@fuego-systems/core 0.1.11 → 0.1.14

package/dist/search/normalizer.js.map

@@ -1 +1 @@
-{"version":3,"file":"normalizer.js","sourceRoot":"","sources":["../../src/search/normalizer.ts"],"names":[],"mappings":";AAAA;;;;;;;;;;GAUG;;AAoDH,8BAuBC;AAzED,yEAAyE;AACzE,MAAM,aAAa,GAAuB;IACxC,CAAC,SAAS,EAAE,sBAAsB,CAAC;IACnC,CAAC,UAAU,EAAE,sBAAsB,CAAC;IACpC,CAAC,UAAU,EAAE,sBAAsB,CAAC;IACpC,CAAC,SAAS,EAAE,aAAa,CAAC;IAC1B,CAAC,UAAU,EAAE,YAAY,CAAC;IAC1B,CAAC,SAAS,EAAE,YAAY,CAAC;IACzB,CAAC,UAAU,EAAE,WAAW,CAAC;IACzB,CAAC,SAAS,EAAE,WAAW,CAAC;IACxB,CAAC,UAAU,EAAE,SAAS,CAAC;IACvB,CAAC,UAAU,EAAE,WAAW,CAAC;IACzB,CAAC,UAAU,EAAE,SAAS,CAAC;IACvB,CAAC,UAAU,EAAE,SAAS,CAAC;IACvB,CAAC,UAAU,EAAE,UAAU,CAAC;IACxB,CAAC,UAAU,EAAE,UAAU,CAAC;IACxB,CAAC,UAAU,EAAE,UAAU,CAAC;IACxB,CAAC,WAAW,EAAE,WAAW,CAAC;IAC1B,CAAC,UAAU,EAAE,QAAQ,CAAC;IACtB,CAAC,UAAU,EAAE,YAAY,CAAC;IAC1B,CAAC,WAAW,EAAE,aAAa,CAAC;IAC5B,CAAC,YAAY,EAAE,cAAc,CAAC;IAC9B,CAAC,cAAc,EAAE,eAAe,CAAC;IACjC,CAAC,aAAa,EAAE,YAAY,CAAC;IAC7B,CAAC,aAAa,EAAE,aAAa,CAAC;IAC9B,CAAC,YAAY,EAAE,cAAc,CAAC;IAC9B,CAAC,YAAY,EAAE,WAAW,CAAC;IAC3B,CAAC,YAAY,EAAE,WAAW,CAAC;IAC3B,CAAC,UAAU,EAAE,SAAS,CAAC;IACvB,CAAC,YAAY,EAAE,cAAc,CAAC;IAC9B,CAAC,YAAY,EAAE,YAAY,CAAC;IAC5B,CAAC,YAAY,EAAE,aAAa,CAAC;IAC7B,CAAC,YAAY,EAAE,WAAW,CAAC;IAC3B,CAAC,UAAU,EAAE,aAAa,CAAC;IAC3B,CAAC,UAAU,EAAE,WAAW,CAAC;IACzB,CAAC,UAAU,EAAE,YAAY,CAAC;CAC3B,CAAA;AAED;;;;;;;;;;;GAWG;AACH,SAAgB,SAAS,CAAE,IAA+B;IACxD,IAAI,CAAC,IAAI,EAAE,CAAC;QACV,OAAO,EAAE,CAAA;IACX,CAAC;IAED,qEAAqE;IACrE,IAAI,MAAM,GAAG,IAAI;SACd,SAAS,CAAC,KAAK,CAAC;SAChB,OAAO,CAAC,kBAAkB,EAAE,EAAE,CAAC,CAAA;IAElC,qBAAqB;IACrB,MAAM,GAAG,MAAM,CAAC,WAAW,EAAE,CAAA;IAE7B,+BAA+B;IAC/B,KAAK,MAAM,CAAC,OAAO,EAAE,SAAS,CAAC,IAAI,aAAa,EAAE,CAAC;QACjD,MAAM,GAAG,MAAM,CAAC,OAAO,CAAC,OAAO,EAAE,SAAS,CAAC,CAAA;IAC7C,CAAC;IAED,2DAA2D;IAC3D,MAAM,GAAG,MAAM,CAAC,OAAO,CAAC,cAAc,EAAE,GAAG,CAAC,CAAA;IAE5C,uCAAuC;IACvC,OAAO,MAAM,CAAC,OAAO,CAAC,MAAM,EAAE,GAAG,CAAC,CAAC,IAAI,EAAE,CAAA;AAC3C,CAAC","sourcesContent":["/**\n * Text normaliser for the HealthcareService search index.\n *\n * Produces a lower-cased, accent-stripped, abbreviation-expanded string\n * suitable for trigram similarity matching via pg_trgm.\n *\n * The normalisation here must mirror what the Postgres migration achieves with\n * `unaccent()` for query-time comparisons.  By pre-normalising at index time we\n * avoid calling `unaccent()` on every row during search, which would prevent\n * the GIN index from being used.