Bug in the JSON parser with JavaScript code

[josdejong]

First, thanks for the excellent article JSON Parser with JavaScript. I like the straightforward and easy to understand code, and it was faster than my original code. I used it as a basis to simplify and improve my own lossless-json parser , the parse code in v2 code is largely yours 😄 .

The full CodeSandbox example in the article contains a bug though when parsing escaped control characters like \n and \t (line 100-111). These are replaced with regular word characters n and t instead of the real control characters:

        if (str[i] === "\\") {
          const char = str[i + 1];
          if (
            char === '"' ||
            char === "\\" ||
            char === "/" ||
            char === "b" ||
            char === "f" ||
            char === "n" ||
            char === "r" ||
            char === "t"
          ) {
            result += char; // Oops! Here we should append the real control character
            i++;
          }

In lossless-json I solved this by having a map escapeCharacters and using the value from the map, holding the real control characters, see code:

        if (text[i] === '\\') {
          const char = text[i + 1]
          const escapeChar = escapeCharacters[char]
          if (escapeChar !== undefined) {
            result += escapeChar
            i++

const escapeCharacters = {
  '"': '"',
  '\\': '\\',
  '/': '/',
  b: '\b',
  f: '\f',
  n: '\n',
  r: '\r',
  t: '\t'
}