@natlibfi/marc-record-validators-melinda 11.6.7 → 12.0.0-alpha.12

package/dist/normalizeFieldForComparison.js

@@ -1,103 +1,63 @@
-"use strict";
-
-Object.defineProperty(exports, "__esModule", {
-  value: true
-});
-exports.cloneAndNormalizeFieldForComparison = cloneAndNormalizeFieldForComparison;
-exports.cloneAndRemovePunctuation = cloneAndRemovePunctuation;
-exports.fieldTrimSubfieldValues = fieldTrimSubfieldValues;
-exports.isEnnakkotietoSubfieldG = isEnnakkotietoSubfieldG;
-exports.tagAndSubfieldCodeReferToIsbn = tagAndSubfieldCodeReferToIsbn;
-var _clone = _interopRequireDefault(require("clone"));
-var _punctuation = require("./punctuation2");
-var _utils = require("./utils.js");
-var _normalizeIdentifiers = require("./normalize-identifiers");
-var _debug = _interopRequireDefault(require("debug"));
-var _normalizeSubfieldValueForComparison = require("./normalizeSubfieldValueForComparison");
-function _interopRequireDefault(e) { return e && e.__esModule ? e : { default: e }; }
-/*
-  Note that this file contains very powerful normalizations and spells that are:
-  - meant for comparing similarity/mergability of two fields (clone, normalize, compare),
-  - and NOT for modifying the actual field!
-
-  This is mainly used by melinda-marc-record-merge-reducers. However, also removeInferiorDataFields fixer also used this.
-  Thus it is here. However, most of the testing is done via merge-reducers...
-*/
-
-const debug = (0, _debug.default)('@natlibfi/melinda-marc-record-merge-reducers:normalizeFieldForComparison');
-//const debugData = debug.extend('data');
-const debugDev = debug.extend('dev');
-function isEnnakkotietoSubfieldG(subfield) {
-  if (subfield.code !== 'g') {
+import clone from "clone";
+import { fieldStripPunctuation } from "./punctuation2.js";
+import { fieldToString, isControlSubfieldCode } from "./utils.js";
+import { fieldNormalizeControlNumbers } from "./normalize-identifiers.js";
+import createDebugLogger from "debug";
+import { normalizePartData, subfieldContainsPartData } from "./normalizeSubfieldValueForComparison.js";
+const debug = createDebugLogger("@natlibfi/melinda-marc-record-merge-reducers:normalizeFieldForComparison");
+const debugDev = debug.extend("dev");
+export function isEnnakkotietoSubfieldG(subfield) {
+  if (valuelessSubfield(subfield)) {
+    return false;
+  }
+  if (subfield.code !== "g") {
     return false;
   }
   return subfield.value.match(/^ENNAKKOTIETO\.?$/gui);
 }
 function debugFieldComparison(oldField, newField) {
-  // NB: Debug-only function!
-  /*
-  // We may drop certain subfields:
-  if (oldField.subfields.length === newField.subfields.length) {
-    oldField.subfields.forEach((subfield, index) => {
-      const newValue = newField.subfields[index].value;
-      if (subfield.value !== newValue) {
-        nvdebug(`NORMALIZE SUBFIELD: '${subfield.value}' => '${newValue}'`, debugDev);
-      }
-    });
-  }
-  */
-  const oldString = (0, _utils.fieldToString)(oldField);
-  const newString = (0, _utils.fieldToString)(newField);
+  const oldString = fieldToString(oldField);
+  const newString = fieldToString(newField);
   if (oldString === newString) {
     return;
   }
-  //nvdebug(`NORMALIZE FIELD:\n '${fieldToString(oldField)}' =>\n '${fieldToString(newField)}'`, debugDev);
 }
-function containsHumanName(tag = '???', subfieldCode = undefined) {
-  // NB! This set is for bibs! Auth has 400... What else...
-  if (['100', '600', '700', '800'].includes(tag)) {
-    if (subfieldCode === undefined || subfieldCode === 'a') {
+function containsHumanName(tag = "???", subfieldCode = void 0) {
+  if (["100", "600", "700", "800"].includes(tag)) {
+    if (subfieldCode === void 0 || subfieldCode === "a") {
       return true;
     }
   }
-  // Others?
   return false;
 }
-function containsCorporateName(tag = '???', subfieldCode = undefined) {
-  // NB! This set is for bibs! Auth has 400... What else...
-  if (['110', '610', '710', '810'].includes(tag)) {
-    if (subfieldCode === undefined || subfieldCode === 'a') {
+function containsCorporateName(tag = "???", subfieldCode = void 0) {
+  if (["110", "610", "710", "810"].includes(tag)) {
+    if (subfieldCode === void 0 || subfieldCode === "a") {
       return true;
     }
   }
-  // Others?
   return false;
 }
 function skipAllSubfieldNormalizations(value, subfieldCode, tag) {
-  if (isEnnakkotietoSubfieldG({
-    'code': subfieldCode,
-    value
-  })) {
+  if (isEnnakkotietoSubfieldG({ "code": subfieldCode, value })) {
     return true;
   }
-  if (tag === '035' && ['a', 'z'].includes(subfieldCode)) {
-    // A
+  if (tag === "035" && ["a", "z"].includes(subfieldCode)) {
     return true;
   }
-  if ((0, _utils.isControlSubfieldCode)(subfieldCode)) {
+  if (isControlSubfieldCode(subfieldCode)) {
     return true;
   }
   return false;
 }
 function skipSubfieldLowercase(value, subfieldCode, tag) {
-  // These may contain Roman Numerals...
-  if ((0, _normalizeSubfieldValueForComparison.subfieldContainsPartData)(tag, subfieldCode)) {
+  if (subfieldContainsPartData(tag, subfieldCode)) {
     return true;
   }
   return skipAllSubfieldNormalizations(value, subfieldCode, tag);
 }
 function skipAllFieldNormalizations(tag) {
-  if (['LOW', 'SID'].includes(tag)) {
+  if (["LOW", "SID"].includes(tag)) {
     return true;
   }
   return false;
@@ -106,66 +66,62 @@ function subfieldValueLowercase(value, subfieldCode, tag) {
   if (skipSubfieldLowercase(value, subfieldCode, tag)) {
     return value;
   }
-
-  //return value.toLowerCase();
   const newValue = value.toLowerCase();
   if (newValue !== value) {
-    //nvdebug(`SVL ${tag} $${subfieldCode} '${value}' =>`, debugDev);
-    //nvdebug(`SVL ${tag} $${subfieldCode} '${newValue}'`, debugDev);
     return newValue;
   }
   return value;
 }
 function subfieldLowercase(sf, tag) {
+  if (valuelessSubfield(sf)) {
+    return;
+  }
   sf.value = subfieldValueLowercase(sf.value, sf.code, tag);
 }
 function fieldLowercase(field) {
   if (skipFieldLowercase(field)) {
     return;
   }
-  field.subfields.forEach(sf => subfieldLowercase(sf, field.tag)); // eslint-disable-line array-callback-return
-
-  function skipFieldLowercase(field) {
-    if (skipAllFieldNormalizations(field.tag)) {
+  field.subfields.forEach((sf) => subfieldLowercase(sf, field.tag));
+  function skipFieldLowercase(field2) {
+    if (skipAllFieldNormalizations(field2.tag)) {
       return true;
     }
-    // Skip non-interesting fields
-    if (!containsHumanName(field.tag) && !containsCorporateName(field.tag) && !['240', '245', '630'].includes(field.tag)) {
+    if (!containsHumanName(field2.tag) && !containsCorporateName(field2.tag) && !["240", "245", "630"].includes(field2.tag)) {
       return true;
     }
     return false;
   }
 }
 function hack490SubfieldA(field) {
-  if (field.tag !== '490') {
+  if (field.tag !== "490") {
     return;
   }
-  field.subfields.forEach(sf => removeSarja(sf)); // eslint-disable-line array-callback-return
-
-  // NB! This won't work, if the punctuation has not been stripped beforehand!
+  field.subfields.forEach((sf) => removeSarja(sf));
   function removeSarja(subfield) {
-    if (subfield.code !== 'a') {
+    if (valuelessSubfield(subfield)) {
       return;
     }
-    const tmp = subfield.value.replace(/ ?-(?:[a-z]|ä|ö)*sarja$/u, '');
+    if (subfield.code !== "a") {
+      return;
+    }
+    const tmp = subfield.value.replace(/ ?-(?:[a-z]|ä|ö)*sarja$/u, "");
     if (tmp.length > 0) {
       subfield.value = tmp;
       return;
     }
   }
 }
-function tagAndSubfieldCodeReferToIsbn(tag, subfieldCode) {
-  // NB! We don't do this to 020$z!
