Priorities for this week: Plan our Testing Infrastructure (choose lead person), Contact all GSoC students | Post closed tickets in your report. | Note: This channel is for our Tuesday status meetings (at 20:30 UTC); you probably want #parrot instead. | irclog: irclog.perlgeek.de/
Set by moderator on 10 May 2011.
02:07 whiteknight left 03:32 bluescreen left 07:52 contingencyplan joined 10:44 lucian joined 10:50 lucian left 10:51 lucian joined 11:37 lucian_ joined 11:41 lucian left 12:39 darbelo joined 13:00 lucian_ is now known as lucian 13:35 bluescreen joined 15:36 whiteknight joined 16:34 kid51 joined
kid51 kid51's report 16:34
* DONE
** Cage-cleaning of 20+ Trac tickets; led to closing of trac.parrot.org/parrot/ticket/2115 and discussion of others. 16:35
** Took and closed trac.parrot.org/parrot/ticket/2119.
** Almost succeeded in closing trac.parrot.org/parrot/ticket/540, but Taptinder reported build error on Cygwin64.
* DOING
** Looking into trac.parrot.org/parrot/ticket/1761. How do I get access to parrot.org?
** Drafting presentation for YAPC::NA::2011 in Asheville, Monday, June 27.
** Trying to encourage participation in FOSSCON in Philadelphia, Saturday, July 23.
EOR
17:04 atrodo joined
cotto_work *did: 17:10
- reviewed the Select dynpmc in tewk/select (#2034)
-- summary:
--- not a fan of the interface
--- am willing to update it myself, accept a patch or commit as-is experimentally
--- I'll do something once I get some kind of feedback.
- booked hotel for YAPC::NA 17:11
- M0 progress:
-- got a copy of Appel's Compiling with Continuations
--- It's dense and delicious, not unlike granola.
-- wrote some more tests
-- figured out a workable syntax for labels
- GSoC progress:
-- helped soh_cah_toa get a minimal frontend building
-- helped him navigate Parrot's guts
-- the debugger is now known as "hbdb"
*will do:
- get the M0 prototypes as far along as possible by Saturday
- write a bunch of example M0 code and tests
- make a list of questions
- meet with allison, dukeleto and chromatic on Saturday to (hopefully) hammer out the last bits of M0
- help soh_cah_toa as needed
*eor
note: I may miss a subset of #ps today. dukeleto++ said he'd lead.
whiteknight WHAT I DID: 17:40
* Was without internet access for most of the week.
* Worked on Rosella. The Query, File, String and Memoize libraries, and the REPL. Query and Memoize are now listed as "stable"
* Playing around with my Parrot-in-Xchat plugin. nothing working yet, but less hopeless now than it was.
* Ran callgrind on an unoptimized Parrot building Rakudo's core.pm. 4 hours later, I uploaded the results file to a repo on github for public analysis. I've already found some goodies in it.
* Looked at profiling issues in general. Starting to form a vague idea about what to do with it.
* Finally picked up my Python book and am starting to learn it in earnest.
* GSOC Started! We have 6 active students starting their projects as of yesterday. I'm following along anxiously.
WHAT I WILL DO:
* Looking at profiling, PCC, and optimization opportunities. May run a few benchmarks if I can fix anything. 17:41
* Maybe more Rosella, but I've scratched that itch a lot lately.
EOR
17:41 lucian left 18:05 benabik joined 18:16 NotFound joined 18:25 mikehh joined
NotFound What I did: 18:26
-parrot
* Added a --help option to the create_language tool.
* Minor fixes.
-winxed
* Improved compile time scope search.
* Put some examples up to date with recent NCI changes.
* Mysql example now allows to specify connection encoding.
* Created module WinxedGtk to acces Gtk+ via NCI.
github.com/NotFound/WinxedGtk
Written in winxed, but should work from pir and HLLs.
What I will do;
* Improve WinxedGtk.
