On 30/09/2012, at 2:00 PM, Guillaume Laforge wrote:

> On Sun, Sep 30, 2012 at 2:23 PM, Luke Daley <[email protected]> wrote:
> 
> On 30/09/2012, at 1:11 PM, Guillaume Laforge wrote:
> 
> > On Sun, Sep 30, 2012 at 12:33 PM, Luke Daley <[email protected]> 
> > wrote:
> > [...]
> > > when I then use that JAR
> >
> > What do you mean here? Are you grabbing the jar out of the maven local repo?
> >
> > Yes, I was copying that jar from the maven local repo manually to the 
> > WEB-INF/lib folder of a web application (a Gaelyk app in my case).
> 
> Why not copy it from the build?
> 
> I ended up doing clean builds and copying from the build, but that's not 
> solving the problem :-)
> 
>  
> >  [...]git clean -f -d did the trick.
> >
> > I still can't reproduce. The jar is written into the local maven cache 
> > every time for me.
> >
> > You're sure it's an updated JAR that is pushed in the local maven cache?
> > That it really contains changes made?
> 
> I'm using…
> 
> shasum 
> subprojects/groovy-servlet/target/libs/groovy-servlet-2.1.0-SNAPSHOT.jar && 
> shasum 
> ~/.m2/repository/org/codehaus/groovy/groovy-servlet/2.1.0-SNAPSHOT/groovy-servlet-2.1.0-SNAPSHOT.jar
> 
> To check the content. It's definitely changing in sync.
> 
> Well, in the case of Groovy, it doesn't prove anything

If there was a problem where changes that were made to Groovy files were not 
picked up on compilation, we'd definitely know about it. Everyone would have 
the problem. Therefore, the fact that the jar in the build is the same as what 
is in the maven local cache is a good indicator that the change has been 
applied.

> , as the groovy compiled classes contained a kind of timestamp field that 
> changes upon recompilation, even if no changes to the code have been made. So 
> you could think the JAR is indeed updated, which it is, but perhaps it may 
> not include the actual changes.
> 
> I'll try to have another look at it and see what's going on, when I have some 
> time.

-- 
Luke Daley
Principal Engineer, Gradleware 
http://gradleware.com


---------------------------------------------------------------------
To unsubscribe from this list, please visit:

    http://xircles.codehaus.org/manage_email


Reply via email to