Perl 6 - the future is here, just unevenly distributed

IRC log for #parrotsketch, 2008-07-08

| Channels | #parrotsketch index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
00:48 contingencyplan joined #parrotsketch
00:51 contingencyplan joined #parrotsketch
07:04 contingencyplan joined #parrotsketch
09:14 cognominal joined #parrotsketch
13:26 pmichaud joined #parrotsketch
13:39 rdice joined #parrotsketch
15:41 tewk joined #parrotsketch
16:17 tewk joined #parrotsketch
16:29 Whiteknight joined #parrotsketch
16:40 paco joined #parrotsketch
17:22 shamu joined #parrotsketch
18:12 barney joined #parrotsketch
18:12 particle joined #parrotsketch
18:13 Auzon joined #parrotsketch
18:14 jhorwitz joined #parrotsketch
18:24 NotFound joined #parrotsketch
18:25 allison joined #parrotsketch
18:26 cotto_work joined #parrotsketch
18:26 magnachef_away joined #parrotsketch
18:28 jonathan joined #parrotsketch
18:28 rgrjr joined #parrotsketch
18:30 cotto_work joined #parrotsketch
18:30 chromatic joined #parrotsketch
18:30 * moritz greets the assembled parrot painters - er, sketchers ;-)
18:31 chromatic ROLL CALL
18:31 kj joined #parrotsketch
18:31 jonathan HERE!
18:31 pmichaud bagel!
18:31 * barney is here
18:31 kj here
18:32 allison here
18:32 chromatic Alphabetical order?  You bet.  allison?
18:32 Whiteknight here
18:32 allison I spent the week travelling.
18:32 allison Met with Jim Keenan and NY Perl/Parroteers
18:33 * jhorwitz is here
18:33 allison Worked on pdd25cx branch, but haven't yet committed context/scoped handlers fix.
18:33 allison EOR
18:33 chromatic barney?
18:34 barney Renamed Plumhead to Pipp.
18:34 barney Tidied languages/json.
18:34 barney Added make-target 'pipp'.
18:34 barney Removed 'pipp.pl'.
18:34 barney Use optok-parsing in Pipp.
18:34 barney Run the tests in 'languages/pipp'.
18:34 barney Start on OO-Support of PHP.
18:34 barney file one q
18:34 barney .eor
18:35 chromatic I fixed some bugs.
18:35 chromatic I closed some tickets.
18:35 chromatic I'm supervising Whiteknight's work, but mostly that's just giving some advice and approving his mid-term report.
18:35 chromatic I'm happy to look at any new bugs anyone files, if they can reproduce them with a little PIR.
18:35 chromatic I hope to have more time in the next week.  We'll see.  Probably not until next Monday.
18:35 chromatic cotto_work, are you here?
18:36 cotto_work yes
18:36 cotto_work no report, though
18:36 chromatic japhb?
18:37 chromatic jhorwitz?
18:37 jhorwitz got mod_perl6 working again
18:37 jhorwitz worked on new rakudo patches for mod_perl6 to keep up with jonathan's latest changes
18:37 jhorwitz would liek to commit PERL6LIB implementation for 'require' today
18:37 jhorwitz also would like some review on partial ::() implementation before committing (RT #56700)
18:37 jhorwitz EOR
18:37 chromatic jonathan?
18:38 jonathan * Last week, time to do much at all was non-existent.
18:38 jonathan * This week, things are better and I'm making up last week's Rakudo day
18:38 jonathan * First Rakudo day of the week is today
18:38 jonathan ** Working on signature objects
18:38 jonathan ** We're now building a basic signature object, but lots of details missing
18:38 jonathan ** Got the grammar a little closer to STD.pm
18:38 jonathan ** my ($a, $b); now works, but only that case! Lots more work that I'm doing here at the moment.
18:38 jonathan * Will continue getting the type info into signatures, in prep for getting Perl 6 MMD in place later in the month
18:38 jonathan * Also want to get enums implemented - they're also on my hit-list for the near future.
