@natlibfi/marc-record-validators-melinda 12.1.0-alpha.1 → 13.0.0

package/dist/normalize-dashes.js

@@ -1,5 +1,8 @@
 import clone from "clone";
-import { fieldToString, isControlSubfieldCode, nvdebug } from "./utils.js";
+import { fieldToString, isContentSubfieldCode, nvdebug } from "./utils.js";
+import createDebugLogger from "debug";
+const debug = createDebugLogger("@natlibfi/marc-record-validators-melinda:normaliza-dashes");
+const debugDev = debug.extend("dev");
 export default function() {
   return {
     description: 'Normalize various dashes to "-"',
@@ -7,7 +10,7 @@ export default function() {
     fix
   };
   function fix(record) {
-    nvdebug(`FIX ME`);
+    nvdebug(`FIX ME`, debugDev);
     record.fields.forEach((field) => {
       fixDashes(field);
     });
@@ -16,7 +19,7 @@ export default function() {
   }
   function validate(record) {
     const res = { message: [] };
-    nvdebug(`VALIDATE ME`);
+    nvdebug(`VALIDATE ME`, debugDev);
     record.fields?.forEach((field) => {
       validateField(field, res);
     });
@@ -25,7 +28,7 @@ export default function() {
   }
   function validateField(field, res) {
     const orig = fieldToString(field);
-    nvdebug(` VALIDATE FIELD '${orig}'`);
+    nvdebug(` VALIDATE FIELD '${orig}'`, debugDev);
     const normalizedField = fixDashes(clone(field));
     const mod = fieldToString(normalizedField);
     if (orig === mod) {
@@ -39,11 +42,11 @@ function fixDashes(field) {
   if (!field.subfields) {
     return field;
   }
-  nvdebug(`Dashing ${fieldToString(field)}`);
+  nvdebug(`Dashing ${fieldToString(field)}`, debugDev);
   field.subfields.forEach((sf) => subfieldFixDashes(sf));
   return field;
   function subfieldFixDashes(subfield) {
-    if (isControlSubfieldCode(subfield.code)) {
+    if (!isContentSubfieldCode(subfield.code, field.tag)) {
       return;
     }
     subfield.value = subfield.value.replace(/[\u2010\u2011\u2012\u2013\u2014\u2015]/ug, "-");

package/dist/normalize-dashes.js.map

@@ -1,7 +1,7 @@
 {
   "version": 3,
   "sources": ["../src/normalize-dashes.js"],
-  "sourcesContent": ["//import createDebugLogger from 'debug';\nimport clone from 'clone';\nimport {fieldToString, isControlSubfieldCode, nvdebug} from './utils.js';\n\n// Author(s): Nicholas Volk\nexport default function () {\n\n  return {\n    description: 'Normalize various dashes to \"-\"',\n    validate, fix\n  };\n\n  function fix(record) {\n    nvdebug(`FIX ME`);\n    record.fields.forEach(field => {\n      fixDashes(field);\n    });\n\n    const res = {message: [], fix: [], valid: true};\n    return res;\n  }\n\n  function validate(record) {\n    const res = {message: []};\n\n    nvdebug(`VALIDATE ME`);\n    record.fields?.forEach(field => {\n      validateField(field, res);\n    });\n\n    res.valid = !(res.message.length >= 1);\n    return res;\n  }\n\n  function validateField(field, res) {\n    const orig = fieldToString(field);\n    nvdebug(` VALIDATE FIELD '${orig}'`);\n\n    const normalizedField = fixDashes(clone(field));\n    const mod = fieldToString(normalizedField);\n    if (orig === mod) { // Fail as the input is \"broken\"/\"crap\"/sumthing\n      return;\n    }\n    res.message.push(`'TODO: ${orig}' => '${mod}'`);\n    return;\n  }\n}\n\n\nfunction fixDashes(field) {\n  if (!field.subfields) {\n    return field;\n  }\n\n  nvdebug(`Dashing ${fieldToString(field)}`);\n\n  field.subfields.forEach(sf => subfieldFixDashes(sf));\n\n  return field;\n\n  function subfieldFixDashes(subfield) {\n    if (isControlSubfieldCode(subfield.code)) {\n      return;\n    }\n    // Normalize dashes U+2010 ... U+2015 to '-':\n    subfield.value = subfield.value.replace(/[\\u2010\\u2011\\u2012\\u2013\\u2014\\u2015]/ug, '-');\n  }\n\n}\n"],
-  "mappings": "AACA,OAAO,WAAW;AAClB,SAAQ,eAAe,uBAAuB,eAAc;AAG5D,0BAA2B;AAEzB,SAAO;AAAA,IACL,aAAa;AAAA,IACb;AAAA,IAAU;AAAA,EACZ;AAEA,WAAS,IAAI,QAAQ;AACnB,YAAQ,QAAQ;AAChB,WAAO,OAAO,QAAQ,WAAS;AAC7B,gBAAU,KAAK;AAAA,IACjB,CAAC;AAED,UAAM,MAAM,EAAC,SAAS,CAAC,GAAG,KAAK,CAAC,GAAG,OAAO,KAAI;AAC9C,WAAO;AAAA,EACT;AAEA,WAAS,SAAS,QAAQ;AACxB,UAAM,MAAM,EAAC,SAAS,CAAC,EAAC;AAExB,YAAQ,aAAa;AACrB,WAAO,QAAQ,QAAQ,WAAS;AAC9B,oBAAc,OAAO,GAAG;AAAA,IAC1B,CAAC;AAED,QAAI,QAAQ,EAAE,IAAI,QAAQ,UAAU;AACpC,WAAO;AAAA,EACT;AAEA,WAAS,cAAc,OAAO,KAAK;AACjC,UAAM,OAAO,cAAc,KAAK;AAChC,YAAQ,oBAAoB,IAAI,GAAG;AAEnC,UAAM,kBAAkB,UAAU,MAAM,KAAK,CAAC;AAC9C,UAAM,MAAM,cAAc,eAAe;AACzC,QAAI,SAAS,KAAK;AAChB;AAAA,IACF;AACA,QAAI,QAAQ,KAAK,UAAU,IAAI,SAAS,GAAG,GAAG;AAC9C;AAAA,EACF;AACF;AAGA,SAAS,UAAU,OAAO;AACxB,MAAI,CAAC,MAAM,WAAW;AACpB,WAAO;AAAA,EACT;AAEA,UAAQ,WAAW,cAAc,KAAK,CAAC,EAAE;AAEzC,QAAM,UAAU,QAAQ,QAAM,kBAAkB,EAAE,CAAC;AAEnD,SAAO;AAEP,WAAS,kBAAkB,UAAU;AACnC,QAAI,sBAAsB,SAAS,IAAI,GAAG;AACxC;AAAA,IACF;AAEA,aAAS,QAAQ,SAAS,MAAM,QAAQ,4CAA4C,GAAG;AAAA,EACzF;AAEF;",
+  "sourcesContent": ["import clone from 'clone';\nimport {fieldToString, isContentSubfieldCode, nvdebug} from './utils.js';\n\nimport createDebugLogger from 'debug';\n\nconst debug = createDebugLogger('@natlibfi/marc-record-validators-melinda:normaliza-dashes');\n//const debugData = debug.extend('data');\nconst debugDev = debug.extend('dev');\n\n\n// Author(s): Nicholas Volk\nexport default function () {\n\n  return {\n    description: 'Normalize various dashes to \"-\"',\n    validate, fix\n  };\n\n  function fix(record) {\n    nvdebug(`FIX ME`, debugDev);\n    record.fields.forEach(field => {\n      fixDashes(field);\n    });\n\n    const res = {message: [], fix: [], valid: true};\n    return res;\n  }\n\n  function validate(record) {\n    const res = {message: []};\n\n    nvdebug(`VALIDATE ME`, debugDev);\n    record.fields?.forEach(field => {\n      validateField(field, res);\n    });\n\n    res.valid = !(res.message.length >= 1);\n    return res;\n  }\n\n  function validateField(field, res) {\n    const orig = fieldToString(field);\n    nvdebug(` VALIDATE FIELD '${orig}'`, debugDev);\n\n    const normalizedField = fixDashes(clone(field));\n    const mod = fieldToString(normalizedField);\n    if (orig === mod) { // Fail as the input is \"broken\"/\"crap\"/sumthing\n      return;\n    }\n    res.message.push(`'TODO: ${orig}' => '${mod}'`);\n    return;\n  }\n}\n\n\nfunction fixDashes(field) {\n  if (!field.subfields) {\n    return field;\n  }\n\n  nvdebug(`Dashing ${fieldToString(field)}`, debugDev);\n\n  field.subfields.forEach(sf => subfieldFixDashes(sf));\n\n  return field;\n\n  function subfieldFixDashes(subfield) {\n    if (!isContentSubfieldCode(subfield.code, field.tag)) {\n      return;\n    }\n    // Normalize dashes U+2010 ... U+2015 to '-':\n    subfield.value = subfield.value.replace(/[\\u2010\\u2011\\u2012\\u2013\\u2014\\u2015]/ug, '-');\n  }\n\n}\n"],
+  "mappings": "AAAA,OAAO,WAAW;AAClB,SAAQ,eAAe,uBAAuB,eAAc;AAE5D,OAAO,uBAAuB;AAE9B,MAAM,QAAQ,kBAAkB,2DAA2D;AAE3F,MAAM,WAAW,MAAM,OAAO,KAAK;AAInC,0BAA2B;AAEzB,SAAO;AAAA,IACL,aAAa;AAAA,IACb;AAAA,IAAU;AAAA,EACZ;AAEA,WAAS,IAAI,QAAQ;AACnB,YAAQ,UAAU,QAAQ;AAC1B,WAAO,OAAO,QAAQ,WAAS;AAC7B,gBAAU,KAAK;AAAA,IACjB,CAAC;AAED,UAAM,MAAM,EAAC,SAAS,CAAC,GAAG,KAAK,CAAC,GAAG,OAAO,KAAI;AAC9C,WAAO;AAAA,EACT;AAEA,WAAS,SAAS,QAAQ;AACxB,UAAM,MAAM,EAAC,SAAS,CAAC,EAAC;AAExB,YAAQ,eAAe,QAAQ;AAC/B,WAAO,QAAQ,QAAQ,WAAS;AAC9B,oBAAc,OAAO,GAAG;AAAA,IAC1B,CAAC;AAED,QAAI,QAAQ,EAAE,IAAI,QAAQ,UAAU;AACpC,WAAO;AAAA,EACT;AAEA,WAAS,cAAc,OAAO,KAAK;AACjC,UAAM,OAAO,cAAc,KAAK;AAChC,YAAQ,oBAAoB,IAAI,KAAK,QAAQ;AAE7C,UAAM,kBAAkB,UAAU,MAAM,KAAK,CAAC;AAC9C,UAAM,MAAM,cAAc,eAAe;AACzC,QAAI,SAAS,KAAK;AAChB;AAAA,IACF;AACA,QAAI,QAAQ,KAAK,UAAU,IAAI,SAAS,GAAG,GAAG;AAC9C;AAAA,EACF;AACF;AAGA,SAAS,UAAU,OAAO;AACxB,MAAI,CAAC,MAAM,WAAW;AACpB,WAAO;AAAA,EACT;AAEA,UAAQ,WAAW,cAAc,KAAK,CAAC,IAAI,QAAQ;AAEnD,QAAM,UAAU,QAAQ,QAAM,kBAAkB,EAAE,CAAC;AAEnD,SAAO;AAEP,WAAS,kBAAkB,UAAU;AACnC,QAAI,CAAC,sBAAsB,SAAS,MAAM,MAAM,GAAG,GAAG;AACpD;AAAA,IACF;AAEA,aAAS,QAAQ,SAAS,MAAM,QAAQ,4CAA4C,GAAG;AAAA,EACzF;AAEF;",
   "names": []
 }

package/dist/normalize-dashes.test.js

@@ -3,7 +3,6 @@ import { MarcRecord } from "@natlibfi/marc-record";
 import validatorFactory from "./normalize-dashes.js";
 import { READERS } from "@natlibfi/fixura";
 import generateTests from "@natlibfi/fixugen";
-import createDebugLogger from "debug";
 generateTests({
   callback,
   path: [import.meta.dirname, "..", "test-fixtures", "normalize-dashes"],
@@ -18,18 +17,13 @@ generateTests({
     }
   }
 });
-const debug = createDebugLogger("@natlibfi/marc-record-validators-melinda/normalize-dashes:test");
 async function testValidatorFactory() {
   const validator = await validatorFactory();
   assert.equal(typeof validator, "object");
   assert.equal(typeof validator.description, "string");
   assert.equal(typeof validator.validate, "function");
 }
-async function callback({ getFixture, enabled = true, fix = false }) {
-  if (enabled === false) {
-    debug("TEST SKIPPED!");
-    return;
-  }
+async function callback({ getFixture, fix = false }) {
   const validator = await validatorFactory();
   const record = new MarcRecord(getFixture("record.json"));
   const expectedResult = getFixture("expectedResult.json");

