phpwiki-talk Mailing List for PhpWiki (Page 7)

Hello Reini,
I fixed a problem in CreateTocPlugin. When displaying an old
revision of the page, the TOC was always taken from the
latest revision.
What is still incorrect in that when you preview the
modifications and have added, removed or otherwise changed
the headers, the TOC is calculated from the latest saved
revision.
This confuses our users.
Can you please help me to fix this? I have no idea.
Best regards,
Marc-Etienne
-- 
Marc-Etienne Vargenau
Alcatel-Lucent France, Route de Villejust, 91620 NOZAY, FRANCE
+33 (0)1 30 77 28 33, Mar...@al...

>
> I think using "name" should work in all cases, except in very
> old browsers (NS4). It works in IE6.
Oops...
I meant 'using "id" should work in all cases'...
Sorry about it.
Marc-Etienne
-- 
Marc-Etienne Vargenau
Alcatel-Lucent France, Route de Villejust, 91620 NOZAY, FRANCE
+33 (0)1 30 77 28 33, Mar...@al...

Le 21/09/2010 16:18, Reini Urban a écrit :
> 2010年9月21日 Marc-Etienne Vargenau<Mar...@al...>:
>> Le 21/09/2010 09:08, ru...@us... a écrit :
>>> Revision: 7704
>>> http://phpwiki.svn.sourceforge.net/phpwiki/?rev=7704&view=rev
>>> Author: rurban
>>> Date: 2010年09月21日 07:08:56 +0000 (2010年9月21日)
>>>
>>> Log Message:
>>> -----------
>>> revert 7551-7553. we use named anchors to jump to, not internal ids
>>>
>>> Modified Paths:
>>> --------------
>>> trunk/lib/plugin/CreateToc.php
>>>
>>> Modified: trunk/lib/plugin/CreateToc.php
>>
>> Hello Reini,
>>
>> I do not really agree with this one.
>> Both ids and names can be used for anchors, and name is
>> deprecated for<a>. It no longer exists in XHTML 1.1 and HTML 5.
>
> hmm, so if we are in quirks mode we should use name to be able to jump
> to #anchor,
> and if strict xhtml (ENABLE_XHTML_XML) we should use id, right?
I think using "name" should work in all cases, except in very
old browsers (NS4). It works in IE6.
Regards,
Marc-Etienne
-- 
Marc-Etienne Vargenau
Alcatel-Lucent France, Route de Villejust, 91620 NOZAY, FRANCE
+33 (0)1 30 77 28 33, Mar...@al...

2010年9月21日 Marc-Etienne Vargenau <Mar...@al...>:
> Le 21/09/2010 09:08, ru...@us... a écrit :
>> Revision: 7704
>>      http://phpwiki.svn.sourceforge.net/phpwiki/?rev=7704&view=rev
>> Author:  rurban
>> Date:   2010年09月21日 07:08:56 +0000 (2010年9月21日)
>>
>> Log Message:
>> -----------
>> revert 7551-7553. we use named anchors to jump to, not internal ids
>>
>> Modified Paths:
>> --------------
>>   trunk/lib/plugin/CreateToc.php
>>
>> Modified: trunk/lib/plugin/CreateToc.php
>
> Hello Reini,
>
> I do not really agree with this one.
> Both ids and names can be used for anchors, and name is
> deprecated for <a>. It no longer exists in XHTML 1.1 and HTML 5.
hmm, so if we are in quirks mode we should use name to be able to jump
to #anchor,
and if strict xhtml (ENABLE_XHTML_XML) we should use id, right?
-- 
Reini Urban

Le 21/09/2010 09:08, ru...@us... a écrit :
> Revision: 7704
> http://phpwiki.svn.sourceforge.net/phpwiki/?rev=7704&view=rev
> Author: rurban
> Date: 2010年09月21日 07:08:56 +0000 (2010年9月21日)
>
> Log Message:
> -----------
> revert 7551-7553. we use named anchors to jump to, not internal ids
>
> Modified Paths:
> --------------
> trunk/lib/plugin/CreateToc.php
>
> Modified: trunk/lib/plugin/CreateToc.php
Hello Reini,
I do not really agree with this one.
Both ids and names can be used for anchors, and name is
deprecated for <a>. It no longer exists in XHTML 1.1 and HTML 5.
Best regards,
Marc-Etienne
-- 
Marc-Etienne Vargenau
Alcatel-Lucent France, Route de Villejust, 91620 NOZAY, FRANCE
+33 (0)1 30 77 28 33, Mar...@al...

