Question

For example this:

<!-- All the characters are going to be converted into a Hex values depending the encoding used -->

<meta http-equiv="Content-Type" content="text/html; charset=UTF-8" /> <!-- It Just interpret the Hex values that are going to be displayed -->


<?php

/* PHP Strings are bytestream */
/* PHP treat the strings as a Hex values from the econding used */

$string="€"; // Hex value from the Encoding Method(UTF-8). [U+20AC][E2|82|AC]
if(preg_match('/\xE2\x82\xAC/',$string,$m)){
    echo "Match<br>";
    print_r($m);
    }
else{
    echo "Don't Match";
    }

?>

As long as you use correct bytes secuences to Match Unicode Characters.
Is not needed to use Unicode Support?

or is that I’m thinking wrong?

Answer 1

For that particular match, you don’t need Unicode support. Any simple direct string match will work for two UTF-8 strings​—that was a deliberate design feature of UTF-8—but then you wouldn’t be using regex if all you needed was a direct string match: for your example you’d be better off with strpos.

Many other regex features will behave unexpectedly without Unicode support. For example:

/€*/

with Unicode support, that’s multiple € signs (\xE2\x82\xAC\xE2\x82\xAC\xE2\x82\xAC...). Without it, that’s the first two bytes of a € symbol then any number of 0xAC bytes (\xE2\x82\xAC\xAC\xAC\xAC...), so the only valid UTF-8 sequence it would match would be a single €.

/[x€]/

with Unicode support, matches x or a euro. Without Unicode support, matching x or the byte 0xE2 or the byte 0x82 or the byte 0xAC.

And so on.