package/dist/normalize-dashes.test.js.map

@@ -1,7 +1,7 @@
 {
   "version": 3,
   "sources": ["../src/normalize-dashes.test.js"],
-  "sourcesContent": ["import assert from 'node:assert';\nimport {MarcRecord} from '@natlibfi/marc-record';\nimport validatorFactory from './normalize-dashes.js';\nimport {READERS} from '@natlibfi/fixura';\nimport generateTests from '@natlibfi/fixugen';\nimport createDebugLogger from 'debug';\n\ngenerateTests({\n  callback,\n  path: [import.meta.dirname, '..', 'test-fixtures', 'normalize-dashes'],\n  useMetadataFile: true,\n  recurse: false,\n  fixura: {\n    reader: READERS.JSON\n  },\n  hooks: {\n    before: async () => {\n      testValidatorFactory();\n    }\n  }\n});\nconst debug = createDebugLogger('@natlibfi/marc-record-validators-melinda/normalize-dashes:test');\n\nasync function testValidatorFactory() {\n  const validator = await validatorFactory();\n\n  assert.equal(typeof validator, 'object');\n  assert.equal(typeof validator.description, 'string');\n  assert.equal(typeof validator.validate, 'function');\n}\n\nasync function callback({getFixture, enabled = true, fix = false}) {\n  if (enabled === false) {\n    debug('TEST SKIPPED!');\n    return;\n  }\n\n  const validator = await validatorFactory();\n  const record = new MarcRecord(getFixture('record.json'));\n  const expectedResult = getFixture('expectedResult.json');\n  // console.log(expectedResult); // eslint-disable-line\n\n  if (!fix) {\n    const result = await validator.validate(record);\n    assert.deepEqual(result, expectedResult);\n    return;\n  }\n\n  await validator.fix(record);\n  assert.deepEqual(record, expectedResult);\n}\n"],
-  "mappings": "AAAA,OAAO,YAAY;AACnB,SAAQ,kBAAiB;AACzB,OAAO,sBAAsB;AAC7B,SAAQ,eAAc;AACtB,OAAO,mBAAmB;AAC1B,OAAO,uBAAuB;AAE9B,cAAc;AAAA,EACZ;AAAA,EACA,MAAM,CAAC,YAAY,SAAS,MAAM,iBAAiB,kBAAkB;AAAA,EACrE,iBAAiB;AAAA,EACjB,SAAS;AAAA,EACT,QAAQ;AAAA,IACN,QAAQ,QAAQ;AAAA,EAClB;AAAA,EACA,OAAO;AAAA,IACL,QAAQ,YAAY;AAClB,2BAAqB;AAAA,IACvB;AAAA,EACF;AACF,CAAC;AACD,MAAM,QAAQ,kBAAkB,gEAAgE;AAEhG,eAAe,uBAAuB;AACpC,QAAM,YAAY,MAAM,iBAAiB;AAEzC,SAAO,MAAM,OAAO,WAAW,QAAQ;AACvC,SAAO,MAAM,OAAO,UAAU,aAAa,QAAQ;AACnD,SAAO,MAAM,OAAO,UAAU,UAAU,UAAU;AACpD;AAEA,eAAe,SAAS,EAAC,YAAY,UAAU,MAAM,MAAM,MAAK,GAAG;AACjE,MAAI,YAAY,OAAO;AACrB,UAAM,eAAe;AACrB;AAAA,EACF;AAEA,QAAM,YAAY,MAAM,iBAAiB;AACzC,QAAM,SAAS,IAAI,WAAW,WAAW,aAAa,CAAC;AACvD,QAAM,iBAAiB,WAAW,qBAAqB;AAGvD,MAAI,CAAC,KAAK;AACR,UAAM,SAAS,MAAM,UAAU,SAAS,MAAM;AAC9C,WAAO,UAAU,QAAQ,cAAc;AACvC;AAAA,EACF;AAEA,QAAM,UAAU,IAAI,MAAM;AAC1B,SAAO,UAAU,QAAQ,cAAc;AACzC;",
+  "sourcesContent": ["import assert from 'node:assert';\nimport {MarcRecord} from '@natlibfi/marc-record';\nimport validatorFactory from './normalize-dashes.js';\nimport {READERS} from '@natlibfi/fixura';\nimport generateTests from '@natlibfi/fixugen';\n//import createDebugLogger from 'debug';\n\ngenerateTests({\n  callback,\n  path: [import.meta.dirname, '..', 'test-fixtures', 'normalize-dashes'],\n  useMetadataFile: true,\n  recurse: false,\n  fixura: {\n    reader: READERS.JSON\n  },\n  hooks: {\n    before: async () => {\n      testValidatorFactory();\n    }\n  }\n});\n//const debug = createDebugLogger('@natlibfi/marc-record-validators-melinda/normalize-dashes:test');\n\nasync function testValidatorFactory() {\n  const validator = await validatorFactory();\n\n  assert.equal(typeof validator, 'object');\n  assert.equal(typeof validator.description, 'string');\n  assert.equal(typeof validator.validate, 'function');\n}\n\nasync function callback({getFixture, fix = false}) {\n  const validator = await validatorFactory();\n  const record = new MarcRecord(getFixture('record.json'));\n  const expectedResult = getFixture('expectedResult.json');\n  // console.log(expectedResult); // eslint-disable-line\n\n  if (!fix) {\n    const result = await validator.validate(record);\n    assert.deepEqual(result, expectedResult);\n    return;\n  }\n\n  await validator.fix(record);\n  assert.deepEqual(record, expectedResult);\n}\n"],
+  "mappings": "AAAA,OAAO,YAAY;AACnB,SAAQ,kBAAiB;AACzB,OAAO,sBAAsB;AAC7B,SAAQ,eAAc;AACtB,OAAO,mBAAmB;AAG1B,cAAc;AAAA,EACZ;AAAA,EACA,MAAM,CAAC,YAAY,SAAS,MAAM,iBAAiB,kBAAkB;AAAA,EACrE,iBAAiB;AAAA,EACjB,SAAS;AAAA,EACT,QAAQ;AAAA,IACN,QAAQ,QAAQ;AAAA,EAClB;AAAA,EACA,OAAO;AAAA,IACL,QAAQ,YAAY;AAClB,2BAAqB;AAAA,IACvB;AAAA,EACF;AACF,CAAC;AAGD,eAAe,uBAAuB;AACpC,QAAM,YAAY,MAAM,iBAAiB;AAEzC,SAAO,MAAM,OAAO,WAAW,QAAQ;AACvC,SAAO,MAAM,OAAO,UAAU,aAAa,QAAQ;AACnD,SAAO,MAAM,OAAO,UAAU,UAAU,UAAU;AACpD;AAEA,eAAe,SAAS,EAAC,YAAY,MAAM,MAAK,GAAG;AACjD,QAAM,YAAY,MAAM,iBAAiB;AACzC,QAAM,SAAS,IAAI,WAAW,WAAW,aAAa,CAAC;AACvD,QAAM,iBAAiB,WAAW,qBAAqB;AAGvD,MAAI,CAAC,KAAK;AACR,UAAM,SAAS,MAAM,UAAU,SAAS,MAAM;AAC9C,WAAO,UAAU,QAAQ,cAAc;AACvC;AAAA,EACF;AAEA,QAAM,UAAU,IAAI,MAAM;AAC1B,SAAO,UAAU,QAAQ,cAAc;AACzC;",
   "names": []
 }

