@natlibfi/melinda-record-matching 6.0.0-alpha.1 → 6.0.1-alpha.1

package/dist/candidate-search/candidate-search-utils.js

@@ -1,15 +1,33 @@
 import createDebugLogger from "debug";
+import { uniqArray } from "@natlibfi/marc-record-validators-melinda/dist/utils.js";
 export function toQueries(identifiers, queryString) {
   const debug = createDebugLogger("@natlibfi/melinda-record-matching:toQueries");
   const debugData = debug.extend("data");
-  const quotedIdentifiers = identifiers.map((identifier) => identifier.match(/\//u) || identifier.match(/\^/u) ? `"${identifier}"` : `${identifier}`);
+  const debugDev = debug.extend("dev");
+  const useMaxLength = queryString === "dc.identifier" ? true : false;
+  const maxLength = 40;
+  debugData(`Adding ${queryString} to ${JSON.stringify(identifiers)}`);
+  debugDev(`We are ${useMaxLength ? `using maxLength ${maxLength}, queryString ${queryString}` : `not using maxLength, queryString ${queryString}`}`);
+  const croppedIdentifiers = uniqArray(useMaxLength ? identifiers.map((identifier) => identifier.substring(0, maxLength)) : identifiers);
+  const quotedIdentifiers = croppedIdentifiers.map((identifier) => identifier.match(/\//u) || identifier.match(/\^/u) ? `"${identifier}"` : `${identifier}`);
   const caretPairs = toPairs(quotedIdentifiers.filter((identifier) => identifier.match(/\^/u)));
   const nonCaretPairs = toPairs(quotedIdentifiers.filter((identifier) => !identifier.match(/\^/u)));
   const pairs = nonCaretPairs.concat(caretPairs);
   debugData(`Pairs (${pairs.length}): ${JSON.stringify(pairs)}`);
-  const queries = pairs.map(([a, b]) => b ? `${queryString}=${a} or ${queryString}=${b}` : `${queryString}=${a}`);
+  const queries = pairs.map(
+    ([a, b]) => {
+      const lengths = a.length + (b ? b.length : 0);
+      debugDev(`Lengths: A: ${a} (${a.length}) + B: ${b} (${b ? b.length : 0}) = ${lengths}`);
+      if (useMaxLength && lengths > maxLength && a && b) {
+        return [`${queryString}=${a}`, `${queryString}=${b}`];
+      }
+      return b ? `${queryString}=${a} or ${queryString}=${b}` : `${queryString}=${a}`;
+    }
+  );
   debugData(`Queries (${queries.length}): ${JSON.stringify(queries)}`);
-  return queries;
+  const flatQueries = queries.flat();
+  debugData(`FlatQueries (${flatQueries.length}): ${JSON.stringify(flatQueries)}`);
+  return flatQueries;
 }
 function toPairs(array) {
   if (array.length === 0) {

package/dist/candidate-search/candidate-search-utils.js.map

@@ -1,7 +1,7 @@
 {
   "version": 3,
   "sources": ["../../src/candidate-search/candidate-search-utils.js"],
-  "sourcesContent": ["import createDebugLogger from 'debug';\n\nexport function toQueries(identifiers, queryString) {\n\n  const debug = createDebugLogger('@natlibfi/melinda-record-matching:toQueries');\n  const debugData = debug.extend('data');\n\n  // We quote the identifier, if it contains a slash! (Slash in non-quoted SRU-search breaks search.)\n  // We also quote the identifier, if it starts with caret (f028 searches fail without caret and quotes...)\n  const quotedIdentifiers = identifiers.map(identifier => identifier.match(/\\//u) || identifier.match(/\\^/u) ? `\"${identifier}\"` : `${identifier}`);\n\n  // We can't pair queries with starting caret and without (ie. left anchored queries with non-left anchored queries)\n  const caretPairs = toPairs(quotedIdentifiers.filter(identifier => identifier.match(/\\^/u)));\n  const nonCaretPairs = toPairs(quotedIdentifiers.filter(identifier => !identifier.match(/\\^/u)));\n\n  const pairs = nonCaretPairs.concat(caretPairs);\n  debugData(`Pairs (${pairs.length}): ${JSON.stringify(pairs)}`);\n\n  // Aleph supports only two queries with or -operator (This is not actually true)\n  const queries = pairs.map(([a, b]) => b ? `${queryString}=${a} or ${queryString}=${b}` : `${queryString}=${a}`);\n  debugData(`Queries (${queries.length}): ${JSON.stringify(queries)}`);\n\n  return queries;\n}\n\nfunction toPairs(array) {\n  if (array.length === 0) {\n    return [];\n  }\n  return [array.slice(0, 2)].concat(toPairs(array.slice(2), 2));\n}\n"],
-  "mappings": "AAAA,OAAO,uBAAuB;AAEvB,gBAAS,UAAU,aAAa,aAAa;AAElD,QAAM,QAAQ,kBAAkB,6CAA6C;AAC7E,QAAM,YAAY,MAAM,OAAO,MAAM;AAIrC,QAAM,oBAAoB,YAAY,IAAI,gBAAc,WAAW,MAAM,KAAK,KAAK,WAAW,MAAM,KAAK,IAAI,IAAI,UAAU,MAAM,GAAG,UAAU,EAAE;AAGhJ,QAAM,aAAa,QAAQ,kBAAkB,OAAO,gBAAc,WAAW,MAAM,KAAK,CAAC,CAAC;AAC1F,QAAM,gBAAgB,QAAQ,kBAAkB,OAAO,gBAAc,CAAC,WAAW,MAAM,KAAK,CAAC,CAAC;AAE9F,QAAM,QAAQ,cAAc,OAAO,UAAU;AAC7C,YAAU,UAAU,MAAM,MAAM,MAAM,KAAK,UAAU,KAAK,CAAC,EAAE;AAG7D,QAAM,UAAU,MAAM,IAAI,CAAC,CAAC,GAAG,CAAC,MAAM,IAAI,GAAG,WAAW,IAAI,CAAC,OAAO,WAAW,IAAI,CAAC,KAAK,GAAG,WAAW,IAAI,CAAC,EAAE;AAC9G,YAAU,YAAY,QAAQ,MAAM,MAAM,KAAK,UAAU,OAAO,CAAC,EAAE;AAEnE,SAAO;AACT;AAEA,SAAS,QAAQ,OAAO;AACtB,MAAI,MAAM,WAAW,GAAG;AACtB,WAAO,CAAC;AAAA,EACV;AACA,SAAO,CAAC,MAAM,MAAM,GAAG,CAAC,CAAC,EAAE,OAAO,QAAQ,MAAM,MAAM,CAAC,GAAG,CAAC,CAAC;AAC9D;",
+  "sourcesContent": ["import createDebugLogger from 'debug';\nimport {uniqArray} from '@natlibfi/marc-record-validators-melinda/dist/utils.js';\n\nexport function toQueries(identifiers, queryString) {\n\n  const debug = createDebugLogger('@natlibfi/melinda-record-matching:toQueries');\n  const debugData = debug.extend('data');\n  const debugDev = debug.extend('dev');\n\n  // For dc.identifier search use max length of 40 characters to avoid dc.identifier -> 1007 mapped query from crashing\n  // DEVELOP: check if the length crash is related to mapping the query to multiple Aleph indexes\n  const useMaxLength = queryString === 'dc.identifier' ? true : false;\n  const maxLength = 40;\n  debugData(`Adding ${queryString} to ${JSON.stringify(identifiers)}`)\n  debugDev(`We are ${useMaxLength ? `using maxLength ${maxLength}, queryString ${queryString}` : `not using maxLength, queryString ${queryString}`}`);\n  const croppedIdentifiers = uniqArray(useMaxLength ? identifiers.map((identifier) => identifier.substring(0, maxLength)) : identifiers);\n\n  // We quote the identifier, if it contains a slash! (Slash in non-quoted SRU-search breaks search.)\n  // We also quote the identifier, if it starts with caret (f028 searches fail without caret and quotes...)\n  const quotedIdentifiers = croppedIdentifiers.map(identifier => identifier.match(/\\//u) || identifier.match(/\\^/u) ? `\"${identifier}\"` : `${identifier}`);\n\n  // We can't pair queries with starting caret and without (ie. left anchored queries with non-left anchored queries)\n  const caretPairs = toPairs(quotedIdentifiers.filter(identifier => identifier.match(/\\^/u)));\n  const nonCaretPairs = toPairs(quotedIdentifiers.filter(identifier => !identifier.match(/\\^/u)));\n\n  const pairs = nonCaretPairs.concat(caretPairs);\n  debugData(`Pairs (${pairs.length}): ${JSON.stringify(pairs)}`);\n\n  // Aleph supports only two queries with or -operator (This is not actually true)\n  const queries = pairs.map(([a, b]) => {\n    const lengths = a.length + (b ? b.length : 0);\n    debugDev(`Lengths: A: ${a} (${a.length}) + B: ${b} (${b ? b.length : 0}) = ${lengths}`);\n\n    // Do not create a paired query if query length would be too long\n    if (useMaxLength && lengths > maxLength && a && b) {\n      return [`${queryString}=${a}`, `${queryString}=${b}`];\n    }\n    return b ? `${queryString}=${a} or ${queryString}=${b}` : `${queryString}=${a}`}\n  );\n\n  debugData(`Queries (${queries.length}): ${JSON.stringify(queries)}`);\n  const flatQueries = queries.flat();\n  debugData(`FlatQueries (${flatQueries.length}): ${JSON.stringify(flatQueries)}`);\n\n  return flatQueries;\n}\n\nfunction toPairs(array) {\n  if (array.length === 0) {\n    return [];\n  }\n  return [array.slice(0, 2)].concat(toPairs(array.slice(2), 2));\n}\n\n"],
+  "mappings": "AAAA,OAAO,uBAAuB;AAC9B,SAAQ,iBAAgB;AAEjB,gBAAS,UAAU,aAAa,aAAa;AAElD,QAAM,QAAQ,kBAAkB,6CAA6C;AAC7E,QAAM,YAAY,MAAM,OAAO,MAAM;AACrC,QAAM,WAAW,MAAM,OAAO,KAAK;AAInC,QAAM,eAAe,gBAAgB,kBAAkB,OAAO;AAC9D,QAAM,YAAY;AAClB,YAAU,UAAU,WAAW,OAAO,KAAK,UAAU,WAAW,CAAC,EAAE;AACnE,WAAS,UAAU,eAAe,mBAAmB,SAAS,iBAAiB,WAAW,KAAK,oCAAoC,WAAW,EAAE,EAAE;AAClJ,QAAM,qBAAqB,UAAU,eAAe,YAAY,IAAI,CAAC,eAAe,WAAW,UAAU,GAAG,SAAS,CAAC,IAAI,WAAW;AAIrI,QAAM,oBAAoB,mBAAmB,IAAI,gBAAc,WAAW,MAAM,KAAK,KAAK,WAAW,MAAM,KAAK,IAAI,IAAI,UAAU,MAAM,GAAG,UAAU,EAAE;AAGvJ,QAAM,aAAa,QAAQ,kBAAkB,OAAO,gBAAc,WAAW,MAAM,KAAK,CAAC,CAAC;AAC1F,QAAM,gBAAgB,QAAQ,kBAAkB,OAAO,gBAAc,CAAC,WAAW,MAAM,KAAK,CAAC,CAAC;AAE9F,QAAM,QAAQ,cAAc,OAAO,UAAU;AAC7C,YAAU,UAAU,MAAM,MAAM,MAAM,KAAK,UAAU,KAAK,CAAC,EAAE;AAG7D,QAAM,UAAU,MAAM;AAAA,IAAI,CAAC,CAAC,GAAG,CAAC,MAAM;AACpC,YAAM,UAAU,EAAE,UAAU,IAAI,EAAE,SAAS;AAC3C,eAAS,eAAe,CAAC,KAAK,EAAE,MAAM,UAAU,CAAC,KAAK,IAAI,EAAE,SAAS,CAAC,OAAO,OAAO,EAAE;AAGtF,UAAI,gBAAgB,UAAU,aAAa,KAAK,GAAG;AACjD,eAAO,CAAC,GAAG,WAAW,IAAI,CAAC,IAAI,GAAG,WAAW,IAAI,CAAC,EAAE;AAAA,MACtD;AACA,aAAO,IAAI,GAAG,WAAW,IAAI,CAAC,OAAO,WAAW,IAAI,CAAC,KAAK,GAAG,WAAW,IAAI,CAAC;AAAA,IAAE;AAAA,EACjF;AAEA,YAAU,YAAY,QAAQ,MAAM,MAAM,KAAK,UAAU,OAAO,CAAC,EAAE;AACnE,QAAM,cAAc,QAAQ,KAAK;AACjC,YAAU,gBAAgB,YAAY,MAAM,MAAM,KAAK,UAAU,WAAW,CAAC,EAAE;AAE/E,SAAO;AACT;AAEA,SAAS,QAAQ,OAAO;AACtB,MAAI,MAAM,WAAW,GAAG;AACtB,WAAO,CAAC;AAAA,EACV;AACA,SAAO,CAAC,MAAM,MAAM,GAAG,CAAC,CAAC,EAAE,OAAO,QAAQ,MAAM,MAAM,CAAC,GAAG,CAAC,CAAC;AAC9D;",
   "names": []
 }

package/dist/candidate-search/query-list/bib.js

@@ -4,7 +4,9 @@ import { getMelindaIdsF035, validateSidFieldSubfieldCounts, getSubfieldValues, t
 import { extractHostMelindaIdsFromField, extractPublicationYearFrom773, getHostMelindaFields } from "./component.js";
 import { fieldToString } from "@natlibfi/marc-record-validators-melinda";
 import { getTitle } from "../../match-detection/features/bib/title.js";
+import { uniqArray } from "@natlibfi/marc-record-validators-melinda/dist/utils.js";
 const debug = createDebugLogger("@natlibfi/melinda-record-matching:candidate-search:query");
+const debugDev = debug.extend("dev");
 const IS_OK_ALONE_THRESHOLD = 5;
 export function bibSourceIds(record) {
   const debug2 = createDebugLogger("@natlibfi/melinda-record-matching:candidate-search:query:source-ids");
@@ -101,6 +103,7 @@ export function bibTitleAuthorYearAlternates(record) {
 }
 function getTitleForQuery(record) {
   const title = getTitle(record, ["a"]);
+  debug(`getTitleForQuery: title ${title}`);
   if (title) {
     const nWords = title.split(" ");
     if (nWords < 3 || title.length < 12) {
@@ -115,15 +118,20 @@ function getTitleForQuery(record) {
 }
 function dcTitle(record, onlyTitleLength, alternates = false) {
   const title = getTitleForQuery(record);
+  debug(`dcTitle: title: ${title}`);
   if (!testStringOrNumber(title)) {
     return [];
   }
   const formatted = getFormattedTitle(title);
+  debug(`dcTitle: formatted: ${formatted}`);
   const useWordSearch = checkUseWordSearch(formatted);
   if (formatted.length >= onlyTitleLength && !alternates) {
     return [`dc.title="${useWordSearch ? "" : "^"}${formatted}*"`, formatted, true];
   }
   const queryIsOkAlone = !useWordSearch && formatted.length >= IS_OK_ALONE_THRESHOLD;
+  if (formatted.length < 1) {
+    return ["", formatted, queryIsOkAlone];
+  }
   return [`dc.title="${useWordSearch || !queryIsOkAlone ? "" : "^"}${formatted}${queryIsOkAlone ? "*" : ""}"`, formatted, queryIsOkAlone];
   function getFormattedTitle(title2) {
     const formatted2 = String(title2).replace(/[\-+ !"(){}\[\]<>;:.?/@*%=^_`~]/gu, " ").replace(/[^\w\s\p{Alphabetic}]/gu, "").replace(/  +/gu, " ").trim().replace(/^(.{30}\S*) .*$/, "$1").trim();
@@ -133,6 +141,7 @@ function dcTitle(record, onlyTitleLength, alternates = false) {
 export function bibTitleAuthorPublisher({ record, onlyTitleLength, addYear = false, alternates = false, alternateQueries = [] }) {
   debug(`bibTitleAuthorPublisher, onlyTitleLength: ${onlyTitleLength}, addYear: ${addYear}, alternates: ${alternates}`);
   const [query, formatted, queryIsOkAlone] = dcTitle(record, onlyTitleLength, alternates);
+  debug(`query ${JSON.stringify(query)}, formatted: ${JSON.stringify(formatted)}, isOKAlone: ${JSON.stringify(queryIsOkAlone)}`);
   if (query === void 0) {
     return [];
   }
@@ -146,22 +155,24 @@ export function bibTitleAuthorPublisher({ record, onlyTitleLength, addYear = fal
     debug("addAuthorsToSearch");
     const [authorQuery] = bibAuthors(record);
     if (authorQuery !== void 0) {
+      const combinedQuery = query2.length > 0 ? `${authorQuery} AND ${query2}` : `${authorQuery}`;
       if (addYear2) {
-        const newAlternateQueries2 = alternates2 ? [...alternateQueries2, `${authorQuery} AND ${query2}`] : alternateQueries2;
-        return addYearToSearch({ query: `${authorQuery} AND ${query2}`, queryIsOkAlone: true, alternates: alternates2, alternateQueries: newAlternateQueries2 });
+        const newAlternateQueries2 = alternates2 ? [...alternateQueries2, combinedQuery] : alternateQueries2;
+        return addYearToSearch({ query: combinedQuery, queryIsOkAlone: true, alternates: alternates2, alternateQueries: newAlternateQueries2 });
       }
-      return alternates2 ? alternateQueries2 : [`${authorQuery} AND ${query2}`];
+      return alternates2 ? alternateQueries2 : [combinedQuery];
     }
     return addPublisherToSearch({ query: query2, queryIsOkAlone: queryIsOkAlone2, addYear: addYear2, alternates: alternates2, alternateQueries: alternateQueries2 });
   }
   function addPublisherToSearch({ query: query2, queryIsOkAlone: queryIsOkAlone2 = false, addYear: addYear2 = false, alternates: alternates2 = false, alternateQueries: alternateQueries2 = [] }) {
     const [publisherQuery] = bibPublishers(record);
     if (publisherQuery !== void 0) {
+      const combinedQuery = query2.length > 0 ? `${publisherQuery} AND ${query2}` : `${publisherQuery}`;
       if (addYear2) {
-        const newAlternateQueries2 = alternates2 ? [...alternateQueries2, `${publisherQuery} AND ${query2}`] : alternateQueries2;
-        return addYearToSearch({ query: `${publisherQuery} AND ${query2}`, queryIsOkAlone: true, alternates: alternates2, alternateQueries: newAlternateQueries2 });
+        const newAlternateQueries2 = alternates2 ? [...alternateQueries2, combinedQuery] : alternateQueries2;
+        return addYearToSearch({ query: combinedQuery, queryIsOkAlone: true, alternates: alternates2, alternateQueries: newAlternateQueries2 });
       }
-      return alternates2 ? alternateQueries2 : [`${publisherQuery} AND ${query2}`];
+      return alternates2 ? alternateQueries2 : [combinedQuery];
     }
     if (queryIsOkAlone2 && !addYear2) {
       return alternates2 ? alternateQueries2 : [`${query2}`];
@@ -171,7 +182,7 @@ export function bibTitleAuthorPublisher({ record, onlyTitleLength, addYear = fal
   function addYearToSearch({ query: query2, queryIsOkAlone: queryIsOkAlone2 = false, alternates: alternates2 = false, alternateQueries: alternateQueries2 = [] }) {
     const [yearQuery] = bibYear(record);
     if (yearQuery !== void 0) {
-      const yearAndOtherQueries = `${yearQuery} AND ${query2}`;
+      const yearAndOtherQueries = query2.length > 0 ? `${yearQuery} AND ${query2}` : `${yearQuery}`;
       return alternates2 ? [...alternateQueries2, yearAndOtherQueries] : [yearAndOtherQueries];
     }
     if (queryIsOkAlone2) {
@@ -244,10 +255,12 @@ export function bibYear(record) {
     }
     debugData(`f008: ${JSON.stringify(f008)}`);
     const { value } = f008;
-    if (value === "||||" || value === "    ") {
+    const year2 = value.slice(7, 11);
+    if (year2 === "||||" || year2 === "    ") {
+      debug2("Meaningless year in f008");
       return false;
     }
-    return value.slice(7, 11);
+    return year2;
   }
 }
 export function checkUseWordSearch(formatted) {
@@ -258,6 +271,7 @@ export function checkUseWordSearch(formatted) {
 export function bibStandardIdentifiers(record) {
   const debug2 = createDebugLogger("@natlibfi/melinda-record-matching:candidate-search:query:bibStandardIdentifiers");
   const debugData = debug2.extend("data");
+  const debugDev2 = debug2.extend("dev");
   debug2(`Creating queries for standard identifiers`);
   const fields = record.get(/^(?<def>015|020|022|024|028)$/u);
   const identifiers = [].concat(...fields.map(toIdentifiers));
@@ -283,14 +297,49 @@ export function bibStandardIdentifiers(record) {
       return subfields.filter((sub) => ["a", "z", "y"].includes(sub.code) && testStringOrNumber(sub.value) && issnIsbnReqExp.test(String(sub.value))).map(({ value }) => String(value));
     }
     if (tag === "028") {
+      let cleanQualifiers = function(val) {
+        if (val !== void 0) {
+          const cleanVal = val.replace(/ +\(.*\)/gu, "").replace(/ +/ug, " ").replace(/^ +/ug, "").replace(/ +$/ug, "");
+          return uniqArray([val, cleanVal]);
+        }
+        return [];
+      }, normalizeF028ForSearch = function(val) {
+        const f028NormCharsRegex = /[-"<>;:%=~`!\(\)\{\}\.\?\/\@\*\^]/g;
+        const normVal = val.replace(f028NormCharsRegex, " ").replace(/ +/ug, " ").replace(/^ +/ug, "").replace(/ +$/ug, "").toLowerCase();
+        return normVal !== " " ? normVal : "";
+      };
       const a = subfields.find((sf) => sf.code === "a");
-      if (a) {
-        const b = subfields.find((sf) => sf.code === "b");
-        if (b) {
-          return [`^${a.value}`, `^${b.value} ${a.value}`];
+      const b = subfields.find((sf) => sf.code === "b");
+      const cleanAs = cleanQualifiers(a?.value);
+      const cleanBs = cleanQualifiers(b?.value);
+      debugDev2(JSON.stringify(cleanAs));
+      debugDev2(JSON.stringify(cleanBs));
+      const splitAs = cleanAs.map((elem) => elem.split(",")).concat(cleanAs).flat();
+      const splitBs = cleanBs.map((elem) => elem.split(",")).concat(cleanBs).flat();
+      debugDev2(JSON.stringify(splitAs));
+      debugDev2(JSON.stringify(splitBs));
+      const normAs = splitAs.map((elem) => normalizeF028ForSearch(elem));
+      const normBs = splitBs.map((elem) => normalizeF028ForSearch(elem));
+      debugDev2(JSON.stringify(normAs));
+      debugDev2(JSON.stringify(normBs));
+      const identifiers2 = normAs.map((normA, index) => {
+        debugDev2(`Handling ${normA} from index: ${index}`);
+        if (normA.length > 0 && normBs[index]?.length > 0) {
+          debugDev2(`Found a (${normA}) + b (${normBs[index]})`);
+          return [`^${normBs[index]} ${normA}`, `^${normA} ${normBs[index]}`, `^${normA}`];
         }
-        return [`^${a.value}`];
-      }
+        if (normA.length > 0 && normBs[0]?.length > 0) {
+          debugDev2(`Found a (${normA}) + b (${normBs[0]})`);
+          return [`^${normBs[0]} ${normA}`, `^${normA} ${normBs[0]}`, `^${normA}`];
+        }
+        if (normA.length > 0) {
+          debugDev2(`Found a (${normA}) but no b ${normBs[index]}`);
+          return [`^${normA}`];
+        }
+        return [];
+      });
+      debugDev2(`Current identifiers from f028: ${JSON.stringify(identifiers2)}`);
+      return uniqArray(identifiers2.flat());
     }
     return subfields.filter((sub) => ["a", "z"].includes(sub.code) && testStringOrNumber(sub.value) && otherIdReqExp.test(String(sub.value))).map(({ value }) => String(value));
   }

