Replacing characters, preg_replace_callback?

[ZiLenT-X]

Does anybody out there have a clue how this could be done?

I have a page that reads out HTML from files and echo the HTML into the site.

$handle = fopen($FilePath, "r");
$contents = fread($handle, filesize($FilePath));
fclose($handle);

In this HTML page there can be img tags with norwegian charracters (øæå) i want to rewrite these img tags so that the norwegian charracters are replaced with charracters that work with Firefox. It works in IE and Opera.

If i do this to the file that is read out to $contents

$contents = str_replace("å", "%c3%a5", $contents);
$contents = str_replace("ø", "%c3%b8", $contents);
$contents = str_replace("æ", "%c3%a6", $contents); 

i get the img tags to work. But then all the text in the document is also turned into %c3%a6 and not readable.. :-(

So i only want to change the charracters inside the img tag.
Want the link from this:

<img width=209 height=149
src="Trådløst nettverk/Trådløs%20guide%20windows%20XP-filer/image002.jpg" align=left hspace=12>

to this

<img width=209 height=149
src="Tr%c3%a5dl%c3%b8st nettverk/Tr%c3%a5dl%c3%b8s%20guide%20windows%20XP-filer/image002.jpg" align=left hspace=12>

i have tryed this

$contents = preg_replace('/\<img (.*?)\>/is', "<img ".urlencode("$1").">", $contents);

and this

function url_encode($matches){
return "<img$matches[1] src=\"". str_replace(array("æ", "ø", "å"), array("%c3%a5", "%c3%b8","%c3%b8"), $matches[2]) ."\"$matches[3]>";
}

$contents = preg_replace_callback("#\<img(.*?) src=\"(.*?)\"(.*?)\>#", "url_encode", $contents);

and it did not work..

This is starting to "bug" me bigtime, so if anyone have a clue PLEASE HELP!

[NogDog]

EDIT: Never mind, this would change the HTML markup tags, too. 🙁

It might be simpler to just use the htmlentities() function.

[NogDog]

How about this?

echo preg_replace('/[\x80-\xFF]/e', "htmlentities($0)", $text);

[ZiLenT-X]

Thanks! But it did not work.

Now the link looks like this:

<img width=209 height=149
src="Tr&Atilde;&yen;dl&Atilde;&cedil;st nettverk/Tr&aring;dl&oslash;s%20guide%20windows%20XP-filer/image002.jpg" align=left hspace=12>

[bpat1434]

I think you'd need to use the preg_replace_callback() function like this:

function translate($matches){
	$string = $matches[0];
	$string = str_replace($matches[2], '{REPLACE}', $string);

$temp = urlencode($matches[2]);

$string = str_replace('{REPLACE}', $temp, $string);

return $string;
}

$pattern = "|<img(.*?)src=\"(.*?)\"([^>]*)>|";
$contents2 = preg_replace_callback($pattern,
					'translate',
					$contents);

That works for me and outputs:

Original:
<img width=209 height=149 src="Trådløst nettverk/Trådløs guide windows XP-filer/image002.jpg" align=left hspace=12>Trådløst nettverk
Trådløs guide windows XP-filer

Transformed:
<img width=209 height=149 src="Tr%E5dl%F8st+nettverk%2FTr%E5dl%F8s+guide+windows+XP-filer%2Fimage002.jpg" align=left hspace=12>Trådløst nettverk
Trådløs guide windows XP-filer

---

Difference: Different

EDIT:
You can probably trim it down to something like this (the translate function that is):

function translate($matches){
	$string = $matches[0];
	return str_replace($matches[2], urlencode($matches[2]), $string);
}

Either way works just as well.

[ZiLenT-X]

Thanks. But, the function removes the slashes from the link.

<img width=209 height=149 src="Dok%2FStudenter%2FTr%C3%A5dl%C3%B8st+nettverk%2FTr%E5dl%F8s%2520guide%2520windows%2520XP-filer%2Fimage002.jpg" align=left hspace=12>

So it doesent work. Is there any way to do this whitout the use for urlencode?

And just replace the norwegian characters?

$contents = str_replace("å", "%c3%a5", $contents);
$contents = str_replace("ø", "%c3%b8", $contents);
$contents = str_replace("æ", "%c3%a6", $contents); 

[bpat1434]

In my original function just replace the one urlencode() line with your three str_replace lines 😉