-  if (subfieldCode === 'z' && ['765', '767', '770', '772', '773', '774', '776', '777', '780', '785', '786', '787'].includes(tag)) {
+export function tagAndSubfieldCodeReferToIsbn(tag, subfieldCode) {
+  if (subfieldCode === "z" && ["765", "767", "770", "772", "773", "774", "776", "777", "780", "785", "786", "787"].includes(tag)) {
     return true;
   }
-  if (tag === '020' && subfieldCode === 'a') {
+  if (tag === "020" && subfieldCode === "a") {
     return true;
   }
   return false;
 }
 function looksLikeIsbn(value) {
-  // Does not check validity!
   if (value.match(/^(?:[0-9]-?){9}(?:[0-9]-?[0-9]-?[0-9]-?)?[0-9Xx]$/u)) {
     return true;
   }
@@ -175,111 +131,86 @@ function normalizeISBN(field) {
   if (!field.subfields) {
     return;
   }
-
-  //nvdebug(`ISBN-field? ${fieldToString(field)}`);
-  const relevantSubfields = field.subfields.filter(sf => tagAndSubfieldCodeReferToIsbn(field.tag, sf.code) && looksLikeIsbn(sf.value));
-  relevantSubfields.forEach(sf => normalizeIsbnSubfield(sf)); // eslint-disable-line array-callback-return
-
+  const relevantSubfields = field.subfields.filter((sf) => tagAndSubfieldCodeReferToIsbn(field.tag, sf.code) && looksLikeIsbn(sf.value));
+  relevantSubfields.forEach((sf) => normalizeIsbnSubfield(sf));
   function normalizeIsbnSubfield(sf) {
-    //nvdebug(` ISBN-subfield? ${subfieldToString(sf)}`);
-    sf.value = sf.value.replace(/-/ug, '');
-    sf.value = sf.value.replace(/x/u, 'X');
+    if (valuelessSubfield(sf)) {
+      return;
+    }
+    sf.value = sf.value.replace(/-/ug, "");
+    sf.value = sf.value.replace(/x/u, "X");
   }
 }
 function fieldSpecificHacks(field) {
-  normalizeISBN(field); // 020$a, not $z!
+  normalizeISBN(field);
   hack490SubfieldA(field);
 }
-function fieldTrimSubfieldValues(field) {
-  field.subfields?.forEach(sf => {
-    sf.value = sf.value.replace(/^[ \t\n]+/u, '');
-    sf.value = sf.value.replace(/[ \t\n]+$/u, '');
-    sf.value = sf.value.replace(/[ \t\n]+/gu, ' ');
+export function fieldTrimSubfieldValues(field) {
+  field.subfields?.forEach((sf) => {
+    if (valuelessSubfield(sf)) {
+      return;
+    }
+    sf.value = sf.value.replace(/^[ \t\n]+/u, "");
+    sf.value = sf.value.replace(/[ \t\n]+$/u, "");
+    sf.value = sf.value.replace(/[ \t\n]+/gu, " ");
   });
 }
 function fieldRemoveDecomposedDiacritics(field) {
-  // Raison d'être/motivation: "Sirén" and diacriticless "Siren" might refer to a same surname, so this normalization
-  // allows us to compare authors and avoid duplicate fields.
-  field.subfields.forEach(sf => {
+  field.subfields.forEach((sf) => {
+    if (valuelessSubfield(sf)) {
+      return;
+    }
     sf.value = removeDecomposedDiacritics(sf.value);
   });
 }
-function removeDecomposedDiacritics(value = '') {
-  // NB #1: Does nothing to precomposed letters. Do String.normalize('NFD') first, if you want to handle them.
-  // NB #2: Finnish letters 'å', 'ä', 'ö', 'Å', Ä', and 'Ö' should be handled (=precomposed) before calling this. (= keep them as is)
-  // NB #3: Calling our very own fixComposition() before this function handles both #1 and #2.
-  return String(value).replace(/\p{Diacritic}/gu, '');
+function removeDecomposedDiacritics(value = "") {
+  return String(value).replace(/\p{Diacritic}/gu, "");
 }
 function normalizeSubfieldValue(value, subfieldCode, tag) {
-  // NB! For comparison of values only
-  /* eslint-disable */
   value = subfieldValueLowercase(value, subfieldCode, tag);
-
-  // Normalize: s. = sivut = pp.
-  value = (0, _normalizeSubfieldValueForComparison.normalizePartData)(value, subfieldCode, tag);
-  value = value.replace(/^\[([^[\]]+)\]/gu, '$1'); // eslint-disable-line functional/immutable-data
-
-  if (['130', '730'].includes(tag) && subfieldCode === 'a') {
-    value = value.replace(' : ', ', '); // "Halloween ends (elokuva, 2022)" vs "Halloween ends (elokuva : 2023)"
+  value = normalizePartData(value, subfieldCode, tag);
+  value = value.replace(/^\[([^[\]]+)\]/gu, "$1");
+  if (["130", "730"].includes(tag) && subfieldCode === "a") {
+    value = value.replace(" : ", ", ");
   }
-  /* eslint-enable */
-
-  // Not going to do these in the foreseeable future, but keeping them here for discussion:
-  // Possible normalizations include but are not limited to:
-  // ø => ö? Might be language dependent: 041 $a fin => ö, 041 $a eng => o?
-  // Ø => Ö?
-  // ß => ss
-  // þ => th (NB! Both upper and lower case)
-  // ...
-  // Probably nots:
-  // ü => y (probably not, though this correlates with Finnish letter-to-sound rules)
-  // w => v (OK for Finnish sorting in certain cases, but we are not here, are we?)
-  // I guess we should use decomposed values in code here. (Not sure what composition my examples above use.)
   return value;
 }
-function cloneAndRemovePunctuation(field) {
-  const clonedField = (0, _clone.default)(field);
+export function cloneAndRemovePunctuation(field) {
+  const clonedField = clone(field);
   if (fieldSkipNormalization(field)) {
     return clonedField;
   }
-  (0, _punctuation.fieldStripPunctuation)(clonedField);
+  fieldStripPunctuation(clonedField);
   fieldTrimSubfieldValues(clonedField);
-  debugDev('PUNC');
+  debugDev("PUNC");
   debugFieldComparison(field, clonedField);
   return clonedField;
 }
 function removeCharsThatDontCarryMeaning(value, tag, subfieldCode) {
-  if (tag === '080') {
+  if (tag === "080") {
     return value;
   }
-
-  // 3" refers to inches, but as this is for comparison only we don't mind...
-  value = value.replace(/['‘’"„“”«»]/gu, ''); // MET-570 et al. Subset of https://hexdocs.pm/ex_unicode/Unicode.Category.QuoteMarks.html
-  // MRA-273: Handle X00$a name initials.
-  // NB #1: that we remove spaces for comparison (as it simpler), though actually space should be used. Doesn't matter as this is comparison only.
-  // NB #2: we might/should eventually write a validator/fixer that adds those spaces. After that point, this expection should become obsolete.
-  if (subfieldCode === 'a' && ['100', '400', '600', '700', '800'].includes(tag)) {
-    // 400 is used in auth records. It's not a bib field at all.
-    value = value.replace(/([A-Z]|Å|Ä|Ö)\. +/ugi, '$1.');
+  value = value.replace(/['‘’"„“”«»]/gu, "");
+  if (subfieldCode === "a" && ["100", "400", "600", "700", "800"].includes(tag)) {
+    value = value.replace(/([A-Z]|Å|Ä|Ö)\. +/ugi, "$1.");
   }
   return value;
 }
 function normalizeField(field) {
-  //sf.value = removeDecomposedDiacritics(sf.value);
-  (0, _punctuation.fieldStripPunctuation)(field);
+  fieldStripPunctuation(field);
   fieldLowercase(field);
-  (0, _normalizeIdentifiers.fieldNormalizeControlNumbers)(field); // FIN11 vs FI-MELINDA etc.
+  fieldNormalizeControlNumbers(field);
   return field;
 }
-function cloneAndNormalizeFieldForComparison(field) {
-  // NB! This new field is for comparison purposes only.
-  // Some of the normalizations might be considered a bit overkill for other purposes.
-  const clonedField = (0, _clone.default)(field);
+export function cloneAndNormalizeFieldForComparison(field) {
+  const clonedField = clone(field);
   if (fieldSkipNormalization(field)) {
     return clonedField;
   }
-  clonedField.subfields.forEach(sf => {
-    // Do this for all fields or some fields?