Le 21/09/2010 08:08, ru...@us... a écrit :
> Revision: 7702
> http://phpwiki.svn.sourceforge.net/phpwiki/?rev=7702&view=rev
> Author: rurban
> Date: 2010年09月21日 06:08:32 +0000 (2010年9月21日)
>
> Log Message:
> -----------
> new may return an empty object. we must check for _userid to avoid recursion cycles
>
> Modified Paths:
> --------------
> trunk/lib/WikiUserNew.php
Hello Reini,
Can you please carefully check file WikiUserNew.php?
I reverted some of the changes because I could no longer
save user preferences.
I might have reverted too many things.
Best regards,
Marc-Etienne
-- 
Marc-Etienne Vargenau
Alcatel-Lucent France, Route de Villejust, 91620 NOZAY, FRANCE
+33 (0)1 30 77 28 33, Mar...@al...

2010年9月17日 Marc-Etienne Vargenau <Mar...@al...>:
> Le 17/09/2010 13:49, Reini Urban a écrit :
>> Marc-Etienne,
>> Do you want to formally announce the rc1 release?
>> You did most of the work.
>>
>> As you see I found some minor bugs when installing completely from
>> scratch on windows.
>
> Hello Reini,
>
> Great news.
> It certainly needs some more testing, we will make a
> formal announce for 1.4.0 final.
Sure.
> I do not test on Windows, only Linux and Mac OS X.
I only did some minor testing on Windows yet.
> You did not create a tag and a branch, do we need to do it?
branch not.
tag probably not, only real releases.
> What procedure do you use to create the ZIP file?
on windows after removing config/config.ini:
 zip -r <reldir>.zip <reldir>
> I would like to create an RPM file for Linux, is there a spec
> for that somewhere?
Not that I know of.
But there's a debian deb with a different layout. ini in etc, ...)
admin should be in /usr/lib/phpwiki/admin probably, ...
-- 
Reini Urban
http://phpwiki.org/      http://murbreak.at/

Hi,
Excellent, that’s brilliant news!
Thanks a mil,
Sébastien.
On 17/09/2010 13:51, Marc-Etienne Vargenau wrote:
> Le 17/09/2010 13:49, Reini Urban a écrit :
>> Marc-Etienne,
>> Do you want to formally announce the rc1 release?
>> You did most of the work.
>>
>> As you see I found some minor bugs when installing completely from
>> scratch on windows.
>
> Hello Reini,
>
> Great news.
> It certainly needs some more testing, we will make a
> formal announce for 1.4.0 final.
>
> I do not test on Windows, only Linux and Mac OS X.
>
> You did not create a tag and a branch, do we need to do it?
>
> What procedure do you use to create the ZIP file?
> I would like to create an RPM file for Linux, is there a spec
> for that somewhere?
>
> Best regards,
>
> Marc-Etienne
>

Le 17/09/2010 13:49, Reini Urban a écrit :
> Marc-Etienne,
> Do you want to formally announce the rc1 release?
> You did most of the work.
>
> As you see I found some minor bugs when installing completely from
> scratch on windows.
Hello Reini,
Great news.
It certainly needs some more testing, we will make a
formal announce for 1.4.0 final.
I do not test on Windows, only Linux and Mac OS X.
You did not create a tag and a branch, do we need to do it?
What procedure do you use to create the ZIP file?
I would like to create an RPM file for Linux, is there a spec
for that somewhere?
Best regards,
Marc-Etienne
-- 
Marc-Etienne Vargenau
Alcatel-Lucent France, Route de Villejust, 91620 NOZAY, FRANCE
+33 (0)1 30 77 28 33, Mar...@al...

