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 April 03, 2009
06:05 PM

Perl 6 Design Minutes for 26 March 2009

[ #38748 ]

The Perl 6 design team met by phone on 26 March 2009. Larry, Patrick, Jerry, Nicholas, and chromatic attended.

Larry:

  • writing talks for my upcoming trip
  • leaving for NY and then Boston tomorrow morning
  • talking at Bloomberg, Columbia, Harvard, MIT, and Yale

Patrick:

  • picking up any degrees?

Larry:

  • probably Farenheit
  • speaking of heat, jumped in on the logo discussion
  • installed Camelia
  • the discussion is dying down now
  • worked on the sequential output of the regexp matches with .caps and .chunks methods
  • they now come in key/value pairs
  • they key of each match was the name it was bound to
  • just a list of pairs, so there can be duplicates
  • speculated about where we want to go about type narrowness
  • Jonathan had a thundering apoplexy
  • made that a post-6.0 decision
  • we'll allow only one type constraint parameter for 6.0 to allow us to extend the syntax afterwards
  • to allow multiple constraints, define a named subset type (for now)
  • removed prefix int from the standard grammar
  • most of my actual hacking comes from the desire to rewrite gimme5 in viv
  • lots of cleanup
  • fleshing out a lot of spots where things didn't get attached quite right in the match tree or AST or caps
  • good for finding lots of little bugs
  • the text in the nibbler now comes back as a match object, not just a string
  • easier to call methods on when iterating through caps list
  • nouns and values are now syntax categories
  • derived grammars can more easily add new nouns and values
  • added the first implementation of ww rule
  • still buggy and chews up all of my memory
  • not used in STD yet

Jerry:

  • student application period is open for GSoC
  • closes on Friday, April 3
  • we're looking for students and mentors

Patrick:

  • enjoyed nice weather on the west coast
  • spent several good days at Disneyland
  • caught up on email otherwise
  • released Rakudo Developers Release #15, codename "Oslo"
  • that went smoothly; it's very easy to do releases
  • will extract that into a release managers guide, like Parrot has
  • will delegate releases to other people over the next few months
  • reviewing and commenting on patches from other people
  • mainly preparing for major PGE and PCT refactors
  • will bring them up to date with the Synopses changes
  • that should progress quite a ways on my Hague grant

Nicholas:

  • Rafael's making great progress on smartmatching for Perl 5.10.1

Patrick:

  • if you have a method declared without a parameter list, does it get @_ like a sub, or no parameters?

Larry:

  • I've been thinking it comes in the same way Perl 5 does it
  • hadn't bothered to try to think about it the other way

Patrick:

  • Rakudo assumes @_

Larry:

  • but it leaves out the invocant
  • that's the difference
  • I haven't decided
  • it's further from what a Perl 5 programmer might expect
  • but it might be more useful

Patrick:

  • we have three interpretations
  • the Settings and Synopsis 32 show people treating no parameter list as "no parameters"

Larry:

  • there is an invocant and no additional parameters
  • () is that

Patrick:

  • they treat a missing parameter list as that

Larry:

  • I think that's wrong

Patrick:

  • Rakudo's interpretation is that you get @_ which does not include the invocant
  • the other interpretation is that you get @_ which does include an invocant
  • I need to add parens to those documents then

Larry:

  • probably the correct answer is that it has to include the invocant
  • we've said that the invocant is always just the first parameter
  • also the interchangeability of sub and method bodies, so you can alias a method to a particular multimethod

Patrick:

  • if you have a block with @_ and $_, does the former include the latter which was the first argument to the block?

Larry:

  • they're both implicit?

Patrick:

  • neither explicitly given as parameters, but are mentioned (implicitly or not) in the block
  • does @_ include a $_ parameter?

Larry:

  • and how would you write that signature?

Jerry:

  • \_? yuck

Patrick:

  • I don't need an immediate answer on that
  • I'll turn Moritz's ticket into a spec bug (not specified), not a Rakudo bug
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.