On 10.02.2004 19:44, Alan wrote:

I'm learning forms.

The registration example uses JXTemplate to generate the
confirmation page. Where does JXTemplate stand these days? Is it
a first class citizen of the Repblic of Cocoon?

The best resource on this is the CVS: http://cvs.apache.org/viewcvs.cgi/cocoon-2.1/src/java/org/apache/cocoon/generation/JXTemplateGenerator.java.

Here's another metric:


    [EMAIL PROTECTED] cocoon-2.1.3]$ find . -name \*.xsp | wc
          150     150    9546
    [EMAIL PROTECTED] cocoon-2.1.3]$ find . -name \*.jx | wc
          13      13     800

    It appears that XSP, although folks are moving away from it, still does a
    lot of work in the Cocoon 2.1.3 distribution.

JXTemplate doesn't look like it has as much employment, yet.

Of course. Have a look how long JXTemplate exists. Furthermore finding files by endings is a disadvantage for jx (e.g. http://cvs.apache.org/viewcvs.cgi/cocoon-2.1/src/blocks/woody/samples/forms/form2_jx.xml). Grepping on the namespace URIs should give better results.


  Beside handling form output, where else would I use JXTemplate
  in my sitemap?

You can use it everywhere where you need a template technique. Or shall I say where a template technique is powerful enough for you? In comparison to JSP you have the advantage of applying transformer afterwards.

Is JXTemplate moving into the vacuum created by XSP then? Anyone else want to chime in? It seems to me to be the case if JXTemplate has been moved to the core (from the CVS log) and XSP is moving out.

Not that official as replacement for XSP, but it's one possibility. JXTemplate was probably in the scratchpad before. XSP will probably moved out of core in Cocoon 2.2 in its own block (but no deeper discussion or vote on this have taken place yet).


And to answer your fears on your blog: The problem of XSP is the implementation, not the idea. It allows to much, so that we can't be restrictive on validating or similar. We don't get bugs fixed as we can't hit all use cases (e.g. http://nagoya.apache.org/bugzilla/show_bug.cgi?id=15841).

I'm not sure what I'm looking at in the bug report, but I can see that there is no good way to validate XSP. Especially if you can inject elements from Java snippets.

That's exactly the problem I wanted to point out. There are probably more examples on this.


Joerg

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



Reply via email to