ant-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Erik Hatcher <jakarta-...@ehatchersolutions.com>
Subject Re: RES: Promoting EARs
Date Wed, 11 Sep 2002 16:30:45 GMT
Chapter 11 of Steve and I's book is entirely on XDoclet.  But the 
XDoclet documentation is very good, in fact.  Especially if you build 
the CVS HEAD version of XDoclet - the documentation is very dynamically 
built from source and other descriptors and is very well done.

	Erik


Gilson Nascimento D Elrei wrote:
> Is there a good (and practical) documentation about it ?
> thanks in advance.
> Gilson
> 
> 
> 
>>----- Mensagem original -----
>>De:		Erik Hatcher [SMTP:jakarta-ant@ehatchersolutions.com]
>>Enviada em:		quarta-feira, 11 de setembro de 2002 13:17
>>Para:		Ant Users List
>>Assunto:		Re: Promoting EARs
>>
>>Along with the other folks great advice, I'd highly recommend you use 
>>XDoclet in conjunction with your EJB stuff.  For example, (and I'm only 
>>using Session Beans, Entity Beans proved too painful for us), I 
>>configure the deployment JNDI names custom per the site I'm building by 
>>using an Ant property in an XDoclet tag.
>>
>>XDoclet expands Ant properties in @tags, which is a very powerful feature.
>>
>>FYI: I'm running XDoclet from CVS HEAD, its the only way to fly with it 
>>right now.  Look for an official release in the near future.
>>
>>	Erik
>>
>>
>>Kyle Adams wrote:
>>
>>>I've been asked to write scripts to automate our promotion process (from
>>>Dev to Test to Production).
>>>
>>>We deploy our applications as EARs, typically containing one WAR and
>>>multiple EJB JARs.  Our promotion process consists of extracting the
>>>deployment descriptors, making necessary changes to move from one
>>>environment to the next, placing the modified DDs back in the archives,
>>>and FTPing everything up.
>>>
>>>Currently we do this by hand.  Looking at the core set of Ant tasks, I
>>>see some problems with automating the process - I can unear the
>>>application, then unzip all of the archives it contains, and use a
>>><patternset> to only extract deployment descriptors.  The problem is
>>>that all of our EJB DDs have the same names - ejbjar.xml and
>>>weblogic-ejbjar.xml.  Thus, they end up over-writing each other on
>>>extraction.
>>>
>>>Thus, I'm at a loss as to easy ways to extract files (with the same
>>>name) from multiple archives in such a way that they don't stomp all
>>>over each other.  I could write my own task, creating separate
>>>subdirectories within the destination directory to separate out the EJB
>>>DDs, but I wanted to check if anyone else has addressed this issue
>>>before.
>>>
>>>Thanks,
>>>
>>>Kyle Adams
>>>Java Developer
>>>Gordon Food Service
>>>
>>>--
>>>To unsubscribe, e-mail:
>>
>><mailto:ant-user-unsubscribe@jakarta.apache.org>
>>
>>>For additional commands, e-mail:
>>
>><mailto:ant-user-help@jakarta.apache.org>
>>
>>>
>>>
>>
>>
>>--
>>To unsubscribe, e-mail:   <mailto:ant-user-unsubscribe@jakarta.apache.org>
>>For additional commands, e-mail: <mailto:ant-user-help@jakarta.apache.org>
> 
> 
> --
> To unsubscribe, e-mail:   <mailto:ant-user-unsubscribe@jakarta.apache.org>
> For additional commands, e-mail: <mailto:ant-user-help@jakarta.apache.org>
> 
> 
> 



--
To unsubscribe, e-mail:   <mailto:ant-user-unsubscribe@jakarta.apache.org>
For additional commands, e-mail: <mailto:ant-user-help@jakarta.apache.org>


Mime
View raw message