@natlibfi/marc-record-validators-melinda 10.12.0-alpha.4 → 10.13.0

package/dist/normalizeFieldForComparison.js

@@ -0,0 +1,292 @@
+"use strict";
+
+Object.defineProperty(exports, "__esModule", {
+  value: true
+});
+exports.cloneAndNormalizeFieldForComparison = cloneAndNormalizeFieldForComparison;
+exports.cloneAndRemovePunctuation = cloneAndRemovePunctuation;
+exports.fieldTrimSubfieldValues = fieldTrimSubfieldValues;
+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(obj) { return obj && obj.__esModule ? obj : { default: obj }; }
+/*
+  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 data!
+*/
+
+const debug = (0, _debug.default)('@natlibfi/melinda-marc-record-merge-reducers:normalize');
+//const debugData = debug.extend('data');
+const debugDev = debug.extend('dev');
+function debugFieldComparison(oldField, newField) {
+  // NB: Debug-only function!
+  /*
+  // We may drop certain subfields:
+  if (oldField.subfields.length === newField.subfields.length) { // eslint-disable-line functional/no-conditional-statements
+    oldField.subfields.forEach((subfield, index) => {
+      const newValue = newField.subfields[index].value;
+      if (subfield.value !== newValue) { // eslint-disable-line functional/no-conditional-statements
+        nvdebug(`NORMALIZE SUBFIELD: '${subfield.value}' => '${newValue}'`, debugDev);
+      }
+    });
+  }
+  */
+  const oldString = (0, _utils.fieldToString)(oldField);
+  const newString = (0, _utils.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') {
+      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') {
+      return true;
+    }
+  }
+  // Others?
+  return false;
+}
+function skipAllSubfieldNormalizations(value, subfieldCode, tag) {
+  if (subfieldCode === 'g' && value === 'ENNAKKOTIETO.') {
+    return true;
+  }
+  if (tag === '035' && ['a', 'z'].includes(subfieldCode)) {
+    // A
+    return true;
+  }
+  if ((0, _utils.isControlSubfieldCode)(subfieldCode)) {
+    return true;
+  }
+  return false;
+}
+function skipSubfieldLowercase(value, subfieldCode, tag) {
+  // These may contain Roman Numerals...
+  if ((0, _normalizeSubfieldValueForComparison.subfieldContainsPartData)(tag, subfieldCode)) {
+    return true;
+  }
+  return skipAllSubfieldNormalizations(value, subfieldCode, tag);
+}
+function skipAllFieldNormalizations(tag) {
+  if (['LOW', 'SID'].includes(tag)) {
+    return true;
+  }
+  return false;
+}
+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} '${newValue}'`, debugDev);
+    return newValue;
+  }
+  return value;
+}
+function subfieldLowercase(sf, tag) {
+  sf.value = subfieldValueLowercase(sf.value, sf.code, tag); // eslint-disable-line functional/immutable-data
+}
+
+function fieldLowercase(field) {
+  if (skipFieldLowercase(field)) {
+    return;
+  }
+  field.subfields.forEach(sf => subfieldLowercase(sf, field.tag));
+  function skipFieldLowercase(field) {
+    if (skipAllFieldNormalizations(field.tag)) {
+      return true;
+    }
+    // Skip non-interesting fields
+    if (!containsHumanName(field.tag) && !containsCorporateName(field.tag) && !['240', '245', '630'].includes(field.tag)) {
+      return true;
+    }
+    return false;
+  }
+}
+function hack490SubfieldA(field) {
+  if (field.tag !== '490') {
+    return;
+  }
+  field.subfields.forEach(sf => removeSarja(sf));
+
+  // NB! This won't work, if the punctuation has not been stripped beforehand!
+  function removeSarja(subfield) {
+    if (subfield.code !== 'a') {
+      return;
+    }
+    const tmp = subfield.value.replace(/ ?-(?:[a-z]|ä|ö)*sarja$/u, '');
+    if (tmp.length > 0) {
+      subfield.value = tmp; // eslint-disable-line functional/immutable-data
+      return;
+    }
+  }
+}
+function tagAndSubfieldCodeReferToIsbn(tag, subfieldCode) {
+  // NB! We don't do this to 020$z!
+  if (subfieldCode === 'z' && ['765', '767', '770', '772', '773', '774', '776', '777', '780', '785', '786', '787'].includes(tag)) {
+    return true;
+  }
+  if (tag === '020' && subfieldCode === 'a') {
+    return true;
+  }
+  return false;
+}
+function looksLikeIsbn(value) {
+  // Does not check validity!
+  if (value.match(/^(?:[0-9]-?){9}(?:[0-9]-?[0-9]-?[0-9]-?)?[0-9Xx]$/u)) {
+    return true;
+  }
+  return false;
+}
+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));
+  function normalizeIsbnSubfield(sf) {
+    //nvdebug(` ISBN-subfield? ${subfieldToString(sf)}`);
+    sf.value = sf.value.replace(/-/ug, ''); // eslint-disable-line functional/immutable-data
+    sf.value = sf.value.replace(/x/u, 'X'); // eslint-disable-line functional/immutable-data
+  }
+}
+
+function fieldSpecificHacks(field) {
+  normalizeISBN(field); // 020$a, not $z!
+  hack490SubfieldA(field);
+}
+function fieldTrimSubfieldValues(field) {
+  field.subfields?.forEach(sf => {
+    sf.value = sf.value.replace(/^[ \t\n]+/u, ''); // eslint-disable-line functional/immutable-data
+    sf.value = sf.value.replace(/[ \t\n]+$/u, ''); // eslint-disable-line functional/immutable-data
+    sf.value = sf.value.replace(/[ \t\n]+/gu, ' '); // eslint-disable-line functional/immutable-data
+  });
+}
+
+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 => {
+    sf.value = removeDecomposedDiacritics(sf.value); // eslint-disable-line functional/immutable-data
+  });
+}
+
+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 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, prefer-named-capture-group
+
+  if (['130', '730'].includes(tag) && subfieldCode === 'a') {
+    value = value.replace(' : ', ', '); // "Halloween ends (elokuva, 2022)" vs "Halloween ends (elokuva : 2023)"
+  }
+  /* 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 $a eng => o?