18:38 jonathan EOR
18:38 chromatic kj, are you here?
18:38 kj tried to debug PIRC's heredoc preprocessor. Found the weak spot, and fixed it, but the solutions seems very hacky
18:39 kj and (just) did some updates to languages/PIR's action methods.
18:39 kj .eor
18:39 chromatic moritz?
18:39 moritz my $report = q:pseudo_yaml[ - Added 18 test files to Rakudo's spectest_regression target - applied some small patches, resolved tickets - Auzon++ for tests, and bacek++ for both tests and patches - filled out GSOC mid term evaluation for Auzon - Test progress is now plotted on http://rakudo.de/ and regularly updated
18:39 moritz ];
18:39 moritz where did those line breaks go? ;-)
18:39 chromatic It's not always easy to write YAML by hand.
18:39 moritz ;)
18:40 chromatic NotFound?
18:40 NotFound Almost nothing, looking at some details.
18:40 chromatic particle?
18:41 particle reviewed tewk's gsoc progress
18:41 particle submitted his mid-term eval
18:41 particle still having trouble getting my main dev env working, due to hw problems
18:41 particle switching to linux for now
18:41 particle .end
18:42 chromatic pmichaud?
18:42 pmichaud ** Rakudo spectest_regression:  93 files, 1672 passing tests  (+14, +432 from 07-01)
18:42 pmichaud == Overall
18:42 pmichaud == Parrot stuff
18:42 pmichaud : chromatic++ finished the lexicals work -- now things work properly
18:42 pmichaud : investigated some .HLL stuff for P6object -- have some blockers there
18:42 pmichaud == PCT stuff
18:42 pmichaud : worked on updating P6object to understand HLL
18:42 pmichaud : eliminated use of implicit newclosure from PCT
18:43 pmichaud : updated PCT to use :lexid
18:43 pmichaud : allow :name attribute to be a PAST tree
18:43 pmichaud : fix arity handling on 'for' loops
18:43 pmichaud == NQP stuff
18:43 pmichaud == Rakudo stuff
18:43 pmichaud : refactor builtins into src/builtins/any-*.pir
18:43 pmichaud : refactored lexical handling, especially $_, $!, $/
18:43 pmichaud : fixed 'given' statement
18:43 pmichaud : refactored placeholder variables
18:43 pmichaud : refactor .?, .*, .+
18:43 pmichaud : eliminated a lot of inline PIR
18:43 pmichaud : applied patches (bacek++, masak++)
18:43 pmichaud : fixed some bugs (cjfields++, masak++, moritz++)
18:43 pmichaud == Blocker status
18:43 pmichaud : Rakudo .HLL mapping is blocking on ability to create classes in other HLL namespaces (RT #56616 and others)
18:43 pmichaud : gather/take is blocking on resumable exceptions (pdd25cx)
18:43 pmichaud : using pre-compiled Perl 6 modules needs some design work, and likely some fixes to :load handling and lexicals (RT #47956)
18:43 pmichaud : list assignment is blocking on some design issues, and possibly PGE improvements
18:43 pmichaud : Thanks again to chromatic++ and jonathan++ for removing blockers in lexical variable handling
18:43 pmichaud Queue 1 question
18:43 pmichaud .end
18:43 chromatic rgrjr, are you here?
18:44 rgrjr yup; not much to say though.
18:44 chromatic tewk?  particle says you owe us some status updates.
18:44 rgrjr * working on test case for continuation/eh interaction in pdd25cx
18:45 rgrjr - found a regression(?) in trunk, but no fix yet.
18:45 rgrjr one Q (or 2)
18:45 rgrjr .eor
18:48 chromatic Whiteknight?
18:48 Whiteknight * Lots of GC-related refactoring and bugfixing
18:48 Whiteknight * Stuck on a nasty segfault
18:48 Whiteknight EOR
18:49 chromatic Alright, we have two queued questions.
18:49 chromatic barney?
18:50 barney Will next weeks release be 0.6.4 ? Or will pdd25cx be merged ?
