cocoon-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Vadim Gritsenko <>
Subject Re: build system [was: scratchpad blocks]
Date Tue, 29 Oct 2002 04:31:25 GMT
Torsten Curdt wrote:

>while implementing the scratchpad blocks...
>>I am fine with that... So I'll go for
>>         <!-- the following blocks are marked unstable and should be
>>               considered being pre-alpha quality. Their API might change
>>               without notice -->
>>         <myteststuff include="false" status="unstable"/>
>>         <scratchpad include="false" status="unstable"/>
>...I found some things I like to hear some opinions on before changing the 
>whole build system ;-)
>1. I found that the status="unstable" should be not in the properties.xml file 
>but the modules.xml file. since that's the file that describes our blocks
>2. Why is this file called modules.xml?

AFAIU, module.xml is for the gump.

And, while we are on the subject, why, why gump projects / cocoon blocks 
named as "fop-block", and not "cocoon-fop-block"???

Just look at - isn't it 
confusing? One might think that FOP guys come up with some block idea...


> Shouldn't it be better named 
>blocks.xml? But it also includes the core... Should we have a core block?
>3. No matter what I select in the properties.xml (include or not) it seem 
>every block gets included (for installwar) - someone else experienced that 
>before? (I haven't had time to have a closer look yet)
>4. For me it seems like javadocs/documentation as well as the blocks are 
>*always* build (for installwar). no matter what is selected in the 
>properties.xml. this might of course also be related to 3.
>5. Shouldn't we have also a lib directory inside blocks and move the optional 
>jars into those directories? So we can easily copy them only if they are 
>really needed inside the webapp
>6. javadoc spits out dozens of full text lines... warnings and stuff. is that 
>really necessary? Cannot remember to have seen it that verbose before...
>Maybe because I switched to jdk1.4 now? (stupid GLIBC_2.0 bug in 1.3)
>Anyone: before making bigger changes of the build system please talk back to 
>me since I have some things waiting to be committed ;-)
>Fire off!
>it's late - good night :-)

To unsubscribe, e-mail:
For additional commands, email:

View raw message