Marc-Etienne,
Do you want to formally announce the rc1 release?
You did most of the work.
As you see I found some minor bugs when installing completely from
scratch on windows.
-- 
Reini Urban
http://phpwiki.org/      http://murbreak.at/

Le 13/09/2010 17:44, Sébastien Le Callonnec a écrit :
> Hi Marc-Étienne,
>
> Ok, thanks, I’ll have a look! Is that happening in specific conditions?
Hello Sébastien,
As indicated in the error message, I am using Pear to
access the database and My SQL.
PHP 5.2.6 under Linux.
Best regards,
Marc-Etienne
-- 
Marc-Etienne Vargenau
Alcatel-Lucent France, Route de Villejust, 91620 NOZAY, FRANCE
+33 (0)1 30 77 28 33, Mar...@al...

Hi Marc-Étienne,
Ok, thanks, I’ll have a look! Is that happening in specific conditions?
Regards,
Sébastien.
On 13/09/2010 16:21, Marc-Etienne Vargenau wrote:
> Le 11/09/2010 00:02, Sébastien Le Callonnec a écrit :
>> Hi Marc-Étienne,
>>
>>
>> I’m curious as to what error this patch caused so that I can have a
>> look, rather than blindly reverting it -- the "AllPages" plugin is now
>> again broken in my environments, which doesn’t seem to me like a
>> convergent approach.
>
> Hello Sébastien,
>
> First let me say that you are very welcome to contribute to Phpwiki.
>
> For me, your patch produces the following error, that is why I reverted.
>
> Fatal Error:
>
> lib/WikiDB/backend/PearDB.php (In template 'body' < 'html'):1061 Error:
> WikiDB_backend_PearDB_mysql: fatal database error
>
> * DB Error: syntax error
> * (SELECT page.id AS id, page.pagename AS pagename, page.hits AS hits
> FROM page WHERE AND page.pagename NOT IN ('') ORDER BY pagename ASC
> LIMIT 0, 1000 [nativecode=1064 ** You have an error in your SQL syntax;
> check the manual that corresponds to your MySQL server version for the
> right syntax to use near 'AND page.pagename NOT IN ('') ORDER BY
> pagename ASC LIMIT 0, 1000' at line 1])
>
> Reini, can you please check what is going on?
>
> Best regards,
>
> Marc-Etienne
>

Le 11/09/2010 00:02, Sébastien Le Callonnec a écrit :
> Hi Marc-Étienne,
>
>
> I’m curious as to what error this patch caused so that I can have a
> look, rather than blindly reverting it -- the "AllPages" plugin is now
> again broken in my environments, which doesn’t seem to me like a
> convergent approach.
Hello Sébastien,
First let me say that you are very welcome to contribute to Phpwiki.
For me, your patch produces the following error, that is why I reverted.
Fatal Error:
lib/WikiDB/backend/PearDB.php (In template 'body' < 'html'):1061 Error: 
WikiDB_backend_PearDB_mysql: fatal database error
 * DB Error: syntax error
 * (SELECT page.id AS id, page.pagename AS pagename, page.hits AS 
hits FROM page WHERE AND page.pagename NOT IN ('') ORDER BY pagename ASC 
LIMIT 0, 1000 [nativecode=1064 ** You have an error in your SQL syntax; 
check the manual that corresponds to your MySQL server version for the 
right syntax to use near 'AND page.pagename NOT IN ('') ORDER BY 
pagename ASC LIMIT 0, 1000' at line 1])
Reini, can you please check what is going on?
Best regards,
Marc-Etienne
-- 
Marc-Etienne Vargenau
Alcatel-Lucent France, Route de Villejust, 91620 NOZAY, FRANCE
+33 (0)1 30 77 28 33, Mar...@al...

