The Perl 6 Design Team met on 16 May 2007 by phone. Larry, Damian, Allison, Patrick, Jesse, Nicholas, and chromatic attended. These are the minutes.
Allison:
- resting up and feel better
- working on the PMC PDD
- checked in the first draft, but it still needs some work
- spent the weekend reviewing patches and resolving bugs
Jesse:
- anything standing in your way?
Allison:
- a handful of C developers would be nice
c:
- released Parrot 0.4.12
- didn't get as much coding done as I wanted
- it's a solid release though
- right now I'm cleaning up some code files
- will write up a getting started guide soon though
- intended to help people interested in any language to start hacking on Parrot or write languages
- send people with questions my way
Larry:
- last week's What's Wrong Perl 6 meditation on PerlMonks solicited information
- seemed well-intentioned
- it did provide a place for people to link to if there are good responses
- I tried to respond to the honest questions and ignore the disingenuous ones
- more clarifications as requested
- seems like I've been answering a lot of questions on IRC about what semantics are explained where
- mostly new people trying to come up to speed
- we'll have a bunch more once Pugs 6.28 is out and we start getting a compiler bootstrapped in Perl 6
- mostly now just doing research and thinking about my talks
Damian:
- still occupied with family issues
Patrick:
- checked in a draft of the abc grammar
- have some comments and feedback on that
- seems to have universal appeal
- studied Larry's STD grammar in more detail
- expect to have a working implementation of abc using those details by this time next week
- also should have a fairly significant start of the Perl 6 grammar too
- turns out to be easier than I anticipated
- it's exciting
- the embedded actions can become a token in PGE
- it can execute those actions at runtime
- you have a ParseToAST object you pass to the grammar
- it'll fire off those actions at runtime
- that makes it pluggable
- should be straightforward enough that other people can contribute patches
- especially with a standard grammar that people can work from
- I'll plug rules from that into the existing grammar
- goal is to use that grammar outright for everything
- I may propose patches to the standard grammar too
Larry:
Patrick:
- my biggest work at the moment is looking at the bottom-up parser
- figuring out how to do something similar in PGE
- PGE's parser might be somewhat different
- a lot of languages won't need what Perl 6 does
- I'll keep a simplified version around for abc, mini Perl 6, etc
- it should look pretty good
- expect to have something workable by this time next week
Larry:
- there are certainly bugs in what I've written
Patrick:
- I've found and ignored a few
Larry:
- there may even be egregious design errors
Patrick:
- I'll bring them up
- having the grammar in full is helpful
- seeing contextual variables is a ton of help
Damian:
- that's excellent and very exciting
Jesse:
- now that you have the release out c, are you still looking at Patrick's stuff?
c:
- after I get the "getting started" guide done, I'll look at PGE and parsing and PAST
- need to port Pheme to HLLCompiler, for example
- intended as better tutorial documentation
Patrick:
- I'm glad to see that
- lots of people could use it
Jesse:
- the two Perl 6 microgrants seem to be making progress
- even if they've been slow the past weeks
- we'll try to choose smaller grants next time
- maybe more visible progress
c:
- I like the work Steve Peters has done
- it's led to some cleanups outside of his patches
- I use it as examples for some of my cleanups
Larry:
- with Perl 5 state variables, if you clone a closure do you get a clone of the state variable as well?
Nicholas:
Larry:
- the correct behavior is that every time you create a clone of the closure, you get a new initialized state variable
- if you treat it like a C static, you get the wrong behavior
Nicholas:
- I think it's like a
my variable in the lexical pad
- there was something weird with
pp_assign if the rvalue is a function for example
Larry:
- I just want to make sure there aren't a bunch of broken Perl 5 scripts
Nicholas:
