Stories
Slash Boxes
Comments
NOTE: use Perl; is on undef hiatus. You can read content, but you can't post it. More info will be forthcoming forthcomingly.

All the Perl that's Practical to Extract and Report

use Perl Log In

Log In

[ Create a new account ]

chromatic (983)

chromatic
  (email not shown publicly)
http://wgz.org/chromatic/

Blog Information [technorati.com] Profile for chr0matic [technorati.com]

Journal of chromatic (983)

Friday September 14, 2007
05:09 PM

Perl 6 Design Minutes for 12 September 2007

[ #34455 ]

The Perl 6 design team met by phone on 12 September 2007. Larry, Allison, Patrick, Jesse, Nicholas, and chromatic attended.

Patrick:

  • discussing various S05 changes with Larry
  • implemented some of them to keep Parrot up to date
  • I expect the new compiler tools to use updated S05 versions
  • better to get the updates in before other people start using them
  • the changes look really good
  • they simplify things a lot

Larry:

  • with carping in the backchannels

Patrick:

  • all of my carping is in public on IRC

Larry:

  • unlogged IRC

Patrick:

  • well yeah
  • continuing to work on the revisions to namespace support in PCT
  • I have a design laid out in my head
  • hope to get it implemented this afternoon
  • otherwise just watching the commit list

Jesse:

  • how is the hacking of other people going?

Patrick:

  • Colin has a commit bit and has already done a couple of commits
  • there are other people doing documentation and checking out other features
  • seems to be going well

Larry:

  • mostly re-rationalizing the various tradeoffs in S05
  • still ramifications of the decision to reserve all punctuation
  • that changes the desirability of using puncutation in various ways

Patrick:

  • it removes one of the constraints
  • frees us to do things in nicer ways

Larry:

  • we have to decide whether they're actually nicer or just look nicer
  • may or may not end up with a shorter way of representing the whitespace rule
  • don't have to put .ws all over the place
  • currently under discussion
  • there are n different ways to do it
  • on PerlMonks, someone pointed out that there's no good way of having a temporary binding that doesn't end up in the final match
  • you could use it for repetitions
  • perhaps a rule that begins with an underscore might end up not getting matchified
  • if the whitespace rule ended up being underscore, it might not capture
  • other than that, work's been fairly hairy
  • still thinking belatedly about the questions for an interview

Allison:

  • created a branch for new OO in Parrot
  • lion-taming to try to get it to compile
  • ran into one spot, getting PGE to compile
  • had to fix the subclass opcode
  • currently looking at a problem in GetOpt::Object
  • can't find a method in that oddly-defined class
  • just need to standardize its use of OO
  • also spent some time creating a Parrot timeline
  • want to make it public after we meet a few deadlines
  • it's given me more concrete ideas of what I work on every weeek

Patrick:

  • we should have five or six done by the end of October

Allison:

  • basically one deadline every two weeks
  • lots of them are parallelizable

c:

  • found some bugs
  • fixed some bugs
  • having lots of trouble with platform-specific things
  • mostly the unavailability of certain platforms I don't have access to
  • if we had a smoke farm or something that we could get results from various platforms, we could get more results

Nicholas:

  • sounds like a reasonably good cage-cleaner or newbie task
  • someone popped up with a relatively unused IA-64 machine and offered to help with bleadperl

Jesse:

  • working with Richard Dice to get the microgrants paid
  • Adriano Ferreira has put in a proposal for a new grant
  • we're talking about that

Patrick:

  • feels like we ought to use lexicals for backrefs without captures
  • whenever I've read S05, that seemed to be the natural way to capture something into a variable that doesn't end up into a match object

Larry:

  • that's one way to work it
  • seems like the default is backwards though
  • usually the reason you bind something is because you already have the name somewhere, and you want to capture it

c:

  • not sure I agree with that
  • naming a thing gives you an abstraction, but doesn't necessarily imply that you want to use that abstraction elsewhere
  • that's why we have lexicals and not just dynamic scope everywhere

Larry:

  • almost wish I had a twigil that was shorter than greater-than/less-than

Patrick:

  • maybe add it to everything else you're juggling at the moment
  • seems straightforward and natural to me though
The Fine Print: The following comments are owned by whoever posted them. We are not responsible for them in any way.
 Full
 Abbreviated
 Hidden
More | Login | Reply
Loading... please wait.