package/dist/candidate-search/query-list/bib.js.map

@@ -1,7 +1,7 @@
 {
   "version": 3,
   "sources": ["../../../src/candidate-search/query-list/bib.js"],
-  "sourcesContent": ["import createDebugLogger from 'debug';\nimport {toQueries} from '../candidate-search-utils.js';\nimport {getMelindaIdsF035, validateSidFieldSubfieldCounts, getSubfieldValues, testStringOrNumber} from '../../matching-utils.js';\nimport {extractHostMelindaIdsFromField, extractPublicationYearFrom773, getHostMelindaFields} from './component.js';\nimport {fieldToString} from '@natlibfi/marc-record-validators-melinda';\nimport {getTitle} from '../../match-detection/features/bib/title.js';\n\nconst debug = createDebugLogger('@natlibfi/melinda-record-matching:candidate-search:query');\n\nconst IS_OK_ALONE_THRESHOLD = 5;\n\nexport function bibSourceIds(record) {\n\n  /* Melinda's SRU-index melinda.sourceid includes source IDs from SID fields in Melinda records\n     SID-fields in Melinda have sf $c with local id and sf $b with a code for the local db:\n\n     SID__ $c 123457 $b helka\n     SID__ $c (ANDL100020)1077305 $b sata\n     SID__ $c VER999999 $ FI-KV\n     SID__ $c /10024/508126 $ REPO_THESEUS\n\n    In melinda.sourceid -index case is kept, sourceprefixes in brackets and hyphens are normalized away:\n    Note: slashes are not normalized away, but a SRU-search-string including slashes needs to be quoted\n\n    1234567helka\n    1077305sata\n    VER999999FIKV\n    /10024/508126REPO_THESEUS\n\n    Note: All Melinda records that have a matching records in a local db do NOT have SID for that local records,\n          existence of a SID field depends on how the record has been added to Melinda and how it has been handled\n          afterwards. SIDs are also not reliably maintained. A record might or might not have a SID for a local db\n          after the matching record is removed from the local db.\n\n   */\n\n\n  const debug = createDebugLogger('@natlibfi/melinda-record-matching:candidate-search:query:source-ids');\n  const debugData = debug.extend('data');\n  //const debugInfo = debug.extend('info');\n\n  debug(`Creating queries for sourceid's`);\n\n  const fSids = record.get('SID');\n  debugData(`SID-fields (${fSids.length}): ${JSON.stringify(fSids)}`);\n\n  return fSids.length > 0 ? toSidQueries(fSids) : [];\n\n  function toSidQueries(fSids) {\n    debug(`Creating actual queries for sourceid's`);\n\n    const sidStrings = getSidStrings(fSids);\n\n    if (sidStrings.length < 1) {\n      debug(`No identifiers found, no queries created.`);\n      return [];\n    }\n\n    const sidQueries = toQueries(sidStrings, 'melinda.sourceid');\n\n    return sidQueries;\n\n    function getSidStrings(fSids) {\n      debug(`Getting Sid strings from SID fields`);\n\n      // Map SID fields to valid sidStrings, filter out empty strings\n      const sidStrings = fSids.map(toSidString).filter(nonEmptySid => nonEmptySid);\n      return sidStrings;\n\n      function toSidString(field) {\n        debug(`Getting string from a field`);\n\n        return validateSidFieldSubfieldCounts(field) ? createSidString(field) : '';\n\n        function createSidString(field) {\n          debug(`Creating string from a field`);\n          const [sfC] = getSubfieldValues(field, 'c');\n          const [sfB] = getSubfieldValues(field, 'b');\n\n          const cleanedSfC = removeSourcePrefix(normalizeSidSubfieldValue(sfC));\n          const cleanedSfB = normalizeSidSubfieldValue(sfB);\n\n          debugData(`${JSON.stringify(sfC)} + ${JSON.stringify(sfB)}`);\n          return cleanedSfC.concat(cleanedSfB);\n        }\n\n        function removeSourcePrefix(subfieldValue) {\n          const sourcePrefixRegex = (/^(?<sourcePrefix>\\([A-Za-z0-9-]+\\))(?<id>.+)$/u);\n          const normalizedValue = testStringOrNumber(subfieldValue) ? String(subfieldValue).replace(sourcePrefixRegex, '$<id>') : '';\n          debugData(`Normalized ${subfieldValue} to ${normalizedValue}`);\n          return normalizedValue;\n        }\n\n        function normalizeSidSubfieldValue(subfieldValue) {\n          debugData(`Normalizing ${subfieldValue}`);\n          const normalizeAwayRegex = (/[- ]/u);\n          return testStringOrNumber(subfieldValue) ? String(subfieldValue).replace(normalizeAwayRegex, '') : '';\n        }\n\n      }\n    }\n  }\n}\n\nexport function bibMelindaIds(record) {\n  // Melinda's SRU-index melinda.melindaid includes f001 controlnumbers and old Melinda-IDs from f035z's for all non-deleted Melinda-records\n\n  const debug = createDebugLogger('@natlibfi/melinda-record-matching:candidate-search:query:bibMelindaIds');\n  const debugData = debug.extend('data');\n  debug(`Creating queries for MelindaIds`);\n\n  // Note: Melinda-ID's for search queries are created just from records f035a's and f035z's\n  // Both (FI-MELINDA)- and FCC-prefixed forms are found\n  // f001 controlnumber is not currently included, even if record's f003 is FI-MELINDA\n  const melindaIds = getMelindaIdsF035(record);\n\n  debugData(`Unique identifiers (${melindaIds.length}): ${JSON.stringify(melindaIds)}`);\n\n  if (melindaIds.length < 1) {\n    debug(`No identifiers found, no queries created.`);\n    return [];\n  }\n\n  return toQueries(melindaIds, 'melinda.melindaid');\n}\n\n\n// bibHostComponents returns host id from the first subfield $w of first field f773, see test-fixtures 04 and 05\n// bibHostComponents should search all 773 $ws for possible host id, but what should it do in case of multiple host ids?\nexport function bibHostComponents(record) {\n  const debug = createDebugLogger('@natlibfi/melinda-record-matching:candidate-search:query:bibHostComponents');\n  const debugData = debug.extend('data');\n  debug(`Creating queries for hostIds`);\n  const f773s = getHostMelindaFields(record);\n  if (f773s.length !== 1) { // This needs some thinking...\n    return [];\n  }\n\n  const [id] = extractHostMelindaIdsFromField(f773s[0]);\n  debug(`Found id: ${JSON.stringify(id)}`);\n\n  // NB! record.isCR() does not work if LDR/07=a, so we get year from 773$g!\n  const year = extractPublicationYearFrom773(f773s[0]);\n  debug(`${fieldToString(f773s[0])} => ${year ? year : 'N/A'}`);\n  if (year) {\n    return [`dc.date=${year} AND melinda.partsofhost=${id}`];\n  }\n\n  return [`melinda.partsofhost=${id}`];\n\n}\n\n\n\n\n// SRU search dc.title with a search phrase starting with ^ maps currently in Melinda to (probably) to *headings* index TIT\n// - Aleph cannot currently handle headings searches starting with a boolean - in these cases use word search\n\nexport function bibTitle(record) {\n  // We get author/publisher only when formatted title is shorter than 5 chars\n  return bibTitleAuthorPublisher({record, onlyTitleLength: 5});\n}\n\nexport function bibTitleAuthor(record) {\n  debug('bibTitleAuthor');\n  // We use onlyTitleLength that is longer than our formatted length to\n  // get an author or an publisher always\n  return bibTitleAuthorPublisher({record, onlyTitleLength: 100});\n}\n\nexport function bibTitleAuthorYear(record) {\n  debug('bibTitleAuthorYearAlternates');\n  // We use onlyTitleLength that is longer than our formatted length to\n  // get an author or an publisher always\n\n  return bibTitleAuthorPublisher({record, onlyTitleLength: 100, addYear: true});\n}\n\nexport function bibTitleAuthorYearAlternates(record) {\n  debug('bibTitleAuthorYearAlternates');\n  // We use onlyTitleLength that is longer than our formatted length to\n  // get an author or an publisher always\n  const origQueryList = bibTitleAuthorPublisher({record, onlyTitleLength: 100, addYear: true, alternates: true, alternateQueries: []});\n  debug(`${JSON.stringify(origQueryList)}`);\n  return {queryList: Array.from(origQueryList).reverse(), queryListType: 'alternates'};\n}\n\nfunction getTitleForQuery(record) {\n  const title = getTitle(record, ['a']);\n  if (title) {\n    const nWords = title.split(' ');\n    // If lone $a is deemed too short, fetch $b as well:\n    // (NV: I've seen pairs with 245$a-only vs 245$a$b, so I'd like to use $a only if it is long enough)\n\n    if (nWords < 3 || title.length < 12) {\n      const [f245] = record.get('245');\n      // If punctuation is ' =' I think that f245$a is good enough despite shortness. Trying to balance between two bad situations...\n      // \"Suo (= short title) siell\u00E4, vetel\u00E4 (= missing $b name in another language) t\u00E4\u00E4ll\u00E4...\"\n      if (f245 && f245.subfields.find(sf => sf.code === 'a' && sf.value.match(/ =$/u)) && title.length >= IS_OK_ALONE_THRESHOLD) {\n        return title;\n      }\n      return getTitle(record, ['a', 'b']); // Try to get a longer title\n    }\n  }\n  return title;\n}\n\nfunction dcTitle(record, onlyTitleLength, alternates = false) {\n  const title = getTitleForQuery(record);\n  if (!testStringOrNumber(title)) {\n    return [];\n  }\n\n  const formatted = getFormattedTitle(title);\n\n  // use word search for titles starting with a boolean\n  const useWordSearch = checkUseWordSearch(formatted);\n  // Prevent too many matches / SRU crashing by having a minimum length\n  // Note that currently this fails matching if there are no matches from previous matchers\n  if (formatted.length >= onlyTitleLength && !alternates) {\n    return [`dc.title=\"${useWordSearch ? '' : '^'}${formatted}*\"`, formatted, true];\n  }\n\n  const queryIsOkAlone = !useWordSearch && formatted.length >= IS_OK_ALONE_THRESHOLD;\n\n  // use word search without ending * also in combination searches to avoid SRU-server crashes [MRA-189]\n  return [`dc.title=\"${useWordSearch || !queryIsOkAlone ? '' : '^'}${formatted}${queryIsOkAlone ? '*' : ''}\"`, formatted, queryIsOkAlone];\n\n  function getFormattedTitle(title) {\n    const formatted = String(title)\n      .replace(/[\\-+ !\"(){}\\[\\]<>;:.?/@*%=^_`~]/gu, ' ')\n      .replace(/[^\\w\\s\\p{Alphabetic}]/gu, '') // Apparently matches to/works with non-aplhabetic scripts such as Chinese as well\n      .replace(/  +/gu, ' ') // Clean up concurrent spaces from eg. subfield changes\n      .trim()\n      .replace(/^(.{30}\\S*) .*$/, \"$1\")\n      .trim();\n\n    return formatted;\n  }\n}\n\nexport function bibTitleAuthorPublisher({record, onlyTitleLength, addYear = false, alternates = false, alternateQueries = []}) {\n  debug(`bibTitleAuthorPublisher, onlyTitleLength: ${onlyTitleLength}, addYear: ${addYear}, alternates: ${alternates}`);\n  const [query, formatted, queryIsOkAlone] = dcTitle(record, onlyTitleLength, alternates);\n  if (query === undefined) {\n    return [];\n  }\n\n  debug(`query: ${query}`);\n\n  // Prevent too many matches / SRU crashing by having a minimum length\n  // Note that currently this fails matching if there are no matches from previous matchers\n  if (formatted.length >= onlyTitleLength && !alternates) {\n    return [query];\n  }\n\n  const newAlternateQueries = alternates ? [...alternateQueries, query] : alternateQueries;\n\n  return addAuthorsToSearch({query, queryIsOkAlone, addYear, alternates, alternateQueries: newAlternateQueries});\n\n  function addAuthorsToSearch({query, queryIsOkAlone = false, addYear = false, alternates = false, alternateQueries = []}) {\n    debug('addAuthorsToSearch');\n    const [authorQuery] = bibAuthors(record);\n    if (authorQuery !== undefined) {\n      if (addYear) {\n        const newAlternateQueries = alternates ? [...alternateQueries, `${authorQuery} AND ${query}`] : alternateQueries;\n        return addYearToSearch({query: `${authorQuery} AND ${query}`, queryIsOkAlone: true, alternates, alternateQueries: newAlternateQueries});\n      }\n      return alternates ? alternateQueries : [`${authorQuery} AND ${query}`];\n    }\n    return addPublisherToSearch({query, queryIsOkAlone, addYear, alternates, alternateQueries});\n    //return [];\n  }\n\n  function addPublisherToSearch({query, queryIsOkAlone = false, addYear = false, alternates = false, alternateQueries = []}) {\n    const [publisherQuery] = bibPublishers(record);\n    if (publisherQuery !== undefined) {\n      if (addYear) {\n        const newAlternateQueries = alternates ? [...alternateQueries, `${publisherQuery} AND ${query}`] : alternateQueries;\n        return addYearToSearch({query: `${publisherQuery} AND ${query}`, queryIsOkAlone: true, alternates, alternateQueries: newAlternateQueries});\n      }\n      return alternates ? alternateQueries : [`${publisherQuery} AND ${query}`];\n    }\n    if (queryIsOkAlone && !addYear) {\n      return alternates ? alternateQueries : [`${query}`];\n    }\n    return addYearToSearch({query, queryIsOkAlone, alternates, alternateQueries});\n  }\n\n  function addYearToSearch({query, queryIsOkAlone = false, alternates = false, alternateQueries = []}) {\n    const [yearQuery] = bibYear(record);\n    if (yearQuery !== undefined) {\n      const yearAndOtherQueries = `${yearQuery} AND ${query}`;\n      return alternates ? [...alternateQueries, yearAndOtherQueries] : [yearAndOtherQueries];\n    }\n    if (queryIsOkAlone) {\n      return alternates ? alternateQueries : [`${query}`];\n    }\n    return [];\n  }\n\n}\n\nexport function bibAuthors(record) {\n  const debug = createDebugLogger('@natlibfi/melinda-record-matching:candidate-search:query:bibAuthors');\n  const debugData = debug.extend('data');\n  debug(`Creating query for the first author`);\n  //debugData(record);\n\n  const author = getAuthor(record);\n\n  if (testStringOrNumber(author)) {\n    const formatted = String(author)\n      .replace(/[^\\w\\s\\p{Alphabetic}]/gu, '')\n      // Clean up concurrent spaces from fe. subfield changes\n      .replace(/ +/gu, ' ')\n      .trim()\n      .slice(0, 30)\n      .trim();\n\n    // use word search for authors starting with a boolean\n    const useWordSearch = checkUseWordSearch(formatted);\n    // Prevent too many matches by having a minimum length\n    debugData(`Author string: ${formatted}`);\n    if (formatted.length >= 5) {\n      return [`dc.author=\"${useWordSearch ? '' : '^'}${formatted}*\"`];\n    }\n    return [];\n  }\n\n  return [];\n\n  function getAuthor(record) {\n    //debugData(record);\n    const [field] = record.get(/^(?:100|110|111|700|710|711)$/u);\n    //debugData(field);\n\n    if (field) {\n      const authorString = field.subfields\n        .filter(({code}) => ['a'].includes(code)) // We might use different subfield code sets for X00, X10 and X11?\n        .map(({value}) => testStringOrNumber(value) ? String(value) : '')\n        .filter(value => value)\n        // In Melinda's index subfield separators are indexed as ' '\n        .join(' ');\n      return authorString;\n    }\n    return false;\n  }\n}\n\nexport function bibPublishers(record) {\n  const debug = createDebugLogger('@natlibfi/melinda-record-matching:candidate-search:query:bibPublishers');\n  const debugData = debug.extend('data');\n  debug(`Creating query for the first publisher`);\n  //debugData(record);\n\n  const publisher = getPublisher(record);\n  if (testStringOrNumber(publisher)) {\n    const formatted = String(publisher)\n      .replace(/[^\\w\\s\\p{Alphabetic}]/gu, '')\n      // Clean up concurrent spaces from fe. subfield changes\n      .replace(/ +/gu, ' ')\n      .trim()\n      .slice(0, 30)\n      .trim();\n\n    debugData(`Publisher string: '${formatted}'`);\n    if (formatted.length === 0) {\n      return [];\n    }\n    // use non-wildcard word search from dc.publisher\n    return [`dc.publisher=\"${formatted}\"`];\n  }\n\n  return [];\n\n  function getPublisher(record) {\n    //debugData(record);\n    const [field] = record.get(/^26[04]$/u).filter(f => f.subfields.some(sf => sf.code === 'b')); // Filter removes copyright-only fields\n    //debugData(field);\n\n    if (field) {\n      const publisherString = field.subfields\n        .filter(({code}) => ['b'].includes(code))\n        .map(({value}) => testStringOrNumber(value) ? String(value) : '')\n        .filter(value => value)\n        // In Melinda's index subfield separators are indexed as ' '\n        .join(' ')\n        .trim();\n      return publisherString;\n    }\n    return false;\n  }\n}\n\nexport function bibYear(record) {\n  const debug = createDebugLogger('@natlibfi/melinda-record-matching:candidate-search:query:bibYear');\n  const debugData = debug.extend('data');\n  debug(`Creating query for the publishing year`);\n\n  const year = getYear(record);\n  if (year) {\n    return [`dc.date=\"${year}\"`];\n  }\n  return [];\n\n  function getYear(record) {\n    const [f008] = record.get(/^008$/u);\n    if (!f008 || !f008.value || f008.value.length < 11) {\n      debug('f008 missing/crappy');\n      return false;\n    }\n\n    debugData(`f008: ${JSON.stringify(f008)}`);\n    const {value} = f008;\n    if (value === '||||' || value === '    ') { // Meaningless values\n      return false;\n    }\n    return value.slice(7, 11);\n  }\n}\n\nexport function checkUseWordSearch(formatted) {\n  const lowercased = formatted.toLowerCase();\n  // Note: add a space to startWords to catch just actual boolean words\n  const booleanStartWords = ['and ', 'or ', 'nor ', 'not '];\n  return booleanStartWords.some(word => lowercased.startsWith(word));\n}\n\nexport function bibStandardIdentifiers(record) {\n\n  const debug = createDebugLogger('@natlibfi/melinda-record-matching:candidate-search:query:bibStandardIdentifiers');\n  const debugData = debug.extend('data');\n  debug(`Creating queries for standard identifiers`);\n\n  // DEVELOP: should we query also f015 and f028?\n\n  // const fields = record.get(/^(?<def>020|022|024)$/u);\n  const fields = record.get(/^(?<def>015|020|022|024|028)$/u);\n  const identifiers = [].concat(...fields.map(toIdentifiers));\n  const uniqueIdentifiers = [...new Set(identifiers)];\n\n  debugData(`Standard identifier fields: ${JSON.stringify(fields)}`);\n  debugData(`Identifiers (${identifiers.length}): ${JSON.stringify(identifiers)}`);\n  debugData(`Unique identifiers (${uniqueIdentifiers.length}): ${JSON.stringify(uniqueIdentifiers)}`);\n\n  if (uniqueIdentifiers.length < 1) {\n    debug(`No identifiers found, no queries created.`);\n    return [];\n  }\n\n  return toQueries(uniqueIdentifiers, 'dc.identifier');\n\n  function toIdentifiers({tag, subfields}) {\n    const issnIsbnReqExp = (/^[A-Za-z0-9-]+$/u);\n    const otherIdReqExp = (/^[A-Za-z0-9-:]+$/u);\n\n    if (tag === '015') { // TODO: test (does this use the right index etc.)\n      return subfields\n        .filter(sf => sf.code === 'a' && sf.value.match(/^[A-Z0-9\\-]+$/ui))\n        .map(sf => sf.value)\n    }\n\n    if (tag === '020') {\n      return subfields\n        .filter(sub => ['a', 'z'].includes(sub.code) && testStringOrNumber(sub.value) && issnIsbnReqExp.test(String(sub.value)))\n        .map(({value}) => String(value));\n    }\n\n    if (tag === '022') {\n      return subfields\n        .filter(sub => ['a', 'z', 'y'].includes(sub.code) && testStringOrNumber(sub.value) && issnIsbnReqExp.test(String(sub.value)))\n        .map(({value}) => String(value));\n    }\n\n    if (tag === '028') { // TODO: test\n      const a = subfields.find(sf => sf.code === 'a');\n      if (a) {\n        const b = subfields.find(sf => sf.code === 'b');\n        // TODO: normalize?\n        if (b) {\n          return [`^${a.value}`, `^${b.value} ${a.value}`];\n        }\n        return [`^${a.value}`];\n      }\n    }\n\n    // Default seems to be 024:\n    return subfields\n      .filter(sub => ['a', 'z'].includes(sub.code) && testStringOrNumber(sub.value) && otherIdReqExp.test(String(sub.value)))\n      .map(({value}) => String(value));\n  }\n}\n"],