Hi Marc-Étienne,
I’m curious as to what error this patch caused so that I can have a 
look, rather than blindly reverting it -- the "AllPages" plugin is now 
again broken in my environments, which doesn’t seem to me like a 
convergent approach.
Thanks,
Sébastien.
On 10/09/2010 12:31, var...@us... wrote:
> Revision: 7681
> http://phpwiki.svn.sourceforge.net/phpwiki/?rev=7681&view=rev
> Author: vargenau
> Date: 2010年09月10日 11:31:37 +0000 (2010年9月10日)
>
> Log Message:
> -----------
> revert of Patch 3024787 by S?195円?169円bastien Le Callonnec (caused an SQL syntax error)
>
> Modified Paths:
> --------------
> trunk/TODO
> trunk/lib/plugin/AllPages.php
>
> Modified: trunk/TODO
> ===================================================================
> --- trunk/TODO	2010年09月10日 11:27:57 UTC (rev 7680)
> +++ trunk/TODO	2010年09月10日 11:31:37 UTC (rev 7681)
> @@ -4,6 +4,7 @@
> * AllPages give too many pages
> * WantedPages give too many pages (from interwiki map)
> (to be tested in both Pear and ADODB)
> +* revert of Patch 3024787 by Sébastien Le Callonnec (caused an SQL syntax error)
>
> == Others==
> * Minimizer for JS/CSS
>
> Modified: trunk/lib/plugin/AllPages.php
> ===================================================================
> --- trunk/lib/plugin/AllPages.php	2010年09月10日 11:27:57 UTC (rev 7680)
> +++ trunk/lib/plugin/AllPages.php	2010年09月10日 11:31:37 UTC (rev 7681)
> @@ -67,18 +67,17 @@
> $timer = new DebugTimer;
> }
> $caption = _("All pages in this wiki (%d total):");
> - $exclude_empty_pagename = array('');
>
> if ( !empty($args['userpages']) ) {
> $pages = PageList::allUserPages($args['include_empty'],
> - $args['sortby'], '',
> - $exclude_empty_pagename);
> + $args['sortby'], ''
> + );
> $caption = _("List of user-created pages (%d total):");
> $args['count'] = $request->getArg('count');
> } elseif ( !empty($args['owner']) ) {
> $pages = PageList::allPagesByOwner($args['owner'], $args['include_empty'],
> - $args['sortby'], '',
> - $exclude_empty_pagename);
> + $args['sortby'], ''
> + );
> $args['count'] = $request->getArg('count');
> if (!$args['count'])
> $args['count'] = $dbi->numPages($args['include_empty'], $args['exclude']);
> @@ -90,8 +89,8 @@
> $pages->_options['count'] = $args['count'];
> } elseif ( !empty($args['author']) ) {
> $pages = PageList::allPagesByAuthor($args['author'], $args['include_empty'],
> - $args['sortby'], '',
> - $exclude_empty_pagename);
> + $args['sortby'], ''
> + );
> $args['count'] = $request->getArg('count');
> if (!$args['count'])
> $args['count'] = $dbi->numPages($args['include_empty'], $args['exclude']);
> @@ -103,8 +102,7 @@
> $pages->_options['count'] = $args['count'];
> } elseif ( !empty($args['creator']) ) {
> $pages = PageList::allPagesByCreator($args['creator'], $args['include_empty'],
> - $args['sortby'], '',
> - $exclude_empty_pagename
> + $args['sortby'], ''
> );
> $args['count'] = $request->getArg('count');
> if (!$args['count'])
> @@ -115,6 +113,8 @@
> : $args['creator'],
> 'if_known'), $args['count']);
> $pages->_options['count'] = $args['count'];
> + //} elseif ($pages) {
> + // $args['count'] = count($pages);
> } else {
> if (! $request->getArg('count'))
> $args['count'] = $dbi->numPages($args['include_empty'], $args['exclude']);
> @@ -126,17 +126,15 @@
> $pagelist = new PageList($args['info'], $args['exclude'], $args);
> if (!$args['noheader']) $pagelist->setCaption($caption);
>
> - if ($pages !== false) {
> + // deleted pages show up as version 0.
> + //if ($args['include_empty'])
> + // $pagelist->_addColumn('version');
> +
> + if ($pages !== false)
> $pagelist->addPageList($pages);
> - }
> - else {
> - // Clear count as previous value was for non-empty pages.
> - $pagelist->clearArg('count');
> - $pagelist->addPages($dbi->getAllPages($args['include_empty'],
> - $args['sortby'],
> - $args['limit'],
> - $exclude_empty_pagename));
> - }
> + else
> + $pagelist->addPages( $dbi->getAllPages($args['include_empty'], $args['sortby'],
> + $args['limit']) );
> if (DEBUG&& $args['debug']) {
> return HTML($pagelist,
> HTML::p(fmt("Elapsed time: %s s", $timer->getStats())));
>
>
> This was sent by the SourceForge.net collaborative development platform, the world's largest Open Source development site.
>
> ------------------------------------------------------------------------------
> Automate Storage Tiering Simply
> Optimize IT performance and efficiency through flexible, powerful,
> automated storage tiering capabilities. View this brief to learn how
> you can reduce costs and improve performance.
> http://p.sf.net/sfu/dell-sfdev2dev
> _______________________________________________
> phpwiki-checkins mailing list
> php...@li...
> https://lists.sourceforge.net/lists/listinfo/phpwiki-checkins