+  clonedField.subfields.forEach((sf) => {
+    if (valuelessSubfield(sf)) {
+      return;
+    }
     sf.value = normalizeSubfieldValue(sf.value, sf.code, field.tag);
     sf.value = removeCharsThatDontCarryMeaning(sf.value, field.tag, sf.code);
   });
@@ -287,14 +218,16 @@ function cloneAndNormalizeFieldForComparison(field) {
   fieldRemoveDecomposedDiacritics(clonedField);
   fieldSpecificHacks(clonedField);
   fieldTrimSubfieldValues(clonedField);
-  debugFieldComparison(field, clonedField); // For debugging purposes only
-
+  debugFieldComparison(field, clonedField);
   return clonedField;
 }
 function fieldSkipNormalization(field) {
-  if (!field.subfields || ['018', '066', '080', '083'].includes(field.tag)) {
+  if (!field.subfields || ["018", "066", "080", "083"].includes(field.tag)) {
     return true;
   }
   return false;
 }
-//# sourceMappingURL=normalizeFieldForComparison.js.map
+function valuelessSubfield(sf) {
+  return sf.value === void 0;
+}
+//# sourceMappingURL=normalizeFieldForComparison.js.map

package/dist/normalizeFieldForComparison.js.map

@@ -1 +1,7 @@
-{"version":3,"file":"normalizeFieldForComparison.js","names":["_clone","_interopRequireDefault","require","_punctuation","_utils","_normalizeIdentifiers","_debug","_normalizeSubfieldValueForComparison","e","__esModule","default","debug","createDebugLogger","debugDev","extend","isEnnakkotietoSubfieldG","subfield","code","value","match","debugFieldComparison","oldField","newField","oldString","fieldToString","newString","containsHumanName","tag","subfieldCode","undefined","includes","containsCorporateName","skipAllSubfieldNormalizations","isControlSubfieldCode","skipSubfieldLowercase","subfieldContainsPartData","skipAllFieldNormalizations","subfieldValueLowercase","newValue","toLowerCase","subfieldLowercase","sf","fieldLowercase","field","skipFieldLowercase","subfields","forEach","hack490SubfieldA","removeSarja","tmp","replace","length","tagAndSubfieldCodeReferToIsbn","looksLikeIsbn","normalizeISBN","relevantSubfields","filter","normalizeIsbnSubfield","fieldSpecificHacks","fieldTrimSubfieldValues","fieldRemoveDecomposedDiacritics","removeDecomposedDiacritics","String","normalizeSubfieldValue","normalizePartData","cloneAndRemovePunctuation","clonedField","clone","fieldSkipNormalization","fieldStripPunctuation","removeCharsThatDontCarryMeaning","normalizeField","fieldNormalizeControlNumbers","cloneAndNormalizeFieldForComparison"],"sources":["../src/normalizeFieldForComparison.js"],"sourcesContent":["/*\n  Note that this file contains very powerful normalizations and spells that are:\n  - meant for comparing similarity/mergability of two fields (clone, normalize, compare),\n  - and NOT for modifying the actual field!\n\n  This is mainly used by melinda-marc-record-merge-reducers. However, also removeInferiorDataFields fixer also used this.\n  Thus it is here. However, most of the testing is done via merge-reducers...\n*/\nimport clone from 'clone';\nimport {fieldStripPunctuation} from './punctuation2';\nimport {fieldToString, isControlSubfieldCode} from './utils.js';\n\nimport {fieldNormalizeControlNumbers/*, normalizeControlSubfieldValue*/} from './normalize-identifiers';\nimport createDebugLogger from 'debug';\nimport {normalizePartData, subfieldContainsPartData} from './normalizeSubfieldValueForComparison';\n\nconst debug = createDebugLogger('@natlibfi/melinda-marc-record-merge-reducers:normalizeFieldForComparison');\n//const debugData = debug.extend('data');\nconst debugDev = debug.extend('dev');\n\nexport function isEnnakkotietoSubfieldG(subfield) {\n  if (subfield.code !== 'g') {\n    return false;\n  }\n  return subfield.value.match(/^ENNAKKOTIETO\\.?$/gui);\n}\n\nfunction debugFieldComparison(oldField, newField) { // NB: Debug-only function!\n  /*\n  // We may drop certain subfields:\n  if (oldField.subfields.length === newField.subfields.length) {\n    oldField.subfields.forEach((subfield, index) => {\n      const newValue = newField.subfields[index].value;\n      if (subfield.value !== newValue) {\n        nvdebug(`NORMALIZE SUBFIELD: '${subfield.value}' => '${newValue}'`, debugDev);\n      }\n    });\n  }\n  */\n  const oldString = fieldToString(oldField);\n  const newString = fieldToString(newField);\n  if (oldString === newString) {\n    return;\n  }\n  //nvdebug(`NORMALIZE FIELD:\\n '${fieldToString(oldField)}' =>\\n '${fieldToString(newField)}'`, debugDev);\n}\n\nfunction containsHumanName(tag = '???', subfieldCode = undefined) {\n  // NB! This set is for bibs! Auth has 400... What else...\n  if (['100', '600', '700', '800'].includes(tag)) {\n    if (subfieldCode === undefined || subfieldCode === 'a') {\n      return true;\n    }\n  }\n  // Others?\n  return false;\n}\n\nfunction containsCorporateName(tag = '???', subfieldCode = undefined) {\n  // NB! This set is for bibs! Auth has 400... What else...\n  if (['110', '610', '710', '810'].includes(tag)) {\n    if (subfieldCode === undefined || subfieldCode === 'a') {\n      return true;\n    }\n  }\n  // Others?\n  return false;\n}\n\nfunction skipAllSubfieldNormalizations(value, subfieldCode, tag) {\n\n  if (isEnnakkotietoSubfieldG({'code': subfieldCode, value})) {\n    return true;\n  }\n\n  if (tag === '035' && ['a', 'z'].includes(subfieldCode)) { // A\n    return true;\n  }\n\n  if (isControlSubfieldCode(subfieldCode)) {\n    return true;\n  }\n  return false;\n}\n\nfunction skipSubfieldLowercase(value, subfieldCode, tag) {\n  // These may contain Roman Numerals...\n  if (subfieldContainsPartData(tag, subfieldCode)) {\n    return true;\n  }\n\n  return skipAllSubfieldNormalizations(value, subfieldCode, tag);\n}\n\nfunction skipAllFieldNormalizations(tag) {\n  if (['LOW', 'SID'].includes(tag)) {\n    return true;\n  }\n  return false;\n}\n\n\nfunction subfieldValueLowercase(value, subfieldCode, tag) {\n  if (skipSubfieldLowercase(value, subfieldCode, tag)) {\n    return value;\n  }\n\n  //return value.toLowerCase();\n  const newValue = value.toLowerCase();\n  if (newValue !