This page is no longer maintained — Please continue to the home page at www.scala-lang.org

Scala Meeting Report 2011/10/11

3 replies
adriaanm
Joined: 2010-02-08,
User offline. Last seen 31 weeks 4 days ago.
This summary is intended for Scala contributors and maintainers, and assumes a certain familiarity with Scala's internals.
It is for your information only, provided “AS IS” with no warranties and it confers no rights (but it doesn't hurt to ask).

Scala Meeting Report 2011/10/11 (apologies for the delay)

Attending

Tiark: paper submissions, scala-virtualized tutorials
Chris: project slick -- a better linq
Vlad: paper submission PEPM2012
Adriaan: paper submission PEPM2012, met with Barcelona guys, resubmitting project, abstract due end of the month
Martin: string interpolation!
Toni: company work
Philipp: wrote tutorial on SourceContexts, papers (PEPM, NIPS) , work on the Delite profiler
Manohar: talk with Tiark and Stanford guys - aim for a longer term paper - so looking into delite
Eugene: talked to people about macros. There's a frightening pattern: usually everyone is against that stuff, but when I show the examples from our proposals - the attitude changes to "wow, neat idea". That's a good sign, I think =)
Miguel: CLR generics - details in Scala compiler corner
Lukas: more work on effects plugin, fixed many bugs. checks collections successfully now (builders etc). spent some time trying to get it to run in eclipse (build and/or presentation compiler).
Vojin: Working on loop fusion for PLDI submission. Fixed 2 minor bugs. Written failing test cases which I am now fixing.
Nada: Added more primitives (variables, arrays, for loops) to LMS JS; with Greg, built a convenient way using tuple packing/unpacking to handle multiple-arity LMS functions uniformly
Greg: Building IDE for Scala Virtualized and Scala+GWT, hacking js+lms stuff with Nada
Alex: staged collections for PLDI submission. Currently on parallel hash tables.
Heather: little NIPS workshop paper, profiler visualization, started over on doc-site- lots of progress.
Hubert: stuck with typer errors and enjoying it
Paul: Mastering method synthesis, finishing the pattern matcher parserwise, chasing performance mysteries, knocking over old dominoes like varargs eta expansion.
josh: Github move


Discussion

Pattern matching syntax [paul]
def f(x: Any) = x match {
 case _: Array[_ <: AnyRef] => true
 case _ => false
}
def f(x: Any) = x match {
 case x: Array[t <: Foo] => x(0)
}
def f(x: Any) = x match {
 case x: t forSome { type t <: String } =>
}
object Foo { def unapply[T, U](x: Any): Option[(T, U)]
def f(x: Any) = x match {
 case Foo[t, u](x) => (x: t) => 5
}

this currently does not parse, but we need it to distinguish various Array cases e.g., Array[_ <: String] and Array[_ <: AnyRef] denote different cases

Documentation
Using github pages (jekyll) -- moving away from the pythonic stuff


GitHub repo for SIDs
Summary of the discussion from scala-devel:
  • A single repo at github, multiple branches to store SID-specific code
  • Proposals are in plain text format, somehow (?) allow to attach google docs
  • Fork and pull request - bread and butter for collaborations, since everything is plain text. Also, we can use code reviews and inline comments for both the code and the proposals.
  • New mailing list: scala-sips, maybe, issue tracker + comments at github will do?
  • Disqus comments
  • Looking for someone who will make this beautiful. One of the design suggestions: http://scalagwt.github.com/samples/Mnemonics.html
Simon Ochsenreither
Joined: 2011-07-17,
User offline. Last seen 42 years 45 weeks ago.
Re: Scala Meeting Report 2011/10/11
Thanks for the update!

Regarding the SIP page:

Any reason why all comments were deleted? Imho it is not nice towards those who thought about the proposals and spent time writing down their comments.

Regarding the work mentioned in “Attending”:

Is there anyone overseeing the evolution of the standard library?
Imho things should start getting more professional than “PhD student finishes thesis, dumps source without any ScalaDoc into the source tree and runs as fast as he can”.
Would be nice to have someone who looks after the state and the further development of the library solely on technical merit and user requirements (and not on “Can we turn this new feature into a paper?”) and has abilities to let people form JSR-like expert groups who can work on a topic.

Thanks and bye,


Simon


Joshua.Suereth
Joined: 2008-09-02,
User offline. Last seen 32 weeks 5 days ago.
Re: Re: Scala Meeting Report 2011/10/11


On Tue, Oct 18, 2011 at 11:59 AM, Simon Ochsenreither <simon.ochsenreither@googlemail.com> wrote:
Thanks for the update!

Regarding the SIP page:

Any reason why all comments were deleted? Imho it is not nice towards those who thought about the proposals and spent time writing down their comments.

Ouch - This is an issue with our Disqus integration.    The URLs for these SIPs had double / in them.   Disqus commenting uses the URL to uniquely identiy a website for comments.   All the comments are still on disqus.  I'm working on pulling them back.  Thanks for mentioning this, although It'd be nice if you had opened a ticket on the github project sooner so I would have known to look into this.
I'll revert the page to it's previous (ugly url) state so the comments come back and work on the Disqus integration before fixing that request.
<snip/>
Simon Ochsenreither
Joined: 2011-07-17,
User offline. Last seen 42 years 45 weeks ago.
Re: Re: Scala Meeting Report 2011/10/11
Thanks!

Copyright © 2012 École Polytechnique Fédérale de Lausanne (EPFL), Lausanne, Switzerland