+  // Ø => Ö?
+  // ß => ss
+  // þ => th (NB! Both upper and lower case)
+  // ...
+  // Probably nots:
+  // ü => y (probably not, though this correlates with Finnish letter-to-sound rules)
+  // w => v (OK for Finnish sorting in certain cases, but we are not here, are we?)
+  // I guess we should use decomposed values in code here. (Not sure what composition my examples above use.)
+  return value;
+}
+function cloneAndRemovePunctuation(field) {
+  const clonedField = (0, _clone.default)(field);
+  if (fieldSkipNormalization(field)) {
+    return clonedField;
+  }
+  (0, _punctuation.fieldStripPunctuation)(clonedField);
+  fieldTrimSubfieldValues(clonedField);
+  debugDev('PUNC');
+  debugFieldComparison(field, clonedField);
+  return clonedField;
+}
+function removeCharsThatDontCarryMeaning(value, tag, subfieldCode) {
+  if (tag === '080') {
+    return value;
+  }
+  /* eslint-disable */
+  // 3" refers to inches, but as this is for comparison only we don't mind...
+  value = value.replace(/['"]/gu, '');
+  // MRA-273: Handle X00$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.');
+  }
+  /* eslint-enable */
+  return value;
+}
+function normalizeField(field) {
+  //sf.value = removeDecomposedDiacritics(sf.value); // eslint-disable-line functional/immutable-data
+  (0, _punctuation.fieldStripPunctuation)(field);
+  fieldLowercase(field);
+  (0, _normalizeIdentifiers.fieldNormalizeControlNumbers)(field); // FIN11 vs FI-MELINDA etc.
+  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);
+  if (fieldSkipNormalization(field)) {
+    return clonedField;
+  }
+  clonedField.subfields.forEach(sf => {
+    // Do this for all fields or some fields?
+    sf.value = normalizeSubfieldValue(sf.value, sf.code, field.tag); // eslint-disable-line functional/immutable-data
+    sf.value = removeCharsThatDontCarryMeaning(sf.value, field.tag, sf.code); // eslint-disable-line functional/immutable-data
+  });
+
+  normalizeField(clonedField); // eslint-disable-line functional/immutable-data
+  fieldRemoveDecomposedDiacritics(clonedField);
+  fieldSpecificHacks(clonedField);
+  fieldTrimSubfieldValues(clonedField);
+  debugFieldComparison(field, clonedField); // For debugging purposes only
+
+  return clonedField;
+}
+function fieldSkipNormalization(field) {
+  if (!field.subfields || ['018', '066', '080', '083'].includes(field.tag)) {
+    return true;
+  }
+  return false;
+}
+//# sourceMappingURL=normalizeFieldForComparison.js.map

package/dist/normalizeFieldForComparison.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"normalizeFieldForComparison.js","names":["_clone","_interopRequireDefault","require","_punctuation","_utils","_normalizeIdentifiers","_debug","_normalizeSubfieldValueForComparison","obj","__esModule","default","debug","createDebugLogger","debugDev","extend","debugFieldComparison","oldField","newField","oldString","fieldToString","newString","containsHumanName","tag","subfieldCode","undefined","includes","containsCorporateName","skipAllSubfieldNormalizations","value","isControlSubfieldCode","skipSubfieldLowercase","subfieldContainsPartData","skipAllFieldNormalizations","subfieldValueLowercase","newValue","toLowerCase","subfieldLowercase","sf","code","fieldLowercase","field","skipFieldLowercase","subfields","forEach","hack490SubfieldA","removeSarja","subfield","tmp","replace","length","tagAndSubfieldCodeReferToIsbn","looksLikeIsbn","match","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 data!\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:normalize');\n//const debugData = debug.extend('data');\nconst debugDev = debug.extend('dev');\n\nfunction debugFieldComparison(oldField, newField) { // NB: Debug-only function!