package/dist/normalize-identifiers.js.map

@@ -1,7 +1,7 @@
 {
   "version": 3,
   "sources": ["../src/normalize-identifiers.js"],
-  "sourcesContent": ["// Relocated from melinda-marc-record-merge-reducers (and renamed)\n//import createDebugLogger from 'debug';\nimport clone from 'clone';\nimport {fieldToString} from './utils.js';\n//const debug = createDebugLogger('@natlibfi/marc-record-validators-melinda:normalize-identifiers');\n\n/*\nfunction nvdebug(message, func) {\n  if (func) {\n    func(message);\n  }\n  console.info(message); // eslint-disable-line no-console\n}\n*/\n\nexport default function () {\n\n  // NB! We should and could handle ISNIs here as well.\n  return {\n    description: 'Normalizes identifiers such as subfield $0 values',\n    validate, fix\n  };\n\n  function fix(record) {\n    const res = {message: [], fix: [], valid: true};\n    //message.fix = [];\n\n    // Actual parsing of all fields\n    /*\n      if (!record.fields) {\n        return false;\n      }\n      */\n\n    //nvdebug(`NORMALIZE CONTROL NUMBER FIX`, debug);\n    record.fields.forEach(field => {\n      //nvdebug(` NORMALIZE CONTROL NUMBER FIX ${fieldToString(field)}`, debug);\n\n      fieldNormalizeControlNumbers(field);\n      //validateField(field, true, message);\n    });\n\n    // message.valid = !(message.message.length >= 1);\n    return res;\n  }\n\n  function validate(record) {\n    const res = {message: []};\n    //nvdebug(`NORMALIZE CONTROL NUMBER VALIDATE`, debug);\n    // Actual parsing of all fields\n    /*\n      if (!record.fields) {\n        return false;\n      }\n      */\n\n    record.fields.forEach(field => {\n      //nvdebug(` NORMALIZE CONTROL NUMBER VALIDATE ${fieldToString(field)}`, debug);\n      validateField(field, res);\n    });\n\n    res.valid = !(res.message.length >= 1);\n    return res;\n  }\n\n  function validateField(field, res) {\n    if (!field.subfields) {\n      return;\n    }\n\n    const normalizedField = clone(field);\n    fieldNormalizeControlNumbers(normalizedField);\n\n    const orig = fieldToString(field);\n    const mod = fieldToString(normalizedField);\n    if (orig !== mod) { // Fail as the input is \"broken\"/\"crap\"/sumthing\n      res.message.push(`'${orig}' could do with control number identifier normalization`);\n      return;\n    }\n\n    return;\n  }\n}\n\nconst alephInternal01 = '(FIN01)';\nconst alephInternal10 = '(FIN10)';\nconst alephInternal11 = '(FIN11)';\nconst alephInternal12 = '(FIN12)';\nconst alephInternal13 = '(FIN13)';\n\nconst standard01 = '(FI-MELINDA)';\nconst standard10 = '(FI-ASTERI-S)';\nconst standard11 = '(FI-ASTERI-N)';\nconst standard12 = '(FI-ASTERI-A)';\nconst standard13 = '(FI-ASTERI-W)';\n\nconst both01 = {ALEPH_INTERNAL: alephInternal01, STANDARD: standard01};\nconst both10 = {ALEPH_INTERNAL: alephInternal10, STANDARD: standard10};\nconst both11 = {ALEPH_INTERNAL: alephInternal11, STANDARD: standard11};\nconst both12 = {ALEPH_INTERNAL: alephInternal12, STANDARD: standard12};\nconst both13 = {ALEPH_INTERNAL: alephInternal13, STANDARD: standard13};\n\nconst mappings = {\n  'FCC': both01,\n  '(FI-ASTERI-A)': both12,\n  '(FI-ASTERI-N)': both11,\n  '(FI-ASTERI-S)': both10,\n  '(FI-ASTERI-W)': both13,\n  '(FI-MELINDA)': both01,\n  '(FIN01)': both01,\n  '(FIN10)': both10,\n  '(FIN11)': both11,\n  '(FIN12)': both12,\n  '(FIN13)': both13,\n  'http://urn.fi/URN:NBN:fi:au:finaf:': both11,\n  'https://urn.fi/URN:NBN:fi:au:finaf:': both11\n};\n\nfunction normalizeNineDigitIDs(value, targetFormat = 'ALEPH_INTERNAL') {\n  // $value should be prefix + nine-digits. Do nothing if nine-digit tail condition is not met:\n  const nineDigitTail = value.slice(-9);\n  if (!(/^[0-9]{9}$/u).test(nineDigitTail)) {\n    return value;\n  }\n  // Normalize prefix:\n  const currPrefix = value.slice(0, -9);\n\n  if (currPrefix in mappings) {\n    //nvdebug(`${currPrefix}, TF:${targetFormat}...`);\n    //nvdebug(`${JSON.stringify(mappings[currPrefix])}`);\n    return `${mappings[currPrefix][targetFormat]}${nineDigitTail}`;\n  }\n  return value;\n}\n\nexport function isIsni(value) {\n  if ((/^\\(isni\\)(?: ?[0-9]{4}){4}$/u).test(value)) {\n    return true;\n  }\n  if ((/^https:\\/\\/isni.org\\/isni\\/[0-9]{16}$/u).test(value)) {\n    return true;\n  }\n  return false;\n}\n\nexport function normalizeIsni(value) {\n  if (isIsni(value)) {\n    return `https://isni.org/isni/${value.replace(/[^0-9]/gu, '')}`;\n  }\n  return value;\n}\n\n\nexport function normalizeControlSubfieldValue(value = '', targetFormat = 'ALEPH_INTERNAL') {\n  if (isIsni(value)) {\n    return normalizeIsni(value);\n  }\n\n  const normalizedValue = normalizeNineDigitIDs(value, targetFormat);\n  if (normalizedValue !== value) {\n    return normalizedValue;\n  }\n\n  // Something for isni IDs?\n  return value;\n}\n\n//export function normalizableSubfieldPrefix(tag, sf) {\nexport function normalizeAs(tag, subfieldCode) {\n  //nvdebug(`nAs ${tag}, ${subfieldCode}`);\n  if (subfieldCode === '0' || subfieldCode === '1' || subfieldCode === 'w') {\n    return 'ALEPH_INTERNAL';\n  }\n\n  if (tag === '035' && ['a', 'z'].includes(subfieldCode)) {\n    return 'STANDARD';\n  }\n  return undefined;\n}\n\nexport function fieldNormalizeControlNumbers(field) {\n  // Rename \"Prefixes\" as \"ControlNumberIdentifiers\"?\n  // No, since isni etc...  however, just \"ControlNumber\" would do...\n  // \"identifiers\" ?\n  if (!field.subfields) {\n    return;\n  }\n\n  field.subfields.forEach(sf => {\n    const targetFormat = normalizeAs(field.tag, sf.code);\n    if (targetFormat !== undefined) {\n      //nvdebug(`NORMALIZE SUBFIELD $${sf.code} IN FIELD: '${fieldToString(field)}' TO ${targetFormat}`);\n      sf.value = normalizeControlSubfieldValue(sf.value, targetFormat);\n      return;\n    }\n  });\n}\n"],
-  "mappings": "AAEA,OAAO,WAAW;AAClB,SAAQ,qBAAoB;AAY5B,0BAA2B;AAGzB,SAAO;AAAA,IACL,aAAa;AAAA,IACb;AAAA,IAAU;AAAA,EACZ;AAEA,WAAS,IAAI,QAAQ;AACnB,UAAM,MAAM,EAAC,SAAS,CAAC,GAAG,KAAK,CAAC,GAAG,OAAO,KAAI;AAW9C,WAAO,OAAO,QAAQ,WAAS;AAG7B,mCAA6B,KAAK;AAAA,IAEpC,CAAC;AAGD,WAAO;AAAA,EACT;AAEA,WAAS,SAAS,QAAQ;AACxB,UAAM,MAAM,EAAC,SAAS,CAAC,EAAC;AASxB,WAAO,OAAO,QAAQ,WAAS;AAE7B,oBAAc,OAAO,GAAG;AAAA,IAC1B,CAAC;AAED,QAAI,QAAQ,EAAE,IAAI,QAAQ,UAAU;AACpC,WAAO;AAAA,EACT;AAEA,WAAS,cAAc,OAAO,KAAK;AACjC,QAAI,CAAC,MAAM,WAAW;AACpB;AAAA,IACF;AAEA,UAAM,kBAAkB,MAAM,KAAK;AACnC,iCAA6B,eAAe;AAE5C,UAAM,OAAO,cAAc,KAAK;AAChC,UAAM,MAAM,cAAc,eAAe;AACzC,QAAI,SAAS,KAAK;AAChB,UAAI,QAAQ,KAAK,IAAI,IAAI,yDAAyD;AAClF;AAAA,IACF;AAEA;AAAA,EACF;AACF;AAEA,MAAM,kBAAkB;AACxB,MAAM,kBAAkB;AACxB,MAAM,kBAAkB;AACxB,MAAM,kBAAkB;AACxB,MAAM,kBAAkB;AAExB,MAAM,aAAa;AACnB,MAAM,aAAa;AACnB,MAAM,aAAa;AACnB,MAAM,aAAa;AACnB,MAAM,aAAa;AAEnB,MAAM,SAAS,EAAC,gBAAgB,iBAAiB,UAAU,WAAU;AACrE,MAAM,SAAS,EAAC,gBAAgB,iBAAiB,UAAU,WAAU;AACrE,MAAM,SAAS,EAAC,gBAAgB,iBAAiB,UAAU,WAAU;AACrE,MAAM,SAAS,EAAC,gBAAgB,iBAAiB,UAAU,WAAU;AACrE,MAAM,SAAS,EAAC,gBAAgB,iBAAiB,UAAU,WAAU;AAErE,MAAM,WAAW;AAAA,EACf,OAAO;AAAA,EACP,iBAAiB;AAAA,EACjB,iBAAiB;AAAA,EACjB,iBAAiB;AAAA,EACjB,iBAAiB;AAAA,EACjB,gBAAgB;AAAA,EAChB,WAAW;AAAA,EACX,WAAW;AAAA,EACX,WAAW;AAAA,EACX,WAAW;AAAA,EACX,WAAW;AAAA,EACX,sCAAsC;AAAA,EACtC,uCAAuC;AACzC;AAEA,SAAS,sBAAsB,OAAO,eAAe,kBAAkB;AAErE,QAAM,gBAAgB,MAAM,MAAM,EAAE;AACpC,MAAI,CAAE,cAAe,KAAK,aAAa,GAAG;AACxC,WAAO;AAAA,EACT;AAEA,QAAM,aAAa,MAAM,MAAM,GAAG,EAAE;AAEpC,MAAI,cAAc,UAAU;AAG1B,WAAO,GAAG,SAAS,UAAU,EAAE,YAAY,CAAC,GAAG,aAAa;AAAA,EAC9D;AACA,SAAO;AACT;AAEO,gBAAS,OAAO,OAAO;AAC5B,MAAK,+BAAgC,KAAK,KAAK,GAAG;AAChD,WAAO;AAAA,EACT;AACA,MAAK,yCAA0C,KAAK,KAAK,GAAG;AAC1D,WAAO;AAAA,EACT;AACA,SAAO;AACT;AAEO,gBAAS,cAAc,OAAO;AACnC,MAAI,OAAO,KAAK,GAAG;AACjB,WAAO,yBAAyB,MAAM,QAAQ,YAAY,EAAE,CAAC;AAAA,EAC/D;AACA,SAAO;AACT;AAGO,gBAAS,8BAA8B,QAAQ,IAAI,eAAe,kBAAkB;AACzF,MAAI,OAAO,KAAK,GAAG;AACjB,WAAO,cAAc,KAAK;AAAA,EAC5B;AAEA,QAAM,kBAAkB,sBAAsB,OAAO,YAAY;AACjE,MAAI,oBAAoB,OAAO;AAC7B,WAAO;AAAA,EACT;AAGA,SAAO;AACT;AAGO,gBAAS,YAAY,KAAK,cAAc;AAE7C,MAAI,iBAAiB,OAAO,iBAAiB,OAAO,iBAAiB,KAAK;AACxE,WAAO;AAAA,EACT;AAEA,MAAI,QAAQ,SAAS,CAAC,KAAK,GAAG,EAAE,SAAS,YAAY,GAAG;AACtD,WAAO;AAAA,EACT;AACA,SAAO;AACT;AAEO,gBAAS,6BAA6B,OAAO;AAIlD,MAAI,CAAC,MAAM,WAAW;AACpB;AAAA,EACF;AAEA,QAAM,UAAU,QAAQ,QAAM;AAC5B,UAAM,eAAe,YAAY,MAAM,KAAK,GAAG,IAAI;AACnD,QAAI,iBAAiB,QAAW;AAE9B,SAAG,QAAQ,8BAA8B,GAAG,OAAO,YAAY;AAC/D;AAAA,IACF;AAAA,EACF,CAAC;AACH;",
+  "sourcesContent": ["// Relocated from melinda-marc-record-merge-reducers (and renamed)\nimport clone from 'clone';\nimport {fieldToString, /*nvdebug*/} from './utils.js';\n\n//const debug = createDebugLogger('@natlibfi/marc-record-validators-melinda:normalize-identifiers');\n//const debugData = debug.extend('data');\n//const debugDev = debug.extend('dev');\n\nexport default function () {\n\n  // NB! We should and could handle ISNIs here as well.\n  return {\n    description: 'Normalizes identifiers such as subfield $0 values',\n    validate, fix\n  };\n\n  function fix(record) {\n    const res = {message: [], fix: [], valid: true};\n    //message.fix = [];\n\n    // Actual parsing of all fields\n    /*\n      if (!record.fields) {\n        return false;\n      }\n      */\n\n    //nvdebug(`NORMALIZE CONTROL NUMBER FIX`, debugDev);\n    record.fields.forEach(field => {\n      //nvdebug(` NORMALIZE CONTROL NUMBER FIX ${fieldToString(field)}`, debugDev);\n\n      fieldNormalizeControlNumbers(field);\n      //validateField(field, true, message);\n    });\n\n    // message.valid = !(message.message.length >= 1);\n    return res;\n  }\n\n  function validate(record) {\n    const res = {message: []};\n    //nvdebug(`NORMALIZE CONTROL NUMBER VALIDATE`, debugDev);\n    // Actual parsing of all fields\n    /*\n      if (!record.fields) {\n        return false;\n      }\n      */\n\n    record.fields.forEach(field => {\n      //nvdebug(` NORMALIZE CONTROL NUMBER VALIDATE ${fieldToString(field)}`, debugDev);\n      validateField(field, res);\n    });\n\n    res.valid = !(res.message.length >= 1);\n    return res;\n  }\n\n  function validateField(field, res) {\n    if (!field.subfields) {\n      return;\n    }\n\n    const normalizedField = clone(field);\n    fieldNormalizeControlNumbers(normalizedField);\n\n    const orig = fieldToString(field);\n    const mod = fieldToString(normalizedField);\n    if (orig !== mod) { // Fail as the input is \"broken\"/\"crap\"/sumthing\n      res.message.push(`'${orig}' could do with control number identifier normalization`);\n      return;\n    }\n\n    return;\n  }\n}\n\nconst alephInternal01 = '(FIN01)';\nconst alephInternal10 = '(FIN10)';\nconst alephInternal11 = '(FIN11)';\nconst alephInternal12 = '(FIN12)';\nconst alephInternal13 = '(FIN13)';\n\nconst standard01 = '(FI-MELINDA)';\nconst standard10 = '(FI-ASTERI-S)';\nconst standard11 = '(FI-ASTERI-N)';\nconst standard12 = '(FI-ASTERI-A)';\nconst standard13 = '(FI-ASTERI-W)';\n\nconst both01 = {ALEPH_INTERNAL: alephInternal01, STANDARD: standard01};\nconst both10 = {ALEPH_INTERNAL: alephInternal10, STANDARD: standard10};\nconst both11 = {ALEPH_INTERNAL: alephInternal11, STANDARD: standard11};\nconst both12 = {ALEPH_INTERNAL: alephInternal12, STANDARD: standard12};\nconst both13 = {ALEPH_INTERNAL: alephInternal13, STANDARD: standard13};\n\nconst mappings = {\n  'FCC': both01,\n  '(FI-ASTERI-A)': both12,\n  '(FI-ASTERI-N)': both11,\n  '(FI-ASTERI-S)': both10,\n  '(FI-ASTERI-W)': both13,\n  '(FI-MELINDA)': both01,\n  '(FIN01)': both01,\n  '(FIN10)': both10,\n  '(FIN11)': both11,\n  '(FIN12)': both12,\n  '(FIN13)': both13,\n  'http://urn.fi/URN:NBN:fi:au:finaf:': both11,\n  'https://urn.fi/URN:NBN:fi:au:finaf:': both11\n};\n\nfunction normalizeNineDigitIDs(value, targetFormat = 'ALEPH_INTERNAL') {\n  // $value should be prefix + nine-digits. Do nothing if nine-digit tail condition is not met:\n  const nineDigitTail = value.slice(-9);\n  if (!(/^[0-9]{9}$/u).test(nineDigitTail)) {\n    return value;\n  }\n  // Normalize prefix:\n  const currPrefix = value.slice(0, -9);\n\n  if (currPrefix in mappings) {\n    //nvdebug(`${currPrefix}, TF:${targetFormat}...`, debugDev);\n    //nvdebug(`${JSON.stringify(mappings[currPrefix])}`, debugDev);\n    return `${mappings[currPrefix][targetFormat]}${nineDigitTail}`;\n  }\n  return value;\n}\n\nexport function isIsni(value) {\n  if ((/^\\(isni\\)(?: ?[0-9]{4}){4}$/u).test(value)) {\n    return true;\n  }\n  if ((/^https:\\/\\/isni.org\\/isni\\/[0-9]{16}$/u).test(value)) {\n    return true;\n  }\n  return false;\n}\n\nexport function normalizeIsni(value) {\n  if (isIsni(value)) {\n    return `https://isni.org/isni/${value.replace(/[^0-9]/gu, '')}`;\n  }\n  return value;\n}\n\n\nexport function normalizeControlSubfieldValue(value = '', targetFormat = 'ALEPH_INTERNAL') {\n  if (isIsni(value)) {\n    return normalizeIsni(value);\n  }\n\n  const normalizedValue = normalizeNineDigitIDs(value, targetFormat);\n  if (normalizedValue !== value) {\n    return normalizedValue;\n  }\n\n  // Something for isni IDs?\n  return value;\n}\n\n//export function normalizableSubfieldPrefix(tag, sf) {\nexport function normalizeAs(tag, subfieldCode) {\n  //nvdebug(`nAs ${tag}, ${subfieldCode}`, debugDev);\n  if (subfieldCode === '0' || subfieldCode === '1' || subfieldCode === 'w') {\n    return 'ALEPH_INTERNAL';\n  }\n\n  if (tag === '035' && ['a', 'z'].includes(subfieldCode)) {\n    return 'STANDARD';\n  }\n  return undefined;\n}\n\nexport function fieldNormalizeControlNumbers(field) {\n  // Rename \"Prefixes\" as \"ControlNumberIdentifiers\"?\n  // No, since isni etc...  however, just \"ControlNumber\" would do...\n  // \"identifiers\" ?\n  if (!field.subfields) {\n    return;\n  }\n\n  field.subfields.forEach(sf => {\n    const targetFormat = normalizeAs(field.tag, sf.code);\n    if (targetFormat !== undefined) {\n      //nvdebug(`NORMALIZE SUBFIELD $${sf.code} IN FIELD: '${fieldToString(field)}' TO ${targetFormat}`, debugDev);\n      sf.value = normalizeControlSubfieldValue(sf.value, targetFormat);\n      return;\n    }\n  });\n}\n"],
+  "mappings": "AACA,OAAO,WAAW;AAClB,SAAQ,qBAAiC;AAMzC,0BAA2B;AAGzB,SAAO;AAAA,IACL,aAAa;AAAA,IACb;AAAA,IAAU;AAAA,EACZ;AAEA,WAAS,IAAI,QAAQ;AACnB,UAAM,MAAM,EAAC,SAAS,CAAC,GAAG,KAAK,CAAC,GAAG,OAAO,KAAI;AAW9C,WAAO,OAAO,QAAQ,WAAS;AAG7B,mCAA6B,KAAK;AAAA,IAEpC,CAAC;AAGD,WAAO;AAAA,EACT;AAEA,WAAS,SAAS,QAAQ;AACxB,UAAM,MAAM,EAAC,SAAS,CAAC,EAAC;AASxB,WAAO,OAAO,QAAQ,WAAS;AAE7B,oBAAc,OAAO,GAAG;AAAA,IAC1B,CAAC;AAED,QAAI,QAAQ,EAAE,IAAI,QAAQ,UAAU;AACpC,WAAO;AAAA,EACT;AAEA,WAAS,cAAc,OAAO,KAAK;AACjC,QAAI,CAAC,MAAM,WAAW;AACpB;AAAA,IACF;AAEA,UAAM,kBAAkB,MAAM,KAAK;AACnC,iCAA6B,eAAe;AAE5C,UAAM,OAAO,cAAc,KAAK;AAChC,UAAM,MAAM,cAAc,eAAe;AACzC,QAAI,SAAS,KAAK;AAChB,UAAI,QAAQ,KAAK,IAAI,IAAI,yDAAyD;AAClF;AAAA,IACF;AAEA;AAAA,EACF;AACF;AAEA,MAAM,kBAAkB;AACxB,MAAM,kBAAkB;AACxB,MAAM,kBAAkB;AACxB,MAAM,kBAAkB;AACxB,MAAM,kBAAkB;AAExB,MAAM,aAAa;AACnB,MAAM,aAAa;AACnB,MAAM,aAAa;AACnB,MAAM,aAAa;AACnB,MAAM,aAAa;AAEnB,MAAM,SAAS,EAAC,gBAAgB,iBAAiB,UAAU,WAAU;AACrE,MAAM,SAAS,EAAC,gBAAgB,iBAAiB,UAAU,WAAU;AACrE,MAAM,SAAS,EAAC,gBAAgB,iBAAiB,UAAU,WAAU;AACrE,MAAM,SAAS,EAAC,gBAAgB,iBAAiB,UAAU,WAAU;AACrE,MAAM,SAAS,EAAC,gBAAgB,iBAAiB,UAAU,WAAU;AAErE,MAAM,WAAW;AAAA,EACf,OAAO;AAAA,EACP,iBAAiB;AAAA,EACjB,iBAAiB;AAAA,EACjB,iBAAiB;AAAA,EACjB,iBAAiB;AAAA,EACjB,gBAAgB;AAAA,EAChB,WAAW;AAAA,EACX,WAAW;AAAA,EACX,WAAW;AAAA,EACX,WAAW;AAAA,EACX,WAAW;AAAA,EACX,sCAAsC;AAAA,EACtC,uCAAuC;AACzC;AAEA,SAAS,sBAAsB,OAAO,eAAe,kBAAkB;AAErE,QAAM,gBAAgB,MAAM,MAAM,EAAE;AACpC,MAAI,CAAE,cAAe,KAAK,aAAa,GAAG;AACxC,WAAO;AAAA,EACT;AAEA,QAAM,aAAa,MAAM,MAAM,GAAG,EAAE;AAEpC,MAAI,cAAc,UAAU;AAG1B,WAAO,GAAG,SAAS,UAAU,EAAE,YAAY,CAAC,GAAG,aAAa;AAAA,EAC9D;AACA,SAAO;AACT;AAEO,gBAAS,OAAO,OAAO;AAC5B,MAAK,+BAAgC,KAAK,KAAK,GAAG;AAChD,WAAO;AAAA,EACT;AACA,MAAK,yCAA0C,KAAK,KAAK,GAAG;AAC1D,WAAO;AAAA,EACT;AACA,SAAO;AACT;AAEO,gBAAS,cAAc,OAAO;AACnC,MAAI,OAAO,KAAK,GAAG;AACjB,WAAO,yBAAyB,MAAM,QAAQ,YAAY,EAAE,CAAC;AAAA,EAC/D;AACA,SAAO;AACT;AAGO,gBAAS,8BAA8B,QAAQ,IAAI,eAAe,kBAAkB;AACzF,MAAI,OAAO,KAAK,GAAG;AACjB,WAAO,cAAc,KAAK;AAAA,EAC5B;AAEA,QAAM,kBAAkB,sBAAsB,OAAO,YAAY;AACjE,MAAI,oBAAoB,OAAO;AAC7B,WAAO;AAAA,EACT;AAGA,SAAO;AACT;AAGO,gBAAS,YAAY,KAAK,cAAc;AAE7C,MAAI,iBAAiB,OAAO,iBAAiB,OAAO,iBAAiB,KAAK;AACxE,WAAO;AAAA,EACT;AAEA,MAAI,QAAQ,SAAS,CAAC,KAAK,GAAG,EAAE,SAAS,YAAY,GAAG;AACtD,WAAO;AAAA,EACT;AACA,SAAO;AACT;AAEO,gBAAS,6BAA6B,OAAO;AAIlD,MAAI,CAAC,MAAM,WAAW;AACpB;AAAA,EACF;AAEA,QAAM,UAAU,QAAQ,QAAM;AAC5B,UAAM,eAAe,YAAY,MAAM,KAAK,GAAG,IAAI;AACnD,QAAI,iBAAiB,QAAW;AAE9B,SAAG,QAAQ,8BAA8B,GAAG,OAAO,YAAY;AAC/D;AAAA,IACF;AAAA,EACF,CAAC;AACH;",
   "names": []
 }

