euredact 0.2.0 → 0.3.1

package/README.md

@@ -166,7 +166,7 @@ interface Detection {
 String enum with all supported PII categories:
 
 ```
-NAME              ADDRESS           IBAN              BIC
+NAME              ADDRESS           BANK_ACCOUNT      BIC
 CREDIT_CARD       PHONE             EMAIL             DOB
 DATE_OF_DEATH     NATIONAL_ID       SSN               TAX_ID
 PASSPORT          DRIVERS_LICENSE   RESIDENCE_PERMIT  LICENSE_PLATE
@@ -178,6 +178,76 @@ SOCIAL_HANDLE     SECRET            OTHER
 
 For custom patterns, `entityType` is a plain string (e.g. `"EMPLOYEE_ID"`).
 
+## Country codes
+
+`countries` accepts **ISO 3166-1 alpha-2** codes. The two EU/VAT spellings are
+accepted as equivalents: `GB`/`UK` and `GR`/`EL`. Codes are case-insensitive
+and whitespace-tolerant.
+
+An **unrecognised** code does not throw — it logs an `[euredact]` warning and
+detection continues with the shared, country-independent patterns. Throwing on
+an unknown locale invites callers to wrap the call in `try/catch` and skip
+redaction entirely, failing open with unredacted PII.
+
+## Country-independent detection
+
+`countries` narrows which country-specific patterns run. It does **not** gate
+identifiers that carry their own evidence:
+
+| Type | Why it ignores `countries` |
+|---|---|
+| `BANK_ACCOUNT` (IBAN) | carries an ISO 3166 country code and a mod-97 checksum |
+| `PHONE` (with `+` prefix) | an E.164 country prefix is self-identifying |
+| `EMAIL`, `CREDIT_CARD`, `VIN`, `IMEI`, `IP_ADDRESS`, `UUID`, `SECRET` | no national form to gate on |
+
+```ts
+redact("Rekening: BE68 5390 0754 7034", { countries: ["AT"] });
+// -> 'Rekening: [BANK_ACCOUNT]'
+```
+
+## BIC detection
+
+BIC is the only bank identifier here with **no check digit**, and characters
+5-6 of ordinary uppercase words are frequently valid ISO 3166 country codes
+(`DRINGEND` -> `GE`, `HOSPITAL` -> `IT`). Detection is therefore gated:
+
+| Stage | Condition | Result |
+|---|---|---|
+| Gate 0 | the token also occurs as an ordinary lowercase word in the same document | never emitted |
+| Tier 1 | registry hit on the BIC6 institution+country prefix | emitted |
+| Gate 2 | heading / shouted-word shape | never emitted |
+| Tier 2 | `BIC`/`SWIFT` keyword, an IBAN, or a bank block in the enclosing line, record or paragraph | emitted |
+| — | none of the above | never emitted |
+
+The package bundles **no licensed BIC data** — only a small seed list of BIC6
+prefixes compiled from publicly published bank data. Deployments holding a
+licensed directory install it at startup:
+
+```ts
+import { setBicRegistry } from "euredact";
+
+setBicRegistry(["ABNANL2A", "INGBNL2A", "BBRUBE"]);   // iterable
+setBicRegistry(bic => myDirectory.has(bic));          // membership callable
+setBicRegistry(null);                                 // remove
+```
+
+The registry is an **accept** signal, never a filter: a code missing from it
+falls through to the context gate and is still detected when banking context is
+present. A stale list costs a little recall on bare, contextless BICs — it
+never causes a leak. Annual review is sufficient.
+
+## `IBAN` is now `BANK_ACCOUNT`
+
+The canonical type name is `BANK_ACCOUNT`; `IBAN` was a legacy alias.
+`detection.entityType` is now `"BANK_ACCOUNT"` and the placeholder written into
+redacted text is `[BANK_ACCOUNT]`.
+
+`EntityType.IBAN` is kept as an alias with the same value, so
+`EntityType.IBAN === EntityType.BANK_ACCOUNT` and code referring to the member
+keeps working. Code matching the *string* `"IBAN"` — or the `[IBAN]`
+placeholder — must be updated. `LEGACY_TYPE_ALIASES` publishes the mapping.