18:51 * particle votes merge + todo/skip before release
18:52 pmichaud I think merge depends on allison+chromatic+other schedule though, and allison is likely fairly busy this week and next
18:52 chromatic Depends on how bad the breakage is.
18:53 chromatic If we can't get it merged a week before the release, I'd feel nervous about merging before the release.
18:53 pmichaud agreed
18:53 particle meetoo
18:53 chromatic As it is, we don't know if any language which uses PGE will run on the branch right now without modifications.
18:53 barney me too
18:54 pmichaud I can throw tuits at any PGE issues, no problem.  But if there are core issues I'm not likely to be as much help there.
18:54 rgrjr fwiw, pdd25cx breaks Kea-CL (IIRC), so I would rather not merge.  (But I don't think that should stop anyone.)
18:54 chromatic rgrjr, do you use PGE there?
18:54 rgrjr chromatic: no.
18:54 chromatic Is it the continuation promotion problem you reported earlier?
18:54 allison Without substantial language testing, we shouldn't merge the branch before the release.
18:55 pmichaud are the "PGE issues" really within PGE, or is it just that PGE is the first major PIR/object component in the Parrot build?
18:55 rgrjr chromatic: sorry; I'll have to rerun the tests and get back to you.
18:55 particle so, we need allison's patch, and to have the branch brought up to trunk head
18:55 allison pmichaud: the parrot build is successful now, including PGE
18:55 particle then we can request heavy hll testing
18:56 allison pmichaud: it's just that PGE is 2 of the remaining 4 failing test files
18:56 pmichaud okay.
18:56 pmichaud I'll do those immediately after #ps.
18:56 allison my patch may fix a number of them
18:57 allison I'll try to commit that right after parrotsketch, but I'll be in meetings the rest of the afternoon
18:57 pmichaud no problem.  if I get stuck I'll just send an "I'm stuck" exception your way :-)
18:57 chromatic If you can't commit it, can you send it my way so that I can commit it?
18:57 barney I'm taking a trip to the mountains till Sunday. So I can prepare the release starting on monday.
18:58 allison chromatic: not sure how that would help, as I need to finish the code. I mean, I can commit what I have now, but partially completed it'll make things worse, not better.
18:59 chromatic Completely uncommitted, no one can help you.
18:59 japhb branch the branch?
18:59 particle fail early, fail often
19:00 Whiteknight ...or in my case, fail early, fail every time apparently :(
19:00 allison it'll take me as long to explain what needs to be done as it'll take me to do it
19:00 pmichaud note that rakudo counts only its successes, not its failures :-)
19:00 allison I don't have anything scheduled tonight, so I'll work on it then. It should only be a couple hours of work left.
19:01 chromatic Alright then, let's move on to pmichaud's question.
19:02 pmichaud my question was also about the release.  But I'm also looking for help with #56616 if anyone is interested.
19:02 pmichaud essentially I'm not sure how to create classes in other hll namespaces
19:02 pmichaud at least, whatever I try seems to break.
19:03 chromatic The method form seems reasonable to me.
19:03 pmichaud oh, yes, there's that also.
19:04 pmichaud right now the problem is that I can create classes using namespaces from other HLLs, and that works, but then Parrot MMD doesn't work properly
19:04 pmichaud (and 'isa' gives some strange results)
19:04 pmichaud oops, I reported the wrong ticket
19:05 pmichaud #56650
19:05 pmichaud *that's* the real blocker  (sorry for the misdirection)
19:06 chromatic Alright, anyone want to look into that?
19:06 Tene I might look into it.
19:06 pmichaud if #56616 is okay with others, I'll go ahead and apply it in trunk (I have it in a branch right now)
19:08 allison yes, I'm fine with making them methods
19:08 pmichaud done, thanks.
19:08 Tene I have no idea what's happening after work for me tonight.  If nobody else has dealt with 56650 by then, I'll try.
19:08 pmichaud EOQ from me, Tene++ (or anyone++) if they can resolve 56650 for me.  :-)
19:08 chromatic Are there any other questions or blockers?