package/dist/normalize-identifiers.test.js

@@ -3,7 +3,6 @@ import { MarcRecord } from "@natlibfi/marc-record";
 import validatorFactory from "./normalize-identifiers.js";
 import { READERS } from "@natlibfi/fixura";
 import generateTests from "@natlibfi/fixugen";
-import createDebugLogger from "debug";
 generateTests({
   callback,
   path: [import.meta.dirname, "..", "test-fixtures", "normalize-identifiers"],
@@ -18,18 +17,13 @@ generateTests({
     }
   }
 });
-const debug = createDebugLogger("@natlibfi/marc-record-validators-melinda/normalize-identifiers:test");
 async function testValidatorFactory() {
   const validator = await validatorFactory();
   assert.equal(typeof validator, "object");
   assert.equal(typeof validator.description, "string");
   assert.equal(typeof validator.validate, "function");
 }
-async function callback({ getFixture, enabled = true, fix = false }) {
-  if (enabled === false) {
-    debug("TEST SKIPPED!");
-    return;
-  }
+async function callback({ getFixture, fix = false }) {
   const validator = await validatorFactory();
   const record = new MarcRecord(getFixture("record.json"));
   const expectedResult = getFixture("expectedResult.json");

package/dist/normalize-identifiers.test.js.map

@@ -1,7 +1,7 @@
 {
   "version": 3,
   "sources": ["../src/normalize-identifiers.test.js"],
-  "sourcesContent": ["import assert from 'node:assert';\nimport {MarcRecord} from '@natlibfi/marc-record';\nimport validatorFactory from './normalize-identifiers.js';\nimport {READERS} from '@natlibfi/fixura';\nimport generateTests from '@natlibfi/fixugen';\nimport createDebugLogger from 'debug';\n\ngenerateTests({\n  callback,\n  path: [import.meta.dirname, '..', 'test-fixtures', 'normalize-identifiers'],\n  useMetadataFile: true,\n  recurse: false,\n  fixura: {\n    reader: READERS.JSON\n  },\n  hooks: {\n    before: async () => {\n      testValidatorFactory();\n    }\n  }\n});\nconst debug = createDebugLogger('@natlibfi/marc-record-validators-melinda/normalize-identifiers:test');\n\nasync function testValidatorFactory() {\n  const validator = await validatorFactory();\n\n  assert.equal(typeof validator, 'object');\n  assert.equal(typeof validator.description, 'string');\n  assert.equal(typeof validator.validate, 'function');\n}\n\nasync function callback({getFixture, enabled = true, fix = false}) {\n  if (enabled === false) {\n    debug('TEST SKIPPED!');\n    return;\n  }\n\n  const validator = await validatorFactory();\n  const record = new MarcRecord(getFixture('record.json'));\n  const expectedResult = getFixture('expectedResult.json');\n  // console.log(expectedResult); // eslint-disable-line\n\n  if (!fix) {\n    const result = await validator.validate(record);\n    assert.deepEqual(result, expectedResult);\n    return;\n  }\n\n  await validator.fix(record);\n  assert.deepEqual(record, expectedResult);\n}\n"],
-  "mappings": "AAAA,OAAO,YAAY;AACnB,SAAQ,kBAAiB;AACzB,OAAO,sBAAsB;AAC7B,SAAQ,eAAc;AACtB,OAAO,mBAAmB;AAC1B,OAAO,uBAAuB;AAE9B,cAAc;AAAA,EACZ;AAAA,EACA,MAAM,CAAC,YAAY,SAAS,MAAM,iBAAiB,uBAAuB;AAAA,EAC1E,iBAAiB;AAAA,EACjB,SAAS;AAAA,EACT,QAAQ;AAAA,IACN,QAAQ,QAAQ;AAAA,EAClB;AAAA,EACA,OAAO;AAAA,IACL,QAAQ,YAAY;AAClB,2BAAqB;AAAA,IACvB;AAAA,EACF;AACF,CAAC;AACD,MAAM,QAAQ,kBAAkB,qEAAqE;AAErG,eAAe,uBAAuB;AACpC,QAAM,YAAY,MAAM,iBAAiB;AAEzC,SAAO,MAAM,OAAO,WAAW,QAAQ;AACvC,SAAO,MAAM,OAAO,UAAU,aAAa,QAAQ;AACnD,SAAO,MAAM,OAAO,UAAU,UAAU,UAAU;AACpD;AAEA,eAAe,SAAS,EAAC,YAAY,UAAU,MAAM,MAAM,MAAK,GAAG;AACjE,MAAI,YAAY,OAAO;AACrB,UAAM,eAAe;AACrB;AAAA,EACF;AAEA,QAAM,YAAY,MAAM,iBAAiB;AACzC,QAAM,SAAS,IAAI,WAAW,WAAW,aAAa,CAAC;AACvD,QAAM,iBAAiB,WAAW,qBAAqB;AAGvD,MAAI,CAAC,KAAK;AACR,UAAM,SAAS,MAAM,UAAU,SAAS,MAAM;AAC9C,WAAO,UAAU,QAAQ,cAAc;AACvC;AAAA,EACF;AAEA,QAAM,UAAU,IAAI,MAAM;AAC1B,SAAO,UAAU,QAAQ,cAAc;AACzC;",
+  "sourcesContent": ["import assert from 'node:assert';\nimport {MarcRecord} from '@natlibfi/marc-record';\nimport validatorFactory from './normalize-identifiers.js';\nimport {READERS} from '@natlibfi/fixura';\nimport generateTests from '@natlibfi/fixugen';\n//import createDebugLogger from 'debug';\n\ngenerateTests({\n  callback,\n  path: [import.meta.dirname, '..', 'test-fixtures', 'normalize-identifiers'],\n  useMetadataFile: true,\n  recurse: false,\n  fixura: {\n    reader: READERS.JSON\n  },\n  hooks: {\n    before: async () => {\n      testValidatorFactory();\n    }\n  }\n});\n//const debug = createDebugLogger('@natlibfi/marc-record-validators-melinda/normalize-identifiers:test');\n\nasync function testValidatorFactory() {\n  const validator = await validatorFactory();\n\n  assert.equal(typeof validator, 'object');\n  assert.equal(typeof validator.description, 'string');\n  assert.equal(typeof validator.validate, 'function');\n}\n\nasync function callback({getFixture, fix = false}) {\n  const validator = await validatorFactory();\n  const record = new MarcRecord(getFixture('record.json'));\n  const expectedResult = getFixture('expectedResult.json');\n  // console.log(expectedResult); // eslint-disable-line\n\n  if (!fix) {\n    const result = await validator.validate(record);\n    assert.deepEqual(result, expectedResult);\n    return;\n  }\n\n  await validator.fix(record);\n  assert.deepEqual(record, expectedResult);\n}\n"],
+  "mappings": "AAAA,OAAO,YAAY;AACnB,SAAQ,kBAAiB;AACzB,OAAO,sBAAsB;AAC7B,SAAQ,eAAc;AACtB,OAAO,mBAAmB;AAG1B,cAAc;AAAA,EACZ;AAAA,EACA,MAAM,CAAC,YAAY,SAAS,MAAM,iBAAiB,uBAAuB;AAAA,EAC1E,iBAAiB;AAAA,EACjB,SAAS;AAAA,EACT,QAAQ;AAAA,IACN,QAAQ,QAAQ;AAAA,EAClB;AAAA,EACA,OAAO;AAAA,IACL,QAAQ,YAAY;AAClB,2BAAqB;AAAA,IACvB;AAAA,EACF;AACF,CAAC;AAGD,eAAe,uBAAuB;AACpC,QAAM,YAAY,MAAM,iBAAiB;AAEzC,SAAO,MAAM,OAAO,WAAW,QAAQ;AACvC,SAAO,MAAM,OAAO,UAAU,aAAa,QAAQ;AACnD,SAAO,MAAM,OAAO,UAAU,UAAU,UAAU;AACpD;AAEA,eAAe,SAAS,EAAC,YAAY,MAAM,MAAK,GAAG;AACjD,QAAM,YAAY,MAAM,iBAAiB;AACzC,QAAM,SAAS,IAAI,WAAW,WAAW,aAAa,CAAC;AACvD,QAAM,iBAAiB,WAAW,qBAAqB;AAGvD,MAAI,CAAC,KAAK;AACR,UAAM,SAAS,MAAM,UAAU,SAAS,MAAM;AAC9C,WAAO,UAAU,QAAQ,cAAc;AACvC;AAAA,EACF;AAEA,QAAM,UAAU,IAAI,MAAM;AAC1B,SAAO,UAAU,QAAQ,cAAc;AACzC;",
   "names": []
 }

package/dist/normalize-qualifying-information.test.js

@@ -3,7 +3,6 @@ import { MarcRecord } from "@natlibfi/marc-record";
 import validatorFactory from "./normalize-qualifying-information.js";
 import { READERS } from "@natlibfi/fixura";
 import generateTests from "@natlibfi/fixugen";
-import createDebugLogger from "debug";
 generateTests({
   callback,
   path: [import.meta.dirname, "..", "test-fixtures", "normalize-qualifying-information"],
@@ -18,18 +17,13 @@ generateTests({
     }
   }
 });
-const debug = createDebugLogger("@natlibfi/marc-record-validators-melinda/normalize-qualifying-information:test");
 async function testValidatorFactory() {
   const validator = await validatorFactory();
   assert.equal(typeof validator, "object");
   assert.equal(typeof validator.description, "string");
   assert.equal(typeof validator.validate, "function");
 }
-async function callback({ getFixture, enabled = true, fix = false }) {
-  if (enabled === false) {
-    debug("TEST SKIPPED!");
-    return;
-  }
+async function callback({ getFixture, fix = false }) {
   const validator = await validatorFactory();
   const record = new MarcRecord(getFixture("record.json"));
   const expectedResult = getFixture("expectedResult.json");

