weekly meeting at 18:30 UTC, Tuesday
Set by moderator on 28 August 2008.
01:26 davidfetter joined 12:57 kid51 joined
kid51 Pre-reporting ... 12:58
* Cleanup of errors in configuration system and tests reported in the wake of 12:59
consolidation of t/steps/ tests into fewer files.
* Elimination of import of init::defaults from approx. 24 t/steps/ tests.
* Merge branch implementing file-based configuration.
* Some consolidation of t/configure tests into fewer files.
* No large projects pending.
EOR 13:00
13:58 clinton joined 14:07 clinton left 15:05 davidfetter joined 16:54 cotto_w0rk joined 17:17 pmichaud joined
pmichaud I will miss #parrotsketch today 17:17
little to report for this week -- mainly did some thinking about some rakudo refactors and looking at rt tickets 17:18
tewk window 11 17:49
18:17 allison joined 18:23 NotFound joined
moritz hello 18:31
Tene RAWR!
cotto_work o hai
allison ni hao 18:33
NotFound H
Tene So, alphabetical! 18:34
Allison, why don't you go first?
allison - Spent half of last week wrestling with the guts of Parrot's dispatch system and won. The new unified MMD system is now fully functional. 18:35
- (Based on the experience, started on a task list for cleaning up Parrot's dispatch system.)
- Introduced a new way of invoking PIR/NCI subroutines from C: 'Parrot_pcc_invoke_sub_from_sig_object' that takes a subroutine object and a CallSignature PMC object, and performs a PCC-style invocation.
- I'm working on updating all MMD calls to the new syntax (have also delegated tasks for parts of this).
- The pdd27mmd branch will be ready for initial language testing tomorrow.
EOR
Tene cotto_work: you next
cotto_work * rewrote several t/pmc/foo.t tests in pure PIR
* got rid of several dozen unneeded invocations of Parrot during make test 18:36
* hoped to speed up make test, but it's actually slower now
* will look into this to make sure it's not my fault
end of line
Tene Not sure who's around. Moritz next?
moritz just the usual testing and applying patches
PIR level coverage testing would ROCK 18:37
eor
Tene particle?
pmichaud isn't here.
tewk?
Me!
* Many new tests for cardinal from Damian Knopp
* Many newly passing tests in cardinal
* Copied rakudo's 'use' for cardinal's 'require'
* Precompiled libraries don't work in cardinal due to a weird issue with :outer (rt #58506)
Did I miss anyone?
NotFound Me 18:38
cotto_work q 1 q
Tene Oh, hi!
moritz pmichaud and tewk posted earlier
Tene orite.
NotFound: report!
NotFound More work on parrot debugger
Some fixes, closing some tickets 18:39
EOR
Tene Anyone else?
18:39 rurban joined
Tene Hi, rurban! you have a report for us? 18:39
'kay, question time now. cotto.
rurban Just finished the last logical errors in the conditionals.
found more make install -c languages bugs and features to add. 18:40
added the advice from Allison to the pdd30
eof
Tene prods cotto_work. 18:41
cotto_work k
I'd like to move the macros generated by pasm_fp_equality_macro into an include file 18:42
what'd be the best place for it?
(lib/Parrot/Test.pm:1125)
particle runtime/parrot/include/fp_equality.pasm 18:43
cotto_work thanks
eoq
Tene Any other questions? 18:44
Looks like no.
Thanks for coming, everyone! 18:45
Same time next week.
18:45 chromatic joined
rurban I have a minor one: just for forth 18:45
Tene Hi chromatic! Do you have a report? 18:46
rurban where to put languages specific pbcs when installed?
WMLScript, forth, perl6, tcl and such
chromatic I'm still working on being able to remove PARROT_API from vtable entries. 18:47
I ran into MMD problems, so the ultimate fix may have to wait until after the MMD merge.
I'll try it for only dynpmcs right now (as that's the real problem), but I'm keeping an open mind.
Otherwise, we need to get our open bugs count way down again.
EOR
Tene rurban: I think I saw someone mention /usr/lib/parrot/languages/
rurban hmm. I dont like that too much. the searchpath in library.c needs to be changed
I propose lib/parrot/library/<language>/some.pbc 18:48
cotto_work chromatic, do you have a target in mind?
particle runtime/<language>/library/ was the intended dir some time ago
rurban if the pbc should not got the root namespace, such as perl6 or tcllib.pbc
particle as to where they're installed on posix systems, i have no idea 18:49
chromatic 850 would be nice before the next release.
800 would be better.
rurban pipp also has a lot of such pbc's to install somewhere 18:51
Tene On rt.perl.org, it looks like there are 728 open tickets in the parrot and perl6 queues?
724, I mean. 18:52
moritz the count in the perl6 queue isn't going down any time soon, I fear
quite many of them are longer term TODO items
chromatic Er, yeah. Sorry. Ticket Counts: 37 new + 657 open = 694
I mean 650 and 600 for that final number.
Tene ah, new doesn't count in open.
rurban forth wants e.g. "load_bytecode" couldn't find file 'languages/forth/tokenstream.pbc'. So lib/parrot/library/languages/forth/ could also make some sense (would work out of the box), but library/language/forth/ can be simplified to library/forth/. 18:53
21:25 rurban_ joined 21:26 rurban_ left 21:58 chromatic left 22:12 cotto_work left 23:46 NotFound left