#41980 [Fbk->Opn]: encoding problem

borys dot forytarz at gmail dot com Thu, 12 Jul 2007 13:48:21 -0700

 ID:               41980
 User updated by:  borys dot forytarz at gmail dot com
 Reported By:      borys dot forytarz at gmail dot com
-Status:           Feedback
+Status:           Open
 Bug Type:         DOM XML related
 Operating System: Linux
 PHP Version:      5.2.0
 New Comment:


I have checked about files encodings.

mb_detect_encoding() returns, that they are ASCII-encoded (!?). So I
wrote a simple script to convert them to utf-8:

<?php
$cont = file_get_contents('login.php.tpl');
$f = fopen('login.php.tpl','w');
echo "\n".mb_detect_encoding('login.php.tpl').' > ';
fwrite($f,mb_convert_encoding($cont,'utf-8'));
echo mb_detect_encoding('login.php.tpl')."\n";
fclose($f);
?>

and the output is: ASCII > ASCII (I expected ASCII > UTF-8)

result of using iconv instead of mb_convert_encoding is the same

what's going on?


Previous Comments:
------------------------------------------------------------------------

[2007-07-12 20:38:33] [EMAIL PROTECTED]

Please try using this CVS snapshot:

  http://snaps.php.net/php5.2-latest.tar.gz
 
For Windows (zip):
 
  http://snaps.php.net/win32/php5.2-win32-latest.zip

For Windows (installer):

  http://snaps.php.net/win32/php5.2-win32-installer-latest.msi



------------------------------------------------------------------------

[2007-07-12 19:58:58] borys dot forytarz at gmail dot com

there should be:

...
foreach($content->childNodes as $child) {
...

sorry

------------------------------------------------------------------------

[2007-07-12 19:55:58] borys dot forytarz at gmail dot com

Here is an example:

At first, source files (both encoded with UTF-8)

First file (main.tpl):

<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN"
"http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd";>
<html>

<head>
        <title>Some title</title>
        <meta http-equiv="content-type" content="text/html; charset=utf-8" />
</head>

<body>
Some polish letters: &#281; ó &#261; &#347; &#263; &#380; &#378; &#324;
- they are encoded correctly and displays correctly.
</body>
</html>

Second file (contents.tpl):

<content>
<h1>some polish letters, like: &#281; ó &#322; &#261; &#347; &#263;
&#378; &#324; &#380; - they are not encoded correctly and does not
display correctly.</h1>
</content>



PHP file:
<?php
$dom = new DOMDocument('1.0','UTF-8');
$dom->loadHtmlFile('main.tpl');

$dom2 = new DOMDocument('1.0','UTF-8');
$dom2->loadHTMLFile('contents.tpl');

$contents = $dom2->getElementsByTagName('content');
$body = $dom->getElementsByTagName('body')->items(0);

foreach($contents as $content) {
    foreach($content as $child) {
        $imp = $dom->importNode($child,true);
        $body->appendChild($imp);
    }
}

$dom->saveXML();
?>

It is something like above. I was writing from memory because the real
script is really huge. But it demonstrates the idea and what is going
not properly.

------------------------------------------------------------------------

[2007-07-12 19:24:45] borys dot forytarz at gmail dot com

Description:
------------
There is a problem with DOM and encoding. I have two separate files,
one full XHTML code (DTD, head, meta, body and more contents) saved in
UTF-8. Meta declaration is UTF-8, server sends the code in UTF-8 too.
The second file is a simple file without any DTD, head, meta and body.
Saved in UTF-8 too. The problem is, when I import nodes from the second
file using importNode(), in the output there are invalid encoded
characters (those who were declared in the second file). It is strange
because as I read, DOM works in UTF-8 so there should be not such a
problem.

What is more, I was debugging the properties such as actualEncoding and
they shown me that there is UTF-8...

If it's not a bug, but I think it is, how to fix that? I can't declare
in the second file DTD, head and body elements.

Reproduce code:
---------------
$this->dom = new DOMDocument('1.0','UTF-8');
$this->dom->encoding = 'UTF-8';

$this->dom->formatOutput = self::$formatOutput;
$this->dom->preserveWhiteSpace = self::$preserveWhiteSpace;
@$this->dom->loadHtmlFile($html);

...

echo $this->dom->saveXML();

The above works well for the complete XHTML file. But when I load an
incomplete file (encoded in UTF-8) I don't see properly encoded
characters when I import nodes from the second document to the first
one.

I tried to convert the whole output with iconv() and
mb_convert_encoding() but it seems not to make any difference at all.

Expected result:
----------------
Properly encoded characters from both complete XHTML file and second
"poor" file. The second file is such as follows:

<content id="something">
   <h1>some string</h1>
</content>

Actual result:
--------------
Not properly encoded characters from between <content> tag.


------------------------------------------------------------------------


-- 
Edit this bug report at http://bugs.php.net/?id=41980&edit=1

#41980 [Fbk->Opn]: encoding problem

Reply via email to