axyseo 2.1.8 → 2.1.10

package/build/languageProcessing/languages/zh/helpers/splitIntoTokensCustom.js

@@ -0,0 +1,41 @@
+/**
+ * Custom tokenizer for Chinese language.
+ * Chinese doesn't use spaces between words, so we split by characters
+ * while preserving punctuation and spaces as separate tokens.
+ *
+ * @param {string} text The text to tokenize.
+ *
+ * @returns {string[]} An array of tokens.
+ */
+export default function splitIntoTokensCustom(text) {
+  if (!text) {
+    return [];
+  }
+  const tokens = [];
+  const chineseCharRegex = /[\u4e00-\u9fff]/;
+  const whitespaceRegex = /\s/;
+  const punctuationRegex = /[.,;:!?()[\]{}'"]/;
+  for (let i = 0; i < text.length; i++) {
+    const char = text[i];
+    if (chineseCharRegex.test(char)) {
+      // Chinese character - add as individual token
+      tokens.push(char);
+    } else if (whitespaceRegex.test(char)) {
+      // Whitespace - skip
+      continue;
+    } else if (punctuationRegex.test(char)) {
+      // Punctuation - add as individual token
+      tokens.push(char);
+    } else {
+      // Other characters (like English letters, numbers) - group together
+      let word = char;
+      while (i + 1 < text.length && !chineseCharRegex.test(text[i + 1]) && !whitespaceRegex.test(text[i + 1]) && !punctuationRegex.test(text[i + 1])) {
+        i++;
+        word += text[i];
+      }
+      tokens.push(word);
+    }
+  }
+  return tokens.filter(token => token.trim() !== '');
+}
+//# sourceMappingURL=splitIntoTokensCustom.js.map

package/build/languageProcessing/languages/zh/helpers/splitIntoTokensCustom.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"splitIntoTokensCustom.js","names":["splitIntoTokensCustom","text","tokens","chineseCharRegex","whitespaceRegex","punctuationRegex","i","length","char","test","push","word","filter","token","trim"],"sources":["../../../../../src/languageProcessing/languages/zh/helpers/splitIntoTokensCustom.js"],"sourcesContent":["/**\n * Custom tokenizer for Chinese language.\n * Chinese doesn't use spaces between words, so we split by characters\n * while preserving punctuation and spaces as separate tokens.\n *\n * @param {string} text The text to tokenize.\n *\n * @returns {string[]} An array of tokens.\n */\nexport default function splitIntoTokensCustom(text) {\n  if (!text) {\n    return [];\n  }\n\n  const tokens = [];\n  const chineseCharRegex = /[\\u4e00-\\u9fff]/;\n  const whitespaceRegex = /\\s/;\n  const punctuationRegex = /[.,;:!?()[\\]{}'\"]/;\n\n  for (let i = 0; i < text.length; i++) {\n    const char = text[i];\n\n    if (chineseCharRegex.test(char)) {\n      // Chinese character - add as individual token\n      tokens.push(char);\n    } else if (whitespaceRegex.test(char)) {\n      // Whitespace - skip\n      continue;\n    } else if (punctuationRegex.test(char)) {\n      // Punctuation - add as individual token\n      tokens.push(char);\n    } else {\n      // Other characters (like English letters, numbers) - group together\n      let word = char;\n      while (\n        i + 1 < text.length &&\n        !chineseCharRegex.test(text[i + 1]) &&\n        !whitespaceRegex.test(text[i + 1]) &&\n        !punctuationRegex.test(text[i + 1])\n      ) {\n        i++;\n        word += text[i];\n      }\n      tokens.push(word);\n    }\n  }\n\n  return tokens.filter(token => token.trim() !== '');\n}\n"],"mappings":"AAAA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA,eAAe,SAASA,qBAAqBA,CAACC,IAAI,EAAE;EAClD,IAAI,CAACA,IAAI,EAAE;IACT,OAAO,EAAE;EACX;EAEA,MAAMC,MAAM,GAAG,EAAE;EACjB,MAAMC,gBAAgB,GAAG,iBAAiB;EAC1C,MAAMC,eAAe,GAAG,IAAI;EAC5B,MAAMC,gBAAgB,GAAG,mBAAmB;EAE5C,KAAK,IAAIC,CAAC,GAAG,CAAC,EAAEA,CAAC,GAAGL,IAAI,CAACM,MAAM,EAAED,CAAC,EAAE,EAAE;IACpC,MAAME,IAAI,GAAGP,IAAI,CAACK,CAAC,CAAC;IAEpB,IAAIH,gBAAgB,CAACM,IAAI,CAACD,IAAI,CAAC,EAAE;MAC/B;MACAN,MAAM,CAACQ,IAAI,CAACF,IAAI,CAAC;IACnB,CAAC,MAAM,IAAIJ,eAAe,CAACK,IAAI,CAACD,IAAI,CAAC,EAAE;MACrC;MACA;IACF,CAAC,MAAM,IAAIH,gBAAgB,CAACI,IAAI,CAACD,IAAI,CAAC,EAAE;MACtC;MACAN,MAAM,CAACQ,IAAI,CAACF,IAAI,CAAC;IACnB,CAAC,MAAM;MACL;MACA,IAAIG,IAAI,GAAGH,IAAI;MACf,OACEF,CAAC,GAAG,CAAC,GAAGL,IAAI,CAACM,MAAM,IACnB,CAACJ,gBAAgB,CAACM,IAAI,CAACR,IAAI,CAACK,CAAC,GAAG,CAAC,CAAC,CAAC,IACnC,CAACF,eAAe,CAACK,IAAI,CAACR,IAAI,CAACK,CAAC,GAAG,CAAC,CAAC,CAAC,IAClC,CAACD,gBAAgB,CAACI,IAAI,CAACR,IAAI,CAACK,CAAC,GAAG,CAAC,CAAC,CAAC,EACnC;QACAA,CAAC,EAAE;QACHK,IAAI,IAAIV,IAAI,CAACK,CAAC,CAAC;MACjB;MACAJ,MAAM,CAACQ,IAAI,CAACC,IAAI,CAAC;IACnB;EACF;EAEA,OAAOT,MAAM,CAACU,MAAM,CAACC,KAAK,IAAIA,KAAK,CAACC,IAAI,CAAC,CAAC,KAAK,EAAE,CAAC;AACpD","ignoreList":[]}

package/build/languageProcessing/researches/findKeywordInFirstParagraph.js

@@ -3,6 +3,7 @@ import { inRange, isEmpty } from 'lodash';
 import { findTopicFormsInString } from "../helpers/match/findKeywordFormsInString.js";
 import { getParentNode } from "../helpers/sentence/getSentencesFromTree";
 import { createShortcodeTagsRegex } from "../helpers";
+import isChineseText from "../helpers/language/isChineseText";
 
 /**
  * Checks if the introductory paragraph contains keyphrase or synonyms.
@@ -36,8 +37,29 @@ export default function (paper, researcher) {
   });
   const firstParagraph = paragraphs[0];
   const topicForms = researcher.getResearch('morphology');
-  const matchWordCustomHelper = researcher.getHelper('matchWordCustomHelper');
+  let matchWordCustomHelper = researcher.getHelper('matchWordCustomHelper');
   const locale = paper.getLocale();
+
+  // Auto-detect Chinese and use Chinese helper if not already available
+  if (!matchWordCustomHelper && isChineseText(paper.getKeyword() + ' ' + (firstParagraph ? firstParagraph.innerText() : ''))) {
+    // Use Chinese word matching for Chinese text
+    matchWordCustomHelper = function (text, word) {
+      const textToSearch = typeof text === 'string' ? text : text.text || text;
+      const matches = [];
+      if (!textToSearch || !word) {
+        return matches;
+      }
+      const lowerText = textToSearch.toLowerCase();
+      const lowerWord = word.toLowerCase();
+      let startIndex = 0;
+      let index;
+      while ((index = lowerText.indexOf(lowerWord, startIndex)) !== -1) {
+        matches.push(word);
+        startIndex = index + lowerWord.length;
+      }
+      return matches;
+    };
+  }
   const startOffset = firstParagraph && firstParagraph.sourceCodeLocation.startOffset;
   const mappedBlocks = paper._attributes.wpBlocks;
   const filteredIntroductionBlock = mappedBlocks && mappedBlocks.filter(block => inRange(startOffset, block.startOffset, block.endOffset))[0];

package/build/languageProcessing/researches/findKeywordInFirstParagraph.js.map

@@ -1 +1 @@
-{"version":3,"file":"findKeywordInFirstParagraph.js","names":["inRange","isEmpty","findTopicFormsInString","getParentNode","createShortcodeTagsRegex","paper","researcher","paragraphs","getResearch","filter","paragraph","parentNode","isImplicit","name","childNodes","test","value","firstParagraph","topicForms","matchWordCustomHelper","getHelper","locale","getLocale","startOffset","sourceCodeLocation","mappedBlocks","_attributes","wpBlocks","filteredIntroductionBlock","block","endOffset","result","foundInOneSentence","foundInParagraph","keyphraseOrSynonym","introduction","parentBlock","sentences","map","sentence","text","useSynonyms","firstResultSentence","find","resultSentence","percentWordMatches","resultParagraph","innerText"],"sources":["../../../src/languageProcessing/researches/findKeywordInFirstParagraph.js"],"sourcesContent":["/** @module analyses/findKeywordInFirstParagraph */\nimport {inRange, isEmpty} from 'lodash';\n\nimport {findTopicFormsInString} from '../helpers/match/findKeywordFormsInString.js';\nimport {getParentNode} from '../helpers/sentence/getSentencesFromTree';\nimport {createShortcodeTagsRegex} from '../helpers';\n\n/**\n * Checks if the introductory paragraph contains keyphrase or synonyms.\n * First splits the first paragraph by sentences. Finds the first paragraph which contains sentences e.g., not an image).\n * (1) Tries to find all (content) words from the keyphrase or a synonym phrase within one sentence.\n * If found all words within one sentence, returns an object with foundInOneSentence = true and keyphraseOrSynonym = \"keyphrase\"\n * or \"synonym\".\n * If it did not find all words within one sentence, goes ahead with matching the keyphrase with the entire first paragraph.\n * (2) Tries to find all (content) words from the keyphrase or a synonym phrase within the paragraph.\n * If found all words within the paragraph, returns an object with foundInOneSentence = false, foundInParagraph = true,\n * and keyphraseOrSynonym = \"keyphrase\" or \"synonym\".\n * If found not all words within the paragraph of nothing at all, returns an object with foundInOneSentence = false,\n * foundInParagraph = false, and keyphraseOrSynonym = \"\".\n *\n * @param {Paper} paper The text to check for paragraphs.\n * @param {Researcher} researcher The researcher to use for analysis.\n *\n * @returns {Object} Whether the keyphrase words were found in one sentence, whether the keyphrase words were found in\n * the paragraph, whether a keyphrase or a synonym phrase was matched.\n */\nexport default function(paper, researcher) {\n  let paragraphs = researcher.getResearch('getParagraphs');\n  // Filter captions from non-Classic editors.\n  paragraphs = paragraphs.filter(paragraph => {\n    const parentNode = getParentNode(paper, paragraph);\n    return !(paragraph.isImplicit && parentNode && parentNode.name === 'figcaption');\n  });\n  // Filter captions from Classic editor and from classic block inside Block editor.\n  paragraphs = paragraphs.filter(paragraph => {\n    return !(\n      paragraph.childNodes &&\n      paragraph.childNodes[0] &&\n      createShortcodeTagsRegex(['caption']).test(paragraph.childNodes[0].value)\n    );\n  });\n  const firstParagraph = paragraphs[0];\n\n  const topicForms = researcher.getResearch('morphology');\n  const matchWordCustomHelper = researcher.getHelper('matchWordCustomHelper');\n  const locale = paper.getLocale();\n  const startOffset = firstParagraph && firstParagraph.sourceCodeLocation.startOffset;\n\n  const mappedBlocks = paper._attributes.wpBlocks;\n  const filteredIntroductionBlock =\n    mappedBlocks &&\n    mappedBlocks.filter(block => inRange(startOffset, block.startOffset, block.endOffset))[0];\n  const result = {\n    foundInOneSentence: false,\n    foundInParagraph: false,\n    keyphraseOrSynonym: '',\n    introduction: firstParagraph,\n    parentBlock: filteredIntroductionBlock || null\n  };\n\n  if (isEmpty(firstParagraph)) {\n    return result;\n  }\n\n  const sentences = firstParagraph.sentences.map(sentence => sentence.text);\n  // Use both keyphrase and synonyms to match topic words in the first paragraph.\n  const useSynonyms = true;\n\n  if (!isEmpty(sentences)) {\n    const firstResultSentence = sentences\n      .map(sentence =>\n        findTopicFormsInString(topicForms, sentence, useSynonyms, locale, matchWordCustomHelper)\n      )\n      .find(resultSentence => resultSentence.percentWordMatches === 100);\n\n    if (firstResultSentence) {\n      result.foundInOneSentence = true;\n      result.foundInParagraph = true;\n      result.keyphraseOrSynonym = firstResultSentence.keyphraseOrSynonym;\n      return result;\n    }\n\n    const resultParagraph = findTopicFormsInString(\n      topicForms,\n      firstParagraph.innerText(),\n      useSynonyms,\n      locale,\n      matchWordCustomHelper\n    );\n    if (resultParagraph.percentWordMatches === 100) {\n      result.foundInParagraph = true;\n      result.keyphraseOrSynonym = resultParagraph.keyphraseOrSynonym;\n      return result;\n    }\n  }\n\n  return result;\n}\n"],"mappings":"AAAA;AACA,SAAQA,OAAO,EAAEC,OAAO,QAAO,QAAQ;AAEvC,SAAQC,sBAAsB;AAC9B,SAAQC,aAAa;AACrB,SAAQC,wBAAwB;;AAEhC;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA,eAAe,UAASC,KAAK,EAAEC,UAAU,EAAE;EACzC,IAAIC,UAAU,GAAGD,UAAU,CAACE,WAAW,CAAC,eAAe,CAAC;EACxD;EACAD,UAAU,GAAGA,UAAU,CAACE,MAAM,CAACC,SAAS,IAAI;IAC1C,MAAMC,UAAU,GAAGR,aAAa,CAACE,KAAK,EAAEK,SAAS,CAAC;IAClD,OAAO,EAAEA,SAAS,CAACE,UAAU,IAAID,UAAU,IAAIA,UAAU,CAACE,IAAI,KAAK,YAAY,CAAC;EAClF,CAAC,CAAC;EACF;EACAN,UAAU,GAAGA,UAAU,CAACE,MAAM,CAACC,SAAS,IAAI;IAC1C,OAAO,EACLA,SAAS,CAACI,UAAU,IACpBJ,SAAS,CAACI,UAAU,CAAC,CAAC,CAAC,IACvBV,wBAAwB,CAAC,CAAC,SAAS,CAAC,CAAC,CAACW,IAAI,CAACL,SAAS,CAACI,UAAU,CAAC,CAAC,CAAC,CAACE,KAAK,CAAC,CAC1E;EACH,CAAC,CAAC;EACF,MAAMC,cAAc,GAAGV,UAAU,CAAC,CAAC,CAAC;EAEpC,MAAMW,UAAU,GAAGZ,UAAU,CAACE,WAAW,CAAC,YAAY,CAAC;EACvD,MAAMW,qBAAqB,GAAGb,UAAU,CAACc,SAAS,CAAC,uBAAuB,CAAC;EAC3E,MAAMC,MAAM,GAAGhB,KAAK,CAACiB,SAAS,CAAC,CAAC;EAChC,MAAMC,WAAW,GAAGN,cAAc,IAAIA,cAAc,CAACO,kBAAkB,CAACD,WAAW;EAEnF,MAAME,YAAY,GAAGpB,KAAK,CAACqB,WAAW,CAACC,QAAQ;EAC/C,MAAMC,yBAAyB,GAC7BH,YAAY,IACZA,YAAY,CAAChB,MAAM,CAACoB,KAAK,IAAI7B,OAAO,CAACuB,WAAW,EAAEM,KAAK,CAACN,WAAW,EAAEM,KAAK,CAACC,SAAS,CAAC,CAAC,CAAC,CAAC,CAAC;EAC3F,MAAMC,MAAM,GAAG;IACbC,kBAAkB,EAAE,KAAK;IACzBC,gBAAgB,EAAE,KAAK;IACvBC,kBAAkB,EAAE,EAAE;IACtBC,YAAY,EAAElB,cAAc;IAC5BmB,WAAW,EAAER,yBAAyB,IAAI;EAC5C,CAAC;EAED,IAAI3B,OAAO,CAACgB,cAAc,CAAC,EAAE;IAC3B,OAAOc,MAAM;EACf;EAEA,MAAMM,SAAS,GAAGpB,cAAc,CAACoB,SAAS,CAACC,GAAG,CAACC,QAAQ,IAAIA,QAAQ,CAACC,IAAI,CAAC;EACzE;EACA,MAAMC,WAAW,GAAG,IAAI;EAExB,IAAI,CAACxC,OAAO,CAACoC,SAAS,CAAC,EAAE;IACvB,MAAMK,mBAAmB,GAAGL,SAAS,CAClCC,GAAG,CAACC,QAAQ,IACXrC,sBAAsB,CAACgB,UAAU,EAAEqB,QAAQ,EAAEE,WAAW,EAAEpB,MAAM,EAAEF,qBAAqB,CACzF,CAAC,CACAwB,IAAI,CAACC,cAAc,IAAIA,cAAc,CAACC,kBAAkB,KAAK,GAAG,CAAC;IAEpE,IAAIH,mBAAmB,EAAE;MACvBX,MAAM,CAACC,kBAAkB,GAAG,IAAI;MAChCD,MAAM,CAACE,gBAAgB,GAAG,IAAI;MAC9BF,MAAM,CAACG,kBAAkB,GAAGQ,mBAAmB,CAACR,kBAAkB;MAClE,OAAOH,MAAM;IACf;IAEA,MAAMe,eAAe,GAAG5C,sBAAsB,CAC5CgB,UAAU,EACVD,cAAc,CAAC8B,SAAS,CAAC,CAAC,EAC1BN,WAAW,EACXpB,MAAM,EACNF,qBACF,CAAC;IACD,IAAI2B,eAAe,CAACD,kBAAkB,KAAK,GAAG,EAAE;MAC9Cd,MAAM,CAACE,gBAAgB,GAAG,IAAI;MAC9BF,MAAM,CAACG,kBAAkB,GAAGY,eAAe,CAACZ,kBAAkB;MAC9D,OAAOH,MAAM;IACf;EACF;EAEA,OAAOA,MAAM;AACf","ignoreList":[]}
+{"version":3,"file":"findKeywordInFirstParagraph.js","names":["inRange","isEmpty","findTopicFormsInString","getParentNode","createShortcodeTagsRegex","isChineseText","paper","researcher","paragraphs","getResearch","filter","paragraph","parentNode","isImplicit","name","childNodes","test","value","firstParagraph","topicForms","matchWordCustomHelper","getHelper","locale","getLocale","getKeyword","innerText","text","word","textToSearch","matches","lowerText","toLowerCase","lowerWord","startIndex","index","indexOf","push","length","startOffset","sourceCodeLocation","mappedBlocks","_attributes","wpBlocks","filteredIntroductionBlock","block","endOffset","result","foundInOneSentence","foundInParagraph","keyphraseOrSynonym","introduction","parentBlock","sentences","map","sentence","useSynonyms","firstResultSentence","find","resultSentence","percentWordMatches","resultParagraph"],"sources":["../../../src/languageProcessing/researches/findKeywordInFirstParagraph.js"],"sourcesContent":["/** @module analyses/findKeywordInFirstParagraph */\nimport {inRange, isEmpty} from 'lodash';\n\nimport {findTopicFormsInString} from '../helpers/match/findKeywordFormsInString.js';\nimport {getParentNode} from '../helpers/sentence/getSentencesFromTree';\nimport {createShortcodeTagsRegex} from '../helpers';\nimport isChineseText from '../helpers/language/isChineseText';\n\n/**\n * Checks if the introductory paragraph contains keyphrase or synonyms.\n * First splits the first paragraph by sentences. Finds the first paragraph which contains sentences e.g., not an image).\n * (1) Tries to find all (content) words from the keyphrase or a synonym phrase within one sentence.\n * If found all words within one sentence, returns an object with foundInOneSentence = true and keyphraseOrSynonym = \"keyphrase\"\n * or \"synonym\".\n * If it did not find all words within one sentence, goes ahead with matching the keyphrase with the entire first paragraph.\n * (2) Tries to find all (content) words from the keyphrase or a synonym phrase within the paragraph.\n * If found all words within the paragraph, returns an object with foundInOneSentence = false, foundInParagraph = true,\n * and keyphraseOrSynonym = \"keyphrase\" or \"synonym\".\n * If found not all words within the paragraph of nothing at all, returns an object with foundInOneSentence = false,\n * foundInParagraph = false, and keyphraseOrSynonym = \"\".\n *\n * @param {Paper} paper The text to check for paragraphs.\n * @param {Researcher} researcher The researcher to use for analysis.\n *\n * @returns {Object} Whether the keyphrase words were found in one sentence, whether the keyphrase words were found in\n * the paragraph, whether a keyphrase or a synonym phrase was matched.\n */\nexport default function(paper, researcher) {\n  let paragraphs = researcher.getResearch('getParagraphs');\n  // Filter captions from non-Classic editors.\n  paragraphs = paragraphs.filter(paragraph => {\n    const parentNode = getParentNode(paper, paragraph);\n    return !(paragraph.isImplicit && parentNode && parentNode.name === 'figcaption');\n  });\n  // Filter captions from Classic editor and from classic block inside Block editor.\n  paragraphs = paragraphs.filter(paragraph => {\n    return !(\n      paragraph.childNodes &&\n      paragraph.childNodes[0] &&\n      createShortcodeTagsRegex(['caption']).test(paragraph.childNodes[0].value)\n    );\n  });\n  const firstParagraph = paragraphs[0];\n\n  const topicForms = researcher.getResearch('morphology');\n  let matchWordCustomHelper = researcher.getHelper('matchWordCustomHelper');\n  const locale = paper.getLocale();\n\n  // Auto-detect Chinese and use Chinese helper if not already available\n  if (\n    !matchWordCustomHelper &&\n    isChineseText(paper.getKeyword() + ' ' + (firstParagraph ? firstParagraph.innerText() : ''))\n  ) {\n    // Use Chinese word matching for Chinese text\n    matchWordCustomHelper = function(text, word) {\n      const textToSearch = typeof text === 'string' ? text : text.text || text;\n      const matches = [];\n\n      if (!textToSearch || !word) {\n        return matches;\n      }\n\n      const lowerText = textToSearch.toLowerCase();\n      const lowerWord = word.toLowerCase();\n\n      let startIndex = 0;\n      let index;\n\n      while ((index = lowerText.indexOf(lowerWord, startIndex)) !== -1) {\n        matches.push(word);\n        startIndex = index + lowerWord.length;\n      }\n\n      return matches;\n    };\n  }\n  const startOffset = firstParagraph && firstParagraph.sourceCodeLocation.startOffset;\n\n  const mappedBlocks = paper._attributes.wpBlocks;\n  const filteredIntroductionBlock =\n    mappedBlocks &&\n    mappedBlocks.filter(block => inRange(startOffset, block.startOffset, block.endOffset))[0];\n  const result = {\n    foundInOneSentence: false,\n    foundInParagraph: false,\n    keyphraseOrSynonym: '',\n    introduction: firstParagraph,\n    parentBlock: filteredIntroductionBlock || null\n  };\n\n  if (isEmpty(firstParagraph)) {\n    return result;\n  }\n\n  const sentences = firstParagraph.sentences.map(sentence => sentence.text);\n  // Use both keyphrase and synonyms to match topic words in the first paragraph.\n  const useSynonyms = true;\n\n  if (!isEmpty(sentences)) {\n    const firstResultSentence = sentences\n      .map(sentence =>\n        findTopicFormsInString(topicForms, sentence, useSynonyms, locale, matchWordCustomHelper)\n      )\n      .find(resultSentence => resultSentence.percentWordMatches === 100);\n\n    if (firstResultSentence) {\n      result.foundInOneSentence = true;\n      result.foundInParagraph = true;\n      result.keyphraseOrSynonym = firstResultSentence.keyphraseOrSynonym;\n      return result;\n    }\n\n    const resultParagraph = findTopicFormsInString(\n      topicForms,\n      firstParagraph.innerText(),\n      useSynonyms,\n      locale,\n      matchWordCustomHelper\n    );\n    if (resultParagraph.percentWordMatches === 100) {\n      result.foundInParagraph = true;\n      result.keyphraseOrSynonym = resultParagraph.keyphraseOrSynonym;\n      return result;\n    }\n  }\n\n  return result;\n}\n"],"mappings":"AAAA;AACA,SAAQA,OAAO,EAAEC,OAAO,QAAO,QAAQ;AAEvC,SAAQC,sBAAsB;AAC9B,SAAQC,aAAa;AACrB,SAAQC,wBAAwB;AAChC,OAAOC,aAAa;;AAEpB;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA,eAAe,UAASC,KAAK,EAAEC,UAAU,EAAE;EACzC,IAAIC,UAAU,GAAGD,UAAU,CAACE,WAAW,CAAC,eAAe,CAAC;EACxD;EACAD,UAAU,GAAGA,UAAU,CAACE,MAAM,CAACC,SAAS,IAAI;IAC1C,MAAMC,UAAU,GAAGT,aAAa,CAACG,KAAK,EAAEK,SAAS,CAAC;IAClD,OAAO,EAAEA,SAAS,CAACE,UAAU,IAAID,UAAU,IAAIA,UAAU,CAACE,IAAI,KAAK,YAAY,CAAC;EAClF,CAAC,CAAC;EACF;EACAN,UAAU,GAAGA,UAAU,CAACE,MAAM,CAACC,SAAS,IAAI;IAC1C,OAAO,EACLA,SAAS,CAACI,UAAU,IACpBJ,SAAS,CAACI,UAAU,CAAC,CAAC,CAAC,IACvBX,wBAAwB,CAAC,CAAC,SAAS,CAAC,CAAC,CAACY,IAAI,CAACL,SAAS,CAACI,UAAU,CAAC,CAAC,CAAC,CAACE,KAAK,CAAC,CAC1E;EACH,CAAC,CAAC;EACF,MAAMC,cAAc,GAAGV,UAAU,CAAC,CAAC,CAAC;EAEpC,MAAMW,UAAU,GAAGZ,UAAU,CAACE,WAAW,CAAC,YAAY,CAAC;EACvD,IAAIW,qBAAqB,GAAGb,UAAU,CAACc,SAAS,CAAC,uBAAuB,CAAC;EACzE,MAAMC,MAAM,GAAGhB,KAAK,CAACiB,SAAS,CAAC,CAAC;;EAEhC;EACA,IACE,CAACH,qBAAqB,IACtBf,aAAa,CAACC,KAAK,CAACkB,UAAU,CAAC,CAAC,GAAG,GAAG,IAAIN,cAAc,GAAGA,cAAc,CAACO,SAAS,CAAC,CAAC,GAAG,EAAE,CAAC,CAAC,EAC5F;IACA;IACAL,qBAAqB,GAAG,SAAAA,CAASM,IAAI,EAAEC,IAAI,EAAE;MAC3C,MAAMC,YAAY,GAAG,OAAOF,IAAI,KAAK,QAAQ,GAAGA,IAAI,GAAGA,IAAI,CAACA,IAAI,IAAIA,IAAI;MACxE,MAAMG,OAAO,GAAG,EAAE;MAElB,IAAI,CAACD,YAAY,IAAI,CAACD,IAAI,EAAE;QAC1B,OAAOE,OAAO;MAChB;MAEA,MAAMC,SAAS,GAAGF,YAAY,CAACG,WAAW,CAAC,CAAC;MAC5C,MAAMC,SAAS,GAAGL,IAAI,CAACI,WAAW,CAAC,CAAC;MAEpC,IAAIE,UAAU,GAAG,CAAC;MAClB,IAAIC,KAAK;MAET,OAAO,CAACA,KAAK,GAAGJ,SAAS,CAACK,OAAO,CAACH,SAAS,EAAEC,UAAU,CAAC,MAAM,CAAC,CAAC,EAAE;QAChEJ,OAAO,CAACO,IAAI,CAACT,IAAI,CAAC;QAClBM,UAAU,GAAGC,KAAK,GAAGF,SAAS,CAACK,MAAM;MACvC;MAEA,OAAOR,OAAO;IAChB,CAAC;EACH;EACA,MAAMS,WAAW,GAAGpB,cAAc,IAAIA,cAAc,CAACqB,kBAAkB,CAACD,WAAW;EAEnF,MAAME,YAAY,GAAGlC,KAAK,CAACmC,WAAW,CAACC,QAAQ;EAC/C,MAAMC,yBAAyB,GAC7BH,YAAY,IACZA,YAAY,CAAC9B,MAAM,CAACkC,KAAK,IAAI5C,OAAO,CAACsC,WAAW,EAAEM,KAAK,CAACN,WAAW,EAAEM,KAAK,CAACC,SAAS,CAAC,CAAC,CAAC,CAAC,CAAC;EAC3F,MAAMC,MAAM,GAAG;IACbC,kBAAkB,EAAE,KAAK;IACzBC,gBAAgB,EAAE,KAAK;IACvBC,kBAAkB,EAAE,EAAE;IACtBC,YAAY,EAAEhC,cAAc;IAC5BiC,WAAW,EAAER,yBAAyB,IAAI;EAC5C,CAAC;EAED,IAAI1C,OAAO,CAACiB,cAAc,CAAC,EAAE;IAC3B,OAAO4B,MAAM;EACf;EAEA,MAAMM,SAAS,GAAGlC,cAAc,CAACkC,SAAS,CAACC,GAAG,CAACC,QAAQ,IAAIA,QAAQ,CAAC5B,IAAI,CAAC;EACzE;EACA,MAAM6B,WAAW,GAAG,IAAI;EAExB,IAAI,CAACtD,OAAO,CAACmD,SAAS,CAAC,EAAE;IACvB,MAAMI,mBAAmB,GAAGJ,SAAS,CAClCC,GAAG,CAACC,QAAQ,IACXpD,sBAAsB,CAACiB,UAAU,EAAEmC,QAAQ,EAAEC,WAAW,EAAEjC,MAAM,EAAEF,qBAAqB,CACzF,CAAC,CACAqC,IAAI,CAACC,cAAc,IAAIA,cAAc,CAACC,kBAAkB,KAAK,GAAG,CAAC;IAEpE,IAAIH,mBAAmB,EAAE;MACvBV,MAAM,CAACC,kBAAkB,GAAG,IAAI;MAChCD,MAAM,CAACE,gBAAgB,GAAG,IAAI;MAC9BF,MAAM,CAACG,kBAAkB,GAAGO,mBAAmB,CAACP,kBAAkB;MAClE,OAAOH,MAAM;IACf;IAEA,MAAMc,eAAe,GAAG1D,sBAAsB,CAC5CiB,UAAU,EACVD,cAAc,CAACO,SAAS,CAAC,CAAC,EAC1B8B,WAAW,EACXjC,MAAM,EACNF,qBACF,CAAC;IACD,IAAIwC,eAAe,CAACD,kBAAkB,KAAK,GAAG,EAAE;MAC9Cb,MAAM,CAACE,gBAAgB,GAAG,IAAI;MAC9BF,MAAM,CAACG,kBAAkB,GAAGW,eAAe,CAACX,kBAAkB;MAC9D,OAAOH,MAAM;IACf;EACF;EAEA,OAAOA,MAAM;AACf","ignoreList":[]}

package/build/languageProcessing/researches/getAnchorsWithKeyphrase.js

@@ -1,4 +1,4 @@
-import { flatten, uniq } from "lodash";
+import { flatten, uniq } from 'lodash';
 import filterWordsFromArray from "../helpers/word/filterWordsFromArray";
 import { findTopicFormsInString } from "../helpers/match/findKeywordFormsInString";
 import getWords from "../helpers/word/getWords";
@@ -7,6 +7,7 @@ import parseSynonyms from "../helpers/sanitize/parseSynonyms";
 import processExactMatchRequest from "../helpers/match/processExactMatchRequest";
 import urlHelper from "../helpers/url/url.js";
 import { WORD_BOUNDARY_WITH_HYPHEN, WORD_BOUNDARY_WITHOUT_HYPHEN } from "../../config/wordBoundariesWithoutPunctuation";
+import { getMatchHelper } from "../helpers/language/chineseHelperFactory.js";
 let functionWords = [];
 
 /**
@@ -86,14 +87,14 @@ function getAnchorsWithSameTextAsTopic(anchors, topicForms, locale, customHelper
     const currentAnchorText = currentAnchor.innerText();
 
     /*
-          * For keyphrase matching, we want to split words on hyphens and en-dashes, except for languages where hyphens shouldn't
-          * be treated as word boundaries. Currently, the latter only applies to Indonesian, where hyphens are used to create plural forms of nouns,
-          * such as "buku-buku" being a plural form of "buku". We want to treat forms like "buku-buku" as one word, so we shouldn't
-          * split words on hyphens in Indonesian.
-          * For languages where hyphens are treated as word boundaries we pass a custom word boundary regex string to the getWords helper
-          * that includes whitespaces, hyphens (u002d), and en-dashes (u2013). Otherwise, we pass a word boundary regex that only includes
-          * whitespaces and en-dashes.
-          */
+     * For keyphrase matching, we want to split words on hyphens and en-dashes, except for languages where hyphens shouldn't
+     * be treated as word boundaries. Currently, the latter only applies to Indonesian, where hyphens are used to create plural forms of nouns,
+     * such as "buku-buku" being a plural form of "buku". We want to treat forms like "buku-buku" as one word, so we shouldn't
+     * split words on hyphens in Indonesian.
+     * For languages where hyphens are treated as word boundaries we pass a custom word boundary regex string to the getWords helper
+     * that includes whitespaces, hyphens (u002d), and en-dashes (u2013). Otherwise, we pass a word boundary regex that only includes
+     * whitespaces and en-dashes.
+     */
     let anchorWords;
     if (getWordsCustomHelper) {
       anchorWords = uniq(getWordsCustomHelper(currentAnchorText));
@@ -143,15 +144,15 @@ function getAnchorsWithSameTextAsTopic(anchors, topicForms, locale, customHelper
  * @returns {Object} The amount of anchor texts whose content words are the same as the keyphrase or synonyms' content words.
  */
 export default function (paper, researcher) {
-  functionWords = researcher.getConfig("functionWords");
-  const areHyphensWordBoundaries = researcher.getConfig("areHyphensWordBoundaries");
+  functionWords = researcher.getConfig('functionWords');
+  const areHyphensWordBoundaries = researcher.getConfig('areHyphensWordBoundaries');
   const result = {
     anchorsWithKeyphrase: [],
     anchorsWithKeyphraseCount: 0
   };
   // STEP 1.
   // If the paper's text is empty, return empty result.
-  if (paper.getText() === "") {
+  if (paper.getText() === '') {
     return result;
   }
 
@@ -162,7 +163,7 @@ export default function (paper, researcher) {
    * This is a conscious decision where we won't assess the paper if the keyphrase is not set.
    * This includes a case where only the synonym is set but not the keyphrase.
    */
-  if (keyphrase === "") {
+  if (keyphrase === '') {
     return result;
   }
   /*
@@ -173,7 +174,7 @@ export default function (paper, researcher) {
   originalTopics.push(keyphrase);
 
   // Retrieve the anchors.
-  let anchors = paper.getTree().findAll(treeNode => treeNode.name === "a");
+  let anchors = paper.getTree().findAll(treeNode => treeNode.name === 'a');
   /*
    * We get the site's URL (e.g., https://yoast.com) or domain (e.g., yoast.com) from the paper.
    * In case of WordPress, the variable is a URL. In case of Shopify, it is a domain.
@@ -188,10 +189,14 @@ export default function (paper, researcher) {
     return result;
   }
   const locale = paper.getLocale();
-  const topicForms = researcher.getResearch("morphology");
+  const topicForms = researcher.getResearch('morphology');
+
+  // Auto-detect Chinese and enhance helper if needed
+  const text = paper.getText();
+  const enhancedMatchHelper = getMatchHelper(text, keyphrase, researcher.getHelper('matchWordCustomHelper'));
   const customHelpers = {
-    matchWordCustomHelper: researcher.getHelper("matchWordCustomHelper"),
-    getWordsCustomHelper: researcher.getHelper("getWordsCustomHelper")
+    matchWordCustomHelper: enhancedMatchHelper,
+    getWordsCustomHelper: researcher.getHelper('getWordsCustomHelper')
   };
 
   // STEP 4.

package/build/languageProcessing/researches/getAnchorsWithKeyphrase.js.map

@@ -1 +1 @@
-{"version":3,"file":"getAnchorsWithKeyphrase.js","names":["flatten","uniq","filterWordsFromArray","findTopicFormsInString","getWords","matchTextWithArray","parseSynonyms","processExactMatchRequest","urlHelper","WORD_BOUNDARY_WITH_HYPHEN","WORD_BOUNDARY_WITHOUT_HYPHEN","functionWords","isLinkingToSelf","anchorLink","siteUrlOrDomain","Boolean","areEqual","isRelativeFragmentURL","getAnchorsLinkingToSelf","anchors","anchorsLinkingToSelf","map","anchor","attributes","href","filter","index","getAnchorsContainingTopic","topicForms","locale","matchWordCustomHelper","anchorsContainingTopic","anchorText","innerText","percentWordMatches","getAnchorsWithSameTextAsTopic","customHelpers","exactMatchRequest","areHyphensWordBoundaries","getWordsCustomHelper","keyphraseAndSynonymsWords","keyphraseForms","synonymsForms","forEach","form","push","anchorsContainedInTopic","currentAnchor","currentAnchorText","anchorWords","filteredAnchorWords","length","request","exactMatchRequested","every","anchorWord","keyphrase","includes","i","topicForm","count","paper","researcher","getConfig","result","anchorsWithKeyphrase","anchorsWithKeyphraseCount","getText","getKeyword","originalTopics","getSynonyms","getTree","findAll","treeNode","name","getPermalink","getLocale","getResearch","getHelper","isExactMatchRequested","originalTopic"],"sources":["../../../src/languageProcessing/researches/getAnchorsWithKeyphrase.js"],"sourcesContent":["import { flatten, uniq } from \"lodash\";\nimport filterWordsFromArray from \"../helpers/word/filterWordsFromArray\";\nimport { findTopicFormsInString } from \"../helpers/match/findKeywordFormsInString\";\nimport getWords from \"../helpers/word/getWords\";\nimport matchTextWithArray from \"../helpers/match/matchTextWithArray\";\nimport parseSynonyms from \"../helpers/sanitize/parseSynonyms\";\nimport processExactMatchRequest from \"../helpers/match/processExactMatchRequest\";\nimport urlHelper from \"../helpers/url/url.js\";\nimport { WORD_BOUNDARY_WITH_HYPHEN, WORD_BOUNDARY_WITHOUT_HYPHEN } from \"../../config/wordBoundariesWithoutPunctuation\";\n\nlet functionWords = [];\n\n/**\n * Checks whether the anchor's link is a relative fragment or the same as the site url/domain.\n * Relative fragment links always point to the page itself.\n *\n * @param {String} anchorLink       The link anchor.\n * @param {String} siteUrlOrDomain  The site URL or domain of the paper.\n *\n * @returns {boolean} Whether the anchor's link is a relative fragment or the same as the site url/domain.\n */\nfunction isLinkingToSelf( anchorLink, siteUrlOrDomain ) {\n\treturn Boolean( urlHelper.areEqual( anchorLink, siteUrlOrDomain ) || urlHelper.isRelativeFragmentURL( anchorLink ) );\n}\n\n/**\n * Gets the anchors whose url is not linking at the current site url/domain.\n *\n * @param {Array}   anchors         An array with all anchors from the paper.\n * @param {String}  siteUrlOrDomain The site URL or domain of the paper.\n *\n * @returns {Array} The array of all anchors whose url is not linking at the current site url/domain.\n */\nfunction getAnchorsLinkingToSelf( anchors, siteUrlOrDomain ) {\n\tconst anchorsLinkingToSelf = anchors.map( function( anchor ) {\n\t\tconst anchorLink = anchor.attributes.href;\n\t\t// Return false if there is no href attribute.\n\t\treturn anchorLink ? isLinkingToSelf( anchorLink, siteUrlOrDomain ) : false;\n\t} );\n\n\treturn  anchors.filter( ( anchor, index ) => ! anchorsLinkingToSelf[ index ] );\n}\n\n/**\n * Gets the anchors with text that contains all content words of the topic (i.e. keyphrase or synonyms).\n *\n * @param {Array}       anchors         An array with all anchors from the paper\n * @param {Object}      topicForms      The object with topicForms. It contains all forms of the keyphrase and synonyms.\n * @param {String}      locale          The locale of the paper.\n * @param {Function}    matchWordCustomHelper The helper function to match word in text.\n *\n * @returns {String[]} The array of all anchors with text that contains all content words of the keyphrase or synonyms.\n */\nfunction getAnchorsContainingTopic( anchors, topicForms, locale, matchWordCustomHelper ) {\n\tconst anchorsContainingTopic = anchors.map( function( anchor ) {\n\t\t// Only retrieve the anchor's text. This is because we only use the anchor text for the following check.\n\t\tconst anchorText = anchor.innerText();\n\t\treturn findTopicFormsInString( topicForms, anchorText, true, locale, matchWordCustomHelper  ).percentWordMatches === 100;\n\t} );\n\n\treturn anchors.filter( ( anchor, index ) => anchorsContainingTopic[ index ] );\n}\n\n/**\n * Gets the anchors with text that has the same content words as the keyphrase or synonyms.\n *\n * @param {Array}       anchors             \t\tAn array with all anchors from the paper.\n * @param {Object}      topicForms          \t\tThe object with topicForms. It contains all forms of the keyphrase and synonyms.\n * @param {string}      locale              \t\tThe locale of the paper.\n * @param {Object}      customHelpers       \t\tAn object containing custom helpers.\n * @param {Object[]}    exactMatchRequest   \t\tAn array of objects containing the keyphrase and information on whether\n * \t\t\t\t\t\t\t\t\t\t\t\t\tthe exact match has been requested.\n * @param {boolean}\t \tareHyphensWordBoundaries\tWhether hyphens should be treated as word boundaries.\n *\n * @returns {Array} The array of all anchors with text that has the same content words as the keyphrase/synonyms.\n */\nfunction getAnchorsWithSameTextAsTopic( anchors, topicForms, locale, customHelpers, exactMatchRequest, areHyphensWordBoundaries  ) {\n\tconst matchWordCustomHelper = customHelpers.matchWordCustomHelper;\n\tconst getWordsCustomHelper = customHelpers.getWordsCustomHelper;\n\n\t// Prepare keyphrase and synonym forms for comparison with anchors.\n\tconst keyphraseAndSynonymsWords = [ flatten( topicForms.keyphraseForms ) ];\n\tconst synonymsForms = topicForms.synonymsForms;\n\tsynonymsForms.forEach( form => keyphraseAndSynonymsWords.push( flatten( form ) ) );\n\n\t// The variable that will save all the anchors with text that has the same content words as the keyphrase/synonyms.\n\tconst anchorsContainedInTopic = [];\n\n\tanchors.forEach( function( currentAnchor ) {\n\t\tconst currentAnchorText = currentAnchor.innerText();\n\n\t\t/*\n        * For keyphrase matching, we want to split words on hyphens and en-dashes, except for languages where hyphens shouldn't\n        * be treated as word boundaries. Currently, the latter only applies to Indonesian, where hyphens are used to create plural forms of nouns,\n        * such as \"buku-buku\" being a plural form of \"buku\". We want to treat forms like \"buku-buku\" as one word, so we shouldn't\n        * split words on hyphens in Indonesian.\n        * For languages where hyphens are treated as word boundaries we pass a custom word boundary regex string to the getWords helper\n        * that includes whitespaces, hyphens (u002d), and en-dashes (u2013). Otherwise, we pass a word boundary regex that only includes\n        * whitespaces and en-dashes.\n        */\n\t\tlet anchorWords;\n\t\tif ( getWordsCustomHelper ) {\n\t\t\tanchorWords = uniq( getWordsCustomHelper( currentAnchorText ) );\n\t\t} else if ( areHyphensWordBoundaries ) {\n\t\t\tanchorWords = uniq( getWords( currentAnchorText, WORD_BOUNDARY_WITH_HYPHEN ) );\n\t\t} else {\n\t\t\tanchorWords = uniq( getWords( currentAnchorText, WORD_BOUNDARY_WITHOUT_HYPHEN ) );\n\t\t}\n\n\t\t/*\n\t\t * Filter function words out of the anchor text.\n\t\t * If the anchor text contains only function words, we keep them.\n\t\t */\n\t\tconst filteredAnchorWords = filterWordsFromArray( anchorWords, functionWords );\n\t\tif ( filteredAnchorWords.length > 0 ) {\n\t\t\tanchorWords = filteredAnchorWords;\n\t\t}\n\n\t\texactMatchRequest.forEach( request => {\n\t\t\t/*\n\t\t\t * Check a) if the exact match is requested for the keyphrase, and\n\t\t\t * b) if every content word in the anchor text is included in the keyphrase or synonym.\n\t\t\t */\n\t\t\tif ( request.exactMatchRequested &&\n\t\t\t\tanchorWords.every( anchorWord => request.keyphrase.includes( anchorWord ) ) ) {\n\t\t\t\tanchorsContainedInTopic.push( true );\n\t\t\t}\n\t\t} );\n\n\t\t// Check if every word in the anchor text is also present in the keyphrase/synonym.\n\t\tfor ( let i = 0; i < keyphraseAndSynonymsWords.length; i++ ) {\n\t\t\tconst topicForm =  keyphraseAndSynonymsWords[ i ];\n\n\t\t\tif ( anchorWords.every( anchorWord => matchTextWithArray( anchorWord, topicForm, locale, matchWordCustomHelper ).count > 0 ) ) {\n\t\t\t\tanchorsContainedInTopic.push( true );\n\t\t\t\tbreak;\n\t\t\t}\n\t\t}\n\t} );\n\n\treturn anchors.filter( ( anchor, index ) => anchorsContainedInTopic[ index ] );\n}\n\n/**\n * Checks whether the content words of the anchor text are the same as the content words of the keyphrase or synonym.\n * Also includes different word forms if the morphology is available.\n *\n * @param {Paper}       paper       The paper to research.\n * @param {Researcher}  researcher  The researcher to use.\n *\n * @returns {Object} The amount of anchor texts whose content words are the same as the keyphrase or synonyms' content words.\n */\nexport default function( paper, researcher ) {\n\tfunctionWords = researcher.getConfig( \"functionWords\" );\n\n\tconst areHyphensWordBoundaries = researcher.getConfig( \"areHyphensWordBoundaries\" );\n\n\tconst result = {\n\t\tanchorsWithKeyphrase: [],\n\t\tanchorsWithKeyphraseCount: 0,\n\t};\n\t// STEP 1.\n\t// If the paper's text is empty, return empty result.\n\tif ( paper.getText() === \"\" ) {\n\t\treturn result;\n\t}\n\n\t// STEP 2.\n\tconst keyphrase = paper.getKeyword();\n\t/*\n\t * If no keyphrase is set, return empty result.\n\t * This is a conscious decision where we won't assess the paper if the keyphrase is not set.\n\t * This includes a case where only the synonym is set but not the keyphrase.\n\t */\n\tif ( keyphrase === \"\" ) {\n\t\treturn result;\n\t}\n\t/*\n\t * When the keyphrase is set, also retrieve the synonyms and save them in \"topics\" array.\n\t * Eventually, the term topics here refers to either keyphrase or synonyms.\n\t */\n\tconst originalTopics = parseSynonyms( paper.getSynonyms() );\n\toriginalTopics.push( keyphrase );\n\n\t// Retrieve the anchors.\n\tlet anchors = paper.getTree().findAll( treeNode => treeNode.name === \"a\" );\n\t/*\n\t * We get the site's URL (e.g., https://yoast.com) or domain (e.g., yoast.com) from the paper.\n\t * In case of WordPress, the variable is a URL. In case of Shopify, it is a domain.\n\t */\n\tconst siteUrlOrDomain = paper.getPermalink();\n\n\t// STEP 3.\n\t// Get the anchors with urls that are not linking to the current site url/domain.\n\tanchors = getAnchorsLinkingToSelf( anchors, siteUrlOrDomain );\n\t// If all anchor urls are linking to the current site url/domain, return empty result.\n\tif ( anchors.length === 0 ) {\n\t\treturn result;\n\t}\n\n\tconst locale = paper.getLocale();\n\tconst topicForms = researcher.getResearch( \"morphology\" );\n\tconst customHelpers = {\n\t\tmatchWordCustomHelper: researcher.getHelper( \"matchWordCustomHelper\" ),\n\t\tgetWordsCustomHelper: researcher.getHelper( \"getWordsCustomHelper\" ),\n\t};\n\n\t// STEP 4.\n\t// Get the anchors with text that contains the keyphrase/synonyms' content words.\n\tanchors = getAnchorsContainingTopic( anchors, topicForms, locale, customHelpers.matchWordCustomHelper );\n\t// If all anchor texts do not contain the keyphrase/synonyms' content words, return empty result.\n\tif ( anchors.length === 0 ) {\n\t\treturn result;\n\t}\n\n\t// STEP 5.\n\t// Check if exact match is requested for every topic (keyphrase or synonym).\n\tconst isExactMatchRequested = originalTopics.map( originalTopic => processExactMatchRequest( originalTopic ) );\n\t// Get the anchors with text that has the same content words as the keyphrase/synonyms.\n\tanchors = getAnchorsWithSameTextAsTopic( anchors, topicForms, locale, customHelpers, isExactMatchRequested, areHyphensWordBoundaries );\n\n\treturn {\n\t\tanchorsWithKeyphrase: anchors,\n\t\tanchorsWithKeyphraseCount: anchors.length,\n\t};\n}\n\n"],"mappings":"AAAA,SAASA,OAAO,EAAEC,IAAI,QAAQ,QAAQ;AACtC,OAAOC,oBAAoB;AAC3B,SAASC,sBAAsB;AAC/B,OAAOC,QAAQ;AACf,OAAOC,kBAAkB;AACzB,OAAOC,aAAa;AACpB,OAAOC,wBAAwB;AAC/B,OAAOC,SAAS;AAChB,SAASC,yBAAyB,EAAEC,4BAA4B;AAEhE,IAAIC,aAAa,GAAG,EAAE;;AAEtB;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA,SAASC,eAAeA,CAAEC,UAAU,EAAEC,eAAe,EAAG;EACvD,OAAOC,OAAO,CAAEP,SAAS,CAACQ,QAAQ,CAAEH,UAAU,EAAEC,eAAgB,CAAC,IAAIN,SAAS,CAACS,qBAAqB,CAAEJ,UAAW,CAAE,CAAC;AACrH;;AAEA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA,SAASK,uBAAuBA,CAAEC,OAAO,EAAEL,eAAe,EAAG;EAC5D,MAAMM,oBAAoB,GAAGD,OAAO,CAACE,GAAG,CAAE,UAAUC,MAAM,EAAG;IAC5D,MAAMT,UAAU,GAAGS,MAAM,CAACC,UAAU,CAACC,IAAI;IACzC;IACA,OAAOX,UAAU,GAAGD,eAAe,CAAEC,UAAU,EAAEC,eAAgB,CAAC,GAAG,KAAK;EAC3E,CAAE,CAAC;EAEH,OAAQK,OAAO,CAACM,MAAM,CAAE,CAAEH,MAAM,EAAEI,KAAK,KAAM,CAAEN,oBAAoB,CAAEM,KAAK,CAAG,CAAC;AAC/E;;AAEA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA,SAASC,yBAAyBA,CAAER,OAAO,EAAES,UAAU,EAAEC,MAAM,EAAEC,qBAAqB,EAAG;EACxF,MAAMC,sBAAsB,GAAGZ,OAAO,CAACE,GAAG,CAAE,UAAUC,MAAM,EAAG;IAC9D;IACA,MAAMU,UAAU,GAAGV,MAAM,CAACW,SAAS,CAAC,CAAC;IACrC,OAAO9B,sBAAsB,CAAEyB,UAAU,EAAEI,UAAU,EAAE,IAAI,EAAEH,MAAM,EAAEC,qBAAuB,CAAC,CAACI,kBAAkB,KAAK,GAAG;EACzH,CAAE,CAAC;EAEH,OAAOf,OAAO,CAACM,MAAM,CAAE,CAAEH,MAAM,EAAEI,KAAK,KAAMK,sBAAsB,CAAEL,KAAK,CAAG,CAAC;AAC9E;;AAEA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA,SAASS,6BAA6BA,CAAEhB,OAAO,EAAES,UAAU,EAAEC,MAAM,EAAEO,aAAa,EAAEC,iBAAiB,EAAEC,wBAAwB,EAAI;EAClI,MAAMR,qBAAqB,GAAGM,aAAa,CAACN,qBAAqB;EACjE,MAAMS,oBAAoB,GAAGH,aAAa,CAACG,oBAAoB;;EAE/D;EACA,MAAMC,yBAAyB,GAAG,CAAExC,OAAO,CAAE4B,UAAU,CAACa,cAAe,CAAC,CAAE;EAC1E,MAAMC,aAAa,GAAGd,UAAU,CAACc,aAAa;EAC9CA,aAAa,CAACC,OAAO,CAAEC,IAAI,IAAIJ,yBAAyB,CAACK,IAAI,CAAE7C,OAAO,CAAE4C,IAAK,CAAE,CAAE,CAAC;;EAElF;EACA,MAAME,uBAAuB,GAAG,EAAE;EAElC3B,OAAO,CAACwB,OAAO,CAAE,UAAUI,aAAa,EAAG;IAC1C,MAAMC,iBAAiB,GAAGD,aAAa,CAACd,SAAS,CAAC,CAAC;;IAEnD;AACF;AACA;AACA;AACA;AACA;AACA;AACA;AACA;IACE,IAAIgB,WAAW;IACf,IAAKV,oBAAoB,EAAG;MAC3BU,WAAW,GAAGhD,IAAI,CAAEsC,oBAAoB,CAAES,iBAAkB,CAAE,CAAC;IAChE,CAAC,MAAM,IAAKV,wBAAwB,EAAG;MACtCW,WAAW,GAAGhD,IAAI,CAAEG,QAAQ,CAAE4C,iBAAiB,EAAEvC,yBAA0B,CAAE,CAAC;IAC/E,CAAC,MAAM;MACNwC,WAAW,GAAGhD,IAAI,CAAEG,QAAQ,CAAE4C,iBAAiB,EAAEtC,4BAA6B,CAAE,CAAC;IAClF;;IAEA;AACF;AACA;AACA;IACE,MAAMwC,mBAAmB,GAAGhD,oBAAoB,CAAE+C,WAAW,EAAEtC,aAAc,CAAC;IAC9E,IAAKuC,mBAAmB,CAACC,MAAM,GAAG,CAAC,EAAG;MACrCF,WAAW,GAAGC,mBAAmB;IAClC;IAEAb,iBAAiB,CAACM,OAAO,CAAES,OAAO,IAAI;MACrC;AACH;AACA;AACA;MACG,IAAKA,OAAO,CAACC,mBAAmB,IAC/BJ,WAAW,CAACK,KAAK,CAAEC,UAAU,IAAIH,OAAO,CAACI,SAAS,CAACC,QAAQ,CAAEF,UAAW,CAAE,CAAC,EAAG;QAC9ET,uBAAuB,CAACD,IAAI,CAAE,IAAK,CAAC;MACrC;IACD,CAAE,CAAC;;IAEH;IACA,KAAM,IAAIa,CAAC,GAAG,CAAC,EAAEA,CAAC,GAAGlB,yBAAyB,CAACW,MAAM,EAAEO,CAAC,EAAE,EAAG;MAC5D,MAAMC,SAAS,GAAInB,yBAAyB,CAAEkB,CAAC,CAAE;MAEjD,IAAKT,WAAW,CAACK,KAAK,CAAEC,UAAU,IAAIlD,kBAAkB,CAAEkD,UAAU,EAAEI,SAAS,EAAE9B,MAAM,EAAEC,qBAAsB,CAAC,CAAC8B,KAAK,GAAG,CAAE,CAAC,EAAG;QAC9Hd,uBAAuB,CAACD,IAAI,CAAE,IAAK,CAAC;QACpC;MACD;IACD;EACD,CAAE,CAAC;EAEH,OAAO1B,OAAO,CAACM,MAAM,CAAE,CAAEH,MAAM,EAAEI,KAAK,KAAMoB,uBAAuB,CAAEpB,KAAK,CAAG,CAAC;AAC/E;;AAEA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA,eAAe,UAAUmC,KAAK,EAAEC,UAAU,EAAG;EAC5CnD,aAAa,GAAGmD,UAAU,CAACC,SAAS,CAAE,eAAgB,CAAC;EAEvD,MAAMzB,wBAAwB,GAAGwB,UAAU,CAACC,SAAS,CAAE,0BAA2B,CAAC;EAEnF,MAAMC,MAAM,GAAG;IACdC,oBAAoB,EAAE,EAAE;IACxBC,yBAAyB,EAAE;EAC5B,CAAC;EACD;EACA;EACA,IAAKL,KAAK,CAACM,OAAO,CAAC,CAAC,KAAK,EAAE,EAAG;IAC7B,OAAOH,MAAM;EACd;;EAEA;EACA,MAAMR,SAAS,GAAGK,KAAK,CAACO,UAAU,CAAC,CAAC;EACpC;AACD;AACA;AACA;AACA;EACC,IAAKZ,SAAS,KAAK,EAAE,EAAG;IACvB,OAAOQ,MAAM;EACd;EACA;AACD;AACA;AACA;EACC,MAAMK,cAAc,GAAG/D,aAAa,CAAEuD,KAAK,CAACS,WAAW,CAAC,CAAE,CAAC;EAC3DD,cAAc,CAACxB,IAAI,CAAEW,SAAU,CAAC;;EAEhC;EACA,IAAIrC,OAAO,GAAG0C,KAAK,CAACU,OAAO,CAAC,CAAC,CAACC,OAAO,CAAEC,QAAQ,IAAIA,QAAQ,CAACC,IAAI,KAAK,GAAI,CAAC;EAC1E;AACD;AACA;AACA;EACC,MAAM5D,eAAe,GAAG+C,KAAK,CAACc,YAAY,CAAC,CAAC;;EAE5C;EACA;EACAxD,OAAO,GAAGD,uBAAuB,CAAEC,OAAO,EAAEL,eAAgB,CAAC;EAC7D;EACA,IAAKK,OAAO,CAACgC,MAAM,KAAK,CAAC,EAAG;IAC3B,OAAOa,MAAM;EACd;EAEA,MAAMnC,MAAM,GAAGgC,KAAK,CAACe,SAAS,CAAC,CAAC;EAChC,MAAMhD,UAAU,GAAGkC,UAAU,CAACe,WAAW,CAAE,YAAa,CAAC;EACzD,MAAMzC,aAAa,GAAG;IACrBN,qBAAqB,EAAEgC,UAAU,CAACgB,SAAS,CAAE,uBAAwB,CAAC;IACtEvC,oBAAoB,EAAEuB,UAAU,CAACgB,SAAS,CAAE,sBAAuB;EACpE,CAAC;;EAED;EACA;EACA3D,OAAO,GAAGQ,yBAAyB,CAAER,OAAO,EAAES,UAAU,EAAEC,MAAM,EAAEO,aAAa,CAACN,qBAAsB,CAAC;EACvG;EACA,IAAKX,OAAO,CAACgC,MAAM,KAAK,CAAC,EAAG;IAC3B,OAAOa,MAAM;EACd;;EAEA;EACA;EACA,MAAMe,qBAAqB,GAAGV,cAAc,CAAChD,GAAG,CAAE2D,aAAa,IAAIzE,wBAAwB,CAAEyE,aAAc,CAAE,CAAC;EAC9G;EACA7D,OAAO,GAAGgB,6BAA6B,CAAEhB,OAAO,EAAES,UAAU,EAAEC,MAAM,EAAEO,aAAa,EAAE2C,qBAAqB,EAAEzC,wBAAyB,CAAC;EAEtI,OAAO;IACN2B,oBAAoB,EAAE9C,OAAO;IAC7B+C,yBAAyB,EAAE/C,OAAO,CAACgC;EACpC,CAAC;AACF","ignoreList":[]}
+{"version":3,"file":"getAnchorsWithKeyphrase.js","names":["flatten","uniq","filterWordsFromArray","findTopicFormsInString","getWords","matchTextWithArray","parseSynonyms","processExactMatchRequest","urlHelper","WORD_BOUNDARY_WITH_HYPHEN","WORD_BOUNDARY_WITHOUT_HYPHEN","getMatchHelper","functionWords","isLinkingToSelf","anchorLink","siteUrlOrDomain","Boolean","areEqual","isRelativeFragmentURL","getAnchorsLinkingToSelf","anchors","anchorsLinkingToSelf","map","anchor","attributes","href","filter","index","getAnchorsContainingTopic","topicForms","locale","matchWordCustomHelper","anchorsContainingTopic","anchorText","innerText","percentWordMatches","getAnchorsWithSameTextAsTopic","customHelpers","exactMatchRequest","areHyphensWordBoundaries","getWordsCustomHelper","keyphraseAndSynonymsWords","keyphraseForms","synonymsForms","forEach","form","push","anchorsContainedInTopic","currentAnchor","currentAnchorText","anchorWords","filteredAnchorWords","length","request","exactMatchRequested","every","anchorWord","keyphrase","includes","i","topicForm","count","paper","researcher","getConfig","result","anchorsWithKeyphrase","anchorsWithKeyphraseCount","getText","getKeyword","originalTopics","getSynonyms","getTree","findAll","treeNode","name","getPermalink","getLocale","getResearch","text","enhancedMatchHelper","getHelper","isExactMatchRequested","originalTopic"],"sources":["../../../src/languageProcessing/researches/getAnchorsWithKeyphrase.js"],"sourcesContent":["import {flatten, uniq} from 'lodash';\nimport filterWordsFromArray from '../helpers/word/filterWordsFromArray';\nimport {findTopicFormsInString} from '../helpers/match/findKeywordFormsInString';\nimport getWords from '../helpers/word/getWords';\nimport matchTextWithArray from '../helpers/match/matchTextWithArray';\nimport parseSynonyms from '../helpers/sanitize/parseSynonyms';\nimport processExactMatchRequest from '../helpers/match/processExactMatchRequest';\nimport urlHelper from '../helpers/url/url.js';\nimport {\n  WORD_BOUNDARY_WITH_HYPHEN,\n  WORD_BOUNDARY_WITHOUT_HYPHEN\n} from '../../config/wordBoundariesWithoutPunctuation';\nimport {getMatchHelper} from '../helpers/language/chineseHelperFactory.js';\n\nlet functionWords = [];\n\n/**\n * Checks whether the anchor's link is a relative fragment or the same as the site url/domain.\n * Relative fragment links always point to the page itself.\n *\n * @param {String} anchorLink       The link anchor.\n * @param {String} siteUrlOrDomain  The site URL or domain of the paper.\n *\n * @returns {boolean} Whether the anchor's link is a relative fragment or the same as the site url/domain.\n */\nfunction isLinkingToSelf(anchorLink, siteUrlOrDomain) {\n  return Boolean(\n    urlHelper.areEqual(anchorLink, siteUrlOrDomain) || urlHelper.isRelativeFragmentURL(anchorLink)\n  );\n}\n\n/**\n * Gets the anchors whose url is not linking at the current site url/domain.\n *\n * @param {Array}   anchors         An array with all anchors from the paper.\n * @param {String}  siteUrlOrDomain The site URL or domain of the paper.\n *\n * @returns {Array} The array of all anchors whose url is not linking at the current site url/domain.\n */\nfunction getAnchorsLinkingToSelf(anchors, siteUrlOrDomain) {\n  const anchorsLinkingToSelf = anchors.map(function(anchor) {\n    const anchorLink = anchor.attributes.href;\n    // Return false if there is no href attribute.\n    return anchorLink ? isLinkingToSelf(anchorLink, siteUrlOrDomain) : false;\n  });\n\n  return anchors.filter((anchor, index) => !anchorsLinkingToSelf[index]);\n}\n\n/**\n * Gets the anchors with text that contains all content words of the topic (i.e. keyphrase or synonyms).\n *\n * @param {Array}       anchors         An array with all anchors from the paper\n * @param {Object}      topicForms      The object with topicForms. It contains all forms of the keyphrase and synonyms.\n * @param {String}      locale          The locale of the paper.\n * @param {Function}    matchWordCustomHelper The helper function to match word in text.\n *\n * @returns {String[]} The array of all anchors with text that contains all content words of the keyphrase or synonyms.\n */\nfunction getAnchorsContainingTopic(anchors, topicForms, locale, matchWordCustomHelper) {\n  const anchorsContainingTopic = anchors.map(function(anchor) {\n    // Only retrieve the anchor's text. This is because we only use the anchor text for the following check.\n    const anchorText = anchor.innerText();\n    return (\n      findTopicFormsInString(topicForms, anchorText, true, locale, matchWordCustomHelper)\n        .percentWordMatches === 100\n    );\n  });\n\n  return anchors.filter((anchor, index) => anchorsContainingTopic[index]);\n}\n\n/**\n * Gets the anchors with text that has the same content words as the keyphrase or synonyms.\n *\n * @param {Array}       anchors             \t\tAn array with all anchors from the paper.\n * @param {Object}      topicForms          \t\tThe object with topicForms. It contains all forms of the keyphrase and synonyms.\n * @param {string}      locale              \t\tThe locale of the paper.\n * @param {Object}      customHelpers       \t\tAn object containing custom helpers.\n * @param {Object[]}    exactMatchRequest   \t\tAn array of objects containing the keyphrase and information on whether\n * \t\t\t\t\t\t\t\t\t\t\t\t\tthe exact match has been requested.\n * @param {boolean}\t \tareHyphensWordBoundaries\tWhether hyphens should be treated as word boundaries.\n *\n * @returns {Array} The array of all anchors with text that has the same content words as the keyphrase/synonyms.\n */\nfunction getAnchorsWithSameTextAsTopic(\n  anchors,\n  topicForms,\n  locale,\n  customHelpers,\n  exactMatchRequest,\n  areHyphensWordBoundaries\n) {\n  const matchWordCustomHelper = customHelpers.matchWordCustomHelper;\n  const getWordsCustomHelper = customHelpers.getWordsCustomHelper;\n\n  // Prepare keyphrase and synonym forms for comparison with anchors.