mask-privacy 3.1.0 → 3.2.0

package/README.md

@@ -87,12 +87,6 @@ Mask prevents the misidentification of real data as tokens by using universally
 
 This prefix-based approach ensures that the SDK does not inadvertently process valid PII as an existing token.
 
-Additional collision-proof prefixes for international identifiers:
-* Turkish TCID tokens use the `990000` prefix (no valid Kimlik number starts with `99`).
-* Saudi NID tokens use the `100000` prefix (length-constrained to avoid overlap with real IDs).
-* UAE Emirates ID tokens use the `784-0000-` prefix (zeroed sub-fields are structurally invalid).
-* IBAN tokens zero the check digits (`XX00...`), which always fails ISO 7064 Mod-97 verification.
-
 ### 4. Enterprise Async Support
 Mask is built from the ground up for high-concurrency Node.js environments. All core operations are asynchronous and promised-based. Calling `encode()`, `decode()`, or `scanAndTokenize()` allows your event loop to remain unblocked while handling PII tokenization tasks.
 
@@ -135,7 +129,7 @@ Mask includes the ability to detokenize PII embedded within larger text blocks (
 
 ## Multilingual PII Detection (Waterfall Pipeline)
 
-Mask is built for the global enterprise. While many privacy tools are English-centric, the TypeScript SDK implements a **3-Tier Waterfall Detection** strategy designed for high-performance PII detection across 8 major languages using local ONNX models.
+Mask is built for the global enterprise. The TypeScript SDK implements a **3-Tier Waterfall Detection** strategy for high-precision PII detection in **English and Spanish** using local ONNX models.
 
 ### Supported Language Matrix
 
@@ -145,12 +139,6 @@ Mask provides first-class support for the following languages:
 | :--- | :--- | :--- | :--- |
 | **English** | `en` | ✅ Full | DistilBERT (Simple) |
 | **Spanish** | `es` | ✅ Full | BERT Multilingual |
-| **French** | `fr` | ✅ Full | BERT Multilingual |
-| **German** | `de` | ✅ Full | BERT Multilingual |
-| **Turkish** | `tr` | ✅ Full | BERT Multilingual |
-| **Arabic** | `ar` | ✅ Full | BERT Multilingual |
-| **Japanese** | `ja` | ✅ Full | BERT Multilingual |
-| **Chinese** | `zh` | ✅ Full | BERT Multilingual |
 
 ### How the Waterfall Works: The Excising Mechanism
 
@@ -165,11 +153,11 @@ To maintain high performance, the TypeScript SDK does not simply run three separ
 
 ### Configuration & Environment Variables
 
-Configure your multilingual environment using standard variables. These are parsed at runtime when the `LocalTransformersScanner` is initialized.
+Configure your language environment using standard variables.
 
 | Variable | Default | Description |
 | :--- | :--- | :--- |
-| `MASK_LANGUAGES` | `en` | Comma-separated list of languages (e.g., `en,es,fr,ar`). |
+| `MASK_LANGUAGES` | `en` | Comma-separated language codes. Supported: `en`, `es`. |
 | `MASK_NLP_MODEL` | *(varies)* | Override the default model (e.g., `Xenova/bert-base-multilingual-cased-ner-hrl`). |
 | `MASK_MODEL_CACHE_DIR` | `~/.cache` | Local directory for storing serialized ONNX models. |
 | `MASK_NLP_MAX_WORKERS` | `4` | Number of worker processes/threads for NLP analysis. |
@@ -221,8 +209,8 @@ The TypeScript SDK manages AI models automatically via **Transformers.js**. For
 ```bash
 npm install @huggingface/transformers # Required extra
 
-# Pre-cache models for your required languages
-export MASK_LANGUAGES="en,es,fr"
+# Pre-cache models for English and Spanish
+export MASK_LANGUAGES="en,es"
 npx mask-privacy cache-models
 ```
 