-  "mappings": "AAAA,OAAO,uBAAuB;AAC9B,SAAQ,iBAAgB;AACxB,SAAQ,mBAAmB,gCAAgC,mBAAmB,0BAAyB;AACvG,SAAQ,gCAAgC,+BAA+B,4BAA2B;AAClG,SAAQ,qBAAoB;AAC5B,SAAQ,gBAAe;AAEvB,MAAM,QAAQ,kBAAkB,0DAA0D;AAE1F,MAAM,wBAAwB;AAEvB,gBAAS,aAAa,QAAQ;AA0BnC,QAAMA,SAAQ,kBAAkB,qEAAqE;AACrG,QAAM,YAAYA,OAAM,OAAO,MAAM;AAGrC,EAAAA,OAAM,iCAAiC;AAEvC,QAAM,QAAQ,OAAO,IAAI,KAAK;AAC9B,YAAU,eAAe,MAAM,MAAM,MAAM,KAAK,UAAU,KAAK,CAAC,EAAE;AAElE,SAAO,MAAM,SAAS,IAAI,aAAa,KAAK,IAAI,CAAC;AAEjD,WAAS,aAAaC,QAAO;AAC3B,IAAAD,OAAM,wCAAwC;AAE9C,UAAM,aAAa,cAAcC,MAAK;AAEtC,QAAI,WAAW,SAAS,GAAG;AACzB,MAAAD,OAAM,2CAA2C;AACjD,aAAO,CAAC;AAAA,IACV;AAEA,UAAM,aAAa,UAAU,YAAY,kBAAkB;AAE3D,WAAO;AAEP,aAAS,cAAcC,QAAO;AAC5B,MAAAD,OAAM,qCAAqC;AAG3C,YAAME,cAAaD,OAAM,IAAI,WAAW,EAAE,OAAO,iBAAe,WAAW;AAC3E,aAAOC;AAEP,eAAS,YAAY,OAAO;AAC1B,QAAAF,OAAM,6BAA6B;AAEnC,eAAO,+BAA+B,KAAK,IAAI,gBAAgB,KAAK,IAAI;AAExE,iBAAS,gBAAgBG,QAAO;AAC9B,UAAAH,OAAM,8BAA8B;AACpC,gBAAM,CAAC,GAAG,IAAI,kBAAkBG,QAAO,GAAG;AAC1C,gBAAM,CAAC,GAAG,IAAI,kBAAkBA,QAAO,GAAG;AAE1C,gBAAM,aAAa,mBAAmB,0BAA0B,GAAG,CAAC;AACpE,gBAAM,aAAa,0BAA0B,GAAG;AAEhD,oBAAU,GAAG,KAAK,UAAU,GAAG,CAAC,MAAM,KAAK,UAAU,GAAG,CAAC,EAAE;AAC3D,iBAAO,WAAW,OAAO,UAAU;AAAA,QACrC;AAEA,iBAAS,mBAAmB,eAAe;AACzC,gBAAM,oBAAqB;AAC3B,gBAAM,kBAAkB,mBAAmB,aAAa,IAAI,OAAO,aAAa,EAAE,QAAQ,mBAAmB,OAAO,IAAI;AACxH,oBAAU,cAAc,aAAa,OAAO,eAAe,EAAE;AAC7D,iBAAO;AAAA,QACT;AAEA,iBAAS,0BAA0B,eAAe;AAChD,oBAAU,eAAe,aAAa,EAAE;AACxC,gBAAM,qBAAsB;AAC5B,iBAAO,mBAAmB,aAAa,IAAI,OAAO,aAAa,EAAE,QAAQ,oBAAoB,EAAE,IAAI;AAAA,QACrG;AAAA,MAEF;AAAA,IACF;AAAA,EACF;AACF;AAEO,gBAAS,cAAc,QAAQ;AAGpC,QAAMH,SAAQ,kBAAkB,wEAAwE;AACxG,QAAM,YAAYA,OAAM,OAAO,MAAM;AACrC,EAAAA,OAAM,iCAAiC;AAKvC,QAAM,aAAa,kBAAkB,MAAM;AAE3C,YAAU,uBAAuB,WAAW,MAAM,MAAM,KAAK,UAAU,UAAU,CAAC,EAAE;AAEpF,MAAI,WAAW,SAAS,GAAG;AACzB,IAAAA,OAAM,2CAA2C;AACjD,WAAO,CAAC;AAAA,EACV;AAEA,SAAO,UAAU,YAAY,mBAAmB;AAClD;AAKO,gBAAS,kBAAkB,QAAQ;AACxC,QAAMA,SAAQ,kBAAkB,4EAA4E;AAC5G,QAAM,YAAYA,OAAM,OAAO,MAAM;AACrC,EAAAA,OAAM,8BAA8B;AACpC,QAAM,QAAQ,qBAAqB,MAAM;AACzC,MAAI,MAAM,WAAW,GAAG;AACtB,WAAO,CAAC;AAAA,EACV;AAEA,QAAM,CAAC,EAAE,IAAI,+BAA+B,MAAM,CAAC,CAAC;AACpD,EAAAA,OAAM,aAAa,KAAK,UAAU,EAAE,CAAC,EAAE;AAGvC,QAAM,OAAO,8BAA8B,MAAM,CAAC,CAAC;AACnD,EAAAA,OAAM,GAAG,cAAc,MAAM,CAAC,CAAC,CAAC,OAAO,OAAO,OAAO,KAAK,EAAE;AAC5D,MAAI,MAAM;AACR,WAAO,CAAC,WAAW,IAAI,4BAA4B,EAAE,EAAE;AAAA,EACzD;AAEA,SAAO,CAAC,uBAAuB,EAAE,EAAE;AAErC;AAQO,gBAAS,SAAS,QAAQ;AAE/B,SAAO,wBAAwB,EAAC,QAAQ,iBAAiB,EAAC,CAAC;AAC7D;AAEO,gBAAS,eAAe,QAAQ;AACrC,QAAM,gBAAgB;AAGtB,SAAO,wBAAwB,EAAC,QAAQ,iBAAiB,IAAG,CAAC;AAC/D;AAEO,gBAAS,mBAAmB,QAAQ;AACzC,QAAM,8BAA8B;AAIpC,SAAO,wBAAwB,EAAC,QAAQ,iBAAiB,KAAK,SAAS,KAAI,CAAC;AAC9E;AAEO,gBAAS,6BAA6B,QAAQ;AACnD,QAAM,8BAA8B;AAGpC,QAAM,gBAAgB,wBAAwB,EAAC,QAAQ,iBAAiB,KAAK,SAAS,MAAM,YAAY,MAAM,kBAAkB,CAAC,EAAC,CAAC;AACnI,QAAM,GAAG,KAAK,UAAU,aAAa,CAAC,EAAE;AACxC,SAAO,EAAC,WAAW,MAAM,KAAK,aAAa,EAAE,QAAQ,GAAG,eAAe,aAAY;AACrF;AAEA,SAAS,iBAAiB,QAAQ;AAChC,QAAM,QAAQ,SAAS,QAAQ,CAAC,GAAG,CAAC;AACpC,MAAI,OAAO;AACT,UAAM,SAAS,MAAM,MAAM,GAAG;AAI9B,QAAI,SAAS,KAAK,MAAM,SAAS,IAAI;AACnC,YAAM,CAAC,IAAI,IAAI,OAAO,IAAI,KAAK;AAG/B,UAAI,QAAQ,KAAK,UAAU,KAAK,QAAM,GAAG,SAAS,OAAO,GAAG,MAAM,MAAM,MAAM,CAAC,KAAK,MAAM,UAAU,uBAAuB;AACzH,eAAO;AAAA,MACT;AACA,aAAO,SAAS,QAAQ,CAAC,KAAK,GAAG,CAAC;AAAA,IACpC;AAAA,EACF;AACA,SAAO;AACT;AAEA,SAAS,QAAQ,QAAQ,iBAAiB,aAAa,OAAO;AAC5D,QAAM,QAAQ,iBAAiB,MAAM;AACrC,MAAI,CAAC,mBAAmB,KAAK,GAAG;AAC9B,WAAO,CAAC;AAAA,EACV;AAEA,QAAM,YAAY,kBAAkB,KAAK;AAGzC,QAAM,gBAAgB,mBAAmB,SAAS;AAGlD,MAAI,UAAU,UAAU,mBAAmB,CAAC,YAAY;AACtD,WAAO,CAAC,aAAa,gBAAgB,KAAK,GAAG,GAAG,SAAS,MAAM,WAAW,IAAI;AAAA,EAChF;AAEA,QAAM,iBAAiB,CAAC,iBAAiB,UAAU,UAAU;AAG7D,SAAO,CAAC,aAAa,iBAAiB,CAAC,iBAAiB,KAAK,GAAG,GAAG,SAAS,GAAG,iBAAiB,MAAM,EAAE,KAAK,WAAW,cAAc;AAEtI,WAAS,kBAAkBI,QAAO;AAChC,UAAMC,aAAY,OAAOD,MAAK,EAC3B,QAAQ,qCAAqC,GAAG,EAChD,QAAQ,2BAA2B,EAAE,EACrC,QAAQ,SAAS,GAAG,EACpB,KAAK,EACL,QAAQ,mBAAmB,IAAI,EAC/B,KAAK;AAER,WAAOC;AAAA,EACT;AACF;AAEO,gBAAS,wBAAwB,EAAC,QAAQ,iBAAiB,UAAU,OAAO,aAAa,OAAO,mBAAmB,CAAC,EAAC,GAAG;AAC7H,QAAM,6CAA6C,eAAe,cAAc,OAAO,iBAAiB,UAAU,EAAE;AACpH,QAAM,CAAC,OAAO,WAAW,cAAc,IAAI,QAAQ,QAAQ,iBAAiB,UAAU;AACtF,MAAI,UAAU,QAAW;AACvB,WAAO,CAAC;AAAA,EACV;AAEA,QAAM,UAAU,KAAK,EAAE;AAIvB,MAAI,UAAU,UAAU,mBAAmB,CAAC,YAAY;AACtD,WAAO,CAAC,KAAK;AAAA,EACf;AAEA,QAAM,sBAAsB,aAAa,CAAC,GAAG,kBAAkB,KAAK,IAAI;AAExE,SAAO,mBAAmB,EAAC,OAAO,gBAAgB,SAAS,YAAY,kBAAkB,oBAAmB,CAAC;AAE7G,WAAS,mBAAmB,EAAC,OAAAC,QAAO,gBAAAC,kBAAiB,OAAO,SAAAC,WAAU,OAAO,YAAAC,cAAa,OAAO,kBAAAC,oBAAmB,CAAC,EAAC,GAAG;AACvH,UAAM,oBAAoB;AAC1B,UAAM,CAAC,WAAW,IAAI,WAAW,MAAM;AACvC,QAAI,gBAAgB,QAAW;AAC7B,UAAIF,UAAS;AACX,cAAMG,uBAAsBF,cAAa,CAAC,GAAGC,mBAAkB,GAAG,WAAW,QAAQJ,MAAK,EAAE,IAAII;AAChG,eAAO,gBAAgB,EAAC,OAAO,GAAG,WAAW,QAAQJ,MAAK,IAAI,gBAAgB,MAAM,YAAAG,aAAY,kBAAkBE,qBAAmB,CAAC;AAAA,MACxI;AACA,aAAOF,cAAaC,oBAAmB,CAAC,GAAG,WAAW,QAAQJ,MAAK,EAAE;AAAA,IACvE;AACA,WAAO,qBAAqB,EAAC,OAAAA,QAAO,gBAAAC,iBAAgB,SAAAC,UAAS,YAAAC,aAAY,kBAAAC,kBAAgB,CAAC;AAAA,EAE5F;AAEA,WAAS,qBAAqB,EAAC,OAAAJ,QAAO,gBAAAC,kBAAiB,OAAO,SAAAC,WAAU,OAAO,YAAAC,cAAa,OAAO,kBAAAC,oBAAmB,CAAC,EAAC,GAAG;AACzH,UAAM,CAAC,cAAc,IAAI,cAAc,MAAM;AAC7C,QAAI,mBAAmB,QAAW;AAChC,UAAIF,UAAS;AACX,cAAMG,uBAAsBF,cAAa,CAAC,GAAGC,mBAAkB,GAAG,cAAc,QAAQJ,MAAK,EAAE,IAAII;AACnG,eAAO,gBAAgB,EAAC,OAAO,GAAG,cAAc,QAAQJ,MAAK,IAAI,gBAAgB,MAAM,YAAAG,aAAY,kBAAkBE,qBAAmB,CAAC;AAAA,MAC3I;AACA,aAAOF,cAAaC,oBAAmB,CAAC,GAAG,cAAc,QAAQJ,MAAK,EAAE;AAAA,IAC1E;AACA,QAAIC,mBAAkB,CAACC,UAAS;AAC9B,aAAOC,cAAaC,oBAAmB,CAAC,GAAGJ,MAAK,EAAE;AAAA,IACpD;AACA,WAAO,gBAAgB,EAAC,OAAAA,QAAO,gBAAAC,iBAAgB,YAAAE,aAAY,kBAAAC,kBAAgB,CAAC;AAAA,EAC9E;AAEA,WAAS,gBAAgB,EAAC,OAAAJ,QAAO,gBAAAC,kBAAiB,OAAO,YAAAE,cAAa,OAAO,kBAAAC,oBAAmB,CAAC,EAAC,GAAG;AACnG,UAAM,CAAC,SAAS,IAAI,QAAQ,MAAM;AAClC,QAAI,cAAc,QAAW;AAC3B,YAAM,sBAAsB,GAAG,SAAS,QAAQJ,MAAK;AACrD,aAAOG,cAAa,CAAC,GAAGC,mBAAkB,mBAAmB,IAAI,CAAC,mBAAmB;AAAA,IACvF;AACA,QAAIH,iBAAgB;AAClB,aAAOE,cAAaC,oBAAmB,CAAC,GAAGJ,MAAK,EAAE;AAAA,IACpD;AACA,WAAO,CAAC;AAAA,EACV;AAEF;AAEO,gBAAS,WAAW,QAAQ;AACjC,QAAMN,SAAQ,kBAAkB,qEAAqE;AACrG,QAAM,YAAYA,OAAM,OAAO,MAAM;AACrC,EAAAA,OAAM,qCAAqC;AAG3C,QAAM,SAAS,UAAU,MAAM;AAE/B,MAAI,mBAAmB,MAAM,GAAG;AAC9B,UAAM,YAAY,OAAO,MAAM,EAC5B,QAAQ,2BAA2B,EAAE,EAErC,QAAQ,QAAQ,GAAG,EACnB,KAAK,EACL,MAAM,GAAG,EAAE,EACX,KAAK;AAGR,UAAM,gBAAgB,mBAAmB,SAAS;AAElD,cAAU,kBAAkB,SAAS,EAAE;AACvC,QAAI,UAAU,UAAU,GAAG;AACzB,aAAO,CAAC,cAAc,gBAAgB,KAAK,GAAG,GAAG,SAAS,IAAI;AAAA,IAChE;AACA,WAAO,CAAC;AAAA,EACV;AAEA,SAAO,CAAC;AAER,WAAS,UAAUY,SAAQ;AAEzB,UAAM,CAAC,KAAK,IAAIA,QAAO,IAAI,gCAAgC;AAG3D,QAAI,OAAO;AACT,YAAM,eAAe,MAAM,UACxB,OAAO,CAAC,EAAC,KAAI,MAAM,CAAC,GAAG,EAAE,SAAS,IAAI,CAAC,EACvC,IAAI,CAAC,EAAC,MAAK,MAAM,mBAAmB,KAAK,IAAI,OAAO,KAAK,IAAI,EAAE,EAC/D,OAAO,WAAS,KAAK,EAErB,KAAK,GAAG;AACX,aAAO;AAAA,IACT;AACA,WAAO;AAAA,EACT;AACF;AAEO,gBAAS,cAAc,QAAQ;AACpC,QAAMZ,SAAQ,kBAAkB,wEAAwE;AACxG,QAAM,YAAYA,OAAM,OAAO,MAAM;AACrC,EAAAA,OAAM,wCAAwC;AAG9C,QAAM,YAAY,aAAa,MAAM;AACrC,MAAI,mBAAmB,SAAS,GAAG;AACjC,UAAM,YAAY,OAAO,SAAS,EAC/B,QAAQ,2BAA2B,EAAE,EAErC,QAAQ,QAAQ,GAAG,EACnB,KAAK,EACL,MAAM,GAAG,EAAE,EACX,KAAK;AAER,cAAU,sBAAsB,SAAS,GAAG;AAC5C,QAAI,UAAU,WAAW,GAAG;AAC1B,aAAO,CAAC;AAAA,IACV;AAEA,WAAO,CAAC,iBAAiB,SAAS,GAAG;AAAA,EACvC;AAEA,SAAO,CAAC;AAER,WAAS,aAAaY,SAAQ;AAE5B,UAAM,CAAC,KAAK,IAAIA,QAAO,IAAI,WAAW,EAAE,OAAO,OAAK,EAAE,UAAU,KAAK,QAAM,GAAG,SAAS,GAAG,CAAC;AAG3F,QAAI,OAAO;AACT,YAAM,kBAAkB,MAAM,UAC3B,OAAO,CAAC,EAAC,KAAI,MAAM,CAAC,GAAG,EAAE,SAAS,IAAI,CAAC,EACvC,IAAI,CAAC,EAAC,MAAK,MAAM,mBAAmB,KAAK,IAAI,OAAO,KAAK,IAAI,EAAE,EAC/D,OAAO,WAAS,KAAK,EAErB,KAAK,GAAG,EACR,KAAK;AACR,aAAO;AAAA,IACT;AACA,WAAO;AAAA,EACT;AACF;AAEO,gBAAS,QAAQ,QAAQ;AAC9B,QAAMZ,SAAQ,kBAAkB,kEAAkE;AAClG,QAAM,YAAYA,OAAM,OAAO,MAAM;AACrC,EAAAA,OAAM,wCAAwC;AAE9C,QAAM,OAAO,QAAQ,MAAM;AAC3B,MAAI,MAAM;AACR,WAAO,CAAC,YAAY,IAAI,GAAG;AAAA,EAC7B;AACA,SAAO,CAAC;AAER,WAAS,QAAQY,SAAQ;AACvB,UAAM,CAAC,IAAI,IAAIA,QAAO,IAAI,QAAQ;AAClC,QAAI,CAAC,QAAQ,CAAC,KAAK,SAAS,KAAK,MAAM,SAAS,IAAI;AAClD,MAAAZ,OAAM,qBAAqB;AAC3B,aAAO;AAAA,IACT;AAEA,cAAU,SAAS,KAAK,UAAU,IAAI,CAAC,EAAE;AACzC,UAAM,EAAC,MAAK,IAAI;AAChB,QAAI,UAAU,UAAU,UAAU,QAAQ;AACxC,aAAO;AAAA,IACT;AACA,WAAO,MAAM,MAAM,GAAG,EAAE;AAAA,EAC1B;AACF;AAEO,gBAAS,mBAAmB,WAAW;AAC5C,QAAM,aAAa,UAAU,YAAY;AAEzC,QAAM,oBAAoB,CAAC,QAAQ,OAAO,QAAQ,MAAM;AACxD,SAAO,kBAAkB,KAAK,UAAQ,WAAW,WAAW,IAAI,CAAC;AACnE;AAEO,gBAAS,uBAAuB,QAAQ;AAE7C,QAAMA,SAAQ,kBAAkB,iFAAiF;AACjH,QAAM,YAAYA,OAAM,OAAO,MAAM;AACrC,EAAAA,OAAM,2CAA2C;AAKjD,QAAM,SAAS,OAAO,IAAI,gCAAgC;AAC1D,QAAM,cAAc,CAAC,EAAE,OAAO,GAAG,OAAO,IAAI,aAAa,CAAC;AAC1D,QAAM,oBAAoB,CAAC,GAAG,IAAI,IAAI,WAAW,CAAC;AAElD,YAAU,+BAA+B,KAAK,UAAU,MAAM,CAAC,EAAE;AACjE,YAAU,gBAAgB,YAAY,MAAM,MAAM,KAAK,UAAU,WAAW,CAAC,EAAE;AAC/E,YAAU,uBAAuB,kBAAkB,MAAM,MAAM,KAAK,UAAU,iBAAiB,CAAC,EAAE;AAElG,MAAI,kBAAkB,SAAS,GAAG;AAChC,IAAAA,OAAM,2CAA2C;AACjD,WAAO,CAAC;AAAA,EACV;AAEA,SAAO,UAAU,mBAAmB,eAAe;AAEnD,WAAS,cAAc,EAAC,KAAK,UAAS,GAAG;AACvC,UAAM,iBAAkB;AACxB,UAAM,gBAAiB;AAEvB,QAAI,QAAQ,OAAO;AACjB,aAAO,UACJ,OAAO,QAAM,GAAG,SAAS,OAAO,GAAG,MAAM,MAAM,iBAAiB,CAAC,EACjE,IAAI,QAAM,GAAG,KAAK;AAAA,IACvB;AAEA,QAAI,QAAQ,OAAO;AACjB,aAAO,UACJ,OAAO,SAAO,CAAC,KAAK,GAAG,EAAE,SAAS,IAAI,IAAI,KAAK,mBAAmB,IAAI,KAAK,KAAK,eAAe,KAAK,OAAO,IAAI,KAAK,CAAC,CAAC,EACtH,IAAI,CAAC,EAAC,MAAK,MAAM,OAAO,KAAK,CAAC;AAAA,IACnC;AAEA,QAAI,QAAQ,OAAO;AACjB,aAAO,UACJ,OAAO,SAAO,CAAC,KAAK,KAAK,GAAG,EAAE,SAAS,IAAI,IAAI,KAAK,mBAAmB,IAAI,KAAK,KAAK,eAAe,KAAK,OAAO,IAAI,KAAK,CAAC,CAAC,EAC3H,IAAI,CAAC,EAAC,MAAK,MAAM,OAAO,KAAK,CAAC;AAAA,IACnC;AAEA,QAAI,QAAQ,OAAO;AACjB,YAAM,IAAI,UAAU,KAAK,QAAM,GAAG,SAAS,GAAG;AAC9C,UAAI,GAAG;AACL,cAAM,IAAI,UAAU,KAAK,QAAM,GAAG,SAAS,GAAG;AAE9C,YAAI,GAAG;AACL,iBAAO,CAAC,IAAI,EAAE,KAAK,IAAI,IAAI,EAAE,KAAK,IAAI,EAAE,KAAK,EAAE;AAAA,QACjD;AACA,eAAO,CAAC,IAAI,EAAE,KAAK,EAAE;AAAA,MACvB;AAAA,IACF;AAGA,WAAO,UACJ,OAAO,SAAO,CAAC,KAAK,GAAG,EAAE,SAAS,IAAI,IAAI,KAAK,mBAAmB,IAAI,KAAK,KAAK,cAAc,KAAK,OAAO,IAAI,KAAK,CAAC,CAAC,EACrH,IAAI,CAAC,EAAC,MAAK,MAAM,OAAO,KAAK,CAAC;AAAA,EACnC;AACF;",
-  "names": ["debug", "fSids", "sidStrings", "field", "title", "formatted", "query", "queryIsOkAlone", "addYear", "alternates", "alternateQueries", "newAlternateQueries", "record"]
+  "sourcesContent": ["import createDebugLogger from 'debug';\nimport {toQueries} from '../candidate-search-utils.js';\nimport {getMelindaIdsF035, validateSidFieldSubfieldCounts, getSubfieldValues, testStringOrNumber} from '../../matching-utils.js';\nimport {extractHostMelindaIdsFromField, extractPublicationYearFrom773, getHostMelindaFields} from './component.js';\nimport {fieldToString} from '@natlibfi/marc-record-validators-melinda';\nimport {getTitle} from '../../match-detection/features/bib/title.js';\nimport {uniqArray} from '@natlibfi/marc-record-validators-melinda/dist/utils.js';\n\nconst debug = createDebugLogger('@natlibfi/melinda-record-matching:candidate-search:query');\nconst debugDev = debug.extend('dev');\n\n\nconst IS_OK_ALONE_THRESHOLD = 5;\n\nexport function bibSourceIds(record) {\n\n  /* Melinda's SRU-index melinda.sourceid includes source IDs from SID fields in Melinda records\n     SID-fields in Melinda have sf $c with local id and sf $b with a code for the local db:\n\n     SID__ $c 123457 $b helka\n     SID__ $c (ANDL100020)1077305 $b sata\n     SID__ $c VER999999 $ FI-KV\n     SID__ $c /10024/508126 $ REPO_THESEUS\n\n    In melinda.sourceid -index case is kept, sourceprefixes in brackets and hyphens are normalized away:\n    Note: slashes are not normalized away, but a SRU-search-string including slashes needs to be quoted\n\n    1234567helka\n    1077305sata\n    VER999999FIKV\n    /10024/508126REPO_THESEUS\n\n    Note: All Melinda records that have a matching records in a local db do NOT have SID for that local records,\n          existence of a SID field depends on how the record has been added to Melinda and how it has been handled\n          afterwards. SIDs are also not reliably maintained. A record might or might not have a SID for a local db\n          after the matching record is removed from the local db.\n\n   */\n\n\n  const debug = createDebugLogger('@natlibfi/melinda-record-matching:candidate-search:query:source-ids');\n  const debugData = debug.extend('data');\n  //const debugInfo = debug.extend('info');\n\n  debug(`Creating queries for sourceid's`);\n\n  const fSids = record.get('SID');\n  debugData(`SID-fields (${fSids.length}): ${JSON.stringify(fSids)}`);\n\n  return fSids.length > 0 ? toSidQueries(fSids) : [];\n\n  function toSidQueries(fSids) {\n    debug(`Creating actual queries for sourceid's`);\n\n    const sidStrings = getSidStrings(fSids);\n\n    if (sidStrings.length < 1) {\n      debug(`No identifiers found, no queries created.`);\n      return [];\n    }\n\n    const sidQueries = toQueries(sidStrings, 'melinda.sourceid');\n\n    return sidQueries;\n\n    function getSidStrings(fSids) {\n      debug(`Getting Sid strings from SID fields`);\n\n      // Map SID fields to valid sidStrings, filter out empty strings\n      const sidStrings = fSids.map(toSidString).filter(nonEmptySid => nonEmptySid);\n      return sidStrings;\n\n      function toSidString(field) {\n        debug(`Getting string from a field`);\n\n        return validateSidFieldSubfieldCounts(field) ? createSidString(field) : '';\n\n        function createSidString(field) {\n          debug(`Creating string from a field`);\n          const [sfC] = getSubfieldValues(field, 'c');\n          const [sfB] = getSubfieldValues(field, 'b');\n\n          const cleanedSfC = removeSourcePrefix(normalizeSidSubfieldValue(sfC));\n          const cleanedSfB = normalizeSidSubfieldValue(sfB);\n\n          debugData(`${JSON.stringify(sfC)} + ${JSON.stringify(sfB)}`);\n          return cleanedSfC.concat(cleanedSfB);\n        }\n\n        function removeSourcePrefix(subfieldValue) {\n          const sourcePrefixRegex = (/^(?<sourcePrefix>\\([A-Za-z0-9-]+\\))(?<id>.+)$/u);\n          const normalizedValue = testStringOrNumber(subfieldValue) ? String(subfieldValue).replace(sourcePrefixRegex, '$<id>') : '';\n          debugData(`Normalized ${subfieldValue} to ${normalizedValue}`);\n          return normalizedValue;\n        }\n\n        function normalizeSidSubfieldValue(subfieldValue) {\n          debugData(`Normalizing ${subfieldValue}`);\n          const normalizeAwayRegex = (/[- ]/u);\n          return testStringOrNumber(subfieldValue) ? String(subfieldValue).replace(normalizeAwayRegex, '') : '';\n        }\n\n      }\n    }\n  }\n}\n\nexport function bibMelindaIds(record) {\n  // Melinda's SRU-index melinda.melindaid includes f001 controlnumbers and old Melinda-IDs from f035z's for all non-deleted Melinda-records\n\n  const debug = createDebugLogger('@natlibfi/melinda-record-matching:candidate-search:query:bibMelindaIds');\n  const debugData = debug.extend('data');\n  debug(`Creating queries for MelindaIds`);\n\n  // Note: Melinda-ID's for search queries are created just from records f035a's and f035z's\n  // Both (FI-MELINDA)- and FCC-prefixed forms are found\n  // f001 controlnumber is not currently included, even if record's f003 is FI-MELINDA\n  const melindaIds = getMelindaIdsF035(record);\n\n  debugData(`Unique identifiers (${melindaIds.length}): ${JSON.stringify(melindaIds)}`);\n\n  if (melindaIds.length < 1) {\n    debug(`No identifiers found, no queries created.`);\n    return [];\n  }\n\n  return toQueries(melindaIds, 'melinda.melindaid');\n}\n\n\n// bibHostComponents returns host id from the first subfield $w of first field f773, see test-fixtures 04 and 05\n// bibHostComponents should search all 773 $ws for possible host id, but what should it do in case of multiple host ids?\nexport function bibHostComponents(record) {\n  const debug = createDebugLogger('@natlibfi/melinda-record-matching:candidate-search:query:bibHostComponents');\n  const debugData = debug.extend('data');\n  debug(`Creating queries for hostIds`);\n  const f773s = getHostMelindaFields(record);\n  if (f773s.length !== 1) { // This needs some thinking...\n    return [];\n  }\n\n  const [id] = extractHostMelindaIdsFromField(f773s[0]);\n  debug(`Found id: ${JSON.stringify(id)}`);\n\n  // NB! record.isCR() does not work if LDR/07=a, so we get year from 773$g!\n  const year = extractPublicationYearFrom773(f773s[0]);\n  debug(`${fieldToString(f773s[0])} => ${year ? year : 'N/A'}`);\n  if (year) {\n    return [`dc.date=${year} AND melinda.partsofhost=${id}`];\n  }\n\n  return [`melinda.partsofhost=${id}`];\n\n}\n\n\n\n\n// SRU search dc.title with a search phrase starting with ^ maps currently in Melinda to (probably) to *headings* index TIT\n// - Aleph cannot currently handle headings searches starting with a boolean - in these cases use word search\n\nexport function bibTitle(record) {\n  // We get author/publisher only when formatted title is shorter than 5 chars\n  return bibTitleAuthorPublisher({record, onlyTitleLength: 5});\n}\n\nexport function bibTitleAuthor(record) {\n  debug('bibTitleAuthor');\n  // We use onlyTitleLength that is longer than our formatted length to\n  // get an author or an publisher always\n  return bibTitleAuthorPublisher({record, onlyTitleLength: 100});\n}\n\nexport function bibTitleAuthorYear(record) {\n  debug('bibTitleAuthorYearAlternates');\n  // We use onlyTitleLength that is longer than our formatted length to\n  // get an author or an publisher always\n\n  return bibTitleAuthorPublisher({record, onlyTitleLength: 100, addYear: true});\n}\n\nexport function bibTitleAuthorYearAlternates(record) {\n  debug('bibTitleAuthorYearAlternates');\n  // We use onlyTitleLength that is longer than our formatted length to\n  // get an author or an publisher always\n  const origQueryList = bibTitleAuthorPublisher({record, onlyTitleLength: 100, addYear: true, alternates: true, alternateQueries: []});\n  debug(`${JSON.stringify(origQueryList)}`);\n  return {queryList: Array.from(origQueryList).reverse(), queryListType: 'alternates'};\n}\n\nfunction getTitleForQuery(record) {\n  const title = getTitle(record, ['a']);\n  debug(`getTitleForQuery: title ${title}`);\n  if (title) {\n    const nWords = title.split(' ');\n    // If lone $a is deemed too short, fetch $b as well:\n    // (NV: I've seen pairs with 245$a-only vs 245$a$b, so I'd like to use $a only if it is long enough)\n\n    if (nWords < 3 || title.length < 12) {\n      const [f245] = record.get('245');\n      // If punctuation is ' =' I think that f245$a is good enough despite shortness. Trying to balance between two bad situations...\n      // \"Suo (= short title) siell\u00E4, vetel\u00E4 (= missing $b name in another language) t\u00E4\u00E4ll\u00E4...\"\n      if (f245 && f245.subfields.find(sf => sf.code === 'a' && sf.value.match(/ =$/u)) && title.length >= IS_OK_ALONE_THRESHOLD) {\n        return title;\n      }\n      return getTitle(record, ['a', 'b']); // Try to get a longer title\n    }\n  }\n  return title;\n}\n\nfunction dcTitle(record, onlyTitleLength, alternates = false) {\n  const title = getTitleForQuery(record);\n  debug(`dcTitle: title: ${title}`);\n  if (!testStringOrNumber(title)) {\n    return [];\n  }\n\n  const formatted = getFormattedTitle(title);\n  debug(`dcTitle: formatted: ${formatted}`);\n\n  // use word search for titles starting with a boolean\n  const useWordSearch = checkUseWordSearch(formatted);\n  // Prevent too many matches / SRU crashing by having a minimum length\n  // Note that currently this fails matching if there are no matches from previous matchers\n  if (formatted.length >= onlyTitleLength && !alternates) {\n    return [`dc.title=\"${useWordSearch ? '' : '^'}${formatted}*\"`, formatted, true];\n  }\n\n  const queryIsOkAlone = !useWordSearch && formatted.length >= IS_OK_ALONE_THRESHOLD;\n\n  if (formatted.length<1) {\n    return [\"\", formatted, queryIsOkAlone];\n  }\n\n  // use word search without ending * also in combination searches to avoid SRU-server crashes [MRA-189]\n  return [`dc.title=\"${useWordSearch || !queryIsOkAlone ? '' : '^'}${formatted}${queryIsOkAlone ? '*' : ''}\"`, formatted, queryIsOkAlone];\n\n  function getFormattedTitle(title) {\n    const formatted = String(title)\n      .replace(/[\\-+ !\"(){}\\[\\]<>;:.?/@*%=^_`~]/gu, ' ')\n      .replace(/[^\\w\\s\\p{Alphabetic}]/gu, '') // Apparently matches to/works with non-aplhabetic scripts such as Chinese as well\n      .replace(/  +/gu, ' ') // Clean up concurrent spaces from eg. subfield changes\n      .trim()\n      .replace(/^(.{30}\\S*) .*$/, \"$1\")\n      .trim();\n\n    return formatted;\n  }\n}\n\nexport function bibTitleAuthorPublisher({record, onlyTitleLength, addYear = false, alternates = false, alternateQueries = []}) {\n  debug(`bibTitleAuthorPublisher, onlyTitleLength: ${onlyTitleLength}, addYear: ${addYear}, alternates: ${alternates}`);\n  const [query, formatted, queryIsOkAlone] = dcTitle(record, onlyTitleLength, alternates);\n  debug(`query ${JSON.stringify(query)}, formatted: ${JSON.stringify(formatted)}, isOKAlone: ${JSON.stringify(queryIsOkAlone)}`);\n  if (query === undefined) {\n    return [];\n  }\n\n  debug(`query: ${query}`);\n\n  // Prevent too many matches / SRU crashing by having a minimum length\n  // Note that currently this fails matching if there are no matches from previous matchers\n  if (formatted.length >= onlyTitleLength && !alternates) {\n    return [query];\n  }\n\n  const newAlternateQueries = alternates ? [...alternateQueries, query] : alternateQueries;\n\n  return addAuthorsToSearch({query, queryIsOkAlone, addYear, alternates, alternateQueries: newAlternateQueries});\n\n  function addAuthorsToSearch({query, queryIsOkAlone = false, addYear = false, alternates = false, alternateQueries = []}) {\n    debug('addAuthorsToSearch');\n    const [authorQuery] = bibAuthors(record);\n    if (authorQuery !== undefined) {\n      const combinedQuery = query.length > 0 ? `${authorQuery} AND ${query}` : `${authorQuery}`;\n      if (addYear) {\n        const newAlternateQueries = alternates ? [...alternateQueries, combinedQuery] : alternateQueries;\n        return addYearToSearch({query: combinedQuery, queryIsOkAlone: true, alternates, alternateQueries: newAlternateQueries});\n      }\n      return alternates ? alternateQueries : [combinedQuery];\n    }\n    return addPublisherToSearch({query, queryIsOkAlone, addYear, alternates, alternateQueries});\n    //return [];\n  }\n\n  function addPublisherToSearch({query, queryIsOkAlone = false, addYear = false, alternates = false, alternateQueries = []}) {\n    const [publisherQuery] = bibPublishers(record);\n    if (publisherQuery !== undefined) {\n      const combinedQuery = query.length > 0 ? `${publisherQuery} AND ${query}` : `${publisherQuery}`;\n      if (addYear) {\n        const newAlternateQueries = alternates ? [...alternateQueries, combinedQuery] : alternateQueries;\n        return addYearToSearch({query: combinedQuery, queryIsOkAlone: true, alternates, alternateQueries: newAlternateQueries});\n      }\n      return alternates ? alternateQueries : [combinedQuery];\n    }\n    if (queryIsOkAlone && !addYear) {\n      return alternates ? alternateQueries : [`${query}`];\n    }\n    return addYearToSearch({query, queryIsOkAlone, alternates, alternateQueries});\n  }\n\n  function addYearToSearch({query, queryIsOkAlone = false, alternates = false, alternateQueries = []}) {\n    const [yearQuery] = bibYear(record);\n    if (yearQuery !