-Re: Bumping up version of Groovy in Bigtop
Stephen Chu 2012-10-27, 23:00
+1 on bumping up Groovy version to 1.8.
On Sat, Oct 27, 2012 at 1:51 PM, Wing Yew Poon <[EMAIL PROTECTED]> wrote:
> It's good to hear from someone who's been using 1.8.
> I was looking recently at the version of groovy used in itest, and
> surprised that it was such an ancient version (1.6). I'm in favor of
> moving to a more recent stable version like 1.8. I am undecided if we
> should move to 2.x. Since we are not currently using Java 1.7 yet (for
> hadoop etc), I don't see the benefit as far as invokedynamic is
> On Sat, Oct 27, 2012 at 4:44 AM, Steve Loughran <[EMAIL PROTECTED]>
> > On 26 October 2012 19:54, Roman Shaposhnik <[EMAIL PROTECTED]> wrote:
> >> On Fri, Oct 26, 2012 at 11:45 AM, Johnny Zhang <[EMAIL PROTECTED]>
> >> wrote:
> >> > Roman, how is your experience with Groovy's back compatibility? If we
> >> bump
> >> > to 1.8 or 2.0, do we expect any current tests failures?
> > 1.8.x is pretty good, no problems I've encountered except that some of
> > maven plugins are hard-coded to older ones (that's maven for you)
> > I haven't used 2.x. What it does promise is invokedynamic on Java7 and
> > performance gains from that
> > http://dist.groovy.codehaus.org/JVM%20Summit%202012%20-%20Groovy.pdf
> > -plus a static compile option.
> > for using Groovy in apps (as opposed to test code or other stuff where
> > want high-perf groovy, this is all good).
> > For tests? I don't see Groovy the bottleneck in my own code, though apart
> > from some GUI stuff.
> > Groovy actually makes writing Swing apps something that you'd actually
> > consider doing:
> > You still need a reason to do so (here the monitoring/demo tool for HDFS
> > and JT failure), but once you've made that decision, Groovy is the
> > to start with -at least until Java adds closures.