regex 在Qualtrics中实现时，字符串变量从JavaScript正则表达式中消失

dbf7pr2w  于 2023-04-22  发布在  Java
关注(0)|答案(1)|浏览(77)

我试图计算Qualtrics上的调查响应中与一组正则表达式匹配的数量（例如，每当人们使用“我认为...”，“在我看来”等时，计数增加1）。
下面是我写的JavaScript代码：

const DONT_PHRASES_ARR = ["dont"," don't"," do not"," can not"," cant"," can't"];
const DONT_PHRASES = DONT_PHRASES_ARR.join("|");
const PRONOUNS_ARR = ["he","she","it","they"];
const PRONOUNS = PRONOUNS_ARR.join("|");
const PRESIDENT_NAMES_ARR = ["candidate","clinton","donald","gop","hillary","hilary","trump","trum"];
const PRESIDENT_NAMES = PRESIDENT_NAMES_ARR.join("|");
const SKIP_WORDS_ARR = ["also"," really"," very much"];
const SKIP_WORDS = SKIP_WORDS_ARR.join("|");

const AMBIGUOUS_WORDS_ARR = ["seemed","prefer"];
const AMBIGUOUS_WORDS = AMBIGUOUS_WORDS_ARR.join("|");
const I_OPINION_WORDS_ARR = ["agree","believe","consider","disagree","hope","feel","felt","find","oppose","think","thought","support"];
const I_OPINION_WORDS = I_OPINION_WORDS_ARR.join("|");
const OPINION_PHRASES_ARR = ["in my opinion","it seems to me","from my perspective","in my view","from my view","from my standpoint","for me"];
const OPINION_PHRASES = OPINION_PHRASES_ARR.join("|");
const OPINION_PHRASE_REGEXES = [
  {label: "op1", regex: new RegExp(`(?:i(?:${DONT_PHRASES}|${SKIP_WORDS})? (?:\b${I_OPINION_WORDS}\b) )`, 'i')},
  {label: "op2", regex: new RegExp(`(?:i'm [a-z]+ to (?:\b${I_OPINION_WORDS}\b) )`, 'i')},
  {label: "op3", regex: new RegExp(`(?:${OPINION_PHRASES},? )`, 'i')}
];

const STRONG_FACT_WORDS_ARR = ["are","can't","demonstrate","demonstrates","did","had","is","needs","should","will","would"];
const STRONG_FACT_WORDS = STRONG_FACT_WORDS_ARR.join("|");
const WEAKER_FACT_WORDS_ARR = ["were","was","has"];
const WEAKER_FACT_WORDS = WEAKER_FACT_WORDS_ARR.join("|");
const FACT_WORDS_ARR = WEAKER_FACT_WORDS_ARR.concat(STRONG_FACT_WORDS_ARR);
const FACT_WORDS = FACT_WORDS_ARR.join("|");
const FACT_PHRASES = new RegExp();
const FACT_PHRASE_REGEXES = [
  {label: "fp1", regex: new RegExp(`(?:[tT]he [A-Z][a-z]+ (?:\b${FACT_WORDS}\b) )`, 'gmi')},
  {label: "fp2", regex: new RegExp(`(?:(?:^|.+\. )[A-Z][a-z]+ (?:\b${FACT_WORDS}\b) )`, 'gmi')},
  {label: "fp3", regex: new RegExp(`(?:[tT]he [^\.]*[A-Z][a-z]+\'s? [a-z]+ (?:\b${FACT_WORDS}\b) )`, 'gmi')},
  {label: "fp4", regex: new RegExp(`(?:(?:[^\.]*${PRONOUNS}) (?:\b${STRONG_FACT_WORDS}\b) )`, 'gi')},
  {label: "fp5", regex: new RegExp(`(?:(?:^|\. )${PRONOUNS} (?:\b${FACT_WORDS}\b) )`, 'gi')},
  {label: "fp6", regex: new RegExp(`(?:(?:^|[^.]* )(?:\b${PRESIDENT_NAMES}\b) (?:\b${FACT_WORDS}\b) )`, 'gi')},
  {label: "fp7", regex: new RegExp(`(?:(?:^|[^.]* )(?:${PRONOUNS}|${PRESIDENT_NAMES}) [a-z]+(?:ed|[^ia]s) )`, 'gi')},
  {label: "fp8", regex: new RegExp(`(?:(?:^|[^.]* )(?:${PRONOUNS}|${PRESIDENT_NAMES}) [a-z]+ [a-z]+(?:ed|[^ia]s) )`, 'gi')},
  {label: "fp9", regex: new RegExp(`(?:(?:^|\. )(?:She\'s|He\'s) )`, 'gi')}
];