== undefined) {\n      const yearAndOtherQueries = query.length > 0 ? `${yearQuery} AND ${query}` : `${yearQuery}`;\n      return alternates ? [...alternateQueries, yearAndOtherQueries] : [yearAndOtherQueries];\n    }\n    if (queryIsOkAlone) {\n      return alternates ? alternateQueries : [`${query}`];\n    }\n    return [];\n  }\n\n}\n\nexport function bibAuthors(record) {\n  const debug = createDebugLogger('@natlibfi/melinda-record-matching:candidate-search:query:bibAuthors');\n  const debugData = debug.extend('data');\n  debug(`Creating query for the first author`);\n  //debugData(record);\n\n  const author = getAuthor(record);\n\n  if (testStringOrNumber(author)) {\n    const formatted = String(author)\n      .replace(/[^\\w\\s\\p{Alphabetic}]/gu, '')\n      // Clean up concurrent spaces from fe. subfield changes\n      .replace(/ +/gu, ' ')\n      .trim()\n      .slice(0, 30)\n      .trim();\n\n    // use word search for authors starting with a boolean\n    const useWordSearch = checkUseWordSearch(formatted);\n    // Prevent too many matches by having a minimum length\n    debugData(`Author string: ${formatted}`);\n    if (formatted.length >= 5) {\n      return [`dc.author=\"${useWordSearch ? '' : '^'}${formatted}*\"`];\n    }\n    return [];\n  }\n\n  return [];\n\n  function getAuthor(record) {\n    //debugData(record);\n    const [field] = record.get(/^(?:100|110|111|700|710|711)$/u);\n    //debugData(field);\n\n    if (field) {\n      const authorString = field.subfields\n        .filter(({code}) => ['a'].includes(code)) // We might use different subfield code sets for X00, X10 and X11?\n        .map(({value}) => testStringOrNumber(value) ? String(value) : '')\n        .filter(value => value)\n        // In Melinda's index subfield separators are indexed as ' '\n        .join(' ');\n      return authorString;\n    }\n    return false;\n  }\n}\n\nexport function bibPublishers(record) {\n  const debug = createDebugLogger('@natlibfi/melinda-record-matching:candidate-search:query:bibPublishers');\n  const debugData = debug.extend('data');\n  debug(`Creating query for the first publisher`);\n  //debugData(record);\n\n  const publisher = getPublisher(record);\n  if (testStringOrNumber(publisher)) {\n    const formatted = String(publisher)\n      .replace(/[^\\w\\s\\p{Alphabetic}]/gu, '')\n      // Clean up concurrent spaces from fe. subfield changes\n      .replace(/ +/gu, ' ')\n      .trim()\n      .slice(0, 30)\n      .trim();\n\n    debugData(`Publisher string: '${formatted}'`);\n    if (formatted.length === 0) {\n      return [];\n    }\n    // use non-wildcard word search from dc.publisher\n    return [`dc.publisher=\"${formatted}\"`];\n  }\n\n  return [];\n\n  function getPublisher(record) {\n    //debugData(record);\n    const [field] = record.get(/^26[04]$/u).filter(f => f.subfields.some(sf => sf.code === 'b')); // Filter removes copyright-only fields\n    //debugData(field);\n\n    if (field) {\n      const publisherString = field.subfields\n        .filter(({code}) => ['b'].includes(code))\n        .map(({value}) => testStringOrNumber(value) ? String(value) : '')\n        .filter(value => value)\n        // In Melinda's index subfield separators are indexed as ' '\n        .join(' ')\n        .trim();\n      return publisherString;\n    }\n    return false;\n  }\n}\n\nexport function bibYear(record) {\n  const debug = createDebugLogger('@natlibfi/melinda-record-matching:candidate-search:query:bibYear');\n  const debugData = debug.extend('data');\n  debug(`Creating query for the publishing year`);\n\n  const year = getYear(record);\n  if (year) {\n    return [`dc.date=\"${year}\"`];\n  }\n  return [];\n\n  function getYear(record) {\n    const [f008] = record.get(/^008$/u);\n    if (!f008 || !f008.value || f008.value.length < 11) {\n      debug('f008 missing/crappy');\n      return false;\n    }\n\n    debugData(`f008: ${JSON.stringify(f008)}`);\n    const {value} = f008;\n    const year = value.slice(7, 11);\n    if (year === '||||' || year === '    ') { // Meaningless values\n      debug('Meaningless year in f008');\n      return false;\n    }\n    return year;\n  }\n}\n\nexport function checkUseWordSearch(formatted) {\n  const lowercased = formatted.toLowerCase();\n  // Note: add a space to startWords to catch just actual boolean words\n  const booleanStartWords = ['and ', 'or ', 'nor ', 'not '];\n  return booleanStartWords.some(word => lowercased.startsWith(word));\n}\n\nexport function bibStandardIdentifiers(record) {\n\n  const debug = createDebugLogger('@natlibfi/melinda-record-matching:candidate-search:query:bibStandardIdentifiers');\n  const debugData = debug.extend('data');\n  const debugDev = debug.extend('dev');\n\n  debug(`Creating queries for standard identifiers`);\n\n  // DEVELOP: query identifiers from their own indexes (this needs some mapping work in SRU/Aleph)\n\n  // const fields = record.get(/^(?<def>020|022|024)$/u);\n  const fields = record.get(/^(?<def>015|020|022|024|028)$/u);\n  const identifiers = [].concat(...fields.map(toIdentifiers));\n  const uniqueIdentifiers = [...new Set(identifiers)];\n\n  debugData(`Standard identifier fields: ${JSON.stringify(fields)}`);\n  debugData(`Identifiers (${identifiers.length}): ${JSON.stringify(identifiers)}`);\n  debugData(`Unique identifiers (${uniqueIdentifiers.length}): ${JSON.stringify(uniqueIdentifiers)}`);\n\n  if (uniqueIdentifiers.length < 1) {\n    debug(`No identifiers found, no queries created.`);\n    return [];\n  }\n\n  return toQueries(uniqueIdentifiers, 'dc.identifier');\n\n  function toIdentifiers({tag, subfields}) {\n    const issnIsbnReqExp = (/^[A-Za-z0-9-]+$/u);\n    const otherIdReqExp = (/^[A-Za-z0-9-:]+$/u);\n\n    if (tag === '015') { // TODO: test (does this use the right index etc.)\n      return subfields\n        .filter(sf => sf.code === 'a' && sf.value.match(/^[A-Z0-9\\-]+$/ui))\n        .map(sf => sf.value)\n    }\n\n    if (tag === '020') {\n      return subfields\n        .filter(sub => ['a', 'z'].includes(sub.code) && testStringOrNumber(sub.value) && issnIsbnReqExp.test(String(sub.value)))\n        .map(({value}) => String(value));\n    }\n\n    if (tag === '022') {\n      return subfields\n        .filter(sub => ['a', 'z', 'y'].includes(sub.code) && testStringOrNumber(sub.value) && issnIsbnReqExp.test(String(sub.value)))\n        .map(({value}) => String(value));\n    }\n\n    if (tag === '028') { // TODO: test\n      const a = subfields.find(sf => sf.code === 'a');\n      const b = subfields.find(sf => sf.code === 'b');\n\n      // Handle cases where there is a qualifier in brackets after value\n      const cleanAs = cleanQualifiers(a?.value);\n      const cleanBs = cleanQualifiers(b?.value);\n      debugDev(JSON.stringify(cleanAs));\n      debugDev(JSON.stringify(cleanBs));\n\n      // Handle cases where there are several values separated by a comma in subfield\n      const splitAs = cleanAs.map(elem => elem.split(',')).concat(cleanAs).flat();\n      const splitBs = cleanBs.map(elem => elem.split(',')).concat(cleanBs).flat();\n      debugDev(JSON.stringify(splitAs));\n      debugDev(JSON.stringify(splitBs));\n\n      const normAs = splitAs.map(elem => normalizeF028ForSearch(elem));\n      const normBs = splitBs.map(elem => normalizeF028ForSearch(elem));\n      debugDev(JSON.stringify(normAs));\n      debugDev(JSON.stringify(normBs));\n\n      const identifiers = normAs.map((normA, index) => {\n        debugDev(`Handling ${normA} from index: ${index}`);\n        // Paired $a + $b\n        if (normA.length > 0 && normBs[index]?.length > 0) {\n          debugDev(`Found a (${normA}) + b (${normBs[index]})`);\n          // let's not return plain $b\n          return [`^${normBs[index]} ${normA}`, `^${normA} ${normBs[index]}`, `^${normA}`];\n          //return [`^${normA}`, `^${normBs[index]}`, `^${normBs[index]} ${normA}`, `^${normA} ${normBs[index]}`];\n\n        }\n        // if there is no pair, get $a + first $b\n        // NOTE: we do not get all possible pairs\n        if (normA.length > 0 && normBs[0]?.length > 0) {\n          debugDev(`Found a (${normA}) + b (${normBs[0]})`);\n          // let's not return plain $b\n          return [`^${normBs[0]} ${normA}`, `^${normA} ${normBs[0]}`, `^${normA}`];\n          //return [`^${normA}`, `^${normBs[index]}`, `^${normBs[index]} ${normA}`, `^${normA} ${normBs[index]}`];\n\n        }\n        // Otherwise just $a\n        if (normA.length > 0) {\n          debugDev(`Found a (${normA}) but no b ${normBs[index]}`);\n          return [`^${normA}`];\n        }\n        return [];\n      });\n\n      debugDev(`Current identifiers from f028: ${JSON.stringify(identifiers)}`);\n      return uniqArray(identifiers.flat());\n\n      function cleanQualifiers(val) {\n        if (val !== undefined) {\n        const cleanVal = val.replace(/ +\\(.*\\)/gu, '').replace(/ +/ug, ' ').replace(/^ +/ug, '').replace(/ +$/ug, '');\n        return uniqArray([val, cleanVal]);\n        }\n        return [];\n      }\n\n      function normalizeF028ForSearch(val) {\n        // see normalization routine 34 used for 028-index in Aleph\n        const f028NormCharsRegex = /[-\"<>;:%=~`!\\(\\)\\{\\}\\.\\?\\/\\@\\*\\^]/g;\n        const normVal = val.replace(f028NormCharsRegex, ' ').replace(/ +/ug, ' ').replace(/^ +/ug, '').replace(/ +$/ug, '').toLowerCase();\n        // Do not return a value normalized to just a space\n        return normVal !== \" \" ? normVal : \"\";\n      }\n\n    }\n\n    // Default seems to be 024:\n    return subfields\n      .filter(sub => ['a', 'z'].includes(sub.code) && testStringOrNumber(sub.value) && otherIdReqExp.test(String(sub.value)))\n      .map(({value}) => String(value));\n  }\n}\n"],
+  "mappings": "AAAA,OAAO,uBAAuB;AAC9B,SAAQ,iBAAgB;AACxB,SAAQ,mBAAmB,gCAAgC,mBAAmB,0BAAyB;AACvG,SAAQ,gCAAgC,+BAA+B,4BAA2B;AAClG,SAAQ,qBAAoB;AAC5B,SAAQ,gBAAe;AACvB,SAAQ,iBAAgB;AAExB,MAAM,QAAQ,kBAAkB,0DAA0D;AAC1F,MAAM,WAAW,MAAM,OAAO,KAAK;AAGnC,MAAM,wBAAwB;AAEvB,gBAAS,aAAa,QAAQ;AA0BnC,QAAMA,SAAQ,kBAAkB,qEAAqE;AACrG,QAAM,YAAYA,OAAM,OAAO,MAAM;AAGrC,EAAAA,OAAM,iCAAiC;AAEvC,QAAM,QAAQ,OAAO,IAAI,KAAK;AAC9B,YAAU,eAAe,MAAM,MAAM,MAAM,KAAK,UAAU,KAAK,CAAC,EAAE;AAElE,SAAO,MAAM,SAAS,IAAI,aAAa,KAAK,IAAI,CAAC;AAEjD,WAAS,aAAaC,QAAO;AAC3B,IAAAD,OAAM,wCAAwC;AAE9C,UAAM,aAAa,cAAcC,MAAK;AAEtC,QAAI,WAAW,SAAS,GAAG;AACzB,MAAAD,OAAM,2CAA2C;AACjD,aAAO,CAAC;AAAA,IACV;AAEA,UAAM,aAAa,UAAU,YAAY,kBAAkB;AAE3D,WAAO;AAEP,aAAS,cAAcC,QAAO;AAC5B,MAAAD,OAAM,qCAAqC;AAG3C,YAAME,cAAaD,OAAM,IAAI,WAAW,EAAE,OAAO,iBAAe,WAAW;AAC3E,aAAOC;AAEP,eAAS,YAAY,OAAO;AAC1B,QAAAF,OAAM,6BAA6B;AAEnC,eAAO,+BAA+B,KAAK,IAAI,gBAAgB,KAAK,IAAI;AAExE,iBAAS,gBAAgBG,QAAO;AAC9B,UAAAH,OAAM,8BAA8B;AACpC,gBAAM,CAAC,GAAG,IAAI,kBAAkBG,QAAO,GAAG;AAC1C,gBAAM,CAAC,GAAG,IAAI,kBAAkBA,QAAO,GAAG;AAE1C,gBAAM,aAAa,mBAAmB,0BAA0B,GAAG,CAAC;AACpE,gBAAM,aAAa,0BAA0B,GAAG;AAEhD,oBAAU,GAAG,KAAK,UAAU,GAAG,CAAC,MAAM,KAAK,UAAU,GAAG,CAAC,EAAE;AAC3D,iBAAO,WAAW,OAAO,UAAU;AAAA,QACrC;AAEA,iBAAS,mBAAmB,eAAe;AACzC,gBAAM,oBAAqB;AAC3B,gBAAM,kBAAkB,mBAAmB,aAAa,IAAI,OAAO,aAAa,EAAE,QAAQ,mBAAmB,OAAO,IAAI;AACxH,oBAAU,cAAc,aAAa,OAAO,eAAe,EAAE;AAC7D,iBAAO;AAAA,QACT;AAEA,iBAAS,0BAA0B,eAAe;AAChD,oBAAU,eAAe,aAAa,EAAE;AACxC,gBAAM,qBAAsB;AAC5B,iBAAO,mBAAmB,aAAa,IAAI,OAAO,aAAa,EAAE,QAAQ,oBAAoB,EAAE,IAAI;AAAA,QACrG;AAAA,MAEF;AAAA,IACF;AAAA,EACF;AACF;AAEO,gBAAS,cAAc,QAAQ;AAGpC,QAAMH,SAAQ,kBAAkB,wEAAwE;AACxG,QAAM,YAAYA,OAAM,OAAO,MAAM;AACrC,EAAAA,OAAM,iCAAiC;AAKvC,QAAM,aAAa,kBAAkB,MAAM;AAE3C,YAAU,uBAAuB,WAAW,MAAM,MAAM,KAAK,UAAU,UAAU,CAAC,EAAE;AAEpF,MAAI,WAAW,SAAS,GAAG;AACzB,IAAAA,OAAM,2CAA2C;AACjD,WAAO,CAAC;AAAA,EACV;AAEA,SAAO,UAAU,YAAY,mBAAmB;AAClD;AAKO,gBAAS,kBAAkB,QAAQ;AACxC,QAAMA,SAAQ,kBAAkB,4EAA4E;AAC5G,QAAM,YAAYA,OAAM,OAAO,MAAM;AACrC,EAAAA,OAAM,8BAA8B;AACpC,QAAM,QAAQ,qBAAqB,MAAM;AACzC,MAAI,MAAM,WAAW,GAAG;AACtB,WAAO,CAAC;AAAA,EACV;AAEA,QAAM,CAAC,EAAE,IAAI,+BAA+B,MAAM,CAAC,CAAC;AACpD,EAAAA,OAAM,aAAa,KAAK,UAAU,EAAE,CAAC,EAAE;AAGvC,QAAM,OAAO,8BAA8B,MAAM,CAAC,CAAC;AACnD,EAAAA,OAAM,GAAG,cAAc,MAAM,CAAC,CAAC,CAAC,OAAO,OAAO,OAAO,KAAK,EAAE;AAC5D,MAAI,MAAM;AACR,WAAO,CAAC,WAAW,IAAI,4BAA4B,EAAE,EAAE;AAAA,EACzD;AAEA,SAAO,CAAC,uBAAuB,EAAE,EAAE;AAErC;AAQO,gBAAS,SAAS,QAAQ;AAE/B,SAAO,wBAAwB,EAAC,QAAQ,iBAAiB,EAAC,CAAC;AAC7D;AAEO,gBAAS,eAAe,QAAQ;AACrC,QAAM,gBAAgB;AAGtB,SAAO,wBAAwB,EAAC,QAAQ,iBAAiB,IAAG,CAAC;AAC/D;AAEO,gBAAS,mBAAmB,QAAQ;AACzC,QAAM,8BAA8B;AAIpC,SAAO,wBAAwB,EAAC,QAAQ,iBAAiB,KAAK,SAAS,KAAI,CAAC;AAC9E;AAEO,gBAAS,6BAA6B,QAAQ;AACnD,QAAM,8BAA8B;AAGpC,QAAM,gBAAgB,wBAAwB,EAAC,QAAQ,iBAAiB,KAAK,SAAS,MAAM,YAAY,MAAM,kBAAkB,CAAC,EAAC,CAAC;AACnI,QAAM,GAAG,KAAK,UAAU,aAAa,CAAC,EAAE;AACxC,SAAO,EAAC,WAAW,MAAM,KAAK,aAAa,EAAE,QAAQ,GAAG,eAAe,aAAY;AACrF;AAEA,SAAS,iBAAiB,QAAQ;AAChC,QAAM,QAAQ,SAAS,QAAQ,CAAC,GAAG,CAAC;AACpC,QAAM,2BAA2B,KAAK,EAAE;AACxC,MAAI,OAAO;AACT,UAAM,SAAS,MAAM,MAAM,GAAG;AAI9B,QAAI,SAAS,KAAK,MAAM,SAAS,IAAI;AACnC,YAAM,CAAC,IAAI,IAAI,OAAO,IAAI,KAAK;AAG/B,UAAI,QAAQ,KAAK,UAAU,KAAK,QAAM,GAAG,SAAS,OAAO,GAAG,MAAM,MAAM,MAAM,CAAC,KAAK,MAAM,UAAU,uBAAuB;AACzH,eAAO;AAAA,MACT;AACA,aAAO,SAAS,QAAQ,CAAC,KAAK,GAAG,CAAC;AAAA,IACpC;AAAA,EACF;AACA,SAAO;AACT;AAEA,SAAS,QAAQ,QAAQ,iBAAiB,aAAa,OAAO;AAC5D,QAAM,QAAQ,iBAAiB,MAAM;AACrC,QAAM,mBAAmB,KAAK,EAAE;AAChC,MAAI,CAAC,mBAAmB,KAAK,GAAG;AAC9B,WAAO,CAAC;AAAA,EACV;AAEA,QAAM,YAAY,kBAAkB,KAAK;AACzC,QAAM,uBAAuB,SAAS,EAAE;AAGxC,QAAM,gBAAgB,mBAAmB,SAAS;AAGlD,MAAI,UAAU,UAAU,mBAAmB,CAAC,YAAY;AACtD,WAAO,CAAC,aAAa,gBAAgB,KAAK,GAAG,GAAG,SAAS,MAAM,WAAW,IAAI;AAAA,EAChF;AAEA,QAAM,iBAAiB,CAAC,iBAAiB,UAAU,UAAU;AAE7D,MAAI,UAAU,SAAO,GAAG;AACtB,WAAO,CAAC,IAAI,WAAW,cAAc;AAAA,EACvC;AAGA,SAAO,CAAC,aAAa,iBAAiB,CAAC,iBAAiB,KAAK,GAAG,GAAG,SAAS,GAAG,iBAAiB,MAAM,EAAE,KAAK,WAAW,cAAc;AAEtI,WAAS,kBAAkBI,QAAO;AAChC,UAAMC,aAAY,OAAOD,MAAK,EAC3B,QAAQ,qCAAqC,GAAG,EAChD,QAAQ,2BAA2B,EAAE,EACrC,QAAQ,SAAS,GAAG,EACpB,KAAK,EACL,QAAQ,mBAAmB,IAAI,EAC/B,KAAK;AAER,WAAOC;AAAA,EACT;AACF;AAEO,gBAAS,wBAAwB,EAAC,QAAQ,iBAAiB,UAAU,OAAO,aAAa,OAAO,mBAAmB,CAAC,EAAC,GAAG;AAC7H,QAAM,6CAA6C,eAAe,cAAc,OAAO,iBAAiB,UAAU,EAAE;AACpH,QAAM,CAAC,OAAO,WAAW,cAAc,IAAI,QAAQ,QAAQ,iBAAiB,UAAU;AACtF,QAAM,SAAS,KAAK,UAAU,KAAK,CAAC,gBAAgB,KAAK,UAAU,SAAS,CAAC,gBAAgB,KAAK,UAAU,cAAc,CAAC,EAAE;AAC7H,MAAI,UAAU,QAAW;AACvB,WAAO,CAAC;AAAA,EACV;AAEA,QAAM,UAAU,KAAK,EAAE;AAIvB,MAAI,UAAU,UAAU,mBAAmB,CAAC,YAAY;AACtD,WAAO,CAAC,KAAK;AAAA,EACf;AAEA,QAAM,sBAAsB,aAAa,CAAC,GAAG,kBAAkB,KAAK,IAAI;AAExE,SAAO,mBAAmB,EAAC,OAAO,gBAAgB,SAAS,YAAY,kBAAkB,oBAAmB,CAAC;AAE7G,WAAS,mBAAmB,EAAC,OAAAC,QAAO,gBAAAC,kBAAiB,OAAO,SAAAC,WAAU,OAAO,YAAAC,cAAa,OAAO,kBAAAC,oBAAmB,CAAC,EAAC,GAAG;AACvH,UAAM,oBAAoB;AAC1B,UAAM,CAAC,WAAW,IAAI,WAAW,MAAM;AACvC,QAAI,gBAAgB,QAAW;AAC7B,YAAM,gBAAgBJ,OAAM,SAAS,IAAI,GAAG,WAAW,QAAQA,MAAK,KAAK,GAAG,WAAW;AACvF,UAAIE,UAAS;AACX,cAAMG,uBAAsBF,cAAa,CAAC,GAAGC,mBAAkB,aAAa,IAAIA;AAChF,eAAO,gBAAgB,EAAC,OAAO,eAAe,gBAAgB,MAAM,YAAAD,aAAY,kBAAkBE,qBAAmB,CAAC;AAAA,MACxH;AACA,aAAOF,cAAaC,oBAAmB,CAAC,aAAa;AAAA,IACvD;AACA,WAAO,qBAAqB,EAAC,OAAAJ,QAAO,gBAAAC,iBAAgB,SAAAC,UAAS,YAAAC,aAAY,kBAAAC,kBAAgB,CAAC;AAAA,EAE5F;AAEA,WAAS,qBAAqB,EAAC,OAAAJ,QAAO,gBAAAC,kBAAiB,OAAO,SAAAC,WAAU,OAAO,YAAAC,cAAa,OAAO,kBAAAC,oBAAmB,CAAC,EAAC,GAAG;AACzH,UAAM,CAAC,cAAc,IAAI,cAAc,MAAM;AAC7C,QAAI,mBAAmB,QAAW;AAChC,YAAM,gBAAgBJ,OAAM,SAAS,IAAI,GAAG,cAAc,QAAQA,MAAK,KAAK,GAAG,cAAc;AAC7F,UAAIE,UAAS;AACX,cAAMG,uBAAsBF,cAAa,CAAC,GAAGC,mBAAkB,aAAa,IAAIA;AAChF,eAAO,gBAAgB,EAAC,OAAO,eAAe,gBAAgB,MAAM,YAAAD,aAAY,kBAAkBE,qBAAmB,CAAC;AAAA,MACxH;AACA,aAAOF,cAAaC,oBAAmB,CAAC,aAAa;AAAA,IACvD;AACA,QAAIH,mBAAkB,CAACC,UAAS;AAC9B,aAAOC,cAAaC,oBAAmB,CAAC,GAAGJ,MAAK,EAAE;AAAA,IACpD;AACA,WAAO,gBAAgB,EAAC,OAAAA,QAAO,gBAAAC,iBAAgB,YAAAE,aAAY,kBAAAC,kBAAgB,CAAC;AAAA,EAC9E;AAEA,WAAS,gBAAgB,EAAC,OAAAJ,QAAO,gBAAAC,kBAAiB,OAAO,YAAAE,cAAa,OAAO,kBAAAC,oBAAmB,CAAC,EAAC,GAAG;AACnG,UAAM,CAAC,SAAS,IAAI,QAAQ,MAAM;AAClC,QAAI,cAAc,QAAW;AAC3B,YAAM,sBAAsBJ,OAAM,SAAS,IAAI,GAAG,SAAS,QAAQA,MAAK,KAAK,GAAG,SAAS;AACzF,aAAOG,cAAa,CAAC,GAAGC,mBAAkB,mBAAmB,IAAI,CAAC,mBAAmB;AAAA,IACvF;AACA,QAAIH,iBAAgB;AAClB,aAAOE,cAAaC,oBAAmB,CAAC,GAAGJ,MAAK,EAAE;AAAA,IACpD;AACA,WAAO,CAAC;AAAA,EACV;AAEF;AAEO,gBAAS,WAAW,QAAQ;AACjC,QAAMN,SAAQ,kBAAkB,qEAAqE;AACrG,QAAM,YAAYA,OAAM,OAAO,MAAM;AACrC,EAAAA,OAAM,qCAAqC;AAG3C,QAAM,SAAS,UAAU,MAAM;AAE/B,MAAI,mBAAmB,MAAM,GAAG;AAC9B,UAAM,YAAY,OAAO,MAAM,EAC5B,QAAQ,2BAA2B,EAAE,EAErC,QAAQ,QAAQ,GAAG,EACnB,KAAK,EACL,MAAM,GAAG,EAAE,EACX,KAAK;AAGR,UAAM,gBAAgB,mBAAmB,SAAS;AAElD,cAAU,kBAAkB,SAAS,EAAE;AACvC,QAAI,UAAU,UAAU,GAAG;AACzB,aAAO,CAAC,cAAc,gBAAgB,KAAK,GAAG,GAAG,SAAS,IAAI;AAAA,IAChE;AACA,WAAO,CAAC;AAAA,EACV;AAEA,SAAO,CAAC;AAER,WAAS,UAAUY,SAAQ;AAEzB,UAAM,CAAC,KAAK,IAAIA,QAAO,IAAI,gCAAgC;AAG3D,QAAI,OAAO;AACT,YAAM,eAAe,MAAM,UACxB,OAAO,CAAC,EAAC,KAAI,MAAM,CAAC,GAAG,EAAE,SAAS,IAAI,CAAC,EACvC,IAAI,CAAC,EAAC,MAAK,MAAM,mBAAmB,KAAK,IAAI,OAAO,KAAK,IAAI,EAAE,EAC/D,OAAO,WAAS,KAAK,EAErB,KAAK,GAAG;AACX,aAAO;AAAA,IACT;AACA,WAAO;AAAA,EACT;AACF;AAEO,gBAAS,cAAc,QAAQ;AACpC,QAAMZ,SAAQ,kBAAkB,wEAAwE;AACxG,QAAM,YAAYA,OAAM,OAAO,MAAM;AACrC,EAAAA,OAAM,wCAAwC;AAG9C,QAAM,YAAY,aAAa,MAAM;AACrC,MAAI,mBAAmB,SAAS,GAAG;AACjC,UAAM,YAAY,OAAO,SAAS,EAC/B,QAAQ,2BAA2B,EAAE,EAErC,QAAQ,QAAQ,GAAG,EACnB,KAAK,EACL,MAAM,GAAG,EAAE,EACX,KAAK;AAER,cAAU,sBAAsB,SAAS,GAAG;AAC5C,QAAI,UAAU,WAAW,GAAG;AAC1B,aAAO,CAAC;AAAA,IACV;AAEA,WAAO,CAAC,iBAAiB,SAAS,GAAG;AAAA,EACvC;AAEA,SAAO,CAAC;AAER,WAAS,aAAaY,SAAQ;AAE5B,UAAM,CAAC,KAAK,IAAIA,QAAO,IAAI,WAAW,EAAE,OAAO,OAAK,EAAE,UAAU,KAAK,QAAM,GAAG,SAAS,GAAG,CAAC;AAG3F,QAAI,OAAO;AACT,YAAM,kBAAkB,MAAM,UAC3B,OAAO,CAAC,EAAC,KAAI,MAAM,CAAC,GAAG,EAAE,SAAS,IAAI,CAAC,EACvC,IAAI,CAAC,EAAC,MAAK,MAAM,mBAAmB,KAAK,IAAI,OAAO,KAAK,IAAI,EAAE,EAC/D,OAAO,WAAS,KAAK,EAErB,KAAK,GAAG,EACR,KAAK;AACR,aAAO;AAAA,IACT;AACA,WAAO;AAAA,EACT;AACF;AAEO,gBAAS,QAAQ,QAAQ;AAC9B,QAAMZ,SAAQ,kBAAkB,kEAAkE;AAClG,QAAM,YAAYA,OAAM,OAAO,MAAM;AACrC,EAAAA,OAAM,wCAAwC;AAE9C,QAAM,OAAO,QAAQ,MAAM;AAC3B,MAAI,MAAM;AACR,WAAO,CAAC,YAAY,IAAI,GAAG;AAAA,EAC7B;AACA,SAAO,CAAC;AAER,WAAS,QAAQY,SAAQ;AACvB,UAAM,CAAC,IAAI,IAAIA,QAAO,IAAI,QAAQ;AAClC,QAAI,CAAC,QAAQ,CAAC,KAAK,SAAS,KAAK,MAAM,SAAS,IAAI;AAClD,MAAAZ,OAAM,qBAAqB;AAC3B,aAAO;AAAA,IACT;AAEA,cAAU,SAAS,KAAK,UAAU,IAAI,CAAC,EAAE;AACzC,UAAM,EAAC,MAAK,IAAI;AAChB,UAAMa,QAAO,MAAM,MAAM,GAAG,EAAE;AAC9B,QAAIA,UAAS,UAAUA,UAAS,QAAQ;AACtC,MAAAb,OAAM,0BAA0B;AAChC,aAAO;AAAA,IACT;AACA,WAAOa;AAAA,EACT;AACF;AAEO,gBAAS,mBAAmB,WAAW;AAC5C,QAAM,aAAa,UAAU,YAAY;AAEzC,QAAM,oBAAoB,CAAC,QAAQ,OAAO,QAAQ,MAAM;AACxD,SAAO,kBAAkB,KAAK,UAAQ,WAAW,WAAW,IAAI,CAAC;AACnE;AAEO,gBAAS,uBAAuB,QAAQ;AAE7C,QAAMb,SAAQ,kBAAkB,iFAAiF;AACjH,QAAM,YAAYA,OAAM,OAAO,MAAM;AACrC,QAAMc,YAAWd,OAAM,OAAO,KAAK;AAEnC,EAAAA,OAAM,2CAA2C;AAKjD,QAAM,SAAS,OAAO,IAAI,gCAAgC;AAC1D,QAAM,cAAc,CAAC,EAAE,OAAO,GAAG,OAAO,IAAI,aAAa,CAAC;AAC1D,QAAM,oBAAoB,CAAC,GAAG,IAAI,IAAI,WAAW,CAAC;AAElD,YAAU,+BAA+B,KAAK,UAAU,MAAM,CAAC,EAAE;AACjE,YAAU,gBAAgB,YAAY,MAAM,MAAM,KAAK,UAAU,WAAW,CAAC,EAAE;AAC/E,YAAU,uBAAuB,kBAAkB,MAAM,MAAM,KAAK,UAAU,iBAAiB,CAAC,EAAE;AAElG,MAAI,kBAAkB,SAAS,GAAG;AAChC,IAAAA,OAAM,2CAA2C;AACjD,WAAO,CAAC;AAAA,EACV;AAEA,SAAO,UAAU,mBAAmB,eAAe;AAEnD,WAAS,cAAc,EAAC,KAAK,UAAS,GAAG;AACvC,UAAM,iBAAkB;AACxB,UAAM,gBAAiB;AAEvB,QAAI,QAAQ,OAAO;AACjB,aAAO,UACJ,OAAO,QAAM,GAAG,SAAS,OAAO,GAAG,MAAM,MAAM,iBAAiB,CAAC,EACjE,IAAI,QAAM,GAAG,KAAK;AAAA,IACvB;AAEA,QAAI,QAAQ,OAAO;AACjB,aAAO,UACJ,OAAO,SAAO,CAAC,KAAK,GAAG,EAAE,SAAS,IAAI,IAAI,KAAK,mBAAmB,IAAI,KAAK,KAAK,eAAe,KAAK,OAAO,IAAI,KAAK,CAAC,CAAC,EACtH,IAAI,CAAC,EAAC,MAAK,MAAM,OAAO,KAAK,CAAC;AAAA,IACnC;AAEA,QAAI,QAAQ,OAAO;AACjB,aAAO,UACJ,OAAO,SAAO,CAAC,KAAK,KAAK,GAAG,EAAE,SAAS,IAAI,IAAI,KAAK,mBAAmB,IAAI,KAAK,KAAK,eAAe,KAAK,OAAO,IAAI,KAAK,CAAC,CAAC,EAC3H,IAAI,CAAC,EAAC,MAAK,MAAM,OAAO,KAAK,CAAC;AAAA,IACnC;AAEA,QAAI,QAAQ,OAAO;AAmDjB,UAAS,kBAAT,SAAyB,KAAK;AAC5B,YAAI,QAAQ,QAAW;AACvB,gBAAM,WAAW,IAAI,QAAQ,cAAc,EAAE,EAAE,QAAQ,QAAQ,GAAG,EAAE,QAAQ,SAAS,EAAE,EAAE,QAAQ,SAAS,EAAE;AAC5G,iBAAO,UAAU,CAAC,KAAK,QAAQ,CAAC;AAAA,QAChC;AACA,eAAO,CAAC;AAAA,MACV,GAES,yBAAT,SAAgC,KAAK;AAEnC,cAAM,qBAAqB;AAC3B,cAAM,UAAU,IAAI,QAAQ,oBAAoB,GAAG,EAAE,QAAQ,QAAQ,GAAG,EAAE,QAAQ,SAAS,EAAE,EAAE,QAAQ,SAAS,EAAE,EAAE,YAAY;AAEhI,eAAO,YAAY,MAAM,UAAU;AAAA,MACrC;AAhEA,YAAM,IAAI,UAAU,KAAK,QAAM,GAAG,SAAS,GAAG;AAC9C,YAAM,IAAI,UAAU,KAAK,QAAM,GAAG,SAAS,GAAG;AAG9C,YAAM,UAAU,gBAAgB,GAAG,KAAK;AACxC,YAAM,UAAU,gBAAgB,GAAG,KAAK;AACxC,MAAAc,UAAS,KAAK,UAAU,OAAO,CAAC;AAChC,MAAAA,UAAS,KAAK,UAAU,OAAO,CAAC;AAGhC,YAAM,UAAU,QAAQ,IAAI,UAAQ,KAAK,MAAM,GAAG,CAAC,EAAE,OAAO,OAAO,EAAE,KAAK;AAC1E,YAAM,UAAU,QAAQ,IAAI,UAAQ,KAAK,MAAM,GAAG,CAAC,EAAE,OAAO,OAAO,EAAE,KAAK;AAC1E,MAAAA,UAAS,KAAK,UAAU,OAAO,CAAC;AAChC,MAAAA,UAAS,KAAK,UAAU,OAAO,CAAC;AAEhC,YAAM,SAAS,QAAQ,IAAI,UAAQ,uBAAuB,IAAI,CAAC;AAC/D,YAAM,SAAS,QAAQ,IAAI,UAAQ,uBAAuB,IAAI,CAAC;AAC/D,MAAAA,UAAS,KAAK,UAAU,MAAM,CAAC;AAC/B,MAAAA,UAAS,KAAK,UAAU,MAAM,CAAC;AAE/B,YAAMC,eAAc,OAAO,IAAI,CAAC,OAAO,UAAU;AAC/C,QAAAD,UAAS,YAAY,KAAK,gBAAgB,KAAK,EAAE;AAEjD,YAAI,MAAM,SAAS,KAAK,OAAO,KAAK,GAAG,SAAS,GAAG;AACjD,UAAAA,UAAS,YAAY,KAAK,UAAU,OAAO,KAAK,CAAC,GAAG;AAEpD,iBAAO,CAAC,IAAI,OAAO,KAAK,CAAC,IAAI,KAAK,IAAI,IAAI,KAAK,IAAI,OAAO,KAAK,CAAC,IAAI,IAAI,KAAK,EAAE;AAAA,QAGjF;AAGA,YAAI,MAAM,SAAS,KAAK,OAAO,CAAC,GAAG,SAAS,GAAG;AAC7C,UAAAA,UAAS,YAAY,KAAK,UAAU,OAAO,CAAC,CAAC,GAAG;AAEhD,iBAAO,CAAC,IAAI,OAAO,CAAC,CAAC,IAAI,KAAK,IAAI,IAAI,KAAK,IAAI,OAAO,CAAC,CAAC,IAAI,IAAI,KAAK,EAAE;AAAA,QAGzE;AAEA,YAAI,MAAM,SAAS,GAAG;AACpB,UAAAA,UAAS,YAAY,KAAK,cAAc,OAAO,KAAK,CAAC,EAAE;AACvD,iBAAO,CAAC,IAAI,KAAK,EAAE;AAAA,QACrB;AACA,eAAO,CAAC;AAAA,MACV,CAAC;AAED,MAAAA,UAAS,kCAAkC,KAAK,UAAUC,YAAW,CAAC,EAAE;AACxE,aAAO,UAAUA,aAAY,KAAK,CAAC;AAAA,IAkBrC;AAGA,WAAO,UACJ,OAAO,SAAO,CAAC,KAAK,GAAG,EAAE,SAAS,IAAI,IAAI,KAAK,mBAAmB,IAAI,KAAK,KAAK,cAAc,KAAK,OAAO,IAAI,KAAK,CAAC,CAAC,EACrH,IAAI,CAAC,EAAC,MAAK,MAAM,OAAO,KAAK,CAAC;AAAA,EACnC;AACF;",
+  "names": ["debug", "fSids", "sidStrings", "field", "title", "formatted", "query", "queryIsOkAlone", "addYear", "alternates", "alternateQueries", "newAlternateQueries", "record", "year", "debugDev", "identifiers"]
 }

