Priorities for this week: Pre-post/plan PDS topics & attend PDS; Damn the build, full Awesome ahead! | Post closed tickets in your report. | Note: This channel is for our weekly status meetings (Tuesdays at 19:30 UTC); you probably want #parrot instead. | irclog: irclog.perlgeek.de/
Set by moderator on 26 July 2011.
00:02 whiteknight left 01:50 kid51 joined 02:49 kid51 left 07:42 contingencyplan joined 08:13 dafrito joined 08:16 lucian joined 09:58 contingencyplan left 10:07 lucian left 10:24 lucian joined 11:10 darbelo left 12:32 Coke left, Coke joined 12:54 darbelo joined 13:00 darbelo left 13:21 darbelo joined 13:27 lucian_ joined 13:33 lucian left 16:08 lucian_ left 16:10 spinclad_ left 16:28 darbelo left 16:44 lucian joined 17:12 darbelo joined 18:00 kid51 joined
kid51 kid51's report 18:00
* DONE 18:01
** Will miss first 30 min of #ps today.
** Participated in online PDS Sat July 30. Posted summary with new roadmap goals.
** Favorable test results enabled closing of TT #2096.
** Overall count (Trac report #10): 511 (it was 510 one week ago).
* WILL BE DOING
** Planning Parrot Foundation annual meeting and election. 18:02
** Hope to find some time to think about the benchmarking goal.
** Taking additional look at docs/project/release_manager_guide.pod.
* EOR
18:02 kid51 left
lucian Report 18:04
Done:
- fixed a horrible object system __getattribute__ bug
- minor cleanups
- some incremental improvements in the compiler
Todo:
- make all object system tests pass again
- refactor compiler to handle scopes better (with a general Codegen class) 18:05
EOR
18:16 lucian left 18:17 NotFound joined 18:18 lucian joined 18:24 darbelo left 18:25 darbelo joined 18:28 lucian left 18:31 contingencyplan joined 18:32 lucian joined 18:42 lucian left
NotFound What I did: 18:48
-parrot
* Fixed a test and added a missing vtable pod doc.
* Worried about NCI and string encodings.
-winxed
* Added __CLASS__ and __NAMESPACE__ predefined constants.
* Resurrected winxed.org website after some problems with my hosting.
Thanks to my friend Luis Villaverde for his free hosting, BTW.
What I will do:
* Keep looking for NCI and string encodings.
EOR
18:50 lucian joined
Util # Done: Attended Parrot Developer Summit. 18:56
# Plan to do: Nothing ($WORK).
.end
19:14 benabik joined, atrodo joined, whiteknight joined 19:15 bubaflub joined
whiteknight WHAT I DID 19:19
* Moved. Huge pain in the butt. Still have more to do. Have very little free time and shoddy internet access
* Couldn't be at PDS. Backscrolled. Seemed like a good, productive meeting.
* Trying to help soh_cah_toa with his debugger. Did some testing, need to meet with him more to start moving forward. Looked at IMCC line numbers and other horrors.
* Lots of fixes to Rosella. I had pushed a lot of disruptive changes to master instead of a branch, and got lots of bug reports.
* Some parrot branchwork. Added a cache for load_bytecode_p_s, and some other fixes. Worked more on whiteknight/imcc_tag branch, and got some feedback from several people. Wrote a few quick tests for the new syntax, need more. 19:20
WHAT I WILL DO
* Help soh_cah_toa with debugger. Look more at IMCC line numbers
* Preparations for 6model work with NotFound++ and other interested volunteers
* Some Rosella work.
EOR
tcurtis DID:
* Worked on DPDA interpreter.
* Committed stuff.
* Tests.
* Start on epsilon-transition elimination.
WILL DO:
* Blog.
* Fix bugs.
* Finish LR(0) parser generation and DPDA interpreter.
* Comments/tests.
EOR
bubaflub DID:
* Finished moving 19:21
* Emailed GMP people about licensing stuff
19:21 kid51 joined
bubaflub * Some docs 19:21
WILL DO:
* Finish docs for project
EOR
benabik DID: 19:22
* Separated out PIR and PBC generation
* Decided to test each step of making PBC separately
* Immediatly discoverd bugs in newPOST
WILL DO:
* Debug newPOST
* Blog (honest)
BLOCKERS:
* Subtle differences between newPOST and existing POST classes
* newPOST not generating PIR
* newPOST not generating PBC
EOR
kid51 afk; will backscroll 19:24
cotto_work *did: 19:26
- PDS - kid51++ for leading
-- got a goodly batch of new items on my todo list
- started organizing soh_cah_toa's project based on issues he's run into with imcc
- M0/mole progress:
-- got M0 tests passing on all platforms
-- tightened some loose ends of the spec (still needs review though)
- profiling
-- nothing
*todo
- turn docs/dev/profiling.pod into some we can point users at
- M0 spec, Mole planning
- help soh_cah_toa figure out a good path forward
- YAPC::EU prep
.end
whiteknight hello 19:30