== value) {\n    //nvdebug(`SVL ${tag} $${subfieldCode} '${value}' =>`, debugDev);\n    //nvdebug(`SVL ${tag} $${subfieldCode} '${newValue}'`, debugDev);\n    return newValue;\n  }\n  return value;\n}\n\nfunction subfieldLowercase(sf, tag) {\n  sf.value = subfieldValueLowercase(sf.value, sf.code, tag);\n}\n\nfunction fieldLowercase(field) {\n  if (skipFieldLowercase(field)) {\n    return;\n  }\n\n  field.subfields.forEach(sf => subfieldLowercase(sf, field.tag)); // eslint-disable-line array-callback-return\n\n  function skipFieldLowercase(field) {\n    if (skipAllFieldNormalizations(field.tag)) {\n      return true;\n    }\n    // Skip non-interesting fields\n    if (!containsHumanName(field.tag) && !containsCorporateName(field.tag) && !['240', '245', '630'].includes(field.tag)) {\n      return true;\n    }\n\n    return false;\n  }\n}\n\n\nfunction hack490SubfieldA(field) {\n  if (field.tag !== '490') {\n    return;\n  }\n  field.subfields.forEach(sf => removeSarja(sf)); // eslint-disable-line array-callback-return\n\n  // NB! This won't work, if the punctuation has not been stripped beforehand!\n  function removeSarja(subfield) {\n    if (subfield.code !== 'a') {\n      return;\n    }\n    const tmp = subfield.value.replace(/ ?-(?:[a-z]|ä|ö)*sarja$/u, '');\n    if (tmp.length > 0) {\n      subfield.value = tmp;\n      return;\n    }\n  }\n}\n\nexport function tagAndSubfieldCodeReferToIsbn(tag, subfieldCode) {\n  // NB! We don't do this to 020$z!\n  if (subfieldCode === 'z' && ['765', '767', '770', '772', '773', '774', '776', '777', '780', '785', '786', '787'].includes(tag)) {\n    return true;\n  }\n  if (tag === '020' && subfieldCode === 'a') {\n    return true;\n  }\n  return false;\n}\n\nfunction looksLikeIsbn(value) {\n  // Does not check validity!\n  if (value.match(/^(?:[0-9]-?){9}(?:[0-9]-?[0-9]-?[0-9]-?)?[0-9Xx]$/u)) {\n    return true;\n  }\n  return false;\n}\n\nfunction normalizeISBN(field) {\n  if (!field.subfields) {\n    return;\n  }\n\n  //nvdebug(`ISBN-field? ${fieldToString(field)}`);\n  const relevantSubfields = field.subfields.filter(sf => tagAndSubfieldCodeReferToIsbn(field.tag, sf.code) && looksLikeIsbn(sf.value));\n  relevantSubfields.forEach(sf => normalizeIsbnSubfield(sf)); // eslint-disable-line array-callback-return\n\n  function normalizeIsbnSubfield(sf) {\n    //nvdebug(` ISBN-subfield? ${subfieldToString(sf)}`);\n    sf.value = sf.value.replace(/-/ug, '');\n    sf.value = sf.value.replace(/x/u, 'X');\n  }\n\n}\n\nfunction fieldSpecificHacks(field) {\n  normalizeISBN(field); // 020$a, not $z!\n  hack490SubfieldA(field);\n}\n\nexport function fieldTrimSubfieldValues(field) {\n  field.subfields?.forEach((sf) => {\n    sf.value = sf.value.replace(/^[ \\t\\n]+/u, '');\n    sf.value = sf.value.replace(/[ \\t\\n]+$/u, '');\n    sf.value = sf.value.replace(/[ \\t\\n]+/gu, ' ');\n  });\n}\n\nfunction fieldRemoveDecomposedDiacritics(field) {\n  // Raison d'être/motivation: \"Sirén\" and diacriticless \"Siren\" might refer to a same surname, so this normalization\n  // allows us to compare authors and avoid duplicate fields.\n  field.subfields.forEach((sf) => {\n    sf.value = removeDecomposedDiacritics(sf.value);\n  });\n}\n\nfunction removeDecomposedDiacritics(value = '') {\n  // NB #1: Does nothing to precomposed letters. Do String.normalize('NFD') first, if you want to handle them.\n  // NB #2: Finnish letters 'å', 'ä', 'ö', 'Å', Ä', and 'Ö' should be handled (=precomposed) before calling this. (= keep them as is)\n  // NB #3: Calling our very own fixComposition() before this function handles both #1 and #2.\n  return String(value).replace(/\\p{Diacritic}/gu, '');\n}\n\nfunction normalizeSubfieldValue(value, subfieldCode, tag) {\n  // NB! For comparison of values only\n  /* eslint-disable */\n  value = subfieldValueLowercase(value, subfieldCode, tag);\n\n  // Normalize: s. = sivut = pp.\n  value = normalizePartData(value, subfieldCode, tag);\n  value = value.replace(/^\\[([^[\\]]+)\\]/gu, '$1'); // eslint-disable-line functional/immutable-data\n\n  if (['130', '730'].includes(tag) && subfieldCode === 'a') {\n    value = value.replace(' : ', ', '); // \"Halloween ends (elokuva, 2022)\" vs \"Halloween ends (elokuva : 2023)\"\n  }\n  /* eslint-enable */\n\n  // Not going to do these in the foreseeable future, but keeping them here for discussion:\n  // Possible normalizations include but are not limited to:\n  // ø => ö? Might be language dependent: 041 $a fin => ö, 041 $a eng => o?\n  // Ø => Ö?\n  // ß => ss\n  // þ => th (NB! Both upper and lower case)\n  // ...\n  // Probably nots:\n  // ü => y (probably not, though this correlates with Finnish letter-to-sound rules)\n  // w => v (OK for Finnish sorting in certain cases, but we are not here, are we?)\n  // I guess we should use decomposed values in code here. (Not sure what composition my examples above use.)\n  return value;\n}\n\nexport function cloneAndRemovePunctuation(field) {\n  const clonedField = clone(field);\n  if (fieldSkipNormalization(field)) {\n    return clonedField;\n  }\n  fieldStripPunctuation(clonedField);\n  fieldTrimSubfieldValues(clonedField);\n  debugDev('PUNC');\n  debugFieldComparison(field, clonedField);\n\n  return clonedField;\n}\n\nfunction removeCharsThatDontCarryMeaning(value, tag, subfieldCode) {\n  if (tag === '080') {\n    return value;\n  }\n\n  // 3\" refers to inches, but as this is for comparison only we don't mind...\n  value = value.replace(/['‘’\"„“”«»]/gu, ''); // MET-570 et al. Subset of https://hexdocs.pm/ex_unicode/Unicode.Category.QuoteMarks.html\n  // MRA-273: Handle X00$a name initials.\n  // NB #1: that we remove spaces for comparison (as it simpler), though actually space should be used. Doesn't matter as this is comparison only.\n  // NB #2: we might/should eventually write a validator/fixer that adds those spaces. After that point, this expection should become obsolete.\n  if (subfieldCode === 'a' && ['100', '400', '600', '700', '800'].includes(tag)) { // 400 is used in auth records. It's not a bib field at all.\n    value = value.replace(/([A-Z]|Å|Ä|Ö)\\. +/ugi, '$1.');\n  }\n\n  return value;\n}\n\nfunction normalizeField(field) {\n  //sf.value = removeDecomposedDiacritics(sf.value);\n  fieldStripPunctuation(field);\n  fieldLowercase(field);\n  fieldNormalizeControlNumbers(field); // FIN11 vs FI-MELINDA etc.\n  return field;\n}\n\nexport function cloneAndNormalizeFieldForComparison(field) {\n  // NB! This new field is for comparison purposes only.