Util Method For Lucene Escape

Often I find myself escaping the queries prior to executing them. I was using lucene-escape-query, but realized it isn’t quite right. I was going to create a pull request for that repo, but it seems like such a common thing that it would be beneficial to be included in this client library.

How are others handling this right now? Is there existing way or another package util people are using to do this?

What I was thinking is a simple util method as follows could be added to this client library…

// Based on https://www.elastic.co/guide/en/elasticsearch/reference/1.7/query-dsl-query-string-query.html#_reserved_characters
// TODO: This might not be exactly right
function escape (query) {
  return query.replace(/([\!\*\+\&\|\(\)\[\]\{\}\^\~\?\:\"\/])/g, "\\$1");
}

Thanks!

[vitorbaptista]

Thanks a lot, @spalger. You’ve given me lots of food for thought 😄

I still need to use the query string for now, because the API doesn’t receive the filters separated like that, but inside a q parameter. However, I’ll probably start adding those filters explictly as API parameters, then your suggestion will come in handy 👍

[GuillaumeCisco]

Done simply with lucene by using :

// https://lucene.apache.org/core/6_0_0/queryparser/org/apache/lucene/queryparser/classic/package-summary.html#Escaping_Special_Characters
const toEscape = [
  '+',
  '-',
  '&',
  '|',
  '!',
  '(',
  ')',
  '{',
  '}',
  '[',
  ']',
  '^',
  '"',
  '~',
  '*',
  '?',
  ':',
  '\\',
  '/',
]

const luceneEscape = (str) => {
  let res = ''
  for (let i = 0; i < str.length; i += 1) {
    const char = str.charAt(i)
    if (toEscape.includes(char)) {
      // handle special case && and ||
      if (['&', '|'].includes(char)) {
        if (char === str.charAt(i + 1)) {
          res = `${res}\\${char}\\${str.charAt(i + 1)}`
          i += 1
        } else res = `${res}${char}`
      } else res = `${res}\\${char}`
    } else res = `${res}${char}`
  }

  return res
}