\n  const keyphraseAndSynonymsWords = [flatten(topicForms.keyphraseForms)];\n  const synonymsForms = topicForms.synonymsForms;\n  synonymsForms.forEach(form => keyphraseAndSynonymsWords.push(flatten(form)));\n\n  // The variable that will save all the anchors with text that has the same content words as the keyphrase/synonyms.\n  const anchorsContainedInTopic = [];\n\n  anchors.forEach(function(currentAnchor) {\n    const currentAnchorText = currentAnchor.innerText();\n\n    /*\n     * For keyphrase matching, we want to split words on hyphens and en-dashes, except for languages where hyphens shouldn't\n     * be treated as word boundaries. Currently, the latter only applies to Indonesian, where hyphens are used to create plural forms of nouns,\n     * such as \"buku-buku\" being a plural form of \"buku\". We want to treat forms like \"buku-buku\" as one word, so we shouldn't\n     * split words on hyphens in Indonesian.\n     * For languages where hyphens are treated as word boundaries we pass a custom word boundary regex string to the getWords helper\n     * that includes whitespaces, hyphens (u002d), and en-dashes (u2013). Otherwise, we pass a word boundary regex that only includes\n     * whitespaces and en-dashes.\n     */\n    let anchorWords;\n    if (getWordsCustomHelper) {\n      anchorWords = uniq(getWordsCustomHelper(currentAnchorText));\n    } else if (areHyphensWordBoundaries) {\n      anchorWords = uniq(getWords(currentAnchorText, WORD_BOUNDARY_WITH_HYPHEN));\n    } else {\n      anchorWords = uniq(getWords(currentAnchorText, WORD_BOUNDARY_WITHOUT_HYPHEN));\n    }\n\n    /*\n     * Filter function words out of the anchor text.\n     * If the anchor text contains only function words, we keep them.\n     */\n    const filteredAnchorWords = filterWordsFromArray(anchorWords, functionWords);\n    if (filteredAnchorWords.length > 0) {\n      anchorWords = filteredAnchorWords;\n    }\n\n    exactMatchRequest.forEach(request => {\n      /*\n       * Check a) if the exact match is requested for the keyphrase, and\n       * b) if every content word in the anchor text is included in the keyphrase or synonym.\n       */\n      if (\n        request.exactMatchRequested &&\n        anchorWords.every(anchorWord => request.keyphrase.includes(anchorWord))\n      ) {\n        anchorsContainedInTopic.push(true);\n      }\n    });\n\n    // Check if every word in the anchor text is also present in the keyphrase/synonym.\n    for (let i = 0; i < keyphraseAndSynonymsWords.length; i++) {\n      const topicForm = keyphraseAndSynonymsWords[i];\n\n      if (\n        anchorWords.every(\n          anchorWord =>\n            matchTextWithArray(anchorWord, topicForm, locale, matchWordCustomHelper).count > 0\n        )\n      ) {\n        anchorsContainedInTopic.push(true);\n        break;\n      }\n    }\n  });\n\n  return anchors.filter((anchor, index) => anchorsContainedInTopic[index]);\n}\n\n/**\n * Checks whether the content words of the anchor text are the same as the content words of the keyphrase or synonym.\n * Also includes different word forms if the morphology is available.\n *\n * @param {Paper}       paper       The paper to research.\n * @param {Researcher}  researcher  The researcher to use.\n *\n * @returns {Object} The amount of anchor texts whose content words are the same as the keyphrase or synonyms' content words.\n */\nexport default function(paper, researcher) {\n  functionWords = researcher.getConfig('functionWords');\n\n  const areHyphensWordBoundaries = researcher.getConfig('areHyphensWordBoundaries');\n\n  const result = {\n    anchorsWithKeyphrase: [],\n    anchorsWithKeyphraseCount: 0\n  };\n  // STEP 1.\n  // If the paper's text is empty, return empty result.\n  if (paper.getText() === '') {\n    return result;\n  }\n\n  // STEP 2.\n  const keyphrase = paper.getKeyword();\n  /*\n   * If no keyphrase is set, return empty result.\n   * This is a conscious decision where we won't assess the paper if the keyphrase is not set.\n   * This includes a case where only the synonym is set but not the keyphrase.\n   */\n  if (keyphrase === '') {\n    return result;\n  }\n  /*\n   * When the keyphrase is set, also retrieve the synonyms and save them in \"topics\" array.\n   * Eventually, the term topics here refers to either keyphrase or synonyms.\n   */\n  const originalTopics = parseSynonyms(paper.getSynonyms());\n  originalTopics.push(keyphrase);\n\n  // Retrieve the anchors.\n  let anchors = paper.getTree().findAll(treeNode => treeNode.name === 'a');\n  /*\n   * We get the site's URL (e.g., https://yoast.com) or domain (e.g., yoast.com) from the paper.\n   * In case of WordPress, the variable is a URL. In case of Shopify, it is a domain.\n   */\n  const siteUrlOrDomain = paper.getPermalink();\n\n  // STEP 3.\n  // Get the anchors with urls that are not linking to the current site url/domain.\n  anchors = getAnchorsLinkingToSelf(anchors, siteUrlOrDomain);\n  // If all anchor urls are linking to the current site url/domain, return empty result.\n  if (anchors.length === 0) {\n    return result;\n  }\n\n  const locale = paper.getLocale();\n  const topicForms = researcher.getResearch('morphology');\n\n  // Auto-detect Chinese and enhance helper if needed\n  const text = paper.getText();\n  const enhancedMatchHelper = getMatchHelper(\n    text,\n    keyphrase,\n    researcher.getHelper('matchWordCustomHelper')\n  );\n\n  const customHelpers = {\n    matchWordCustomHelper: enhancedMatchHelper,\n    getWordsCustomHelper: researcher.getHelper('getWordsCustomHelper')\n  };\n\n  // STEP 4.\n  // Get the anchors with text that contains the keyphrase/synonyms' content words.\n  anchors = getAnchorsContainingTopic(\n    anchors,\n    topicForms,\n    locale,\n    customHelpers.matchWordCustomHelper\n  );\n  // If all anchor texts do not contain the keyphrase/synonyms' content words, return empty result.\n  if (anchors.length === 0) {\n    return result;\n  }\n\n  // STEP 5.\n  // Check if exact match is requested for every topic (keyphrase or synonym).\n  const isExactMatchRequested = originalTopics.map(originalTopic =>\n    processExactMatchRequest(originalTopic)\n  );\n  // Get the anchors with text that has the same content words as the keyphrase/synonyms.\n  anchors = getAnchorsWithSameTextAsTopic(\n    anchors,\n    topicForms,\n    locale,\n    customHelpers,\n    isExactMatchRequested,\n    areHyphensWordBoundaries\n  );\n\n  return {\n    anchorsWithKeyphrase: anchors,\n    anchorsWithKeyphraseCount: anchors.length\n  };\n}\n"],"mappings":"AAAA,SAAQA,OAAO,EAAEC,IAAI,QAAO,QAAQ;AACpC,OAAOC,oBAAoB;AAC3B,SAAQC,sBAAsB;AAC9B,OAAOC,QAAQ;AACf,OAAOC,kBAAkB;AACzB,OAAOC,aAAa;AACpB,OAAOC,wBAAwB;AAC/B,OAAOC,SAAS;AAChB,SACEC,yBAAyB,EACzBC,4BAA4B;AAE9B,SAAQC,cAAc;AAEtB,IAAIC,aAAa,GAAG,EAAE;;AAEtB;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA,SAASC,eAAeA,CAACC,UAAU,EAAEC,eAAe,EAAE;EACpD,OAAOC,OAAO,CACZR,SAAS,CAACS,QAAQ,CAACH,UAAU,EAAEC,eAAe,CAAC,IAAIP,SAAS,CAACU,qBAAqB,CAACJ,UAAU,CAC/F,CAAC;AACH;;AAEA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA,SAASK,uBAAuBA,CAACC,OAAO,EAAEL,eAAe,EAAE;EACzD,MAAMM,oBAAoB,GAAGD,OAAO,CAACE,GAAG,CAAC,UAASC,MAAM,EAAE;IACxD,MAAMT,UAAU,GAAGS,MAAM,CAACC,UAAU,CAACC,IAAI;IACzC;IACA,OAAOX,UAAU,GAAGD,eAAe,CAACC,UAAU,EAAEC,eAAe,CAAC,GAAG,KAAK;EAC1E,CAAC,CAAC;EAEF,OAAOK,OAAO,CAACM,MAAM,CAAC,CAACH,MAAM,EAAEI,KAAK,KAAK,CAACN,oBAAoB,CAACM,KAAK,CAAC,CAAC;AACxE;;AAEA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA,SAASC,yBAAyBA,CAACR,OAAO,EAAES,UAAU,EAAEC,MAAM,EAAEC,qBAAqB,EAAE;EACrF,MAAMC,sBAAsB,GAAGZ,OAAO,CAACE,GAAG,CAAC,UAASC,MAAM,EAAE;IAC1D;IACA,MAAMU,UAAU,GAAGV,MAAM,CAACW,SAAS,CAAC,CAAC;IACrC,OACE/B,sBAAsB,CAAC0B,UAAU,EAAEI,UAAU,EAAE,IAAI,EAAEH,MAAM,EAAEC,qBAAqB,CAAC,CAChFI,kBAAkB,KAAK,GAAG;EAEjC,CAAC,CAAC;EAEF,OAAOf,OAAO,CAACM,MAAM,CAAC,CAACH,MAAM,EAAEI,KAAK,KAAKK,sBAAsB,CAACL,KAAK,CAAC,CAAC;AACzE;;AAEA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA,SAASS,6BAA6BA,CACpChB,OAAO,EACPS,UAAU,EACVC,MAAM,EACNO,aAAa,EACbC,iBAAiB,EACjBC,wBAAwB,EACxB;EACA,MAAMR,qBAAqB,GAAGM,aAAa,CAACN,qBAAqB;EACjE,MAAMS,oBAAoB,GAAGH,aAAa,CAACG,oBAAoB;;EAE/D;EACA,MAAMC,yBAAyB,GAAG,CAACzC,OAAO,CAAC6B,UAAU,CAACa,cAAc,CAAC,CAAC;EACtE,MAAMC,aAAa,GAAGd,UAAU,CAACc,aAAa;EAC9CA,aAAa,CAACC,OAAO,CAACC,IAAI,IAAIJ,yBAAyB,CAACK,IAAI,CAAC9C,OAAO,CAAC6C,IAAI,CAAC,CAAC,CAAC;;EAE5E;EACA,MAAME,uBAAuB,GAAG,EAAE;EAElC3B,OAAO,CAACwB,OAAO,CAAC,UAASI,aAAa,EAAE;IACtC,MAAMC,iBAAiB,GAAGD,aAAa,CAACd,SAAS,CAAC,CAAC;;IAEnD;AACJ;AACA;AACA;AACA;AACA;AACA;AACA;AACA;IACI,IAAIgB,WAAW;IACf,IAAIV,oBAAoB,EAAE;MACxBU,WAAW,GAAGjD,IAAI,CAACuC,oBAAoB,CAACS,iBAAiB,CAAC,CAAC;IAC7D,CAAC,MAAM,IAAIV,wBAAwB,EAAE;MACnCW,WAAW,GAAGjD,IAAI,CAACG,QAAQ,CAAC6C,iBAAiB,EAAExC,yBAAyB,CAAC,CAAC;IAC5E,CAAC,MAAM;MACLyC,WAAW,GAAGjD,IAAI,CAACG,QAAQ,CAAC6C,iBAAiB,EAAEvC,4BAA4B,CAAC,CAAC;IAC/E;;IAEA;AACJ;AACA;AACA;IACI,MAAMyC,mBAAmB,GAAGjD,oBAAoB,CAACgD,WAAW,EAAEtC,aAAa,CAAC;IAC5E,IAAIuC,mBAAmB,CAACC,MAAM,GAAG,CAAC,EAAE;MAClCF,WAAW,GAAGC,mBAAmB;IACnC;IAEAb,iBAAiB,CAACM,OAAO,CAACS,OAAO,IAAI;MACnC;AACN;AACA;AACA;MACM,IACEA,OAAO,CAACC,mBAAmB,IAC3BJ,WAAW,CAACK,KAAK,CAACC,UAAU,IAAIH,OAAO,CAACI,SAAS,CAACC,QAAQ,CAACF,UAAU,CAAC,CAAC,EACvE;QACAT,uBAAuB,CAACD,IAAI,CAAC,IAAI,CAAC;MACpC;IACF,CAAC,CAAC;;IAEF;IACA,KAAK,IAAIa,CAAC,GAAG,CAAC,EAAEA,CAAC,GAAGlB,yBAAyB,CAACW,MAAM,EAAEO,CAAC,EAAE,EAAE;MACzD,MAAMC,SAAS,GAAGnB,yBAAyB,CAACkB,CAAC,CAAC;MAE9C,IACET,WAAW,CAACK,KAAK,CACfC,UAAU,IACRnD,kBAAkB,CAACmD,UAAU,EAAEI,SAAS,EAAE9B,MAAM,EAAEC,qBAAqB,CAAC,CAAC8B,KAAK,GAAG,CACrF,CAAC,EACD;QACAd,uBAAuB,CAACD,IAAI,CAAC,IAAI,CAAC;QAClC;MACF;IACF;EACF,CAAC,CAAC;EAEF,OAAO1B,OAAO,CAACM,MAAM,CAAC,CAACH,MAAM,EAAEI,KAAK,KAAKoB,uBAAuB,CAACpB,KAAK,CAAC,CAAC;AAC1E;;AAEA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA,eAAe,UAASmC,KAAK,EAAEC,UAAU,EAAE;EACzCnD,aAAa,GAAGmD,UAAU,CAACC,SAAS,CAAC,eAAe,CAAC;EAErD,MAAMzB,wBAAwB,GAAGwB,UAAU,CAACC,SAAS,CAAC,0BAA0B,CAAC;EAEjF,MAAMC,MAAM,GAAG;IACbC,oBAAoB,EAAE,EAAE;IACxBC,yBAAyB,EAAE;EAC7B,CAAC;EACD;EACA;EACA,IAAIL,KAAK,CAACM,OAAO,CAAC,CAAC,KAAK,EAAE,EAAE;IAC1B,OAAOH,MAAM;EACf;;EAEA;EACA,MAAMR,SAAS,GAAGK,KAAK,CAACO,UAAU,CAAC,CAAC;EACpC;AACF;AACA;AACA;AACA;EACE,IAAIZ,SAAS,KAAK,EAAE,EAAE;IACpB,OAAOQ,MAAM;EACf;EACA;AACF;AACA;AACA;EACE,MAAMK,cAAc,GAAGhE,aAAa,CAACwD,KAAK,CAACS,WAAW,CAAC,CAAC,CAAC;EACzDD,cAAc,CAACxB,IAAI,CAACW,SAAS,CAAC;;EAE9B;EACA,IAAIrC,OAAO,GAAG0C,KAAK,CAACU,OAAO,CAAC,CAAC,CAACC,OAAO,CAACC,QAAQ,IAAIA,QAAQ,CAACC,IAAI,KAAK,GAAG,CAAC;EACxE;AACF;AACA;AACA;EACE,MAAM5D,eAAe,GAAG+C,KAAK,CAACc,YAAY,CAAC,CAAC;;EAE5C;EACA;EACAxD,OAAO,GAAGD,uBAAuB,CAACC,OAAO,EAAEL,eAAe,CAAC;EAC3D;EACA,IAAIK,OAAO,CAACgC,MAAM,KAAK,CAAC,EAAE;IACxB,OAAOa,MAAM;EACf;EAEA,MAAMnC,MAAM,GAAGgC,KAAK,CAACe,SAAS,CAAC,CAAC;EAChC,MAAMhD,UAAU,GAAGkC,UAAU,CAACe,WAAW,CAAC,YAAY,CAAC;;EAEvD;EACA,MAAMC,IAAI,GAAGjB,KAAK,CAACM,OAAO,CAAC,CAAC;EAC5B,MAAMY,mBAAmB,GAAGrE,cAAc,CACxCoE,IAAI,EACJtB,SAAS,EACTM,UAAU,CAACkB,SAAS,CAAC,uBAAuB,CAC9C,CAAC;EAED,MAAM5C,aAAa,GAAG;IACpBN,qBAAqB,EAAEiD,mBAAmB;IAC1CxC,oBAAoB,EAAEuB,UAAU,CAACkB,SAAS,CAAC,sBAAsB;EACnE,CAAC;;EAED;EACA;EACA7D,OAAO,GAAGQ,yBAAyB,CACjCR,OAAO,EACPS,UAAU,EACVC,MAAM,EACNO,aAAa,CAACN,qBAChB,CAAC;EACD;EACA,IAAIX,OAAO,CAACgC,MAAM,KAAK,CAAC,EAAE;IACxB,OAAOa,MAAM;EACf;;EAEA;EACA;EACA,MAAMiB,qBAAqB,GAAGZ,cAAc,CAAChD,GAAG,CAAC6D,aAAa,IAC5D5E,wBAAwB,CAAC4E,aAAa,CACxC,CAAC;EACD;EACA/D,OAAO,GAAGgB,6BAA6B,CACrChB,OAAO,EACPS,UAAU,EACVC,MAAM,EACNO,aAAa,EACb6C,qBAAqB,EACrB3C,wBACF,CAAC;EAED,OAAO;IACL2B,oBAAoB,EAAE9C,OAAO;IAC7B+C,yBAAyB,EAAE/C,OAAO,CAACgC;EACrC,CAAC;AACH","ignoreList":[]}

