forrest-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Robert Koberg <...@koberg.com>
Subject Re: Tabs and menu (was: RE: bert skin added)
Date Mon, 03 Jun 2002 10:12:57 GMT
Steven Noels wrote:

>On Mon, 3 Jun 2002, Bert Van Kets wrote:
>
>  
>
>>Thanks David.
>>I'm working on the menu and tabs now.  The problem I'm having is in knowing 
>>what file is selected in the sub menu's.  Seems the Cocoon site is still 
>>having the same problem.
>>
>>Steven, is your "book.xml on steroids" coping with this problem, or need I 
>>go further on this?
>>    
>>
>
>the idea would be to pass the current request URI as a parameter to the
>book2menu.xsl (nothing special, you get the picture). Why not move the
>tabs in a separate pipeline, too, with an accompagnying book2tabs.xsl.
>

Can this be used to build a snail trail (the linked path to the page, 
e.g.: cocoon : docs : a page)? Does the snail trail have to be hardcoded 
on a content piece?

>
>Using the steroid stuff, we'd move from book.xml to libre.xml - I'll 
>commit a sample of its current configuration this afternoon. We're still 
>in the process of deciding where libre should go, since it has a scope 
>which is broader then forrest/cocoon - it could be possible we post it at 
>SF and include a libre.jar in forrest.
>
>  
>
>>The Tabs have the same problem.  It's not clear to me whether tabs span a 
>>sub-directory or a certain number of book.xml files.  I'll have to check 
>>the archives, unless someone can fill me in here.
>>What's the selection criteria for the tab?  The most obvious one is "the 
>>user selected it", but that needs to be stored somewhere.  Where????
>>    
>>
>
>An extra pipeline to start with IMO. I need to run right now, but maybe
>the top level directories in src/documentation/content/xdocs should be
>reflected as tabs... opinions on this?
>  
>
Does this do anything to address the issue of internal links in content? 
What happens if libre.xml (or any site) gets reorganized? Do you then 
have to go through each content piece and update the links? Using 
chained transformations this way seems to only serve dogma. There is 
much benefit that could be achieved by re-using the libre.xml data (and 
it would be nice if it maintained the same strucutre...) during the 
entire page's transformation.

best,
-Rob


Mime
View raw message