const history = [];

function analyze() {
  const response = document.getElementById('input').value.trim();
     
    const response_sentences = response.match(/[^\.!\?]+[ \.!\?]+/g);

    var i = 0
        while (i < response_sentences.length){
          let response_words = response_sentences[i].toString().split(' ')
               .map((w) => w.toLowerCase().replace(/[\W]/g, ''));
            i++;
          }
    
  
let fact_phrases = [];
let opinion_phrases = [];
var j = 0;
while (j < response_sentences.length) {
  // Check for opinion phrases
  let opinion_match = false;
  for (let {label, regex} of OPINION_PHRASE_REGEXES) {
    if (response_sentences[j].match(regex)) {
      opinion_match = true;
      op_match_phr = response_sentences[j].match(regex);
      opinion_phrases.push({match: op_match_phr[0], label: label, sent: j});
      break;
    }
  }
  if (!opinion_match) {
    // This sentence does not contain an opinion phrase
    let fact_match = false;
    for (let {label, regex} of FACT_PHRASE_REGEXES) {
      if (response_sentences[j].match(regex)) {
        fact_match = true;
        fact_match_phr = response_sentences[j].match(regex);
        fact_phrases.push({match: fact_match_phr[0], label: label, sent: j});
        break;
      }
    }
  }
  j++;
}

    const opinion_phrases_uniq = opinion_phrases.filter((obj, index) =>
      opinion_phrases.findIndex((item) => item.sent === obj.sent) === index
    );
   
    let opinion_phrases_uniq_t2 = opinion_phrases_uniq.length;
  
    let fact_phrases_t2 = fact_phrases.length;
    
    const fact_phrases_uniq = fact_phrases.filter((obj, index) =>
      fact_phrases.findIndex((item) => item.sent === obj.sent) === index
    );
    console.log('FACT PHRASES UNIQUE:', fact_phrases_uniq)
  

    let fact_phrases_uniq_t2 = fact_phrases_uniq.length;
    
    let net_score = opinion_phrases_uniq_t2 - fact_phrases_uniq_t2;

    if (response !== 'NA') {
      console.log('-------------------------');
      // console.log('RESPONSE: ', response);
      console.log('OPINION PHRASES:', opinion_phrases_uniq_t2);
      console.log('FACT PHRASES:', fact_phrases_uniq_t2);
      console.log('NET SCORE:', net_score);
    }
  
  
 const result = {
      input: response,
      net_score: net_score,
      opinion_phrases_t2: opinion_phrases_uniq_t2,
      fact_phrases_t2: fact_phrases_uniq_t2,
      opinion_phrases: opinion_phrases,
      fact_phrases: fact_phrases
    };
  
    // Display result
    const output = document.getElementById('output');
    output.textContent = `Net score: ${net_score}\nOpinion phrases: ${opinion_phrases_uniq_t2}\nFact phrases: ${fact_phrases_uniq_t2}`;
  }

我试着在Qualtrics中实现这个脚本：

{
    /*Place your JavaScript here to run when the page loads*/

});

Qualtrics.SurveyEngine.addOnReady(function analyze() {
  
  const DONT_PHRASES_ARR = ["dont"," don't"," do not"," can not"," cant"," can't"];
  const DONT_PHRASES = DONT_PHRASES_ARR.join("|");
  const PRONOUNS_ARR = ["he","she","it","they"];
  const PRONOUNS = PRONOUNS_ARR.join("|");
  const PRESIDENT_NAMES_ARR = ["candidate","clinton","donald","gop","hillary","hilary","trump","trum"];
  const PRESIDENT_NAMES = PRESIDENT_NAMES_ARR.join("|");
  const SKIP_WORDS_ARR = ["also"," really"," very much"];
  const SKIP_WORDS = SKIP_WORDS_ARR.join("|");

