Normalize all UTF quotes in Javascript

stdquote.js

/**
 * Will normalize quotes in a given string. There are many variations of quotes
 * in the unicode character set, this function attempts to convert any variation
 * of quote to the standard Quotation Mark - U+0022 Standard Universal: "
 *
 * @param {string} str The string to normalize
 * @return {string} Normalized string.
 * @see https://unicode-table.com/en/sets/quotation-marks/
 */
helpers.stdQuote = (str) => {
  const allQuotes = [
    '“', // U+201c
    '”', // U+201d
    '«', // U+00AB
    '»', // U+00BB
    '„', // U+201E
    '“', // U+201C
    '‟', // U+201F
    '”', // U+201D
    '❝', // U+275D
    '❞', // U+275E
    '〝', // U+301D
    '〞', // U+301E
    '〟', // U+301F
    '＂', // U+FF02
  ];

  const stdQuote = '"'; // U+0022

  const normalized = allQuotes.reduce((strNorm, quoteChar) => {
    // eslint-disable-next-line security/detect-non-literal-regexp
    const re = new RegExp(quoteChar, 'g');
    return strNorm.replace(re, stdQuote);
  }, str);

  return normalized;
};

Great! Have you got a single-quote version that takes backticks etc into account?

This gave me a good start. I also needed single quotes and dashes covered too. I made a more expansive list and made sure to use the Unicode identifiers. It is very prone to human error to just use the visual rendering, especially when it comes to the dashes.

const sanitizeChars: Array<[string, '-' | "'" | '"']> = Object.entries({
  // EN DASH / HYPHEN (U+002D)
  '\u1806': '\u002D', // '᠆'
  '\u2010': '\u002D', // '‐'
  '\u2011': '\u002D', // '‑'
  '\u2012': '\u002D', // '‒'
  '\u2013': '\u002D', // '–'
  '\uFE58': '\u002D', // '﹘'
  '\uFE63': '\u002D', // '﹣'
  '\uFF0D': '\u002D', // '－'

  // SINGLE QUOTES (U+0027)
  '\u003C': '\u0027', // '<'
  '\u003E': '\u0027', // '>'
  '\u2018': '\u0027', // '‘'
  '\u2019': '\u0027', // '’'
  '\u201A': '\u0027', // '‚'
  '\u201B': '\u0027', // '‛'
  '\u2039': '\u0027', // '‹'
  '\u203A': '\u0027', // '›'
  '\u275B': '\u0027', // '❛'
  '\u275C': '\u0027', // '❜'
  '\u276E': '\u0027', // '❮'
  '\u276F': '\u0027', // '❯'
  '\uFF07': '\u0027', // '＇'
  '\u300C': '\u0027', // '「'
  '\u300D': '\u0027', // '」'

  // // DOUBLE QUOTES (U+0022)
  '\u00AB': '\u0022', // '«'
  '\u00BB': '\u0022', // '»'
  '\u201C': '\u0022', // '“'
  '\u201D': '\u0022', // '”'
  '\u201E': '\u0022', // '„'
  '\u201F': '\u0022', // '‟'
  '\u275D': '\u0022', // '❝'
  '\u275E': '\u0022', // '❞'
  '\u2E42': '\u0022', // '⹂'
  '\u301D': '\u0022', // '〝'
  '\u301E': '\u0022', // '〞'
  '\u301F': '\u0022', // '〟'
  '\uFF02': '\u0022', // '＂'
  '\u300E': '\u0022', // '『'
  '\u300F': '\u0022', // '』'
});

/** Normalizes non-standard quotes and dashes into a uniform format.  */
const sanitizeString = (query: string) => {
  return sanitizeChars.reduce((acc, [char, stdChar]) => {
    return acc.replaceAll(char, stdChar);
  }, query);
};