\n  /*\n  // We may drop certain subfields:\n  if (oldField.subfields.length === newField.subfields.length) { // eslint-disable-line functional/no-conditional-statements\n    oldField.subfields.forEach((subfield, index) => {\n      const newValue = newField.subfields[index].value;\n      if (subfield.value !== newValue) { // eslint-disable-line functional/no-conditional-statements\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\n  if (subfieldCode === 'g' && value === 'ENNAKKOTIETO.') {\n    return true;\n  }\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} '${newValue}'`, debugDev);\n    return newValue;\n  }\n  return value;\n}\n\nfunction subfieldLowercase(sf, tag) {\n  sf.value = subfieldValueLowercase(sf.value, sf.code, tag); // eslint-disable-line functional/immutable-data\n}\n\nfunction fieldLowercase(field) {\n  if (skipFieldLowercase(field)) {\n    return;\n  }\n\n  field.subfields.forEach(sf => subfieldLowercase(sf, field.tag));\n\n  function skipFieldLowercase(field) {\n    if (skipAllFieldNormalizations(field.tag)) {\n      return true;\n    }\n    // Skip non-interesting fields\n    if (!containsHumanName(field.tag) && !containsCorporateName(field.tag) && !['240', '245', '630'].includes(field.tag)) {\n      return true;\n    }\n\n    return false;\n  }\n}\n\n\nfunction hack490SubfieldA(field) {\n  if (field.tag !== '490') {\n    return;\n  }\n  field.subfields.forEach(sf => removeSarja(sf));\n\n  // NB! This won't work, if the punctuation has not been stripped beforehand!\n  function removeSarja(subfield) {\n    if (subfield.code !== 'a') {\n      return;\n    }\n    const tmp = subfield.value.replace(/ ?-(?:[a-z]|ä|ö)*sarja$/u, '');\n    if (tmp.length > 0) {\n      subfield.value = tmp; // eslint-disable-line functional/immutable-data\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    //nvdebug(` ISBN-subfield? ${subfieldToString(sf)}`);\n    sf.value = sf.value.replace(/-/ug, ''); // eslint-disable-line functional/immutable-data\n    sf.value = sf.value.replace(/x/u, 'X'); // eslint-disable-line functional/immutable-data\n  }\n\n}\n\nfunction fieldSpecificHacks(field) {\n  normalizeISBN(field); // 020$a, not $z!\n  hack490SubfieldA(field);\n}\n\nexport function fieldTrimSubfieldValues(field) {\n  field.subfields?.forEach((sf) => {\n    sf.value = sf.value.replace(/^[ \\t\\n]+/u, ''); // eslint-disable-line functional/immutable-data\n    sf.value = sf.value.replace(/[ \\t\\n]+$/u, ''); // eslint-disable-line functional/immutable-data\n    sf.value = sf.value.replace(/[ \\t\\n]+/gu, ' '); // eslint-disable-line functional/immutable-data\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); // eslint-disable-line functional/immutable-data\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, prefer-named-capture-group\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 $a eng => o?\n  // Ø => Ö?\n  // ß => ss\n  // þ => th (NB! Both upper and lower case)\n  // ...\n  // Probably nots:\n  // ü => y (probably not, though this correlates with Finnish letter-to-sound rules)\n  // w => v (OK for Finnish sorting in certain cases, but we are not here, are we?)\n  // I guess we should use decomposed values in code here. (Not sure what composition my examples above use.)\n  return value;\n}\n\nexport function cloneAndRemovePunctuation(field) {\n  const clonedField = clone(field);\n  if (fieldSkipNormalization(field)) {\n    return clonedField;\n  }\n  fieldStripPunctuation(clonedField);\n  fieldTrimSubfieldValues(clonedField);\n  debugDev('PUNC');\n  debugFieldComparison(field, clonedField);\n\n  return clonedField;\n}\n\nfunction removeCharsThatDontCarryMeaning(value, tag, subfieldCode) {\n  if (tag === '080') {\n    return value;\n  }\n  /* eslint-disable */\n  // 3\" refers to inches, but as this is for comparison only we don't mind...