package/dist/normalize-qualifying-information.test.js.map

@@ -1,7 +1,7 @@
 {
   "version": 3,
   "sources": ["../src/normalize-qualifying-information.test.js"],
-  "sourcesContent": ["import assert from 'node:assert';\nimport {MarcRecord} from '@natlibfi/marc-record';\nimport validatorFactory from './normalize-qualifying-information.js';\nimport {READERS} from '@natlibfi/fixura';\nimport generateTests from '@natlibfi/fixugen';\nimport createDebugLogger from 'debug';\n\ngenerateTests({\n  callback,\n  path: [import.meta.dirname, '..', 'test-fixtures', 'normalize-qualifying-information'],\n  useMetadataFile: true,\n  recurse: false,\n  fixura: {\n    reader: READERS.JSON\n  },\n  hooks: {\n    before: async () => {\n      testValidatorFactory();\n    }\n  }\n});\nconst debug = createDebugLogger('@natlibfi/marc-record-validators-melinda/normalize-qualifying-information:test');\n\nasync function testValidatorFactory() {\n  const validator = await validatorFactory();\n\n  assert.equal(typeof validator, 'object');\n  assert.equal(typeof validator.description, 'string');\n  assert.equal(typeof validator.validate, 'function');\n}\n\nasync function callback({getFixture, enabled = true, fix = false}) {\n  if (enabled === false) {\n    debug('TEST SKIPPED!');\n    return;\n  }\n\n  const validator = await validatorFactory();\n  const record = new MarcRecord(getFixture('record.json'));\n  const expectedResult = getFixture('expectedResult.json');\n  // console.log(expectedResult); // eslint-disable-line\n\n  if (!fix) {\n    const result = await validator.validate(record);\n    assert.deepEqual(result, expectedResult);\n    return;\n  }\n\n  await validator.fix(record);\n  assert.deepEqual(record, expectedResult);\n}\n"],
-  "mappings": "AAAA,OAAO,YAAY;AACnB,SAAQ,kBAAiB;AACzB,OAAO,sBAAsB;AAC7B,SAAQ,eAAc;AACtB,OAAO,mBAAmB;AAC1B,OAAO,uBAAuB;AAE9B,cAAc;AAAA,EACZ;AAAA,EACA,MAAM,CAAC,YAAY,SAAS,MAAM,iBAAiB,kCAAkC;AAAA,EACrF,iBAAiB;AAAA,EACjB,SAAS;AAAA,EACT,QAAQ;AAAA,IACN,QAAQ,QAAQ;AAAA,EAClB;AAAA,EACA,OAAO;AAAA,IACL,QAAQ,YAAY;AAClB,2BAAqB;AAAA,IACvB;AAAA,EACF;AACF,CAAC;AACD,MAAM,QAAQ,kBAAkB,gFAAgF;AAEhH,eAAe,uBAAuB;AACpC,QAAM,YAAY,MAAM,iBAAiB;AAEzC,SAAO,MAAM,OAAO,WAAW,QAAQ;AACvC,SAAO,MAAM,OAAO,UAAU,aAAa,QAAQ;AACnD,SAAO,MAAM,OAAO,UAAU,UAAU,UAAU;AACpD;AAEA,eAAe,SAAS,EAAC,YAAY,UAAU,MAAM,MAAM,MAAK,GAAG;AACjE,MAAI,YAAY,OAAO;AACrB,UAAM,eAAe;AACrB;AAAA,EACF;AAEA,QAAM,YAAY,MAAM,iBAAiB;AACzC,QAAM,SAAS,IAAI,WAAW,WAAW,aAAa,CAAC;AACvD,QAAM,iBAAiB,WAAW,qBAAqB;AAGvD,MAAI,CAAC,KAAK;AACR,UAAM,SAAS,MAAM,UAAU,SAAS,MAAM;AAC9C,WAAO,UAAU,QAAQ,cAAc;AACvC;AAAA,EACF;AAEA,QAAM,UAAU,IAAI,MAAM;AAC1B,SAAO,UAAU,QAAQ,cAAc;AACzC;",
+  "sourcesContent": ["import assert from 'node:assert';\nimport {MarcRecord} from '@natlibfi/marc-record';\nimport validatorFactory from './normalize-qualifying-information.js';\nimport {READERS} from '@natlibfi/fixura';\nimport generateTests from '@natlibfi/fixugen';\n//import createDebugLogger from 'debug';\n\ngenerateTests({\n  callback,\n  path: [import.meta.dirname, '..', 'test-fixtures', 'normalize-qualifying-information'],\n  useMetadataFile: true,\n  recurse: false,\n  fixura: {\n    reader: READERS.JSON\n  },\n  hooks: {\n    before: async () => {\n      testValidatorFactory();\n    }\n  }\n});\n//const debug = createDebugLogger('@natlibfi/marc-record-validators-melinda/normalize-qualifying-information:test');\n\nasync function testValidatorFactory() {\n  const validator = await validatorFactory();\n\n  assert.equal(typeof validator, 'object');\n  assert.equal(typeof validator.description, 'string');\n  assert.equal(typeof validator.validate, 'function');\n}\n\nasync function callback({getFixture, fix = false}) {\n  const validator = await validatorFactory();\n  const record = new MarcRecord(getFixture('record.json'));\n  const expectedResult = getFixture('expectedResult.json');\n  // console.log(expectedResult); // eslint-disable-line\n\n  if (!fix) {\n    const result = await validator.validate(record);\n    assert.deepEqual(result, expectedResult);\n    return;\n  }\n\n  await validator.fix(record);\n  assert.deepEqual(record, expectedResult);\n}\n"],
+  "mappings": "AAAA,OAAO,YAAY;AACnB,SAAQ,kBAAiB;AACzB,OAAO,sBAAsB;AAC7B,SAAQ,eAAc;AACtB,OAAO,mBAAmB;AAG1B,cAAc;AAAA,EACZ;AAAA,EACA,MAAM,CAAC,YAAY,SAAS,MAAM,iBAAiB,kCAAkC;AAAA,EACrF,iBAAiB;AAAA,EACjB,SAAS;AAAA,EACT,QAAQ;AAAA,IACN,QAAQ,QAAQ;AAAA,EAClB;AAAA,EACA,OAAO;AAAA,IACL,QAAQ,YAAY;AAClB,2BAAqB;AAAA,IACvB;AAAA,EACF;AACF,CAAC;AAGD,eAAe,uBAAuB;AACpC,QAAM,YAAY,MAAM,iBAAiB;AAEzC,SAAO,MAAM,OAAO,WAAW,QAAQ;AACvC,SAAO,MAAM,OAAO,UAAU,aAAa,QAAQ;AACnD,SAAO,MAAM,OAAO,UAAU,UAAU,UAAU;AACpD;AAEA,eAAe,SAAS,EAAC,YAAY,MAAM,MAAK,GAAG;AACjD,QAAM,YAAY,MAAM,iBAAiB;AACzC,QAAM,SAAS,IAAI,WAAW,WAAW,aAAa,CAAC;AACvD,QAAM,iBAAiB,WAAW,qBAAqB;AAGvD,MAAI,CAAC,KAAK;AACR,UAAM,SAAS,MAAM,UAAU,SAAS,MAAM;AAC9C,WAAO,UAAU,QAAQ,cAAc;AACvC;AAAA,EACF;AAEA,QAAM,UAAU,IAAI,MAAM;AAC1B,SAAO,UAAU,QAAQ,cAAc;AACzC;",
   "names": []
 }

package/dist/normalize-utf8-diacritics.js.map

@@ -1,7 +1,7 @@
 {
   "version": 3,
   "sources": ["../src/normalize-utf8-diacritics.js"],
-  "sourcesContent": ["//import createDebugLogger from 'debug';\nimport clone from 'clone';\nimport {convert as nongenericNormalization} from './unicode-decomposition.js';\nimport {fieldToString} from './utils.js';\n\n// Note that https://github.com/NatLibFi/marc-record-validators-melinda/blob/master/src/unicode-decomposition.js contains\n// similar functionalities. It's less generic and lacks diacritic removal but has it advantages as well.\n\n//const debug = createDebugLogger('@natlibfi/marc-record-validators-melinda/normalize-utf-diacritics');\n\n// See also https://github.com/NatLibFi/marc-record-validators-melinda/blob/master/src/unicode-decomposition.js .\n// It uses a list of convertable characters whilst this uses a generic stuff as well.\n// It handles various '.' and '\u00A9' type normalizations as well.\n// NB! This version has minor bug/feature issue regarding fixComposition()\n\n// Author(s): Nicholas Volk\nexport default function () {\n\n  return {\n    description: 'Generic normalization of latin UTF-8 diacritics. Precompose Finnish \u00E5, \u00E4 and \u00F6. Decompose others.',\n    validate, fix\n  };\n\n  function fix(record) {\n    const res = {message: [], fix: [], valid: true};\n    //message.fix = [];\n\n    // Actual parsing of all fields\n    /*\n    if (!record.fields) {\n      return false;\n    }\n    */\n\n    record.fields.forEach(field => {\n      fieldFixComposition(field);\n      //validateField(field, true, message);\n    });\n\n    // message.valid = !(message.message.length >= 1);\n    return res;\n  }\n\n  function validate(record) {\n    const res = {message: []};\n\n    // Actual parsing of all fields\n    /*\n    if (!record.fields) {\n      return false;\n    }\n    */\n\n    record.fields.forEach(field => {\n      validateField(field, res);\n    });\n\n    res.valid = !(res.message.length >= 1);\n    return res;\n  }\n\n  function validateField(field, res) {\n    if (!field.subfields) {\n      return;\n    }\n    const orig = fieldToString(field);\n\n    const normalizedField = fieldFixComposition(clone(field));\n    const mod = fieldToString(normalizedField);\n    if (orig !== mod) { // Fail as the input is \"broken\"/\"crap\"/sumthing\n      res.message.push(`'${orig}' requires normalization`);\n      return;\n    }\n    return;\n  }\n}\n\n\n// Traditionally these six are precomposed and all the rest decomposed\nexport function precomposeFinnishLetters(value = '') {\n  return value.\n    replace(/a\u030A/gu, '\u00E5').\n    replace(/a\u0308/gu, '\u00E4').\n    replace(/o\u0308/gu, '\u00F6').\n    replace(/A\u030A/gu, '\u00C5').\n    replace(/A\u0308/gu, '\u00C4').\n    replace(/O\u0308/gu, '\u00D6');\n}\n\nexport function fixComposition(value = '') {\n  // Target: Diacritics use Melinda internal notation.\n  // General solution: Decompose everything and then compose '\u00E5', '\u00E4', '\u00F6', '\u00C5', '\u00C4' and '\u00D6'.\n  // https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Global_Objects/String/normalize\n  // Bug/Feature: the generic normalize() function also normalizes non-latin encodings as well, is this ok?\n  // Exception: Input contains non-Latin script letters: don't decompose (see field 880 tests):\n  if (value.match(/[^\\p{Script=Latin}\\p{Script=Common}\\p{Script=Inherited}]/u)) {\n    // Problem with this approach: mixed language content (eg. cyrillic + latin) won't get normalized.\n    // Hack/Damage control: we might add decomposition rules for most common diacritis here (eg. \u00FC, \u00E9...).\n    // OR we could split input to words and handle them separately?\n    // NB! Hack not implemented yet. The main source of problematic case would probably be greek characters\n    // within texts, that are written with latin alphabet.\n    //return precomposeFinnishLetters(value);\n    return nongenericNormalization(value);\n  }\n  return precomposeFinnishLetters(String(value).normalize('NFD'));\n}\n\n\nexport function fieldFixComposition(field) {\n  if (!field.subfields) {\n    return field;\n  }\n  //const originalValue = fieldToString(field);\n  //nvdebug(`fFC: '${originalValue}'`, debug);\n  field.subfields.forEach((subfield, index) => {\n    field.subfields[index].value = fixComposition(subfield.value);\n  });\n  //const newValue = fieldToString(field);\n  //if (originalValue !== newValue) {\n  //  debug(`FIXCOMP: '${originalValue}' => '${newValue}'`);\n  //}\n  return field;\n}\n\n/*\nexport function 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    sf.value = removeDecomposedDiacritics(sf.value);\n  });\n\n  function removeDecomposedDiacritics(value = '') {\n    // NB #1: Does nothing to precomposed letters. String.normalize('NFD') can handle them.\n    // NB #2: Finnish letters '\u00E5', '\u00E4', '\u00F6', '\u00C5', \u00C4', and '\u00D6' should be handled (=precomposed) before calling this.\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}\n*/\n\n"],
-  "mappings": "AACA,OAAO,WAAW;AAClB,SAAQ,WAAW,+BAA8B;AACjD,SAAQ,qBAAoB;AAa5B,0BAA2B;AAEzB,SAAO;AAAA,IACL,aAAa;AAAA,IACb;AAAA,IAAU;AAAA,EACZ;AAEA,WAAS,IAAI,QAAQ;AACnB,UAAM,MAAM,EAAC,SAAS,CAAC,GAAG,KAAK,CAAC,GAAG,OAAO,KAAI;AAU9C,WAAO,OAAO,QAAQ,WAAS;AAC7B,0BAAoB,KAAK;AAAA,IAE3B,CAAC;AAGD,WAAO;AAAA,EACT;AAEA,WAAS,SAAS,QAAQ;AACxB,UAAM,MAAM,EAAC,SAAS,CAAC,EAAC;AASxB,WAAO,OAAO,QAAQ,WAAS;AAC7B,oBAAc,OAAO,GAAG;AAAA,IAC1B,CAAC;AAED,QAAI,QAAQ,EAAE,IAAI,QAAQ,UAAU;AACpC,WAAO;AAAA,EACT;AAEA,WAAS,cAAc,OAAO,KAAK;AACjC,QAAI,CAAC,MAAM,WAAW;AACpB;AAAA,IACF;AACA,UAAM,OAAO,cAAc,KAAK;AAEhC,UAAM,kBAAkB,oBAAoB,MAAM,KAAK,CAAC;AACxD,UAAM,MAAM,cAAc,eAAe;AACzC,QAAI,SAAS,KAAK;AAChB,UAAI,QAAQ,KAAK,IAAI,IAAI,0BAA0B;AACnD;AAAA,IACF;AACA;AAAA,EACF;AACF;AAIO,gBAAS,yBAAyB,QAAQ,IAAI;AACnD,SAAO,MACL,QAAQ,QAAQ,MAAG,EACnB,QAAQ,QAAQ,MAAG,EACnB,QAAQ,QAAQ,MAAG,EACnB,QAAQ,QAAQ,MAAG,EACnB,QAAQ,QAAQ,MAAG,EACnB,QAAQ,QAAQ,MAAG;AACvB;AAEO,gBAAS,eAAe,QAAQ,IAAI;AAMzC,MAAI,MAAM,MAAM,2DAA2D,GAAG;AAO5E,WAAO,wBAAwB,KAAK;AAAA,EACtC;AACA,SAAO,yBAAyB,OAAO,KAAK,EAAE,UAAU,KAAK,CAAC;AAChE;AAGO,gBAAS,oBAAoB,OAAO;AACzC,MAAI,CAAC,MAAM,WAAW;AACpB,WAAO;AAAA,EACT;AAGA,QAAM,UAAU,QAAQ,CAAC,UAAU,UAAU;AAC3C,UAAM,UAAU,KAAK,EAAE,QAAQ,eAAe,SAAS,KAAK;AAAA,EAC9D,CAAC;AAKD,SAAO;AACT;",
+  "sourcesContent": ["//import createDebugLogger from 'debug';\nimport clone from 'clone';\nimport {convert as nongenericNormalization} from './unicode-decomposition.js';\nimport {fieldToString /*, nvdebug */} from './utils.js';\n\n// Note that https://github.com/NatLibFi/marc-record-validators-melinda/blob/master/src/unicode-decomposition.js contains\n// similar functionalities. It's less generic and lacks diacritic removal but has it advantages as well.\n\n//const debug = createDebugLogger('@natlibfi/marc-record-validators-melinda/normalize-utf-diacritics');\n//const debugData = debug.extend('data');\n//const debugDev = debug.extend('dev');\n\n\n// See also https://github.com/NatLibFi/marc-record-validators-melinda/blob/master/src/unicode-decomposition.js .\n// It uses a list of convertable characters whilst this uses a generic stuff as well.\n// It handles various '.' and '\u00A9' type normalizations as well.\n// NB! This version has minor bug/feature issue regarding fixComposition()\n\n// Author(s): Nicholas Volk\nexport default function () {\n\n  return {\n    description: 'Generic normalization of latin UTF-8 diacritics. Precompose Finnish \u00E5, \u00E4 and \u00F6. Decompose others.',\n    validate, fix\n  };\n\n  function fix(record) {\n    const res = {message: [], fix: [], valid: true};\n    //message.fix = [];\n\n    // Actual parsing of all fields\n    /*\n    if (!record.fields) {\n      return false;\n    }\n    */\n\n    record.fields.forEach(field => {\n      fieldFixComposition(field);\n      //validateField(field, true, message);\n    });\n\n    // message.valid = !(message.message.length >= 1);\n    return res;\n  }\n\n  function validate(record) {\n    const res = {message: []};\n\n    // Actual parsing of all fields\n    /*\n    if (!record.fields) {\n      return false;\n    }\n    */\n\n    record.fields.forEach(field => {\n      validateField(field, res);\n    });\n\n    res.valid = !(res.message.length >= 1);\n    return res;\n  }\n\n  function validateField(field, res) {\n    if (!field.subfields) {\n      return;\n    }\n    const orig = fieldToString(field);\n\n    const normalizedField = fieldFixComposition(clone(field));\n    const mod = fieldToString(normalizedField);\n    if (orig !== mod) { // Fail as the input is \"broken\"/\"crap\"/sumthing\n      res.message.push(`'${orig}' requires normalization`);\n      return;\n    }\n    return;\n  }\n}\n\n\n// Traditionally these six are precomposed and all the rest decomposed\nexport function precomposeFinnishLetters(value = '') {\n  return value.\n    replace(/a\u030A/gu, '\u00E5').\n    replace(/a\u0308/gu, '\u00E4').\n    replace(/o\u0308/gu, '\u00F6').\n    replace(/A\u030A/gu, '\u00C5').\n    replace(/A\u0308/gu, '\u00C4').\n    replace(/O\u0308/gu, '\u00D6');\n}\n\nexport function fixComposition(value = '') {\n  // Target: Diacritics use Melinda internal notation.\n  // General solution: Decompose everything and then compose '\u00E5', '\u00E4', '\u00F6', '\u00C5', '\u00C4' and '\u00D6'.\n  // https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Global_Objects/String/normalize\n  // Bug/Feature: the generic normalize() function also normalizes non-latin encodings as well, is this ok?\n  // Exception: Input contains non-Latin script letters: don't decompose (see field 880 tests):\n  if (value.match(/[^\\p{Script=Latin}\\p{Script=Common}\\p{Script=Inherited}]/u)) {\n    // Problem with this approach: mixed language content (eg. cyrillic + latin) won't get normalized.\n    // Hack/Damage control: we might add decomposition rules for most common diacritis here (eg. \u00FC, \u00E9...).\n    // OR we could split input to words and handle them separately?\n    // NB! Hack not implemented yet. The main source of problematic case would probably be greek characters\n    // within texts, that are written with latin alphabet.\n    //return precomposeFinnishLetters(value);\n    return nongenericNormalization(value);\n  }\n  return precomposeFinnishLetters(String(value).normalize('NFD'));\n}\n\n\nexport function fieldFixComposition(field) {\n  if (!field.subfields) {\n    return field;\n  }\n  //const originalValue = fieldToString(field);\n  //nvdebug(`fFC: '${originalValue}'`, debugDev);\n  field.subfields.forEach((subfield, index) => {\n    field.subfields[index].value = fixComposition(subfield.value);\n  });\n  //const newValue = fieldToString(field);\n  //if (originalValue !== newValue) {\n  //  nvdebug(`FIXCOMP: '${originalValue}' => '${newValue}'`, debugDev);\n  //}\n  return field;\n}\n\n/*\nexport function 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    sf.value = removeDecomposedDiacritics(sf.value);\n  });\n\n  function removeDecomposedDiacritics(value = '') {\n    // NB #1: Does nothing to precomposed letters. String.normalize('NFD') can handle them.\n    // NB #2: Finnish letters '\u00E5', '\u00E4', '\u00F6', '\u00C5', \u00C4', and '\u00D6' should be handled (=precomposed) before calling this.\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}\n*/\n\n"],
+  "mappings": "AACA,OAAO,WAAW;AAClB,SAAQ,WAAW,+BAA8B;AACjD,SAAQ,qBAAmC;AAgB3C,0BAA2B;AAEzB,SAAO;AAAA,IACL,aAAa;AAAA,IACb;AAAA,IAAU;AAAA,EACZ;AAEA,WAAS,IAAI,QAAQ;AACnB,UAAM,MAAM,EAAC,SAAS,CAAC,GAAG,KAAK,CAAC,GAAG,OAAO,KAAI;AAU9C,WAAO,OAAO,QAAQ,WAAS;AAC7B,0BAAoB,KAAK;AAAA,IAE3B,CAAC;AAGD,WAAO;AAAA,EACT;AAEA,WAAS,SAAS,QAAQ;AACxB,UAAM,MAAM,EAAC,SAAS,CAAC,EAAC;AASxB,WAAO,OAAO,QAAQ,WAAS;AAC7B,oBAAc,OAAO,GAAG;AAAA,IAC1B,CAAC;AAED,QAAI,QAAQ,EAAE,IAAI,QAAQ,UAAU;AACpC,WAAO;AAAA,EACT;AAEA,WAAS,cAAc,OAAO,KAAK;AACjC,QAAI,CAAC,MAAM,WAAW;AACpB;AAAA,IACF;AACA,UAAM,OAAO,cAAc,KAAK;AAEhC,UAAM,kBAAkB,oBAAoB,MAAM,KAAK,CAAC;AACxD,UAAM,MAAM,cAAc,eAAe;AACzC,QAAI,SAAS,KAAK;AAChB,UAAI,QAAQ,KAAK,IAAI,IAAI,0BAA0B;AACnD;AAAA,IACF;AACA;AAAA,EACF;AACF;AAIO,gBAAS,yBAAyB,QAAQ,IAAI;AACnD,SAAO,MACL,QAAQ,QAAQ,MAAG,EACnB,QAAQ,QAAQ,MAAG,EACnB,QAAQ,QAAQ,MAAG,EACnB,QAAQ,QAAQ,MAAG,EACnB,QAAQ,QAAQ,MAAG,EACnB,QAAQ,QAAQ,MAAG;AACvB;AAEO,gBAAS,eAAe,QAAQ,IAAI;AAMzC,MAAI,MAAM,MAAM,2DAA2D,GAAG;AAO5E,WAAO,wBAAwB,KAAK;AAAA,EACtC;AACA,SAAO,yBAAyB,OAAO,KAAK,EAAE,UAAU,KAAK,CAAC;AAChE;AAGO,gBAAS,oBAAoB,OAAO;AACzC,MAAI,CAAC,MAAM,WAAW;AACpB,WAAO;AAAA,EACT;AAGA,QAAM,UAAU,QAAQ,CAAC,UAAU,UAAU;AAC3C,UAAM,UAAU,KAAK,EAAE,QAAQ,eAAe,SAAS,KAAK;AAAA,EAC9D,CAAC;AAKD,SAAO;AACT;",
   "names": []
 }

