reg exp question: removing class and style from html

**Justin Koivisto** · Jul 17 '05, 07:07 AM

Re: reg exp question: removing class and style from html

chotiwallah wrote:
[color=blue]
> i have a little database driven content managment system. people can
> load up html-docs. some of them use ms word as their html-editor,
> which resultes in loads of "class" and "style" attributes - like this:
>
> Some text
>
> now i'd like to remove them (the attributes, not the people, that is).
> i know reg exp is the way, but somehow the solution avoids me. just
> pointing me to some more advanced tutorial (pref. in german) would
> help a lot.
>
> any help appreciated, micha[/color]

Something like this may help you:

$pattern='`(cla ss|style)=(\\\' |\\").*\\2/Ui';
$str=preg_repla ce($pattern,'', $str);

Here are some of the things I've been using:

Regular-Expressions.info - Regex Tutorial, Examples and Reference - Regexp Patterns

http://www.regular-expressions.info/

At Regular-Expressions.info you will find a wide range of in-depth information about a powerful search pattern language called regular expressions.

Faculty of Engineering and Physical Sciences | Page not found

http://www.comp.leeds.ac.uk/Perl/matching.html

Perl regular expressions for the common man

http://www.anaesthetist.com/mnm/perl/regex.htm

Perl regex, regular expression, extracting information, text processing

And there are also tools like these:

The Regex Coach - interactive regular expressions

http://www.weitz.de/regex-coach/

Visual REGEXP : a graphical explorer for your regexps

http://laurent.riesterer.free.fr/regexp/

regexp, regular, expression, debugging, graphical

Sorry, none are German, but you may be able to translate withBabblefish:

http://babblefish.com/babblefish/language_webt.htm

--
Justin Koivisto - spam@koivi.com
PHP POSTERS: Please use comp.lang.php for PHP related questions,
alt.php* groups are not recommended.

**Josip** · Jul 17 '05, 07:07 AM

Re: reg exp question: removing class and style from html

On 12 Jul 2004 09:07:23 -0700, chotiwallah@web .de (chotiwallah) wrote:
[color=blue]
>i have a little database driven content managment system. people can
>load up html-docs. some of them use ms word as their html-editor,
>which resultes in loads of "class" and "style" attributes - like this:
>[/color]

This is written in JavaScript... It works great!!!!

if ( (D.indexOf('cla ss=Mso') >= 0) || (D.indexOf('cla ss="Mso') >=
0) ) {

// make one line
D = D.replace(/\r\n/g, ' ').
replace(/\n/g, ' ').
replace(/\r/g, ' ').
replace(/\&nbsp\;/g,' ');

// keep tags, strip attributes
D = D.replace(/ class=[^\s|>]*/gi,'').

replace(/ style=\"[^>]*\"/gi,'').
replace(/ align=[^\s|>]*/gi,'');

//clean up tags
D = D.replace(/]*>/gi,'').
replace(/ ]*>/gi,' ').
replace(/]*>/gi,'').
replace(/<li [^>]*>/gi,'<li>').
replace(/<ul [^>]*>/gi,'<ul>');

// replace outdated tags
D = D.replace(//gi,'').
replace(/<\/b>/gi,'');

// mozilla doesn't like tags
D = D.replace(//gi,'').
replace(/<\/em>/gi,'');

// kill unwanted tags
D = D.replace(/<\?xml:[^>]*>/g, ''). // Word xml
replace(/<\/?st1:[^>]*>/g,''). // Word SmartTags
replace(/<\/?[a-z]\:[^>]*>/g,''). // All other funny Word
//non-HTML stuff
replace(/<\/?font[^>]*>/gi,''). // Disable if you want to
//keep font formatting
replace(/<\/?span[^>]*>/gi,' ').
replace(/<\/?div[^>]*>/gi,' ').
replace(/<\/?pre[^>]*>/gi,' ').
replace(/<\/?h[1-6][^>]*>/gi,' ');

//remove empty tags
//D = D.replace(/<\/strong>/gi,'').
//replace(/<\/i>/gi,'').
//replace(/<P[^>]*><\/P>/gi,'');