+
+
 ## Custom Patterns
 
 Register domain-specific PII patterns at runtime:

package/dist/cjs/cache.js

@@ -21,10 +21,10 @@ class ResultCache {
     key(text, countries, mode) {
         const sorted = [...countries].sort();
         const raw = `${text}|${sorted.join("|")}|${mode}`;
-        // Use multiple FNV-1a hashes on different slices for reduced collisions
+        // Include raw length to reduce collision risk alongside dual FNV-1a hashes
         const h1 = fnv1aHash(raw);
         const h2 = fnv1aHash(raw.slice(Math.floor(raw.length / 2)) + raw.slice(0, Math.floor(raw.length / 2)));
-        return h1 + h2;
+        return `${raw.length}:${h1}${h2}`;
     }
     get(key) {
         if (!this.enabled)

package/dist/cjs/index.js

@@ -1,17 +1,23 @@
 "use strict";
 Object.defineProperty(exports, "__esModule", { value: true });
-exports.COUNTRY_CONFIGS = exports.EuRedact = exports.DetectionSource = exports.EntityType = void 0;
+exports.SEED_BIC6_PREFIXES = exports.getBicRegistry = exports.setBicRegistry = exports.COUNTRY_CONFIGS = exports.EuRedact = exports.LEGACY_TYPE_ALIASES = exports.DetectionSource = exports.EntityType = void 0;
 exports.availableCountries = availableCountries;
 exports.redact = redact;
 exports.addCustomPattern = addCustomPattern;
+exports.clear = clear;
 exports.redactBatch = redactBatch;
 var types_js_1 = require("./types.js");
 Object.defineProperty(exports, "EntityType", { enumerable: true, get: function () { return types_js_1.EntityType; } });
 Object.defineProperty(exports, "DetectionSource", { enumerable: true, get: function () { return types_js_1.DetectionSource; } });
+Object.defineProperty(exports, "LEGACY_TYPE_ALIASES", { enumerable: true, get: function () { return types_js_1.LEGACY_TYPE_ALIASES; } });
 var sdk_js_1 = require("./sdk.js");
 Object.defineProperty(exports, "EuRedact", { enumerable: true, get: function () { return sdk_js_1.EuRedact; } });
 var index_js_1 = require("./rules/countries/index.js");
 Object.defineProperty(exports, "COUNTRY_CONFIGS", { enumerable: true, get: function () { return index_js_1.COUNTRY_CONFIGS; } });
+var bicRegistry_js_1 = require("./rules/bicRegistry.js");
+Object.defineProperty(exports, "setBicRegistry", { enumerable: true, get: function () { return bicRegistry_js_1.setBicRegistry; } });
+Object.defineProperty(exports, "getBicRegistry", { enumerable: true, get: function () { return bicRegistry_js_1.getBicRegistry; } });
+Object.defineProperty(exports, "SEED_BIC6_PREFIXES", { enumerable: true, get: function () { return bicRegistry_js_1.SEED_BIC6_PREFIXES; } });
 const sdk_js_2 = require("./sdk.js");
 const index_js_2 = require("./rules/countries/index.js");
 let _instance = null;
@@ -29,6 +35,9 @@ function redact(text, options) {
 function addCustomPattern(name, pattern) {
     getInstance().addCustomPattern(name, pattern);
 }
+function clear() {
+    getInstance().clear();
+}
 function redactBatch(texts, options) {
     return getInstance().redactBatch(texts, options);
 }

package/dist/cjs/normalizer.js

@@ -9,27 +9,25 @@ function normalize(text) {
     return [normalized, buildOffsetMapping(text, normalized)];
 }
 function buildOffsetMapping(original, normalized) {
-    const nfdOriginal = original.normalize("NFD");
-    const nfdNormalized = normalized.normalize("NFD");
-    // Build: normalized index -> original index via NFD as intermediate
+    // Precompute cumulative NFD lengths for normalized characters
+    const normNfdCum = [0];
+    for (let i = 0; i < normalized.length; i++) {
+        normNfdCum.push(normNfdCum[i] + normalized[i].normalize("NFD").length);
+    }
+    // Precompute cumulative NFD lengths for original characters
+    const origNfdCum = [0];
+    for (let i = 0; i < original.length; i++) {
+        origNfdCum.push(origNfdCum[i] + original[i].normalize("NFD").length);
+    }
+    // Two-pointer scan: map each normalized position to its original position
     const normToOrig = [];
+    let origPtr = 0;
     for (let normIdx = 0; normIdx < normalized.length; normIdx++) {
-        const nfdOfNormChar = normalized[normIdx].normalize("NFD");
-        let nfdStart = 0;
-        for (let i = 0; i < normIdx; i++) {
-            nfdStart += normalized[i].normalize("NFD").length;
-        }
-        let targetOrig = 0;
-        let running = 0;
-        for (let oi = 0; oi < original.length; oi++) {
-            const nfdOc = original[oi].normalize("NFD");
-            if (running + nfdOc.length > nfdStart) {
-                targetOrig = oi;
-                break;
-            }
-            running += nfdOc.length;
+        const nfdStart = normNfdCum[normIdx];
+        while (origPtr + 1 < original.length && origNfdCum[origPtr + 1] <= nfdStart) {
+            origPtr++;
         }
-        normToOrig.push(targetOrig);
+        normToOrig.push(origPtr);
     }
     return normToOrig;
 }

package/dist/cjs/rules/bicRegistry.js

@@ -0,0 +1,239 @@
+"use strict";
+/**
+ * BIC registry — tier-1 membership testing for ISO 9362 bank identifier codes.
