In my experience with Maven, there is no such thing as a "minor
change" :)
On Fri, 2010-02-12 at 12:12 -0300, Hardy Ferentschik wrote:
> As much as I see the potential of gradle I don't want to be the one to
> push this forward
> right now. A multi-module setup is a minor change in this case.
>
As much as I see the potential of gradle I don't want to be the one to
push this forward
right now. A multi-module setup is a minor change in this case.
--Hardy
On Fri, 12 Feb 2010 11:57:17 -0300, Steve Ebersole
wrote:
> Might I suggest y'all look at gradle since this is something you re ju
Ahahah
I am a big believer in Ricardian's competitive advantage. In this theory, Maven
and I are on a different side of the fence.
Emmanuel
On 12 févr. 2010, at 15:55, Hardy Ferentschik wrote:
> It seems to me that there are potential other benefits we could gain from a
> multi-module
> setup.
Might I suggest y'all look at gradle since this is something you re just
getting ready to start?
On Fri, 2010-02-12 at 11:55 -0300, Hardy Ferentschik wrote:
> It seems to me that there are potential other benefits we could gain from
> a multi-module
> setup.
>
> Maybe Emmanuel should actually d
Solr is not a good use case. We should have make it mandatory in the first
place.
Anyway, we probably need multiple modules in the future but here is the caveat:
- in an ideal world we would split hsearch-core from hsearch-hibernate but
people will then be forced to put these two dependencies t
It seems to me that there are potential other benefits we could gain from
a multi-module
setup.
Maybe Emmanuel should actually do it, since he loves maven so much ;-)
On Fri, 12 Feb 2010 11:46:32 -0300, Sanne Grinovero
wrote:
> it's ok for me;
> the best benefit I expect is to split the tes
I exclude it too :)
On 12 févr. 2010, at 15:42, Hardy Ferentschik wrote:
> Of course it does not offer any additional functionality, but when deployed
> as archetype
> into the maven repo it allows to "bootstrap" a Search enabled maven project
> using the
> maven archetype plugin
> (http://mav
it's ok for me;
the best benefit I expect is to split the testssuite and clean that,
opening the doors to a performance&stress testsuite.
Also the Solr "optional" dependencies should be actually mandatory if
you use the analyzerdefs, so this could eventually be moved to a
separate search module; t
Of course it does not offer any additional functionality, but when
deployed as archetype
into the maven repo it allows to "bootstrap" a Search enabled maven
project using the
maven archetype plugin
(http://maven.apache.org/plugins/maven-archetype-plugin).
I agree in current setup the quickst
Is the quickstart archetype really needed?
On my side, I always exclude it as it goes in my way.
On 12 févr. 2010, at 15:25, Hardy Ferentschik wrote:
> Hi,
>
> For HSEARCH-459 I would like to move Search to a multi-module setup
> (similar to Validator).
> This way the archetype sources will ac
Hi,
For HSEARCH-459 I would like to move Search to a multi-module setup
(similar to Validator).
This way the archetype sources will actually always get compiled and are
less likely to get outdated.
I think we talked before about potential benefits of having a multi-module
build for Search.
11 matches
Mail list logo