package/dist/normalize-utf8-diacritics.test.js

@@ -3,7 +3,6 @@ import { MarcRecord } from "@natlibfi/marc-record";
 import validatorFactory from "./normalize-utf8-diacritics.js";
 import { READERS } from "@natlibfi/fixura";
 import generateTests from "@natlibfi/fixugen";
-import createDebugLogger from "debug";
 generateTests({
   callback,
   path: [import.meta.dirname, "..", "test-fixtures", "normalize-utf8-diacritics"],
@@ -18,18 +17,13 @@ generateTests({
     }
   }
 });
-const debug = createDebugLogger("@natlibfi/marc-record-validators-melinda/normalize-utf8-diacritics:test");
 async function testValidatorFactory() {
   const validator = await validatorFactory();
   assert.equal(typeof validator, "object");
   assert.equal(typeof validator.description, "string");
   assert.equal(typeof validator.validate, "function");
 }
-async function callback({ getFixture, enabled = true, fix = false }) {
-  if (enabled === false) {
-    debug("TEST SKIPPED!");
-    return;
-  }
+async function callback({ getFixture, fix = false }) {
   const validator = await validatorFactory();
   const record = new MarcRecord(getFixture("record.json"));
   const expectedResult = getFixture("expectedResult.json");

package/dist/normalize-utf8-diacritics.test.js.map

@@ -1,7 +1,7 @@
 {
   "version": 3,
   "sources": ["../src/normalize-utf8-diacritics.test.js"],
-  "sourcesContent": ["import assert from 'node:assert';\nimport {MarcRecord} from '@natlibfi/marc-record';\nimport validatorFactory from './normalize-utf8-diacritics.js';\nimport {READERS} from '@natlibfi/fixura';\nimport generateTests from '@natlibfi/fixugen';\nimport createDebugLogger from 'debug';\n\ngenerateTests({\n  callback,\n  path: [import.meta.dirname, '..', 'test-fixtures', 'normalize-utf8-diacritics'],\n  useMetadataFile: true,\n  recurse: false,\n  fixura: {\n    reader: READERS.JSON\n  },\n  hooks: {\n    before: async () => {\n      testValidatorFactory();\n    }\n  }\n});\n\nconst debug = createDebugLogger('@natlibfi/marc-record-validators-melinda/normalize-utf8-diacritics:test');\n\nasync function testValidatorFactory() {\n  const validator = await validatorFactory();\n\n  assert.equal(typeof validator, 'object');\n  assert.equal(typeof validator.description, 'string');\n  assert.equal(typeof validator.validate, 'function');\n}\n\nasync function callback({getFixture, enabled = true, fix = false}) {\n  if (enabled === false) {\n    debug('TEST SKIPPED!');\n    return;\n  }\n\n  const validator = await validatorFactory();\n  const record = new MarcRecord(getFixture('record.json'));\n  const expectedResult = getFixture('expectedResult.json');\n  // console.log(expectedResult); // eslint-disable-line\n\n  if (!fix) {\n    const result = await validator.validate(record);\n    assert.deepEqual(result, expectedResult);\n    return;\n  }\n\n  await validator.fix(record);\n  assert.deepEqual(record, expectedResult);\n}\n"],
-  "mappings": "AAAA,OAAO,YAAY;AACnB,SAAQ,kBAAiB;AACzB,OAAO,sBAAsB;AAC7B,SAAQ,eAAc;AACtB,OAAO,mBAAmB;AAC1B,OAAO,uBAAuB;AAE9B,cAAc;AAAA,EACZ;AAAA,EACA,MAAM,CAAC,YAAY,SAAS,MAAM,iBAAiB,2BAA2B;AAAA,EAC9E,iBAAiB;AAAA,EACjB,SAAS;AAAA,EACT,QAAQ;AAAA,IACN,QAAQ,QAAQ;AAAA,EAClB;AAAA,EACA,OAAO;AAAA,IACL,QAAQ,YAAY;AAClB,2BAAqB;AAAA,IACvB;AAAA,EACF;AACF,CAAC;AAED,MAAM,QAAQ,kBAAkB,yEAAyE;AAEzG,eAAe,uBAAuB;AACpC,QAAM,YAAY,MAAM,iBAAiB;AAEzC,SAAO,MAAM,OAAO,WAAW,QAAQ;AACvC,SAAO,MAAM,OAAO,UAAU,aAAa,QAAQ;AACnD,SAAO,MAAM,OAAO,UAAU,UAAU,UAAU;AACpD;AAEA,eAAe,SAAS,EAAC,YAAY,UAAU,MAAM,MAAM,MAAK,GAAG;AACjE,MAAI,YAAY,OAAO;AACrB,UAAM,eAAe;AACrB;AAAA,EACF;AAEA,QAAM,YAAY,MAAM,iBAAiB;AACzC,QAAM,SAAS,IAAI,WAAW,WAAW,aAAa,CAAC;AACvD,QAAM,iBAAiB,WAAW,qBAAqB;AAGvD,MAAI,CAAC,KAAK;AACR,UAAM,SAAS,MAAM,UAAU,SAAS,MAAM;AAC9C,WAAO,UAAU,QAAQ,cAAc;AACvC;AAAA,EACF;AAEA,QAAM,UAAU,IAAI,MAAM;AAC1B,SAAO,UAAU,QAAQ,cAAc;AACzC;",
+  "sourcesContent": ["import assert from 'node:assert';\nimport {MarcRecord} from '@natlibfi/marc-record';\nimport validatorFactory from './normalize-utf8-diacritics.js';\nimport {READERS} from '@natlibfi/fixura';\nimport generateTests from '@natlibfi/fixugen';\n//import createDebugLogger from 'debug';\n\ngenerateTests({\n  callback,\n  path: [import.meta.dirname, '..', 'test-fixtures', 'normalize-utf8-diacritics'],\n  useMetadataFile: true,\n  recurse: false,\n  fixura: {\n    reader: READERS.JSON\n  },\n  hooks: {\n    before: async () => {\n      testValidatorFactory();\n    }\n  }\n});\n\n//const debug = createDebugLogger('@natlibfi/marc-record-validators-melinda/normalize-utf8-diacritics:test');\n\nasync function testValidatorFactory() {\n  const validator = await validatorFactory();\n\n  assert.equal(typeof validator, 'object');\n  assert.equal(typeof validator.description, 'string');\n  assert.equal(typeof validator.validate, 'function');\n}\n\nasync function callback({getFixture, fix = false}) {\n  const validator = await validatorFactory();\n  const record = new MarcRecord(getFixture('record.json'));\n  const expectedResult = getFixture('expectedResult.json');\n  // console.log(expectedResult); // eslint-disable-line\n\n  if (!fix) {\n    const result = await validator.validate(record);\n    assert.deepEqual(result, expectedResult);\n    return;\n  }\n\n  await validator.fix(record);\n  assert.deepEqual(record, expectedResult);\n}\n"],
+  "mappings": "AAAA,OAAO,YAAY;AACnB,SAAQ,kBAAiB;AACzB,OAAO,sBAAsB;AAC7B,SAAQ,eAAc;AACtB,OAAO,mBAAmB;AAG1B,cAAc;AAAA,EACZ;AAAA,EACA,MAAM,CAAC,YAAY,SAAS,MAAM,iBAAiB,2BAA2B;AAAA,EAC9E,iBAAiB;AAAA,EACjB,SAAS;AAAA,EACT,QAAQ;AAAA,IACN,QAAQ,QAAQ;AAAA,EAClB;AAAA,EACA,OAAO;AAAA,IACL,QAAQ,YAAY;AAClB,2BAAqB;AAAA,IACvB;AAAA,EACF;AACF,CAAC;AAID,eAAe,uBAAuB;AACpC,QAAM,YAAY,MAAM,iBAAiB;AAEzC,SAAO,MAAM,OAAO,WAAW,QAAQ;AACvC,SAAO,MAAM,OAAO,UAAU,aAAa,QAAQ;AACnD,SAAO,MAAM,OAAO,UAAU,UAAU,UAAU;AACpD;AAEA,eAAe,SAAS,EAAC,YAAY,MAAM,MAAK,GAAG;AACjD,QAAM,YAAY,MAAM,iBAAiB;AACzC,QAAM,SAAS,IAAI,WAAW,WAAW,aAAa,CAAC;AACvD,QAAM,iBAAiB,WAAW,qBAAqB;AAGvD,MAAI,CAAC,KAAK;AACR,UAAM,SAAS,MAAM,UAAU,SAAS,MAAM;AAC9C,WAAO,UAAU,QAAQ,cAAc;AACvC;AAAA,EACF;AAEA,QAAM,UAAU,IAAI,MAAM;AAC1B,SAAO,UAAU,QAAQ,cAAc;AACzC;",
   "names": []
 }

