PHP regex doesn't working in my website but works on regular expression editor

Question

my aim

for the "author name input", to be sure that only spaces and utf-8 letters are inputted. my website lang is Turkish, Turkish alphabet has non-English characters.

my weird issue

this regex works on rubular.com
if input string : "Selim Çınar" result : matches
if input string : "Selim Çınar 12" result : don't match
regex: /^[\p{L} ]+$/u

then I created trial.php on my website and run codes below

1

echo '<br /><br /><br />';
    $str ='Selim Cinar';
    if (!preg_match("/^[\p{L} ]+$/u", $str))
    {echo 'no, not only utf-8 letters and spaces';}
    else {$str.' yes utf-8 letters and spaces';}
echo '<br /><br /><br />';

result for code above : empty page with only <br /> tags at source page

2

echo '<br /><br /><br />';
    $str ='Selim Çınar'; //includes Tr characters
    if (!preg_match("/^[\p{L} ]+$/u", $str))
    {echo 'no, not only utf-8 letters and spaces';}
    else {$str.' yes utf-8 letters and spaces';}
echo '<br /><br /><br />';

result for code above : empty page with only <br /> tags at source page

3

code source: http://php.net/manual/en/function.ctype-alpha.php

    $str ='Selim Çınar'; //includes Tr characters
    $str =trim($str);
    $str = str_replace(' ', '', $str);
    setLocale(LC_CTYPE, 'TR_tr.UTF-8');
    if (ctype_alpha($str)) {echo 'yes utf-8 letters';}
    else {echo 'no, not only utf-8 letters';}

result for code above : no, not only utf-8 letters

4

code source: http://php.net/manual/en/function.ctype-alpha.php

    $str ='Selim Cinar';
    $str =trim($str);
    $str = str_replace(' ', '', $str);
    setLocale(LC_CTYPE, 'TR_tr.UTF-8');
    if (ctype_alpha($str)) {echo 'yes utf-8 letters';}
    else {echo 'no, not only utf-8 letters';}

result for code above : yes utf-8 letters

my phpinfo

PHP Version 5.4.10
Apache 2.0 Handler
Apache API version: 20051115
PCRE (Perl Compatible Regular Expressions) Support enabled
PCRE Library Version 8.20 2011-10-21

about trial.php

trial.php is pure php. no html header declaration.

My Questions

1. why am I getting empty page for case 1 and case 2 {! UPDATE : SOLVED by MikeM below }
2. Why case 3 doesn't understand "SelimÇınar" as utf-8? Is my code false (setLocale part maybe)?

UPDATE

question 1 is solved by MikeM.

**question 2 still exists as an question.

Answer 1

You are only getting an empty page for cases 1 and 2 because the regex successfully matches $str and therefore the else branch is executed, but there is no echo there so nothing is printed.

I don't know the answer to your second question. The setLocale looks okay to me, but its behaviour is system dependent.

Answer 2

Do not use setlocale or utf8_decode, your problem is very simple in that your php source files are not saved in UTF-8. This depends on your text editor.

This is what WILL work when you saved your file correctly:

$str = 'Selim Çınar'; //Since this is a string literal, its encoding is determined by
                     //how this source file was saved
if (preg_match("/^[\p{L} ]+$/u", $str)) {
    echo 'yes only utf-8 letters or spaces';
} else {
    echo 'no, not only utf-8 letters or spaces';
}