+ *
+ * BIC is the only bank identifier in the supported set with **no check digit**
+ * (IBAN has mod-97, VAT has country-specific checksums). Structure alone cannot
+ * separate a genuine BIC from an ordinary 8- or 11-character uppercase word,
+ * because characters 5-6 of everyday words are frequently valid ISO 3166
+ * country codes (`DRINGEND` -> `GE`, `HOSPITAL` -> `IT`). Registry membership
+ * is therefore the strongest precision signal available.
+ *
+ * Two sources are consulted, in order:
+ *
+ * 1. A deployment-supplied registry, installed via {@link setBicRegistry}.
+ *    This is the path for anyone holding a licensed directory (SWIFTRef).
+ * 2. {@link SEED_BIC6_PREFIXES} — a small bundled set of institution+country
+ *    prefixes for major European banks, compiled from publicly published bank
+ *    data. No licensed directory is bundled.
+ *
+ * Matching is on the **BIC6 prefix** (4-letter institution + 2-letter country),
+ * so any location and branch code on a known institution is accepted. The
+ * prefix is the stable part of a BIC; branch codes churn as branches open and
+ * close.
+ *
+ * The registry is used as an *accept* signal, never as a filter: a candidate
+ * that misses it falls through to the context gate in `suppressors.ts` and is
+ * still detected when banking context is present. A stale list therefore costs
+ * a little recall on bare, contextless BICs — it never causes a leak. Annual
+ * review is sufficient.
+ */
+Object.defineProperty(exports, "__esModule", { value: true });
+exports.SEED_BIC6_PREFIXES = void 0;
+exports.setBicRegistry = setBicRegistry;
+exports.getBicRegistry = getBicRegistry;
+exports.isRegisteredBic = isRegisteredBic;
+// BIC6 prefixes (institution + country) of major European institutions,
+// compiled from publicly published bank data — not from a licensed directory.
+// Kept deliberately small: this covers the banks that appear bare and
+// contextless in real documents. Anything outside this list is still detected
+// via the context gate.
+exports.SEED_BIC6_PREFIXES = new Set([
+    // --- Netherlands ---
+    "ABNANL", // ABN AMRO
+    "INGBNL", // ING
+    "RABONL", // Rabobank
+    "SNSBNL", // SNS Bank
+    "TRIONL", // Triodos
+    "BUNQNL", // bunq
+    "KNABNL", // Knab
+    "ASNBNL", // ASN Bank
+    "RBRBNL", // RegioBank
+    "FVLBNL", // Van Lanschot
+    "BITSNL", // BinckBank
+    "DEUTNL", // Deutsche Bank NL
+    // --- Belgium ---
+    "BBRUBE", // BNP Paribas Fortis
+    "GEBABE", // BNP Paribas Fortis (legacy)
+    "KREDBE", // KBC
+    "GKCCBE", // Belfius
+    "BPOTBE", // bpost bank
+    "ARSPBE", // Argenta
+    "AXABBE", // AXA Bank Belgium
+    "CREGBE", // Belfius (legacy)
+    "NICABE", // Crelan
+    "TRIOBE", // Triodos Belgium
+    "CTBKBE", // Banca Monte Paschi Belgio
+    "JVBABE", // Van Breda
+    // --- Germany ---
+    "DEUTDE", // Deutsche Bank
+    "COBADE", // Commerzbank
+    "PBNKDE", // Postbank
+    "GENODE", // Volksbanken / Raiffeisen
+    "MALADE", // Sparkassen
+    "HELADE", // Helaba / Sparkassen
+    "BYLADE", // BayernLB / Sparkassen
+    "SOLADE", // Sparkassen (Baden-Wurttemberg)
+    "WELADE", // Sparkassen (Westfalen)
+    "NOLADE", // Norddeutsche Landesbank
+    "SPKKDE", // Sparkasse
+    "HYVEDE", // UniCredit / HypoVereinsbank
+    "DRESDE", // Dresdner (legacy Commerzbank)
+    "INGDDE", // ING-DiBa
+    "NTSBDE", // N26
+    "SFRTDE", // Sutor / fintech BaaS
+    "TEDADE", // Telda / BaaS
+    // --- France ---
+    "BNPAFR", // BNP Paribas
+    "SOGEFR", // Societe Generale
+    "AGRIFR", // Credit Agricole
+    "CCBPFR", // Banque Populaire
+    "CEPAFR", // Caisse d'Epargne
+    "CMCIFR", // Credit Mutuel / CIC
+    "PSSTFR", // La Banque Postale
+    "BREDFR", // BRED Banque Populaire
+    "CRLYFR", // LCL
+    "HSBCFR", // HSBC France
+    "TRZOFR", // Treezor
+    "QNTOFR", // Qonto
+    // --- Austria ---
+    "GIBAAT", // Erste Bank / Sparkassen
+    "BKAUAT", // UniCredit Bank Austria
+    "RZBAAT", // Raiffeisen Bank International
+    "RLNWAT", // Raiffeisen NO-Wien
+    "OBKLAT", // Oberbank
+    "BAWAAT", // BAWAG P.S.K.