\n  value = value.replace(/['\"]/gu, '');\n  // MRA-273: Handle X00$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  /* eslint-enable */\n  return value;\n}\n\nfunction normalizeField(field) {\n  //sf.value = removeDecomposedDiacritics(sf.value); // eslint-disable-line functional/immutable-data\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); // eslint-disable-line functional/immutable-data\n    sf.value = removeCharsThatDontCarryMeaning(sf.value, field.tag, sf.code);// eslint-disable-line functional/immutable-data\n  });\n\n  normalizeField(clonedField); // eslint-disable-line functional/immutable-data\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":";;;;;;;;;AAKA,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,GAAA,WAAAA,GAAA,IAAAA,GAAA,CAAAC,UAAA,GAAAD,GAAA,KAAAE,OAAA,EAAAF,GAAA;AAXlG;AACA;AACA;AACA;AACA;;AASA,MAAMG,KAAK,GAAG,IAAAC,cAAiB,EAAC,wDAAwD,CAAC;AACzF;AACA,MAAMC,QAAQ,GAAGF,KAAK,CAACG,MAAM,CAAC,KAAK,CAAC;AAEpC,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,CAACC,KAAK,EAAEL,YAAY,EAAED,GAAG,EAAE;EAG/D,IAAIC,YAAY,KAAK,GAAG,IAAIK,KAAK,KAAK,eAAe,EAAE;IACrD,OAAO,IAAI;EACb;EAGA,IAAIN,GAAG,KAAK,KAAK,IAAI,CAAC,GAAG,EAAE,GAAG,CAAC,CAACG,QAAQ,CAACF,YAAY,CAAC,EAAE;IAAE;IACxD,OAAO,IAAI;EACb;EAEA,IAAI,IAAAM,4BAAqB,EAACN,YAAY,CAAC,EAAE;IACvC,OAAO,IAAI;EACb;EACA,OAAO,KAAK;AACd;AAEA,SAASO,qBAAqBA,CAACF,KAAK,EAAEL,YAAY,EAAED,GAAG,EAAE;EACvD;EACA,IAAI,IAAAS,6DAAwB,EAACT,GAAG,EAAEC,YAAY,CAAC,EAAE;IAC/C,OAAO,IAAI;EACb;EAEA,OAAOI,6BAA6B,CAACC,KAAK,EAAEL,YAAY,EAAED,GAAG,CAAC;AAChE;AAEA,SAASU,0BAA0BA,CAACV,GAAG,EAAE;EACvC,IAAI,CAAC,KAAK,EAAE,KAAK,CAAC,CAACG,QAAQ,CAACH,GAAG,CAAC,EAAE;IAChC,OAAO,IAAI;EACb;EACA,OAAO,KAAK;AACd;AAGA,SAASW,sBAAsBA,CAACL,KAAK,EAAEL,YAAY,EAAED,GAAG,EAAE;EACxD,IAAIQ,qBAAqB,CAACF,KAAK,EAAEL,YAAY,EAAED,GAAG,CAAC,EAAE;IACnD,OAAOM,KAAK;EACd;;EAEA;EACA,MAAMM,QAAQ,GAAGN,KAAK,CAACO,WAAW,CAAC,CAAC;EACpC,IAAID,QAAQ,KAAKN,KAAK,EAAE;IACtB;IACA;IACA,OAAOM,QAAQ;EACjB;EACA,OAAON,KAAK;AACd;AAEA,SAASQ,iBAAiBA,CAACC,EAAE,EAAEf,GAAG,EAAE;EAClCe,EAAE,CAACT,KAAK,GAAGK,sBAAsB,CAACI,EAAE,CAACT,KAAK,EAAES,EAAE,CAACC,IAAI,EAAEhB,GAAG,CAAC,CAAC,CAAC;AAC7D;;AAEA,SAASiB,cAAcA,CAACC,KAAK,EAAE;EAC7B,IAAIC,kBAAkB,CAACD,KAAK,CAAC,EAAE;IAC7B;EACF;EAEAA,KAAK,CAACE,SAAS,CAACC,OAAO,CAACN,EAAE,IAAID,iBAAiB,CAACC,EAAE,EAAEG,KAAK,CAAClB,GAAG,CAAC,CAAC;EAE/D,SAASmB,kBAAkBA,CAACD,KAAK,EAAE;IACjC,IAAIR,0BAA0B,CAACQ,KAAK,CAAClB,GAAG,CAAC,EAAE;MACzC,OAAO,IAAI;IACb;IACA;IACA,IAAI,CAACD,iBAAiB,CAACmB,KAAK,CAAClB,GAAG,CAAC,IAAI,CAACI,qBAAqB,CAACc,KAAK,CAAClB,GAAG,CAAC,IAAI,CAAC,CAAC,KAAK,EAAE,KAAK,EAAE,KAAK,CAAC,CAACG,QAAQ,CAACe,KAAK,CAAClB,GAAG,CAAC,EAAE;MACpH,OAAO,IAAI;IACb;IAEA,OAAO,KAAK;EACd;AACF;AAGA,SAASsB,gBAAgBA,CAACJ,KAAK,EAAE;EAC/B,IAAIA,KAAK,CAAClB,GAAG,KAAK,KAAK,EAAE;IACvB;EACF;EACAkB,KAAK,CAACE,SAAS,CAACC,OAAO,CAACN,EAAE,IAAIQ,WAAW,CAACR,EAAE,CAAC,CAAC;;EAE9C;EACA,SAASQ,WAAWA,CAACC,QAAQ,EAAE;IAC7B,IAAIA,QAAQ,CAACR,IAAI,KAAK,GAAG,EAAE;MACzB;IACF;IACA,MAAMS,GAAG,GAAGD,QAAQ,CAAClB,KAAK,CAACoB,OAAO,CAAC,0BAA0B,EAAE,EAAE,CAAC;IAClE,IAAID,GAAG,CAACE,MAAM,GAAG,CAAC,EAAE;MAClBH,QAAQ,CAAClB,KAAK,GAAGmB,GAAG,CAAC,CAAC;MACtB;IACF;EACF;AACF;AAEO,SAASG,6BAA6BA,CAAC5B,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,SAAS4B,aAAaA,CAACvB,KAAK,EAAE;EAC5B;EACA,IAAIA,KAAK,CAACwB,KAAK,CAAC,oDAAoD,CAAC,EAAE;IACrE,OAAO,IAAI;EACb;EACA,OAAO,KAAK;AACd;AAEA,SAASC,aAAaA,CAACb,KAAK,EAAE;EAC5B,IAAI,CAACA,KAAK,CAACE,SAAS,EAAE;IACpB;EACF;;EAEA;EACA,MAAMY,iBAAiB,GAAGd,KAAK,CAACE,SAAS,CAACa,MAAM,CAAClB,EAAE,IAAIa,6BAA6B,CAACV,KAAK,CAAClB,GAAG,EAAEe,EAAE,CAACC,IAAI,CAAC,IAAIa,aAAa,CAACd,EAAE,CAACT,KAAK,CAAC,CAAC;EACpI0B,iBAAiB,CAACX,OAAO,CAACN,EAAE,IAAImB,qBAAqB,CAACnB,EAAE,CAAC,CAAC;EAE1D,SAASmB,qBAAqBA,CAACnB,EAAE,EAAE;IACjC;IACAA,EAAE,CAACT,KAAK,GAAGS,EAAE,CAACT,KAAK,CAACoB,OAAO,CAAC,KAAK,EAAE,EAAE,CAAC,CAAC,CAAC;IACxCX,EAAE,CAACT,KAAK,GAAGS,EAAE,CAACT,KAAK,CAACoB,OAAO,CAAC,IAAI,EAAE,GAAG,CAAC,CAAC,CAAC;EAC1C;AAEF;;AAEA,SAASS,kBAAkBA,CAACjB,KAAK,EAAE;EACjCa,aAAa,CAACb,KAAK,CAAC,CAAC,CAAC;EACtBI,gBAAgB,CAACJ,KAAK,CAAC;AACzB;AAEO,SAASkB,uBAAuBA,CAAClB,KAAK,EAAE;EAC7CA,KAAK,CAACE,SAAS,EAAEC,OAAO,CAAEN,EAAE,IAAK;IAC/BA,EAAE,CAACT,KAAK,GAAGS,EAAE,CAACT,KAAK,CAACoB,OAAO,CAAC,YAAY,EAAE,EAAE,CAAC,CAAC,CAAC;IAC/CX,EAAE,CAACT,KAAK,GAAGS,EAAE,CAACT,KAAK,CAACoB,OAAO,CAAC,YAAY,EAAE,EAAE,CAAC,CAAC,CAAC;IAC/CX,EAAE,CAACT,KAAK,GAAGS,EAAE,CAACT,KAAK,CAACoB,OAAO,CAAC,YAAY,EAAE,GAAG,CAAC,CAAC,CAAC;EAClD,CAAC,CAAC;AACJ;;AAEA,SAASW,+BAA+BA,CAACnB,KAAK,EAAE;EAC9C;EACA;EACAA,KAAK,CAACE,SAAS,CAACC,OAAO,CAAEN,EAAE,IAAK;IAC9BA,EAAE,CAACT,KAAK,GAAGgC,0BAA0B,CAACvB,EAAE,CAACT,KAAK,CAAC,CAAC,CAAC;EACnD,CAAC,CAAC;AACJ;;AAEA,SAASgC,0BAA0BA,CAAChC,KAAK,GAAG,EAAE,EAAE;EAC9C;EACA;EACA;EACA,OAAOiC,MAAM,CAACjC,KAAK,CAAC,CAACoB,OAAO,CAAC,iBAAiB,EAAE,EAAE,CAAC;AACrD;AAEA,SAASc,sBAAsBA,CAAClC,KAAK,EAAEL,YAAY,EAAED,GAAG,EAAE;EACxD;EACA;EACAM,KAAK,GAAGK,sBAAsB,CAACL,KAAK,EAAEL,YAAY,EAAED,GAAG,CAAC;;EAExD;EACAM,KAAK,GAAG,IAAAmC,sDAAiB,EAACnC,KAAK,EAAEL,YAAY,EAAED,GAAG,CAAC;EACnDM,KAAK,GAAGA,KAAK,CAACoB,OAAO,CAAC,kBAAkB,EAAE,IAAI,CAAC,CAAC,CAAC;;EAEjD,IAAI,CAAC,KAAK,EAAE,KAAK,CAAC,CAACvB,QAAQ,CAACH,GAAG,CAAC,IAAIC,YAAY,KAAK,GAAG,EAAE;IACxDK,KAAK,GAAGA,KAAK,CAACoB,OAAO,CAAC,KAAK,EAAE,IAAI,CAAC,CAAC,CAAC;EACtC;EACA;;EAEA;EACA;EACA;EACA;EACA;EACA;EACA;EACA;EACA;EACA;EACA;EACA,OAAOpB,KAAK;AACd;AAEO,SAASoC,yBAAyBA,CAACxB,KAAK,EAAE;EAC/C,MAAMyB,WAAW,GAAG,IAAAC,cAAK,EAAC1B,KAAK,CAAC;EAChC,IAAI2B,sBAAsB,CAAC3B,KAAK,CAAC,EAAE;IACjC,OAAOyB,WAAW;EACpB;EACA,IAAAG,kCAAqB,EAACH,WAAW,CAAC;EAClCP,uBAAuB,CAACO,WAAW,CAAC;EACpCpD,QAAQ,CAAC,MAAM,CAAC;EAChBE,oBAAoB,CAACyB,KAAK,EAAEyB,WAAW,CAAC;EAExC,OAAOA,WAAW;AACpB;AAEA,SAASI,+BAA+BA,CAACzC,KAAK,EAAEN,GAAG,EAAEC,YAAY,EAAE;EACjE,IAAID,GAAG,KAAK,KAAK,EAAE;IACjB,OAAOM,KAAK;EACd;EACA;EACA;EACAA,KAAK,GAAGA,KAAK,CAACoB,OAAO,CAAC,QAAQ,EAAE,EAAE,CAAC;EACnC;EACA;EACA;EACA,IAAIzB,YAAY,KAAK,GAAG,IAAI,CAAC,KAAK,EAAE,KAAK,EAAE,KAAK,EAAE,KAAK,EAAE,KAAK,CAAC,CAACE,QAAQ,CAACH,GAAG,CAAC,EAAE;IAAE;IAC/EM,KAAK,GAAGA,KAAK,CAACoB,OAAO,CAAC,sBAAsB,EAAE,KAAK,CAAC;EACtD;EACA;EACA,OAAOpB,KAAK;AACd;AAEA,SAAS0C,cAAcA,CAAC9B,KAAK,EAAE;EAC7B;EACA,IAAA4B,kCAAqB,EAAC5B,KAAK,CAAC;EAC5BD,cAAc,CAACC,KAAK,CAAC;EACrB,IAAA+B,kDAA4B,EAAC/B,KAAK,CAAC,CAAC,CAAC;EACrC,OAAOA,KAAK;AACd;AAEO,SAASgC,mCAAmCA,CAAChC,KAAK,EAAE;EACzD;EACA;EACA,MAAMyB,WAAW,GAAG,IAAAC,cAAK,EAAC1B,KAAK,CAAC;EAChC,IAAI2B,sBAAsB,CAAC3B,KAAK,CAAC,EAAE;IACjC,OAAOyB,WAAW;EACpB;EACAA,WAAW,CAACvB,SAAS,CAACC,OAAO,CAAEN,EAAE,IAAK;IAAE;IACtCA,EAAE,CAACT,KAAK,GAAGkC,sBAAsB,CAACzB,EAAE,CAACT,KAAK,EAAES,EAAE,CAACC,IAAI,EAAEE,KAAK,CAAClB,GAAG,CAAC,CAAC,CAAC;IACjEe,EAAE,CAACT,KAAK,GAAGyC,+BAA+B,CAAChC,EAAE,CAACT,KAAK,EAAEY,KAAK,CAAClB,GAAG,EAAEe,EAAE,CAACC,IAAI,CAAC,CAAC;EAC3E,CAAC,CAAC;;EAEFgC,cAAc,CAACL,WAAW,CAAC,CAAC,CAAC;EAC7BN,+BAA+B,CAACM,WAAW,CAAC;EAC5CR,kBAAkB,CAACQ,WAAW,CAAC;EAC/BP,uBAAuB,CAACO,WAAW,CAAC;EAGpClD,oBAAoB,CAACyB,KAAK,EAAEyB,WAAW,CAAC,CAAC,CAAC;;EAE1C,OAAOA,WAAW;AACpB;AAEA,SAASE,sBAAsBA,CAAC3B,KAAK,EAAE;EACrC,IAAI,CAACA,KAAK,CAACE,SAAS,IAAI,CAAC,KAAK,EAAE,KAAK,EAAE,KAAK,EAAE,KAAK,CAAC,CAACjB,QAAQ,CAACe,KAAK,CAAClB,GAAG,CAAC,EAAE;IACxE,OAAO,IAAI;EACb;EACA,OAAO,KAAK;AACd"}

package/dist/normalizeSubfieldValueForComparison.js

@@ -0,0 +1,103 @@
+"use strict";
+
+Object.defineProperty(exports, "__esModule", {
+  value: true
+});
+exports.normalizePartData = normalizePartData;
+exports.partsAgree = partsAgree;
+exports.subfieldContainsPartData = subfieldContainsPartData;
+var _utils = require("./utils");
+var _debug = _interopRequireDefault(require("debug"));
+function _interopRequireDefault(obj) { return obj && obj.__esModule ? obj : { default: obj }; }
+// Normalizes at least 490$v and 773$g which contain information such as "Raita 5" vs "5", and "Osa 3" vs "Osa III".