package/dist/match-detection/features/bib/publisherNumber.js

@@ -31,7 +31,7 @@ export default () => ({
         return fieldsToPublisherNumber(remainingFields, [...result, baval, aval]);
       }
       function normalizePublisherNumber(val) {
-        return val.replace(/[- .,:;]/ug, "");
+        return val.replace(/[- .,:;]/ug, "").toLowerCase();
       }
     }
   },

package/dist/match-detection/features/bib/publisherNumber.js.map

@@ -1,7 +1,7 @@
 {
   "version": 3,
   "sources": ["../../../../src/match-detection/features/bib/publisherNumber.js"],
-  "sourcesContent": ["// Compare publusher number (028$b$a or 028$a)\nimport {uniqArray} from '@natlibfi/marc-record-validators-melinda/dist/utils.js';\nimport createDebugLogger from 'debug';\n\nconst debug = createDebugLogger('@natlibfi/melinda-record-matching:match-detection:features:publisher-number');\nconst debugData = debug.extend('data');\n\nexport default () => ({\n  name: 'Publisher or Distributor Number',\n  extract: ({record/*, recordExternal*/}) => {\n    //const label = recordExternal && recordExternal.label ? recordExternal.label : 'record';\n\n    return getIdentifiers();\n\n    function getIdentifiers() {\n      const values = fieldsToPublisherNumber(record.get(/^028$/u));\n\n      debug(`\\t ${values.length} potential publisher identifiers: ${values.join(', ')}`);\n\n      return uniqArray(values);\n\n      function fieldsToPublisherNumber(fields, result = []) {\n        const [field, ...remainingFields] = fields;\n        if (!field) {\n          return result;\n        }\n        const a = field.subfields.find(sf => sf.code === 'a');\n        // NB! Publisher number must contain a digit! :-)\n        if (!a || !a.value || !a.value.match(/[0-9]/u)) {\n          return fieldsToPublisherNumber(remainingFields, result);\n        }\n        const aval = normalizePublisherNumber(`${a.value}`);\n\n        const b = field.subfields.find(sf => sf.code === 'b');\n        if (!b || !b.value) {\n          return fieldsToPublisherNumber(remainingFields, [...result, aval]);\n        }\n        const baval = normalizePublisherNumber(`${b.value}${a.value}`);\n        return fieldsToPublisherNumber(remainingFields, [...result, baval, aval]);\n      }\n\n      function normalizePublisherNumber(val) {\n        return val.replace(/[- .,:;]/ug, '');\n      }\n    }\n  },\n  compare: (aa, bb) => {\n    debugData(`Comparing identifier sets ${JSON.stringify(aa)} and ${JSON.stringify(bb)}`);\n    if (aa.length === 0 || bb.length === 0) {\n      // No data for decision\n      return 0;\n    }\n    const sharedIdentifiers = aa.filter(val => bb.includes(val));\n    if (sharedIdentifiers.length > 0) {\n      const goodMatch = sharedIdentifiers.find(identifier => identifier.match(/[^0-9]/u));\n      if (goodMatch) {\n        debug(`\\tShared identifier found: '${goodMatch}'`);\n        return 0.5;\n      }\n      debug(`\\tShared identifier (numbers only) found: '${sharedIdentifiers[0]}'`);\n      return 0.2;\n    }\n\n    const aa2 = aa.map(a => a.replace(/[^0-9]/ug, ''));\n    const bb2 = bb.map(b => b.replace(/[^0-9]/ug, ''));\n\n    // Numbers matches (aka don't penalize \"Fazer F-123\" vs \"F-123\")\n    // Note that FOOLP-123 vs FOOCD-123 will now also return 0.0. Still nothing positive is returned, so no damage done.\n    if (aa2.find(val => val !== '' && bb2.includes(val))) {\n      return 0.0;\n    }\n\n    return -0.2;\n\n  }\n});\n\n"],
-  "mappings": "AACA,SAAQ,iBAAgB;AACxB,OAAO,uBAAuB;AAE9B,MAAM,QAAQ,kBAAkB,6EAA6E;AAC7G,MAAM,YAAY,MAAM,OAAO,MAAM;AAErC,eAAe,OAAO;AAAA,EACpB,MAAM;AAAA,EACN,SAAS,CAAC;AAAA,IAAC;AAAA;AAAA,EAA0B,MAAM;AAGzC,WAAO,eAAe;AAEtB,aAAS,iBAAiB;AACxB,YAAM,SAAS,wBAAwB,OAAO,IAAI,QAAQ,CAAC;AAE3D,YAAM,KAAM,OAAO,MAAM,qCAAqC,OAAO,KAAK,IAAI,CAAC,EAAE;AAEjF,aAAO,UAAU,MAAM;AAEvB,eAAS,wBAAwB,QAAQ,SAAS,CAAC,GAAG;AACpD,cAAM,CAAC,OAAO,GAAG,eAAe,IAAI;AACpC,YAAI,CAAC,OAAO;AACV,iBAAO;AAAA,QACT;AACA,cAAM,IAAI,MAAM,UAAU,KAAK,QAAM,GAAG,SAAS,GAAG;AAEpD,YAAI,CAAC,KAAK,CAAC,EAAE,SAAS,CAAC,EAAE,MAAM,MAAM,QAAQ,GAAG;AAC9C,iBAAO,wBAAwB,iBAAiB,MAAM;AAAA,QACxD;AACA,cAAM,OAAO,yBAAyB,GAAG,EAAE,KAAK,EAAE;AAElD,cAAM,IAAI,MAAM,UAAU,KAAK,QAAM,GAAG,SAAS,GAAG;AACpD,YAAI,CAAC,KAAK,CAAC,EAAE,OAAO;AAClB,iBAAO,wBAAwB,iBAAiB,CAAC,GAAG,QAAQ,IAAI,CAAC;AAAA,QACnE;AACA,cAAM,QAAQ,yBAAyB,GAAG,EAAE,KAAK,GAAG,EAAE,KAAK,EAAE;AAC7D,eAAO,wBAAwB,iBAAiB,CAAC,GAAG,QAAQ,OAAO,IAAI,CAAC;AAAA,MAC1E;AAEA,eAAS,yBAAyB,KAAK;AACrC,eAAO,IAAI,QAAQ,cAAc,EAAE;AAAA,MACrC;AAAA,IACF;AAAA,EACF;AAAA,EACA,SAAS,CAAC,IAAI,OAAO;AACnB,cAAU,6BAA6B,KAAK,UAAU,EAAE,CAAC,QAAQ,KAAK,UAAU,EAAE,CAAC,EAAE;AACrF,QAAI,GAAG,WAAW,KAAK,GAAG,WAAW,GAAG;AAEtC,aAAO;AAAA,IACT;AACA,UAAM,oBAAoB,GAAG,OAAO,SAAO,GAAG,SAAS,GAAG,CAAC;AAC3D,QAAI,kBAAkB,SAAS,GAAG;AAChC,YAAM,YAAY,kBAAkB,KAAK,gBAAc,WAAW,MAAM,SAAS,CAAC;AAClF,UAAI,WAAW;AACb,cAAM,8BAA+B,SAAS,GAAG;AACjD,eAAO;AAAA,MACT;AACA,YAAM,6CAA8C,kBAAkB,CAAC,CAAC,GAAG;AAC3E,aAAO;AAAA,IACT;AAEA,UAAM,MAAM,GAAG,IAAI,OAAK,EAAE,QAAQ,YAAY,EAAE,CAAC;AACjD,UAAM,MAAM,GAAG,IAAI,OAAK,EAAE,QAAQ,YAAY,EAAE,CAAC;AAIjD,QAAI,IAAI,KAAK,SAAO,QAAQ,MAAM,IAAI,SAAS,GAAG,CAAC,GAAG;AACpD,aAAO;AAAA,IACT;AAEA,WAAO;AAAA,EAET;AACF;",
+  "sourcesContent": ["// Compare publusher number (028$b$a or 028$a)\nimport {uniqArray} from '@natlibfi/marc-record-validators-melinda/dist/utils.js';\nimport createDebugLogger from 'debug';\n\nconst debug = createDebugLogger('@natlibfi/melinda-record-matching:match-detection:features:publisher-number');\nconst debugData = debug.extend('data');\n\nexport default () => ({\n  name: 'Publisher or Distributor Number',\n  extract: ({record/*, recordExternal*/}) => {\n    //const label = recordExternal && recordExternal.label ? recordExternal.label : 'record';\n\n    return getIdentifiers();\n\n    function getIdentifiers() {\n      const values = fieldsToPublisherNumber(record.get(/^028$/u));\n\n      debug(`\\t ${values.length} potential publisher identifiers: ${values.join(', ')}`);\n\n      return uniqArray(values);\n\n      function fieldsToPublisherNumber(fields, result = []) {\n        const [field, ...remainingFields] = fields;\n        if (!field) {\n          return result;\n        }\n        const a = field.subfields.find(sf => sf.code === 'a');\n        // NB! Publisher number must contain a digit! :-)\n        // NOTE: we might have some rare cases with publisher numbers that do not have numbers...\n        if (!a || !a.value || !a.value.match(/[0-9]/u)) {\n          return fieldsToPublisherNumber(remainingFields, result);\n        }\n        const aval = normalizePublisherNumber(`${a.value}`);\n\n        const b = field.subfields.find(sf => sf.code === 'b');\n        if (!b || !b.value) {\n          return fieldsToPublisherNumber(remainingFields, [...result, aval]);\n        }\n        const baval = normalizePublisherNumber(`${b.value}${a.value}`);\n        return fieldsToPublisherNumber(remainingFields, [...result, baval, aval]);\n      }\n\n      function normalizePublisherNumber(val) {\n        return val.replace(/[- .,:;]/ug, '').toLowerCase();\n      }\n    }\n  },\n  compare: (aa, bb) => {\n    debugData(`Comparing identifier sets ${JSON.stringify(aa)} and ${JSON.stringify(bb)}`);\n    if (aa.length === 0 || bb.length === 0) {\n      // No data for decision\n      return 0;\n    }\n    const sharedIdentifiers = aa.filter(val => bb.includes(val));\n    if (sharedIdentifiers.length > 0) {\n      const goodMatch = sharedIdentifiers.find(identifier => identifier.match(/[^0-9]/u));\n      if (goodMatch) {\n        debug(`\\tShared identifier found: '${goodMatch}'`);\n        return 0.5;\n      }\n      debug(`\\tShared identifier (numbers only) found: '${sharedIdentifiers[0]}'`);\n      return 0.2;\n    }\n\n    const aa2 = aa.map(a => a.replace(/[^0-9]/ug, ''));\n    const bb2 = bb.map(b => b.replace(/[^0-9]/ug, ''));\n\n    // Numbers matches (aka don't penalize \"Fazer F-123\" vs \"F-123\")\n    // Note that FOOLP-123 vs FOOCD-123 will now also return 0.0. Still nothing positive is returned, so no damage done.\n    if (aa2.find(val => val !== '' && bb2.includes(val))) {\n      return 0.0;\n    }\n\n    return -0.2;\n\n  }\n});\n\n"],
+  "mappings": "AACA,SAAQ,iBAAgB;AACxB,OAAO,uBAAuB;AAE9B,MAAM,QAAQ,kBAAkB,6EAA6E;AAC7G,MAAM,YAAY,MAAM,OAAO,MAAM;AAErC,eAAe,OAAO;AAAA,EACpB,MAAM;AAAA,EACN,SAAS,CAAC;AAAA,IAAC;AAAA;AAAA,EAA0B,MAAM;AAGzC,WAAO,eAAe;AAEtB,aAAS,iBAAiB;AACxB,YAAM,SAAS,wBAAwB,OAAO,IAAI,QAAQ,CAAC;AAE3D,YAAM,KAAM,OAAO,MAAM,qCAAqC,OAAO,KAAK,IAAI,CAAC,EAAE;AAEjF,aAAO,UAAU,MAAM;AAEvB,eAAS,wBAAwB,QAAQ,SAAS,CAAC,GAAG;AACpD,cAAM,CAAC,OAAO,GAAG,eAAe,IAAI;AACpC,YAAI,CAAC,OAAO;AACV,iBAAO;AAAA,QACT;AACA,cAAM,IAAI,MAAM,UAAU,KAAK,QAAM,GAAG,SAAS,GAAG;AAGpD,YAAI,CAAC,KAAK,CAAC,EAAE,SAAS,CAAC,EAAE,MAAM,MAAM,QAAQ,GAAG;AAC9C,iBAAO,wBAAwB,iBAAiB,MAAM;AAAA,QACxD;AACA,cAAM,OAAO,yBAAyB,GAAG,EAAE,KAAK,EAAE;AAElD,cAAM,IAAI,MAAM,UAAU,KAAK,QAAM,GAAG,SAAS,GAAG;AACpD,YAAI,CAAC,KAAK,CAAC,EAAE,OAAO;AAClB,iBAAO,wBAAwB,iBAAiB,CAAC,GAAG,QAAQ,IAAI,CAAC;AAAA,QACnE;AACA,cAAM,QAAQ,yBAAyB,GAAG,EAAE,KAAK,GAAG,EAAE,KAAK,EAAE;AAC7D,eAAO,wBAAwB,iBAAiB,CAAC,GAAG,QAAQ,OAAO,IAAI,CAAC;AAAA,MAC1E;AAEA,eAAS,yBAAyB,KAAK;AACrC,eAAO,IAAI,QAAQ,cAAc,EAAE,EAAE,YAAY;AAAA,MACnD;AAAA,IACF;AAAA,EACF;AAAA,EACA,SAAS,CAAC,IAAI,OAAO;AACnB,cAAU,6BAA6B,KAAK,UAAU,EAAE,CAAC,QAAQ,KAAK,UAAU,EAAE,CAAC,EAAE;AACrF,QAAI,GAAG,WAAW,KAAK,GAAG,WAAW,GAAG;AAEtC,aAAO;AAAA,IACT;AACA,UAAM,oBAAoB,GAAG,OAAO,SAAO,GAAG,SAAS,GAAG,CAAC;AAC3D,QAAI,kBAAkB,SAAS,GAAG;AAChC,YAAM,YAAY,kBAAkB,KAAK,gBAAc,WAAW,MAAM,SAAS,CAAC;AAClF,UAAI,WAAW;AACb,cAAM,8BAA+B,SAAS,GAAG;AACjD,eAAO;AAAA,MACT;AACA,YAAM,6CAA8C,kBAAkB,CAAC,CAAC,GAAG;AAC3E,aAAO;AAAA,IACT;AAEA,UAAM,MAAM,GAAG,IAAI,OAAK,EAAE,QAAQ,YAAY,EAAE,CAAC;AACjD,UAAM,MAAM,GAAG,IAAI,OAAK,EAAE,QAAQ,YAAY,EAAE,CAAC;AAIjD,QAAI,IAAI,KAAK,SAAO,QAAQ,MAAM,IAAI,SAAS,GAAG,CAAC,GAAG;AACpD,aAAO;AAAA,IACT;AAEA,WAAO;AAAA,EAET;AACF;",
   "names": []
 }