+    "VBOEAT", // Volksbank
+    "HYPTAT", // Hypo Tirol
+    "SPIHAT", // Sparkasse
+    // --- Luxembourg ---
+    "BCEELU", // Banque et Caisse d'Epargne de l'Etat
+    "BILLLU", // Banque Internationale a Luxembourg
+    "BGLLLU", // BGL BNP Paribas
+    "CCRALU", // Banque Raiffeisen
+    "BLUXLU", // Banque de Luxembourg
+    // --- Ireland / UK ---
+    "AIBKIE", // Allied Irish Banks
+    "BOFIIE", // Bank of Ireland
+    "ULSBIE", // Ulster Bank
+    "CITIIE", // Citibank Ireland
+    "BARCGB", // Barclays
+    "HBUKGB", // HSBC UK
+    "MIDLGB", // HSBC (legacy Midland)
+    "LOYDGB", // Lloyds
+    "NWBKGB", // NatWest
+    "RBOSGB", // Royal Bank of Scotland
+    "TSBSGB", // TSB
+    "REVOGB", // Revolut
+    "MONZGB", // Monzo
+    "STARGB", // Starling
+    "CHASGB", // JP Morgan Chase UK
+    // --- Nordics ---
+    "NDEADK", // Nordea Denmark
+    "DABADK", // Danske Bank
+    "JYBADK", // Jyske Bank
+    "SYBKDK", // Sydbank
+    "NDEASE", // Nordea Sweden
+    "ESSESE", // SEB
+    "HANDSE", // Handelsbanken
+    "SWEDSE", // Swedbank
+    "DNBANO", // DNB
+    "NDEANO", // Nordea Norway
+    "NDEAFI", // Nordea Finland
+    "OKOYFI", // OP Corporate Bank
+    "HELSFI", // Aktia
+    "NBIIIS", // Landsbankinn (NBIIISRE)
+    "ESJAIS", // Arion Bank (ESJAISRE)
+    "GLITIS", // Islandsbanki (GLITISRE)
+    "MPBAIS", // Kvika banki (MPBAISRE)
+    "KBNONO", // BNbank ASA (KBNONO22)
+    "PRBONO", // Boligbanken ASA (PRBONO22)
+    "AARBDE", // Aareal Bank AG (AARBDE5W)
+    // --- Southern / Central Europe ---
+    "UNCRIT", // UniCredit Italy
+    "BCITIT", // Intesa Sanpaolo
+    "PASCIT", // Monte dei Paschi
+    "BPMOIT", // Banco BPM
+    "CRPPIT", // Cassa di Risparmio
+    "BSCHES", // Banco Santander
+    "BBVAES", // BBVA
+    "CAIXES", // CaixaBank
+    "SABBES", // Banco Sabadell
+    "BKBKES", // Bankinter
+    "BCOMPT", // Millennium BCP
+    "CGDIPT", // Caixa Geral de Depositos
+    "BPIPPT", // Banco BPI
+    "NOVBPT", // Novo Banco
+    "BPKOPL", // PKO Bank Polski
+    "BREXPL", // mBank
+    "INGBPL", // ING Bank Slaski
+    "WBKPPL", // Santander Bank Polska
+    "KOMBCZ", // Komercni banka
+    "CEKOCZ", // CSOB
+    "GIBACZ", // Ceska sporitelna
+    "TATRSK", // Tatra banka
+    "SUBASK", // Vseobecna uverova banka
+    "GIBASK", // Slovenska sporitelna
+    "OTPVHU", // OTP Bank
+    "BUDAHU", // Budapest Bank
+    "MKKBHU", // MKB Bank
+    // --- Switzerland ---
+    "UBSWCH", // UBS
+    "CRESCH", // Credit Suisse
+    "POFICH", // PostFinance
+    "ZKBKCH", // Zurcher Kantonalbank
+    "RAIFCH", // Raiffeisen Switzerland
+    "MIGRCH", // Migros Bank
+]);
+let provider = null;
+/**
+ * Install a deployment-supplied BIC registry, consulted ahead of the seed list.
+ *
+ * Accepts a membership callable, any iterable of BIC strings, or `null` to
+ * remove a previously installed registry. Entries may be full BIC8/BIC11 codes
+ * or bare BIC6 prefixes; both are matched, case-insensitively and ignoring
+ * spaces.