\n */\n\n/** Expanded abbreviations map (Spanish medical/radiological context). */\nconst ABBREVIATIONS: [RegExp, string][] = [\n  [/\\brm\\b/g, 'resonancia magnetica'],\n  [/\\brmn\\b/g, 'resonancia magnetica'],\n  [/\\brnm\\b/g, 'resonancia magnetica'],\n  [/\\brx\\b/g, 'radiografia'],\n  [/\\btac\\b/g, 'tomografia'],\n  [/\\btc\\b/g, 'tomografia'],\n  [/\\beco\\b/g, 'ecografia'],\n  [/\\bus\\b/g, 'ecografia'],\n  [/\\bder\\b/g, 'derecha'],\n  [/\\bizq\\b/g, 'izquierda'],\n  [/\\bint\\b/g, 'interno'],\n  [/\\bext\\b/g, 'externo'],\n  [/\\bsup\\b/g, 'superior'],\n  [/\\binf\\b/g, 'inferior'],\n  [/\\bant\\b/g, 'anterior'],\n  [/\\bpost\\b/g, 'posterior'],\n  [/\\bmed\\b/g, 'medico'],\n  [/\\bped\\b/g, 'pediatrico'],\n  [/\\bgine\\b/g, 'ginecologia'],\n  [/\\boftal\\b/g, 'oftalmologia'],\n  [/\\btraumat\\b/g, 'traumatologia'],\n  [/\\bneurol\\b/g, 'neurologia'],\n  [/\\bcardio\\b/g, 'cardiologia'],\n  [/\\bderma\\b/g, 'dermatologia'],\n  [/\\boncol\\b/g, 'oncologia'],\n  [/\\burgen\\b/g, 'urgencias'],\n  [/\\bcir\\b/g, 'cirugia'],\n  [/\\bfisio\\b/g, 'fisioterapia'],\n  [/\\bpsico\\b/g, 'psicologia'],\n  [/\\bodont\\b/g, 'odontologia'],\n  [/\\bnutri\\b/g, 'nutricion'],\n  [/\\blab\\b/g, 'laboratorio'],\n  [/\\bpat\\b/g, 'patologia'],\n  [/\\brad\\b/g, 'radiologia']\n]\n\n/**\n * Normalise a text string for storage or query comparison.\n *\n * Steps:\n * 1. NFD decompose + strip combining diacritical marks (accent removal)\n * 2. Lower-case\n * 3. Expand known medical abbreviations\n * 4. Remove non-alphanumeric characters (keep spaces)\n * 5. Collapse whitespace and trim\n * @param text - Input text to normalize.\n * @returns Normalized text suitable for trigram index lookup.\n */\nexport function normalize (text: string | null | undefined): string {\n  if (!text) {\n    return ''\n  }\n\n  // Step 1: Remove diacritics (accent stripping via NFD decomposition)\n  let result = text\n    .normalize('NFD')\n    .replace(/[\\u0300-\\u036f]/g, '')\n\n  // Step 2: Lower-case\n  result = result.toLowerCase()\n\n  // Step 3: Expand abbreviations\n  for (const [pattern, expansion] of ABBREVIATIONS) {\n    result = result.replace(pattern, expansion)\n  }\n\n  // Step 4: Remove non-alphanumeric characters (keep spaces)\n  result = result.replace(/[^a-z0-9\\s]/g, ' ')\n\n  // Step 5: Collapse whitespace and trim\n  return result.replace(/\\s+/g, ' ').trim()\n}\n"]}
