Vision for 2.0: Production Users | Priority for 1.5: Remove Deprecated Features | trac.parrot.org/parrot/wiki/Propos...chProtocol | Note: This channel is only for our Tuesday status meetings; you probably want #parrot instead.
Set by moderator on 4 August 2009.
03:03 japhb joined 04:50 eternaleye joined 05:01 eternaleye joined 08:35 masak joined 08:40 cotto joined 09:43 cotto joined 15:34 davidfetter joined 15:45 rdice joined 17:06 mikehh joined
cotto # What I did: 17:19
* repurposed the old profiling runcore in the pluggable_runcore branch
- found out that nested runloops add significant complexity to the profiling runcore and postprocessing script
- progress beyond that has not been of a tangible nature
* removed deprecated branch_cs opcode
# What I hope to do and how many tuits I expect to have:
* figure out and implement something to deal with inner runloops
# What could block my progress:
* nothing external
That's the end of my report.
17:25 whiteknight joined 17:26 PacoLinux joined 17:38 NotFound joined
japhb * What I did: 17:45
+ Removed my one deprecated file as per release priority.
+ Lots of discussions and design proposals for Parrot module ecosystem ("Aviary")
* What I hope to do:
+ Continue moving Aviary design along
+ Need to sleep occasionally
* Comments:
+ For now Aviary is higher priority than OpenGL and HLL interop, but it would be nice to get back to them at some point.
* EOR
whiteknight What I did: 17:46
* Worked more on the lazy PObj allocator. Very close to having a working implementation for testing
* Moved some stuff around to prepare for the PCT book.
What I will do:
* Helping out with TT #895
* Working on the lazy allocator more, hoping to get it wrapped up and benchmarked
* Still hoping to do IO stuff.
What I am blocking on:
* Not enough time.
NotFound What I did: 18:00
* Created branch auto_attrs to work on TT #895:
** Changed the way of setting the attr size in thee vtables.
** Changed a lot of pmc and dynpmc to auto attrs.
** Built and tested partcl and rakudo with the branch.
* Throw a more descriptive exception when throwing something invalid.
* Minor fixes and cleanups.
What I will do:
* More work on the auto_attrs branch.
18:10 allison joined
allison - Getting back into the swing of things now that I have time again. Catching up on tickets, half-way through a review of DEPRECATED.pod, catching up on the mailing list. 18:16
- Started a fresh branch for the PCC refactors ('pcc_arg_unify'), extracted and applied changes from the old branch. Working my way through the issues there, got it all the way through to the PGE build.
- On the branch, added a 'make corevm' target, so 'make coretest' doesn't have to build PGE/PCT/etc before running the core tests. Very helpful in debugging big refactors to find problems in the small test cases, before trying to throw the massive PCT codebase at it.
- Put out a call for nominations of directors for the Parrot Foundation 2009-2010. (If you're not a member and have thoughts, mention them to a current member. Membership for this election is just committers, we'll get the membership application review going after the first annual meeting.)
EOR
japhb allison: the most important nomination is the one you already made (pmichaud) 18:23
18:27 Util joined
Util # Done: 18:27
* NULL
# Plan for next week:
* Resolve discrepancies in PDD13 vs actual Packfile {.c,PMC} implementations.
# Blockers:
* None.
.end
whiteknight Hello 18:34
Util hi
NotFound hola
18:34 l3t0 joined
spinclad waves 18:34
japhb o/
allison or chromatic? 18:36
fearless leaders?
allison no chromatic?
then let's get started 18:37
any missed reports?
any questions?
japhb q1q
allison q1q
japhb: go ahead
NotFound q1q
japhb Particle pointed out that it's time to put the Aviary proposal somewhere hack-friendly. HE suggested either a POD committed to the repo, or to the wiki. 18:38
Any preference?
NotFound wiki 18:39
allison wiki probably makes the most sense
japhb OK, will do.
particle pod-ify it when it settles down
allison wiki gives you a wider field of contributors
japhb allison: if you can comment on my email from last night, I can do some cleanup en passant
allison will do 18:40
japhb I'm done.
(I mean, asking questions)
allison NotFound: your question?
NotFound TT #895, auto_attrs branch. Not sure when the branch can be merged. Now, after next release, a full deprecation cycle? 18:41
allison NotFound: depends on what kind of changes it has
do you have a diff that can be reviewed? 18:42
NotFound I've do my best to make sure that it doesn't affect foreign PMC, but the posibiity exists.
allison (trac makes one easily, if there haven't been any updates from trunk since the original branch)
NotFound No, I do several updates from trunk. 18:43
I've put a short report on the ticket.
whiteknight allison: we'll put together a diff
allison that makes it a little harder to review, but still possible
since the next release is next week, let's definitely hold off until after that 18:44
but, we can merge right after, as long as you're not changing behavior
that is, it's okay if we need to track down a few bugs
as long as the intention is to still keep supporting the same behavior 18:45
NotFound Behavior is maybe changed if someone is doing something really tricky to initialize or destroy his pmc.
allison (with an agreement that different behavior is a bug)
NotFound partcl and rakudo builds fine and pass tests.
allison if they're not respecting the standard API, we're allowed to break things
(just help them figure out how to use the standard API) 18:46
All that to say, yes, it's likely we can merge right after 1.5. I'd like to review it.