+ */
+function setBicRegistry(registry) {
+    if (registry === null) {
+        provider = null;
+        return;
+    }
+    if (typeof registry === "function") {
+        provider = registry;
+        return;
+    }
+    const known = new Set();
+    for (const entry of registry) {
+        const e = String(entry).trim();
+        if (e && !e.startsWith("#"))
+            known.add(e.replace(/ /g, "").toUpperCase());
+    }
+    provider = (bic) => known.has(bic);
+}
+/** Return the installed registry provider, or null if none is installed. */
+function getBicRegistry() {
+    return provider;
+}
+/**
+ * Tier 1: is `candidate` a BIC of a known institution?
+ *
+ * Checks the deployment-supplied registry first (full code, then BIC6 prefix),
+ * then the bundled seed prefixes.
+ */
+function isRegisteredBic(candidate) {
+    const c = candidate.replace(/ /g, "").toUpperCase();
+    if (c.length !== 8 && c.length !== 11)
+        return false;
+    const prefix = c.slice(0, 6);
+    if (provider !== null) {
+        try {
+            if (provider(c) || provider(prefix))
+                return true;
+        }
+        catch {
+            // a broken provider must not break detection
+        }
+    }
+    return exports.SEED_BIC6_PREFIXES.has(prefix);
+}

package/dist/cjs/rules/countries/index.js

@@ -2,6 +2,26 @@
 Object.defineProperty(exports, "__esModule", { value: true });
 exports.COUNTRY_CONFIGS = void 0;
 const types_js_1 = require("../../types.js");
+const validators_js_1 = require("../validators.js");
+/**
+ * Build a country-independent IBAN pattern from the ISO 13616 length table.
+ *
+ * An IBAN carries its own country code and a mod-97 checksum, so it is
+ * self-identifying and must never depend on the caller's `countries` setting —
+ * a Belgian account in an Austrian file is still an account.
+ *
+ * Each alternative pins the *exact* character count for its country, which
+ * stops the match absorbing a following word: without it, greedy matching
+ * turns `BE68 5390 0754 7034 KBC` into one over-long candidate that fails
+ * mod-97 and is lost entirely.
+ */
+function buildIbanPattern() {
+    const alternatives = Object.entries(validators_js_1.IBAN_LENGTHS)
+        .sort(([a], [b]) => a.localeCompare(b))
+        .map(([code, length]) => `${code}\\d{2}(?:[ \\-]?[A-Z0-9]){${length - 4}}`);
+    return String.raw `\b(?:` + alternatives.join("|") + String.raw `)\b`;
+}
+const GENERIC_IBAN = buildIbanPattern();
 /**
  * Helper to create a PatternDef with sensible defaults.
  */
@@ -9,6 +29,45 @@ function p(entityType, pattern, validator = null, description = "", contextKeywo
     return { entityType, pattern, validator, description, contextKeywords, requiresContext };
 }
 // ---------------------------------------------------------------------------
+// Secret context keywords (consolidated, reused across SECRET patterns)
+// ---------------------------------------------------------------------------
+const SECRET_CONTEXT = [
+    // English
+    "key", "token", "secret", "password", "credential",
+    "api_key", "apikey", "api-key", "auth", "bearer",
+    "aws_secret_access_key", "access_key", "private_key",
+    "AES_KEY", "ENCRYPTION_KEY", "SIGNING_KEY",
+    "ACCOUNT_SID", "AUTH_TOKEN", "TWILIO",
+    "RAILS_SECRET", "SECRET_KEY_BASE", "DATABASE_URL",
+    "NPM_TOKEN", "DIGITALOCEAN", "GOOGLE_API",
+    "DB_PASS", "DB_PASSWORD", "DATASOURCE_PASSWORD",
+    "JWT_SIGNING", "passwd", "DISCORD_TOKEN",
+    // Dutch
+    "wachtwoord", "sleutel", "geheim",
+    // French
+    "mot de passe", "mot_de_passe", "clé", "jeton",
+    // German
+    "Passwort", "Kennwort", "Schlüssel", "Geheimnis",
+    // Italian
+    "chiave", "segreto",
+    // Spanish
+    "contraseña", "clave",
+    // Portuguese
+    "senha", "chave",
+    // Polish
+    "hasło", "klucz",
+    // Czech/Slovak
+    "heslo", "kľúč",
+    // Hungarian
+    "jelszó", "kulcs",
+    // Romanian
+    "parola",
+    // Nordic
+    "lösenord", "nøkkel", "nyckel", "avain", "lykill",
+    // Greek
+    "κωδικός",
+];
+// ---------------------------------------------------------------------------
 // DOB context keywords (reused across SHARED patterns)
 // ---------------------------------------------------------------------------
 const DOB_CONTEXT = [
@@ -36,8 +95,28 @@ const SHARED = {
     code: "SHARED",
     name: "Shared EU Patterns",
     patterns: [
-        p(types_js_1.EntityType.EMAIL, String.raw `\b[\w._%+\-]+@[\w.\-]+\.[a-zA-Z]{2,}\b`),
+        // JavaScript's \w and \b are ASCII-only, so the Python engine's \w (which
+        // is Unicode-aware) matched local parts like "petras_ž@..." while this one
+        // did not. \p{L}\p{N} with the `u` flag restores parity; the leading guard
+        // is a Unicode-safe word boundary.
