recognizers_conf:
         patterns=patterns,
         context=recognizer_conf.get('context')
     )
-    
     registry.add_recognizer(custom_recognizer)
     logger.info(f"Loaded custom recognizer: {custom_recognizer.name}")
 
-# Préparation de l'allow_list
+# Préparation de l'allow_list (simple liste de mots)
 allow_list_config = config.get("allow_list", [])
 allow_list_terms = [item if isinstance(item, str) else item.get('text') for item in allow_list_config if item]
 if allow_list_terms:
     logger.info(f"Prepared {len(allow_list_terms)} terms for the allow list.")
 
+# Initialisation de l'application Flask
 app = Flask(__name__)
 
 # Initialisation du moteur Presidio Analyzer
@@ -67,7 +69,7 @@ logger.info("Initializing AnalyzerEngine with custom configuration...")
 analyzer = AnalyzerEngine(
     nlp_engine=nlp_engine,
     registry=registry,
-    supported_languages=supported_languages_from_config, # On utilise la variable ici aussi
+    supported_languages=supported_languages_from_config, # On s'assure de la cohérence ici aussi
     default_score_threshold=config.get("ner_model_configuration", {}).get("confidence_threshold", {}).get("default", 0.35)
 )
 logger.info("AnalyzerEngine initialized successfully.")
@@ -82,6 +84,7 @@ def analyze_text():
         if not text_to_analyze:
              return jsonify({"error": "text field is missing or empty"}), 400
         
+        # On passe directement la liste de mots à ignorer au paramètre 'allow_list'
         results = analyzer.analyze(
             text=text_to_analyze,
             language=language,