+{"version":3,"file":"normalizer.js","sourceRoot":"","sources":["../../src/search/normalizer.ts"],"names":[],"mappings":";AAAA;;;;;;;;;;GAUG;;AAwDH,8BAuBC;AA7ED,yEAAyE;AACzE,MAAM,aAAa,GAAuB;IACxC,CAAC,SAAS,EAAE,sBAAsB,CAAC;IACnC,CAAC,UAAU,EAAE,sBAAsB,CAAC;IACpC,CAAC,UAAU,EAAE,sBAAsB,CAAC;IACpC,CAAC,SAAS,EAAE,aAAa,CAAC;IAC1B,CAAC,UAAU,EAAE,YAAY,CAAC;IAC1B,CAAC,SAAS,EAAE,YAAY,CAAC;IACzB,CAAC,UAAU,EAAE,WAAW,CAAC;IACzB,CAAC,SAAS,EAAE,WAAW,CAAC;IACxB,CAAC,UAAU,EAAE,SAAS,CAAC;IACvB,CAAC,UAAU,EAAE,WAAW,CAAC;IACzB,CAAC,UAAU,EAAE,SAAS,CAAC;IACvB,CAAC,UAAU,EAAE,SAAS,CAAC;IACvB,CAAC,UAAU,EAAE,UAAU,CAAC;IACxB,CAAC,UAAU,EAAE,UAAU,CAAC;IACxB,CAAC,UAAU,EAAE,UAAU,CAAC;IACxB,CAAC,WAAW,EAAE,WAAW,CAAC;IAC1B,CAAC,UAAU,EAAE,QAAQ,CAAC;IACtB,CAAC,UAAU,EAAE,YAAY,CAAC;IAC1B,CAAC,WAAW,EAAE,aAAa,CAAC;IAC5B,CAAC,YAAY,EAAE,cAAc,CAAC;IAC9B,CAAC,cAAc,EAAE,eAAe,CAAC;IACjC,CAAC,aAAa,EAAE,YAAY,CAAC;IAC7B,CAAC,aAAa,EAAE,aAAa,CAAC;IAC9B,CAAC,YAAY,EAAE,cAAc,CAAC;IAC9B,CAAC,YAAY,EAAE,WAAW,CAAC;IAC3B,CAAC,YAAY,EAAE,WAAW,CAAC;IAC3B,CAAC,UAAU,EAAE,SAAS,CAAC;IACvB,CAAC,YAAY,EAAE,cAAc,CAAC;IAC9B,CAAC,YAAY,EAAE,YAAY,CAAC;IAC5B,CAAC,YAAY,EAAE,aAAa,CAAC;IAC7B,CAAC,YAAY,EAAE,WAAW,CAAC;IAC3B,CAAC,UAAU,EAAE,aAAa,CAAC;IAC3B,CAAC,UAAU,EAAE,WAAW,CAAC;IACzB,CAAC,UAAU,EAAE,YAAY,CAAC;IAC1B,CAAC,SAAS,EAAE,cAAc,CAAC;IAC3B,CAAC,iBAAiB,EAAE,SAAS,CAAC;IAC9B,CAAC,gBAAgB,EAAE,SAAS,CAAC;IAC7B,CAAC,gBAAgB,EAAE,SAAS,CAAC;CAC9B,CAAA;AAED;;;;;;;;;;;GAWG;AACH,SAAgB,SAAS,CAAE,IAA+B;IACxD,IAAI,CAAC,IAAI,EAAE,CAAC;QACV,OAAO,EAAE,CAAA;IACX,CAAC;IAED,qEAAqE;IACrE,IAAI,MAAM,GAAG,IAAI;SACd,SAAS,CAAC,KAAK,CAAC;SAChB,OAAO,CAAC,kBAAkB,EAAE,EAAE,CAAC,CAAA;IAElC,qBAAqB;IACrB,MAAM,GAAG,MAAM,CAAC,WAAW,EAAE,CAAA;IAE7B,+BAA+B;IAC/B,KAAK,MAAM,CAAC,OAAO,EAAE,SAAS,CAAC,IAAI,aAAa,EAAE,CAAC;QACjD,MAAM,GAAG,MAAM,CAAC,OAAO,CAAC,OAAO,EAAE,SAAS,CAAC,CAAA;IAC7C,CAAC;IAED,2DAA2D;IAC3D,MAAM,GAAG,MAAM,CAAC,OAAO,CAAC,cAAc,EAAE,GAAG,CAAC,CAAA;IAE5C,uCAAuC;IACvC,OAAO,MAAM,CAAC,OAAO,CAAC,MAAM,EAAE,GAAG,CAAC,CAAC,IAAI,EAAE,CAAA;AAC3C,CAAC","sourcesContent":["/**\n * Text normaliser for the HealthcareService search index.\n *\n * Produces a lower-cased, accent-stripped, abbreviation-expanded string\n * suitable for trigram similarity matching via pg_trgm.\n *\n * The normalisation here must mirror what the Postgres migration achieves with\n * `unaccent()` for query-time comparisons.  By pre-normalising at index time we\n * avoid calling `unaccent()` on every row during search, which would prevent\n * the GIN index from being used.