package/dist/index.d.mts

@@ -312,14 +312,8 @@ declare class MaskClient {
  * Supported language tags:
  *   en — English (default / Latin-only fallback)
  *   es — Spanish
- *   fr — French
- *   de — German
- *   tr — Turkish
- *   ar — Arabic
- *   zh — Chinese
- *   ja — Japanese
  */
-type LanguageTag = "en" | "es" | "fr" | "de" | "tr" | "ar" | "zh" | "ja";
+type LanguageTag = "en" | "es";
 interface LanguageBreakdown {
     language: LanguageTag;
     breakdown: Record<string, number>;
@@ -332,8 +326,8 @@ interface LanguageBreakdown {
  * @example
  * ```ts
  * const resolver = new LanguageContextResolver();
- * const tag = resolver.resolve("Merhaba, TC Kimlik Numaram 12345678901");
- * // tag === "tr"
+ * const tag = resolver.resolve("Hola, mi DNI es 12345678Z");
+ * // tag === "es"
  * ```
  */
 declare class LanguageContextResolver {
@@ -457,8 +451,8 @@ interface ScoreInput {
  *   baseRisk: 0.92,
  *   matchStart: 10,
  *   matchEnd: 21,
- *   fullText: "TC Kimlik No: 10000000146",
- *   proximityTerms: new Set(["kimlik", "tc"]),
+ *   fullText: "Mi número de DNI es 12345678Z",
+ *   proximityTerms: new Set(["dni", "número"]),
  *   validatorPassed: true,
  * });
  * // score === 0.99 (validator override)

package/dist/index.d.ts

@@ -312,14 +312,8 @@ declare class MaskClient {
  * Supported language tags:
  *   en — English (default / Latin-only fallback)
  *   es — Spanish
- *   fr — French
- *   de — German
- *   tr — Turkish
- *   ar — Arabic
- *   zh — Chinese
- *   ja — Japanese
  */
-type LanguageTag = "en" | "es" | "fr" | "de" | "tr" | "ar" | "zh" | "ja";
+type LanguageTag = "en" | "es";
 interface LanguageBreakdown {
     language: LanguageTag;
     breakdown: Record<string, number>;
@@ -332,8 +326,8 @@ interface LanguageBreakdown {
  * @example
  * ```ts
  * const resolver = new LanguageContextResolver();
- * const tag = resolver.resolve("Merhaba, TC Kimlik Numaram 12345678901");
- * // tag === "tr"
+ * const tag = resolver.resolve("Hola, mi DNI es 12345678Z");
+ * // tag === "es"
  * ```
  */
 declare class LanguageContextResolver {
@@ -457,8 +451,8 @@ interface ScoreInput {
  *   baseRisk: 0.92,
  *   matchStart: 10,
  *   matchEnd: 21,
- *   fullText: "TC Kimlik No: 10000000146",
- *   proximityTerms: new Set(["kimlik", "tc"]),
+ *   fullText: "Mi número de DNI es 12345678Z",
+ *   proximityTerms: new Set(["dni", "número"]),
  *   validatorPassed: true,
  * });
  * // score === 0.99 (validator override)