  const AMBIGUOUS_WORDS_ARR = ["seemed","prefer"];
  const AMBIGUOUS_WORDS = AMBIGUOUS_WORDS_ARR.join("|");
  const I_OPINION_WORDS_ARR = ["agree","believe","consider","disagree","hope","feel","felt","find","oppose","think","thought","support"];
  const I_OPINION_WORDS = I_OPINION_WORDS_ARR.join("|");
  const OPINION_PHRASES_ARR = ["in my opinion","it seems to me","from my perspective","in my view","from my view","from my standpoint","for me"];
  const OPINION_PHRASES = OPINION_PHRASES_ARR.join("|");
  const OPINION_PHRASE_REGEXES = [
    {label: "op1", regex: new RegExp(`(?:i(?:${DONT_PHRASES}|${SKIP_WORDS})? (?:\b${I_OPINION_WORDS}\b) )`, 'i')},
    {label: "op2", regex: new RegExp(`(?:i'm [a-z]+ to (?:\b${I_OPINION_WORDS}\b) )`, 'i')},
    {label: "op3", regex: new RegExp(`(?:${OPINION_PHRASES},? )`, 'i')}
  ];



  const STRONG_FACT_WORDS_ARR = ["are","can't","demonstrate","demonstrates","did","had","is","needs","should","will","would"];
  const STRONG_FACT_WORDS = STRONG_FACT_WORDS_ARR.join("|");
  const WEAKER_FACT_WORDS_ARR = ["were","was","has"];
  const WEAKER_FACT_WORDS = WEAKER_FACT_WORDS_ARR.join("|");
  // const FACT_WORDS = new RegExp(`${STRONG_FACT_WORDS}|${WEAKER_FACT_WORDS}`);
  const FACT_WORDS_ARR = WEAKER_FACT_WORDS_ARR.concat(STRONG_FACT_WORDS_ARR);
  const FACT_WORDS = FACT_WORDS_ARR.join("|");
  const FACT_PHRASES = new RegExp();
  const FACT_PHRASE_REGEXES = [
    {label: "fp1", regex: new RegExp(`(?:[tT]he [A-Z][a-z]+ (?:\b${FACT_WORDS}\b) )`, 'gmi')},
    {label: "fp2", regex: new RegExp(`(?:(?:^|.+\. )[A-Z][a-z]+ (?:\b${FACT_WORDS}\b) )`, 'gmi')},
    {label: "fp3", regex: new RegExp(`(?:[tT]he [^\.]*[A-Z][a-z]+\'s? [a-z]+ (?:\b${FACT_WORDS}\b) )`, 'gmi')},
    {label: "fp4", regex: new RegExp(`(?:(?:[^\.]*${PRONOUNS}) (?:\b${STRONG_FACT_WORDS}\b) )`, 'gi')},
    {label: "fp5", regex: new RegExp(`(?:(?:^|\. )${PRONOUNS} (?:\b${FACT_WORDS}\b) )`, 'gi')},
    {label: "fp6", regex: new RegExp(`(?:(?:^|[^.]* )(?:\b${PRESIDENT_NAMES}\b) (?:\b${FACT_WORDS}\b) )`, 'gi')},
    {label: "fp7", regex: new RegExp(`(?:(?:^|[^.]* )(?:${PRONOUNS}|${PRESIDENT_NAMES}) [a-z]+(?:ed|[^ia]s) )`, 'gi')},
    {label: "fp8", regex: new RegExp(`(?:(?:^|[^.]* )(?:${PRONOUNS}|${PRESIDENT_NAMES}) [a-z]+ [a-z]+(?:ed|[^ia]s) )`, 'gi')},
    {label: "fp9", regex: new RegExp(`(?:(?:^|\. )(?:She\'s|He\'s) )`, 'gi')}
  ];

  const history = [];
  var response = "${q://QID19/QuestionText}"
  var response = response.value.trim();
  const response_sentences = response.match(/[^\.!\?]+[ \.!\?]+/g);

    var i = 0
        while (i < response_sentences.length){
          let response_words = response_sentences[i].toString().split(' ')
               .map((w) => w.toLowerCase().replace(/[\W]/g, ''));
            i++;
          }
  
  let fact_phrases = [];
  let opinion_phrases = [];
  var j = 0;
  while (j < response_sentences.length) {
    // Check for opinion phrases
    let opinion_match = false;
    for (let {label, regex} of OPINION_PHRASE_REGEXES) {
      if (response_sentences[j].match(regex)) {
        opinion_match = true;
        op_match_phr = response_sentences[j].match(regex);
        opinion_phrases.push({match: op_match_phr[0], label: label, sent: j});
        break;
      }
    }
    if (!opinion_match) {
      // This sentence does not contain an opinion phrase
      let fact_match = false;
      for (let {label, regex} of FACT_PHRASE_REGEXES) {
        if (response_sentences[j].match(regex)) {
          fact_match = true;
          fact_match_phr = response_sentences[j].match(regex);
          fact_phrases.push({match: fact_match_phr[0], label: label, sent: j});
          break;
        }
      }
    }
    j++;
  }

  // Do something with fact_sentences and opinion_sentences
    console.log('OP PHRASES :', opinion_phrases)

    let opinion_phrases_t2 = opinion_phrases.length;
    // console.log('OP PHR NUM: ', opinion_phrases_t2)

    const opinion_phrases_uniq = opinion_phrases.filter((obj, index) =>
      opinion_phrases.findIndex((item) => item.sent === obj.sent) === index
    );
     console.log('OP PHRASES UNIQUE:', opinion_phrases_uniq)

    let opinion_phrases_uniq_t2 = opinion_phrases_uniq.length;
    // console.log('OP PHR UNIQUE NUM : ', opinion_phrases_uniq_t2)

    console.log('FACT PHRASES :', fact_phrases)

    let fact_phrases_t2 = fact_phrases.length;
    // console.log('FACT PHR NUM: ', fact_phrases_t2)

    const fact_phrases_uniq = fact_phrases.filter((obj, index) =>
      fact_phrases.findIndex((item) => item.sent === obj.sent) === index
    );
    console.log('FACT PHRASES UNIQUE:', fact_phrases_uniq)
    // Object.keys(fact_phrases[1]).forEach((prop)=> console.log(prop));

    let fact_phrases_uniq_t2 = fact_phrases_uniq.length;
    // console.log('FACT PHR NUM (UNIQUE): ', fact_phrases_uniq_t2)

    let net_score = opinion_phrases_uniq_t2 - fact_phrases_uniq_t2;

    if (response !== 'NA') {
      console.log('-------------------------');
      // console.log('RESPONSE: ', response);
      console.log('OPINION PHRASES:', opinion_phrases_uniq_t2);
      console.log('FACT PHRASES:', fact_phrases_uniq_t2);
      console.log('NET SCORE:', net_score);
    }
  
  
 const result = {
      input: response,
      net_score: net_score,
      opinion_phrases_t2: opinion_phrases_uniq_t2,
      fact_phrases_t2: fact_phrases_uniq_t2,
      opinion_phrases: opinion_phrases,
      fact_phrases: fact_phrases
    };
console.log(result)
  
 Qualtrics.SurveyEngine.setEmbeddedData('feedback', result);

});

Qualtrics.SurveyEngine.addOnUnload(function()
{
    /*Place your JavaScript here to run when the page is unloaded*/

});

当我预览调查并查看控制台时，正则表达式缺少字符串变量（例如SKIP_WORDS）：

const OPINION_PHRASE_REGEXES = [
    {label: "op1", regex: new RegExp(`(?:i(?:|)? (?:\b\b) )`, 'i')},
    {label: "op2", regex: new RegExp(`(?:i'm [a-z]+ to (?:\b\b) )`, 'i')},
    {label: "op3", regex: new RegExp(`(?:,? )`, 'i')}
  ];

这是我希望在那里看到的：

const OPINION_PHRASE_REGEXES = [
    {label: "op1", regex: new RegExp(`(?:i(?:${DONT_PHRASES}|${SKIP_WORDS})? (?:\b${I_OPINION_WORDS}\b) )`, 'i')},
    {label: "op2", regex: new RegExp(`(?:i'm [a-z]+ to (?:\b${I_OPINION_WORDS}\b) )`, 'i')},
    {label: "op3", regex: new RegExp(`(?:${OPINION_PHRASES},? )`, 'i')}
  ];

有没有Qualtrics的语法让我摸不着头脑？我对Stack和JavaScript还是个新手--希望得到帮助！

regex

1条答案

按热度按时间
yqyhoc1h

yqyhoc1h1#

你应该把\b * 放在包含所有你要匹配的替代词的组之外。否则，前面的\b只与第一个词相关联，后面的\b与最后一个词相关联。所有其他词都将被匹配，而不需要在它们周围设置任何词边界。
此外，您需要转义反斜杠，使他们字面;否则，\b是退格字符的转义序列。

(?:\b${FACT_WORDS}\b)

应该是

\\b(?:${FACT_WORDS})\\b
赞(0）分享  回复(0）举报  2023-04-22
我来回答

相关问题

    查看更多

    热门标签

    更多
    JavaquerypythonNode开发语言requestUtil数据库Table后端算法LoggerMessageElementParser

    最新问答

    更多
    • 蜀ICP备2022004421号-2 NULL123 https://www.null123.com © All rights reserved
    • 本站内容来源互联网,如果侵犯您的权益请联系我们删除, 联系方式：448109455@qq.com