cotto_work hi 19:31
benabik hi
bubaflub hello 19:32
lucian waves
cotto_work How this week go? PDS looked good from my perspective.
tcurtis ~~ 19:33
Util Hello 19:34
NotFound Hola
cotto_work any comments on the past week, PDS-related or not? 19:35
whiteknight no 19:37
cotto_work I guess not.
whiteknight PDS was good. I'm sorry I missed it 19:38
cotto_work I don't see any questions queued. This might be a very short #ps.
NotFound q1q
cotto_work NotFound: go ahead
NotFound www.parrot.org/content/we-need-bet...trings-nci 19:39
bubaflub NotFound++
NotFound Someone object to the addition of a get_string method to UnManagedStruct?
whiteknight UnManagedStruct is where it probably needs to go right now, but I don't like it as a long-term solution
cotto_work whiteknight: +1 19:40
19:40 mikehh joined
whiteknight UnManagedStruct is deprecated, after all, even if we're not moving fast about removing it 19:40
NotFound Yeah, but I'm not aware of what willbe its replacement. The same method can be added toit.
Its replacement as the value returned in NCI 'p' return type, I mean. 19:42
whiteknight as a temporary solution, +1. Add an experimental note 19:43
NotFound Well, the point of asking is to not make it experimental. 19:44
I think we need that as a supported feature.
Temporary, the fact of being in UnManagedStruct. Supported, having that method in whatever 'p' returns. 19:46
whiteknight right 19:47
UnManagedStruct is temporary
the replacement will be supported
NotFound As long as UnManagedStruct is deprecated, an experimental note about one method on it is redundant. 19:48
moritz does anybody have a vision on how interaction with NCI should look like? 19:49
NotFound var p = some_call_that_return_p(); string result = p.get_string('utf8');
mikehh We are quite entitled to change our minds about something being deprecated
whiteknight I've been kicking around a few ideas, I'll have to put some stuff to paper eventually 19:50
kid51 is back, > or <
whiteknight mikehh: sure, but I think many people want to see UnManagedStruct and ManagedStruct destroyed
mikehh yeah, but if we find a use for it :-} 19:51
whiteknight better question is, what do we want NCI to look like in the long term, not just what our temporary solutions to current problems are going to look like
a good redesign of the subsystem might obviate all these issues
NotFound I want to show how a good use of NCI will look like. That's why I want that functionality. 19:52
mikehh well that is the purpose of good design
bubaflub i've talked briefly with plobsing++ about the future of NCI - i think most of it is still in his head
NotFound whiteknight: a good redesign will benefit a lot from real use cases. I can't write the use cases I have in mind because of that problem. 19:53
mikehh we seriously need to consider bus numbers there
lucian NotFound++
NotFound Well, I can'r write them... and see lots of errors test in lots of platforms, user locale variants... 19:54
cotto_work Let's go with a method on UnmanagedStruct for the time being.
NotFound The method name I plan to use is get_string because is what ByteBuffer uses for the same. 19:56
cotto_work -1 to the name because of the collision with the VTABLE function
whiteknight as_string 19:57
cotto_work but that's not a hard fix
NotFound No, but if want to make it a feature supported, will be good to fix the name now.
Or at least before the next release. 20:00
cotto_work NotFound: are you happy with "as_string"?
benabik as_encodign?
NotFound cotto_work: yes 20:01
whiteknight +1
cotto_work NotFound: great. I'll look forward to reviewing your branch.
NotFound EOQ then 20:03
cotto_work any other questions or suggestions for goals for the next week? 20:04
whiteknight make code
cotto_work make: *** No rule to make target `code'. Stop.
Do we have any deprecations in need of removing now that we're past the supported release? 20:06
NotFound Maybe we need a deprecation control team. 20:09
cotto_work NotFound: what would that team do? 20:10
wield the deprecation chainsaw to make sure that the policy is followed?
NotFound Don't know... I wish I had a good answer. 20:12
cotto_work if you do, please blog about it
any other questions? 20:14
whiteknight nope 20:18
20:21 kid51 left
cotto_work Let's call it a wrap. 20:27
20:27 whiteknight left 20:31 NotFound left, moritz left 20:47 lucian left 21:09 lucian joined 21:26 lucian left

Powered by Cro and the Raku® Programming Language.

Please report any issues / comments / feature requests as an issue on App::Raku::Log.

Thank you!