2010年8月7日 Matt Brown <ma...@ma...>:
> I was updating the PHPwiki debian package in preparation for the
> upcoming Squeeze release, when I discovered that unfortunately it is
> no longer working at all, as Squeeze is now using PHP 5.3.
>
> I see that there are a number of commits in the subversion repository
> that address many of the issues (deprecated functions, etc).
>
> I've filed https://sourceforge.net/tracker/?func=detail&atid=106121&aid=3041169&group_id=6121
> to ask for a new release. However given that there haven't been any
> releases since the 1.3.14 in 2007 I wanted to send this mail to check
> if releases of PHPwiki are even being made at all...
They are planned *real soon* but as this will be a new major one 1.14
and new blockers popped up and I'm busy with perl and Marc-Etienne is
busy in holidays it will some time.
-- 
Reini Urban
http://phpwiki.org/      http://murbreak.at/

Hi,
I was updating the PHPwiki debian package in preparation for the
upcoming Squeeze release, when I discovered that unfortunately it is
no longer working at all, as Squeeze is now using PHP 5.3.
I see that there are a number of commits in the subversion repository
that address many of the issues (deprecated functions, etc).
I've filed https://sourceforge.net/tracker/?func=detail&atid=106121&aid=3041169&group_id=6121
to ask for a new release. However given that there haven't been any
releases since the 1.3.14 in 2007 I wanted to send this mail to check
if releases of PHPwiki are even being made at all...
Cheers
-- 
Matt Brown
ma...@ma...
Mob +353 86 608 7117 www.mattb.net.nz

No worries, Reini.
I was just getting worried that they might have been overlooked.
Thanks,
Sébastien.
On 15/07/2010 21:24, Reini Urban wrote:
> Sébastien Le Callonnec schrieb:
>> I have a couple of pending patches in the bug tracker; any chance of
>> having them reviewed so that I can apply them? I’m thinking of issue
>> #3024787 in particular, which prevents "AllPages" from working at all.
>
> Sorry, I'm very very busy right now. Marc-Etienne is also waiting for me.
> Marc-Etienne, do you want to look at that?

Sébastien Le Callonnec schrieb:
> I have a couple of pending patches in the bug tracker; any chance of
> having them reviewed so that I can apply them? I’m thinking of issue
> #3024787 in particular, which prevents "AllPages" from working at all.
Sorry, I'm very very busy right now. Marc-Etienne is also waiting for me.
Marc-Etienne, do you want to look at that?
-- 
Reini

Hi,
I have a couple of pending patches in the bug tracker; any chance of 
having them reviewed so that I can apply them? I’m thinking of issue 
#3024787 in particular, which prevents "AllPages" from working at all.
Thanks,
Sébastien.