package/build/languageProcessing/researches/getParagraphs.js

@@ -1,4 +1,4 @@
-import { reject } from "lodash";
+import { reject } from 'lodash';
 
 /**
  * Returns all paragraphs in a given Paper.
@@ -8,11 +8,20 @@ import { reject } from "lodash";
  * @returns {Paragraph[]} All paragraphs in the paper.
  */
 export default function (paper) {
-  let paragraphs = paper.getTree().findAll(node => node.name === "p");
+  let paragraphs = paper.getTree().findAll(node => node.name === 'p');
 
-  // Remove empty paragraphs without sentences and paragraphs only consisting of links.
+  // Remove empty paragraphs without sentences and paragraphs only consisting of empty links.
   paragraphs = reject(paragraphs, paragraph => paragraph.sentences.length === 0);
-  paragraphs = reject(paragraphs, paragraph => paragraph.childNodes.every(node => node.name === "a"));
+  // Don't remove paragraphs with links that contain meaningful text
+  paragraphs = reject(paragraphs, paragraph => {
+    // Only reject if ALL children are links AND those links are empty/meaningless
+    const allLinks = paragraph.childNodes.every(node => node.name === 'a');
+    if (!allLinks) return false;
+
+    // Check if the paragraph has meaningful text content
+    const textContent = paragraph.innerText().trim();
+    return textContent.length === 0; // Only reject if completely empty
+  });
   return paragraphs;
 }
 //# sourceMappingURL=getParagraphs.js.map

package/build/languageProcessing/researches/getParagraphs.js.map

@@ -1 +1 @@
-{"version":3,"file":"getParagraphs.js","names":["reject","paper","paragraphs","getTree","findAll","node","name","paragraph","sentences","length","childNodes","every"],"sources":["../../../src/languageProcessing/researches/getParagraphs.js"],"sourcesContent":["import { reject } from \"lodash\";\n\n/**\n * Returns all paragraphs in a given Paper.\n * Remove paragraphs that do not contain sentences or only consist of links.\n *\n * @param {Paper} paper The current paper.\n * @returns {Paragraph[]} All paragraphs in the paper.\n */\nexport default function( paper ) {\n\tlet paragraphs = paper.getTree().findAll( node => node.name === \"p\" );\n\n\t// Remove empty paragraphs without sentences and paragraphs only consisting of links.\n\tparagraphs = reject( paragraphs, paragraph => paragraph.sentences.length === 0 );\n\tparagraphs = reject( paragraphs, paragraph => paragraph.childNodes.every( node => node.name === \"a\" ) );\n\n\treturn paragraphs;\n}\n"],"mappings":"AAAA,SAASA,MAAM,QAAQ,QAAQ;;AAE/B;AACA;AACA;AACA;AACA;AACA;AACA;AACA,eAAe,UAAUC,KAAK,EAAG;EAChC,IAAIC,UAAU,GAAGD,KAAK,CAACE,OAAO,CAAC,CAAC,CAACC,OAAO,CAAEC,IAAI,IAAIA,IAAI,CAACC,IAAI,KAAK,GAAI,CAAC;;EAErE;EACAJ,UAAU,GAAGF,MAAM,CAAEE,UAAU,EAAEK,SAAS,IAAIA,SAAS,CAACC,SAAS,CAACC,MAAM,KAAK,CAAE,CAAC;EAChFP,UAAU,GAAGF,MAAM,CAAEE,UAAU,EAAEK,SAAS,IAAIA,SAAS,CAACG,UAAU,CAACC,KAAK,CAAEN,IAAI,IAAIA,IAAI,CAACC,IAAI,KAAK,GAAI,CAAE,CAAC;EAEvG,OAAOJ,UAAU;AAClB","ignoreList":[]}
+{"version":3,"file":"getParagraphs.js","names":["reject","paper","paragraphs","getTree","findAll","node","name","paragraph","sentences","length","allLinks","childNodes","every","textContent","innerText","trim"],"sources":["../../../src/languageProcessing/researches/getParagraphs.js"],"sourcesContent":["import {reject} from 'lodash';\n\n/**\n * Returns all paragraphs in a given Paper.\n * Remove paragraphs that do not contain sentences or only consist of links.\n *\n * @param {Paper} paper The current paper.\n * @returns {Paragraph[]} All paragraphs in the paper.\n */\nexport default function(paper) {\n  let paragraphs = paper.getTree().findAll(node => node.name === 'p');\n\n  // Remove empty paragraphs without sentences and paragraphs only consisting of empty links.\n  paragraphs = reject(paragraphs, paragraph => paragraph.sentences.length === 0);\n  // Don't remove paragraphs with links that contain meaningful text\n  paragraphs = reject(paragraphs, paragraph => {\n    // Only reject if ALL children are links AND those links are empty/meaningless\n    const allLinks = paragraph.childNodes.every(node => node.name === 'a');\n    if (!allLinks) return false;\n\n    // Check if the paragraph has meaningful text content\n    const textContent = paragraph.innerText().trim();\n    return textContent.length === 0; // Only reject if completely empty\n  });\n\n  return paragraphs;\n}\n"],"mappings":"AAAA,SAAQA,MAAM,QAAO,QAAQ;;AAE7B;AACA;AACA;AACA;AACA;AACA;AACA;AACA,eAAe,UAASC,KAAK,EAAE;EAC7B,IAAIC,UAAU,GAAGD,KAAK,CAACE,OAAO,CAAC,CAAC,CAACC,OAAO,CAACC,IAAI,IAAIA,IAAI,CAACC,IAAI,KAAK,GAAG,CAAC;;EAEnE;EACAJ,UAAU,GAAGF,MAAM,CAACE,UAAU,EAAEK,SAAS,IAAIA,SAAS,CAACC,SAAS,CAACC,MAAM,KAAK,CAAC,CAAC;EAC9E;EACAP,UAAU,GAAGF,MAAM,CAACE,UAAU,EAAEK,SAAS,IAAI;IAC3C;IACA,MAAMG,QAAQ,GAAGH,SAAS,CAACI,UAAU,CAACC,KAAK,CAACP,IAAI,IAAIA,IAAI,CAACC,IAAI,KAAK,GAAG,CAAC;IACtE,IAAI,CAACI,QAAQ,EAAE,OAAO,KAAK;;IAE3B;IACA,MAAMG,WAAW,GAAGN,SAAS,CAACO,SAAS,CAAC,CAAC,CAACC,IAAI,CAAC,CAAC;IAChD,OAAOF,WAAW,CAACJ,MAAM,KAAK,CAAC,CAAC,CAAC;EACnC,CAAC,CAAC;EAEF,OAAOP,UAAU;AACnB","ignoreList":[]}

package/build/languageProcessing/researches/keywordCount.js

@@ -7,6 +7,7 @@ import isDoubleQuoted from "../helpers/match/isDoubleQuoted";
 import { markWordsInASentence } from "../helpers/word/markWordsInSentences";
 import getSentences from "../helpers/sentence/getSentences";
 import { filterShortcodesFromHTML } from "../helpers";
+import isChineseText from "../helpers/language/isChineseText";
 
 /**
  * Counts the occurrences of the keyphrase in the text and creates the Mark objects for the matches.
@@ -86,12 +87,39 @@ export default function getKeyphraseCount(paper, researcher) {
   if (keyphraseLength === 0) {
     return result;
   }
-  const matchWordCustomHelper = researcher.getHelper('matchWordCustomHelper');
+  let matchWordCustomHelper = researcher.getHelper('matchWordCustomHelper');
   const customSentenceTokenizer = researcher.getHelper('memoizedTokenizer');
   const customSplitIntoTokensHelper = researcher.getHelper('splitIntoTokensCustom');
   const locale = paper.getLocale();
   const text = matchWordCustomHelper ? filterShortcodesFromHTML(paper.getText(), paper._attributes && paper._attributes.shortcodes) : paper.getText();
 
+  // Auto-detect Chinese and use Chinese helper if not already available
+  if (!matchWordCustomHelper && isChineseText(paper.getKeyword() + ' ' + text)) {
+    // Import Chinese helper for Chinese text
+    try {
+      // Use simple Chinese word matching for Chinese text
+      matchWordCustomHelper = function (sentence, word) {
+        const text = typeof sentence === 'string' ? sentence : sentence.text || sentence;
+        const matches = [];
+        if (!text || !word) {
+          return matches;
+        }
+        const lowerText = text.toLowerCase();
+        const lowerWord = word.toLowerCase();
+        let startIndex = 0;
+        let index;
+        while ((index = lowerText.indexOf(lowerWord, startIndex)) !== -1) {
+          matches.push(word);
+          startIndex = index + lowerWord.length;
+        }
+        return matches;
+      };
+    } catch (error) {
+      // If there's any issue, continue with default behavior
+      console.warn('Error setting up Chinese matching, using default:', error.message);
+    }
+  }
+
   // When the custom helper is available, we're using the sentences retrieved from the text for the analysis.
   const sentences = matchWordCustomHelper ? getSentences(text, customSentenceTokenizer) : getSentencesFromTree(paper);
   // Exact matching is requested when the keyphrase is enclosed in double quotes.