// nuke double tags
oldlen = D.length + 1;
while(oldlen > D.length) {
oldlen = D.length;
// join us now and free the tags, we'll be free hackers,
//we'll be free... ;-)
D = D.replace(/<([a-z][a-z]*)> *<\/\1>/gi,' ').
replace(/<([a-z][a-z]*)> *<([a-z][^>]*)> *<\/\1>/gi,'<$2>');
}
D = D.replace(/<([a-z][a-z]*)><\1>/gi,'<$1>').
replace(/<\/([a-z][a-z]*)><\/\1>/gi,'<\/$1>');

// nuke double spaces
D = D.replace(/ */gi,' ');

**Tim Van Wassenhove** · Jul 17 '05, 07:07 AM

Re: reg exp question: removing class and style from html

In article <782d6cb.040712 0807.16d05ffc@p osting.google.c om>, chotiwallah wrote:[color=blue]
> i have a little database driven content managment system. people can
> load up html-docs. some of them use ms word as their html-editor,
> which resultes in loads of "class" and "style" attributes - like this:
>
>Some text
>
> now i'd like to remove them (the attributes, not the people, that is).
> i know reg exp is the way, but somehow the solution avoids me. just
> pointing me to some more advanced tutorial (pref. in german) would
> help a lot.[/color]

As soon as you get to deal with nested tags etc, regular expressions
aren't that handy anymore. You could have a look at Tidy (a websearch
will help you). You buffer all the output, clean it up with tidy, and
output the cleaned up code.

--
Tim Van Wassenhove <http://home.mysth.be/~timvw>

**Michael Austin** · Jul 17 '05, 07:07 AM

Re: reg exp question: removing class and style from html

Josip wrote:
[color=blue]
> On 12 Jul 2004 09:07:23 -0700, chotiwallah@web .de (chotiwallah) wrote:
>
>[color=green]
>>i have a little database driven content managment system. people can
>>load up html-docs. some of them use ms word as their html-editor,
>>which resultes in loads of "class" and "style" attributes - like this:
>>[/color]
>
>
> This is written in JavaScript... It works great!!!!
>
>
>
> if ( (D.indexOf('cla ss=Mso') >= 0) || (D.indexOf('cla ss="Mso') >=
> 0) ) {
>
> // make one line
> D = D.replace(/\r\n/g, ' ').
> replace(/\n/g, ' ').
> replace(/\r/g, ' ').
> replace(/\&nbsp\;/g,' ');
>
> // keep tags, strip attributes
> D = D.replace(/ class=[^\s|>]*/gi,'').
>
> replace(/ style=\"[^>]*\"/gi,'').
> replace(/ align=[^\s|>]*/gi,'');
>
> //clean up tags
> D = D.replace(/]*>/gi,'').
> replace(/ ]*>/gi,' ').
> replace(/]*>/gi,'').
> replace(/<li [^>]*>/gi,'<li>').
> replace(/<ul [^>]*>/gi,'<ul>');
>
> // replace outdated tags
> D = D.replace(//gi,'').
> replace(/<\/b>/gi,'');
>
> // mozilla doesn't like tags
> D = D.replace(//gi,'').
> replace(/<\/em>/gi,'');
>
> // kill unwanted tags
> D = D.replace(/<\?xml:[^>]*>/g, ''). // Word xml
> replace(/<\/?st1:[^>]*>/g,''). // Word SmartTags
> replace(/<\/?[a-z]\:[^>]*>/g,''). // All other funny Word
> //non-HTML stuff
> replace(/<\/?font[^>]*>/gi,''). // Disable if you want to
> //keep font formatting
> replace(/<\/?span[^>]*>/gi,' ').
> replace(/<\/?div[^>]*>/gi,' ').
> replace(/<\/?pre[^>]*>/gi,' ').
> replace(/<\/?h[1-6][^>]*>/gi,' ');
>
> //remove empty tags
> //D = D.replace(/<\/strong>/gi,'').
> //replace(/<\/i>/gi,'').
> //replace(/<P[^>]*><\/P>/gi,'');
>
> // nuke double tags
> oldlen = D.length + 1;
> while(oldlen > D.length) {
> oldlen = D.length;
> // join us now and free the tags, we'll be free hackers,
> //we'll be free... ;-)
> D = D.replace(/<([a-z][a-z]*)> *<\/\1>/gi,' ').
> replace(/<([a-z][a-z]*)> *<([a-z][^>]*)> *<\/\1>/gi,'<$2>');
> }
> D = D.replace(/<([a-z][a-z]*)><\1>/gi,'<$1>').
> replace(/<\/([a-z][a-z]*)><\/\1>/gi,'<\/$1>');
>
> // nuke double spaces
> D = D.replace(/ */gi,' ');
>[/color]

The problem is that you cannot/should not ever "rely" on javascript as
it may be turned off due it's inherent security risks.