package/dist/index.js

@@ -274,21 +274,6 @@ function looksLikeToken(value) {
   if (v7.startsWith("000000") && v7.length === 9) {
     return true;
   }
-  if (v7.startsWith("784-0000-") && v7.length === 18) {
-    return true;
-  }
-  if (v7.length === 11 && v7.startsWith("990000") && /^\d+$/.test(v7) && parseInt(v7[v7.length - 1], 10) % 2 === 0) {
-    return true;
-  }
-  if (v7.length === 10 && v7.startsWith("100000") && /^\d+$/.test(v7)) {
-    return true;
-  }
-  if (v7.length === 18 && v7.startsWith("88000019900101")) {
-    return true;
-  }
-  if (v7.length === 12 && v7.startsWith("000000")) {
-    return true;
-  }
   if (v7.length === 9 && v7.startsWith("000") && /[A-Z]$/.test(v7)) {
     return true;
   }
@@ -307,7 +292,7 @@ var TOKEN_PATTERN;
 var init_fpe_utils = __esm({
   "src/core/fpe_utils.ts"() {
     TOKEN_PATTERN = new RegExp(
-      "tkn-[a-f0-9]{8,64}@[A-Za-z0-9.\\-]+\\.[A-Za-z]{2,}|\\+[1-9]\\d{0,3}-555-\\d{7}|000-00-\\d{4}|4000-0000-0000-\\d{4}|000000\\d{3}|990000\\d{4}[02468]|100000\\d{4}|784-0000-\\d{7}-\\d|88000019900101\\d{3}[0-9X]|000000\\d{6}|000\\d{5}[A-Z]|[A-Z]{2}00[A-F0-9]{4,16}|<(?:PER|LOC|ORG):[^>]+>|\\[TKN-[a-f0-9]{8,64}\\]",
+      "tkn-[a-f0-9]{8,64}@[A-Za-z0-9.\\-]+\\.[A-Za-z]{2,}|\\+[1-9]\\d{0,3}-555-\\d{7}|000-00-\\d{4}|4000-0000-0000-\\d{4}|000000\\d{3}|000\\d{5}[A-Z]|[A-Z]{2}00[A-F0-9]{4,16}|<(?:PER|LOC|ORG):[^>]+>|\\[TKN-[a-f0-9]{8,64}\\]",
       // Opaque
       "g"
     );
@@ -376,24 +361,6 @@ function _computeLuhnDigit(partialNum) {
   }
   return ((10 - sum % 10) % 10).toString();
 }
-function _computeCnIdCheck(partial) {
-  const weights = [7, 9, 10, 5, 8, 4, 2, 1, 6, 3, 7, 9, 10, 5, 8, 4, 2];
-  const checkDigits = "10X98765432";
-  let total = 0;
-  for (let i6 = 0; i6 < 17; i6++) {
-    total += parseInt(partial[i6], 10) * weights[i6];
-  }
-  return checkDigits[total % 11];
-}
-function _computeJaIdCheck(partial) {
-  const weights = [6, 5, 4, 3, 2, 7, 6, 5, 4, 3, 2];
-  let total = 0;
-  for (let i6 = 0; i6 < 11; i6++) {
-    total += parseInt(partial[i6], 10) * weights[i6];
-  }
-  const remainder = total % 11;
-  return remainder <= 1 ? 0 : 11 - remainder;
-}
 function _computeEsIdCheck(num) {
   return "TRWAGMYFPDXBNJZSQVHLCKE"[num % 23];
 }
@@ -405,11 +372,6 @@ async function generateFPEToken(rawText, entityType = "UNKNOWN") {
     else if (_SSN_RE.test(text)) type = "US_SSN";
     else if (_CC_RE.test(text)) type = "CREDIT_CARD";
     else if (_ROUTING_RE.test(text)) type = "US_ROUTING_NUMBER";
-    else if (_TCID_RE.test(text)) type = "TR_TCID";
-    else if (_SAUDI_NID_RE.test(text)) type = "SA_NATIONAL_ID";
-    else if (_UAE_EID_RE.test(text)) type = "UAE_EMIRATES_ID";
-    else if (_CN_ID_RE.test(text)) type = "CN_ID";
-    else if (_JA_ID_RE.test(text)) type = "JA_ID";
     else if (_ES_ID_RE.test(text)) type = "ES_DNI";
     else if (_IBAN_RE.test(text)) type = "INTL_BANK_IBAN";
     else if (_PHONE_RE.test(text)) type = "PHONE_NUMBER";
@@ -436,35 +398,10 @@ async function generateFPEToken(rawText, entityType = "UNKNOWN") {
   if (type === "US_ROUTING_NUMBER" || type === "US_ABA_ROUTING") {
