On Tue, Sep 21, 2010 at 10:30 AM, Stefan Bodewig <bode...@apache.org> wrote:
> On 2010-09-21, Niall Pemberton wrote:
>
>> On Tue, Sep 21, 2010 at 8:28 AM,  <bode...@apache.org> wrote:
>>> Author: bodewig
>>> Date: Tue Sep 21 07:28:24 2010
>>> New Revision: 999257
>
>>> URL: http://svn.apache.org/viewvc?rev=999257&view=rev
>>> Log:
>>> try publishing beanutils' POM as beanutils-core POM, not sure it will work
>
>> It should be OK. It will have an additional (unnecessary) dependency
>> on Commons Collections - but that shouldn't cause any problems.
>
> The reason I added it was
>
> <http://vmgump.apache.org/gump/public/checkstyle/checkstyle/gump_work/build_checkstyle_checkstyle.html>
>
> Which looks like a project using (a released version of) beanutils but
> not depending on commons-collections, like we've seen for
> digester-test.  checkstyle doesn't depend on beanutils but only on
> beantils-core but since we publish the normal beanutils jar as
> beanutils-core in Gump it may be that this dependency on
> commons-collections only exists inside Gump.

BeanUtils used to have a copy of the Collections FastHashMap (and 4
other collections classes). With those BeanUtils core had no
dependency on Commons Collections. There are other clases in BeanUtils
which did require Commons Collections, but they were excluded from the
Core jar. This was a mess so I removed the copied classes and dumped
the BeanUtils core jar. So now there is only one BeanUtils jar and
anything that depends on it requires Commons Collections:

https://issues.apache.org/jira/browse/BEANUTILS-379

So effectively the "core" dependency has disappeared. Not sure how
gump should/can handle that. Perhaps a *packaged* version of beanutils
core. Or as I guess you're trying to do - feed in BeanUtils as core -
but those projects that depend on it will now requires Commons
Collections.

> The "not sure it will work" part is more about the fact that the POM
> will specify a different artifactId than mvn asks for and I don't know
> whether mvn will ignore this or reject the POM or explode or whatever.

OK, me neither - I guess we'll see after the next run.

Niall

> Stefan
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: general-unsubscr...@gump.apache.org
> For additional commands, e-mail: general-h...@gump.apache.org
>
>

---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscr...@gump.apache.org
For additional commands, e-mail: general-h...@gump.apache.org

Reply via email to