**Matthias Esken** · Jul 17 '05, 07:08 AM

Re: reg exp question: removing class and style from html

chotiwallah schrieb:
[color=blue]
> i know reg exp is the way, but somehow the solution avoids me. just
> pointing me to some more advanced tutorial (pref. in german) would
> help a lot.[/color]

Regenechsen

http://www.regenechsen.de/regex_de/

regular expression, regex, TheBat!, macro, search patterns, cryptology, reguläre Ausdrücke, Regex, Kryptologie

Any more questions? Visit the german speaking newsgroup
de.comp.lang.ph p.misc.

Regards,
Matthias

**chotiwallah** · Jul 17 '05, 07:08 AM

Re: reg exp question: removing class and style from html

Josip <josip@sdfs.s d> wrote in message news:<dvg5f09t1 e4r23bl53oieau6 1dntiuf6ks@4ax. com>...[color=blue]
> On 12 Jul 2004 09:07:23 -0700, chotiwallah@web .de (chotiwallah) wrote:
>[color=green]
> >i have a little database driven content managment system. people can
> >load up html-docs. some of them use ms word as their html-editor,
> >which resultes in loads of "class" and "style" attributes - like this:
> >[/color]
>
> This is written in JavaScript... It works great!!!!
>
>
>
> if ( (D.indexOf('cla ss=Mso') >= 0) || (D.indexOf('cla ss="Mso') >=
> 0) ) {
>
> // make one line
> D = D.replace(/\r\n/g, ' ').
> replace(/\n/g, ' ').
> replace(/\r/g, ' ').
> replace(/\&nbsp\;/g,' ');
>
> // keep tags, strip attributes
> D = D.replace(/ class=[^\s|>]*/gi,'').
>
> replace(/ style=\"[^>]*\"/gi,'').
> replace(/ align=[^\s|>]*/gi,'');
>
> //clean up tags
> D = D.replace(/]*>/gi,'').
> replace(/ ]*>/gi,' ').
> replace(/]*>/gi,'').
> replace(/<li [^>]*>/gi,'<li>').
> replace(/<ul [^>]*>/gi,'<ul>');
>
> // replace outdated tags
> D = D.replace(//gi,'').
> replace(/<\/b>/gi,'');
>
> // mozilla doesn't like tags
> D = D.replace(//gi,'').
> replace(/<\/em>/gi,'');
>
> // kill unwanted tags
> D = D.replace(/<\?xml:[^>]*>/g, ''). // Word xml
> replace(/<\/?st1:[^>]*>/g,''). // Word SmartTags
> replace(/<\/?[a-z]\:[^>]*>/g,''). // All other funny Word
> //non-HTML stuff
> replace(/<\/?font[^>]*>/gi,''). // Disable if you want to
> //keep font formatting
> replace(/<\/?span[^>]*>/gi,' ').
> replace(/<\/?div[^>]*>/gi,' ').
> replace(/<\/?pre[^>]*>/gi,' ').
> replace(/<\/?h[1-6][^>]*>/gi,' ');
>
> //remove empty tags
> //D = D.replace(/<\/strong>/gi,'').
> //replace(/<\/i>/gi,'').
> //replace(/<P[^>]*><\/P>/gi,'');
>
> // nuke double tags
> oldlen = D.length + 1;
> while(oldlen > D.length) {
> oldlen = D.length;
> // join us now and free the tags, we'll be free hackers,
> //we'll be free... ;-)
> D = D.replace(/<([a-z][a-z]*)> *<\/\1>/gi,' ').
> replace(/<([a-z][a-z]*)> *<([a-z][^>]*)> *<\/\1>/gi,'<$2>');
> }
> D = D.replace(/<([a-z][a-z]*)><\1>/gi,'<$1>').
> replace(/<\/([a-z][a-z]*)><\/\1>/gi,'<\/$1>');
>
> // nuke double spaces
> D = D.replace(/ */gi,' ');[/color]

brilliant little script, works great, exactly what i needed :-).
i hadn't even thought of doing the whole cleaning clientside.