     return `000000${await _hmacDigits(text, 3)}`;
   }
-  if (type === "TR_TCID") {
-    const core = await _hmacDigits(text, 4);
-    const partial = `990000${core}`;
-    let sum1_10 = 0;
-    for (let i6 = 0; i6 < partial.length; i6++) sum1_10 += parseInt(partial[i6], 10);
-    const d11Raw = sum1_10 % 10;
-    const d11 = d11Raw % 2 === 0 ? d11Raw : (d11Raw + 1) % 10;
-    return `${partial}${d11}`;
-  }
-  if (type === "SA_NATIONAL_ID") {
-    return `100000${await _hmacDigits(text, 4)}`;
-  }
-  if (type === "UAE_EMIRATES_ID") {
-    const base = `7840000${await _hmacDigits(text, 7)}`;
-    const checkDig = _computeLuhnDigit(base);
-    return `784-0000-${base.slice(7, 14)}-${checkDig}`;
-  }
   if (type === "INTL_BANK_IBAN" || type === "IBAN_CODE") {
     const countryCode = text.length >= 2 && /[a-zA-Z]{2}/.test(text.slice(0, 2)) ? text.slice(0, 2).toUpperCase() : "US";
     return `${countryCode}00${(await _hmacHex(text, 8)).toUpperCase()}`;
   }
-  if (type === "CN_ID") {
-    const base = `88000019900101${await _hmacDigits(text, 3)}`;
-    return base + _computeCnIdCheck(base);
-  }
-  if (type === "JA_ID") {
-    const base = `000000${await _hmacDigits(text, 5)}`;
-    return base + _computeJaIdCheck(base).toString();
-  }
   if (type === "ES_DNI") {
     const digits = `000${await _hmacDigits(text, 5)}`;
     return digits + _computeEsIdCheck(parseInt(digits, 10));
@@ -484,7 +421,7 @@ async function generateFPEToken(rawText, entityType = "UNKNOWN") {
   }
   return `[TKN-${await _hmacHex(text)}]`;
 }
-var _masterKey, _EMAIL_RE, _PHONE_RE, _SSN_RE, _CC_RE, _ROUTING_RE, _TCID_RE, _SAUDI_NID_RE, _UAE_EID_RE, _IBAN_RE, _CN_ID_RE, _JA_ID_RE, _ES_ID_RE, _FIRST_NAMES, _LAST_NAMES, _CITIES;
+var _masterKey, _EMAIL_RE, _PHONE_RE, _SSN_RE, _CC_RE, _ROUTING_RE, _ES_ID_RE, _IBAN_RE, _FIRST_NAMES, _LAST_NAMES, _CITIES;
 var init_fpe = __esm({
   "src/core/fpe.ts"() {
     init_config();
@@ -497,13 +434,8 @@ var init_fpe = __esm({
     _SSN_RE = /^\d{3}-\d{2}-\d{4}$/;
     _CC_RE = /^(?:\d{4}[ \-]?){3}\d{4}$/;
     _ROUTING_RE = /^\d{9}$/;
-    _TCID_RE = /^[1-9]\d{9}[02468]$/;
-    _SAUDI_NID_RE = /^1\d{9}$/;
-    _UAE_EID_RE = /^784-\d{4}-\d{7}-\d$/;
-    _IBAN_RE = /^[A-Z]{2}\d{2}[A-Z0-9]{4,30}$/;
-    _CN_ID_RE = /^[1-9]\d{5}(?:18|19|20)\d{2}(?:0[1-9]|1[0-2])(?:0[1-9]|[12]\d|3[01])\d{3}[0-9Xx]$/;
-    _JA_ID_RE = /^\d{12}$/;
     _ES_ID_RE = /^(?:\d{8}[A-Z]|[XYZ]\d{7}[A-Z])$/;
+    _IBAN_RE = /^[A-Z]{2}\d{2}[A-Z0-9]{4,30}$/;
     _FIRST_NAMES = ["Taylor", "Jordan", "Casey", "Morgan", "Riley", "Avery", "Rowan", "Quinn", "Charlie", "Peyton", "Blake", "Dakota", "Reese", "Skyler", "Finley", "Eden", "Harley", "Rory", "Emerson", "Remi"];
     _LAST_NAMES = ["Smith", "Johnson", "Williams", "Brown", "Jones", "Garcia", "Miller", "Davis", "Rodriguez", "Martinez", "Hernandez", "Lopez", "Gonzalez", "Wilson", "Anderson", "Thomas", "Taylor", "Moore", "Jackson", "Martin"];
     _CITIES = ["London", "Paris", "Berlin", "Tokyo", "Rome", "Madrid", "Vienna", "Sydney", "Toronto", "Chicago", "Seattle", "Austin", "Boston", "Denver", "Dallas", "Miami", "Seoul", "Dubai", "Mumbai", "Cairo"];
