[geeklog-devel] [geeklog-cvs] Geeklog-1.x

[Oliver Spiesshofer]

It is astounding to see the way you insult other people who ask to be 
told when they make mistakes.
You seem to be of the opinion that this is a bad thing. Well maybe you 
want to find someone else to discuss technical matters then.
It wont be me for sure.

Oliver

Mark R. Evans wrote:
> Oliver,
>
> I'll see if I can explain this so you can understand it:
>
> You modified COM_siteFooter() to allow the left blocks to be created 
> there instead of COM_siteHeader.
>
> First, let's look at the parameters that are passed to COM_siteFooter():
>
> * @param   boolean     $rightblock     Whether or not to show blocks 
> on right hand side default is no
> * @param   array       $custom         An array defining custom 
> function to be used to format Rightblocks
>
> OK, so we can pass a boolean to turn on / off the right blocks and 
> pass a custom function to format the right blocks.   The first thing 
> you should notice is that there is no information being passed to 
> control the left blocks.  That should be your first red flag that this 
> simply won't work, but let's keep going.
>
> As we move through the code, we get to the part that creates the left 
> blocks, if first checks to see if $custom is defined.  If so, it does:
>
>         if( is_array( $custom ))
>         {
>             $function = $custom[0];
>             if( function_exists( $function ))
>             {
>                 $lblocks = $function( $custom[1], 'left' );
>             }
>         }
>
> But wait, I passed a custom function to handle the right blocks, not 
> the left blocks.  Hmm, this is a problem.
>
> Then, just below that code, you check the following:
>
>     if( $_CONF['right_blocks_in_footer'] == 1 && $rightblock)
>     {
>         $rblocks = '';
>
> OK, makes sense, if there is something in $rightblocks and we should 
> create them in the footer call, we need to process.  But what comes 
> next is more broken code:
>
>         if( is_array( $what ))
>         {
>             $function = $what[0];
>             if( function_exists( $function ))
>             {
>                 $rblocks = $function( $what[1], 'right' );
>             }
>             else
>             {
>                 $rblocks = COM_showBlocks( 'right', $topic );
>             }
>         }
> $what is never defined in COM_siteFooter().  It doesn't exist.  It 
> does exist in COM_siteHeader() and that is where it appears this code 
> was copied.  The desired outcome simply cannot happen because you 
> reference a variable that doesn't exist in the function.
>
> So, let's recap.  What this patch accomplishes is the ability to build 
> the left blocks in the footer, but it removes the ability to have any 
> custom functions to format those left blocks, it removes the ability 
> to dynamically have either left-content-right or left-content or 
> content-right.  For a detailed explanation why this won't work you can 
> see my previous post where all this was detailed out:
>
> http://eight.pairlist.net/pipermail/geeklog-devel/2007-November/002642.html
>
> This post didn't seem to gather much attention as no one bothered to 
> comment.  I detailed pretty well why your approach won't work in that 
> post, maybe I just used too many big words.
>
> Please let me know if you need any additional details or additional 
> explanations.  I'll be happy to walk through the code line by line if 
> necessary.
>
> Thanks!
> Mark
>
> On 11/25/07, *Oliver Spiesshofer* <oliver at spiesshofer.com 
> <mailto:oliver at spiesshofer.com>> wrote:
>
>     Mark R. Evans wrote:
>     > Joe,
>     > It appears the feature freeze was only limited to some, not all.
>     >
>     http://eight.pairlist.net/pipermail/geeklog-devel/2007-September/002325.html
>     <http://eight.pairlist.net/pipermail/geeklog-devel/2007-September/002325.html>
>     >
>     <http://eight.pairlist.net/pipermail/geeklog-devel/2007-September/002325.html>
>     > - Oliver announces he can make a pure CSS layout.   On Sept 27, he
>     > committed the changes:
>     >
>     http://eight.pairlist.net/pipermail/geeklog-cvs/2007-September/003523.html
>     > <
>     http://eight.pairlist.net/pipermail/geeklog-cvs/2007-September/003523.html>
>     > Too bad this particular change doesn't actually work, it just
>     > complicates the code and doesn't produce anything useful.
>     have discussed with Dirk on this and posted it in a much reduced
>     version of what was initially announced, and it works fine. If you
>     have
>     problems to get it run I recommend you to discuss technical issues
>     instead of accusing people. How often do i have to say this?
>     > The XHTML is finally adopted, talk about new features!
>     > Let's see, Blaine committed a new feature just this week.  I'm
>     sure I
>     > could go dig through CVS and find a few more, but why bother...
>     Yeah, why bother if you can just go and complain, why work on
>     something?
>     > The reason I am so passionate about this is that there are a few
>     > things with Geeklog v1.5 that really bring Geeklog to a point
>     that it
>     > is truly an incredible CMS.  The new installation and online
>     > configuration are wonderful enhancements that addressed an area
>     where
>     > Geeklog needed a shot in the arm.
>     > The XHTML capability now opens up Geeklog to some incredible
>     options.
>     Why do you complain that it is in the code then?
>
>     Oliver
>     _______________________________________________
>     geeklog-devel mailing list
>     geeklog-devel at lists.geeklog.net
>     <mailto:geeklog-devel at lists.geeklog.net>
>     http://eight.pairlist.net/mailman/listinfo/geeklog-devel
>
>
> ------------------------------------------------------------------------
>
> _______________________________________________
> geeklog-devel mailing list
> geeklog-devel at lists.geeklog.net
> http://eight.pairlist.net/mailman/listinfo/geeklog-devel
>