+        p(types_js_1.EntityType.EMAIL, String.raw `(?<![\p{L}\p{N}._%+\-])[\p{L}\p{N}._%+\-]+@(?:[\p{L}\p{N}\-]+\.)+\p{L}{2,}\b`),
+        // IBAN (any country) — structure + mod-97 only. Country-specific IBAN
+        // patterns still exist and win on identical spans; this one guarantees an
+        // IBAN is never missed just because its country was not requested.
+        p(types_js_1.EntityType.BANK_ACCOUNT, GENERIC_IBAN, "iban", "IBAN — any country (ISO 13616, mod-97 validated)"),
+        // International phone (E.164, any country). Runs alongside the per-country
+        // patterns, which each hard-code a single grouping and miss the rest:
+        // "+43 664 8213 907" was invisible to the Austrian pattern, which expects
+        // an unbroken subscriber number. A leading "+" is self-identifying, so
+        // this is deliberately not country-gated.
+        p(types_js_1.EntityType.PHONE, String.raw `\+\d{1,3}(?:[\s\-/]?\(?\d{1,4}\)?){2,7}`, "e164", "International phone number (E.164) — any country"),
         p(types_js_1.EntityType.BIC, String.raw `\b[A-Z]{4}[A-Z]{2}[A-Z0-9]{2}(?:[A-Z0-9]{3})?\b`, "bic"),
+        // Space-separated form: "RZBA AT WW", "NICA BE BB". Written as a separate
+        // pattern requiring a space between *every* group, rather than making the
+        // spaces optional above — optional spaces would let "GEBABEBB KBC" match
+        // as one 11-character code and claim the following word. Only a literal
+        // space is allowed, so a match cannot span a line break.
+        p(types_js_1.EntityType.BIC, String.raw `\b[A-Z]{4} [A-Z]{2} [A-Z0-9]{2}(?: [A-Z0-9]{3})?\b`, "bic", "BIC/SWIFT code (space-separated groups)"),
         p(types_js_1.EntityType.CREDIT_CARD, String.raw `\b(?:4[0-9]{3}[\s\-]?[0-9]{4}[\s\-]?[0-9]{4}[\s\-]?[0-9]{4}|5[1-5][0-9]{2}[\s\-]?[0-9]{4}[\s\-]?[0-9]{4}[\s\-]?[0-9]{4}|3[47][0-9]{2}[\s\-]?[0-9]{6}[\s\-]?[0-9]{5})\b`, "luhn"),
         p(types_js_1.EntityType.VIN, String.raw `\b[A-HJ-NPR-Z0-9]{17}\b`, "vin"),
         p(types_js_1.EntityType.IP_ADDRESS, String.raw `\b(?:(?:25[0-5]|2[0-4][0-9]|[01]?[0-9][0-9]?)\.){3}(?:25[0-5]|2[0-4][0-9]|[01]?[0-9][0-9]?)\b`),
@@ -48,7 +127,7 @@ const SHARED = {
         p(types_js_1.EntityType.IMEI, String.raw `\b\d{2}[\-\s]\d{6}[\-\s]\d{6}[\-\s]\d\b`, "imei", "formatted"),
         p(types_js_1.EntityType.GPS_COORDINATES, String.raw `-?(?:[1-8]?\d(?:\.\d{4,})|90(?:\.0{4,}))\s*[,;/]\s*-?(?:1[0-7]\d(?:\.\d{4,})|180(?:\.0{4,})|\d{1,2}(?:\.\d{4,}))`),
         p(types_js_1.EntityType.UUID, String.raw `\b[0-9a-fA-F]{8}-[0-9a-fA-F]{4}-[1-5][0-9a-fA-F]{3}-[89abAB][0-9a-fA-F]{3}-[0-9a-fA-F]{12}\b`),
-        p(types_js_1.EntityType.SOCIAL_HANDLE, String.raw `(?<!\w)@[a-zA-Z][a-zA-Z0-9_.]{1,29}\b`),
+        p(types_js_1.EntityType.SOCIAL_HANDLE, String.raw `(?<![\p{L}\p{N}_])@\p{L}[\p{L}\p{N}_.]{1,29}\b`),
         // --- Secret / API Key (known prefixes — always active) ---
         p(types_js_1.EntityType.SECRET, String.raw `\bAKIA[A-Z0-9]{16}\b`, null, "AWS Access Key ID"),
         p(types_js_1.EntityType.SECRET, String.raw `\bgh[poas]_[a-zA-Z0-9]{36,}\b`, null, "GitHub token"),
@@ -70,35 +149,18 @@ const SHARED = {
         p(types_js_1.EntityType.SECRET, String.raw `\bxapp-\d[\d\-]{20,}`, null, "Slack app numeric token"),
         p(types_js_1.EntityType.SECRET, String.raw `\bxapp-[a-zA-Z0-9\-]{20,}`, null, "Slack app token"),