\n  // Some of the normalizations might be considered a bit overkill for other purposes.\n  const clonedField = clone(field);\n  if (fieldSkipNormalization(field)) {\n    return clonedField;\n  }\n  clonedField.subfields.forEach((sf) => { // Do this for all fields or some fields?\n    sf.value = normalizeSubfieldValue(sf.value, sf.code, field.tag);\n    sf.value = removeCharsThatDontCarryMeaning(sf.value, field.tag, sf.code);\n  });\n\n  normalizeField(clonedField);\n  fieldRemoveDecomposedDiacritics(clonedField);\n  fieldSpecificHacks(clonedField);\n  fieldTrimSubfieldValues(clonedField);\n\n\n  debugFieldComparison(field, clonedField); // For debugging purposes only\n\n  return clonedField;\n}\n\nfunction fieldSkipNormalization(field) {\n  if (!field.subfields || ['018', '066', '080', '083'].includes(field.tag)) {\n    return true;\n  }\n  return false;\n}\n"],"mappings":";;;;;;;;;;AAQA,IAAAA,MAAA,GAAAC,sBAAA,CAAAC,OAAA;AACA,IAAAC,YAAA,GAAAD,OAAA;AACA,IAAAE,MAAA,GAAAF,OAAA;AAEA,IAAAG,qBAAA,GAAAH,OAAA;AACA,IAAAI,MAAA,GAAAL,sBAAA,CAAAC,OAAA;AACA,IAAAK,oCAAA,GAAAL,OAAA;AAAkG,SAAAD,uBAAAO,CAAA,WAAAA,CAAA,IAAAA,CAAA,CAAAC,UAAA,GAAAD,CAAA,KAAAE,OAAA,EAAAF,CAAA;AAdlG;AACA;AACA;AACA;AACA;AACA;AACA;AACA;;AASA,MAAMG,KAAK,GAAG,IAAAC,cAAiB,EAAC,0EAA0E,CAAC;AAC3G;AACA,MAAMC,QAAQ,GAAGF,KAAK,CAACG,MAAM,CAAC,KAAK,CAAC;AAE7B,SAASC,uBAAuBA,CAACC,QAAQ,EAAE;EAChD,IAAIA,QAAQ,CAACC,IAAI,KAAK,GAAG,EAAE;IACzB,OAAO,KAAK;EACd;EACA,OAAOD,QAAQ,CAACE,KAAK,CAACC,KAAK,CAAC,sBAAsB,CAAC;AACrD;AAEA,SAASC,oBAAoBA,CAACC,QAAQ,EAAEC,QAAQ,EAAE;EAAE;EAClD;AACF;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;EACE,MAAMC,SAAS,GAAG,IAAAC,oBAAa,EAACH,QAAQ,CAAC;EACzC,MAAMI,SAAS,GAAG,IAAAD,oBAAa,EAACF,QAAQ,CAAC;EACzC,IAAIC,SAAS,KAAKE,SAAS,EAAE;IAC3B;EACF;EACA;AACF;AAEA,SAASC,iBAAiBA,CAACC,GAAG,GAAG,KAAK,EAAEC,YAAY,GAAGC,SAAS,EAAE;EAChE;EACA,IAAI,CAAC,KAAK,EAAE,KAAK,EAAE,KAAK,EAAE,KAAK,CAAC,CAACC,QAAQ,CAACH,GAAG,CAAC,EAAE;IAC9C,IAAIC,YAAY,KAAKC,SAAS,IAAID,YAAY,KAAK,GAAG,EAAE;MACtD,OAAO,IAAI;IACb;EACF;EACA;EACA,OAAO,KAAK;AACd;AAEA,SAASG,qBAAqBA,CAACJ,GAAG,GAAG,KAAK,EAAEC,YAAY,GAAGC,SAAS,EAAE;EACpE;EACA,IAAI,CAAC,KAAK,EAAE,KAAK,EAAE,KAAK,EAAE,KAAK,CAAC,CAACC,QAAQ,CAACH,GAAG,CAAC,EAAE;IAC9C,IAAIC,YAAY,KAAKC,SAAS,IAAID,YAAY,KAAK,GAAG,EAAE;MACtD,OAAO,IAAI;IACb;EACF;EACA;EACA,OAAO,KAAK;AACd;AAEA,SAASI,6BAA6BA,CAACd,KAAK,EAAEU,YAAY,EAAED,GAAG,EAAE;EAE/D,IAAIZ,uBAAuB,CAAC;IAAC,MAAM,EAAEa,YAAY;IAAEV;EAAK,CAAC,CAAC,EAAE;IAC1D,OAAO,IAAI;EACb;EAEA,IAAIS,GAAG,KAAK,KAAK,IAAI,CAAC,GAAG,EAAE,GAAG,CAAC,CAACG,QAAQ,CAACF,YAAY,CAAC,EAAE;IAAE;IACxD,OAAO,IAAI;EACb;EAEA,IAAI,IAAAK,4BAAqB,EAACL,YAAY,CAAC,EAAE;IACvC,OAAO,IAAI;EACb;EACA,OAAO,KAAK;AACd;AAEA,SAASM,qBAAqBA,CAAChB,KAAK,EAAEU,YAAY,EAAED,GAAG,EAAE;EACvD;EACA,IAAI,IAAAQ,6DAAwB,EAACR,GAAG,EAAEC,YAAY,CAAC,EAAE;IAC/C,OAAO,IAAI;EACb;EAEA,OAAOI,6BAA6B,CAACd,KAAK,EAAEU,YAAY,EAAED,GAAG,CAAC;AAChE;AAEA,SAASS,0BAA0BA,CAACT,GAAG,EAAE;EACvC,IAAI,CAAC,KAAK,EAAE,KAAK,CAAC,CAACG,QAAQ,CAACH,GAAG,CAAC,EAAE;IAChC,OAAO,IAAI;EACb;EACA,OAAO,KAAK;AACd;AAGA,SAASU,sBAAsBA,CAACnB,KAAK,EAAEU,YAAY,EAAED,GAAG,EAAE;EACxD,IAAIO,qBAAqB,CAAChB,KAAK,EAAEU,YAAY,EAAED,GAAG,CAAC,EAAE;IACnD,OAAOT,KAAK;EACd;;EAEA;EACA,MAAMoB,QAAQ,GAAGpB,KAAK,CAACqB,WAAW,CAAC,CAAC;EACpC,IAAID,QAAQ,KAAKpB,KAAK,EAAE;IACtB;IACA;IACA,OAAOoB,QAAQ;EACjB;EACA,OAAOpB,KAAK;AACd;AAEA,SAASsB,iBAAiBA,CAACC,EAAE,EAAEd,GAAG,EAAE;EAClCc,EAAE,CAACvB,KAAK,GAAGmB,sBAAsB,CAACI,EAAE,CAACvB,KAAK,EAAEuB,EAAE,CAACxB,IAAI,EAAEU,GAAG,CAAC;AAC3D;AAEA,SAASe,cAAcA,CAACC,KAAK,EAAE;EAC7B,IAAIC,kBAAkB,CAACD,KAAK,CAAC,EAAE;IAC7B;EACF;EAEAA,KAAK,CAACE,SAAS,CAACC,OAAO,CAACL,EAAE,IAAID,iBAAiB,CAACC,EAAE,EAAEE,KAAK,CAAChB,GAAG,CAAC,CAAC,CAAC,CAAC;;EAEjE,SAASiB,kBAAkBA,CAACD,KAAK,EAAE;IACjC,IAAIP,0BAA0B,CAACO,KAAK,CAAChB,GAAG,CAAC,EAAE;MACzC,OAAO,IAAI;IACb;IACA;IACA,IAAI,CAACD,iBAAiB,CAACiB,KAAK,CAAChB,GAAG,CAAC,IAAI,CAACI,qBAAqB,CAACY,KAAK,CAAChB,GAAG,CAAC,IAAI,CAAC,CAAC,KAAK,EAAE,KAAK,EAAE,KAAK,CAAC,CAACG,QAAQ,CAACa,KAAK,CAAChB,GAAG,CAAC,EAAE;MACpH,OAAO,IAAI;IACb;IAEA,OAAO,KAAK;EACd;AACF;AAGA,SAASoB,gBAAgBA,CAACJ,KAAK,EAAE;EAC/B,IAAIA,KAAK,CAAChB,GAAG,KAAK,KAAK,EAAE;IACvB;EACF;EACAgB,KAAK,CAACE,SAAS,CAACC,OAAO,CAACL,EAAE,IAAIO,WAAW,CAACP,EAAE,CAAC,CAAC,CAAC,CAAC;;EAEhD;EACA,SAASO,WAAWA,CAAChC,QAAQ,EAAE;IAC7B,IAAIA,QAAQ,CAACC,IAAI,KAAK,GAAG,EAAE;MACzB;IACF;IACA,MAAMgC,GAAG,GAAGjC,QAAQ,CAACE,KAAK,CAACgC,OAAO,CAAC,0BAA0B,EAAE,EAAE,CAAC;IAClE,IAAID,GAAG,CAACE,MAAM,GAAG,CAAC,EAAE;MAClBnC,QAAQ,CAACE,KAAK,GAAG+B,GAAG;MACpB;IACF;EACF;AACF;AAEO,SAASG,6BAA6BA,CAACzB,GAAG,EAAEC,YAAY,EAAE;EAC/D;EACA,IAAIA,YAAY,KAAK,GAAG,IAAI,CAAC,KAAK,EAAE,KAAK,EAAE,KAAK,EAAE,KAAK,EAAE,KAAK,EAAE,KAAK,EAAE,KAAK,EAAE,KAAK,EAAE,KAAK,EAAE,KAAK,EAAE,KAAK,EAAE,KAAK,CAAC,CAACE,QAAQ,CAACH,GAAG,CAAC,EAAE;IAC9H,OAAO,IAAI;EACb;EACA,IAAIA,GAAG,KAAK,KAAK,IAAIC,YAAY,KAAK,GAAG,EAAE;IACzC,OAAO,IAAI;EACb;EACA,OAAO,KAAK;AACd;AAEA,SAASyB,aAAaA,CAACnC,KAAK,EAAE;EAC5B;EACA,IAAIA,KAAK,CAACC,KAAK,CAAC,oDAAoD,CAAC,EAAE;IACrE,OAAO,IAAI;EACb;EACA,OAAO,KAAK;AACd;AAEA,SAASmC,aAAaA,CAACX,KAAK,EAAE;EAC5B,IAAI,CAACA,KAAK,CAACE,SAAS,EAAE;IACpB;EACF;;EAEA;EACA,MAAMU,iBAAiB,GAAGZ,KAAK,CAACE,SAAS,CAACW,MAAM,CAACf,EAAE,IAAIW,6BAA6B,CAACT,KAAK,CAAChB,GAAG,EAAEc,EAAE,CAACxB,IAAI,CAAC,IAAIoC,aAAa,CAACZ,EAAE,CAACvB,KAAK,CAAC,CAAC;EACpIqC,iBAAiB,CAACT,OAAO,CAACL,EAAE,IAAIgB,qBAAqB,CAAChB,EAAE,CAAC,CAAC,CAAC,CAAC;;EAE5D,SAASgB,qBAAqBA,CAAChB,EAAE,EAAE;IACjC;IACAA,EAAE,CAACvB,KAAK,GAAGuB,EAAE,CAACvB,KAAK,CAACgC,OAAO,CAAC,KAAK,EAAE,EAAE,CAAC;IACtCT,EAAE,CAACvB,KAAK,GAAGuB,EAAE,CAACvB,KAAK,CAACgC,OAAO,CAAC,IAAI,EAAE,GAAG,CAAC;EACxC;AAEF;AAEA,SAASQ,kBAAkBA,CAACf,KAAK,EAAE;EACjCW,aAAa,CAACX,KAAK,CAAC,CAAC,CAAC;EACtBI,gBAAgB,CAACJ,KAAK,CAAC;AACzB;AAEO,SAASgB,uBAAuBA,CAAChB,KAAK,EAAE;EAC7CA,KAAK,CAACE,SAAS,EAAEC,OAAO,CAAEL,EAAE,IAAK;IAC/BA,EAAE,CAACvB,KAAK,GAAGuB,EAAE,CAACvB,KAAK,CAACgC,OAAO,CAAC,YAAY,EAAE,EAAE,CAAC;IAC7CT,EAAE,CAACvB,KAAK,GAAGuB,EAAE,CAACvB,KAAK,CAACgC,OAAO,CAAC,YAAY,EAAE,EAAE,CAAC;IAC7CT,EAAE,CAACvB,KAAK,GAAGuB,EAAE,CAACvB,KAAK,CAACgC,OAAO,CAAC,YAAY,EAAE,GAAG,CAAC;EAChD,CAAC,CAAC;AACJ;AAEA,SAASU,+BAA+BA,CAACjB,KAAK,EAAE;EAC9C;EACA;EACAA,KAAK,CAACE,SAAS,CAACC,OAAO,CAAEL,EAAE,IAAK;IAC9BA,EAAE,CAACvB,KAAK,GAAG2C,0BAA0B,CAACpB,EAAE,CAACvB,KAAK,CAAC;EACjD,CAAC,CAAC;AACJ;AAEA,SAAS2C,0BAA0BA,CAAC3C,KAAK,GAAG,EAAE,EAAE;EAC9C;EACA