diff --git a/conf/default.yaml b/conf/default.yaml index c3e4680..46ba511 100644 --- a/conf/default.yaml +++ b/conf/default.yaml @@ -1,5 +1,5 @@ # ===================================================================== -# CONFIGURATION PRESIDIO - v20 (FINALE, FIABLE ET COMPLÈTE) +# CONFIGURATION PRESIDIO - v21 (FINALE ET COMPLÈTE) # ===================================================================== supported_languages: [en, fr] @@ -52,17 +52,15 @@ recognizer_registry: patterns: [{name: Email Pattern, regex: "\\b[A-Za-z0-9._%+-]+@[A-Za-z0-9.-]+\\.[A-Z|a-z]{2,}\\b", score: 1.0}] context: ["email", "courriel", "mail"] - # --- CORRECTION FINALE : Règle Téléphone simple et robuste --- - name: BelgianPhoneRecognizer supported_language: fr supported_entity: PHONE_NUMBER patterns: - - name: Phone Pattern BE (+32 + 8-9 chiffres) + - name: Belgian Phone Pattern (+32) regex: "\\b(?:\\+|00)32(?:[\\s.-]?\\d){8,9}\\b" score: 1.0 context: ["téléphone", "tel"] - # --- CORRECTION FINALE : Règle Carte de Crédit fiable --- - name: CustomCreditCardRecognizer supported_language: fr supported_entity: CREDIT_CARD_NUMBER @@ -78,7 +76,6 @@ recognizer_registry: patterns: [{name: IBAN Pattern, regex: "\\b[A-Z]{2}[0-9]{2}\\s?(?:[A-Z0-9]{4}\\s?){2,7}[A-Z0-9]{1,4}\\b", score: 0.95}] context: ["iban", "compte"] - # --- CORRECTION FINALE : Règle SWIFT/BIC ultra-précise --- - name: SWIFTRecognizer supported_language: fr supported_entity: SWIFT_CODE @@ -88,6 +85,15 @@ recognizer_registry: score: 1.0 context: ["bic", "swift"] + - name: SmartOrganizationRecognizer + supported_language: fr + supported_entity: ORGANIZATION + patterns: + - name: Company Name with Legal Form + regex: "\\b([A-Z][a-zà-ÿ]+(?:\\s[A-Z][a-zà-ÿ]+)*)\\s+(SPRL|SARL|SA|SCS|SNC)\\b" + score: 1.0 + context: ["société", "entreprise"] + # --- NUMÉROS D'IDENTIFICATION --- - name: BelgianEnterpriseRecognizer supported_language: fr @@ -101,6 +107,15 @@ recognizer_registry: patterns: [{name: NRN Pattern, regex: "\\b[0-9]{2}\\.[0-9]{2}\\.[0-9]{2}-[0-9]{3}\\.[0-9]{2}\\b", score: 1.0}] context: ["registre national"] + - name: FrenchINSEERecognizer + supported_language: fr + supported_entity: FR_SOCIAL_SECURITY_NUMBER + patterns: + - name: INSEE Pattern with flexible spaces + regex: "\\b[12]\\s*[0-9]{2}\\s*(?:0[1-9]|1[0-2])\\s*(?:2[ABab]|[0-9]{2})\\s*[0-9]{3}\\s*[0-9]{3}\\s*[0-9]{2}\\b" + score: 0.95 + context: ["sécurité sociale", "insee", "nir"] + # 3. LISTE D'EXCLUSION (ALLOW LIST) # ===================================================================== allow_list: @@ -166,6 +181,7 @@ anonymizer_config: IBAN: replace BE_ENTERPRISE_NUMBER: replace BE_NATIONAL_REGISTER_NUMBER: replace + FR_SOCIAL_SECURITY_NUMBER: replace SWIFT_CODE: replace replacements: @@ -181,4 +197,5 @@ anonymizer_config: IBAN: "" BE_ENTERPRISE_NUMBER: "" BE_NATIONAL_REGISTER_NUMBER: "" + FR_SOCIAL_SECURITY_NUMBER: "" SWIFT_CODE: ""