Hi,
Issue #3017253 is back, this time forcing Tahoma on the two links in the 
navbar whereas the others remain with the default "sans-serif" defined 
in body.
(See 
https://sourceforge.net/tracker/?func=detail&aid=3017253&group_id=6121&atid=106121 
for background). This is caused by the fact that these two links are 
actually created by "Button" (which gives them the class wikiadmin), 
whereas the other are plain "WikiLink" (which gives them the class 
named-wiki). Another side effect of the wikiadmin class is that these 2 
links have no "text-decoration," whereas the others do.
Either we set all the links in the navbar to Tahoma (not really my 
favourite option), or leave everything as the default sans-serif (more 
than likely for just these 2 by creating specific CSS styles), but we 
have to decide...
Regards,
Seb

2010年7月3日 Harold Hallikainen <ha...@ha...>:
> Has anyone done an interface to botscout.com to avoid wiki spam? I'm
> seeing a lot of it recently.
No. I haven't seen it yet. But looks very very useful and simple.
I'll definitively add it.
-- 
Reini Urban
http://phpwiki.org/ http://murbreak.at/

Has anyone done an interface to botscout.com to avoid wiki spam? I'm
seeing a lot of it recently.
Thanks!
Harold
-- 
FCC Rules Updated Daily at http://www.hallikainen.com - Advertising
opportunities available!

Hi Wes,
Would you be able to set ENABLE_USER_NEW to "true" (I’m suspecting it is 
set to false?) and see if this solves your problem? I’m also using the 
latest code from Subversion with a similar configuration, and I don’t 
experience that issue (php 5.3.2).
Regards,
Seb
On 25/06/2010 19:10, Wes wrote:
> Hello,
>
> I have a ton of older phpwiki installs running on Postgres. The machine
> they're running on is on it's least breath, so I'm moving them to a new
> machine and converting to MySQL. That's all worked...reasonably well.
> Using a SVN pull from this morning. My config.ini is set to use
> HttpAuth, and the .htaccess file is configured correctly. AuthPolicy is
> set to "first-only". I can log in via the browser/Apache prompt fine,
> but whenever I go to any page, I get this error:
>
> *Fatal error*: Call to a member function get() on a non-object in
> */var/services/old-wikis/SysAdminWiki/lib/WikiUser.php* on line *153*
>
> And "Edit" on any page is asking for a login and pass (meaning it's
> apparently not passing through correctly.)
>
> Thoughts or suggestions would be well received : )
>
> Thanks!
>
> Wes
>
>
>
>
> ------------------------------------------------------------------------------
> ThinkGeek and WIRED's GeekDad team up for the Ultimate
> GeekDad Father's Day Giveaway. ONE MASSIVE PRIZE to the
> lucky parental unit. See the prize list and enter to win:
> http://p.sf.net/sfu/thinkgeek-promo
>
>
>
> _______________________________________________
> Phpwiki-talk mailing list
> Php...@li...
> https://lists.sourceforge.net/lists/listinfo/phpwiki-talk

