inverse function of unraw

[milahu]

what is a fast inverse function of unraw?

the String.raw tagged-template only works with literals, not with dynamic strings

do we have something better than this?

var backslashEncode = (function backslashEncodeFactory() {
  const escapeCode = Array(32);
  for (let i = 0; i < 32; i++)
    escapeCode[i] = '\\x'+i.toString(16).padStart(2, '0');
  escapeCode[0] = '\\0';
  escapeCode[8] = '\\b';
  escapeCode[9] = '\\t';
  escapeCode[10] = '\\n';
  //escapeCode[10] = '\n'; // dont escape newline
  escapeCode[11] = '\\v';
  escapeCode[12] = '\\f';
  escapeCode[13] = '\\r';
  
  return function backslashEncode(str) {
    let res = '';
    for (let i = 0; i < str.length; i++) {
      const char16bit = str[i];
      const code = char16bit.charCodeAt(0);
      res += (
        (code < 32) ? escapeCode[code] : // ascii control
        (code == 92) ? '\\\\' :
        (code < 128) ? char16bit : // ascii printable
        '\\u'+code.toString(16).padStart(4, '0') // unicode
      );
    }
    return res;
  }
})();

fails

String.raw`\u5A5A` == '\\u5A5A'

String.raw({ raw: ['\u5A5A'] }) == '婚'

String.raw({ raw: ['', ''] }, '\u5A5A') == '婚'

JSON.stringify("\u5A5A") == '"婚"'

edit: here is another library with encode + decode fns: Shakeskeyboarde/slashes
but that one fails to encode unicode chars

[Ian Sanders]

A raw function that would work on dynamic strings would necessarily have different behavior from the built-in String.raw because there's no way for JS to distinguish between the string " " and the string "\t" after the string is created. Those become the same string (" " === "\t" is true).

For example (with some theoretical raw function):

const a = "\t";
const b = "	";

raw(a); // "\t"
raw(b); // "\t" even though the string was not made using \t, because there's no way to tell the two apart

// On the other hand:
String.raw`\t`; // "\t"
String.raw`	`; // "	"

So this is not really in the scope of this particular project because this project intends only to invert the String.raw function.

I guess what you are really looking for is a function that will take a UTF-16 string and break it down into the ASCII representation using backslashes. Maybe he satisfies this? You can use the allowUnsafeSymbols to disable encoding of basic symbols like &.

[milahu]

hmm. i thought unraw is just a backslashDecode function
identical to this stripSlashes function, or PHP's stripslashes function

he looks useful, but i dont see any backslashEncode function there ...?

i want to encode strings to their "ascii javascript literal string" representation, similar to JSON.stringify

JSON.stringify("\0 \b \t \n \v \f \r \ud83d\ude0a \" '")
== `"\\u0000 \\b \\t \\n \\u000b \\f \\r 😊 \\" '"`

... but json strings aint ascii, and doublequotes are escaped

so to answer my question

do we have something better than this?

nope

[Ian Sanders]

The problem with string encoding/decoding is there's just so many different use cases and standards. I'm sure what you're looking for is out there, but knowing how to find it is really difficult.