@@ -43559,19 +43491,8 @@ var SCRIPT_SIGNATURES; exports.LanguageContextResolver = void 0;
 var init_assessor = __esm({
   "src/core/dlp/assessor.ts"() {
     SCRIPT_SIGNATURES = [
-      // CJK / East-Asian — checked first because they are unambiguous
-      { tag: "zh", pattern: /[\u4e00-\u9fff\u3400-\u4dbf]/g },
-      { tag: "ja", pattern: /[\u3040-\u309f\u30a0-\u30ff\u31f0-\u31ff]/g },
-      // Arabic script — covers Standard Arabic, Urdu overlap, etc.
-      { tag: "ar", pattern: /[\u0600-\u06ff\u0750-\u077f\u08a0-\u08ff\ufb50-\ufdff\ufe70-\ufeff]/g },
-      // Turkish — distinguished by dotless-i (ı), soft-g (ğ), ş, and cedilla ç
-      { tag: "tr", pattern: /[ğıİşŞ]/g },
-      // German — umlauts and Eszett
-      { tag: "de", pattern: /[äöüÄÖÜß]/g },
       // Spanish — ñ and inverted punctuation
-      { tag: "es", pattern: /[ñÑ¡¿]/g },
-      // French — cedilla, accented vowels with circumflex / diaeresis
-      { tag: "fr", pattern: /[àâçéèêëïîôùûüÿœæ]/gi }
+      { tag: "es", pattern: /[ñÑ¡¿]/g }
     ];
     exports.LanguageContextResolver = class {
       constructor(charThreshold = 1) {
@@ -43635,28 +43556,6 @@ var init_registry = __esm({
       es: [
         /\b[A-Z][a-záéíóúñ\-\']+ [A-Z][a-záéíóúñ\-\']+(?:\s+[A-Z][a-záéíóúñ\-\']+)?\b/g,
         /\b(?:Sr|Sra|Srta)\.?\s+[A-Z][a-záéíóúñ\-\']+\b/g
-      ],
-      fr: [
-        /\b[A-Z][a-zàâçéèêëïîôùûü\-\']+ [A-Z][a-zàâçéèêëïîôùûü\-\']+\b/g,
-        /\b(?:M|Mme|Mlle)\.?\s+[A-Z][a-zàâçéèêëïîôùûü\-\+\']+\b/g
-      ],
-      de: [
-        /\b[A-Z][a-zäöüß\-\']+ [A-Z][a-zäöüß\-\']+\b/g,
-        /\b(?:Herr|Frau)\.?\s+[A-Z][a-zäöüß\-\']+\b/g
-      ],
-      tr: [
-        /\b[A-ZÇĞİÖŞÜ][a-zçğıöşü]+ [A-ZÇĞİÖŞÜ][a-zçğıöşü]+\b/g,
-        /\b(?:Bay|Bayan|Sayın)\.?\s+[A-ZÇĞİÖŞÜ][a-zçğıöşü]+\b/g
-      ],
-      ar: [
-        /[\u0621-\u064a][\u0600-\u06ff]+ [\u0621-\u064a][\u0600-\u06ff]+/g,
-        /(?:أبو|أم|ابن|بنت)\s+[\u0621-\u064a][\u0600-\u06ff]+/gi
-      ],
-      ja: [
-        /\b[A-Z][a-z]+(?:moto|yama|kawa|mura|ta|da|shi|no)\s+[A-Z][a-z]+\b/g
-      ],
-      zh: [
-        /\b[A-Z][a-z]{1,3}\s+[A-Z][a-z]+\b/g
       ]
     };
     LOCALE_ADDRESS_RULES = {
@@ -43664,26 +43563,8 @@ var init_registry = __esm({
         /\b\d{1,5}\s+[A-Z][a-z]+(?:\s+[A-Z][a-z]+)*\s+(?:Street|St|Avenue|Ave|Road|Rd|Boulevard|Blvd|Lane|Ln|Drive|Dr|Court|Ct|Way)\b/g,
         /\b[A-Z][a-z]+(?:\s+[A-Z][a-z]+)*,\s*[A-Z]{2}\s+\d{5}(?:-\d{4})?\b/g
       ],
-      fr: [
-        /\b\d{1,4}\s+(?:rue|avenue|boulevard|place|chemin)\s+[A-ZÀ-ÖØ-Ý][a-zà-öø-ÿ]+\b/gi
-      ],
-      de: [
-        /\b[A-ZÄÖÜa-zäöüß]+(?:straße|strasse|weg|gasse|platz)\s+\d{1,4}\b/g
-      ],
-      tr: [
-        /\b[A-ZÇĞİÖŞÜa-zçğıöşü]+\s+(?:Cad|Sok|Mah)\.?\s+/gi,
-        /\b\d{5}\s+[A-ZÇĞİÖŞÜa-zçğıöşü]+\/[A-ZÇĞİÖŞÜa-zçğıöşü]+\b/g
-      ],
-      ar: [
-        /شارع\s+[\u0600-\u06ff]+/g,
-        /حي\s+[\u0600-\u06ff]+/g,
-        /(?:ص\.ب|P\.?O\.?\s*Box)\s*\d{3,6}/gi
-      ],
-      uk_postcode: [
-        /\b[A-Z]{1,2}\d{1,2}[A-Z]?\s*\d[A-Z]{2}\b/g
-      ],
-      ca_postal: [
-        /\b[A-Z]\d[A-Z]\s*\d[A-Z]\d\b/g
+      es: [
+        /\b(?:Calle|Carrera|Avenida|Paseo|Plaza)\s+[A-ZÀ-ÖØ-Ý][a-zà-öø-ÿ]+\b/gi