package/dist/normalizeFieldForComparison.js

@@ -1,20 +1,13 @@
 import clone from "clone";
 import { fieldStripPunctuation } from "./punctuation2.js";
-import { fieldToString, isControlSubfieldCode } from "./utils.js";
+import { fieldToString, isContentSubfieldCode } from "./utils.js";
 import { fieldNormalizeControlNumbers } from "./normalize-identifiers.js";
 import createDebugLogger from "debug";
 import { normalizePartData, subfieldContainsPartData } from "./normalizeSubfieldValueForComparison.js";
+import { isEnnakkotietoSubfield } from "./prepublicationUtils.js";
+import { getSynonym } from "./merge-fields/worldKnowledge.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\.?$/gui);
-}
 function debugFieldComparison(oldField, newField) {
   const oldString = fieldToString(oldField);
   const newString = fieldToString(newField);
@@ -39,13 +32,13 @@ function containsCorporateName(tag = "???", subfieldCode = void 0) {
   return false;
 }
 function skipAllSubfieldNormalizations(value, subfieldCode, tag) {
-  if (isEnnakkotietoSubfieldG({ "code": subfieldCode, value })) {
+  if (isEnnakkotietoSubfield({ "code": subfieldCode, value })) {
     return true;
   }
   if (tag === "035" && ["a", "z"].includes(subfieldCode)) {
     return true;
   }
-  if (isControlSubfieldCode(subfieldCode)) {
+  if (!isContentSubfieldCode(subfieldCode, tag)) {
     return true;
   }
   return false;
@@ -93,6 +86,21 @@ function fieldLowercase(field) {
     return false;
   }
 }
+function normalize300SubfieldA(field) {
+  if (field.tag !== "300") {
+    return;
+  }
+  field.subfields.forEach((sf) => normalizePages(sf));
+  function normalizePages(subfield) {
+    if (valuelessSubfield(subfield)) {
+      return;
+    }
+    if (subfield.code !== "a") {
+      return;
+    }
+    subfield.value = subfield.value.replace(/(?:\b(?:pages?|pp\.|Seite|sida|sidor|sivu|sivua)\b|\bs$)/g, "s.");
+  }
+}
 function hack490SubfieldA(field) {
   if (field.tag !== "490") {
     return;
@@ -143,6 +151,7 @@ function normalizeISBN(field) {
 }
 function fieldSpecificHacks(field) {
   normalizeISBN(field);
+  normalize300SubfieldA(field);
   hack490SubfieldA(field);
 }
 export function fieldTrimSubfieldValues(field) {
@@ -167,6 +176,8 @@ function removeDecomposedDiacritics(value = "") {
   return String(value).replace(/\p{Diacritic}/gu, "");
 }
 function normalizeSubfieldValue(value, subfieldCode, tag) {
+  value = removeCharsThatDontCarryMeaning(value, tag, subfieldCode);
+  value = getSynonym(tag, subfieldCode, value);
   value = subfieldValueLowercase(value, subfieldCode, tag);
   value = normalizePartData(value, subfieldCode, tag);
   value = value.replace(/^\[([^[\]]+)\]/gu, "$1");
@@ -203,7 +214,7 @@ function normalizeField(field) {
   return field;
 }
 export function cloneAndNormalizeFieldForComparison(field) {
-  const clonedField = clone(field);
+  const clonedField = cloneAndRemovePunctuation(field);
   if (fieldSkipNormalization(field)) {
     return clonedField;
   }
@@ -212,7 +223,6 @@ export function cloneAndNormalizeFieldForComparison(field) {
       return;
     }
     sf.value = normalizeSubfieldValue(sf.value, sf.code, field.tag);
-    sf.value = removeCharsThatDontCarryMeaning(sf.value, field.tag, sf.code);
   });
   normalizeField(clonedField);
   fieldRemoveDecomposedDiacritics(clonedField);

package/dist/normalizeFieldForComparison.js.map

@@ -1,7 +1,7 @@
 {
   "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;",
+  "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, isContentSubfieldCode} from './utils.js';\n\nimport {fieldNormalizeControlNumbers/*, normalizeControlSubfieldValue*/} from './normalize-identifiers.js';\nimport createDebugLogger from 'debug';\nimport {normalizePartData, subfieldContainsPartData} from './normalizeSubfieldValueForComparison.js';\nimport {isEnnakkotietoSubfield} from './prepublicationUtils.js';\nimport {getSynonym} from './merge-fields/worldKnowledge.js';\n\nconst debug = createDebugLogger('@natlibfi/melinda-marc-record-merge-reducers:normalizeFieldForComparison');\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) {\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 410... 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 (isEnnakkotietoSubfield({'code': subfieldCode, value})) {\n    return true;\n  }\n\n  if (tag === '035' && ['a', 'z'].includes(subfieldCode)) {\n    return true;\n  }\n\n  if (!isContentSubfieldCode(subfieldCode, tag)) {\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} '${newValue}'`, debugDev);\n    return newValue;\n  }\n  return value;\n}\n\nfunction subfieldLowercase(sf, tag) {\n  if (valuelessSubfield(sf)) {\n    return;\n  }\n  sf.value = subfieldValueLowercase(sf.value, sf.code, tag);\n}\n\nfunction fieldLowercase(field) {\n  if (skipFieldLowercase(field)) {\n    return;\n  }\n\n  field.subfields.forEach(sf => subfieldLowercase(sf, field.tag));\n\n  function skipFieldLowercase(field) {\n    if (skipAllFieldNormalizations(field.tag)) {\n      return true;\n    }\n    // Skip non-interesting fields\n    if (!containsHumanName(field.tag) && !containsCorporateName(field.tag) && !['240', '245', '630'].includes(field.tag)) {\n      return true;\n    }\n\n    return false;\n  }\n}\n\nfunction normalize300SubfieldA(field) {\n  if (field.tag !== '300') {\n    return;\n  }\n  field.subfields.forEach(sf => normalizePages(sf));\n\n  function normalizePages(subfield) {\n    if (valuelessSubfield(subfield)) {\n      return;\n    }\n\n    if (subfield.code !== 'a') {\n      return;\n    }\n    subfield.value = subfield.value.replace(/(?:\\b(?:pages?|pp\\.|Seite|sida|sidor|sivu|sivua)\\b|\\bs$)/g, 's.');\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)}`, debugDev);\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? ${subfieldToString(sf)}`, debugDev);\n    sf.value = sf.value.replace(/-/ug, '');\n    sf.value = sf.value.replace(/x/u, 'X');\n  }\n\n}\n\nfunction fieldSpecificHacks(field) {\n  normalizeISBN(field); // 020$a, not $z!\n  normalize300SubfieldA(field);\n  hack490SubfieldA(field);\n}\n\nexport function fieldTrimSubfieldValues(field) {\n  field.subfields?.forEach((sf) => {\n    if (valuelessSubfield(sf)) {\n      return;\n    }\n    sf.value = sf.value.replace(/^[ \\t\\n]+/u, '');\n    sf.value = sf.value.replace(/[ \\t\\n]+$/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  /* beslint-disable */\n  value = removeCharsThatDontCarryMeaning(value, tag, subfieldCode);\n  value = getSynonym(tag, subfieldCode, value); // Must be done before punc stripping and lowercasing...\n  value = subfieldValueLowercase(value, subfieldCode, tag);\n\n\n  // Normalize: s. = sivut = pp.\n  value = normalizePartData(value, subfieldCode, tag);\n  value = value.replace(/^\\[([^[\\]]+)\\]/gu, '$1');\n\n  if (['130', '730'].includes(tag) && subfieldCode === 'a') {\n    value = value.replace(' : ', ', '); // \"Halloween ends (elokuva, 2022)\" vs \"Halloween ends (elokuva : 2023)\"\n  }\n  /* beslint-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 = cloneAndRemovePunctuation(field); // was only clone(field)\n\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\n    sf.value = normalizeSubfieldValue(sf.value, sf.code, field.tag);\n    //sf.value = normalizeForSamenessCheck(field.tag, sf.code, sf.value);\n\n\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;AAC1D,SAAQ,8BAA6B;AACrC,SAAQ,kBAAiB;AAEzB,MAAM,QAAQ,kBAAkB,0EAA0E;AAE1G,MAAM,WAAW,MAAM,OAAO,KAAK;AAEnC,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,uBAAuB,EAAC,QAAQ,cAAc,MAAK,CAAC,GAAG;AACzD,WAAO;AAAA,EACT;AAEA,MAAI,QAAQ,SAAS,CAAC,KAAK,GAAG,EAAE,SAAS,YAAY,GAAG;AACtD,WAAO;AAAA,EACT;AAEA,MAAI,CAAC,sBAAsB,cAAc,GAAG,GAAG;AAC7C,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;AAEA,SAAS,sBAAsB,OAAO;AACpC,MAAI,MAAM,QAAQ,OAAO;AACvB;AAAA,EACF;AACA,QAAM,UAAU,QAAQ,QAAM,eAAe,EAAE,CAAC;AAEhD,WAAS,eAAe,UAAU;AAChC,QAAI,kBAAkB,QAAQ,GAAG;AAC/B;AAAA,IACF;AAEA,QAAI,SAAS,SAAS,KAAK;AACzB;AAAA,IACF;AACA,aAAS,QAAQ,SAAS,MAAM,QAAQ,6DAA6D,IAAI;AAAA,EAC3G;AACF;AAEA,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,wBAAsB,KAAK;AAC3B,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,gCAAgC,OAAO,KAAK,YAAY;AAChE,UAAQ,WAAW,KAAK,cAAc,KAAK;AAC3C,UAAQ,uBAAuB,OAAO,cAAc,GAAG;AAIvD,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,0BAA0B,KAAK;AAEnD,MAAI,uBAAuB,KAAK,GAAG;AACjC,WAAO;AAAA,EACT;AACA,cAAY,UAAU,QAAQ,CAAC,OAAO;AACpC,QAAI,kBAAkB,EAAE,GAAG;AACzB;AAAA,IACF;AAEA,OAAG,QAAQ,uBAAuB,GAAG,OAAO,GAAG,MAAM,MAAM,GAAG;AAAA,EAIhE,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"]
 }