         p(types_js_1.EntityType.SECRET, String.raw `\bdapi[a-f0-9]{32,}\b`, null, "Databricks PAT"),
+        // Discord bot token (3-segment base64 with numeric first segment)
+        p(types_js_1.EntityType.SECRET, String.raw `\b[MN][A-Za-z0-9]{23,}\.[A-Za-z0-9_\-]{6}\.[A-Za-z0-9_\-]{27,}\b`, null, "Discord bot token"),
+        // PEM private key headers
+        p(types_js_1.EntityType.SECRET, String.raw `-----BEGIN\s(?:RSA\s)?PRIVATE\sKEY-----[\s\S]{0,16384}?-----END\s(?:RSA\s)?PRIVATE\sKEY-----`, null, "PEM private key"),
+        // AWS Secret Access Key (40 chars, base64 with +/ after an AKIA line)
+        p(types_js_1.EntityType.SECRET, String.raw `(?<=aws_secret_access_key\s{0,4}[:=]\s{0,4})[A-Za-z0-9+/]{40}`, null, "AWS Secret Access Key"),
         // --- Secret / API Key (connection strings with embedded credentials) ---
-        p(types_js_1.EntityType.SECRET, String.raw `(?:mongodb|mysql|postgres(?:ql)?|redis|amqp|rabbitmq):\/\/[^\s:]+:[^\s@]+@[^\s]+`, null, "Connection string with credentials"),
+        p(types_js_1.EntityType.SECRET, String.raw `(?:mongodb|mysql|postgres(?:ql)?|redis|amqp|rabbitmq):\/\/[^\s:]{1,256}:[^\s@]{1,256}@[^\s]+`, null, "Connection string with credentials"),
         // --- Secret / API Key (assignment-based: KEY=value or KEY: value) ---
-        p(types_js_1.EntityType.SECRET, String.raw `(?<=[:=] )\S{8,}|(?<=[:=])\S{8,}`, "high_entropy", "Assigned secret value", [
-            "key", "token", "secret", "password", "credential",
-            "api_key", "apikey", "api-key", "auth", "bearer",
-            "aws_secret_access_key", "access_key", "private_key",
-            "wachtwoord", "mot de passe", "Passwort", "Schlüssel",
-            "clé", "sleutel", "lösenord", "hasło", "heslo",
-            "jelszó", "contraseña", "senha", "parola",
-            "AES_KEY", "ENCRYPTION_KEY", "SIGNING_KEY",
-            "ACCOUNT_SID", "AUTH_TOKEN", "TWILIO",
-            "RAILS_SECRET", "SECRET_KEY_BASE", "DATABASE_URL",
-            "NPM_TOKEN", "DIGITALOCEAN", "GOOGLE_API",
-            "kľúč", "nøkkel", "nyckel", "avain", "lykill",
-            "DB_PASS", "DB_PASSWORD", "DATASOURCE_PASSWORD",
-            "JWT_SIGNING", "SIGNING_KEY",
-            "passwd", "Kennwort",
-        ], true),
+        p(types_js_1.EntityType.SECRET, String.raw `(?<=[:=] )\S{8,}|(?<=[:=])\S{8,}`, "high_entropy", "Assigned secret value", SECRET_CONTEXT, true),
         // --- Secret / API Key (entropy-based fallback for longer tokens) ---
-        p(types_js_1.EntityType.SECRET, String.raw `\b[A-Za-z0-9_\-+/]{24,}[A-Za-z0-9_\-+/=]*\b`, "high_entropy", "High-entropy token", [
-            "key", "token", "secret", "password", "credential",
-            "api_key", "apikey", "api-key", "auth", "bearer",
-            "aws_secret_access_key", "access_key", "private_key",
-            "wachtwoord", "mot de passe", "Passwort", "Schlüssel",
-            "clé", "sleutel", "lösenord", "hasło", "heslo",
-            "jelszó", "contraseña", "senha", "parola",
-            "kľúč", "nøkkel", "nyckel", "avain", "lykill",
-        ], true),
+        p(types_js_1.EntityType.SECRET, String.raw `\b[A-Za-z0-9_\-+/]{24,}[A-Za-z0-9_\-+/=]*\b`, "high_entropy", "High-entropy token", SECRET_CONTEXT, true),