thanks a bundle to everyone, micha

**Michael Austin** · Jul 17 '05, 07:12 AM

Re: reg exp question: removing class and style from html

chotiwallah wrote:[color=blue]
> Josip <josip@sdfs.s d> wrote in message news:<dvg5f09t1 e4r23bl53oieau6 1dntiuf6ks@4ax. com>...
>[color=green]
>>On 12 Jul 2004 09:07:23 -0700, chotiwallah@web .de (chotiwallah) wrote:
>>
>>[color=darkred]
>>>i have a little database driven content managment system. people can
>>>load up html-docs. some of them use ms word as their html-editor,
>>>which resultes in loads of "class" and "style" attributes - like this:
>>>[/color]
>>
>>This is written in JavaScript... It works great!!!!
>>
>>
>>
>> if ( (D.indexOf('cla ss=Mso') >= 0) || (D.indexOf('cla ss="Mso') >=
>>0) ) {
>>
>> // make one line
>> D = D.replace(/\r\n/g, ' ').
>> replace(/\n/g, ' ').
>> replace(/\r/g, ' ').
>> replace(/\&nbsp\;/g,' ');
>>
>> // keep tags, strip attributes
>> D = D.replace(/ class=[^\s|>]*/gi,'').
>>
>> replace(/ style=\"[^>]*\"/gi,'').
>> replace(/ align=[^\s|>]*/gi,'');
>>
>> //clean up tags
>> D = D.replace(/]*>/gi,'').
>> replace(/ ]*>/gi,' ').
>> replace(/]*>/gi,'').
>> replace(/<li [^>]*>/gi,'<li>').
>> replace(/<ul [^>]*>/gi,'<ul>');
>>
>> // replace outdated tags
>> D = D.replace(//gi,'').
>> replace(/<\/b>/gi,'');
>>
>> // mozilla doesn't like tags
>> D = D.replace(//gi,'').
>> replace(/<\/em>/gi,'');
>>
>> // kill unwanted tags
>> D = D.replace(/<\?xml:[^>]*>/g, ''). // Word xml
>> replace(/<\/?st1:[^>]*>/g,''). // Word SmartTags
>> replace(/<\/?[a-z]\:[^>]*>/g,''). // All other funny Word
>> //non-HTML stuff
>> replace(/<\/?font[^>]*>/gi,''). // Disable if you want to
>> //keep font formatting
>> replace(/<\/?span[^>]*>/gi,' ').
>> replace(/<\/?div[^>]*>/gi,' ').
>> replace(/<\/?pre[^>]*>/gi,' ').
>> replace(/<\/?h[1-6][^>]*>/gi,' ');
>>
>> //remove empty tags
>> //D = D.replace(/<\/strong>/gi,'').
>> //replace(/<\/i>/gi,'').
>> //replace(/<P[^>]*><\/P>/gi,'');
>>
>> // nuke double tags
>> oldlen = D.length + 1;
>> while(oldlen > D.length) {
>> oldlen = D.length;
>> // join us now and free the tags, we'll be free hackers,
>> //we'll be free... ;-)
>> D = D.replace(/<([a-z][a-z]*)> *<\/\1>/gi,' ').
>> replace(/<([a-z][a-z]*)> *<([a-z][^>]*)> *<\/\1>/gi,'<$2>');
>> }
>> D = D.replace(/<([a-z][a-z]*)><\1>/gi,'<$1>').
>> replace(/<\/([a-z][a-z]*)><\/\1>/gi,'<\/$1>');
>>
>> // nuke double spaces
>> D = D.replace(/ */gi,' ');[/color]
>
>
>
> brilliant little script, works great, exactly what i needed :-).
> i hadn't even thought of doing the whole cleaning clientside.
>
> thanks a bundle to everyone, micha[/color]

While client-side processing/validation is a good idea, you should no longer
rely on it as not all users turn on javascript or activeX. There are too many
idiots out there writing malicious code that is making a good thing bad.

--
Michael Austin.
Consultant - Available.
Donations welcomed. Http://www.firstdbasource.com/donations.html
:)

reg exp question: removing class and style from html

reg exp question: removing class and style from html

Comment

Comment

Comment

Comment

Comment

Comment

Comment