Hi Tim,

I couldn't wait to try out the latest version of the integration.  It looks 
pretty impressive!

I did bump into a couple of problems, one for which I found a solution, and 
another one for which I didn't.

Here's the one that I did get around solving (because it was simple enough?): 
upon returning from the editor screen the language setting is lost and the 
default language version of the document is being looked up, instead of the 
language that we sterted off with.

I changed line 297 in usecase-tiny_mce.xmap from:
<map:redirect-to 
uri="{request:contextPath}/{page-envelope:publication-id}/authoring{page-envelope:document-id}.html?lenya.usecase=checkin&amp;lenya.step=checkin&amp;backup=false"/>
To
<map:redirect-to 
uri="{request:contextPath}/{page-envelope:publication-id}/authoring{page-envelope:document-id}_{page-envelope:document-language}.html?lenya.usecase=checkin&amp;lenya.step=checkin&amp;backup=false"/>

As far as I can tell, this seems to have solved the problem.

The other problem: aggregate.xsl generates URLs with &amp; entities instead of 
simple ampersand char's.  This basically messes up the parameters that are 
handed over to the server when asking for the image list, link list, and when 
trying to upload images.
There is probably an obvious reason for this, but I fail to see it.

I am running Apache Lenya 1.2.4 on tomcat 5.5.15 (JDK 1.5, using the xalan 
upgrade) and Firefox.  I have not tried this on the recommended tomcat 5.0.* 
(JDK 1.4.*).

Thanks in advance for any help,

Rudi


-----Original Message-----
From: Tim Hannigan [mailto:[EMAIL PROTECTED] 
Sent: Thursday, March 30, 2006 5:14 AM
To: [email protected]
Subject: TinyMCE editor integration update

Hi all,

the integration with Lenya and TinyMCE has been coming along quite well.
I've updated the Wiki with the latest code we're using in out Lenya  
pilot project.

Differences between the previous implementation and the current one  
include:
-the editor is now looking for a css file in each publication (pubid/ 
resources/shared/css) entitled "editor.css" that will contain any CSS  
that the editor should use both in it's display and production; this  
also allows you to setup drop-down menu's directly from your styles

-image and asset browsing are now dynamically generated by a pipeline  
using the xpathdirectory to read files (ie. no more static  
image_list.js and asset_list.js files are needed in the resources  
directory); as a result, we can go back to using the build version of  
the UploadAction class.

-the code generated is now being checked against the full XHTML spec  
and cleaned up by the editor automatically open loading

-most of the files needed for this implementation are bundled within  
the usecase directory; as this implementation continues, we hope to  
roll it entirely into the usecases directory

-cleaned up the sitemaps; this is a continual effort, and will get  
better


If you have any questions regarding this implementation, please post  
to the list. We're using TinyMCE as our full-time editor now, and  
we're having relatively few problems problems with it (across  
browsers). The robustness of the editor is great, and in my opinion  
far exceeds Kupu as an HTML editor.

-Tim



-- 
Tim Hannigan
Manager, Electronic Communications
Marketing & Communications
Queen's University
Kingston ON     K7L 3N6
Phone:  613-533-6000 ext. 74126
Fax:            613-533-6652




---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Reply via email to