mb_check_encoding($in_str,"UTF-8") return different results?

blackhorse

I am running two servers.

Linux server with mysql 4.+ and php 5.

Zend server with mysql 5 and php 5.

I populated the database on both server with the same mysql dump file.

But when I grab the data from the database, and use

mb_check_encoding($in_str,"UTF-8")

on these values.

Zend Server return FALSE which is the right one, but linux server return TRUE.

What is going on? Does it have something to do with php set up? It shouldn't, right?

Thanks!

blackhorse

This is from php manual's posts

I found this little function very useful in fixing strings that are not in utf-8 but need be converted

<?php
// Fixes the encoding to uf8
function fixEncoding($in_str)
{
  $cur_encoding = mb_detect_encoding($in_str) ;
  if($cur_encoding == "UTF-8" && mb_check_encoding($in_str,"UTF-8"))
    return $in_str;
  else
    return utf8_encode($in_str);
} // fixEncoding
?>

I have the data from the database with some utf-8 characters.

On Zend Server

mb_detect_encoding($in_str) returns UTF-8
but
mb_check_encoding($in_str,"UTF-8") returns FALSE.

So this function still calls utf8_encode on these values, and that is the result I want.

But on linux server, same data, mb_check_encoding($in_str,"UTF-8") returns TRUE, so utf8_encode is not called. And the result is not what I wanted.

blackhorse

 $cur_encoding = mb_detect_encoding($in_str) ;
  if($cur_encoding == "UTF-8" && mb_check_encoding($in_str,"UTF-8"))
    return $in_str;
  else
    return utf8_encode($in_str);

First question:

I don't understand why we need use both functions here, mb_detect_encoding($in_str) and mb_check_encoding($in_str,"UTF-8")?

Which case will be caught by mb_detect_encoding($in_str) but not mb_check_encoding($in_str,"UTF-8") and vice versa?

Or will mb_check_encoding($in_str,"UT-8") catch all the

mb_detect_encoding($in_str) == "UTF-8", so we only need mb_check_encoding($in_str,"UT-8") ?

Second question:

If mb_check_encoding($in_str,"UT-8") works, why there are so many people trying to write their own codes to test if a string is utf-8 or not, if you see the posts below http://us2.php.net/manual/en/function.utf8-encode.php? That means mb_check_encoding($in_str,"UT-8") only works on some cases, what is this function's limit?

Third question:

Even with all the limits, still, this function applies on the data from databases created from same mysql dump file, should return the same results, but like in my case, it gives different results in my linux server (php5, mysql 4) and zend server (php5, mysql 5).