HTML to XHTML

poncingbury

Hi there,

I'm new to the forum and was wondering if anyone would be kind enough to help us with this one:

I want to convert any UPPERCASE HTML tag to a lowercase XHTML tag without effecting non-html markup, preferably using a function.

For exmaple:

$input = "This is a tag";

converted to the following:

$output = "This is a tag";

The following works but I need it to convert any tag not just a specific tag:

$input = ereg_replace (" "," ",$input );

All help greatly greatly appreciated.

Cheers

tomkleijkers

Just search for tag:


preg_match("/<(.+)\/>/", $input, $match);
$output = '<' . strtolower($match[1]) . '/>';

I think this is what you want... I don't know if it's the best solution but i guess it works...

Weedpacket

Get friendly with preg_replace and never look at ereg_replace again...

$text = preg_replace("#<(/?)(\w+)\b#e", "'<$1'.strtolower('$2')", $text);

Norman_Graham

You might also be interested in HTML Tidy, which appears to take XHTML into account, although I've never used it myself.

http://www.w3.org/People/Raggett/tidy

HTH

Norm

poncingbury

Hi there guys,

Thanks for the pointers, I tried the following as suggested but it also converts the rest of the text to lower case, I just want it to convert the <HTML> tags and leave the rest of the text as it was.

<?php

$text = "Here is some text to convert ";

function convert_to_xhtml(){
global $text;
if (preg_match("/<(.+)>/", $text, $match)){
$text = '<' . strtolower($match[1]) . '>';
}
}

convert_to_xhtml();
print $text;

Any thoughts?

Cheers

Weedpacket

I think it doesn't work and you maybe should try something else. Maybe that HTML Tidy suggestion. I use it and it generalyl works quite well.

drawmack

the problem is that you're matching the first < and the last > I think Weed's suggestion will work.

drawmack

Originally posted by Weedpacket
Get friendly with preg_replace and never look at ereg_replace again...
$text = preg_replace("#<(/?)(\w+)\b#e", "'<$1'.strtolower('$2')", $text);
[/B]

Weed,
I'd like to addapt this to something that I need but I'm not sure I understand the regular expression well enough to do so. Could you please explain the regex?

poncingbury

Hi all,

The thing is all this is happening dynamically, the HTML is being parsed via a web-based content management system (fed by MySQL D😎 so I can't use TIDY etc...

If it were happening offline I would use dreamweaver's convert to XHTML feature.

Anyway thanks for all your help. Any more cool ideas much appreciated!

Cheers

Weedpacket

Originally posted by drawmack
Weed,
I'd like to addapt this to something that I need but I'm not sure I understand the regular expression well enough to do so. Could you please explain the regex?

Well, everything's covered in the manual, but...

# is the start of the regexp. / is traditional, but I wanted to use that in the regexp itself and I was too lazy to go \/.
< is the start of an HTML tag...
b[/b] ...which might be immediately followed with a /. We want to remember if it is or not, so we color=darkblue[/color] it.
b[/b] One or more wordlike-characters. \w is basically short for [a-zA-Z0-9_]. That covers the names all of the HTML tags. We remember this as well.
\b This is a "zero-width" match. It doesn't actually match any character in the string; instead it matches the boundary between a wordlike character and a nonworlike character - the beginning or end of a word, in other words. May not be strictly necessary in this particular instance, but it's convenient.
# The end of the regexp
e Executable modifier. Take the string produced by the replacement text and run it as PHP code, and use the result of that as the replacement string.

"'<$1'.strtolower('$2')" After the $1 and $2 have been substituted with the bits we remembered (the optional / and the tag name, respectively), this bit of code is evaluated to determine the string that will replace the matched text. In this case, a <, possibly followed by a /, and then a (lower-case) word.

planetsim

regex always confuses me.. i should probably read up on it again sometime soon.. anyone know of good sites...

oh and weedpacket.. nice explanation..

drawmack

thanks weed,
it was the # and # throwing me off.

Norman_Graham

Originally posted by poncingbury
Hi all,

The thing is all this is happening dynamically, the HTML is being parsed via a web-based content management system (fed by MySQL D😎 so I can't use TIDY etc...

If it were happening offline I would use dreamweaver's convert to XHTML feature.

Anyway thanks for all your help. Any more cool ideas much appreciated!

Cheers

B

Yeah, but somewhere, somehow, the XHTML tags are written or produced in some way, either in the PHP or in the DB content - they don't appear purely by magic. Run the W3C XHTML checker on the actual browser output and print out the list of errors - then you'll have a working basis.

HTH

Norm

poncingbury

The HTML is generated via a web-based (WYSIWIG) textfield called HTMLAREA (http://www.interactivetools.com/) and then gets sent to MySQL DB, the only thing is that the HTMLAREA creates HTML not XHTML (can't change this either)...

I need the HTMLAREA and not a traditional form field as it is for clients to edit HTML in WYSIWIG mode via their web browsers.

Basically the HTMLAREA is the root of the problem in that it outputs HTML and not XHTML. The developers say they'll sort this is the next release.

Anyway thanks for all your help guys!

Cheers

Norman_Graham

You could, of course, create your own HTMLAREA - if you look at the reply box on this forum you'll see an example of a DIY version. However, by the time you design and implement this, Interactive Tools may have solved the problem themselves.

Norm