+
+const debug = (0, _debug.default)('@natlibfi/melinda-marc-record-merge-reducers:normalizePart');
+//const debugData = debug.extend('data');
+const debugDev = debug.extend('dev');
+function subfieldContainsPartData(tag, subfieldCode) {
+  if (subfieldCode === 'v' && ['490', '800', '810', '811', '830'].includes(tag)) {
+    return true;
+  }
+  if (tag === '773' && subfieldCode === 'g') {
+    return true;
+  }
+  return false;
+}
+function splitPartData(originalValue) {
+  const value = originalValue.replace(/^\[([0-9]+)\][-.,:; ]*$/ui, '$1'); // eslint-disable-line prefer-named-capture-group
+  const splitPoint = value.lastIndexOf(' ');
+  if (splitPoint === -1) {
+    return [undefined, value];
+  }
+  const lhs = value.substr(0, splitPoint);
+  const rhs = value.substr(splitPoint + 1);
+  return [lhs, rhs];
+}
+function normalizePartType(originalValue) {
+  if (originalValue === undefined) {
+    return undefined;
+  }
+  const value = originalValue.toLowerCase();
+  // Return Finnish singular nominative. Best-ish for debug purposes...
+  if (['osa', 'part', 'teil'].includes(value)) {
+    return 'osa';
+  }
+  if (['p.', 'page', 'pages', 'pp.', 's.', 'sidor', 'sivu', 'sivut'].includes(value)) {
+    return 'sivu';
+  }
+  return value;
+}
+const romanNumbers = {
+  'I': '1',
+  'II': '2',
+  'III': '3',
+  'IV': '4',
+  'V': '5',
+  'VI': '6',
+  'X': '10'
+};
+function normalizePartNumber(value) {
+  // Should we handle all Roman numbers or some range of them?
+  // There's probably a library for our purposes..
+  if (value in romanNumbers) {
+    const arabicValue = romanNumbers[value];
+    (0, _utils.nvdebug)(` MAP ${value} to ${arabicValue}`, debugDev);
+    return arabicValue;
+  }
+  return value.toLowerCase();
+}
+function splitAndNormalizePartData(value) {
+  // This is just a stub. Does not handle eg. "Levy 2, raita 15"
+  const [lhs, rhs] = splitPartData(value);
+  (0, _utils.nvdebug)(`  LHS: '${lhs}'`, debugDev);
+  (0, _utils.nvdebug)(`  RHS: '${rhs}'`, debugDev);
+  const partType = normalizePartType(lhs);
+  const partNumber = normalizePartNumber(rhs);
+  return [partType, partNumber];
+}
+function partsAgree(value1, value2, tag, subfieldCode) {
+  if (!subfieldContainsPartData(tag, subfieldCode)) {
+    return false;
+  }
+  const [partType1, partNumber1] = splitAndNormalizePartData(value1);
+  const [partType2, partNumber2] = splitAndNormalizePartData(value2);
+  if (partNumber1 !== partNumber2) {
+    return false;
+  }
+  if (partType1 === undefined || partType2 === undefined || partType1 === partType2) {
+    return true;
+  }
+  return false;
+}
+function normalizePartData(value, subfieldCode, tag) {
+  // This is for normalizing values for equality comparison only!
+  if (!subfieldContainsPartData(tag, subfieldCode)) {
+    return value;
+  }
+  const [partType, partNumber] = splitAndNormalizePartData(value);
+  if (partType === undefined) {
+    return partNumber;
+  }
+  return `${partType} ${partNumber}`;
+}
+//# sourceMappingURL=normalizeSubfieldValueForComparison.js.map

package/dist/normalizeSubfieldValueForComparison.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"normalizeSubfieldValueForComparison.js","names":["_utils","require","_debug","_interopRequireDefault","obj","__esModule","default","debug","createDebugLogger","debugDev","extend","subfieldContainsPartData","tag","subfieldCode","includes","splitPartData","originalValue","value","replace","splitPoint","lastIndexOf","undefined","lhs","substr","rhs","normalizePartType","toLowerCase","romanNumbers","normalizePartNumber","arabicValue","nvdebug","splitAndNormalizePartData","partType","partNumber","partsAgree","value1","value2","partType1","partNumber1","partType2","partNumber2","normalizePartData"],"sources":["../src/normalizeSubfieldValueForComparison.js"],"sourcesContent":["import {nvdebug} from './utils';\nimport createDebugLogger from 'debug';\n\n// Normalizes at least 490$v and 773$g which contain information such as \"Raita 5\" vs \"5\", and \"Osa 3\" vs \"Osa III\".