       ]
     };
     RAW_PATTERNS = [
@@ -43902,68 +43783,6 @@ var init_registry = __esm({
         "IDENTITY_INTL" /* IDENTITY_INTL */,
         "ca_sin"
       ],
-      [
-        "FR_INSEE_NUM",
-        "\\b[12]\\d{2}[01]\\d\\d{8}\\d{2}\\b",
-        ["insee", "s\xE9curit\xE9 sociale", "france", "num\xE9ro"],
-        0.88,
-        "IDENTITY_INTL" /* IDENTITY_INTL */,
-        "fr_insee"
-      ],
-      [
-        "DE_STEUER_ID",
-        "\\b\\d{2}\\s?\\d{3}\\s?\\d{3}\\s?\\d{3}\\b",
-        ["steuer", "steuernummer", "finanzamt", "deutschland"],
-        0.87,
-        "IDENTITY_INTL" /* IDENTITY_INTL */,
-        null
-      ],
-      [
-        "TR_TCID",
-        "\\b[1-9]\\d{9}[02468]\\b",
-        ["tc", "kimlik", "vatanda\u015Fl\u0131k", "n\xFCfus", "t\xFCrkiye"],
-        0.92,
-        "IDENTITY_INTL" /* IDENTITY_INTL */,
-        "tcid"
-      ],
-      [
-        "SA_NATIONAL_ID",
-        "\\b1\\d{9}\\b",
-        ["\u0647\u0648\u064A\u0629", "\u0631\u0642\u0645 \u0627\u0644\u0647\u0648\u064A\u0629", "saudi", "\u0648\u0637\u0646\u064A\u0629", "identity"],
-        0.91,
-        "IDENTITY_INTL" /* IDENTITY_INTL */,
-        "saudi_nid"
-      ],
-      [
-        "UAE_EMIRATES_ID",
-        "784-\\d{4}-\\d{7}-\\d",
-        ["emirates", "\u0647\u0648\u064A\u0629", "uae", "emirati", "identity"],
-        0.93,
-        "IDENTITY_INTL" /* IDENTITY_INTL */,
-        "luhn",
-        true,
-        ["*", "ar"]
-      ],
-      [
-        "CN_ID",
-        "[1-9]\\d{5}(?:18|19|20)\\d{2}(?:0[1-9]|1[0-2])(0[1-9]|[12]\\d|3[01])\\d{3}[0-9Xx]",
-        ["\u8EAB\u4EFD\u8BC1", "\u8EAB\u4EFD\u53F7\u7801", "id card", "china"],
-        0.95,
-        "IDENTITY_INTL" /* IDENTITY_INTL */,
-        "cn_id",
-        true,
-        ["*", "zh"]
-      ],
-      [
-        "JA_MY_NUMBER",
-        "\\d{12}",
-        ["\u30DE\u30A4\u30CA\u30F3\u30D0\u30FC", "\u500B\u4EBA\u756A\u53F7", "my number", "japan"],
-        0.94,
-        "IDENTITY_INTL" /* IDENTITY_INTL */,
-        "ja_id",
-        true,
-        ["*", "ja"]
-      ],
       [
         "ES_DNI",
         "(?:\\d{8}[A-Z]|[XYZ]\\d{7}[A-Z])",
@@ -43974,16 +43793,6 @@ var init_registry = __esm({
         true,
         ["*", "es"]
       ],
-      [
-        "INTL_PASSPORT",
-        "[A-Z0-9]{6,12}",
-        ["passport", "travel", "immigration", "visa"],
-        0.6,
-        "IDENTITY_INTL" /* IDENTITY_INTL */,
-        null,
-        true,
-        ["*"]
-      ],
       // ── CORPORATE ──────────────────────────────────────────────────────
       [
         "CORP_EMPLOYEE_ID",
@@ -43999,7 +43808,7 @@ var init_registry = __esm({
         this.catalogue = /* @__PURE__ */ new Map();
         this.localeCategoryRegexMap = /* @__PURE__ */ new Map();
         this.buildCatalogue(loadGroups ?? null);
-        for (const loc of ["*", "en", "es", "fr", "de", "tr", "ar", "ja", "zh"]) {
+        for (const loc of ["*", "en", "es"]) {
           this.compileForLocale(loc);
         }
       }
@@ -44177,47 +43986,6 @@ function checkIpv4Octets(raw) {
   }
   return true;
 }
-function checkTcidNumber(raw) {
