- About Scala
- Documentation
- Code Examples
- Software
- Scala Developers
Scala Meeting report, 2009-11-24
Wed, 2009-11-25, 21:30
We are currently publishing a summary of each of the weekly meetings of the
Scala Core Team.
This information is made available as a service to the community. It is by
necessity rather brief and gives only a rough approximation of the main
points of discussions explored during each meeting; it should not be taken
as a source of reliable information, nor as a record of concrete or firm
decisions, nor as anything other than a record of a simple discussion.
The summary that follows is primarily intended for Scala contributors and
maintainers. If you are not a contributor to the Scala system, the
information below is unlikely to be very useful to you, and you might lack
some of the necessary background to make sense of all the discussion items.
We do not have enough people on the team to be able to write a more
complete record, and we might also not have the resources to discuss every
point in detail afterwards. Nevertheless, we hope that this record, cursory
as it is, is better than nothing.
------------------------------------------------------------------------------
Report from the Scala Core Team meeting, 2009-11-24:
People attending the meeting: Martin, Iulian, Donna, Eugene, Alex, Hubert,
Tiark, Gilles, Toni, Adriaan, Lukas, Philipp.
Topics discussed:
* Current work:
- the new scaladoc will be committed soon
- bug fixing
- more work on specialization
- the compiler is now interning symbols at classloading
- scalac can now generate static constructors
* 2.8 Beta:
- Did people in the community test their projects against 2.8.0.beta.RC1?
- Lift: yes, but not the Scalacheck tests; we received no feedback
from other projects
- David Pollak reports they did not set up a testing infrastructure
for other tools. They have not been coordinating with other framework authors.
- We'll release a Beta1-RC2, non-optimized, and tell people to test
their code and let us know whether we can go ahead and release RC3 optimized.
* Eclipse:
- Build Manager is not yet there exactly. Still wrong dependencies from
time to time
- The plugin is basically feature-complete now. It needs more work on
robustness and performance
- Miles needs a couple of weeks
- most user feedback is on unreliable building
- build manager: most of the problems should be represented as tickets
by now
- Miles will prioritise them and we will work on getting the high-prio
ones fixed asap
* Specialisation of pattern matching
- In a class A[T] with "T" specialised for "Int", the following match
can be replaced simply by "1":
(x: T) match {
case _: Int => 1
...
}
- Yes, Iulian will add that optimisation.
- However, the following match will always return "0", including when
the type of "x" is "Int":
(x: T) match {
case _: AnyRef => 0
case _: Int => 1
}
This is because of the boxing of Int that happens as "x" has a
parametric type.
- There should be a warning in such cases (there is none now) and the
specialisation optimisation must not be triggered.
* reverse in SequenceLike:
- In certain situations (for example priority queues) reverse makes no
sense. We'll retrieve the (implicitly supplied) ordering (using the
implicitly method), swap it, and supply explicitly to the resulting collection.
Wed, 2009-11-25, 21:57
#2
Re: Scala Meeting report, 2009-11-24
>>>>> "Antonio" == Antonio Cunei writes:
Antonio> * 2.8 Beta: - Did people in the community test their projects
Antonio> against 2.8.0.beta.RC1?
for whatever it's worth:
both my Project Euler solutions and my main work project (NetLogo) have
been working fine with 2.8.0.beta.RC1. I have filed bug reports on the
few problems I found (none of them showstopping). Our toolchain
includes sbt 0.6.3 and ScalaTest, both working fine, though I'm needing
to use Josh Cough's alternate build of ScalaTest
(1.0.1-2-for-scala-2.8.0.Beta1-RC1-with-test-interfaces-0.1-SNAPSHOT),
not the standard build, in order for it to work with sbt.
Wed, 2009-11-25, 22:17
#3
Re: Scala Meeting report, 2009-11-24
On Wed, Nov 25, 2009 at 12:56 PM, Seth Tisue wrote:
>>>>>> "Antonio" == Antonio Cunei writes:
>
> Antonio> * 2.8 Beta: - Did people in the community test their projects
> Antonio> against 2.8.0.beta.RC1?
