Regex Internationals (round 2)

[Shawazi]

I have enabled people to enter international characters using JS, but am unable to replicate it in PHP.

This Validates

if (!fieldVal.match(/^[\w\d\xC0-\xD6\xD8-\xF6\xF8-\xFF&_@#\.?!:;,\/\(\)\'’ -]+$/)) 
			innerVal = "Only letters, numbers, underscores, dashes, and punctuation may be used.";

This Does Not

if (!preg_match("/^[\w\xC0-\xD6\xD8-\xF6\xF8-\xFF\d&_\.?!@#:;,\/\(\)'’ -]+$/",$thevalue))
		return "Bad";
	else
		return "Good";

[NogDog]

I suspect the problem is somewhere else in the code logic, as this works fine for me:

<?php
function validate($thevalue)
{
   if (!preg_match("/^[\w\xC0-\xD6\xD8-\xF6\xF8-\xFF\d&_\.?!@#:;,\/\(\)'’ -]+$/",$thevalue))
      return "Bad";
   else
      return "Good";
}
$testData = array(
   "Test 1 (good): " . chr(0xC0) . chr(0xD6),
   "Test 2 (bad):  " . chr(0xD7) . chr(0xF7)
);
foreach($testData as $test)
{
   echo $test . ' = ' . validate($test) . "<br>\n";
}

[Shawazi]

Oh Yeah I'm sorry the test string I was using is:

Reykjavík

Notice the i is not a normal one. When I first used the hex code it was able to get the basic international characters but not all of them when I tried random ones throughout the character map - well random to the extend i chose ones that are English-like (excluding chinese, hebrew, etc)

[NogDog]

Just a thought: you might want to request that the form not send any multi-byte characters by using the "accept-charset" attribute in the <form> tag, e.g.:

<form action="form.php" method="post" accept-charset="ISO-8859-1">
...or...
<form action="form.php" method="post" accept-charset="UTF-8">

Otherwise, if you want to be able to accept and process mutli-byte characters, then you'll need to enable the [man]mbstring[/man] extension and use the [man]mb_ereg_match/man.

(I'm by no means sure this is the source of your problem; it's just a hypothesis.)