-  const digitsStr = raw.replace(/\D/g, "");
-  if (digitsStr.length !== 11) return false;
-  const d6 = digitsStr.split("").map(Number);
-  if (d6[0] === 0) return false;
-  if (d6[10] % 2 !== 0) return false;
-  const oddSum = d6[0] + d6[2] + d6[4] + d6[6] + d6[8];
-  const evenSum = d6[1] + d6[3] + d6[5] + d6[7];
-  const computedD10 = ((oddSum * 7 - evenSum) % 10 + 10) % 10;
-  if (computedD10 !== d6[9]) return false;
-  const firstTenSum = d6.slice(0, 10).reduce((a6, b6) => a6 + b6, 0);
-  if (firstTenSum % 10 !== d6[10]) return false;
-  return true;
-}
-function checkSaudiNid(raw) {
-  const digitsStr = raw.replace(/\D/g, "");
-  if (digitsStr.length !== 10) return false;
-  const d6 = digitsStr.split("").map(Number);
-  if (d6[0] !== 1) return false;
-  let total = 0;
-  for (let idx = 0; idx < 10; idx++) {
-    let val = d6[idx];
-    if (idx % 2 === 0) {
-      val *= 2;
-      if (val > 9) val -= 9;
-    }
-    total += val;
-  }
-  return total % 10 === 0;
-}
-function checkFrInsee(raw) {
-  let cleaned = raw.replace(/ /g, "").toUpperCase();
-  if (cleaned.length !== 15) return false;
-  cleaned = cleaned.replace(/2A/g, "19").replace(/2B/g, "18");
-  if (!/^\d+$/.test(cleaned)) return false;
-  const baseNumberStr = cleaned.slice(0, 13);
-  const expectedKey = parseInt(cleaned.slice(13), 10);
-  const baseNumber = BigInt(baseNumberStr);
-  const calculatedKey = 97n - baseNumber % 97n;
-  return Number(calculatedKey) === expectedKey;
-}
 function checkCaSin(raw) {
   const digits = raw.replace(/\D/g, "");
   if (digits.length !== 9) return false;
@@ -44237,30 +44005,6 @@ function checkUkNino(raw) {
   if (cleaned.length !== 9) return false;
   return UK_NINO_REGEX.test(cleaned);
 }
-function checkCnId(raw) {
-  const cleaned = raw.replace(/[^0-9X]/gi, "").toUpperCase();
-  if (cleaned.length !== 18) return false;
-  const weights = [7, 9, 10, 5, 8, 4, 2, 1, 6, 3, 7, 9, 10, 5, 8, 4, 2];
-  const checkDigits = "10X98765432";
-  let total = 0;
-  for (let i6 = 0; i6 < 17; i6++) {
-    total += parseInt(cleaned[i6], 10) * weights[i6];
-  }
-  return cleaned[17] === checkDigits[total % 11];
-}
-function checkJaId(raw) {
-  const digitsStr = raw.replace(/\D/g, "");
-  if (digitsStr.length !== 12) return false;
-  const d6 = digitsStr.split("").map(Number);
-  const weights = [6, 5, 4, 3, 2, 7, 6, 5, 4, 3, 2];
-  let total = 0;
-  for (let i6 = 0; i6 < 11; i6++) {
-    total += d6[i6] * weights[i6];
-  }
-  const remainder = total % 11;
-  const expected = remainder <= 1 ? 0 : 11 - remainder;
-  return d6[11] === expected;
-}
 function checkEsId(raw) {
   const cleaned = raw.replace(/[\s-]/g, "").toUpperCase();
   if (cleaned.length !== 9) return false;
@@ -44391,13 +44135,8 @@ var init_handlers = __esm({
       vin_format: checkVinFormat,
       btc_format: checkBtcFormat,
       ipv4: checkIpv4Octets,
-      tcid: checkTcidNumber,
-      saudi_nid: checkSaudiNid,
-      fr_insee: checkFrInsee,
       ca_sin: checkCaSin,
       uk_nino: checkUkNino,
-      cn_id: checkCnId,
-      ja_id: checkJaId,
       es_id: checkEsId
     };
     exports.DLPValidationEngine = class {