\n */\n\n/** Expanded abbreviations map (Spanish medical/radiological context). */\nconst ABBREVIATIONS: [RegExp, string][] = [\n  [/\\brm\\b/g, 'resonancia magnetica'],\n  [/\\brmn\\b/g, 'resonancia magnetica'],\n  [/\\brnm\\b/g, 'resonancia magnetica'],\n  [/\\brx\\b/g, 'radiografia'],\n  [/\\btac\\b/g, 'tomografia'],\n  [/\\btc\\b/g, 'tomografia'],\n  [/\\beco\\b/g, 'ecografia'],\n  [/\\bus\\b/g, 'ecografia'],\n  [/\\bder\\b/g, 'derecha'],\n  [/\\bizq\\b/g, 'izquierda'],\n  [/\\bint\\b/g, 'interno'],\n  [/\\bext\\b/g, 'externo'],\n  [/\\bsup\\b/g, 'superior'],\n  [/\\binf\\b/g, 'inferior'],\n  [/\\bant\\b/g, 'anterior'],\n  [/\\bpost\\b/g, 'posterior'],\n  [/\\bmed\\b/g, 'medico'],\n  [/\\bped\\b/g, 'pediatrico'],\n  [/\\bgine\\b/g, 'ginecologia'],\n  [/\\boftal\\b/g, 'oftalmologia'],\n  [/\\btraumat\\b/g, 'traumatologia'],\n  [/\\bneurol\\b/g, 'neurologia'],\n  [/\\bcardio\\b/g, 'cardiologia'],\n  [/\\bderma\\b/g, 'dermatologia'],\n  [/\\boncol\\b/g, 'oncologia'],\n  [/\\burgen\\b/g, 'urgencias'],\n  [/\\bcir\\b/g, 'cirugia'],\n  [/\\bfisio\\b/g, 'fisioterapia'],\n  [/\\bpsico\\b/g, 'psicologia'],\n  [/\\bodont\\b/g, 'odontologia'],\n  [/\\bnutri\\b/g, 'nutricion'],\n  [/\\blab\\b/g, 'laboratorio'],\n  [/\\bpat\\b/g, 'patologia'],\n  [/\\brad\\b/g, 'radiologia'],\n  [/\\btv\\b/g, 'transvaginal'],\n  [/\\becodoppler\\b/g, 'doppler'],\n  [/\\becodopler\\b/g, 'doppler'],\n  [/\\becodopple\\b/g, 'doppler']\n]\n\n/**\n * Normalise a text string for storage or query comparison.\n *\n * Steps:\n * 1. NFD decompose + strip combining diacritical marks (accent removal)\n * 2. Lower-case\n * 3. Expand known medical abbreviations\n * 4. Remove non-alphanumeric characters (keep spaces)\n * 5. Collapse whitespace and trim\n * @param text - Input text to normalize.\n * @returns Normalized text suitable for trigram index lookup.\n */\nexport function normalize (text: string | null | undefined): string {\n  if (!text) {\n    return ''\n  }\n\n  // Step 1: Remove diacritics (accent stripping via NFD decomposition)\n  let result = text\n    .normalize('NFD')\n    .replace(/[\\u0300-\\u036f]/g, '')\n\n  // Step 2: Lower-case\n  result = result.toLowerCase()\n\n  // Step 3: Expand abbreviations\n  for (const [pattern, expansion] of ABBREVIATIONS) {\n    result = result.replace(pattern, expansion)\n  }\n\n  // Step 4: Remove non-alphanumeric characters (keep spaces)\n  result = result.replace(/[^a-z0-9\\s]/g, ' ')\n\n  // Step 5: Collapse whitespace and trim\n  return result.replace(/\\s+/g, ' ').trim()\n}\n"]}