package/dist/normalizeSubfieldValueForComparison.js.map

@@ -1,7 +1,7 @@
 {
   "version": 3,
   "sources": ["../src/normalizeSubfieldValueForComparison.js"],
-  "sourcesContent": ["import {nvdebug} from './utils.js';\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:normalizeSubfieldValueForComparison');\n//const debugData = debug.extend('data');\nconst debugDev = debug.extend('dev');\n\nexport function subfieldContainsPartData(tag, subfieldCode) {\n  // NB! Used by reducers' mergeSubield.js\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  // This a very hacky function, but cand really help it, as the the data is very iffy as well...\n  // Remove punctuation and brackets:\n  const value = originalValue.replace(/[-.,:; ]+$/ui, '').replace(/^\\[([0-9]+)\\]$/ui, '$1');\n\n  const [year, rest] = extractYear(value);\n\n  const splitPoint = rest.lastIndexOf(' '); // MRA-627: \"5, 2017\" should be split here. Think of this later on...\n  if (splitPoint === -1) {\n    return [undefined, year, rest];\n  }\n  const lhs = rest.substr(0, splitPoint);\n  const rhs = rest.substr(splitPoint + 1);\n  return [lhs, year, rhs];\n\n  function extractYear(value) {\n    // NB! Note that this is far for perfect. It cover just some very common cases...\n\n    // \"2023, 3\" => [\"2023\", \"3\"]\n    if (value.match(/^(?:1[89][0-9][0-9]|20[012][0-9]), (?:nro |n:o)?[1-9][0-9]{0,2}$/ui)) {\n      return [value.substr(0, 4), value.substr(6)];\n    }\n    // \"2023/12\" => [\"2023\", \"12\"]\n    if (value.match(/^(?:1[89][0-9][0-9]|20[012][0-9])[/:][1-9][0-9]{0,2}$/u)) {\n      return [value.substr(0, 4), value.substr(5)];\n    }\n    // \"Vol. 3/2023\" => [\"2023\", \"Vol. 3\"]\n    if (value.match(/^[^0-9]*[1-9][0-9]{0,2}\\/(?:1[89][0-9][0-9]|20[012][0-9])$/u)) {\n      const len = value.length;\n      return [value.substr(len - 4), value.substr(0, len - 5)];\n    }\n\n\n    return [undefined, value];\n  }\n}\n\nfunction normalizePartType(originalValue) {\n  if (originalValue === undefined) {\n    return undefined;\n  }\n  const value = originalValue.toLowerCase();\n\n  // Return Finnish singular nominative. Choise of language is arbitrary. This is best-ish for debug purposes...\n  if (['n:o', 'no', 'nr', 'nro', 'number', 'numero', 'nummer'].includes(value)) {\n    return 'numero';\n  }\n  if (['band', 'bd', 'h\u00E4fte', 'nide', 'osa', 'part', 'teil', 'vol', 'vol.', 'volume'].includes(value)) {\n    return 'osa';\n  }\n\n  if (['p.', 'page', 'pages', 'pp.', 's.', 'sidor', 'sivu', 'sivut'].includes(value)) {\n    return 'sivu';\n  }\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 ${arabicValue}`, debugDev);\n    return arabicValue;\n  }\n  return value.toLowerCase();\n}\n\nfunction splitAndNormalizePartData(value) {\n  // This is just a stub. Does not handle eg. \"Levy 2, raita 15\"\n  const [partType, partYear, partNumber] = splitPartData(value);\n  //nvdebug(`  LHS: '${lhs}'`, debugDev);\n  //nvdebug(`  RHS: '${rhs}'`, debugDev);\n  return [normalizePartType(partType), partYear, normalizePartNumber(partNumber)];\n}\n\nexport function partsAgree(value1, value2, tag, subfieldCode) {\n  // Note, that parts can not be normalized away, as \"2\" can agree with \"Part 2\" and \"Raita 2\" and \"Volume 2\"...\n  // NB! Used by reducers' mergeSubield.js\n  if (!subfieldContainsPartData(tag, subfieldCode)) {\n    return false;\n  }\n  const [partType1, partYear1, partNumber1] = splitAndNormalizePartData(value1);\n  const [partType2, partYear2, partNumber2] = splitAndNormalizePartData(value2);\n  //nvdebug(`P1: ${partType1} | ${partYear1} | ${partNumber1}`);\n  //nvdebug(`P2: ${partType2} | ${partYear2} | ${partNumber2}`);\n  if (partNumber1 !== partNumber2) {\n    return false;\n  }\n  if (partType1 !== undefined && partType2 !== undefined && partType1 !== partType2) {\n    return false;\n  }\n  if (partYear1 !== undefined && partYear2 !== undefined && partYear1 !== partYear2) {\n    return false;\n  }\n\n\n  return true;\n}\n\nexport function normalizePartData(value, subfieldCode, tag) {\n  // This is for normalizing values for equality comparison only!\n  if (!subfieldContainsPartData(tag, subfieldCode)) {\n    return value;\n  }\n\n  const [partType, partYear, partNumber] = splitAndNormalizePartData(value);\n  if (partType === undefined) {\n    if (partYear === undefined) {\n      return partNumber;\n    }\n    return `${partNumber}/${partYear}`;\n  }\n  if (partYear === undefined) {\n    return `${partType} ${partNumber}`;\n  }\n  return `${partType} ${partNumber}/${partYear}`;\n}\n"],
+  "sourcesContent": ["import {nvdebug} from './utils.js';\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:normalizeSubfieldValueForComparison');\n//const debugData = debug.extend('data');\nconst debugDev = debug.extend('dev');\n\nexport function subfieldContainsPartData(tag, subfieldCode) {\n  // NB! Used by reducers' mergeSubield.js\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  // This a very hacky function, but cand really help it, as the the data is very iffy as well...\n  // Remove punctuation and brackets:\n  const value = originalValue.replace(/[-.,:; ]+$/ui, '').replace(/^\\[([0-9]+)\\]$/ui, '$1');\n\n  const [year, rest] = extractYear(value);\n\n  const splitPoint = rest.lastIndexOf(' '); // MRA-627: \"5, 2017\" should be split here. Think of this later on...\n  if (splitPoint === -1) {\n    return [undefined, year, rest];\n  }\n  const lhs = rest.substr(0, splitPoint);\n  const rhs = rest.substr(splitPoint + 1);\n  return [lhs, year, rhs];\n\n  function extractYear(value) {\n    // NB! Note that this is far for perfect. It cover just some very common cases...\n\n    // \"2023, 3\" => [\"2023\", \"3\"]\n    if (value.match(/^(?:1[89][0-9][0-9]|20[012][0-9]), (?:nro |n:o)?[1-9][0-9]{0,2}$/ui)) {\n      return [value.substr(0, 4), value.substr(6)];\n    }\n    // \"2023/12\" => [\"2023\", \"12\"]\n    if (value.match(/^(?:1[89][0-9][0-9]|20[012][0-9])[/:][1-9][0-9]{0,2}$/u)) {\n      return [value.substr(0, 4), value.substr(5)];\n    }\n    // \"Vol. 3/2023\" => [\"2023\", \"Vol. 3\"]\n    if (value.match(/^[^0-9]*[1-9][0-9]{0,2}\\/(?:1[89][0-9][0-9]|20[012][0-9])$/u)) {\n      const len = value.length;\n      return [value.substr(len - 4), value.substr(0, len - 5)];\n    }\n\n\n    return [undefined, value];\n  }\n}\n\nfunction normalizePartType(originalValue) {\n  if (originalValue === undefined) {\n    return undefined;\n  }\n  const value = originalValue.toLowerCase();\n\n  // Return Finnish singular nominative. Choise of language is arbitrary. This is best-ish for debug purposes...\n  if (['n:o', 'no', 'nr', 'nro', 'number', 'numero', 'nummer'].includes(value)) {\n    return 'numero';\n  }\n  if (['band', 'bd', 'h\u00E4fte', 'nide', 'osa', 'part', 'teil', 'vol', 'vol.', 'volume'].includes(value)) {\n    return 'osa';\n  }\n\n  if (['p.', 'page', 'pages', 'pp.', 's.', 'sidor', 'sivu', 'sivut'].includes(value)) {\n    return 'sivu';\n  }\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 ${arabicValue}`, debugDev);\n    return arabicValue;\n  }\n  return value.toLowerCase();\n}\n\nfunction splitAndNormalizePartData(value) {\n  // This is just a stub. Does not handle eg. \"Levy 2, raita 15\"\n  const [partType, partYear, partNumber] = splitPartData(value);\n  //nvdebug(`  LHS: '${lhs}'`, debugDev);\n  //nvdebug(`  RHS: '${rhs}'`, debugDev);\n  return [normalizePartType(partType), partYear, normalizePartNumber(partNumber)];\n}\n\nexport function partsAgree(value1, value2, tag, subfieldCode) {\n  // Note, that parts can not be normalized away, as \"2\" can agree with \"Part 2\" and \"Raita 2\" and \"Volume 2\"...\n  // NB! Used by reducers' mergeSubield.js\n  if (!subfieldContainsPartData(tag, subfieldCode)) {\n    return false;\n  }\n  const [partType1, partYear1, partNumber1] = splitAndNormalizePartData(value1);\n  const [partType2, partYear2, partNumber2] = splitAndNormalizePartData(value2);\n  //nvdebug(`P1: ${partType1} | ${partYear1} | ${partNumber1}`, debugDev);\n  //nvdebug(`P2: ${partType2} | ${partYear2} | ${partNumber2}`, debugDev);\n  if (partNumber1 !== partNumber2) {\n    return false;\n  }\n  if (partType1 !== undefined && partType2 !== undefined && partType1 !== partType2) {\n    return false;\n  }\n  if (partYear1 !== undefined && partYear2 !== undefined && partYear1 !== partYear2) {\n    return false;\n  }\n\n\n  return true;\n}\n\nexport function normalizePartData(value, subfieldCode, tag) {\n  // This is for normalizing values for equality comparison only!\n  if (!subfieldContainsPartData(tag, subfieldCode)) {\n    return value;\n  }\n\n  const [partType, partYear, partNumber] = splitAndNormalizePartData(value);\n  if (partType === undefined) {\n    if (partYear === undefined) {\n      return partNumber;\n    }\n    return `${partNumber}/${partYear}`;\n  }\n  if (partYear === undefined) {\n    return `${partType} ${partNumber}`;\n  }\n  return `${partType} ${partNumber}/${partYear}`;\n}\n"],
   "mappings": "AAAA,SAAQ,eAAc;AACtB,OAAO,uBAAuB;AAI9B,MAAM,QAAQ,kBAAkB,kFAAkF;AAElH,MAAM,WAAW,MAAM,OAAO,KAAK;AAE5B,gBAAS,yBAAyB,KAAK,cAAc;AAE1D,MAAI,iBAAiB,OAAO,CAAC,OAAO,OAAO,OAAO,OAAO,KAAK,EAAE,SAAS,GAAG,GAAG;AAC7E,WAAO;AAAA,EACT;AACA,MAAI,QAAQ,SAAS,iBAAiB,KAAK;AACzC,WAAO;AAAA,EACT;AACA,SAAO;AACT;AAEA,SAAS,cAAc,eAAe;AAGpC,QAAM,QAAQ,cAAc,QAAQ,gBAAgB,EAAE,EAAE,QAAQ,oBAAoB,IAAI;AAExF,QAAM,CAAC,MAAM,IAAI,IAAI,YAAY,KAAK;AAEtC,QAAM,aAAa,KAAK,YAAY,GAAG;AACvC,MAAI,eAAe,IAAI;AACrB,WAAO,CAAC,QAAW,MAAM,IAAI;AAAA,EAC/B;AACA,QAAM,MAAM,KAAK,OAAO,GAAG,UAAU;AACrC,QAAM,MAAM,KAAK,OAAO,aAAa,CAAC;AACtC,SAAO,CAAC,KAAK,MAAM,GAAG;AAEtB,WAAS,YAAYA,QAAO;AAI1B,QAAIA,OAAM,MAAM,oEAAoE,GAAG;AACrF,aAAO,CAACA,OAAM,OAAO,GAAG,CAAC,GAAGA,OAAM,OAAO,CAAC,CAAC;AAAA,IAC7C;AAEA,QAAIA,OAAM,MAAM,wDAAwD,GAAG;AACzE,aAAO,CAACA,OAAM,OAAO,GAAG,CAAC,GAAGA,OAAM,OAAO,CAAC,CAAC;AAAA,IAC7C;AAEA,QAAIA,OAAM,MAAM,6DAA6D,GAAG;AAC9E,YAAM,MAAMA,OAAM;AAClB,aAAO,CAACA,OAAM,OAAO,MAAM,CAAC,GAAGA,OAAM,OAAO,GAAG,MAAM,CAAC,CAAC;AAAA,IACzD;AAGA,WAAO,CAAC,QAAWA,MAAK;AAAA,EAC1B;AACF;AAEA,SAAS,kBAAkB,eAAe;AACxC,MAAI,kBAAkB,QAAW;AAC/B,WAAO;AAAA,EACT;AACA,QAAM,QAAQ,cAAc,YAAY;AAGxC,MAAI,CAAC,OAAO,MAAM,MAAM,OAAO,UAAU,UAAU,QAAQ,EAAE,SAAS,KAAK,GAAG;AAC5E,WAAO;AAAA,EACT;AACA,MAAI,CAAC,QAAQ,MAAM,YAAS,QAAQ,OAAO,QAAQ,QAAQ,OAAO,QAAQ,QAAQ,EAAE,SAAS,KAAK,GAAG;AACnG,WAAO;AAAA,EACT;AAEA,MAAI,CAAC,MAAM,QAAQ,SAAS,OAAO,MAAM,SAAS,QAAQ,OAAO,EAAE,SAAS,KAAK,GAAG;AAClF,WAAO;AAAA,EACT;AAEA,SAAO;AACT;AAEA,MAAM,eAAe,EAAC,KAAK,KAAK,MAAM,KAAK,OAAO,KAAK,MAAM,KAAK,KAAK,KAAK,MAAM,KAAK,KAAK,KAAI;AAEhG,SAAS,oBAAoB,OAAO;AAGlC,MAAI,SAAS,cAAc;AACzB,UAAM,cAAc,aAAa,KAAK;AACtC,YAAQ,QAAQ,KAAK,OAAO,WAAW,IAAI,QAAQ;AACnD,WAAO;AAAA,EACT;AACA,SAAO,MAAM,YAAY;AAC3B;AAEA,SAAS,0BAA0B,OAAO;AAExC,QAAM,CAAC,UAAU,UAAU,UAAU,IAAI,cAAc,KAAK;AAG5D,SAAO,CAAC,kBAAkB,QAAQ,GAAG,UAAU,oBAAoB,UAAU,CAAC;AAChF;AAEO,gBAAS,WAAW,QAAQ,QAAQ,KAAK,cAAc;AAG5D,MAAI,CAAC,yBAAyB,KAAK,YAAY,GAAG;AAChD,WAAO;AAAA,EACT;AACA,QAAM,CAAC,WAAW,WAAW,WAAW,IAAI,0BAA0B,MAAM;AAC5E,QAAM,CAAC,WAAW,WAAW,WAAW,IAAI,0BAA0B,MAAM;AAG5E,MAAI,gBAAgB,aAAa;AAC/B,WAAO;AAAA,EACT;AACA,MAAI,cAAc,UAAa,cAAc,UAAa,cAAc,WAAW;AACjF,WAAO;AAAA,EACT;AACA,MAAI,cAAc,UAAa,cAAc,UAAa,cAAc,WAAW;AACjF,WAAO;AAAA,EACT;AAGA,SAAO;AACT;AAEO,gBAAS,kBAAkB,OAAO,cAAc,KAAK;AAE1D,MAAI,CAAC,yBAAyB,KAAK,YAAY,GAAG;AAChD,WAAO;AAAA,EACT;AAEA,QAAM,CAAC,UAAU,UAAU,UAAU,IAAI,0BAA0B,KAAK;AACxE,MAAI,aAAa,QAAW;AAC1B,QAAI,aAAa,QAAW;AAC1B,aAAO;AAAA,IACT;AACA,WAAO,GAAG,UAAU,IAAI,QAAQ;AAAA,EAClC;AACA,MAAI,aAAa,QAAW;AAC1B,WAAO,GAAG,QAAQ,IAAI,UAAU;AAAA,EAClC;AACA,SAAO,GAAG,QAAQ,IAAI,UAAU,IAAI,QAAQ;AAC9C;",
   "names": ["value"]
 }

package/dist/prepublicationUtils.js

@@ -136,7 +136,7 @@ export function removeWorsePrepubField594s(record) {
   nonBest.forEach((field) => record.removeField(field));
 }
 export function isEnnakkotietoSubfield(subfield) {
-  if (subfield.code !== "9" && subfield.code !== "g") {
+  if (!["g", "9", "7"].includes(subfield.code)) {
     return false;
   }
   if (subfield.value.length <= 13) {