         p(types_js_1.EntityType.DOB, String.raw `\b(?:0[1-9]|[12][0-9]|3[01])[/.\-](?:0[1-9]|1[0-2])[/.\-](?:19|20)\d{2}\b`, null, "DD/MM/YYYY", DOB_CONTEXT, true),
         p(types_js_1.EntityType.DATE_OF_DEATH, String.raw `\b(?:0[1-9]|[12][0-9]|3[01])[/.\-](?:0[1-9]|1[0-2])[/.\-](?:19|20)\d{2}\b`, null, "", DATE_OF_DEATH_CONTEXT, true),
         p(types_js_1.EntityType.DOB, String.raw `\b(?:19|20)\d{2}[/.\-](?:0[1-9]|1[0-2])[/.\-](?:0[1-9]|[12][0-9]|3[01])\b`, null, "ISO YYYY-MM-DD", DOB_ISO_CONTEXT, true),
@@ -155,7 +217,13 @@ const DE = {
         p(types_js_1.EntityType.PHONE, String.raw `\b0[1-9]\d{1,4}[\s/\-]?\d{3,8}\b`),
         p(types_js_1.EntityType.PHONE, String.raw `\+49\s?\d{2,5}[\s/\-]?\d{3,8}\b`),
         p(types_js_1.EntityType.PASSPORT, String.raw `\b[CFGHJK][0-9A-Z]{8,9}\b`, null, "", ["Reisepass", "passport", "Passnummer", "Reisepassnummer", "Reisepass Nummer", "Reisepass-Nr", "Pass Nr"], true),
-        p(types_js_1.EntityType.LICENSE_PLATE, String.raw `\b[A-ZÄÖÜ]{1,3}[\s\-]?[A-Z]{1,2}[\s\-]?\d{1,4}[EH]?\b`),
+        // The separator after the city code is mandatory. When it was optional, a
+        // contiguous letter run split across both groups and the hyphen was
+        // consumed as the *second* separator, so any "LETTERS-DIGITS" token
+        // matched: ICD-10 parsed as city "IC" + letters "D" + "-" + "10", and
+        // ISO-9001, COVID-19, RFC-2822, POL-2025 likewise. A real plate always
+        // separates the city code: "B-AB 1234", "M-XY 99".
+        p(types_js_1.EntityType.LICENSE_PLATE, String.raw `\b[A-ZÄÖÜ]{1,3}[\s\-][A-Z]{1,2}[\s\-]?\d{1,4}[EH]?\b`),
         p(types_js_1.EntityType.POSTAL_CODE, String.raw `\b\d{5}\b`, null, "", ["PLZ", "Postleitzahl", "postal code", "postcode", "Anschrift", "Adresse", "Wohnort", "Ort", "Stadt", "wohnt", "wohnhaft", "Straße", "Str.", "Weg", "Platz", "Allee", "Ring"], true),
         p(types_js_1.EntityType.CHAMBER_OF_COMMERCE, String.raw `\bHR[AB]\s?\d{4,6}\b`),
         p(types_js_1.EntityType.SSN, String.raw `\b\d{2}[\s]?\d{6}[\s]?[A-Z][\s]?\d{3}\b`, null, "", ["Rentenversicherung", "RV-Nummer", "Sozialversicherung", "SV-Nummer", "Versicherungsnummer", "Rentenversicherungsnummer", "Sozialversicherungsnummer"], true),
@@ -172,6 +240,11 @@ const AT = {
         p(types_js_1.EntityType.IBAN, String.raw `\bAT\d{2}\s?\d{4}\s?\d{4}\s?\d{4}\s?\d{4}\b`, "iban"),
         p(types_js_1.EntityType.VAT, String.raw `\bATU\d{8}\b`),
         p(types_js_1.EntityType.PHONE, String.raw `\b0\d{3,4}[\s\-]?\d{5,8}\b`),
+        // Short area codes (Vienna is "01") split across three groups:
+        // "01 53460 2215". The pattern above requires 3-4 digits after the trunk
+        // 0, so it never matched these. Both separators are mandatory here, which
+        // keeps "01 2025" (a date fragment) out.
+        p(types_js_1.EntityType.PHONE, String.raw `\b0\d{1,4}[\s\-/]\d{3,7}[\s\-/]\d{2,6}\b`, null, "Austrian national phone — short area code, grouped"),
         p(types_js_1.EntityType.PHONE, String.raw `\+43\s?\d{3,4}[\s\-]?\d{5,8}`),
         p(types_js_1.EntityType.LICENSE_PLATE, String.raw `\b[A-ZÄÖÜ]{1,2}\s?\d{1,5}\s?[A-Z]{1,2}\b`, null, "", ["Kennzeichen", "Nummernschild", "Kfz-Kennzeichen"], true),
         p(types_js_1.EntityType.POSTAL_CODE, String.raw `\b[1-9]\d{3}\b`, null, "", ["PLZ", "Postleitzahl", "Adresse", "Anschrift", "Straße", "Str.", "Gasse", "Weg", "Platz", "Postal:", "Wohnort"], true),