package/build/languageProcessing/researches/keywordCount.js.map

@@ -1 +1 @@
-{"version":3,"file":"keywordCount.js","names":["flatten","flattenDeep","getSentencesFromTree","normalizeSingle","getMarkingsInSentence","matchWordFormsWithSentence","isDoubleQuoted","markWordsInASentence","getSentences","filterShortcodesFromHTML","countKeyphraseInText","sentences","keyphraseForms","locale","matchWordCustomHelper","isExactMatchRequested","customSplitIntoTokensHelper","result","count","markings","forEach","sentence","matchesInSentence","map","wordForms","isEachWordFound","every","counts","match","totalMatchCount","Math","min","foundWords","matches","push","getKeyphraseCount","paper","researcher","keyphraseLength","topicForms","getResearch","length","word","form","getHelper","customSentenceTokenizer","getLocale","text","getText","_attributes","shortcodes","getKeyword","keyphraseFound","keywordCount","console","warn"],"sources":["../../../src/languageProcessing/researches/keywordCount.js"],"sourcesContent":["import {flatten, flattenDeep} from 'lodash';\nimport getSentencesFromTree from '../helpers/sentence/getSentencesFromTree';\nimport {normalizeSingle} from '../helpers/sanitize/quotes';\nimport getMarkingsInSentence from '../helpers/highlighting/getMarkingsInSentence';\nimport matchWordFormsWithSentence from '../helpers/match/matchWordFormsWithSentence';\nimport isDoubleQuoted from '../helpers/match/isDoubleQuoted';\nimport {markWordsInASentence} from '../helpers/word/markWordsInSentences';\nimport getSentences from '../helpers/sentence/getSentences';\nimport {filterShortcodesFromHTML} from '../helpers';\n\n/**\n * Counts the occurrences of the keyphrase in the text and creates the Mark objects for the matches.\n *\n * @param {(Sentence|string)[]}\tsentences\t\t\tThe sentences to check.\n * @param {Array}\t\tkeyphraseForms\t\t\t\tThe keyphrase forms.\n * @param {string}\t\tlocale\t\t\t\t\t\tThe locale used in the analysis.\n * @param {function}\tmatchWordCustomHelper\t\tA custom helper to match words with a text.\n * @param {boolean}\t\tisExactMatchRequested\t\tWhether the exact matching is requested.\n * @param {function}\tcustomSplitIntoTokensHelper\tA custom helper to split sentences into tokens.\n *\n * @returns {{markings: Mark[], count: number}} The number of keyphrase occurrences in the text and the Mark objects of the matches.\n */\nexport function countKeyphraseInText(\n  sentences,\n  keyphraseForms,\n  locale,\n  matchWordCustomHelper,\n  isExactMatchRequested,\n  customSplitIntoTokensHelper\n) {\n  const result = {count: 0, markings: []};\n\n  sentences.forEach(sentence => {\n    const matchesInSentence = keyphraseForms.map(wordForms =>\n      matchWordFormsWithSentence(\n        sentence,\n        wordForms,\n        locale,\n        matchWordCustomHelper,\n        isExactMatchRequested,\n        customSplitIntoTokensHelper\n      )\n    );\n    // A sentence has at least one full-match of the keyphrase if each word occurs at least once.\n    const isEachWordFound = matchesInSentence.every(wordForms => wordForms.count > 0);\n\n    if (isEachWordFound) {\n      /*\n       * Retrieve all the occurrences' count of each word of the keyphrase and save it in an array.\n       * matches: [ [ { matches: [\"red\"], count: 1 } ], [ { matches: [\"pandas\"], count: 2 } ] ]\n       * counts: [ 1, 2 ]\n       */\n      const counts = matchesInSentence.map(match => match.count);\n      /*\n       * The number of the full-match count is the lowest count of the occurrences.\n       * counts: [ 1, 2 ]\n       * totalMatchCount: 1\n       *\n       * From the example above, the full-match is 1, because one of the \"pandas\" occurrences is not accompanied by \"red\"\n       * to be counted as a full-match.\n       */\n      const totalMatchCount = Math.min(...counts);\n      const foundWords = flattenDeep(matchesInSentence.map(match => match.matches));\n\n      let markings = [];\n\n      if (matchWordCustomHelper) {\n        // Currently, this check is only applicable for Japanese.\n        markings = markWordsInASentence(sentence, foundWords, matchWordCustomHelper);\n      } else {\n        markings = getMarkingsInSentence(sentence, foundWords);\n      }\n\n      result.count += totalMatchCount;\n      result.markings.push(markings);\n    }\n  });\n\n  return result;\n}\n\n/**\n * Calculates the keyphrase count, takes morphology into account.\n *\n * @param {Paper}       paper       The paper containing keyphrase and text.\n * @param {Researcher}  researcher  The researcher.\n *\n * @returns {{count: number, markings: Mark[], keyphraseLength: number}} An object containing the keyphrase count, markings and the kephrase length.\n */\nexport default function getKeyphraseCount(paper, researcher) {\n  const result = {count: 0, markings: [], keyphraseLength: 0};\n  const topicForms = researcher.getResearch('morphology');\n  let keyphraseForms = topicForms.keyphraseForms;\n  const keyphraseLength = keyphraseForms.length;\n\n  /*\n   * Normalize single quotes so that word form with different type of single quotes can still be matched.\n   * For example, \"key‛word\" should match \"key'word\".\n   */\n  keyphraseForms = keyphraseForms.map(word => word.map(form => normalizeSingle(form)));\n\n  if (keyphraseLength === 0) {\n    return result;\n  }\n\n  const matchWordCustomHelper = researcher.getHelper('matchWordCustomHelper');\n  const customSentenceTokenizer = researcher.getHelper('memoizedTokenizer');\n  const customSplitIntoTokensHelper = researcher.getHelper('splitIntoTokensCustom');\n  const locale = paper.getLocale();\n  const text = matchWordCustomHelper\n    ? filterShortcodesFromHTML(paper.getText(), paper._attributes && paper._attributes.shortcodes)\n    : paper.getText();\n\n  // When the custom helper is available, we're using the sentences retrieved from the text for the analysis.\n  const sentences = matchWordCustomHelper\n    ? getSentences(text, customSentenceTokenizer)\n    : getSentencesFromTree(paper);\n  // Exact matching is requested when the keyphrase is enclosed in double quotes.\n  const isExactMatchRequested = isDoubleQuoted(paper.getKeyword());\n\n  /*\n   * Count the amount of keyphrase occurrences in the sentences.\n   * An occurrence is counted when all words of the keyphrase are contained within the sentence. Each sentence can contain multiple keyphrases.\n   * (e.g. \"The apple potato is an apple and a potato.\" has two occurrences of the keyphrase \"apple potato\").\n   */\n  const keyphraseFound = countKeyphraseInText(\n    sentences,\n    keyphraseForms,\n    locale,\n    matchWordCustomHelper,\n    isExactMatchRequested,\n    customSplitIntoTokensHelper\n  );\n\n  result.count = keyphraseFound.count;\n  result.markings = flatten(keyphraseFound.markings);\n  result.keyphraseLength = keyphraseLength;\n\n  return result;\n}\n\n/**\n * Calculates the keyphrase count, takes morphology into account.\n *\n * @deprecated Use getKeyphraseCount instead.\n *\n * @param {Paper}       paper       The paper containing keyphrase and text.\n * @param {Researcher}  researcher  The researcher.\n *\n * @returns {Object} An array of all the matches, markings and the keyphrase count.\n */\nexport function keywordCount(paper, researcher) {\n  console.warn('This function is deprecated, use getKeyphraseCount instead.');\n  return getKeyphraseCount(paper, researcher);\n}\n"],"mappings":"AAAA,SAAQA,OAAO,EAAEC,WAAW,QAAO,QAAQ;AAC3C,OAAOC,oBAAoB;AAC3B,SAAQC,eAAe;AACvB,OAAOC,qBAAqB;AAC5B,OAAOC,0BAA0B;AACjC,OAAOC,cAAc;AACrB,SAAQC,oBAAoB;AAC5B,OAAOC,YAAY;AACnB,SAAQC,wBAAwB;;AAEhC;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA,OAAO,SAASC,oBAAoBA,CAClCC,SAAS,EACTC,cAAc,EACdC,MAAM,EACNC,qBAAqB,EACrBC,qBAAqB,EACrBC,2BAA2B,EAC3B;EACA,MAAMC,MAAM,GAAG;IAACC,KAAK,EAAE,CAAC;IAAEC,QAAQ,EAAE;EAAE,CAAC;EAEvCR,SAAS,CAACS,OAAO,CAACC,QAAQ,IAAI;IAC5B,MAAMC,iBAAiB,GAAGV,cAAc,CAACW,GAAG,CAACC,SAAS,IACpDnB,0BAA0B,CACxBgB,QAAQ,EACRG,SAAS,EACTX,MAAM,EACNC,qBAAqB,EACrBC,qBAAqB,EACrBC,2BACF,CACF,CAAC;IACD;IACA,MAAMS,eAAe,GAAGH,iBAAiB,CAACI,KAAK,CAACF,SAAS,IAAIA,SAAS,CAACN,KAAK,GAAG,CAAC,CAAC;IAEjF,IAAIO,eAAe,EAAE;MACnB;AACN;AACA;AACA;AACA;MACM,MAAME,MAAM,GAAGL,iBAAiB,CAACC,GAAG,CAACK,KAAK,IAAIA,KAAK,CAACV,KAAK,CAAC;MAC1D;AACN;AACA;AACA;AACA;AACA;AACA;AACA;MACM,MAAMW,eAAe,GAAGC,IAAI,CAACC,GAAG,CAAC,GAAGJ,MAAM,CAAC;MAC3C,MAAMK,UAAU,GAAG/B,WAAW,CAACqB,iBAAiB,CAACC,GAAG,CAACK,KAAK,IAAIA,KAAK,CAACK,OAAO,CAAC,CAAC;MAE7E,IAAId,QAAQ,GAAG,EAAE;MAEjB,IAAIL,qBAAqB,EAAE;QACzB;QACAK,QAAQ,GAAGZ,oBAAoB,CAACc,QAAQ,EAAEW,UAAU,EAAElB,qBAAqB,CAAC;MAC9E,CAAC,MAAM;QACLK,QAAQ,GAAGf,qBAAqB,CAACiB,QAAQ,EAAEW,UAAU,CAAC;MACxD;MAEAf,MAAM,CAACC,KAAK,IAAIW,eAAe;MAC/BZ,MAAM,CAACE,QAAQ,CAACe,IAAI,CAACf,QAAQ,CAAC;IAChC;EACF,CAAC,CAAC;EAEF,OAAOF,MAAM;AACf;;AAEA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA,eAAe,SAASkB,iBAAiBA,CAACC,KAAK,EAAEC,UAAU,EAAE;EAC3D,MAAMpB,MAAM,GAAG;IAACC,KAAK,EAAE,CAAC;IAAEC,QAAQ,EAAE,EAAE;IAAEmB,eAAe,EAAE;EAAC,CAAC;EAC3D,MAAMC,UAAU,GAAGF,UAAU,CAACG,WAAW,CAAC,YAAY,CAAC;EACvD,IAAI5B,cAAc,GAAG2B,UAAU,CAAC3B,cAAc;EAC9C,MAAM0B,eAAe,GAAG1B,cAAc,CAAC6B,MAAM;;EAE7C;AACF;AACA;AACA;EACE7B,cAAc,GAAGA,cAAc,CAACW,GAAG,CAACmB,IAAI,IAAIA,IAAI,CAACnB,GAAG,CAACoB,IAAI,IAAIxC,eAAe,CAACwC,IAAI,CAAC,CAAC,CAAC;EAEpF,IAAIL,eAAe,KAAK,CAAC,EAAE;IACzB,OAAOrB,MAAM;EACf;EAEA,MAAMH,qBAAqB,GAAGuB,UAAU,CAACO,SAAS,CAAC,uBAAuB,CAAC;EAC3E,MAAMC,uBAAuB,GAAGR,UAAU,CAACO,SAAS,CAAC,mBAAmB,CAAC;EACzE,MAAM5B,2BAA2B,GAAGqB,UAAU,CAACO,SAAS,CAAC,uBAAuB,CAAC;EACjF,MAAM/B,MAAM,GAAGuB,KAAK,CAACU,SAAS,CAAC,CAAC;EAChC,MAAMC,IAAI,GAAGjC,qBAAqB,GAC9BL,wBAAwB,CAAC2B,KAAK,CAACY,OAAO,CAAC,CAAC,EAAEZ,KAAK,CAACa,WAAW,IAAIb,KAAK,CAACa,WAAW,CAACC,UAAU,CAAC,GAC5Fd,KAAK,CAACY,OAAO,CAAC,CAAC;;EAEnB;EACA,MAAMrC,SAAS,GAAGG,qBAAqB,GACnCN,YAAY,CAACuC,IAAI,EAAEF,uBAAuB,CAAC,GAC3C3C,oBAAoB,CAACkC,KAAK,CAAC;EAC/B;EACA,MAAMrB,qBAAqB,GAAGT,cAAc,CAAC8B,KAAK,CAACe,UAAU,CAAC,CAAC,CAAC;;EAEhE;AACF;AACA;AACA;AACA;EACE,MAAMC,cAAc,GAAG1C,oBAAoB,CACzCC,SAAS,EACTC,cAAc,EACdC,MAAM,EACNC,qBAAqB,EACrBC,qBAAqB,EACrBC,2BACF,CAAC;EAEDC,MAAM,CAACC,KAAK,GAAGkC,cAAc,CAAClC,KAAK;EACnCD,MAAM,CAACE,QAAQ,GAAGnB,OAAO,CAACoD,cAAc,CAACjC,QAAQ,CAAC;EAClDF,MAAM,CAACqB,eAAe,GAAGA,eAAe;EAExC,OAAOrB,MAAM;AACf;;AAEA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA,OAAO,SAASoC,YAAYA,CAACjB,KAAK,EAAEC,UAAU,EAAE;EAC9CiB,OAAO,CAACC,IAAI,CAAC,6DAA6D,CAAC;EAC3E,OAAOpB,iBAAiB,CAACC,KAAK,EAAEC,UAAU,CAAC;AAC7C","ignoreList":[]}
+{"version":3,"file":"keywordCount.js","names":["flatten","flattenDeep","getSentencesFromTree","normalizeSingle","getMarkingsInSentence","matchWordFormsWithSentence","isDoubleQuoted","markWordsInASentence","getSentences","filterShortcodesFromHTML","isChineseText","countKeyphraseInText","sentences","keyphraseForms","locale","matchWordCustomHelper","isExactMatchRequested","customSplitIntoTokensHelper","result","count","markings","forEach","sentence","matchesInSentence","map","wordForms","isEachWordFound","every","counts","match","totalMatchCount","Math","min","foundWords","matches","push","getKeyphraseCount","paper","researcher","keyphraseLength","topicForms","getResearch","length","word","form","getHelper","customSentenceTokenizer","getLocale","text","getText","_attributes","shortcodes","getKeyword","lowerText","toLowerCase","lowerWord","startIndex","index","indexOf","error","console","warn","message","keyphraseFound","keywordCount"],"sources":["../../../src/languageProcessing/researches/keywordCount.js"],"sourcesContent":["import {flatten, flattenDeep} from 'lodash';\nimport getSentencesFromTree from '../helpers/sentence/getSentencesFromTree';\nimport {normalizeSingle} from '../helpers/sanitize/quotes';\nimport getMarkingsInSentence from '../helpers/highlighting/getMarkingsInSentence';\nimport matchWordFormsWithSentence from '../helpers/match/matchWordFormsWithSentence';\nimport isDoubleQuoted from '../helpers/match/isDoubleQuoted';\nimport {markWordsInASentence} from '../helpers/word/markWordsInSentences';\nimport getSentences from '../helpers/sentence/getSentences';\nimport {filterShortcodesFromHTML} from '../helpers';\nimport isChineseText from '../helpers/language/isChineseText';\n\n/**\n * Counts the occurrences of the keyphrase in the text and creates the Mark objects for the matches.\n *\n * @param {(Sentence|string)[]}\tsentences\t\t\tThe sentences to check.\n * @param {Array}\t\tkeyphraseForms\t\t\t\tThe keyphrase forms.\n * @param {string}\t\tlocale\t\t\t\t\t\tThe locale used in the analysis.\n * @param {function}\tmatchWordCustomHelper\t\tA custom helper to match words with a text.\n * @param {boolean}\t\tisExactMatchRequested\t\tWhether the exact matching is requested.\n * @param {function}\tcustomSplitIntoTokensHelper\tA custom helper to split sentences into tokens.\n *\n * @returns {{markings: Mark[], count: number}} The number of keyphrase occurrences in the text and the Mark objects of the matches.\n */\nexport function countKeyphraseInText(\n  sentences,\n  keyphraseForms,\n  locale,\n  matchWordCustomHelper,\n  isExactMatchRequested,\n  customSplitIntoTokensHelper\n) {\n  const result = {count: 0, markings: []};\n\n  sentences.forEach(sentence => {\n    const matchesInSentence = keyphraseForms.map(wordForms =>\n      matchWordFormsWithSentence(\n        sentence,\n        wordForms,\n        locale,\n        matchWordCustomHelper,\n        isExactMatchRequested,\n        customSplitIntoTokensHelper\n      )\n    );\n    // A sentence has at least one full-match of the keyphrase if each word occurs at least once.\n    const isEachWordFound = matchesInSentence.every(wordForms => wordForms.count > 0);\n\n    if (isEachWordFound) {\n      /*\n       * Retrieve all the occurrences' count of each word of the keyphrase and save it in an array.\n       * matches: [ [ { matches: [\"red\"], count: 1 } ], [ { matches: [\"pandas\"], count: 2 } ] ]\n       * counts: [ 1, 2 ]\n       */\n      const counts = matchesInSentence.map(match => match.count);\n      /*\n       * The number of the full-match count is the lowest count of the occurrences.\n       * counts: [ 1, 2 ]\n       * totalMatchCount: 1\n       *\n       * From the example above, the full-match is 1, because one of the \"pandas\" occurrences is not accompanied by \"red\"\n       * to be counted as a full-match.\n       */\n      const totalMatchCount = Math.min(...counts);\n      const foundWords = flattenDeep(matchesInSentence.map(match => match.matches));\n\n      let markings = [];\n\n      if (matchWordCustomHelper) {\n        // Currently, this check is only applicable for Japanese.\n        markings = markWordsInASentence(sentence, foundWords, matchWordCustomHelper);\n      } else {\n        markings = getMarkingsInSentence(sentence, foundWords);\n      }\n\n      result.count += totalMatchCount;\n      result.markings.push(markings);\n    }\n  });\n\n  return result;\n}\n\n/**\n * Calculates the keyphrase count, takes morphology into account.\n *\n * @param {Paper}       paper       The paper containing keyphrase and text.\n * @param {Researcher}  researcher  The researcher.\n *\n * @returns {{count: number, markings: Mark[], keyphraseLength: number}} An object containing the keyphrase count, markings and the kephrase length.\n */\nexport default function getKeyphraseCount(paper, researcher) {\n  const result = {count: 0, markings: [], keyphraseLength: 0};\n  const topicForms = researcher.getResearch('morphology');\n  let keyphraseForms = topicForms.keyphraseForms;\n  const keyphraseLength = keyphraseForms.length;\n\n  /*\n   * Normalize single quotes so that word form with different type of single quotes can still be matched.\n   * For example, \"key‛word\" should match \"key'word\".\n   */\n  keyphraseForms = keyphraseForms.map(word => word.map(form => normalizeSingle(form)));\n\n  if (keyphraseLength === 0) {\n    return result;\n  }\n\n  let matchWordCustomHelper = researcher.getHelper('matchWordCustomHelper');\n  const customSentenceTokenizer = researcher.getHelper('memoizedTokenizer');\n  const customSplitIntoTokensHelper = researcher.getHelper('splitIntoTokensCustom');\n  const locale = paper.getLocale();\n  const text = matchWordCustomHelper\n    ? filterShortcodesFromHTML(paper.getText(), paper._attributes && paper._attributes.shortcodes)\n    : paper.getText();\n\n  // Auto-detect Chinese and use Chinese helper if not already available\n  if (!matchWordCustomHelper && isChineseText(paper.getKeyword() + ' ' + text)) {\n    // Import Chinese helper for Chinese text\n    try {\n      // Use simple Chinese word matching for Chinese text\n      matchWordCustomHelper = function(sentence, word) {\n        const text = typeof sentence === 'string' ? sentence : sentence.text || sentence;\n        const matches = [];\n\n        if (!text || !word) {\n          return matches;\n        }\n\n        const lowerText = text.toLowerCase();\n        const lowerWord = word.toLowerCase();\n\n        let startIndex = 0;\n        let index;\n\n        while ((index = lowerText.indexOf(lowerWord, startIndex)) !== -1) {\n          matches.push(word);\n          startIndex = index + lowerWord.length;\n        }\n\n        return matches;\n      };\n    } catch (error) {\n      // If there's any issue, continue with default behavior\n      console.warn('Error setting up Chinese matching, using default:', error.message);\n    }\n  }\n\n  // When the custom helper is available, we're using the sentences retrieved from the text for the analysis.\n  const sentences = matchWordCustomHelper\n    ? getSentences(text, customSentenceTokenizer)\n    : getSentencesFromTree(paper);\n  // Exact matching is requested when the keyphrase is enclosed in double quotes.\n  const isExactMatchRequested = isDoubleQuoted(paper.getKeyword());\n\n  /*\n   * Count the amount of keyphrase occurrences in the sentences.\n   * An occurrence is counted when all words of the keyphrase are contained within the sentence. Each sentence can contain multiple keyphrases.\n   * (e.g. \"The apple potato is an apple and a potato.\" has two occurrences of the keyphrase \"apple potato\").\n   */\n  const keyphraseFound = countKeyphraseInText(\n    sentences,\n    keyphraseForms,\n    locale,\n    matchWordCustomHelper,\n    isExactMatchRequested,\n    customSplitIntoTokensHelper\n  );\n\n  result.count = keyphraseFound.count;\n  result.markings = flatten(keyphraseFound.markings);\n  result.keyphraseLength = keyphraseLength;\n\n  return result;\n}\n\n/**\n * Calculates the keyphrase count, takes morphology into account.\n *\n * @deprecated Use getKeyphraseCount instead.\n *\n * @param {Paper}       paper       The paper containing keyphrase and text.\n * @param {Researcher}  researcher  The researcher.\n *\n * @returns {Object} An array of all the matches, markings and the keyphrase count.\n */\nexport function keywordCount(paper, researcher) {\n  console.warn('This function is deprecated, use getKeyphraseCount instead.');\n  return getKeyphraseCount(paper, researcher);\n}\n"],"mappings":"AAAA,SAAQA,OAAO,EAAEC,WAAW,QAAO,QAAQ;AAC3C,OAAOC,oBAAoB;AAC3B,SAAQC,eAAe;AACvB,OAAOC,qBAAqB;AAC5B,OAAOC,0BAA0B;AACjC,OAAOC,cAAc;AACrB,SAAQC,oBAAoB;AAC5B,OAAOC,YAAY;AACnB,SAAQC,wBAAwB;AAChC,OAAOC,aAAa;;AAEpB;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA,OAAO,SAASC,oBAAoBA,CAClCC,SAAS,EACTC,cAAc,EACdC,MAAM,EACNC,qBAAqB,EACrBC,qBAAqB,EACrBC,2BAA2B,EAC3B;EACA,MAAMC,MAAM,GAAG;IAACC,KAAK,EAAE,CAAC;IAAEC,QAAQ,EAAE;EAAE,CAAC;EAEvCR,SAAS,CAACS,OAAO,CAACC,QAAQ,IAAI;IAC5B,MAAMC,iBAAiB,GAAGV,cAAc,CAACW,GAAG,CAACC,SAAS,IACpDpB,0BAA0B,CACxBiB,QAAQ,EACRG,SAAS,EACTX,MAAM,EACNC,qBAAqB,EACrBC,qBAAqB,EACrBC,2BACF,CACF,CAAC;IACD;IACA,MAAMS,eAAe,GAAGH,iBAAiB,CAACI,KAAK,CAACF,SAAS,IAAIA,SAAS,CAACN,KAAK,GAAG,CAAC,CAAC;IAEjF,IAAIO,eAAe,EAAE;MACnB;AACN;AACA;AACA;AACA;MACM,MAAME,MAAM,GAAGL,iBAAiB,CAACC,GAAG,CAACK,KAAK,IAAIA,KAAK,CAACV,KAAK,CAAC;MAC1D;AACN;AACA;AACA;AACA;AACA;AACA;AACA;MACM,MAAMW,eAAe,GAAGC,IAAI,CAACC,GAAG,CAAC,GAAGJ,MAAM,CAAC;MAC3C,MAAMK,UAAU,GAAGhC,WAAW,CAACsB,iBAAiB,CAACC,GAAG,CAACK,KAAK,IAAIA,KAAK,CAACK,OAAO,CAAC,CAAC;MAE7E,IAAId,QAAQ,GAAG,EAAE;MAEjB,IAAIL,qBAAqB,EAAE;QACzB;QACAK,QAAQ,GAAGb,oBAAoB,CAACe,QAAQ,EAAEW,UAAU,EAAElB,qBAAqB,CAAC;MAC9E,CAAC,MAAM;QACLK,QAAQ,GAAGhB,qBAAqB,CAACkB,QAAQ,EAAEW,UAAU,CAAC;MACxD;MAEAf,MAAM,CAACC,KAAK,IAAIW,eAAe;MAC/BZ,MAAM,CAACE,QAAQ,CAACe,IAAI,CAACf,QAAQ,CAAC;IAChC;EACF,CAAC,CAAC;EAEF,OAAOF,MAAM;AACf;;AAEA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA,eAAe,SAASkB,iBAAiBA,CAACC,KAAK,EAAEC,UAAU,EAAE;EAC3D,MAAMpB,MAAM,GAAG;IAACC,KAAK,EAAE,CAAC;IAAEC,QAAQ,EAAE,EAAE;IAAEmB,eAAe,EAAE;EAAC,CAAC;EAC3D,MAAMC,UAAU,GAAGF,UAAU,CAACG,WAAW,CAAC,YAAY,CAAC;EACvD,IAAI5B,cAAc,GAAG2B,UAAU,CAAC3B,cAAc;EAC9C,MAAM0B,eAAe,GAAG1B,cAAc,CAAC6B,MAAM;;EAE7C;AACF;AACA;AACA;EACE7B,cAAc,GAAGA,cAAc,CAACW,GAAG,CAACmB,IAAI,IAAIA,IAAI,CAACnB,GAAG,CAACoB,IAAI,IAAIzC,eAAe,CAACyC,IAAI,CAAC,CAAC,CAAC;EAEpF,IAAIL,eAAe,KAAK,CAAC,EAAE;IACzB,OAAOrB,MAAM;EACf;EAEA,IAAIH,qBAAqB,GAAGuB,UAAU,CAACO,SAAS,CAAC,uBAAuB,CAAC;EACzE,MAAMC,uBAAuB,GAAGR,UAAU,CAACO,SAAS,CAAC,mBAAmB,CAAC;EACzE,MAAM5B,2BAA2B,GAAGqB,UAAU,CAACO,SAAS,CAAC,uBAAuB,CAAC;EACjF,MAAM/B,MAAM,GAAGuB,KAAK,CAACU,SAAS,CAAC,CAAC;EAChC,MAAMC,IAAI,GAAGjC,qBAAqB,GAC9BN,wBAAwB,CAAC4B,KAAK,CAACY,OAAO,CAAC,CAAC,EAAEZ,KAAK,CAACa,WAAW,IAAIb,KAAK,CAACa,WAAW,CAACC,UAAU,CAAC,GAC5Fd,KAAK,CAACY,OAAO,CAAC,CAAC;;EAEnB;EACA,IAAI,CAAClC,qBAAqB,IAAIL,aAAa,CAAC2B,KAAK,CAACe,UAAU,CAAC,CAAC,GAAG,GAAG,GAAGJ,IAAI,CAAC,EAAE;IAC5E;IACA,IAAI;MACF;MACAjC,qBAAqB,GAAG,SAAAA,CAASO,QAAQ,EAAEqB,IAAI,EAAE;QAC/C,MAAMK,IAAI,GAAG,OAAO1B,QAAQ,KAAK,QAAQ,GAAGA,QAAQ,GAAGA,QAAQ,CAAC0B,IAAI,IAAI1B,QAAQ;QAChF,MAAMY,OAAO,GAAG,EAAE;QAElB,IAAI,CAACc,IAAI,IAAI,CAACL,IAAI,EAAE;UAClB,OAAOT,OAAO;QAChB;QAEA,MAAMmB,SAAS,GAAGL,IAAI,CAACM,WAAW,CAAC,CAAC;QACpC,MAAMC,SAAS,GAAGZ,IAAI,CAACW,WAAW,CAAC,CAAC;QAEpC,IAAIE,UAAU,GAAG,CAAC;QAClB,IAAIC,KAAK;QAET,OAAO,CAACA,KAAK,GAAGJ,SAAS,CAACK,OAAO,CAACH,SAAS,EAAEC,UAAU,CAAC,MAAM,CAAC,CAAC,EAAE;UAChEtB,OAAO,CAACC,IAAI,CAACQ,IAAI,CAAC;UAClBa,UAAU,GAAGC,KAAK,GAAGF,SAAS,CAACb,MAAM;QACvC;QAEA,OAAOR,OAAO;MAChB,CAAC;IACH,CAAC,CAAC,OAAOyB,KAAK,EAAE;MACd;MACAC,OAAO,CAACC,IAAI,CAAC,mDAAmD,EAAEF,KAAK,CAACG,OAAO,CAAC;IAClF;EACF;;EAEA;EACA,MAAMlD,SAAS,GAAGG,qBAAqB,GACnCP,YAAY,CAACwC,IAAI,EAAEF,uBAAuB,CAAC,GAC3C5C,oBAAoB,CAACmC,KAAK,CAAC;EAC/B;EACA,MAAMrB,qBAAqB,GAAGV,cAAc,CAAC+B,KAAK,CAACe,UAAU,CAAC,CAAC,CAAC;;EAEhE;AACF;AACA;AACA;AACA;EACE,MAAMW,cAAc,GAAGpD,oBAAoB,CACzCC,SAAS,EACTC,cAAc,EACdC,MAAM,EACNC,qBAAqB,EACrBC,qBAAqB,EACrBC,2BACF,CAAC;EAEDC,MAAM,CAACC,KAAK,GAAG4C,cAAc,CAAC5C,KAAK;EACnCD,MAAM,CAACE,QAAQ,GAAGpB,OAAO,CAAC+D,cAAc,CAAC3C,QAAQ,CAAC;EAClDF,MAAM,CAACqB,eAAe,GAAGA,eAAe;EAExC,OAAOrB,MAAM;AACf;;AAEA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA;AACA,OAAO,SAAS8C,YAAYA,CAAC3B,KAAK,EAAEC,UAAU,EAAE;EAC9CsB,OAAO,CAACC,IAAI,CAAC,6DAA6D,CAAC;EAC3E,OAAOzB,iBAAiB,CAACC,KAAK,EAAEC,UAAU,CAAC;AAC7C","ignoreList":[]}