Why does str_replace not correctly replace these extended ascii characters?

Question

This is a UTF-8 encoded source file. I must be missing something obvious, but I've tried all the permutations I can think of.

<?php

$bad =  array( chr(130), chr(145), chr(146), chr(147), chr(148), chr(150), chr(151), chr(173), chr(160) );
$good = array( chr( 44), chr( 39), chr( 39), chr( 34), chr( 34), chr( 45), chr( 45), chr( 45), chr( 32) );

print_r($bad);
print_r($good);

 $str = <<<EOF
    bad comma ‚
    bad quote ‘
    bad quote ’
    bad quote “
    bad quote ”
    bad dash –
    bad dash —
    bad dash ­
    bad space   

EOF;
    echo $str;
    $clean = str_replace($bad, $good, $str);
    echo "\n";
    echo( $clean);

And when I open it in a browser and view source...

Array
(
    [0] => ‚
    [1] => ‘
    [2] => ’
    [3] => “
    [4] => ”
    [5] => –
    [6] => —
    [7] => ­
    [8] =>  
)
Array
(
    [0] => ,
    [1] => '
    [2] => '
    [3] => "
    [4] => "
    [5] => -
    [6] => -
    [7] => -
    [8] =>  
)
    bad comma ‚
    bad quote ‘
    bad quote ’
    bad quote “
    bad quote â€
    bad dash –
    bad dash —
    bad dash ­
    bad space   

    bad comma ‚
    bad quote ‘
    bad quote ’
    bad quote “
    bad quote â€
    bad dash â€"
    bad dash â€"
    bad dash Â-
    bad space   

Answer 1

There is a difference between ASCII, Multibyte and UTF-8.

In your case those characters are Multibyte Characters. "Multibyte" is just an arbitrary non-UTF-8 encoding.

you could convert them to UTF-8 first or use mb_ereg_replace

Most PHP Devs don't know much about Character Encoding, but it's one of the most important things when working in C/C++.