package/dist/match-detection/features/bib/title.js

@@ -5,6 +5,7 @@ import { testStringOrNumber } from "../../../matching-utils.js";
 import { subfieldToString } from "@natlibfi/marc-record-validators-melinda/dist/utils.js";
 const debug = createDebugLogger("@natlibfi/melinda-record-matching:match-detection:features/bib/title");
 const debugData = debug.extend("data");
+const debugDev = debug.extend("dev");
 const TITLE_MAX = 0.5;
 export default ({ threshold = 0.9 } = {}) => ({
   name: "Title",
@@ -172,12 +173,19 @@ export default ({ threshold = 0.9 } = {}) => ({
 export function getTitle(record, targetSubfieldCodes = ["a", "b", "n", "p"]) {
   const [field] = record.get(/^245$/u);
   debugData(`titleField: ${JSON.stringify(field)}`);
+  debugDev(`targetSubfieldCodes: ${JSON.stringify(targetSubfieldCodes)}`);
   if (field) {
-    const title = field.subfields.filter(({ code }) => targetSubfieldCodes.includes(code)).map(({ value }) => testStringOrNumber(value) ? String(value) : "").join(" ").replace(/\[[^\]]*\]/ug, " ").replace(/ [=\/:](?:$| )/ug, " ").replace(/ +/ug, " ").replace(/^ +/u, "").replace(/ +$/u, "");
+    const title = field.subfields.filter(({ code }) => targetSubfieldCodes.includes(code)).map(({ value }) => testStringOrNumber(value) ? String(value) : "").join(" ").replace(/ [=\/:](?:$| )/ug, " ").replace(/ +/ug, " ").replace(/^ +/u, "").replace(/ +$/u, "");
+    debugDev(`getTitle: title: ${title}`);
+    const cleanTitle = title.replace(/\[[^\]]*\]/ug, " ").replace(/ +/ug, " ").replace(/^ +/u, "").replace(/ +$/u, "");
+    debugDev(`getTitle: cleanTitle: ${cleanTitle} (${cleanTitle.length})`);
+    const usableTitle = cleanTitle.length > 0 ? cleanTitle : title;
+    debugDev(`getTitle: usableTitle: ${usableTitle} (${usableTitle.length})`);
     if (/^[1-8]$/u.test(field.ind2)) {
-      return title.slice(parseInt(field.ind2));
+      const nonFilingTitle = usableTitle.slice(parseInt(field.ind2));
+      if (nonFilingTitle.length > 0) return nonFilingTitle;
     }
-    return title;
+    return usableTitle;
   }
   return false;
 }

