Duffbert's Random Musings

1 - one more option, you might try homesite's "codesweeper" function if you have that around (it comes with dreamweaver). you can tell it how you want the html to look, as in, if you want indents in front of this tag, but not in front of this one, kinda thing. or, tags that are contained inside one another can be indented or not. pretty cool little tool.

2 - The HTML Tidy tool from the W3c.org site does a good job and cleans up the Word tags very well.

3 - Macromedia Dreamweaver also has the ability to quickly clean-up Word html.

4 - So far, the winner seems to be Tidy. The HTMLArea process is very similar to what I use now. When I paste in the Word code, it keeps all the garbage MS-Word HTML there. The Dreamweaver stuff appears to be interesting, but I don't have that package.

I downloaded the Tidy UI app from the SourceForge site, and it rocks! Free is good, of course. I took the HTML from the worst page I have and dropped it into the editor (didn't even need to read the instructions). Tidy did the cleanup and formatting automatically. Even if it didn't do ANY cleanup, the formatting feature allowed me to find what I was looking for much quicker than my prior hunting method...

Cool stuff, everyone, and thanks for the help!

5 - I was wondering if anyone knows how to make Dreamweaver auto-clean it's template comments when doing a "Put" on the server?

The template feature is a huge time and error saver, but having download all resulting HTMLs back from the server, to manually open each one and run the cleanup it's a daunting task.

If anyone has an idea or knows of a Dreamweaver extension that can do on-the-fly comment cleanup when doing a "Put", I'd appreciate a line or two.

6 - (I think I'm starting to sound like a one-trick-pony here) but have you tried copy/pasting the actual page into the htmlarea editor?
http://www.interactivetools.com/products/htmlarea/index.html#demo
I've downloaded the free example, and have been able to copy-paste the webpage itself, hit the <> button and view the raw W3C-compliant HTML. It's worth a shot anyway...

(They should start paying me advert kickbacks )