>
> for whatever it's worth:
>
> both my Project Euler solutions and my main work project (NetLogo) have
> been working fine with 2.8.0.beta.RC1. I have filed bug reports on the
> few problems I found (none of them showstopping). Our toolchain
> includes sbt 0.6.3 and ScalaTest, both working fine, though I'm needing
> to use Josh Cough's alternate build of ScalaTest
> (1.0.1-2-for-scala-2.8.0.Beta1-RC1-with-test-interfaces-0.1-SNAPSHOT),
> not the standard build, in order for it to work with sbt.
Similarly, my personal projects all work, including tests based on
scalacheck and scalatest.
I haven't tried anything with -optimize or -Yspecialize recently though.
Wed, 2009-11-25, 22:27
#4
Re: Scala Meeting report, 2009-11-24
Excellent, thanks!
Keep the reports coming in, it's time for us to get some feedback now! :)
Toni
David Hall wrote:
> On Wed, Nov 25, 2009 at 12:56 PM, Seth Tisue wrote:
>>>>>>> "Antonio" == Antonio Cunei writes:
>> Antonio> * 2.8 Beta: - Did people in the community test their projects
>> Antonio> against 2.8.0.beta.RC1?
>>
>> for whatever it's worth:
>>
>> both my Project Euler solutions and my main work project (NetLogo) have
>> been working fine with 2.8.0.beta.RC1. I have filed bug reports on the
>> few problems I found (none of them showstopping). Our toolchain
>> includes sbt 0.6.3 and ScalaTest, both working fine, though I'm needing
>> to use Josh Cough's alternate build of ScalaTest
>> (1.0.1-2-for-scala-2.8.0.Beta1-RC1-with-test-interfaces-0.1-SNAPSHOT),
>> not the standard build, in order for it to work with sbt.
>
> Similarly, my personal projects all work, including tests based on
> scalacheck and scalatest.
>
> I haven't tried anything with -optimize or -Yspecialize recently though.
>
Wed, 2009-11-25, 23:07
#5
Re: Scala Meeting report, 2009-11-24
Antonio Cunei wrote:
> * 2.8 Beta:
> - Did people in the community test their projects against 2.8.0.beta.RC1?
> - Lift: yes, but not the Scalacheck tests; we received no feedback
> from other projects
I tested Scala Migrations and made a few changes so the code would compile with
both 2.7.x and 2.8.x, but I got blocked at #2635. Internal projects also saw
some work to work with the new collections API.
I'm waiting for the next official release which fixed #2635 before continuing up
the stack of Scala software we use.
Regards,
Blair
Wed, 2009-11-25, 23:27
#6
Re: Scala Meeting report, 2009-11-24
On Wed, 2009-11-25 at 14:06 -0800, Blair Zajac wrote:
> I'm waiting for the next official release which fixed #2635 before continuing up
> the stack of Scala software we use.
You're in luck since Antonio just announced that. :)
Ismael
Wed, 2009-11-25, 23:37
#7
Re: Scala Meeting report, 2009-11-24
On 26/11/2009, at 7:30 AM, Antonio Cunei wrote:
> * 2.8 Beta:
> - Did people in the community test their projects against2.8.0.beta.RC1?
> - Lift: yes, but not the Scalacheck tests; we received no feedback from other projects
As of yesterday, my Kiama library (http://kiama.googlecode.com) is running
properly on 2.8.0.beta.RC1. I'm using
Mac OS X 10.6.2
Java HotSpot(TM) 64-Bit Server VM, Java 1.6.0_15
sbt 0.5.7-p1
scalacheck_2.8.0.Beta1-RC1-1.7-SNAPSHOT
scalatest-1.0.1-for-scala-2.8.0.Beta1-RC1-SNAPSHOT
All of my tests pass. In the course of testing, I discovered one compiler
bug:
https://lampsvn.epfl.ch/trac/scala/ticket/2691
which I have worked around for now.
