0.12.4

[Mark Harrah]

sbt 0.12.4 is now available. This release has two primary purposes: backport cache and incremental compiler fixes from 0.13 and provide better support for cross-building of plugins for 0.13. Details on the latter are in 0.13 related announcements.

See the Changes page[1] and the v0.12.4 commit list[2] for details. This release is intended to be binary and source compatible with 0.12.x. Use it in an existing 0.12 project by setting sbt.version=0.12.4 in project/build.properties. There is a new launcher[3], which includes some important fixes. No changes should be necessary to your project definition and all plugins published for 0.12.x should still work.

Thanks,
Mark

[1] http://www.scala-sbt.org/0.12.4/docs/Community/Changes.html
[2] https://github.com/sbt/sbt/commits/v0.12.4
[3] http://repo.typesafe.com/typesafe/ivy-releases/org.scala-sbt/sbt-launch/0.12.4/sbt-launch.jar

[virtualeyes]

Thanks for the update, Mark.

Just gave 0.12.4 a spin with a multi-project build on Fedora 18 with Scala 2.10.2 and Java 1.7.0_04

On first compilation was horribly slow, and then subsequent compilations were about 50% slower than the previous. Basically takes 3 clean/compile cycles to get to blazing fast cached compilation on 0.12.4, whereas (I believe) 0.12.3 only took 2 clean/compiles to warm up the cache.

Anyway, once warmed up 0.12.4 is as fast or faster than 0.12.3.

Have yet to check out incremental compilation improvements, seemed pretty good on 0.12.3, sub projects mitigate most of the "scala is slow" stigma (for me at any rate).

Thanks for putting in the hard yards, development in Scala would be a mess without SBT.

[Mark Harrah]

On Sun, 30 Jun 2013 07:50:31 -0700 (PDT)
virtualeyes <sit...@gmail.com> wrote:

> Thanks for the update, Mark.
>
> Just gave 0.12.4 a spin with a multi-project build on Fedora 18 with Scala
> 2.10.2 and Java 1.7.0_04
>
> On first compilation was horribly slow, and then subsequent compilations
> were about 50% slower than the previous. Basically takes 3 clean/compile
> cycles to get to blazing fast cached compilation on 0.12.4, whereas (I
> believe) 0.12.3 only took 2 clean/compiles to warm up the cache.
>
> Anyway, once warmed up 0.12.4 is as fast or faster than 0.12.3.

This is somewhat surprising because there weren't big changes and I would expect the opposite result due to the the class loader fix. Can you verify that you used the 0.12.4 launcher? Also, was each compilation a clean and then compile?

> Have yet to check out incremental compilation improvements, seemed pretty
> good on 0.12.3, sub projects mitigate most of the "scala is slow" stigma
> (for me at any rate).

The incremental compilation related fixes were for degenerate cases. There weren't general algorithmic improvements. Those come in 0.13.0. You won't seen an improvement in 0.12.4 unless your code hit one of these bugs that were fixed.

There was a fix for class loading unrelated to incremental compilation, but that should have produced the opposite result.

> Thanks for putting in the hard yards, development in Scala would be a mess
> without SBT.

You're welcome!

-Mark

> On Thursday, June 27, 2013 5:52:34 PM UTC+2, Mark Harrah wrote:
> >
> > sbt 0.12.4 is now available. This release has two primary purposes:
> > backport cache and incremental compiler fixes from 0.13 and provide better
> > support for cross-building of plugins for 0.13. Details on the latter are
> > in 0.13 related announcements.
> >
> > See the Changes page[1] and the v0.12.4 commit list[2] for details. This
> > release is intended to be binary and source compatible with 0.12.x. Use it
> > in an existing 0.12 project by setting sbt.version=0.12.4 in
> > project/build.properties. There is a new launcher[3], which includes some
> > important fixes. No changes should be necessary to your project definition
> > and all plugins published for 0.12.x should still work.
> >
> > Thanks,
> > Mark
> >
> > [1] http://www.scala-sbt.org/0.12.4/docs/Community/Changes.html
> > [2] https://github.com/sbt/sbt/commits/v0.12.4
> > [3]
> > http://repo.typesafe.com/typesafe/ivy-releases/org.scala-sbt/sbt-launch/0.12.4/sbt-launch.jar
> >
>

> --
> You received this message because you are subscribed to the Google Groups "simple-build-tool" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to simple-build-t...@googlegroups.com.
> To post to this group, send email to simple-b...@googlegroups.com.
> Visit this group at http://groups.google.com/group/simple-build-tool.
> For more options, visit https://groups.google.com/groups/opt_out.
>
>

[eugene yokota]

I'm seeing another scripted issue. This time 0.12.4 running 0.12.4 scripts: https://github.com/sbt/sbt/issues/828

-eugene

[virtualeyes]

On Monday, July 1, 2013 4:34:47 PM UTC-4, Mark Harrah wrote:

On Sun, 30 Jun 2013 07:50:31 -0700 (PDT)
virtualeyes <sit...@gmail.com> wrote:

> Thanks for the update, Mark.
>
> Just gave 0.12.4 a spin with a multi-project build on Fedora 18 with Scala
> 2.10.2 and Java 1.7.0_04
>
> On first compilation was horribly slow, and then subsequent compilations
> were about 50% slower than the previous. Basically takes 3 clean/compile
> cycles to get to blazing fast cached compilation on 0.12.4, whereas (I
> believe) 0.12.3 only took 2 clean/compiles to warm up the cache.
>
> Anyway, once warmed up 0.12.4 is as fast or faster than 0.12.3.

This is somewhat surprising because there weren't big changes and I would expect the opposite result due to the the class loader fix.  Can you verify that you used the 0.12.4 launcher?  Also, was each compilation a clean and then compile?

Yes, 0.12.4 launcher.

Test results based on repeated clean/compile:
first
[success] Total time: 68 s

second:
[success] Total time: 38 s

third:
[success] Total time: 17 s ( which is really quite fast ;-) )

[Erlend Hamnaberg]

Something strange has happened with 0.12.4.

After the upgrade and running the scripted plugin, the input prompt disappears.

The history still works, but I cannot see what the command I am about to run, actually is.

Maybe this is already reported.

--

Erlend

[Mark Harrah]

On Wed, 14 Aug 2013 08:56:25 +0200
Erlend Hamnaberg <nga...@gmail.com> wrote:

> Something strange has happened with 0.12.4.
>
> After the upgrade and running the scripted plugin, the input prompt
> disappears.
>
> The history still works, but I cannot see what the command I am about to
> run, actually is.
>
> Maybe this is already reported.

There is this:

https://github.com/sbt/sbt/issues/828

A goal for 0.12.4 was to allow building 0.13 plugins that have 0.12 build definitions. Unfortunately the versions of jline used in each case are incompatible, which results in the prompt issues. It could be my fault somewhere though, so can you at least comment on the issue?

-Mark