\n\nconst debug = createDebugLogger('@natlibfi/melinda-marc-record-merge-reducers:normalizePart');\n//const debugData = debug.extend('data');\nconst debugDev = debug.extend('dev');\n\nexport function subfieldContainsPartData(tag, subfieldCode) {\n  if (subfieldCode === 'v' && ['490', '800', '810', '811', '830'].includes(tag)) {\n    return true;\n  }\n  if (tag === '773' && subfieldCode === 'g') {\n    return true;\n  }\n  return false;\n}\n\nfunction splitPartData(originalValue) {\n  const value = originalValue.replace(/^\\[([0-9]+)\\][-.,:; ]*$/ui, '$1'); // eslint-disable-line prefer-named-capture-group\n  const splitPoint = value.lastIndexOf(' ');\n  if (splitPoint === -1) {\n    return [undefined, value];\n  }\n  const lhs = value.substr(0, splitPoint);\n  const rhs = value.substr(splitPoint + 1);\n  return [lhs, rhs];\n}\n\nfunction normalizePartType(originalValue) {\n  if (originalValue === undefined) {\n    return undefined;\n  }\n  const value = originalValue.toLowerCase();\n  // Return Finnish singular nominative. Best-ish for debug purposes...\n  if (['osa', 'part', 'teil'].includes(value)) {\n    return 'osa';\n  }\n  if (['p.', 'page', 'pages', 'pp.', 's.', 'sidor', 'sivu', 'sivut'].includes(value)) {\n    return 'sivu';\n  }\n  return value;\n}\n\nconst romanNumbers = {'I': '1', 'II': '2', 'III': '3', 'IV': '4', 'V': '5', 'VI': '6', 'X': '10'};\n\nfunction normalizePartNumber(value) {\n  // Should we handle all Roman numbers or some range of them?\n  // There's probably a library for our purposes..\n  if (value in romanNumbers) {\n    const arabicValue = romanNumbers[value];\n    nvdebug(` MAP ${value} to ${arabicValue}`, debugDev);\n    return arabicValue;\n  }\n  return value.toLowerCase();\n}\n\nfunction splitAndNormalizePartData(value) {\n  // This is just a stub. Does not handle eg. \"Levy 2, raita 15\"\n  const [lhs, rhs] = splitPartData(value);\n  nvdebug(`  LHS: '${lhs}'`, debugDev);\n  nvdebug(`  RHS: '${rhs}'`, debugDev);\n  const partType = normalizePartType(lhs);\n  const partNumber = normalizePartNumber(rhs);\n  return [partType, partNumber];\n}\n\nexport function partsAgree(value1, value2, tag, subfieldCode) {\n  if (!subfieldContainsPartData(tag, subfieldCode)) {\n    return false;\n  }\n  const [partType1, partNumber1] = splitAndNormalizePartData(value1);\n  const [partType2, partNumber2] = splitAndNormalizePartData(value2);\n  if (partNumber1 !== partNumber2) {\n    return false;\n  }\n  if (partType1 === undefined || partType2 === undefined || partType1 === partType2) {\n    return true;\n  }\n\n  return false;\n}\n\nexport function normalizePartData(value, subfieldCode, tag) {\n  // This is for normalizing values for equality comparison only!\n  if (!subfieldContainsPartData(tag, subfieldCode)) {\n    return value;\n  }\n\n  const [partType, partNumber] = splitAndNormalizePartData(value);\n  if (partType === undefined) {\n    return partNumber;\n  }\n  return `${partType} ${partNumber}`;\n}\n"],"mappings":";;;;;;;;AAAA,IAAAA,MAAA,GAAAC,OAAA;AACA,IAAAC,MAAA,GAAAC,sBAAA,CAAAF,OAAA;AAAsC,SAAAE,uBAAAC,GAAA,WAAAA,GAAA,IAAAA,GAAA,CAAAC,UAAA,GAAAD,GAAA,KAAAE,OAAA,EAAAF,GAAA;AAEtC;;AAEA,MAAMG,KAAK,GAAG,IAAAC,cAAiB,EAAC,4DAA4D,CAAC;AAC7F;AACA,MAAMC,QAAQ,GAAGF,KAAK,CAACG,MAAM,CAAC,KAAK,CAAC;AAE7B,SAASC,wBAAwBA,CAACC,GAAG,EAAEC,YAAY,EAAE;EAC1D,IAAIA,YAAY,KAAK,GAAG,IAAI,CAAC,KAAK,EAAE,KAAK,EAAE,KAAK,EAAE,KAAK,EAAE,KAAK,CAAC,CAACC,QAAQ,CAACF,GAAG,CAAC,EAAE;IAC7E,OAAO,IAAI;EACb;EACA,IAAIA,GAAG,KAAK,KAAK,IAAIC,YAAY,KAAK,GAAG,EAAE;IACzC,OAAO,IAAI;EACb;EACA,OAAO,KAAK;AACd;AAEA,SAASE,aAAaA,CAACC,aAAa,EAAE;EACpC,MAAMC,KAAK,GAAGD,aAAa,CAACE,OAAO,CAAC,2BAA2B,EAAE,IAAI,CAAC,CAAC,CAAC;EACxE,MAAMC,UAAU,GAAGF,KAAK,CAACG,WAAW,CAAC,GAAG,CAAC;EACzC,IAAID,UAAU,KAAK,CAAC,CAAC,EAAE;IACrB,OAAO,CAACE,SAAS,EAAEJ,KAAK,CAAC;EAC3B;EACA,MAAMK,GAAG,GAAGL,KAAK,CAACM,MAAM,CAAC,CAAC,EAAEJ,UAAU,CAAC;EACvC,MAAMK,GAAG,GAAGP,KAAK,CAACM,MAAM,CAACJ,UAAU,GAAG,CAAC,CAAC;EACxC,OAAO,CAACG,GAAG,EAAEE,GAAG,CAAC;AACnB;AAEA,SAASC,iBAAiBA,CAACT,aAAa,EAAE;EACxC,IAAIA,aAAa,KAAKK,SAAS,EAAE;IAC/B,OAAOA,SAAS;EAClB;EACA,MAAMJ,KAAK,GAAGD,aAAa,CAACU,WAAW,CAAC,CAAC;EACzC;EACA,IAAI,CAAC,KAAK,EAAE,MAAM,EAAE,MAAM,CAAC,CAACZ,QAAQ,CAACG,KAAK,CAAC,EAAE;IAC3C,OAAO,KAAK;EACd;EACA,IAAI,CAAC,IAAI,EAAE,MAAM,EAAE,OAAO,EAAE,KAAK,EAAE,IAAI,EAAE,OAAO,EAAE,MAAM,EAAE,OAAO,CAAC,CAACH,QAAQ,CAACG,KAAK,CAAC,EAAE;IAClF,OAAO,MAAM;EACf;EACA,OAAOA,KAAK;AACd;AAEA,MAAMU,YAAY,GAAG;EAAC,GAAG,EAAE,GAAG;EAAE,IAAI,EAAE,GAAG;EAAE,KAAK,EAAE,GAAG;EAAE,IAAI,EAAE,GAAG;EAAE,GAAG,EAAE,GAAG;EAAE,IAAI,EAAE,GAAG;EAAE,GAAG,EAAE;AAAI,CAAC;AAEjG,SAASC,mBAAmBA,CAACX,KAAK,EAAE;EAClC;EACA;EACA,IAAIA,KAAK,IAAIU,YAAY,EAAE;IACzB,MAAME,WAAW,GAAGF,YAAY,CAACV,KAAK,CAAC;IACvC,IAAAa,cAAO,EAAE,QAAOb,KAAM,OAAMY,WAAY,EAAC,EAAEpB,QAAQ,CAAC;IACpD,OAAOoB,WAAW;EACpB;EACA,OAAOZ,KAAK,CAACS,WAAW,CAAC,CAAC;AAC5B;AAEA,SAASK,yBAAyBA,CAACd,KAAK,EAAE;EACxC;EACA,MAAM,CAACK,GAAG,EAAEE,GAAG,CAAC,GAAGT,aAAa,CAACE,KAAK,CAAC;EACvC,IAAAa,cAAO,EAAE,WAAUR,GAAI,GAAE,EAAEb,QAAQ,CAAC;EACpC,IAAAqB,cAAO,EAAE,WAAUN,GAAI,GAAE,EAAEf,QAAQ,CAAC;EACpC,MAAMuB,QAAQ,GAAGP,iBAAiB,CAACH,GAAG,CAAC;EACvC,MAAMW,UAAU,GAAGL,mBAAmB,CAACJ,GAAG,CAAC;EAC3C,OAAO,CAACQ,QAAQ,EAAEC,UAAU,CAAC;AAC/B;AAEO,SAASC,UAAUA,CAACC,MAAM,EAAEC,MAAM,EAAExB,GAAG,EAAEC,YAAY,EAAE;EAC5D,IAAI,CAACF,wBAAwB,CAACC,GAAG,EAAEC,YAAY,CAAC,EAAE;IAChD,OAAO,KAAK;EACd;EACA,MAAM,CAACwB,SAAS,EAAEC,WAAW,CAAC,GAAGP,yBAAyB,CAACI,MAAM,CAAC;EAClE,MAAM,CAACI,SAAS,EAAEC,WAAW,CAAC,GAAGT,yBAAyB,CAACK,MAAM,CAAC;EAClE,IAAIE,WAAW,KAAKE,WAAW,EAAE;IAC/B,OAAO,KAAK;EACd;EACA,IAAIH,SAAS,KAAKhB,SAAS,IAAIkB,SAAS,KAAKlB,SAAS,IAAIgB,SAAS,KAAKE,SAAS,EAAE;IACjF,OAAO,IAAI;EACb;EAEA,OAAO,KAAK;AACd;AAEO,SAASE,iBAAiBA,CAACxB,KAAK,EAAEJ,YAAY,EAAED,GAAG,EAAE;EAC1D;EACA,IAAI,CAACD,wBAAwB,CAACC,GAAG,EAAEC,YAAY,CAAC,EAAE;IAChD,OAAOI,KAAK;EACd;EAEA,MAAM,CAACe,QAAQ,EAAEC,UAAU,CAAC,GAAGF,yBAAyB,CAACd,KAAK,CAAC;EAC/D,IAAIe,QAAQ,KAAKX,SAAS,EAAE;IAC1B,OAAOY,UAAU;EACnB;EACA,OAAQ,GAAED,QAAS,IAAGC,UAAW,EAAC;AACpC"}

package/dist/punctuation2.js

@@ -232,6 +232,20 @@ const linkingEntryWhatever = [{
   'followedBy': 'abdghiklmnopqrstuwxyz',
   'remove': /\. -$/u
 }];
+
+// '!' means negation, thus '!b' means any other subfield but 'b'.
+// 'followedBy': '#' means that current subfield is the last subfield.
+// NB! Note that control subfields are ignored in punctuation rules.
+// NB #2! Control field ignorance causes issues with field 257: https://wiki.helsinki.fi/display/rdasovellusohje/Loppupisteohje
+//        Might need to work on that at some point. NOT a top priority though.
+// NB #3! Final punctuation creation is/should be handled by endind-punctuation.js validator!
+
+const crappy246 = [{
+  'code': 'abfghinp',
+  'followedBy': '#',
+  'remove': /\.$/u,
+  'context': dotIsProbablyPunc
+}];
 const cleanCrappyPunctuationRules = {
   '100': removeX00Whatever,
   '110': removeX10Whatever,
@@ -245,6 +259,7 @@ const cleanCrappyPunctuationRules = {
     'remove': /\.$/u,
     'context': dotIsProbablyPunc
   }],
+  '246': crappy246,
   '300': [{
     'code': 'a',
     'followedBy': '!b',
@@ -280,7 +295,8 @@ const cleanCrappyPunctuationRules = {
   '776': linkingEntryWhatever,
   '800': removeX00Whatever,
   '810': removeX10Whatever,
-  '830': remove490And830Whatever
+  '830': remove490And830Whatever,
+  '946': crappy246
 };
 const cleanLegalX00Comma = {
   'code': 'abcde',
@@ -343,7 +359,7 @@ const cleanLegalSeriesTitle = [
 const clean24X = [{
   'name': 'I:A',
   'code': 'i',
-  'followedBy': 'i',
+  'followedBy': 'a',
   'remove': / *:$/u
 }, {
   'name': 'A:B',
@@ -614,7 +630,12 @@ const addPairedPunctuationRules = {
   '800': addX00,
   '810': addX10,
   '830': addSeriesTitle,
-  '946': add246
+  '946': [{
+    'code': 'i',
+    'followedBy': 'a',
+    'add': ':',
+    'context': defaultNeedsPuncAfter
+  }]
 };
 function ruleAppliesToSubfieldCode(targetSubfieldCodes, currSubfieldCode) {
   const negation = targetSubfieldCodes.includes('!');