<?php // -*-php-*-
rcs_id('$Id: RichTable.php,v 2.2 2010年06月24日 17:44:24 Stefan Schorn Exp $');
/**
Richtable - Tabellen plugin
*/
class WikiPlugin_RichTable
extends WikiPlugin
{
	function getName() {
		return _("RichTable");
	}
	function getDescription() {
		return _("Layout tables using a very rich markup style.");
	}
	function getDefaultArguments() {
		return array();
	}
	function getVersion() {
		return preg_replace("/[Revision: $]/", '',
		"\$Revision: 1.7 $");
	}
	function run($dbi, $argstr, &$request, $basepage) {
		global $WikiTheme;
		include_once("lib/BlockParser.php");
		// RichTablePlugin markup is new.
		$markup = 2.0;
		$lines = preg_split('/\n/', $argstr);
		$table = HTML::table();
		$atlas = 0;
		$atlasdl = 0;
		if ($lines[0][0] == '*') {
			$line = substr(array_shift($lines),1);
			$attrs = $this->_parse_attr($line);
			foreach ($attrs as $key => $value) {
				if (in_array ($key, array("id", "class", "title", "style",
				"bgcolor", "frame", "rules", "border",
				"cellspacing", "cellpadding",
				"summary", "align", "width"))) {
					if ($key == 'class') $class = $value;
					$table->setAttr($key, $value);
				}
			}
		}
		$r = 0;
		$c = 0;
		$ownclass = 0;
		$ownclasstr = 0;
		$Zelle = array();
		if(is_array($lines)) $lastLine = $lines[count($lines)-1];
		if (!(trim($lastLine) == '-')) { $lines[] = '-';}
		foreach ($lines as $line){					
			if (substr($line,0,1) == "-") {
				$makeit = 0;
				if (isset($row)) {
					if (isset($cell)) {
						if ($r == 1) $cmax = $c;
						if (isset($class) && $class && !$ownclass) {
							if ($c == $cmax)	{
								if ($r == 1) {
									$cell->setAttr('class', $class.'-bn');
								} else {
									$cell->setAttr('class', $class.'-bt');
								}
							} else {
								if ($r == 1) {
									$cell->setAttr('class', $class.'-br');
								} else {
									$cell->setAttr('class', $class.'-brt');
								}
							}
						} 
						
						if (isset($content)) {
							// debug $cell->pushContent(TransformText('<<'.$r.'#'.$c.'>>', $markup, $basepage));
							$cell->pushContent(TransformText($content, $markup, $basepage));
							unset($content);
						}
						$row->pushContent($cell);
						unset($cell);
							$makeit = 1;
					}
					if ($makeit == 1) {
						$table->pushContent($row);
						unset($row);
					}
				}
				$r++;
				$row = HTML::tr();
				$attrs = $this->_parse_attr(substr($line,1));
				$c = 0;
				$ownclasstr = 0;
				foreach ($attrs as $key => $value) {
					if (in_array ($key, array("id", "class", "title", "style",
					"bgcolor", "align", "valign"))) {
						$row->setAttr($key, $value);
						if ($key == 'class' && isset($value) && $value) {
							$ownclasstr = 1;
						}						
					}
				}
				if (!$ownclasstr && isset($class) && $class) 
					$row->setAttr('class', $class);
				continue;
			}
			if (substr($line,0,1) == "|" and isset($row)) {
				if (isset($cell)) {
					if ($class && !$ownclass) {
						if ($r == 1) {
							$cell->setAttr('class', $class.'-br');
						} else {
							$cell->setAttr('class', $class.'-brt');
						}
					}
					if (isset ($content)) {
						//debug $cell->pushContent(TransformText('<<'.$r.'#'.$c.'>>', $markup, $basepage));
						$cell->pushContent(TransformText($content, $markup, $basepage));
						unset($content);
					}
					$row->pushContent($cell);
				}
				$keyZold = $c;
				$go = 1;
				if (isset($Zelle[$r])) {
					foreach ($Zelle[$r] as $keyZ => $valueZ) {
						$diff = $keyZ - $keyZold;
				