package/dist/match-detection/features/bib/title.js.map

@@ -1,7 +1,7 @@
 {
   "version": 3,
   "sources": ["../../../../src/match-detection/features/bib/title.js"],
-  "sourcesContent": ["import createDebugLogger from 'debug';\nimport naturalPkg from 'natural';\nconst {LevenshteinDistance: leven} = naturalPkg;\nimport {testStringOrNumber} from '../../../matching-utils.js';\nimport { subfieldToString } from '@natlibfi/marc-record-validators-melinda/dist/utils.js';\n\n\nconst debug = createDebugLogger('@natlibfi/melinda-record-matching:match-detection:features/bib/title');\nconst debugData = debug.extend('data');\n\nconst TITLE_MAX = 0.5;\nexport default ({threshold = 0.9} = {}) => ({\n  name: 'Title',\n  extract: ({record, recordExternal}) => {\n    const label = recordExternal && recordExternal.label ? recordExternal.label : 'record';\n    const a = getTitle(record, ['a']);\n    debug(`${label}: title: ${a}`);\n\n    if (a) {\n      const b = normalizeTitle(getTitle(record, ['b'])) || '';\n      const n = normalizeTitle(getTitle(record, ['n'])) || '';\n      const p = normalizeTitle(getTitle(record, ['p'])) || '';\n      return [normalizeTitle(a), b, n, p];\n    }\n\n    return ['', '', '', ''];\n\n    function normalizeTitle(title) {\n      return title\n        // decompose unicode diacritics\n        .normalize('NFD')\n        .replace(/\\b(?:ein|ett|I|one|uno|yksi)\\b/uig, '1')\n        .replace(/\\b(?:dos|kaksi|II|tv\u00E5|two|zwei)\\b/uig, '2')\n        .replace(/\\b(?:drei|III|kolme|three|tre|tres)\\b/uig, '3')\n        .replace(/\\b(?:four|fyra|IV|nelj\u00E4|quat+ro|vier)\\b/uig, '4')\n        .replace(/\\b(?:five|f\u00FCnf|fyra|viisi|V)\\b/uig, '5')\n        .replace(/\\b(?:kuusi|sechts|sex|six|VI)\\b/uig, '6')\n        .replace(/\\b(and|ja|och|und)\\b/uig, '&')\n        // strip non-letters/numbers\n        // - note: combined with decomposing unicode diacritics this normalizes both 'saa' and 's\u00E4\u00E4' as 'saa'\n        // - we could precompose the Finnish letters back to avoid this\n        // - see validator normalize-utf8-diacritics for details\n        .replace(/[^\\p{Letter}\\p{Number}& ]/gu, '') // 2026-07-01: keep ' '. We need it for splitToBaseAndNumber()\n        .replace(/\\s+/ug, ' ')\n        .replace(/ $/u, '')\n        .replace(/^ /u, '')\n        .toLowerCase();\n    }\n\n  },\n  compare: (origA, origB) => {\n    return compare2(origA, origB);\n\n    function compare2(a, b) { // Added this wrapper function as I had issues with recursion\n\n      const [aa, ab, an, ap] = a;\n      const [ba, bb, bn, bp] = b;\n      debugData(`COMPARE:\\n['${aa}', '${ab}', '${an}', '${ap}'] VS\\n['${ba}', '${bb}', '${bn}', '${bp}']`);\n\n      const aFull = toFullTitle(a);\n      const bFull = toFullTitle(b);\n\n      // debugData(`JOINED COMPARE:\\n  '${aFull}' vs  \\n  '${bFull}'`);\n\n      if (isEmpty(aa) || isEmpty(ba)) {\n        return -1.0;\n      }\n\n      if (aFull === bFull) {\n        return TITLE_MAX;\n      }\n\n      // MELKEHITYS-3496-ish: move part of $a to $n:\n      const [altAa, altAn, altBa, altBn] = reconstructTitleAndNumber(aa, an, ba, bn);\n      if (altAa) {\n        return compare2([altAa, ab, altAn, ap], [altBa, bb, altBn, bp]);\n      }\n\n      if (ab && ab !== '' && ab === bb) { // Remove $b from equation (MELKEHITYS-3494)\n        debug(`Ignore \\$b ${ab}`);\n        return compare2([aa, '', an, ap], [ba, '', bn, bp]);\n      }\n      if (an && an !== '' && an === bn) { // Remove $n from equation\n        debug(`Ignore \\$n ${an}`);\n        return compare2([aa, ab, '', ap], [ba, bb, '', bp]);\n      }\n      if (ap && ap !== '' && ap === bp) { // Remove $p from equation\n        debug(`Ignore \\$p ${ap}`);\n        return compare2([aa, ab, an, ''], [ba, bb, bn, '']);\n      }\n      // There seems to be wobble between $b and $p:\n      if (!isEmpty(ab) && ab === bp) {\n        return compare2([aa, '', an, ap], [ba, bb, bn, '']);\n      }\n      if (!isEmpty(ap) && ap === bp) {\n        return compare2([aa, ab, an, ''], [ba, '', bn, bp]);\n      }\n\n      if (an && bn && an.match(/[0-9]/u)) {\n        debug(`NUMBER FOUND. Compare ${an} and ${bn}`);\n        const atmp = an.replace(/[^0-9]/ug, '');\n        const btmp = bn.replace(/[^0-9]/ug, '');\n        if (atmp === btmp) {\n          debug(`Ignore \\$n '${an}' vs '${bn}'`);\n          return compare2([aa, ab, '', ap], [ba, bb, '', bp]);\n        }\n      }\n\n\n      // f245$n information is critical; it can not mismatch at all  (exceptions have already been handled above):\n      if (an && an !== bn) {\n        return -1.0;\n      }\n\n      const [distance, maxLength, correctness] = doLevenshtein(aFull.replace(/ /ug, ''), bFull.replace(/ /ug, ''));\n\n      debug(`'${aFull}' vs '${bFull}': Max length = ${maxLength}, distance = ${distance}, correctness = ${correctness}`);\n\n\n      if (correctness >= threshold) {\n        return TITLE_MAX * 0.8;\n      }\n\n      // Subset removal (MELKEHITYS-3498-ish)\n      if (ab && bb) { // At this point ab and bb are never equal...\n        // If X is a subset of Y, then remove X and shorten Y (= remove the shared content)\n        if (ab.indexOf(bb) === 0) {\n          return compare2([aa, ab.substring(bb.length), an, ap], [ba, '', bn, bp]);\n        }\n        if (bb.indexOf(ab) === 0) {\n          return compare2([aa, '', an, ap], [ba, bb.substring(ab.length), bn, bp]);\n        }\n      }\n\n      // Try the same without $p:\n      if (localXor(ap, bp)) {\n        const result = compare2([aa, ab, an, ''], [ba, bb, bn, '']);\n        return result > 0.0 ? result * 0.8 : result;\n      }\n\n      if (aa === ba && an === bn) {\n        const candScore = TITLE_MAX/2;\n        // $b is missing from one:\n        if (ap === bp && localXor(ab, bb)) {\n          debug(`Handle omitted \\$b using x ${candScore}`);\n          return candScore;\n        }\n        if (ab === bb && localXor(ap, bp)) {\n          debug(`Handle omitted \\$p using x ${candScore}`);\n          return candScore;\n        }\n      }\n\n      if (an === bn && ap === bp ) {\n        // $b-less $a is other record's $b, see MELKEHITYS-3485 for motivation (though we skip 246 and 490 here):\n        const candScore = TITLE_MAX/2;\n        if (aa === bb && !ab) {\n          return candScore;\n        }\n        if (ba === ab && !bb) {\n          return candScore;\n        }\n      }\n\n      return -0.5; // Not likely\n    }\n\n    function isEmpty(x) {\n      return !x || x.length === 0;\n    }\n\n    function localXor(x, y) {\n      if (isEmpty(x)) {\n        return !isEmpty(y);\n      }\n      // 'x' exists, thus 'y' can not exist:\n      return isEmpty(y);\n    }\n\n    function doLevenshtein(string1, string2) {\n      const distance = leven(string1, string2);\n      const len = getMaxLength(string1, string2);\n      const correctness = 1.0 - (distance / len);\n      return [distance, len, correctness];\n    }\n\n    function toFullTitle(arr) {\n      const relevant = arr.filter(val => typeof val === 'string' && val.length);\n      // No longer add space between subfields. Due to heavy normalization there are no spaces left in array elements either!\n      return relevant.join('');\n    }\n\n    function getMaxLength(str1, str2) {\n      return str1.length > str2.length ? str1.length : str2.length;\n    }\n\n    function reconstructTitleAndNumber(a1, n1, a2, n2) {\n      if (!n1 && n2) {\n        const [altA1, altN1] = splitToBaseAndNumber(a1);\n        if (altN1) {\n          debugData(`A: Reconstruct:\\n $a '${a1}' =>\\n $a '${altA1}' +\\n $n '${altN1}`);\n          return [altA1, altN1, a2, n2];\n        }\n      }\n      if (n1 && !n2) {\n        const [altA2, altN2] = splitToBaseAndNumber(a2);\n        if (altN2) {\n          debugData(`B: Reconstruct $a '${a2}' as $a '${altA2}' and $n '${altN2}`);\n          return [a1, n1, altA2, altN2];\n        }\n\n      }\n      return [null, null, null, null]; // Fail\n    }\n\n\n    function splitToBaseAndNumber(val) {\n      const words = val.split(' ');\n      const len = words.length;\n      if (len < 2 || !words[len-1].match(/^[1-9][0-9]*$/)) {\n        return [val, null];\n      }\n      if (len > 2 && words[len-2].match(/^(del|osa|vol|volume)$/u)) { // NB! \"vol.\" has lost it's '.' during normalization!\n        const number = `${words[len-2]} ${words[len-1]}`;\n        return [words.slice(0, len-2).join(' '), number];\n      }\n\n      const number = words.pop();\n      return [words.join(' '), number];\n    }\n\n  }\n});\n\nexport function getTitle(record, targetSubfieldCodes = ['a', 'b', 'n', 'p']) {\n  const [field] = record.get(/^245$/u);\n  debugData(`titleField: ${JSON.stringify(field)}`);\n\n  if (field) {\n    const title = field.subfields\n      // get also $n:s and $p:s here\n      .filter(({code}) => targetSubfieldCodes.includes(code))\n      // Would be nice to normalize $n values...\n      .map(({value}) => testStringOrNumber(value) ? String(value) : '')\n      .join(' ')\n      .replace(/\\[[^\\]]*\\]/ug, ' ') // Remove [whatever] stuff. ADD TEST FOR THIS\n      .replace(/ [=\\/:](?:$| )/ug, ' ') // Strip punctuation\n      // Also \u0111 vs d pairs seen:\n      //.replace(/(?:\u0111|\u0227|t\u0304|k\u030C|\u01E7|s\u0306|c\u0306)/ug, '') // Hack. Saamelaisbibliografia has often dropped this wierd characters (oft old articles, no longer used in sami either)\n      // trim:\n      .replace(/ +/ug, ' ')\n      .replace(/^ +/u, '')\n      .replace(/ +$/u, '');\n\n    // Skip non-filing indicator (note that '9' is a magic indicator value, so we don't do it):\n    if (/^[1-8]$/u.test(field.ind2)) { // Skip non-filing characters\n      return title.slice(parseInt(field.ind2));\n    }\n    return title;\n\n  }\n  return false;\n}"],
-  "mappings": "AAAA,OAAO,uBAAuB;AAC9B,OAAO,gBAAgB;AACvB,MAAM,EAAC,qBAAqB,MAAK,IAAI;AACrC,SAAQ,0BAAyB;AACjC,SAAS,wBAAwB;AAGjC,MAAM,QAAQ,kBAAkB,sEAAsE;AACtG,MAAM,YAAY,MAAM,OAAO,MAAM;AAErC,MAAM,YAAY;AAClB,eAAe,CAAC,EAAC,YAAY,IAAG,IAAI,CAAC,OAAO;AAAA,EAC1C,MAAM;AAAA,EACN,SAAS,CAAC,EAAC,QAAQ,eAAc,MAAM;AACrC,UAAM,QAAQ,kBAAkB,eAAe,QAAQ,eAAe,QAAQ;AAC9E,UAAM,IAAI,SAAS,QAAQ,CAAC,GAAG,CAAC;AAChC,UAAM,GAAG,KAAK,YAAY,CAAC,EAAE;AAE7B,QAAI,GAAG;AACL,YAAM,IAAI,eAAe,SAAS,QAAQ,CAAC,GAAG,CAAC,CAAC,KAAK;AACrD,YAAM,IAAI,eAAe,SAAS,QAAQ,CAAC,GAAG,CAAC,CAAC,KAAK;AACrD,YAAM,IAAI,eAAe,SAAS,QAAQ,CAAC,GAAG,CAAC,CAAC,KAAK;AACrD,aAAO,CAAC,eAAe,CAAC,GAAG,GAAG,GAAG,CAAC;AAAA,IACpC;AAEA,WAAO,CAAC,IAAI,IAAI,IAAI,EAAE;AAEtB,aAAS,eAAe,OAAO;AAC7B,aAAO,MAEJ,UAAU,KAAK,EACf,QAAQ,qCAAqC,GAAG,EAChD,QAAQ,wCAAwC,GAAG,EACnD,QAAQ,4CAA4C,GAAG,EACvD,QAAQ,8CAA8C,GAAG,EACzD,QAAQ,qCAAqC,GAAG,EAChD,QAAQ,sCAAsC,GAAG,EACjD,QAAQ,2BAA2B,GAAG,EAKtC,QAAQ,+BAA+B,EAAE,EACzC,QAAQ,SAAS,GAAG,EACpB,QAAQ,OAAO,EAAE,EACjB,QAAQ,OAAO,EAAE,EACjB,YAAY;AAAA,IACjB;AAAA,EAEF;AAAA,EACA,SAAS,CAAC,OAAO,UAAU;AACzB,WAAO,SAAS,OAAO,KAAK;AAE5B,aAAS,SAAS,GAAG,GAAG;AAEtB,YAAM,CAAC,IAAI,IAAI,IAAI,EAAE,IAAI;AACzB,YAAM,CAAC,IAAI,IAAI,IAAI,EAAE,IAAI;AACzB,gBAAU;AAAA,IAAe,EAAE,OAAO,EAAE,OAAO,EAAE,OAAO,EAAE;AAAA,IAAY,EAAE,OAAO,EAAE,OAAO,EAAE,OAAO,EAAE,IAAI;AAEnG,YAAM,QAAQ,YAAY,CAAC;AAC3B,YAAM,QAAQ,YAAY,CAAC;AAI3B,UAAI,QAAQ,EAAE,KAAK,QAAQ,EAAE,GAAG;AAC9B,eAAO;AAAA,MACT;AAEA,UAAI,UAAU,OAAO;AACnB,eAAO;AAAA,MACT;AAGA,YAAM,CAAC,OAAO,OAAO,OAAO,KAAK,IAAI,0BAA0B,IAAI,IAAI,IAAI,EAAE;AAC7E,UAAI,OAAO;AACT,eAAO,SAAS,CAAC,OAAO,IAAI,OAAO,EAAE,GAAG,CAAC,OAAO,IAAI,OAAO,EAAE,CAAC;AAAA,MAChE;AAEA,UAAI,MAAM,OAAO,MAAM,OAAO,IAAI;AAChC,cAAM,aAAc,EAAE,EAAE;AACxB,eAAO,SAAS,CAAC,IAAI,IAAI,IAAI,EAAE,GAAG,CAAC,IAAI,IAAI,IAAI,EAAE,CAAC;AAAA,MACpD;AACA,UAAI,MAAM,OAAO,MAAM,OAAO,IAAI;AAChC,cAAM,aAAc,EAAE,EAAE;AACxB,eAAO,SAAS,CAAC,IAAI,IAAI,IAAI,EAAE,GAAG,CAAC,IAAI,IAAI,IAAI,EAAE,CAAC;AAAA,MACpD;AACA,UAAI,MAAM,OAAO,MAAM,OAAO,IAAI;AAChC,cAAM,aAAc,EAAE,EAAE;AACxB,eAAO,SAAS,CAAC,IAAI,IAAI,IAAI,EAAE,GAAG,CAAC,IAAI,IAAI,IAAI,EAAE,CAAC;AAAA,MACpD;AAEA,UAAI,CAAC,QAAQ,EAAE,KAAK,OAAO,IAAI;AAC7B,eAAO,SAAS,CAAC,IAAI,IAAI,IAAI,EAAE,GAAG,CAAC,IAAI,IAAI,IAAI,EAAE,CAAC;AAAA,MACpD;AACA,UAAI,CAAC,QAAQ,EAAE,KAAK,OAAO,IAAI;AAC7B,eAAO,SAAS,CAAC,IAAI,IAAI,IAAI,EAAE,GAAG,CAAC,IAAI,IAAI,IAAI,EAAE,CAAC;AAAA,MACpD;AAEA,UAAI,MAAM,MAAM,GAAG,MAAM,QAAQ,GAAG;AAClC,cAAM,yBAAyB,EAAE,QAAQ,EAAE,EAAE;AAC7C,cAAM,OAAO,GAAG,QAAQ,YAAY,EAAE;AACtC,cAAM,OAAO,GAAG,QAAQ,YAAY,EAAE;AACtC,YAAI,SAAS,MAAM;AACjB,gBAAM,cAAe,EAAE,SAAS,EAAE,GAAG;AACrC,iBAAO,SAAS,CAAC,IAAI,IAAI,IAAI,EAAE,GAAG,CAAC,IAAI,IAAI,IAAI,EAAE,CAAC;AAAA,QACpD;AAAA,MACF;AAIA,UAAI,MAAM,OAAO,IAAI;AACnB,eAAO;AAAA,MACT;AAEA,YAAM,CAAC,UAAU,WAAW,WAAW,IAAI,cAAc,MAAM,QAAQ,OAAO,EAAE,GAAG,MAAM,QAAQ,OAAO,EAAE,CAAC;AAE3G,YAAM,IAAI,KAAK,SAAS,KAAK,mBAAmB,SAAS,gBAAgB,QAAQ,mBAAmB,WAAW,EAAE;AAGjH,UAAI,eAAe,WAAW;AAC5B,eAAO,YAAY;AAAA,MACrB;AAGA,UAAI,MAAM,IAAI;AAEZ,YAAI,GAAG,QAAQ,EAAE,MAAM,GAAG;AACxB,iBAAO,SAAS,CAAC,IAAI,GAAG,UAAU,GAAG,MAAM,GAAG,IAAI,EAAE,GAAG,CAAC,IAAI,IAAI,IAAI,EAAE,CAAC;AAAA,QACzE;AACA,YAAI,GAAG,QAAQ,EAAE,MAAM,GAAG;AACxB,iBAAO,SAAS,CAAC,IAAI,IAAI,IAAI,EAAE,GAAG,CAAC,IAAI,GAAG,UAAU,GAAG,MAAM,GAAG,IAAI,EAAE,CAAC;AAAA,QACzE;AAAA,MACF;AAGA,UAAI,SAAS,IAAI,EAAE,GAAG;AACpB,cAAM,SAAS,SAAS,CAAC,IAAI,IAAI,IAAI,EAAE,GAAG,CAAC,IAAI,IAAI,IAAI,EAAE,CAAC;AAC1D,eAAO,SAAS,IAAM,SAAS,MAAM;AAAA,MACvC;AAEA,UAAI,OAAO,MAAM,OAAO,IAAI;AAC1B,cAAM,YAAY,YAAU;AAE5B,YAAI,OAAO,MAAM,SAAS,IAAI,EAAE,GAAG;AACjC,gBAAM,6BAA8B,SAAS,EAAE;AAC/C,iBAAO;AAAA,QACT;AACA,YAAI,OAAO,MAAM,SAAS,IAAI,EAAE,GAAG;AACjC,gBAAM,6BAA8B,SAAS,EAAE;AAC/C,iBAAO;AAAA,QACT;AAAA,MACF;AAEA,UAAI,OAAO,MAAM,OAAO,IAAK;AAE3B,cAAM,YAAY,YAAU;AAC5B,YAAI,OAAO,MAAM,CAAC,IAAI;AACpB,iBAAO;AAAA,QACT;AACA,YAAI,OAAO,MAAM,CAAC,IAAI;AACpB,iBAAO;AAAA,QACT;AAAA,MACF;AAEA,aAAO;AAAA,IACT;AAEA,aAAS,QAAQ,GAAG;AAClB,aAAO,CAAC,KAAK,EAAE,WAAW;AAAA,IAC5B;AAEA,aAAS,SAAS,GAAG,GAAG;AACtB,UAAI,QAAQ,CAAC,GAAG;AACd,eAAO,CAAC,QAAQ,CAAC;AAAA,MACnB;AAEA,aAAO,QAAQ,CAAC;AAAA,IAClB;AAEA,aAAS,cAAc,SAAS,SAAS;AACvC,YAAM,WAAW,MAAM,SAAS,OAAO;AACvC,YAAM,MAAM,aAAa,SAAS,OAAO;AACzC,YAAM,cAAc,IAAO,WAAW;AACtC,aAAO,CAAC,UAAU,KAAK,WAAW;AAAA,IACpC;AAEA,aAAS,YAAY,KAAK;AACxB,YAAM,WAAW,IAAI,OAAO,SAAO,OAAO,QAAQ,YAAY,IAAI,MAAM;AAExE,aAAO,SAAS,KAAK,EAAE;AAAA,IACzB;AAEA,aAAS,aAAa,MAAM,MAAM;AAChC,aAAO,KAAK,SAAS,KAAK,SAAS,KAAK,SAAS,KAAK;AAAA,IACxD;AAEA,aAAS,0BAA0B,IAAI,IAAI,IAAI,IAAI;AACjD,UAAI,CAAC,MAAM,IAAI;AACb,cAAM,CAAC,OAAO,KAAK,IAAI,qBAAqB,EAAE;AAC9C,YAAI,OAAO;AACT,oBAAU;AAAA,OAAyB,EAAE;AAAA,OAAc,KAAK;AAAA,OAAa,KAAK,EAAE;AAC5E,iBAAO,CAAC,OAAO,OAAO,IAAI,EAAE;AAAA,QAC9B;AAAA,MACF;AACA,UAAI,MAAM,CAAC,IAAI;AACb,cAAM,CAAC,OAAO,KAAK,IAAI,qBAAqB,EAAE;AAC9C,YAAI,OAAO;AACT,oBAAU,sBAAsB,EAAE,YAAY,KAAK,aAAa,KAAK,EAAE;AACvE,iBAAO,CAAC,IAAI,IAAI,OAAO,KAAK;AAAA,QAC9B;AAAA,MAEF;AACA,aAAO,CAAC,MAAM,MAAM,MAAM,IAAI;AAAA,IAChC;AAGA,aAAS,qBAAqB,KAAK;AACjC,YAAM,QAAQ,IAAI,MAAM,GAAG;AAC3B,YAAM,MAAM,MAAM;AAClB,UAAI,MAAM,KAAK,CAAC,MAAM,MAAI,CAAC,EAAE,MAAM,eAAe,GAAG;AACnD,eAAO,CAAC,KAAK,IAAI;AAAA,MACnB;AACA,UAAI,MAAM,KAAK,MAAM,MAAI,CAAC,EAAE,MAAM,yBAAyB,GAAG;AAC5D,cAAMA,UAAS,GAAG,MAAM,MAAI,CAAC,CAAC,IAAI,MAAM,MAAI,CAAC,CAAC;AAC9C,eAAO,CAAC,MAAM,MAAM,GAAG,MAAI,CAAC,EAAE,KAAK,GAAG,GAAGA,OAAM;AAAA,MACjD;AAEA,YAAM,SAAS,MAAM,IAAI;AACzB,aAAO,CAAC,MAAM,KAAK,GAAG,GAAG,MAAM;AAAA,IACjC;AAAA,EAEF;AACF;AAEO,gBAAS,SAAS,QAAQ,sBAAsB,CAAC,KAAK,KAAK,KAAK,GAAG,GAAG;AAC3E,QAAM,CAAC,KAAK,IAAI,OAAO,IAAI,QAAQ;AACnC,YAAU,eAAe,KAAK,UAAU,KAAK,CAAC,EAAE;AAEhD,MAAI,OAAO;AACT,UAAM,QAAQ,MAAM,UAEjB,OAAO,CAAC,EAAC,KAAI,MAAM,oBAAoB,SAAS,IAAI,CAAC,EAErD,IAAI,CAAC,EAAC,MAAK,MAAM,mBAAmB,KAAK,IAAI,OAAO,KAAK,IAAI,EAAE,EAC/D,KAAK,GAAG,EACR,QAAQ,gBAAgB,GAAG,EAC3B,QAAQ,oBAAoB,GAAG,EAI/B,QAAQ,QAAQ,GAAG,EACnB,QAAQ,QAAQ,EAAE,EAClB,QAAQ,QAAQ,EAAE;AAGrB,QAAI,WAAW,KAAK,MAAM,IAAI,GAAG;AAC/B,aAAO,MAAM,MAAM,SAAS,MAAM,IAAI,CAAC;AAAA,IACzC;AACA,WAAO;AAAA,EAET;AACA,SAAO;AACT;",
+  "sourcesContent": ["import createDebugLogger from 'debug';\nimport naturalPkg from 'natural';\nconst {LevenshteinDistance: leven} = naturalPkg;\nimport {testStringOrNumber} from '../../../matching-utils.js';\nimport { subfieldToString } from '@natlibfi/marc-record-validators-melinda/dist/utils.js';\n\n\nconst debug = createDebugLogger('@natlibfi/melinda-record-matching:match-detection:features/bib/title');\nconst debugData = debug.extend('data');\nconst debugDev = debug.extend('dev');\n\n\nconst TITLE_MAX = 0.5;\nexport default ({threshold = 0.9} = {}) => ({\n  name: 'Title',\n  extract: ({record, recordExternal}) => {\n    const label = recordExternal && recordExternal.label ? recordExternal.label : 'record';\n    const a = getTitle(record, ['a']);\n    debug(`${label}: title: ${a}`);\n\n    if (a) {\n      const b = normalizeTitle(getTitle(record, ['b'])) || '';\n      const n = normalizeTitle(getTitle(record, ['n'])) || '';\n      const p = normalizeTitle(getTitle(record, ['p'])) || '';\n      return [normalizeTitle(a), b, n, p];\n    }\n\n    return ['', '', '', ''];\n\n    function normalizeTitle(title) {\n      return title\n        // decompose unicode diacritics\n        .normalize('NFD')\n        .replace(/\\b(?:ein|ett|I|one|uno|yksi)\\b/uig, '1')\n        .replace(/\\b(?:dos|kaksi|II|tv\u00E5|two|zwei)\\b/uig, '2')\n        .replace(/\\b(?:drei|III|kolme|three|tre|tres)\\b/uig, '3')\n        .replace(/\\b(?:four|fyra|IV|nelj\u00E4|quat+ro|vier)\\b/uig, '4')\n        .replace(/\\b(?:five|f\u00FCnf|fyra|viisi|V)\\b/uig, '5')\n        .replace(/\\b(?:kuusi|sechts|sex|six|VI)\\b/uig, '6')\n        .replace(/\\b(and|ja|och|und)\\b/uig, '&')\n        // strip non-letters/numbers\n        // - note: combined with decomposing unicode diacritics this normalizes both 'saa' and 's\u00E4\u00E4' as 'saa'\n        // - we could precompose the Finnish letters back to avoid this\n        // - see validator normalize-utf8-diacritics for details\n        .replace(/[^\\p{Letter}\\p{Number}& ]/gu, '') // 2026-07-01: keep ' '. We need it for splitToBaseAndNumber()\n        .replace(/\\s+/ug, ' ')\n        .replace(/ $/u, '')\n        .replace(/^ /u, '')\n        .toLowerCase();\n    }\n\n  },\n  compare: (origA, origB) => {\n    return compare2(origA, origB);\n\n    function compare2(a, b) { // Added this wrapper function as I had issues with recursion\n\n      const [aa, ab, an, ap] = a;\n      const [ba, bb, bn, bp] = b;\n      debugData(`COMPARE:\\n['${aa}', '${ab}', '${an}', '${ap}'] VS\\n['${ba}', '${bb}', '${bn}', '${bp}']`);\n\n      const aFull = toFullTitle(a);\n      const bFull = toFullTitle(b);\n\n      // debugData(`JOINED COMPARE:\\n  '${aFull}' vs  \\n  '${bFull}'`);\n\n      if (isEmpty(aa) || isEmpty(ba)) {\n        return -1.0;\n      }\n\n      if (aFull === bFull) {\n        return TITLE_MAX;\n      }\n\n      // MELKEHITYS-3496-ish: move part of $a to $n:\n      const [altAa, altAn, altBa, altBn] = reconstructTitleAndNumber(aa, an, ba, bn);\n      if (altAa) {\n        return compare2([altAa, ab, altAn, ap], [altBa, bb, altBn, bp]);\n      }\n\n      if (ab && ab !== '' && ab === bb) { // Remove $b from equation (MELKEHITYS-3494)\n        debug(`Ignore \\$b ${ab}`);\n        return compare2([aa, '', an, ap], [ba, '', bn, bp]);\n      }\n      if (an && an !== '' && an === bn) { // Remove $n from equation\n        debug(`Ignore \\$n ${an}`);\n        return compare2([aa, ab, '', ap], [ba, bb, '', bp]);\n      }\n      if (ap && ap !== '' && ap === bp) { // Remove $p from equation\n        debug(`Ignore \\$p ${ap}`);\n        return compare2([aa, ab, an, ''], [ba, bb, bn, '']);\n      }\n      // There seems to be wobble between $b and $p:\n      if (!isEmpty(ab) && ab === bp) {\n        return compare2([aa, '', an, ap], [ba, bb, bn, '']);\n      }\n      if (!isEmpty(ap) && ap === bp) {\n        return compare2([aa, ab, an, ''], [ba, '', bn, bp]);\n      }\n\n      if (an && bn && an.match(/[0-9]/u)) {\n        debug(`NUMBER FOUND. Compare ${an} and ${bn}`);\n        const atmp = an.replace(/[^0-9]/ug, '');\n        const btmp = bn.replace(/[^0-9]/ug, '');\n        if (atmp === btmp) {\n          debug(`Ignore \\$n '${an}' vs '${bn}'`);\n          return compare2([aa, ab, '', ap], [ba, bb, '', bp]);\n        }\n      }\n\n\n      // f245$n information is critical; it can not mismatch at all  (exceptions have already been handled above):\n      if (an && an !== bn) {\n        return -1.0;\n      }\n\n      const [distance, maxLength, correctness] = doLevenshtein(aFull.replace(/ /ug, ''), bFull.replace(/ /ug, ''));\n\n      debug(`'${aFull}' vs '${bFull}': Max length = ${maxLength}, distance = ${distance}, correctness = ${correctness}`);\n\n\n      if (correctness >= threshold) {\n        return TITLE_MAX * 0.8;\n      }\n\n      // Subset removal (MELKEHITYS-3498-ish)\n      if (ab && bb) { // At this point ab and bb are never equal...\n        // If X is a subset of Y, then remove X and shorten Y (= remove the shared content)\n        if (ab.indexOf(bb) === 0) {\n          return compare2([aa, ab.substring(bb.length), an, ap], [ba, '', bn, bp]);\n        }\n        if (bb.indexOf(ab) === 0) {\n          return compare2([aa, '', an, ap], [ba, bb.substring(ab.length), bn, bp]);\n        }\n      }\n\n      // Try the same without $p:\n      if (localXor(ap, bp)) {\n        const result = compare2([aa, ab, an, ''], [ba, bb, bn, '']);\n        return result > 0.0 ? result * 0.8 : result;\n      }\n\n      if (aa === ba && an === bn) {\n        const candScore = TITLE_MAX/2;\n        // $b is missing from one:\n        if (ap === bp && localXor(ab, bb)) {\n          debug(`Handle omitted \\$b using x ${candScore}`);\n          return candScore;\n        }\n        if (ab === bb && localXor(ap, bp)) {\n          debug(`Handle omitted \\$p using x ${candScore}`);\n          return candScore;\n        }\n      }\n\n      if (an === bn && ap === bp ) {\n        // $b-less $a is other record's $b, see MELKEHITYS-3485 for motivation (though we skip 246 and 490 here):\n        const candScore = TITLE_MAX/2;\n        if (aa === bb && !ab) {\n          return candScore;\n        }\n        if (ba === ab && !bb) {\n          return candScore;\n        }\n      }\n\n      return -0.5; // Not likely\n    }\n\n    function isEmpty(x) {\n      return !x || x.length === 0;\n    }\n\n    function localXor(x, y) {\n      if (isEmpty(x)) {\n        return !isEmpty(y);\n      }\n      // 'x' exists, thus 'y' can not exist:\n      return isEmpty(y);\n    }\n\n    function doLevenshtein(string1, string2) {\n      const distance = leven(string1, string2);\n      const len = getMaxLength(string1, string2);\n      const correctness = 1.0 - (distance / len);\n      return [distance, len, correctness];\n    }\n\n    function toFullTitle(arr) {\n      const relevant = arr.filter(val => typeof val === 'string' && val.length);\n      // No longer add space between subfields. Due to heavy normalization there are no spaces left in array elements either!\n      return relevant.join('');\n    }\n\n    function getMaxLength(str1, str2) {\n      return str1.length > str2.length ? str1.length : str2.length;\n    }\n\n    function reconstructTitleAndNumber(a1, n1, a2, n2) {\n      if (!n1 && n2) {\n        const [altA1, altN1] = splitToBaseAndNumber(a1);\n        if (altN1) {\n          debugData(`A: Reconstruct:\\n $a '${a1}' =>\\n $a '${altA1}' +\\n $n '${altN1}`);\n          return [altA1, altN1, a2, n2];\n        }\n      }\n      if (n1 && !n2) {\n        const [altA2, altN2] = splitToBaseAndNumber(a2);\n        if (altN2) {\n          debugData(`B: Reconstruct $a '${a2}' as $a '${altA2}' and $n '${altN2}`);\n          return [a1, n1, altA2, altN2];\n        }\n\n      }\n      return [null, null, null, null]; // Fail\n    }\n\n\n    function splitToBaseAndNumber(val) {\n      const words = val.split(' ');\n      const len = words.length;\n      if (len < 2 || !words[len-1].match(/^[1-9][0-9]*$/)) {\n        return [val, null];\n      }\n      if (len > 2 && words[len-2].match(/^(del|osa|vol|volume)$/u)) { // NB! \"vol.\" has lost it's '.' during normalization!\n        const number = `${words[len-2]} ${words[len-1]}`;\n        return [words.slice(0, len-2).join(' '), number];\n      }\n\n      const number = words.pop();\n      return [words.join(' '), number];\n    }\n\n  }\n});\n\nexport function getTitle(record, targetSubfieldCodes = ['a', 'b', 'n', 'p']) {\n  const [field] = record.get(/^245$/u);\n  debugData(`titleField: ${JSON.stringify(field)}`);\n  debugDev(`targetSubfieldCodes: ${JSON.stringify(targetSubfieldCodes)}`);\n\n\n  if (field) {\n    const title = field.subfields\n      // get also $n:s and $p:s here\n      .filter(({code}) => targetSubfieldCodes.includes(code))\n      // Would be nice to normalize $n values...\n      .map(({value}) => testStringOrNumber(value) ? String(value) : '')\n      .join(' ')\n      // Note: we do not want to removed stuff in brackets if its the only stuff in field ... (245 $a [Psi] / ... )\n      //.replace(/\\[[^\\]]*\\]/ug, ' ') // Remove [whatever] stuff. ADD TEST FOR THIS\n      .replace(/ [=\\/:](?:$| )/ug, ' ') // Strip punctuation\n      // Also \u0111 vs d pairs seen:\n      //.replace(/(?:\u0111|\u0227|t\u0304|k\u030C|\u01E7|s\u0306|c\u0306)/ug, '') // Hack. Saamelaisbibliografia has often dropped this wierd characters (oft old articles, no longer used in sami either)\n      // trim:\n      .replace(/ +/ug, ' ')\n      .replace(/^ +/u, '')\n      .replace(/ +$/u, '');\n\n    debugDev(`getTitle: title: ${title}`)\n\n    const cleanTitle = title.replace(/\\[[^\\]]*\\]/ug, ' ') // Remove [whatever] stuff. ADD TEST FOR THIS\n      .replace(/ +/ug, ' ')\n      .replace(/^ +/u, '')\n      .replace(/ +$/u, '');\n\n    debugDev(`getTitle: cleanTitle: ${cleanTitle} (${cleanTitle.length})`);\n    const usableTitle = cleanTitle.length > 0 ? cleanTitle : title;\n    debugDev(`getTitle: usableTitle: ${usableTitle} (${usableTitle.length})`);\n    \n\n    // Skip non-filing indicator (note that '9' is a magic indicator value, so we don't do it):\n    if (/^[1-8]$/u.test(field.ind2)) { // Skip non-filing characters\n      const nonFilingTitle = usableTitle.slice(parseInt(field.ind2));\n      if (nonFilingTitle.length > 0) return nonFilingTitle;\n    }\n    return usableTitle;\n\n  }\n  return false;\n}"],
+  "mappings": "AAAA,OAAO,uBAAuB;AAC9B,OAAO,gBAAgB;AACvB,MAAM,EAAC,qBAAqB,MAAK,IAAI;AACrC,SAAQ,0BAAyB;AACjC,SAAS,wBAAwB;AAGjC,MAAM,QAAQ,kBAAkB,sEAAsE;AACtG,MAAM,YAAY,MAAM,OAAO,MAAM;AACrC,MAAM,WAAW,MAAM,OAAO,KAAK;AAGnC,MAAM,YAAY;AAClB,eAAe,CAAC,EAAC,YAAY,IAAG,IAAI,CAAC,OAAO;AAAA,EAC1C,MAAM;AAAA,EACN,SAAS,CAAC,EAAC,QAAQ,eAAc,MAAM;AACrC,UAAM,QAAQ,kBAAkB,eAAe,QAAQ,eAAe,QAAQ;AAC9E,UAAM,IAAI,SAAS,QAAQ,CAAC,GAAG,CAAC;AAChC,UAAM,GAAG,KAAK,YAAY,CAAC,EAAE;AAE7B,QAAI,GAAG;AACL,YAAM,IAAI,eAAe,SAAS,QAAQ,CAAC,GAAG,CAAC,CAAC,KAAK;AACrD,YAAM,IAAI,eAAe,SAAS,QAAQ,CAAC,GAAG,CAAC,CAAC,KAAK;AACrD,YAAM,IAAI,eAAe,SAAS,QAAQ,CAAC,GAAG,CAAC,CAAC,KAAK;AACrD,aAAO,CAAC,eAAe,CAAC,GAAG,GAAG,GAAG,CAAC;AAAA,IACpC;AAEA,WAAO,CAAC,IAAI,IAAI,IAAI,EAAE;AAEtB,aAAS,eAAe,OAAO;AAC7B,aAAO,MAEJ,UAAU,KAAK,EACf,QAAQ,qCAAqC,GAAG,EAChD,QAAQ,wCAAwC,GAAG,EACnD,QAAQ,4CAA4C,GAAG,EACvD,QAAQ,8CAA8C,GAAG,EACzD,QAAQ,qCAAqC,GAAG,EAChD,QAAQ,sCAAsC,GAAG,EACjD,QAAQ,2BAA2B,GAAG,EAKtC,QAAQ,+BAA+B,EAAE,EACzC,QAAQ,SAAS,GAAG,EACpB,QAAQ,OAAO,EAAE,EACjB,QAAQ,OAAO,EAAE,EACjB,YAAY;AAAA,IACjB;AAAA,EAEF;AAAA,EACA,SAAS,CAAC,OAAO,UAAU;AACzB,WAAO,SAAS,OAAO,KAAK;AAE5B,aAAS,SAAS,GAAG,GAAG;AAEtB,YAAM,CAAC,IAAI,IAAI,IAAI,EAAE,IAAI;AACzB,YAAM,CAAC,IAAI,IAAI,IAAI,EAAE,IAAI;AACzB,gBAAU;AAAA,IAAe,EAAE,OAAO,EAAE,OAAO,EAAE,OAAO,EAAE;AAAA,IAAY,EAAE,OAAO,EAAE,OAAO,EAAE,OAAO,EAAE,IAAI;AAEnG,YAAM,QAAQ,YAAY,CAAC;AAC3B,YAAM,QAAQ,YAAY,CAAC;AAI3B,UAAI,QAAQ,EAAE,KAAK,QAAQ,EAAE,GAAG;AAC9B,eAAO;AAAA,MACT;AAEA,UAAI,UAAU,OAAO;AACnB,eAAO;AAAA,MACT;AAGA,YAAM,CAAC,OAAO,OAAO,OAAO,KAAK,IAAI,0BAA0B,IAAI,IAAI,IAAI,EAAE;AAC7E,UAAI,OAAO;AACT,eAAO,SAAS,CAAC,OAAO,IAAI,OAAO,EAAE,GAAG,CAAC,OAAO,IAAI,OAAO,EAAE,CAAC;AAAA,MAChE;AAEA,UAAI,MAAM,OAAO,MAAM,OAAO,IAAI;AAChC,cAAM,aAAc,EAAE,EAAE;AACxB,eAAO,SAAS,CAAC,IAAI,IAAI,IAAI,EAAE,GAAG,CAAC,IAAI,IAAI,IAAI,EAAE,CAAC;AAAA,MACpD;AACA,UAAI,MAAM,OAAO,MAAM,OAAO,IAAI;AAChC,cAAM,aAAc,EAAE,EAAE;AACxB,eAAO,SAAS,CAAC,IAAI,IAAI,IAAI,EAAE,GAAG,CAAC,IAAI,IAAI,IAAI,EAAE,CAAC;AAAA,MACpD;AACA,UAAI,MAAM,OAAO,MAAM,OAAO,IAAI;AAChC,cAAM,aAAc,EAAE,EAAE;AACxB,eAAO,SAAS,CAAC,IAAI,IAAI,IAAI,EAAE,GAAG,CAAC,IAAI,IAAI,IAAI,EAAE,CAAC;AAAA,MACpD;AAEA,UAAI,CAAC,QAAQ,EAAE,KAAK,OAAO,IAAI;AAC7B,eAAO,SAAS,CAAC,IAAI,IAAI,IAAI,EAAE,GAAG,CAAC,IAAI,IAAI,IAAI,EAAE,CAAC;AAAA,MACpD;AACA,UAAI,CAAC,QAAQ,EAAE,KAAK,OAAO,IAAI;AAC7B,eAAO,SAAS,CAAC,IAAI,IAAI,IAAI,EAAE,GAAG,CAAC,IAAI,IAAI,IAAI,EAAE,CAAC;AAAA,MACpD;AAEA,UAAI,MAAM,MAAM,GAAG,MAAM,QAAQ,GAAG;AAClC,cAAM,yBAAyB,EAAE,QAAQ,EAAE,EAAE;AAC7C,cAAM,OAAO,GAAG,QAAQ,YAAY,EAAE;AACtC,cAAM,OAAO,GAAG,QAAQ,YAAY,EAAE;AACtC,YAAI,SAAS,MAAM;AACjB,gBAAM,cAAe,EAAE,SAAS,EAAE,GAAG;AACrC,iBAAO,SAAS,CAAC,IAAI,IAAI,IAAI,EAAE,GAAG,CAAC,IAAI,IAAI,IAAI,EAAE,CAAC;AAAA,QACpD;AAAA,MACF;AAIA,UAAI,MAAM,OAAO,IAAI;AACnB,eAAO;AAAA,MACT;AAEA,YAAM,CAAC,UAAU,WAAW,WAAW,IAAI,cAAc,MAAM,QAAQ,OAAO,EAAE,GAAG,MAAM,QAAQ,OAAO,EAAE,CAAC;AAE3G,YAAM,IAAI,KAAK,SAAS,KAAK,mBAAmB,SAAS,gBAAgB,QAAQ,mBAAmB,WAAW,EAAE;AAGjH,UAAI,eAAe,WAAW;AAC5B,eAAO,YAAY;AAAA,MACrB;AAGA,UAAI,MAAM,IAAI;AAEZ,YAAI,GAAG,QAAQ,EAAE,MAAM,GAAG;AACxB,iBAAO,SAAS,CAAC,IAAI,GAAG,UAAU,GAAG,MAAM,GAAG,IAAI,EAAE,GAAG,CAAC,IAAI,IAAI,IAAI,EAAE,CAAC;AAAA,QACzE;AACA,YAAI,GAAG,QAAQ,EAAE,MAAM,GAAG;AACxB,iBAAO,SAAS,CAAC,IAAI,IAAI,IAAI,EAAE,GAAG,CAAC,IAAI,GAAG,UAAU,GAAG,MAAM,GAAG,IAAI,EAAE,CAAC;AAAA,QACzE;AAAA,MACF;AAGA,UAAI,SAAS,IAAI,EAAE,GAAG;AACpB,cAAM,SAAS,SAAS,CAAC,IAAI,IAAI,IAAI,EAAE,GAAG,CAAC,IAAI,IAAI,IAAI,EAAE,CAAC;AAC1D,eAAO,SAAS,IAAM,SAAS,MAAM;AAAA,MACvC;AAEA,UAAI,OAAO,MAAM,OAAO,IAAI;AAC1B,cAAM,YAAY,YAAU;AAE5B,YAAI,OAAO,MAAM,SAAS,IAAI,EAAE,GAAG;AACjC,gBAAM,6BAA8B,SAAS,EAAE;AAC/C,iBAAO;AAAA,QACT;AACA,YAAI,OAAO,MAAM,SAAS,IAAI,EAAE,GAAG;AACjC,gBAAM,6BAA8B,SAAS,EAAE;AAC/C,iBAAO;AAAA,QACT;AAAA,MACF;AAEA,UAAI,OAAO,MAAM,OAAO,IAAK;AAE3B,cAAM,YAAY,YAAU;AAC5B,YAAI,OAAO,MAAM,CAAC,IAAI;AACpB,iBAAO;AAAA,QACT;AACA,YAAI,OAAO,MAAM,CAAC,IAAI;AACpB,iBAAO;AAAA,QACT;AAAA,MACF;AAEA,aAAO;AAAA,IACT;AAEA,aAAS,QAAQ,GAAG;AAClB,aAAO,CAAC,KAAK,EAAE,WAAW;AAAA,IAC5B;AAEA,aAAS,SAAS,GAAG,GAAG;AACtB,UAAI,QAAQ,CAAC,GAAG;AACd,eAAO,CAAC,QAAQ,CAAC;AAAA,MACnB;AAEA,aAAO,QAAQ,CAAC;AAAA,IAClB;AAEA,aAAS,cAAc,SAAS,SAAS;AACvC,YAAM,WAAW,MAAM,SAAS,OAAO;AACvC,YAAM,MAAM,aAAa,SAAS,OAAO;AACzC,YAAM,cAAc,IAAO,WAAW;AACtC,aAAO,CAAC,UAAU,KAAK,WAAW;AAAA,IACpC;AAEA,aAAS,YAAY,KAAK;AACxB,YAAM,WAAW,IAAI,OAAO,SAAO,OAAO,QAAQ,YAAY,IAAI,MAAM;AAExE,aAAO,SAAS,KAAK,EAAE;AAAA,IACzB;AAEA,aAAS,aAAa,MAAM,MAAM;AAChC,aAAO,KAAK,SAAS,KAAK,SAAS,KAAK,SAAS,KAAK;AAAA,IACxD;AAEA,aAAS,0BAA0B,IAAI,IAAI,IAAI,IAAI;AACjD,UAAI,CAAC,MAAM,IAAI;AACb,cAAM,CAAC,OAAO,KAAK,IAAI,qBAAqB,EAAE;AAC9C,YAAI,OAAO;AACT,oBAAU;AAAA,OAAyB,EAAE;AAAA,OAAc,KAAK;AAAA,OAAa,KAAK,EAAE;AAC5E,iBAAO,CAAC,OAAO,OAAO,IAAI,EAAE;AAAA,QAC9B;AAAA,MACF;AACA,UAAI,MAAM,CAAC,IAAI;AACb,cAAM,CAAC,OAAO,KAAK,IAAI,qBAAqB,EAAE;AAC9C,YAAI,OAAO;AACT,oBAAU,sBAAsB,EAAE,YAAY,KAAK,aAAa,KAAK,EAAE;AACvE,iBAAO,CAAC,IAAI,IAAI,OAAO,KAAK;AAAA,QAC9B;AAAA,MAEF;AACA,aAAO,CAAC,MAAM,MAAM,MAAM,IAAI;AAAA,IAChC;AAGA,aAAS,qBAAqB,KAAK;AACjC,YAAM,QAAQ,IAAI,MAAM,GAAG;AAC3B,YAAM,MAAM,MAAM;AAClB,UAAI,MAAM,KAAK,CAAC,MAAM,MAAI,CAAC,EAAE,MAAM,eAAe,GAAG;AACnD,eAAO,CAAC,KAAK,IAAI;AAAA,MACnB;AACA,UAAI,MAAM,KAAK,MAAM,MAAI,CAAC,EAAE,MAAM,yBAAyB,GAAG;AAC5D,cAAMA,UAAS,GAAG,MAAM,MAAI,CAAC,CAAC,IAAI,MAAM,MAAI,CAAC,CAAC;AAC9C,eAAO,CAAC,MAAM,MAAM,GAAG,MAAI,CAAC,EAAE,KAAK,GAAG,GAAGA,OAAM;AAAA,MACjD;AAEA,YAAM,SAAS,MAAM,IAAI;AACzB,aAAO,CAAC,MAAM,KAAK,GAAG,GAAG,MAAM;AAAA,IACjC;AAAA,EAEF;AACF;AAEO,gBAAS,SAAS,QAAQ,sBAAsB,CAAC,KAAK,KAAK,KAAK,GAAG,GAAG;AAC3E,QAAM,CAAC,KAAK,IAAI,OAAO,IAAI,QAAQ;AACnC,YAAU,eAAe,KAAK,UAAU,KAAK,CAAC,EAAE;AAChD,WAAS,wBAAwB,KAAK,UAAU,mBAAmB,CAAC,EAAE;AAGtE,MAAI,OAAO;AACT,UAAM,QAAQ,MAAM,UAEjB,OAAO,CAAC,EAAC,KAAI,MAAM,oBAAoB,SAAS,IAAI,CAAC,EAErD,IAAI,CAAC,EAAC,MAAK,MAAM,mBAAmB,KAAK,IAAI,OAAO,KAAK,IAAI,EAAE,EAC/D,KAAK,GAAG,EAGR,QAAQ,oBAAoB,GAAG,EAI/B,QAAQ,QAAQ,GAAG,EACnB,QAAQ,QAAQ,EAAE,EAClB,QAAQ,QAAQ,EAAE;AAErB,aAAS,oBAAoB,KAAK,EAAE;AAEpC,UAAM,aAAa,MAAM,QAAQ,gBAAgB,GAAG,EACjD,QAAQ,QAAQ,GAAG,EACnB,QAAQ,QAAQ,EAAE,EAClB,QAAQ,QAAQ,EAAE;AAErB,aAAS,yBAAyB,UAAU,KAAK,WAAW,MAAM,GAAG;AACrE,UAAM,cAAc,WAAW,SAAS,IAAI,aAAa;AACzD,aAAS,0BAA0B,WAAW,KAAK,YAAY,MAAM,GAAG;AAIxE,QAAI,WAAW,KAAK,MAAM,IAAI,GAAG;AAC/B,YAAM,iBAAiB,YAAY,MAAM,SAAS,MAAM,IAAI,CAAC;AAC7D,UAAI,eAAe,SAAS,EAAG,QAAO;AAAA,IACxC;AACA,WAAO;AAAA,EAET;AACA,SAAO;AACT;",
   "names": ["number"]
 }