EOR 18:27
mikehh What I did since my last report: 18:32
* building and testing parrot on amd64/i386, with gcc/g++
* some fixes
* building and testing rakudo and winxed against latest parrot
* first cut at adding deprecation docs from wiki to docs/deprecations in master
as .pod files. added to html docs as deprecations under developer docs
What I intend to do in the next week:
* testing and fixing
* continue working on deprecation docs as necessary
.eor
benabik DID: 19:28
- got a shiny box from Google
- blogged: parrot.org/content/gsoc-0-schools-out-summer
- twice: www.parrot.org/content/gsoc-1-your-marks-get-set
- Converted PAST::Node and children into NQP with Q:PIR
WILL DO:
- Convert PAST::Compiler similarly
- Start replacing Q:PIR with NQP
- Get work done so far into main repo once bacek looks it over
EOR
Just DID: Booked a room at YAPC::NA 20:01
20:21 lucian joined 20:23 dukeleto joined
Util # Done: 20:24
* Adapted and applied patch from bubaflub++ to close TT#1217.
* Realized I had the wrong date for PDS, a full week after the event :(
# Plan to do:
* Write talk for YAPC::NA::2011
* TT#919
# 7-day ticket report:
11 closed: fixed
1 closed: invalid
7 new 20:25
1 reopened
.end
dukeleto What I did: 20:27
* Hacked on parrot embed tests some more
* Talked with some GSoC students
* Got Jitterbug to play nice with Parrot: jitterbug.leto.net:3000/ 20:28
* Hacked on M0 assembler. It generates partially correct bytecode now.
* Merged some pull requests in Cardinal and Parrot.
What I will do:
* Try to get Rakudo to play nice with Parrot
* Figure out all the other things that I am forgetting that I should do.
* Book tickets to YAPC::NA
Blockers:
* EFINITETIME
.EOR
20:29 whiteknight left
dukeleto hola! 20:31
Util Hello
mikehh hi there
kid51 hello 20:32
cotto_work: Are you present? If not, is dukeleto chairing?
dukeleto kid51: would you like to chair? 20:34
kid51: i was supposed to, but I have unavoidable maintenance/landlord issues
how did we do last week?
NotFound Hola
benabik hi
mikehh moved deprecation stuff to docs/deprecations
dukeleto i see mikehh++ made progress on putting deprecations in the repo
kid51 No net improvement in BZ tickets (per util's report)
dukeleto we need to make sure we update all our documentation to reflect that
BZ tickets? 20:35
kid51: also, you email support@osuosl.org and give them your ssh pub key to get access to parrot.org
cotto_work hi
kid51 thx
Util 12 closed vs 8 new-or-reopened 20:36
dukeleto hands it over to kid51++, but will be here intermittently
kid51 cotto_work: Do you want to chair?
cotto_work kid51: sure
dukeleto kid51: he is at a $work meeting
or not.
dukeleto backs away
cotto_work got back sooner than expected
dukeleto cotto_work: sounds good. all yours
cotto_work Our goals were: 20:37
GOAL 1: Merge/review Select dynpmc (cotto)
GOAL 2: Contact all GSoC students
GOAL 3: Consolidate deprecations data into api.yaml (mikehh)
20:37 soh_cah_toa joined
cotto_work I know that the first one was done. 20:37
dukeleto I have contacted many, but not all gsoc students. 20:38
Anybody heard from rohit lately?
cotto_work Unfortunately we had to disqualify one GSoC student.
dukeleto I skyped with bubaflub++ and soh_cah_toa++
cotto_work dukeleto: weekly blog posts from GSoC students are now mandatory, right?
dukeleto cotto_work: yeps 20:39
cotto_work ok.
I saw some work on deprecations got done by mikehh++ 20:40
mikehh Goal3: I placed the wiki documentation in docs/deprecations
cotto_work mikehh: thanks
q1q 20:41
mikehh not sure about the yaml stuff
I also put those docs into make html, but maybe we want to move it, it is under developer at the moment 20:42
dukeleto mikehh: should be fine for now 20:43
mikehh: did you get to update the instructions about where deps go, so that people are pointed to edit/create those pod files, instead of wiki pages?
cotto_work git is good at tracking files when they get moved
mikehh didn't so much move as copy them (with appropriate changes) but needs checking 20:44
Util q1q 20:45
dukeleto cotto_work: i am talking about documentation
cotto_work: and this stuff moved from outside of git, into git
cotto_work dukeleto: sure. I'm just saying that we can move it around later if needed.