-Chris

7 - Via PHP, rough but it works:
<?php
function decraper($html, $delstyles=false) {
$whitespace = array("\t","\n","\r");
$spaces = array(' ',' ',' ',' ');
$html = str_replace($whitespace, '', $html);
for ($t = 1; $t <= 5; $t++) {
$html = (str_replace($spaces, ' ', $html));
}
$commoncrap = array('&quot;'
,'font-weight: normal;'
,'font-style: normal;'
,'line-height: normal;'
,'font-size-adjust: none;'
,'font-stretch: normal;'); //If it is so normal, why they bother?
$replace = array("'");
$html = str_replace($commoncrap, $replace, $html);
$patterns = array();
$replacements = array();
$patterns[0] = '/(<table\s.*)(width=)(\d+%)(\D)/i'; # Fix unquoted non-alphanumeric characters in table tags
$patterns[1] = '/(<td\s.*)(width=)(\d+%)(\D)/i';
$patterns[2] = '/(<th\s.*)(width=)(\d+%)(\D)/i';
$patterns[3] = '/<td( colspan="[0-9]+")?( rowspan="[0-9]+")?( width="[0-9]+")?( height="[0-9]+")?.*?>/i';
$patterns[4] = '/<tr.*?>/i';
$patterns[5] = '/<\/st1:address>(<\/st1:\w*>)?<\/p>[\n\r\s]*<p[\s\w="\']*>/i';
$patterns[6] = '/<o:p.*?>/i';
$patterns[7] = '/<\/o:p>/i';
$patterns[8] = '/<o:SmartTagType[^>]*>/i';
$patterns[9] = '/<st1:[\w\s"=]*>/i';
$patterns[10] = '/<\/st1:\w*>/i';
$patterns[11] = '/<p class="MsoNormal"[^>]*>(.*?)<\/p>/i';
$patterns[12] = '/ style="margin-top: 0cm;"/i';
$patterns[13] = '/<(\w[^>]*) class=([^ |>]*)([^>]*)/i';
$patterns[14] = '/<ul(.*?)>/i';
$patterns[15] = '/<ol(.*?)>/i';
$patterns[17] = '/<br \/>&nbsp;<br \/>/i';
$patterns[18] = '/&nbsp;<br \/>/i';
$patterns[19] = '/<!-.*?>/';
$patterns[20] = '/\s*style=(""|\'\')/';
$patterns[21] = '/ style=[\'"]tab-interval:[^\'"]*[\'"]/i';
$patterns[22] = '/behavior:[^;\'"]*;*(\n|\r)*/i';
$patterns[23] = '/mso-[^:]*:"[^"]*";/i';
$patterns[24] = '/mso-[^;\'"]*;*(\n|\r)*/i';
$patterns[25] = '/\s*font-family:[^;"]*;?/i';
$patterns[26] = '/margin[^"\';]*;?/i';
$patterns[27] = '/text-indent[^"\';]*;?/i';
$patterns[28] = '/tab-stops:[^\'";]*;?/i';
$patterns[29] = '/border-color: *([^;\'"]*)/i';
$patterns[30] = '/border-collapse: *([^;\'"]*)/i';
$patterns[31] = '/page-break-before: *([^;\'"]*)/i';
$patterns[32] = '/font-variant: *([^;\'"]*)/i';
$patterns[33] = '/<span [^>]*><br \/><\/span><br \/>/i';
$patterns[34] = '/" "/';
$patterns[35] = '/[\t\r\n]/';
$patterns[36] = '/\s\s/s';
$patterns[37] = '/ style=""/';
$patterns[38] = '/<span>(.*?)<\/span>/i';
$patterns[39] = '/<span>(.*?)<\/span>/i';//twice, nested spans
$patterns[40] = '/(;\s|\s;)/';
$patterns[41] = '/;;/';
$patterns[42] = '/";/';
$patterns[43] = '/<li(.*?)>/i';
$replacements[0] = '$1$2"$3"$4';
$replacements[1] = '$1$2"$3"$4';
$replacements[2] = '$1$2"$3"$4';
$replacements[3] = '<td$1$2$3$4>';
$replacements[4] = '<tr>';
$replacements[5] = '<br />';
$replacements[6] = '';
$replacements[7] = '<br />';
$replacements[8] = '';
$replacements[9] = '';
$replacements[10] = '';
$replacements[11] = '$1<br />';
$replacements[12] = '';
$replacements[13] = '<$1$3';
$replacements[14] = '<ul>';
$replacements[15] = '<ol>';
$replacements[17] = '<br />';
$replacements[18] = '<br />';
$replacements[19] = '';
$replacements[20] = '';
$replacements[21] = '';
$replacements[22] = '';
$replacements[23] = '';
$replacements[24] = '';
$replacements[25] = '';
$replacements[26] = '';
$replacements[27] = '';
$replacements[28] = '';
$replacements[29] = '';
$replacements[30] = '';
$replacements[31] = '';
$replacements[32] = '';
$replacements[33] = '<br />';
$replacements[34] = '""';
$replacements[35] = '';
$replacements[36] = '';
$replacements[37] = '';
$replacements[38] = '$1';
$replacements[39] = '$1';
$replacements[40] = ';';
$replacements[41] = ';';
$replacements[42] = '"';
$replacements[43] = '<li>';
if($delstyles===true){
$patterns[44] = '/ style=".*?"/';
$replacements[44] = '';
}
ksort($patterns);
ksort($replacements);
$html = preg_replace($patterns, $replacements, $html);
for ($t=1;$t<=3;$t++) {
$html = (str_replace($spaces, ' ', $html));
}
return $html;
}
or via javascript:
function demoroniser(html) {
html = html.replace(/\r\n/g, ' ').
replace(/\n/g, ' ').
replace(/\r/g, ' ').
replace(/\&nbsp\;/g,' ');
html = html.replace(/ class=[^\s|>]*/gi,'').
//replace(/<p [^>]*TEXT-ALIGN: justify[^>]*>/gi,'<p align="justify">').
replace(/ style=\"[^>]*\"/gi,'').
replace(/ align=[^\s|>]*/gi,'');
html = html.replace(/<b [^>]*>/gi,'<b>').
replace(/<i [^>]*>/gi,'<i>').
replace(/<li [^>]*>/gi,'<li>').
replace(/<ul [^>]*>/gi,'<ul>');
html = html.replace(/<b>/gi,'<strong>').
replace(/<\/b>/gi,'</strong>');
html = html.replace(/<em>/gi,'<i>').
replace(/<\/em>/gi,'</i>');
html = html.replace(/<\?xml:[^>]*>/g, '').
replace(/<\/?st1:[^>]*>/g,'').
replace(/<\/?[a-z]\:[^>]*>/g,'').
replace(/<\/?font[^>]*>/gi,'').
replace(/<\/?span[^>]*>/gi,' ').
replace(/<\/?div[^>]*>/gi,' ').
replace(/<\/?pre[^>]*>/gi,' ').
replace(/<\/?h[1-6][^>]*>/gi,' ');
oldlen = html.length + 1;
while(oldlen > html.length) {
oldlen = html.length;
html = html.replace(/<([a-z][a-z]*)> *<\/\1>/gi,' ').
replace(/<([a-z][a-z]*)> *<([a-z][^>]*)> *<\/\1>/gi,'<$2>');
}
html = html.replace(/<([a-z][a-z]*)><\1>/gi,'<$1>').
replace(/<\/([a-z][a-z]*)><\/\1>/gi,'<\/$1>');
html = html.replace(/ */gi,' ');
return html;
}

8 - Great PHP function for cleaning up Microsoft Work HTML from copy and paste.

VX

9 - I know I'm late to the game here, but Dreamweaver is the best I've seen at cleaning up Microsoft code. They actually have that as a feature built in. Very cool. HTML Tidy is nice too, and much cheaper than Dreamweaver... However, if you want the BEST HTML editor ever created... Dreamweaver is IT.