						$keyZold = $keyZ;
						if ($diff == 1 && $go) {
							$c = $c + $valueZ;						
							unset($Zelle[$r][$keyZ]);
						} else $go = 0;
					}
				} 
				$c++;
				$cell = HTML::td();
				$line = substr($line, 1);
				if ($line[0] == "*" ) {
					$attrs = $this->_parse_attr(substr($line,1));
					$colsp = 0;
					$rowsp = 0;
					$ownclass = 0;
					foreach ($attrs as $key => $value) {
						if (in_array ($key, array("id", "class", "title", "style",
						"colspan", "rowspan", "width", "height",
						"bgcolor", "align", "valign"))) {
							if (!($atlas && $key == 'class'))
							$cell->setAttr($key, $value);
							if ($key == 'colspan' && is_numeric($value)) {
								$colsp = $value;
							}
							if ($key == 'rowspan' && is_numeric($value)) {
								$rowsp = $value;
							}
							if ($key == 'class' && isset($value) && $value) {
								$ownclass = 1;
							}
						}
					}
					if (is_numeric($colsp)) {
						for ($x = 1; $x < $colsp; $x++) {
							$c = $c + 1;
						}
						$colsp = 0;
					}
					if (is_numeric($rowsp)) {
						$endrow = $rowsp + $r;
						for ($x = ($r+1); $x < $endrow; $x++) {
							$Zelle[$x][$c] = 1;
						}
						$rowsp = 0;
					}
					continue;
				}
			}
			if (isset($row) and isset($cell)) {
				$line = str_replace("?\>", "?>", $line);
				$line = str_replace("\~", "~", $line);
				if (empty($content)) $content = '';
				$content .= $line . "\n";
			}
		}
		// $makit - no more empty rows
		$makeit = 0;
		if (isset($row)) {
			if (isset($cell)) {
				if (isset($content))
				$cell->pushContent(TransformText($content));
				$makeit = 1;
				$row->pushContent($cell);
			}
			if ($makeit == 1) {
				$table->pushContent($row);
				unset($row);
			}
		}
		return $table;
	}
	function _parse_attr($line) {
		$attr_chunks = preg_split("/\s*,\s*/", strtolower($line));
		$options = array();
		foreach ($attr_chunks as $attr_pair) {
			if (empty($attr_pair)) continue;
			$key_val = preg_split("/\s*=\s*/", $attr_pair);
			if (!empty($key_val[1]))
			$options[trim($key_val[0])] = trim($key_val[1]);
		}
		return $options;
	}
}
// End:
?>

Hallo,
you are interested in my modified RichTable.php. The following 
fixes/anhancements are established to it.
-W3C proofed (the old plugin only shows pictures in the last cell of the 
table when a "-" (<tr>) is at the end of the plugin. This is not 
nessassary anymore and handled inside the plugin. If allready set, an 
empty row without cell is not printed anymore)
- different styles
 - for first row (first cell till before last cell)
 - frst row last cell
 - every next row (first cell till before last cell)
 - every next row (last cell)
This gives you the chance to make layouts for tables without outer 
border, shadows and round corners etc. It was tricky to handle colspan 
and rowspan to find out which cell is the last one in a row (maybe its 
the one from row before at rowspan ...) but now it works
how does it work?
The plugin
<?plugin RichTableNEW
*class=YOURCLASS, width=100%
-valign=top
|*rowspan=2
1.1
-
|2.1
|2.2
?>
The stylesheet
/* YOURCLASS = class you use in your stylesheet. Here you can use the 
following styles to make round borders and shadow in firefox and chrome, 
safari:
*/
table.YOURCLASS{}
tr.YOURCLASS {
 -moz-box-shadow: 1px 1px 2px 1px #999;
 -moz-border-radius: 4px;
 -khtml-border-radius: 4px;
 -khtml-box-shadow: 1px 1px 2px 1px #999;
 -webkit-border-radius: 4px;
 box-shadow: 1px 1px 2px 1px #999;
 border-radius: 4px;
}
/* bn = border none (right top cell) */
td.YOURCLASS-bn {
 border: none;
}
/* bt = border top (every last cell except top cell)*/
 td.YOURCLASS-bt {
 border-top: 1px #999 solid;
}
/* br = border right (every first row cell except last one */
 td.YOURCLASS-br {
 border-right: 1px #ddd solid;
}
/*brt = border right top (every missing cell) */
 td.YOURCLASS-brt {
 border-right: 1px #ddd solid;
 border-top: 1px #ddd solid;
}
You can see it work here:
http://www.mineralienatlas.de/lexikon/index.php/Geologische%20Zeittafel
Have fun
Stefan