Another notable fact perhaps is that I have switched to use the packrat
parsers implementation in the library. I have sent some more detailed
feedback on that library to Tiark and Adriaan.
I will move to 2.8.0.beta.RC2 once there are builds of sbt, scalatest and
scalacheck.
cheers,
Tony
Thu, 2009-11-26, 00:27
#8
Implicit classes proposal
Hi,
Has anyone given a thought on Jorge's proposal for implicit classes? http://www.scala-lang.org/node/1344That would reduce some boilerplate and common programming errors with respect to the "Pimp my library" pattern, which I expect to be wildly used.
Is it on the roadmap for a next version of Scala? More generally speaking I would like to have some references on the future stability of Scala. Scala-2.8.0 is breaking a lot of stuff (for the greater good of everyone, I agree), but what kind of desirable future features do you envision to be breaking our existing libraries? Another way of asking the question is: if 2.8.x is going to be "the" stable foundation for Scala at large, what may what is likely to be missing until the next "major" release?
Thanks for your insights on this, this is also part of the "Scala story" we can sell to our clients.
Eric. ----------------------------------------------
Eric TORREBORRE
T +61 411 707 402
E etorreborre@yahoo.com
B http://etorreborre.blogspot.com
P http://specs.googlecode.com
----------------------------------------------
Has anyone given a thought on Jorge's proposal for implicit classes? http://www.scala-lang.org/node/1344That would reduce some boilerplate and common programming errors with respect to the "Pimp my library" pattern, which I expect to be wildly used.
Is it on the roadmap for a next version of Scala? More generally speaking I would like to have some references on the future stability of Scala. Scala-2.8.0 is breaking a lot of stuff (for the greater good of everyone, I agree), but what kind of desirable future features do you envision to be breaking our existing libraries? Another way of asking the question is: if 2.8.x is going to be "the" stable foundation for Scala at large, what may what is likely to be missing until the next "major" release?
Thanks for your insights on this, this is also part of the "Scala story" we can sell to our clients.
Eric. ----------------------------------------------
Eric TORREBORRE
T +61 411 707 402
E etorreborre@yahoo.com
B http://etorreborre.blogspot.com
P http://specs.googlecode.com
----------------------------------------------
Thu, 2009-11-26, 01:07
#9
Re: Implicit classes proposal
On Wed, Nov 25, 2009 at 3:22 PM, Eric Torreborre wrote:
> Hi,
> Has anyone given a thought on Jorge's proposal for implicit
> classes? http://www.scala-lang.org/node/1344
Last I recall, it was basically rejected. There were a number of
issues, not least of which is how to handle "top level" implicit
classes, and what the right syntax for importing the
implicitly-generated implicit conversion was.
Tue, 2009-12-01, 17:47
#10
Re: Implicit classes proposal
On Thu, Nov 26, 2009 at 12:22 AM, Eric Torreborre wrote:
> Hi,
> Has anyone given a thought on Jorge's proposal for implicit
> classes? http://www.scala-lang.org/node/1344
It's not been rejected as such, but we did not have the time to
consider it sufficiently. There are a number of tricky issues, most
notably interaction with top-level classes. Now that we have package
objects we might be able to do something about this, but we have to
wait a bit until the details of package objects settle down
sufficiently to do the next step. So, it's not off the map, but also
there's no firm plan to accept them.
Cheers
On Wed, 2009-11-25 at 21:30 +0100, Antonio Cunei wrote:
> * 2.8 Beta:
> - Did people in the community test their projects against 2.8.0.beta.RC1?
I wanted to do this, but we rely on configgy and specs. So, I ported
configgy, which also depends on specs for the tests. It seems like there
are some issues compiling specs with the beta candidate. See (and
follow-ups):
http://groups.google.com/group/specs-users/msg/a77ea850a935ae68
I noticed that Eric later filed (although not sure if it's related or
something else he found):
https://lampsvn.epfl.ch/trac/scala/ticket/2683
I saw builds for scalatest and scalacheck published in the maven repo,
so it seems like things are starting to get into shape.
Best,
Ismael