All the Perl that's Practical to Extract and Report

Perl 6 Design Minutes for 11 October 2006

davorg on 2006-10-13T15:34:00

The Perl 6 design team met by phone on 11 October 2006. The attendees were Larry, Damian, Allison, Jesse, Nicholas, and chromatic. These are the minutes.

Damian:

  • some response to the POD spec
  • mainly the silence was a good response
  • either it's totally wrong or totally right
  • don't think it matters, as long as they're happy
  • the major response was a private one
  • "I wish there were full semantic markup"
  • all of the upper-case stuff is semantic though

Larry:

  • I is for Important

Damian:

  • mostly otherwise just nits on the spelling in other cases
  • some have the desire for an entirely different markup language
  • I've referred them to kwid for that
  • a couple of issues have arisen
  • how can you collect a lot of =CONFIGs into a stylesheet to include them?
  • I assume that there can be an =USE or similar construct
  • will be releasing an alpha of the parser probably next week
  • probably to the CPAN
  • need a namespace for it, maybe initially Perl6::

Allison:

  • worked on the threads PDD
  • pulled in the documentation Nicholas found
  • haven't checked it in yet, but might do it later today
  • expect to have some discussion on it after this

Larry:

  • only redesigning the basic syntax of the language
  • other than that...
  • substitution operators and hyperoperators are now different
  • several other various design things
  • working to help get the next Pugs release out
  • think my current smoke test will be clean
  • we'll probably see a release this weekend

Jesse:

  • random point release?

Larry:

  • not a major digit release
  • but it is getting more usable
  • I'm using a Perl 6 program in my work
  • I upgraded to GHC 6.6 and my program ran 60 times faster
  • can't share the code, but I wish I could

c:

  • worked on an experimental new object library for Parrot
  • noticed a fair bit of unimplemented stuff in the PDDs
  • seems like a good task for the CAGE cleaners
  • also thought you could embed Perl 5 in Parrot

Larry:

  • and get continuations with the runloop

c:

  • and control the runloop from Parrot... hmm...

Damian:

  • with the new POD format, how can you do stylesheet-style things?
  • how can I pull in many =CONFIG blocks and have them affect?
  • do we just use =USE?
  • how does it know that you're pulling in some Perl 6 code to change how the parser works
  • or how you're telling it to pull up some POD to change how the POD works?

Jesse:

  • I assume it would load a piece of code and POD that affects the current unit

Damian:

  • if there's POD in the file being used, that becomes part of the POD
  • if there's code, that code gets passed on to the parser
  • that makes it impossible to document your modules that add behavior

Larry:

  • you probably want some way of importing that
  • might not want to be the default
  • might want to import or not import various things
  • import some streams of POD and not others

Jesse:

  • it'd be neat to use the POD from within code

c:

  • it'd be nice to pull in bits of documentation in class hierarchies

Damian:

  • we definitely want to flatten documentation
  • competing things
  • who is the POD for?
  • is it documentation for the current code?
  • does it supply documentation to something else?
  • sounds like we need at least one more directive to handle this somewhere

Larry:

  • why wouldn't it be more options on =USE?
  • we're shoving a lot more through that mechanism in Perl 6
  • seems to be holding up

Damian:

  • there are two different kinds of things
  • one is a collection of code and possibly documentation of that code
  • that should change how the parser works
  • the second is a collection of effectively just POD
  • includes that POD into the current POD

c:

  • parse and execute versus just insert

Damian:

  • having to specify between those two every time seems difficult

Jesse:

  • does Perl 6 still have require?
  • feels very use versus require

Damian:

  • don't think it does

Jesse:

  • it's just a naïve opinion, based on what I've heard

Larry:

  • Perl has never really had a textual include

Damian:

  • POD already has it for Perl 6
  • we don't have a block form of that, effectively
  • I'll play with it a little bit
  • I'll try to come up with something

Larry:

  • I think the =USE should be the active stuff
  • that'll be the expectation from the Perl people
  • I noticed that you had the ability to change the encoding on the fly
  • the encoding of the source text ought to be unified with the Perl 6 parsing
  • there shouldn't be two different directives: POD is in Big-5, Perl is something else
  • maybe you should use the POD directive to change the encoding of the Perl code too

Damian:

  • sounds the right way around
  • I'll change the description of the encoding

POD Spec?

Theory on 2006-10-12T16:38:17

Where can I read the POD spec? I've spent enough time with POD, Textile, Markdown and the rest to find them all equally annoying for various reasons. I'd love to see Perl 6 (and 5.10?) get it right.

—Theory

Re:POD Spec?

chromatic on 2006-10-12T18:44:22

Damian posted his Synopsis 26 Draft from 07 October 2006 to p6l last week. I expect it to make it to the Perl 6 dev site soon.

Re:POD Spec?

audreyt on 2006-10-13T02:10:39

Currently it's checked in at pugs/docs/Perl6/Spec/Documentation.pod, but it may relocate to svn.perl.org in the future.

Re:POD Spec?

Theory on 2006-10-23T01:23:55

That page comes up empty for me. :-(

—Theory

Re:POD Spec?

audreyt on 2006-10-23T01:47:00

It's down for the weekend; http://svn.perl.org/perl6/pugs/trunk/docs/Perl6/Spec/Documentation.pod is a copy. We're moving the master repo to perlcabal.org over the weekend so it will be more stable soon...