;EACA;EACA,OAAO4C,MAAM,CAAC5C,KAAK,CAAC,CAACgC,OAAO,CAAC,iBAAiB,EAAE,EAAE,CAAC;AACrD;AAEA,SAASa,sBAAsBA,CAAC7C,KAAK,EAAEU,YAAY,EAAED,GAAG,EAAE;EACxD;EACA;EACAT,KAAK,GAAGmB,sBAAsB,CAACnB,KAAK,EAAEU,YAAY,EAAED,GAAG,CAAC;;EAExD;EACAT,KAAK,GAAG,IAAA8C,sDAAiB,EAAC9C,KAAK,EAAEU,YAAY,EAAED,GAAG,CAAC;EACnDT,KAAK,GAAGA,KAAK,CAACgC,OAAO,CAAC,kBAAkB,EAAE,IAAI,CAAC,CAAC,CAAC;;EAEjD,IAAI,CAAC,KAAK,EAAE,KAAK,CAAC,CAACpB,QAAQ,CAACH,GAAG,CAAC,IAAIC,YAAY,KAAK,GAAG,EAAE;IACxDV,KAAK,GAAGA,KAAK,CAACgC,OAAO,CAAC,KAAK,EAAE,IAAI,CAAC,CAAC,CAAC;EACtC;EACA;;EAEA;EACA;EACA;EACA;EACA;EACA;EACA;EACA;EACA;EACA;EACA;EACA,OAAOhC,KAAK;AACd;AAEO,SAAS+C,yBAAyBA,CAACtB,KAAK,EAAE;EAC/C,MAAMuB,WAAW,GAAG,IAAAC,cAAK,EAACxB,KAAK,CAAC;EAChC,IAAIyB,sBAAsB,CAACzB,KAAK,CAAC,EAAE;IACjC,OAAOuB,WAAW;EACpB;EACA,IAAAG,kCAAqB,EAACH,WAAW,CAAC;EAClCP,uBAAuB,CAACO,WAAW,CAAC;EACpCrD,QAAQ,CAAC,MAAM,CAAC;EAChBO,oBAAoB,CAACuB,KAAK,EAAEuB,WAAW,CAAC;EAExC,OAAOA,WAAW;AACpB;AAEA,SAASI,+BAA+BA,CAACpD,KAAK,EAAES,GAAG,EAAEC,YAAY,EAAE;EACjE,IAAID,GAAG,KAAK,KAAK,EAAE;IACjB,OAAOT,KAAK;EACd;;EAEA;EACAA,KAAK,GAAGA,KAAK,CAACgC,OAAO,CAAC,eAAe,EAAE,EAAE,CAAC,CAAC,CAAC;EAC5C;EACA;EACA;EACA,IAAItB,YAAY,KAAK,GAAG,IAAI,CAAC,KAAK,EAAE,KAAK,EAAE,KAAK,EAAE,KAAK,EAAE,KAAK,CAAC,CAACE,QAAQ,CAACH,GAAG,CAAC,EAAE;IAAE;IAC/ET,KAAK,GAAGA,KAAK,CAACgC,OAAO,CAAC,sBAAsB,EAAE,KAAK,CAAC;EACtD;EAEA,OAAOhC,KAAK;AACd;AAEA,SAASqD,cAAcA,CAAC5B,KAAK,EAAE;EAC7B;EACA,IAAA0B,kCAAqB,EAAC1B,KAAK,CAAC;EAC5BD,cAAc,CAACC,KAAK,CAAC;EACrB,IAAA6B,kDAA4B,EAAC7B,KAAK,CAAC,CAAC,CAAC;EACrC,OAAOA,KAAK;AACd;AAEO,SAAS8B,mCAAmCA,CAAC9B,KAAK,EAAE;EACzD;EACA;EACA,MAAMuB,WAAW,GAAG,IAAAC,cAAK,EAACxB,KAAK,CAAC;EAChC,IAAIyB,sBAAsB,CAACzB,KAAK,CAAC,EAAE;IACjC,OAAOuB,WAAW;EACpB;EACAA,WAAW,CAACrB,SAAS,CAACC,OAAO,CAAEL,EAAE,IAAK;IAAE;IACtCA,EAAE,CAACvB,KAAK,GAAG6C,sBAAsB,CAACtB,EAAE,CAACvB,KAAK,EAAEuB,EAAE,CAACxB,IAAI,EAAE0B,KAAK,CAAChB,GAAG,CAAC;IAC/Dc,EAAE,CAACvB,KAAK,GAAGoD,+BAA+B,CAAC7B,EAAE,CAACvB,KAAK,EAAEyB,KAAK,CAAChB,GAAG,EAAEc,EAAE,CAACxB,IAAI,CAAC;EAC1E,CAAC,CAAC;EAEFsD,cAAc,CAACL,WAAW,CAAC;EAC3BN,+BAA+B,CAACM,WAAW,CAAC;EAC5CR,kBAAkB,CAACQ,WAAW,CAAC;EAC/BP,uBAAuB,CAACO,WAAW,CAAC;EAGpC9C,oBAAoB,CAACuB,KAAK,EAAEuB,WAAW,CAAC,CAAC,CAAC;;EAE1C,OAAOA,WAAW;AACpB;AAEA,SAASE,sBAAsBA,CAACzB,KAAK,EAAE;EACrC,IAAI,CAACA,KAAK,CAACE,SAAS,IAAI,CAAC,KAAK,EAAE,KAAK,EAAE,KAAK,EAAE,KAAK,CAAC,CAACf,QAAQ,CAACa,KAAK,CAAChB,GAAG,CAAC,EAAE;IACxE,OAAO,IAAI;EACb;EACA,OAAO,KAAK;AACd","ignoreList":[]}
+{
+  "version": 3,
+  "sources": ["../src/normalizeFieldForComparison.js"],
+  "sourcesContent": ["/*\n  Note that this file contains very powerful normalizations and spells that are:\n  - meant for comparing similarity/mergability of two fields (clone, normalize, compare),\n  - and NOT for modifying the actual field!\n\n  This is mainly used by melinda-marc-record-merge-reducers. However, also removeInferiorDataFields fixer also used this.\n  Thus it is here. However, most of the testing is done via merge-reducers...\n*/\nimport clone from 'clone';\nimport {fieldStripPunctuation} from './punctuation2.js';\nimport {fieldToString, isControlSubfieldCode} from './utils.js';\n\nimport {fieldNormalizeControlNumbers/*, normalizeControlSubfieldValue*/} from './normalize-identifiers.js';\nimport createDebugLogger from 'debug';\nimport {normalizePartData, subfieldContainsPartData} from './normalizeSubfieldValueForComparison.js';\n\nconst debug = createDebugLogger('@natlibfi/melinda-marc-record-merge-reducers:normalizeFieldForComparison');\n//const debugData = debug.extend('data');\nconst debugDev = debug.extend('dev');\n\nexport function isEnnakkotietoSubfieldG(subfield) {\n  if (valuelessSubfield(subfield)) {\n    return false;\n  }\n  if (subfield.code !== 'g') {\n    return false;\n  }\n  return subfield.value.match(/^ENNAKKOTIETO\\.?$/gui);\n}\n\nfunction debugFieldComparison(oldField, newField) { // NB: Debug-only function!\n  /*\n  // We may drop certain subfields:\n  if (oldField.subfields.length === newField.subfields.length) {\n    oldField.subfields.forEach((subfield, index) => {\n      const newValue = newField.subfields[index].value;\n      if (subfield.value !== newValue) {\n        nvdebug(`NORMALIZE SUBFIELD: '${subfield.value}' => '${newValue}'`, debugDev);\n      }\n    });\n  }\n  */\n  const oldString = fieldToString(oldField);\n  const newString = fieldToString(newField);\n  if (oldString === newString) {\n    return;\n  }\n  //nvdebug(`NORMALIZE FIELD:\\n '${fieldToString(oldField)}' =>\\n '${fieldToString(newField)}'`, debugDev);\n}\n\nfunction containsHumanName(tag = '???', subfieldCode = undefined) {\n  // NB! This set is for bibs! Auth has 400... What else...\n  if (['100', '600', '700', '800'].includes(tag)) {\n    if (subfieldCode === undefined || subfieldCode === 'a') {\n      return true;\n    }\n  }\n  // Others?\n  return false;\n}\n\nfunction containsCorporateName(tag = '???', subfieldCode = undefined) {\n  // NB! This set is for bibs! Auth has 400... What else...\n  if (['110', '610', '710', '810'].includes(tag)) {\n    if (subfieldCode === undefined || subfieldCode === 'a') {\n      return true;\n    }\n  }\n  // Others?\n  return false;\n}\n\nfunction skipAllSubfieldNormalizations(value, subfieldCode, tag) {\n\n  if (isEnnakkotietoSubfieldG({'code': subfieldCode, value})) {\n    return true;\n  }\n\n  if (tag === '035' && ['a', 'z'].includes(subfieldCode)) { // A\n    return true;\n  }\n\n  if (isControlSubfieldCode(subfieldCode)) {\n    return true;\n  }\n  return false;\n}\n\nfunction skipSubfieldLowercase(value, subfieldCode, tag) {\n  // These may contain Roman Numerals...\n  if (subfieldContainsPartData(tag, subfieldCode)) {\n    return true;\n  }\n\n  return skipAllSubfieldNormalizations(value, subfieldCode, tag);\n}\n\nfunction skipAllFieldNormalizations(tag) {\n  if (['LOW', 'SID'].includes(tag)) {\n    return true;\n  }\n  return false;\n}\n\n\nfunction subfieldValueLowercase(value, subfieldCode, tag) {\n  if (skipSubfieldLowercase(value, subfieldCode, tag)) {\n    return value;\n  }\n\n  //return value.toLowerCase();\n  const newValue = value.toLowerCase();\n  if (newValue !== value) {\n    //nvdebug(`SVL ${tag} $${subfieldCode} '${value}' =>`, debugDev);\n    //nvdebug(`SVL ${tag} $${subfieldCode} '${newValue}'`, debugDev);\n    return newValue;\n  }\n  return value;\n}\n\nfunction subfieldLowercase(sf, tag) {\n  if (valuelessSubfield(sf)) {\n    return;\n  }\n  sf.value = subfieldValueLowercase(sf.value, sf.code, tag);\n}\n\nfunction fieldLowercase(field) {\n  if (skipFieldLowercase(field)) {\n    return;\n  }\n\n  field.subfields.forEach(sf => subfieldLowercase(sf, field.tag));\n\n  function skipFieldLowercase(field) {\n    if (skipAllFieldNormalizations(field.tag)) {\n      return true;\n    }\n    // Skip non-interesting fields\n    if (!containsHumanName(field.tag) && !containsCorporateName(field.tag) && !['240', '245', '630'].includes(field.tag)) {\n      return true;\n    }\n\n    return false;\n  }\n}\n\n\nfunction hack490SubfieldA(field) {\n  if (field.tag !== '490') {\n    return;\n  }\n  field.subfields.forEach(sf => removeSarja(sf));\n\n  // NB! This won't work, if the punctuation has not been stripped beforehand!\n  function removeSarja(subfield) {\n    if (valuelessSubfield(subfield)) {\n      return;\n    }\n\n    if (subfield.code !== 'a') {\n      return;\n    }\n    const tmp = subfield.value.replace(/ ?-(?:[a-z]|\u00E4|\u00F6)*sarja$/u, '');\n    if (tmp.length > 0) {\n      subfield.value = tmp;\n      return;\n    }\n  }\n}\n\nexport function tagAndSubfieldCodeReferToIsbn(tag, subfieldCode) {\n  // NB! We don't do this to 020$z!\n  if (subfieldCode === 'z' && ['765', '767', '770', '772', '773', '774', '776', '777', '780', '785', '786', '787'].includes(tag)) {\n    return true;\n  }\n  if (tag === '020' && subfieldCode === 'a') {\n    return true;\n  }\n  return false;\n}\n\nfunction looksLikeIsbn(value) {\n  // Does not check validity!\n  if (value.match(/^(?:[0-9]-?){9}(?:[0-9]-?[0-9]-?[0-9]-?)?[0-9Xx]$/u)) {\n    return true;\n  }\n  return false;\n}\n\nfunction normalizeISBN(field) {\n  if (!field.subfields) {\n    return;\n  }\n\n  //nvdebug(`ISBN-field? ${fieldToString(field)}`);\n  const relevantSubfields = field.subfields.filter(sf => tagAndSubfieldCodeReferToIsbn(field.tag, sf.code) && looksLikeIsbn(sf.value));\n  relevantSubfields.forEach(sf => normalizeIsbnSubfield(sf));\n\n  function normalizeIsbnSubfield(sf) {\n    if (valuelessSubfield(sf)) {\n      return;\n    }\n    //nvdebug(` ISBN-subfield? ${subfieldToString(sf)}`);\n    sf.value = sf.value.replace(/-/ug, '');\n    sf.value = sf.value.replace(/x/u, 'X');\n  }\n\n}\n\nfunction fieldSpecificHacks(field) {\n  normalizeISBN(field); // 020$a, not $z!\n  hack490SubfieldA(field);\n}\n\nexport function fieldTrimSubfieldValues(field) {\n  field.subfields?.forEach((sf) => {\n    if (valuelessSubfield(sf)) {\n      return;\n    }\n    sf.value = sf.value.replace(/^[ \\t\\n]+/u, '');\n    sf.value = sf.value.replace(/[ \\t\\n]+$/u, '');\n    sf.value = sf.value.replace(/[ \\t\\n]+/gu, ' ');\n  });\n}\n\nfunction fieldRemoveDecomposedDiacritics(field) {\n  // Raison d'\u00EAtre/motivation: \"Sir\u00E9n\" and diacriticless \"Siren\" might refer to a same surname, so this normalization\n  // allows us to compare authors and avoid duplicate fields.\n  field.subfields.forEach((sf) => {\n    if (valuelessSubfield(sf)) {\n        return;\n    }\n    sf.value = removeDecomposedDiacritics(sf.value);\n  });\n}\n\nfunction removeDecomposedDiacritics(value = '') {\n  // NB #1: Does nothing to precomposed letters. Do String.normalize('NFD') first, if you want to handle them.\n  // NB #2: Finnish letters '\u00E5', '\u00E4', '\u00F6', '\u00C5', \u00C4', and '\u00D6' should be handled (=precomposed) before calling this. (= keep them as is)\n  // NB #3: Calling our very own fixComposition() before this function handles both #1 and #2.\n  return String(value).replace(/\\p{Diacritic}/gu, '');\n}\n\nfunction normalizeSubfieldValue(value, subfieldCode, tag) {\n  // NB! For comparison of values only\n  /* eslint-disable */\n  value = subfieldValueLowercase(value, subfieldCode, tag);\n\n  // Normalize: s. = sivut = pp.\n  value = normalizePartData(value, subfieldCode, tag);\n  value = value.replace(/^\\[([^[\\]]+)\\]/gu, '$1'); // eslint-disable-line functional/immutable-data\n\n  if (['130', '730'].includes(tag) && subfieldCode === 'a') {\n    value = value.replace(' : ', ', '); // \"Halloween ends (elokuva, 2022)\" vs \"Halloween ends (elokuva : 2023)\"\n  }\n  /* eslint-enable */\n\n  // Not going to do these in the foreseeable future, but keeping them here for discussion:\n  // Possible normalizations include but are not limited to:\n  // \u00F8 => \u00F6? Might be language dependent: 041 $a fin => \u00F6, 041 $a eng => o?\n  // \u00D8 => \u00D6?\n  // \u00DF => ss\n  // \u00FE => th (NB! Both upper and lower case)\n  // ...\n  // Probably nots:\n  // \u00FC => y (probably not, though this correlates with Finnish letter-to-sound rules)\n  // w => v (OK for Finnish sorting in certain cases, but we are not here, are we?)\n  // I guess we should use decomposed values in code here. (Not sure what composition my examples above use.)\n  return value;\n}\n\nexport function cloneAndRemovePunctuation(field) {\n  const clonedField = clone(field);\n  if (fieldSkipNormalization(field)) {\n    return clonedField;\n  }\n  fieldStripPunctuation(clonedField);\n  fieldTrimSubfieldValues(clonedField);\n  debugDev('PUNC');\n  debugFieldComparison(field, clonedField);\n\n  return clonedField;\n}\n\nfunction removeCharsThatDontCarryMeaning(value, tag, subfieldCode) {\n  if (tag === '080') {\n    return value;\n  }\n\n  // 3\" refers to inches, but as this is for comparison only we don't mind...\n  value = value.replace(/['\u2018\u2019\"\u201E\u201C\u201D\u00AB\u00BB]/gu, ''); // MET-570 et al. Subset of https://hexdocs.pm/ex_unicode/Unicode.Category.QuoteMarks.html\n  // MRA-273: Handle X00$a name initials.\n  // NB #1: that we remove spaces for comparison (as it simpler), though actually space should be used. Doesn't matter as this is comparison only.\n  // NB #2: we might/should eventually write a validator/fixer that adds those spaces. After that point, this expection should become obsolete.\n  if (subfieldCode === 'a' && ['100', '400', '600', '700', '800'].includes(tag)) { // 400 is used in auth records. It's not a bib field at all.