NotFound Ok.
The key changes are just a few lines, the most of the differences are just updating pmc to use it. 18:47
allison reasonable
any other questions? 18:48
NotFound I'll have to add some comments to clarify things, though.
allison NotFound: that's a good idea for future maintainability 18:49
my question is for japhb
japhb shoot ("bang")
allison You mentioned in your report that Aviary is a higher priority than HLL interop for you at the moment. Is there a list of remaining tasks for HLL interop you could hand off? 18:50
japhb allison: unfortunately, a) I haven't thought about it in a few weeks, so I'd have to go back and look, b) Much of my part in things was "try A B and C, found out B and C don't work, get Tene to work on B and research C further", in other words, iterative experimentation and refinement. 18:51
That could be handed off, but not too many people are experimenting in that space right now.
allison Essentially a problem of creative test cases. 18:52
japhb allison: yup.
And I need Tene or someone of equal knowledge. 18:53
allison Okay, great. If anyone's looking for something to do, that's a good one.
japhb definitely.
allison (Note that hll interop is on the roadmap for 1.6) 18:54
Roadmap review: trac.parrot.org/parrot/report/14
packfile pmcs?
Util I have not formally claimed any tickets on Packfile, but am working on bits of it. I saw that Austin was as well. 18:55
allison do you have a status update? is this task likely to be "complete" for 1.5? 18:56
or, is it more of an ongoing refactor project?
Util Not based on just my own work; AFAIK, you cannot currently build a .pbc from scratch with the Packfile PMCs. 18:57
Austin might be on pace for 1.5, or not, but I definitly am not.
allison okay, unlikely for 1.5, will check with Austin for likely date 18:58
pir profiling tools?
l3t0 allison: i am interested in that if someone can push me in the right direction 18:59
allison chromatic isn't here to report, unlikely for 1.5 19:00
NotFound Looks like none of the people working on that is here today.
allison l3t0: you're in Portland right? we should get you and chromatic together for a hacking session
standard seed libraries? 19:01
the current task is picking one for each month
any preferences for the 1.6 release?
NotFound zlib ? 19:02
19:02 Coke joined
japhb allison: I think that kindof gates on Aviary (dis)agreements 19:02
l3t0 allison: he will probably be at the next pdx.pm meeting, i can ask him about it there
Coke ~~
japhb NotFound: I'm in favor of zlib personally, yes. 19:03
allison japhb: really, the whole question has been submerged into Aviary
japhb allison: ... and that's why Aviary rose to the top of my priority list. :-)
allison since core libraries are pretty irrelevant once you can easily install non-core libraries 19:04
NotFound I'm definitely in favor of dropping postgresql from core and let it to the aviary
japhb ... except for whatever ones you need to do the installs. But we're discussing that on the thread.
NotFound: +1
allison japhb: aye will continue discussing that there 19:05
spinclad .oO{ meanwhile go ahead with zlib, working its interface with Aviary as it develops... }
allison NotFound: will put in a ticket for deprecating postgres from the core
japhb spinclad: that's an excellent idea. I plan to be doing the same with OpenGL too. 19:06
allison aye, there's no reason the "seed" libraries have to be core
and the "core or non-core" debate can come later
japhb nod
allison most important is "get it working"
japhb vigorous nod 19:07
spinclad :)
allison will add a ticket zlib as seed library for 1.6 release. Do we have someone volunteering to implement it?
japhb I *might* be able to work on it or help with it, but I can't own that task right now. 19:09
allison alrighty then, I'll add the ticket anyway, noting that it's non-critical and not a release blocker
NotFound Me more or less the same.
japhb q1q 19:10
allison pluggable runloops?
chromatic not here, but task likely to be scheduled to 2.1 or beyond
prune c data structures?
the task there is currently to decide which data structures are candidates for pruning 19:11
that ends roadmap review
japhb: your question?
japhb You said you were mid-way through reviewing DEPRECATED. What kind of review are you doing?
allison Coke mentioned that there were a number of DEPRECATED items blocking on answers. I'm going through and making sure they have answers. 19:13
Also working on some of them, like the runtime/parrot search path deprecation.
japhb OK, gotcha.
allison Any final questions or comments before we close? 19:15
Coke hurm.
nah. 19:16
19:16 PacoLinux left
allison Coke: there's always #parrot 19:16
thanks all, tty next week!
19:16 Util left
japhb o/ 19:16
19:16 Coke left, NotFound left 19:53 allison joined 20:53 Whiteknight joined 22:10 PacoLinux joined 22:11 PacoLinux 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!