package/package.json

@@ -13,7 +13,7 @@
 		"url": "https://github.com/NatLibFi/melinda-record-matching-js"
 	},
 	"license": "MIT",
-	"version": "6.0.0-alpha.1",
+	"version": "6.0.1-alpha.1",
 	"type": "module",
 	"main": "./dist/index.js",
 	"bin": "./dist/cli.js",
@@ -33,7 +33,7 @@
 		"test:base": "node --test --test-force-exit --experimental-test-coverage --test-reporter=spec 'test/**/*.test.js'",
 		"test": "npm run lint && npm run test:base",
 		"test:only": "npm run lint && npm run test:base:only",
-		"test:base:only": "cross-env DEBUG=@natlibfi/* NODE_ENV=test node --test --test-only --test-force-exit --experimental-test-coverage --test-reporter=spec 'test/**/*.test.js'",
+		"test:base:only": "cross-env DEBUG=@natlibfi/* NODE_ENV=test node --test --test-only --test-force-exit --experimental-test-coverage --test-reporter=spec 'test/*.test.js' 'test/**/*.test.js'",
 		"test:base:only:index": "cross-env DEBUG=@natlibfi/* NODE_ENV=test node --test --test-only --test-force-exit --experimental-test-coverage --test-reporter=spec './test/index.test.js'",
 		"watch:test": "cross-env DEBUG=@natlibfi/* NODE_ENV=test node --watch --test --experimental-test-coverage --test-reporter=spec 'test/**/*.test.js'",
 		"dev": "npm run watch:test",

package/src/candidate-search/candidate-search-utils.js

@@ -1,13 +1,23 @@
 import createDebugLogger from 'debug';
+import {uniqArray} from '@natlibfi/marc-record-validators-melinda/dist/utils.js';
 
 export function toQueries(identifiers, queryString) {
 
   const debug = createDebugLogger('@natlibfi/melinda-record-matching:toQueries');
   const debugData = debug.extend('data');
+  const debugDev = debug.extend('dev');
+
+  // For dc.identifier search use max length of 40 characters to avoid dc.identifier -> 1007 mapped query from crashing
+  // DEVELOP: check if the length crash is related to mapping the query to multiple Aleph indexes
+  const useMaxLength = queryString === 'dc.identifier' ? true : false;
+  const maxLength = 40;
+  debugData(`Adding ${queryString} to ${JSON.stringify(identifiers)}`)
+  debugDev(`We are ${useMaxLength ? `using maxLength ${maxLength}, queryString ${queryString}` : `not using maxLength, queryString ${queryString}`}`);
+  const croppedIdentifiers = uniqArray(useMaxLength ? identifiers.map((identifier) => identifier.substring(0, maxLength)) : identifiers);
 
   // We quote the identifier, if it contains a slash! (Slash in non-quoted SRU-search breaks search.)
   // We also quote the identifier, if it starts with caret (f028 searches fail without caret and quotes...)
-  const quotedIdentifiers = identifiers.map(identifier => identifier.match(/\//u) || identifier.match(/\^/u) ? `"${identifier}"` : `${identifier}`);
+  const quotedIdentifiers = croppedIdentifiers.map(identifier => identifier.match(/\//u) || identifier.match(/\^/u) ? `"${identifier}"` : `${identifier}`);
 
   // We can't pair queries with starting caret and without (ie. left anchored queries with non-left anchored queries)
   const caretPairs = toPairs(quotedIdentifiers.filter(identifier => identifier.match(/\^/u)));
@@ -17,10 +27,22 @@ export function toQueries(identifiers, queryString) {
   debugData(`Pairs (${pairs.length}): ${JSON.stringify(pairs)}`);
 
   // Aleph supports only two queries with or -operator (This is not actually true)
-  const queries = pairs.map(([a, b]) => b ? `${queryString}=${a} or ${queryString}=${b}` : `${queryString}=${a}`);
+  const queries = pairs.map(([a, b]) => {
+    const lengths = a.length + (b ? b.length : 0);
+    debugDev(`Lengths: A: ${a} (${a.length}) + B: ${b} (${b ? b.length : 0}) = ${lengths}`);
+
+    // Do not create a paired query if query length would be too long
+    if (useMaxLength && lengths > maxLength && a && b) {
+      return [`${queryString}=${a}`, `${queryString}=${b}`];
+    }
+    return b ? `${queryString}=${a} or ${queryString}=${b}` : `${queryString}=${a}`}
+  );
+
   debugData(`Queries (${queries.length}): ${JSON.stringify(queries)}`);
+  const flatQueries = queries.flat();
+  debugData(`FlatQueries (${flatQueries.length}): ${JSON.stringify(flatQueries)}`);
 
-  return queries;
+  return flatQueries;
 }
 
 function toPairs(array) {
@@ -29,3 +51,4 @@ function toPairs(array) {
   }
   return [array.slice(0, 2)].concat(toPairs(array.slice(2), 2));
 }
+

package/src/candidate-search/query-list/bib.js

@@ -4,8 +4,11 @@ import {getMelindaIdsF035, validateSidFieldSubfieldCounts, getSubfieldValues, te
 import {extractHostMelindaIdsFromField, extractPublicationYearFrom773, getHostMelindaFields} from './component.js';
 import {fieldToString} from '@natlibfi/marc-record-validators-melinda';
 import {getTitle} from '../../match-detection/features/bib/title.js';
+import {uniqArray} from '@natlibfi/marc-record-validators-melinda/dist/utils.js';
 
 const debug = createDebugLogger('@natlibfi/melinda-record-matching:candidate-search:query');
+const debugDev = debug.extend('dev');
+
 
 const IS_OK_ALONE_THRESHOLD = 5;
 
@@ -187,6 +190,7 @@ export function bibTitleAuthorYearAlternates(record) {
 
 function getTitleForQuery(record) {
   const title = getTitle(record, ['a']);
+  debug(`getTitleForQuery: title ${title}`);
   if (title) {
     const nWords = title.split(' ');
     // If lone $a is deemed too short, fetch $b as well:
@@ -207,11 +211,13 @@ function getTitleForQuery(record) {
 
 function dcTitle(record, onlyTitleLength, alternates = false) {
   const title = getTitleForQuery(record);
+  debug(`dcTitle: title: ${title}`);
   if (!testStringOrNumber(title)) {
     return [];
   }
 
   const formatted = getFormattedTitle(title);
+  debug(`dcTitle: formatted: ${formatted}`);
 
   // use word search for titles starting with a boolean
   const useWordSearch = checkUseWordSearch(formatted);
@@ -223,6 +229,10 @@ function dcTitle(record, onlyTitleLength, alternates = false) {
 
   const queryIsOkAlone = !useWordSearch && formatted.length >= IS_OK_ALONE_THRESHOLD;
 
+  if (formatted.length<1) {
+    return ["", formatted, queryIsOkAlone];
+  }
+
   // use word search without ending * also in combination searches to avoid SRU-server crashes [MRA-189]
   return [`dc.title="${useWordSearch || !queryIsOkAlone ? '' : '^'}${formatted}${queryIsOkAlone ? '*' : ''}"`, formatted, queryIsOkAlone];
 
@@ -242,6 +252,7 @@ function dcTitle(record, onlyTitleLength, alternates = false) {
 export function bibTitleAuthorPublisher({record, onlyTitleLength, addYear = false, alternates = false, alternateQueries = []}) {
   debug(`bibTitleAuthorPublisher, onlyTitleLength: ${onlyTitleLength}, addYear: ${addYear}, alternates: ${alternates}`);
   const [query, formatted, queryIsOkAlone] = dcTitle(record, onlyTitleLength, alternates);
+  debug(`query ${JSON.stringify(query)}, formatted: ${JSON.stringify(formatted)}, isOKAlone: ${JSON.stringify(queryIsOkAlone)}`);
   if (query === undefined) {
     return [];
   }
@@ -262,11 +273,12 @@ export function bibTitleAuthorPublisher({record, onlyTitleLength, addYear = fals
     debug('addAuthorsToSearch');
     const [authorQuery] = bibAuthors(record);
     if (authorQuery !== undefined) {
+      const combinedQuery = query.length > 0 ? `${authorQuery} AND ${query}` : `${authorQuery}`;
       if (addYear) {
-        const newAlternateQueries = alternates ? [...alternateQueries, `${authorQuery} AND ${query}`] : alternateQueries;
-        return addYearToSearch({query: `${authorQuery} AND ${query}`, queryIsOkAlone: true, alternates, alternateQueries: newAlternateQueries});
+        const newAlternateQueries = alternates ? [...alternateQueries, combinedQuery] : alternateQueries;
+        return addYearToSearch({query: combinedQuery, queryIsOkAlone: true, alternates, alternateQueries: newAlternateQueries});
       }
-      return alternates ? alternateQueries : [`${authorQuery} AND ${query}`];
+      return alternates ? alternateQueries : [combinedQuery];
     }
     return addPublisherToSearch({query, queryIsOkAlone, addYear, alternates, alternateQueries});
     //return [];
@@ -275,11 +287,12 @@ export function bibTitleAuthorPublisher({record, onlyTitleLength, addYear = fals
   function addPublisherToSearch({query, queryIsOkAlone = false, addYear = false, alternates = false, alternateQueries = []}) {
     const [publisherQuery] = bibPublishers(record);
     if (publisherQuery !== undefined) {
+      const combinedQuery = query.length > 0 ? `${publisherQuery} AND ${query}` : `${publisherQuery}`;
       if (addYear) {
-        const newAlternateQueries = alternates ? [...alternateQueries, `${publisherQuery} AND ${query}`] : alternateQueries;
-        return addYearToSearch({query: `${publisherQuery} AND ${query}`, queryIsOkAlone: true, alternates, alternateQueries: newAlternateQueries});
+        const newAlternateQueries = alternates ? [...alternateQueries, combinedQuery] : alternateQueries;
+        return addYearToSearch({query: combinedQuery, queryIsOkAlone: true, alternates, alternateQueries: newAlternateQueries});
       }
-      return alternates ? alternateQueries : [`${publisherQuery} AND ${query}`];
+      return alternates ? alternateQueries : [combinedQuery];
     }
     if (queryIsOkAlone && !addYear) {
       return alternates ? alternateQueries : [`${query}`];
@@ -290,7 +303,7 @@ export function bibTitleAuthorPublisher({record, onlyTitleLength, addYear = fals
   function addYearToSearch({query, queryIsOkAlone = false, alternates = false, alternateQueries = []}) {
     const [yearQuery] = bibYear(record);
     if (yearQuery !== undefined) {
-      const yearAndOtherQueries = `${yearQuery} AND ${query}`;
+      const yearAndOtherQueries = query.length > 0 ? `${yearQuery} AND ${query}` : `${yearQuery}`;
       return alternates ? [...alternateQueries, yearAndOtherQueries] : [yearAndOtherQueries];
     }
     if (queryIsOkAlone) {
@@ -413,10 +426,12 @@ export function bibYear(record) {
 
     debugData(`f008: ${JSON.stringify(f008)}`);
     const {value} = f008;
-    if (value === '||||' || value === '    ') { // Meaningless values
+    const year = value.slice(7, 11);
+    if (year === '||||' || year === '    ') { // Meaningless values
+      debug('Meaningless year in f008');
       return false;
     }
-    return value.slice(7, 11);
+    return year;
   }
 }
 
@@ -431,9 +446,11 @@ export function bibStandardIdentifiers(record) {
 
   const debug = createDebugLogger('@natlibfi/melinda-record-matching:candidate-search:query:bibStandardIdentifiers');
   const debugData = debug.extend('data');
+  const debugDev = debug.extend('dev');
+
   debug(`Creating queries for standard identifiers`);
 
-  // DEVELOP: should we query also f015 and f028?
+  // DEVELOP: query identifiers from their own indexes (this needs some mapping work in SRU/Aleph)
 
   // const fields = record.get(/^(?<def>020|022|024)$/u);
   const fields = record.get(/^(?<def>015|020|022|024|028)$/u);
@@ -475,14 +492,71 @@ export function bibStandardIdentifiers(record) {
 
     if (tag === '028') { // TODO: test
       const a = subfields.find(sf => sf.code === 'a');
-      if (a) {
-        const b = subfields.find(sf => sf.code === 'b');
-        // TODO: normalize?
-        if (b) {
-          return [`^${a.value}`, `^${b.value} ${a.value}`];
+      const b = subfields.find(sf => sf.code === 'b');
+
+      // Handle cases where there is a qualifier in brackets after value
+      const cleanAs = cleanQualifiers(a?.value);
+      const cleanBs = cleanQualifiers(b?.value);
+      debugDev(JSON.stringify(cleanAs));
+      debugDev(JSON.stringify(cleanBs));
+
+      // Handle cases where there are several values separated by a comma in subfield
+      const splitAs = cleanAs.map(elem => elem.split(',')).concat(cleanAs).flat();
+      const splitBs = cleanBs.map(elem => elem.split(',')).concat(cleanBs).flat();
+      debugDev(JSON.stringify(splitAs));
+      debugDev(JSON.stringify(splitBs));
+
+      const normAs = splitAs.map(elem => normalizeF028ForSearch(elem));
+      const normBs = splitBs.map(elem => normalizeF028ForSearch(elem));
+      debugDev(JSON.stringify(normAs));
+      debugDev(JSON.stringify(normBs));
+
+      const identifiers = normAs.map((normA, index) => {
+        debugDev(`Handling ${normA} from index: ${index}`);
+        // Paired $a + $b
+        if (normA.length > 0 && normBs[index]?.length > 0) {
+          debugDev(`Found a (${normA}) + b (${normBs[index]})`);
+          // let's not return plain $b
+          return [`^${normBs[index]} ${normA}`, `^${normA} ${normBs[index]}`, `^${normA}`];
+          //return [`^${normA}`, `^${normBs[index]}`, `^${normBs[index]} ${normA}`, `^${normA} ${normBs[index]}`];
+
+        }
+        // if there is no pair, get $a + first $b
+        // NOTE: we do not get all possible pairs
+        if (normA.length > 0 && normBs[0]?.length > 0) {
+          debugDev(`Found a (${normA}) + b (${normBs[0]})`);
+          // let's not return plain $b
+          return [`^${normBs[0]} ${normA}`, `^${normA} ${normBs[0]}`, `^${normA}`];
+          //return [`^${normA}`, `^${normBs[index]}`, `^${normBs[index]} ${normA}`, `^${normA} ${normBs[index]}`];
+
         }
-        return [`^${a.value}`];
+        // Otherwise just $a
+        if (normA.length > 0) {
+          debugDev(`Found a (${normA}) but no b ${normBs[index]}`);
+          return [`^${normA}`];
+        }
+        return [];
+      });
+
+      debugDev(`Current identifiers from f028: ${JSON.stringify(identifiers)}`);
+      return uniqArray(identifiers.flat());
+
+      function cleanQualifiers(val) {
+        if (val !== undefined) {
+        const cleanVal = val.replace(/ +\(.*\)/gu, '').replace(/ +/ug, ' ').replace(/^ +/ug, '').replace(/ +$/ug, '');
+        return uniqArray([val, cleanVal]);
+        }
+        return [];
       }
+
+      function normalizeF028ForSearch(val) {
+        // see normalization routine 34 used for 028-index in Aleph
+        const f028NormCharsRegex = /[-"<>;:%=~`!\(\)\{\}\.\?\/\@\*\^]/g;
+        const normVal = val.replace(f028NormCharsRegex, ' ').replace(/ +/ug, ' ').replace(/^ +/ug, '').replace(/ +$/ug, '').toLowerCase();
+        // Do not return a value normalized to just a space
+        return normVal !== " " ? normVal : "";
+      }
+
     }
 
     // Default seems to be 024:

package/src/match-detection/features/bib/publisherNumber.js

@@ -26,6 +26,7 @@ export default () => ({
         }
         const a = field.subfields.find(sf => sf.code === 'a');
         // NB! Publisher number must contain a digit! :-)
+        // NOTE: we might have some rare cases with publisher numbers that do not have numbers...
         if (!a || !a.value || !a.value.match(/[0-9]/u)) {
           return fieldsToPublisherNumber(remainingFields, result);
         }
@@ -40,7 +41,7 @@ export default () => ({
       }
 
       function normalizePublisherNumber(val) {
-        return val.replace(/[- .,:;]/ug, '');
+        return val.replace(/[- .,:;]/ug, '').toLowerCase();
       }
     }
   },

package/src/match-detection/features/bib/title.js

@@ -7,6 +7,8 @@ import { subfieldToString } from '@natlibfi/marc-record-validators-melinda/dist/
 
 const debug = createDebugLogger('@natlibfi/melinda-record-matching:match-detection:features/bib/title');
 const debugData = debug.extend('data');
+const debugDev = debug.extend('dev');
+
 
 const TITLE_MAX = 0.5;
 export default ({threshold = 0.9} = {}) => ({
@@ -235,6 +237,8 @@ export default ({threshold = 0.9} = {}) => ({
 export function getTitle(record, targetSubfieldCodes = ['a', 'b', 'n', 'p']) {
   const [field] = record.get(/^245$/u);
   debugData(`titleField: ${JSON.stringify(field)}`);
+  debugDev(`targetSubfieldCodes: ${JSON.stringify(targetSubfieldCodes)}`);
+
 
   if (field) {
     const title = field.subfields
@@ -243,7 +247,8 @@ export function getTitle(record, targetSubfieldCodes = ['a', 'b', 'n', 'p']) {
       // Would be nice to normalize $n values...
       .map(({value}) => testStringOrNumber(value) ? String(value) : '')
       .join(' ')
-      .replace(/\[[^\]]*\]/ug, ' ') // Remove [whatever] stuff. ADD TEST FOR THIS
+      // Note: we do not want to removed stuff in brackets if its the only stuff in field ... (245 $a [Psi] / ... )
+      //.replace(/\[[^\]]*\]/ug, ' ') // Remove [whatever] stuff. ADD TEST FOR THIS
       .replace(/ [=\/:](?:$| )/ug, ' ') // Strip punctuation
       // Also đ vs d pairs seen:
       //.replace(/(?:đ|ȧ|t̄|ǩ|ǧ|s̆|c̆)/ug, '') // Hack. Saamelaisbibliografia has often dropped this wierd characters (oft old articles, no longer used in sami either)
@@ -252,11 +257,24 @@ export function getTitle(record, targetSubfieldCodes = ['a', 'b', 'n', 'p']) {
       .replace(/^ +/u, '')
       .replace(/ +$/u, '');
 
+    debugDev(`getTitle: title: ${title}`)
+
+    const cleanTitle = title.replace(/\[[^\]]*\]/ug, ' ') // Remove [whatever] stuff. ADD TEST FOR THIS
+      .replace(/ +/ug, ' ')
+      .replace(/^ +/u, '')
+      .replace(/ +$/u, '');
+
+    debugDev(`getTitle: cleanTitle: ${cleanTitle} (${cleanTitle.length})`);
+    const usableTitle = cleanTitle.length > 0 ? cleanTitle : title;
+    debugDev(`getTitle: usableTitle: ${usableTitle} (${usableTitle.length})`);
+    
+
     // Skip non-filing indicator (note that '9' is a magic indicator value, so we don't do it):
     if (/^[1-8]$/u.test(field.ind2)) { // Skip non-filing characters
-      return title.slice(parseInt(field.ind2));
+      const nonFilingTitle = usableTitle.slice(parseInt(field.ind2));
+      if (nonFilingTitle.length > 0) return nonFilingTitle;
     }
-    return title;
+    return usableTitle;
 
   }
   return false;