\n    value = value.replace(/([A-Z]|\u00C5|\u00C4|\u00D6)\\. +/ugi, '$1.');\n  }\n\n  return value;\n}\n\nfunction normalizeField(field) {\n  //sf.value = removeDecomposedDiacritics(sf.value);\n  fieldStripPunctuation(field);\n  fieldLowercase(field);\n  fieldNormalizeControlNumbers(field); // FIN11 vs FI-MELINDA etc.\n  return field;\n}\n\nexport function cloneAndNormalizeFieldForComparison(field) {\n  // NB! This new field is for comparison purposes only.\n  // Some of the normalizations might be considered a bit overkill for other purposes.\n  const clonedField = clone(field);\n  if (fieldSkipNormalization(field)) {\n    return clonedField;\n  }\n  clonedField.subfields.forEach((sf) => { // Do this for all fields or some fields?\n    if (valuelessSubfield(sf)) {\n      return;\n    }\n    sf.value = normalizeSubfieldValue(sf.value, sf.code, field.tag);\n    sf.value = removeCharsThatDontCarryMeaning(sf.value, field.tag, sf.code);\n  });\n\n  normalizeField(clonedField);\n  fieldRemoveDecomposedDiacritics(clonedField);\n  fieldSpecificHacks(clonedField);\n  fieldTrimSubfieldValues(clonedField);\n\n\n  debugFieldComparison(field, clonedField); // For debugging purposes only\n\n  return clonedField;\n}\n\nfunction fieldSkipNormalization(field) {\n  if (!field.subfields || ['018', '066', '080', '083'].includes(field.tag)) {\n    return true;\n  }\n  return false;\n}\n\nfunction valuelessSubfield(sf) {\n  return sf.value === undefined;\n}"],
+  "mappings": "AAQA,OAAO,WAAW;AAClB,SAAQ,6BAA4B;AACpC,SAAQ,eAAe,6BAA4B;AAEnD,SAAQ,oCAAsE;AAC9E,OAAO,uBAAuB;AAC9B,SAAQ,mBAAmB,gCAA+B;AAE1D,MAAM,QAAQ,kBAAkB,0EAA0E;AAE1G,MAAM,WAAW,MAAM,OAAO,KAAK;AAE5B,gBAAS,wBAAwB,UAAU;AAChD,MAAI,kBAAkB,QAAQ,GAAG;AAC/B,WAAO;AAAA,EACT;AACA,MAAI,SAAS,SAAS,KAAK;AACzB,WAAO;AAAA,EACT;AACA,SAAO,SAAS,MAAM,MAAM,sBAAsB;AACpD;AAEA,SAAS,qBAAqB,UAAU,UAAU;AAYhD,QAAM,YAAY,cAAc,QAAQ;AACxC,QAAM,YAAY,cAAc,QAAQ;AACxC,MAAI,cAAc,WAAW;AAC3B;AAAA,EACF;AAEF;AAEA,SAAS,kBAAkB,MAAM,OAAO,eAAe,QAAW;AAEhE,MAAI,CAAC,OAAO,OAAO,OAAO,KAAK,EAAE,SAAS,GAAG,GAAG;AAC9C,QAAI,iBAAiB,UAAa,iBAAiB,KAAK;AACtD,aAAO;AAAA,IACT;AAAA,EACF;AAEA,SAAO;AACT;AAEA,SAAS,sBAAsB,MAAM,OAAO,eAAe,QAAW;AAEpE,MAAI,CAAC,OAAO,OAAO,OAAO,KAAK,EAAE,SAAS,GAAG,GAAG;AAC9C,QAAI,iBAAiB,UAAa,iBAAiB,KAAK;AACtD,aAAO;AAAA,IACT;AAAA,EACF;AAEA,SAAO;AACT;AAEA,SAAS,8BAA8B,OAAO,cAAc,KAAK;AAE/D,MAAI,wBAAwB,EAAC,QAAQ,cAAc,MAAK,CAAC,GAAG;AAC1D,WAAO;AAAA,EACT;AAEA,MAAI,QAAQ,SAAS,CAAC,KAAK,GAAG,EAAE,SAAS,YAAY,GAAG;AACtD,WAAO;AAAA,EACT;AAEA,MAAI,sBAAsB,YAAY,GAAG;AACvC,WAAO;AAAA,EACT;AACA,SAAO;AACT;AAEA,SAAS,sBAAsB,OAAO,cAAc,KAAK;AAEvD,MAAI,yBAAyB,KAAK,YAAY,GAAG;AAC/C,WAAO;AAAA,EACT;AAEA,SAAO,8BAA8B,OAAO,cAAc,GAAG;AAC/D;AAEA,SAAS,2BAA2B,KAAK;AACvC,MAAI,CAAC,OAAO,KAAK,EAAE,SAAS,GAAG,GAAG;AAChC,WAAO;AAAA,EACT;AACA,SAAO;AACT;AAGA,SAAS,uBAAuB,OAAO,cAAc,KAAK;AACxD,MAAI,sBAAsB,OAAO,cAAc,GAAG,GAAG;AACnD,WAAO;AAAA,EACT;AAGA,QAAM,WAAW,MAAM,YAAY;AACnC,MAAI,aAAa,OAAO;AAGtB,WAAO;AAAA,EACT;AACA,SAAO;AACT;AAEA,SAAS,kBAAkB,IAAI,KAAK;AAClC,MAAI,kBAAkB,EAAE,GAAG;AACzB;AAAA,EACF;AACA,KAAG,QAAQ,uBAAuB,GAAG,OAAO,GAAG,MAAM,GAAG;AAC1D;AAEA,SAAS,eAAe,OAAO;AAC7B,MAAI,mBAAmB,KAAK,GAAG;AAC7B;AAAA,EACF;AAEA,QAAM,UAAU,QAAQ,QAAM,kBAAkB,IAAI,MAAM,GAAG,CAAC;AAE9D,WAAS,mBAAmBA,QAAO;AACjC,QAAI,2BAA2BA,OAAM,GAAG,GAAG;AACzC,aAAO;AAAA,IACT;AAEA,QAAI,CAAC,kBAAkBA,OAAM,GAAG,KAAK,CAAC,sBAAsBA,OAAM,GAAG,KAAK,CAAC,CAAC,OAAO,OAAO,KAAK,EAAE,SAASA,OAAM,GAAG,GAAG;AACpH,aAAO;AAAA,IACT;AAEA,WAAO;AAAA,EACT;AACF;AAGA,SAAS,iBAAiB,OAAO;AAC/B,MAAI,MAAM,QAAQ,OAAO;AACvB;AAAA,EACF;AACA,QAAM,UAAU,QAAQ,QAAM,YAAY,EAAE,CAAC;AAG7C,WAAS,YAAY,UAAU;AAC7B,QAAI,kBAAkB,QAAQ,GAAG;AAC/B;AAAA,IACF;AAEA,QAAI,SAAS,SAAS,KAAK;AACzB;AAAA,IACF;AACA,UAAM,MAAM,SAAS,MAAM,QAAQ,4BAA4B,EAAE;AACjE,QAAI,IAAI,SAAS,GAAG;AAClB,eAAS,QAAQ;AACjB;AAAA,IACF;AAAA,EACF;AACF;AAEO,gBAAS,8BAA8B,KAAK,cAAc;AAE/D,MAAI,iBAAiB,OAAO,CAAC,OAAO,OAAO,OAAO,OAAO,OAAO,OAAO,OAAO,OAAO,OAAO,OAAO,OAAO,KAAK,EAAE,SAAS,GAAG,GAAG;AAC9H,WAAO;AAAA,EACT;AACA,MAAI,QAAQ,SAAS,iBAAiB,KAAK;AACzC,WAAO;AAAA,EACT;AACA,SAAO;AACT;AAEA,SAAS,cAAc,OAAO;AAE5B,MAAI,MAAM,MAAM,oDAAoD,GAAG;AACrE,WAAO;AAAA,EACT;AACA,SAAO;AACT;AAEA,SAAS,cAAc,OAAO;AAC5B,MAAI,CAAC,MAAM,WAAW;AACpB;AAAA,EACF;AAGA,QAAM,oBAAoB,MAAM,UAAU,OAAO,QAAM,8BAA8B,MAAM,KAAK,GAAG,IAAI,KAAK,cAAc,GAAG,KAAK,CAAC;AACnI,oBAAkB,QAAQ,QAAM,sBAAsB,EAAE,CAAC;AAEzD,WAAS,sBAAsB,IAAI;AACjC,QAAI,kBAAkB,EAAE,GAAG;AACzB;AAAA,IACF;AAEA,OAAG,QAAQ,GAAG,MAAM,QAAQ,OAAO,EAAE;AACrC,OAAG,QAAQ,GAAG,MAAM,QAAQ,MAAM,GAAG;AAAA,EACvC;AAEF;AAEA,SAAS,mBAAmB,OAAO;AACjC,gBAAc,KAAK;AACnB,mBAAiB,KAAK;AACxB;AAEO,gBAAS,wBAAwB,OAAO;AAC7C,QAAM,WAAW,QAAQ,CAAC,OAAO;AAC/B,QAAI,kBAAkB,EAAE,GAAG;AACzB;AAAA,IACF;AACA,OAAG,QAAQ,GAAG,MAAM,QAAQ,cAAc,EAAE;AAC5C,OAAG,QAAQ,GAAG,MAAM,QAAQ,cAAc,EAAE;AAC5C,OAAG,QAAQ,GAAG,MAAM,QAAQ,cAAc,GAAG;AAAA,EAC/C,CAAC;AACH;AAEA,SAAS,gCAAgC,OAAO;AAG9C,QAAM,UAAU,QAAQ,CAAC,OAAO;AAC9B,QAAI,kBAAkB,EAAE,GAAG;AACvB;AAAA,IACJ;AACA,OAAG,QAAQ,2BAA2B,GAAG,KAAK;AAAA,EAChD,CAAC;AACH;AAEA,SAAS,2BAA2B,QAAQ,IAAI;AAI9C,SAAO,OAAO,KAAK,EAAE,QAAQ,mBAAmB,EAAE;AACpD;AAEA,SAAS,uBAAuB,OAAO,cAAc,KAAK;AAGxD,UAAQ,uBAAuB,OAAO,cAAc,GAAG;AAGvD,UAAQ,kBAAkB,OAAO,cAAc,GAAG;AAClD,UAAQ,MAAM,QAAQ,oBAAoB,IAAI;AAE9C,MAAI,CAAC,OAAO,KAAK,EAAE,SAAS,GAAG,KAAK,iBAAiB,KAAK;AACxD,YAAQ,MAAM,QAAQ,OAAO,IAAI;AAAA,EACnC;AAcA,SAAO;AACT;AAEO,gBAAS,0BAA0B,OAAO;AAC/C,QAAM,cAAc,MAAM,KAAK;AAC/B,MAAI,uBAAuB,KAAK,GAAG;AACjC,WAAO;AAAA,EACT;AACA,wBAAsB,WAAW;AACjC,0BAAwB,WAAW;AACnC,WAAS,MAAM;AACf,uBAAqB,OAAO,WAAW;AAEvC,SAAO;AACT;AAEA,SAAS,gCAAgC,OAAO,KAAK,cAAc;AACjE,MAAI,QAAQ,OAAO;AACjB,WAAO;AAAA,EACT;AAGA,UAAQ,MAAM,QAAQ,iBAAiB,EAAE;AAIzC,MAAI,iBAAiB,OAAO,CAAC,OAAO,OAAO,OAAO,OAAO,KAAK,EAAE,SAAS,GAAG,GAAG;AAC7E,YAAQ,MAAM,QAAQ,wBAAwB,KAAK;AAAA,EACrD;AAEA,SAAO;AACT;AAEA,SAAS,eAAe,OAAO;AAE7B,wBAAsB,KAAK;AAC3B,iBAAe,KAAK;AACpB,+BAA6B,KAAK;AAClC,SAAO;AACT;AAEO,gBAAS,oCAAoC,OAAO;AAGzD,QAAM,cAAc,MAAM,KAAK;AAC/B,MAAI,uBAAuB,KAAK,GAAG;AACjC,WAAO;AAAA,EACT;AACA,cAAY,UAAU,QAAQ,CAAC,OAAO;AACpC,QAAI,kBAAkB,EAAE,GAAG;AACzB;AAAA,IACF;AACA,OAAG,QAAQ,uBAAuB,GAAG,OAAO,GAAG,MAAM,MAAM,GAAG;AAC9D,OAAG,QAAQ,gCAAgC,GAAG,OAAO,MAAM,KAAK,GAAG,IAAI;AAAA,EACzE,CAAC;AAED,iBAAe,WAAW;AAC1B,kCAAgC,WAAW;AAC3C,qBAAmB,WAAW;AAC9B,0BAAwB,WAAW;AAGnC,uBAAqB,OAAO,WAAW;AAEvC,SAAO;AACT;AAEA,SAAS,uBAAuB,OAAO;AACrC,MAAI,CAAC,MAAM,aAAa,CAAC,OAAO,OAAO,OAAO,KAAK,EAAE,SAAS,MAAM,GAAG,GAAG;AACxE,WAAO;AAAA,EACT;AACA,SAAO;AACT;AAEA,SAAS,kBAAkB,IAAI;AAC7B,SAAO,GAAG,UAAU;AACtB;",
+  "names": ["field"]
+}