19:08 pmichaud or give me another way to create classes in other hll namespaces
19:08 pmichaud rgrjr had a q
19:08 rgrjr Turns out I have two
19:09 chromatic Go ahead.
19:09 rgrjr Let me ask them together:
19:09 rgrjr 1.  The Boston Linux & Unix group asked Uri for a Perl 6 talk on 20, and he posted on boston.pm.  Anyone going to be in the area?
19:09 rgrjr 2.  I don't seem to have privs to change things on RT.  Am I looking in the wrong place?
19:09 rgrjr go!
19:10 chromatic #2, you have to log in, and if you've logged in from an existing ticket, you have to go to the logged-in-ticket view instead.
19:10 rgrjr 1.  "on Aug 20"
19:10 chromatic Confuses me every time.
19:10 rgrjr "logged in ticket view"?
19:11 particle rgrjr has bugadmin rights in rt
19:11 allison rgrjr: it's a different URL than the "public" view
19:11 moritz rgrjr: if it has /Public in the URL, it's the wrong one
19:11 allison someone cleverly just added a "You are currently viewing the Public interface." warning in the past week
19:12 allison now every public ticket page has a link to the full page
19:12 rgrjr aha! somehow I had got un-logged-in.  Seems to work now.  Thanks!
19:12 rgrjr allison:  Cool, a hint! ;-}
19:12 moritz rt's login cookie is only valid as long as your browser is open (at least for me)
19:12 japhb Impatience applies here ... RT should switch for you ...
19:13 tewk * ncigen is working
19:13 tewk * converting src/call_list.txt into tests for dynamic nci stub generation
19:13 tewk * current plan is to revive and extend Parrot_jit_build_call_func
19:13 tewk EOR
19:13 chromatic tewk, any estimate on making a public commit to a branch somewhere?
19:14 tewk I'll do it right now,
19:14 tewk http://tewk.com/nci.diff is always current.
19:15 chromatic Easier to test and review a branch (and incremental changes) than a single patch.
19:16 * particle just added julian albo (notfound) as a new committer
19:16 chromatic ALL HAIL THE COMPUTER
19:16 moritz NotFound++, welcome
19:16 NotFound Thanks to all.
19:20 chromatic Is there anything else?  It sounds like we've finished for the week.
19:20 japhb tewk: Is that call_list that you're testing with including the OpenGL stuff?
19:21 japhb (i.e. did you generate it on a system with OpenGL/GLUT installed? That's a lot of additional stubs ....)
19:23 tewk japhb: no but I can do that.
19:24 japhb tewk: remember that those signatures paper over most pointers by just assuming they will be handled with the {Un,}ManagedStruct interface ....
19:25 japhb Can your code generate the structure definitions needed by ManagedStruct?
19:25 * japhb has been doing them by hand as needed ....
19:26 tewk theoretically yes. I parse the header file and have an ast, so all I need to do is write a generator for ManagedSturct.
19:27 tewk japhb: do you have a pointer to your hand generated code?
19:27 japhb OK, great.  It sounds like you're getting close to the point where I need to pay more attention.  :-)
19:27 japhb Meta: Should we take this back to #parrot?
19:27 tewk ncigen just generates dlfunc pir calls for c functions, but ManagedStructs shouldn't be that hard.
19:28 tewk I already parse and build ast's for structs.
19:28 tewk yes back to #parrot
19:28 japhb excellent
19:30 chromatic Until next week everyone.
19:30 moritz ciao
19:30 jonathan bye bye
19:30 jonathan left #parrotsketch
19:31 jhorwitz left #parrotsketch
19:31 allison bye
19:32 cotto_work left #parrotsketch
19:32 chromatic left #parrotsketch
19:32 shamu left #parrotsketch
19:37 NotFound left #parrotsketch
19:49 Auzon left #parrotsketch
22:48 pmichaud left #parrotsketch
22:57 japhb joined #parrotsketch
22:59 Whiteknight left #parrotsketch

| Channels | #parrotsketch index | Today | | Search | Google Search | Plain-Text | summary