PHP Unicode to character conversion - php

Question

I receive country names like from a library: "\u00c3\u0096sterreich".
How do I convert this to Österreich?
Using PHP 7.3

Answer 1

This one is a lot trickier than it seem, but the below code appears to work.
First we pipe it through the standard regex for Unicode escape sequences, then pack that as a binary string, convert the encoding and finally decode. I cannot promise this is the best way to do this, but it appears to be working correct as far as I can tell.
$str = '\u00c3\u0096sterreich';
$str = preg_replace_callback('/\\\\u([0-9a-fA-F]{4})/', function ($match) {
return utf8_decode(mb_convert_encoding(pack('H*', $match[1]), 'UTF-8', 'UCS-2BE'));
}, $str);
Demo here

Answer 2

The Unicode for the UTF-8 character "Ö" is U+00D6.
This character consists of the 2 hex bytes: c3 and 96.
The representation \u00c3 \u0096 for these 2 bytes is a bit strange. Provided that the multibyte character is represented byte for byte, the following code can also be used.
$str = '\u00c3\u0096sterreich';
$str = preg_replace_callback(
'~\\\\u00([0-9a-f]{2})~i',
function($m){
return hex2bin($m[1]);
},
$str
);
//Test
$expect = "Österreich";
var_dump($str === $expect); //bool(true)

Answer 3

In case anyone else ends up here with a similar issue, I thought I'd try and shed some light on what's going on. Because as mentioned, this is a lot more complicated that it might look.
A string like \u00c3 refers to a Unicode code-point, in hexadecimal. Ö in the Unicode table is character 214, or \u00d6.
The 214 here is not directly related to how Ö is actually stored in any particular encoding (UTF-8, UTF-16, etc), it's just an abstract number in the overall Unicode table that refers to that character. UTF-8, for instance, will store it in two bytes 11000010 10010110 (194 150 in decimal). There's a really good explanation of how this works in this answer, if you're interested in the finer details.
What appears to have happened in your string is that these two bytes have then been encoded back into hexadecimal, and returned as two separate Unicode code points. u00c3 is Ã, and \u0096 is a control character. This is why any standard methods of decoding this (json_decode, etc) won't have worked - ultimately what you have is not a valid representation of the string Österreich.
The other answers should both work perfectly well, but this code snippet might better illustrate the issue with the format your library is using. It specifically matches two consecutive low Unicode code-points, recombines their decimal representations into an unsigned two-byte integer, and then returns the result.
$str = '\u00c3\u0096sterreich';
echo preg_replace_callback('/\\\\u00([0-9a-fA-F]{2})\\\\u00([0-9a-fA-F]{2})/', function ($match) {
$i = (hexdec($match[1]) << 8) + hexdec($match[2]);
return pack('N', $i);
}, $str);
Österreich
See https://3v4l.org/QtUuGD