nacim/Presidio
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Actualiser app.py

Browse Source
This commit is contained in:
nacim
2025-07-28 18:48:44 +00:00
parent 9fcc209efb
commit 5e6ba39ddb
1 changed files with 20 additions and 24 deletions
Download Patch File Download Diff File Expand all files Collapse all files

44
app.py
View File

@@ -11,24 +11,28 @@ logger = logging.getLogger(__name__)
app = Flask(__name__) app = Flask(__name__)
# Initialisation Presidio Analyzer via Provider # Chargement du moteur Presidio via Provider
analyzer = None analyzer = None
try: try:
logger.info("--- Presidio Analyzer Service Starting ---") logger.info("--- Presidio Analyzer Service Starting ---")
CONFIG_FILE_PATH = os.environ.get("PRESIDIO_ANALYZER_CONFIG_FILE", "conf/default.yaml") CONFIG_FILE_PATH = os.environ.get("PRESIDIO_ANALYZER_CONFIG_FILE", "conf/default.yaml")
provider = AnalyzerEngineProvider(analyzer_engine_conf_file=CONFIG_FILE_PATH) provider = AnalyzerEngineProvider(analyzer_engine_conf_file=CONFIG_FILE_PATH)
analyzer = provider.create_engine() analyzer = provider.create_engine()
logger.info(f"Analyzer ready. Supported languages: {analyzer.supported_languages}") logger.info(f"Analyzer ready. Languages: {analyzer.supported_languages}")
except Exception as e: except Exception as e:
logger.exception("Error during AnalyzerEngine initialization.") logger.exception("Error during AnalyzerEngine initialization.")
analyzer = None analyzer = None
# Regex strict pour IBAN belge format attendu
IBAN_REGEX = re.compile(r"\b[A-Z]{2}[0-9]{2}(?:\s[0-9]{4}){3}\b", re.IGNORECASE)
# Regex pour recadrage strict # Regex IPv4
IBAN_REGEX = re.compile(r"\b[A-Z]{2}[0-9]{2}(?:\s[A-Z0-9]{4}){4,7}\b", re.IGNORECASE) IPV4_REGEX = re.compile(
IPV4_REGEX = re.compile(r"\b(?:25[0-5]|2[0-4]\d|1\d{2}|[1-9]?\d)(?:\.(?:25[0-5]|2[0-4]\d|1\d{2}|[1-9]?\d)){3}\b") r"\b(?:(?:25[0-5]|2[0-4][0-9]|1\d{2}|[1-9]?\d)\.){3}"
r"(?:25[0-5]|2[0-4][0-9]|1\d{2}|[1-9]?\d)\b"
)
# Liste des labels/phrases à exclure de l'anonymisation (en minuscules) # Liste des labels/phrases à exclure d’anonymisation (en minuscules)
IGNORE_LABELS = { IGNORE_LABELS = {
"témoins", "témoins",
"témoins clés", "témoins clés",
@@ -66,12 +70,11 @@ def analyze_text():
ent_text = text_to_analyze[res.start:res.end].strip() ent_text = text_to_analyze[res.start:res.end].strip()
ent_text_norm = normalize_label(ent_text) ent_text_norm = normalize_label(ent_text)
# Skip anonymization for labels to keep
if ent_text_norm in IGNORE_LABELS: if ent_text_norm in IGNORE_LABELS:
logger.debug(f"Skipping anonymization of label: '{ent_text}'") logger.debug(f"Skipping anonymization of label: '{ent_text}'")
continue continue
# Recadrage stricte IBAN # Recadrage IBAN strict
if res.entity_type == "IBAN": if res.entity_type == "IBAN":
match = IBAN_REGEX.search(ent_text) match = IBAN_REGEX.search(ent_text)
if match: if match:
@@ -81,16 +84,14 @@ def analyze_text():
old_start, old_end = res.start, res.end old_start, old_end = res.start, res.end
res.start += start_offset res.start += start_offset
res.end = res.start + len(true_iban) res.end = res.start + len(true_iban)
ent_text = true_iban logger.debug(f"Adjusted IBAN span: {old_start}-{old_end} => {res.start}-{res.end}")
logger.debug(f"Adjusted IBAN span from ({old_start}-{old_end}) to ({res.start}-{res.end}): '{ent_text}'")
else: else:
logger.warning(f"Cannot find exact IBAN substring inside entity: '{ent_text}'") logger.warning(f"IBAN regex match but cannot find substring position: '{ent_text}'")
else: else:
# Pas un IBAN valide, ignorer cette entité logger.warning(f"Invalid IBAN detected, skipping: '{ent_text}'")
logger.warning(f"Entity IBAN does not match strict IBAN regex: '{ent_text}'")
continue continue
# Recadrage stricte IP_ADDRESS # Recadrage IP_ADDRESS strict IPv4 (wildcard possible pour IPv6 si besoin)
if res.entity_type == "IP_ADDRESS": if res.entity_type == "IP_ADDRESS":
match = IPV4_REGEX.search(ent_text) match = IPV4_REGEX.search(ent_text)
if match: if match:
@@ -100,27 +101,22 @@ def analyze_text():
old_start, old_end = res.start, res.end old_start, old_end = res.start, res.end
res.start += start_offset res.start += start_offset
res.end = res.start + len(true_ip) res.end = res.start + len(true_ip)
ent_text = true_ip logger.debug(f"Adjusted IP span: {old_start}-{old_end} => {res.start}-{res.end}")
logger.debug(f"Adjusted IP_ADDRESS span from ({old_start}-{old_end}) to ({res.start}-{res.end}): '{ent_text}'")
else: else:
logger.warning(f"Cannot find exact IP substring inside entity: '{ent_text}'") logger.warning(f"IP regex match but cannot find substring position: '{ent_text}'")
else: else:
logger.warning(f"Entity IP_ADDRESS does not match IPv4 regex: '{ent_text}'") logger.warning(f"Invalid IP detected, skipping: '{ent_text}'")
continue continue
filtered_results.append(res) filtered_results.append(res)
# Option: filtrer les chevauchements # Retourner le résultat nettoyé
response_data = [res.to_dict() for res in filtered_results] response_data = [res.to_dict() for res in filtered_results]
return make_response(jsonify(response_data), 200) return make_response(jsonify(response_data), 200)
except Exception as e: except Exception as e:
logger.exception(f"Error during analysis for language '{language}'.") logger.exception("Error processing analysis")
if "No matching recognizers" in str(e):
return jsonify({"error": f"No recognizers available for language '{language}'."}), 400
return jsonify({"error": str(e)}), 500 return jsonify({"error": str(e)}), 500
if __name__ == "__main__": if __name__ == "__main__":
app.run(host="0.0.0.0", port=5001) app.run(host="0.0.0.0", port=5001)