20:45 spinclad joined
dukeleto cotto_work: agreed 20:46
spinclad: welcome
cotto_work We should make sure that the docs updates happen. Having straightforward and accurate deprecation process docs is important.
dukeleto cotto_work: i would so go far as to say it is essential to the health of parrot
deprecations are what allow parrot to move and improve quickly 20:47
mikehh at the moment it would have to be done manually, I don't see a way to automate it
dukeleto mikehh: i am confused at what both you and cotto are talking about
perhaps they are not the same thing. 20:48
mikehh I'll try and keep on top of it
spinclad dukeleto: thank you. (nothing to report, for months now...)
cotto_work I'm talking about updating the docs that were moved into git to say that git (not the wiki) needs to be udpated.
*updated
mikehh I woluld say both need to keep in sync 20:49
if something is added to docs/deprecations, it needs to be added to the wiki and vive versa 20:50
dukeleto mikehh: seems horribly inefficient and just waiting to be out of sink
mikehh: why do we need the trac wiki when we have HTML-formatted POD on Github?
cotto_work One should be canonical and the other should be automatically synced.
dukeleto mikehh: why do we want to create more work for ourselves?
cotto_work (at most. Alternately, we could have just one source.) 20:51
dukeleto I vote for the POD in our repo to be canonical. All external copies of it are the mirrors.
cotto_work I'm fine with that.
dukeleto cotto_work: we could have a www::mech script that updates our trac wiki with the version in our docs/ 20:52
cotto_work dukeleto: yes we could
as long as we're smart about keeping credentials out of cvs
*vcs
dukeleto cotto_work: but i am not writing it, and i think one source is fine for now 20:53
cotto_work dukeleto: Good. You're overloaded already. Volunteers are welcome, but until one shows up we'll keep the canonical version in git.
mikehh I'll look into it if I get some time for it
NotFound Information is what allows parrot to improve quickly. Lack of deprecation information blocks people when his code need updating.
20:54 wknight-phone joined
cotto_work mikehh: thank you. 20:54
lucian is late again. sorry 20:55
mikehh it should move into docs.parrot.org when it gets updated
cotto_work Are there any other comments on the week's goals before we move to questions?
I guess not. 20:56
y question: It's becoming clear that our trac setup is making it difficult for people outside of Parrot to submit tickets (among other things). Can we have a volunteer to look into alternative ways of preventing trac spam that will allow humans to continue posting bug reports?
*my
Util Just use parrotbug? 20:58
cotto_work except that parrotbug doesn't actually post bugs
mikehh whatis to prevent spam coming in from there?
kid51 Some data: Rarely a day goes by when a bot creates a Trac user account. But because we deleted the 'auth' status, that account -- or any valid new user account -- can't do anything until we/I grant them 'developer' privileges. 20:59
The ratio of bot accounts to real-human accounts is upwards of 10:1.
20:59 bubaflub joined
cotto_work mikehh: parrotbug doesn't currently interact directly with trac. 20:59
kid51 A lot of work for me.
lucian someone suggested moving to github issues, forgot who
kid51 But it has kept us free of spam Trac tickets since we made that change. 21:00
wknight-phone me
cotto_work I'd be +1 if we had the migration to do over, but another ticket migration sounds like a huge time sink.
mikehh if we set parrotbug to post to trac or github, will that stop spam? 21:01
cotto_work I would much rather make it easy for humans to set up usable trac accounts without admin intervention. 21:02
I'll take a shot at finding something viable. Others are encouraged to jump in and help. 21:04
eoq 21:05
Util: what was your question?
Util 0% coverage of src/list.c: tapir2.ro.vutbr.cz/cover/cover-resu...3/c_cover/
Looks like current GCs no longer use list.*. Is the code dead now?
cotto_work Util: I believe that's bacek's code. 21:06
Util OK; I will ask him. Thanks. 21:07
cotto_work Are there any other questions? 21:08
kid51 cotto_work: Yours 21:09
cotto_work kid51: mine was answered (enough)
21:10 wknight-phone left
kid51 has to reboot 21:13
21:13 kid51 left
cotto_work Let's call it a wrap. 21:14
mikehh Goals?
cotto_work ah
GOAL 1: Make git the canonical source of deprecation information (mikehh)
GOAL 2: Find a way to make trac human-friendly and bot-hostile (cotto)
(others are welcome)
moderator Priorities for this week: irclog.perlgeek.de/parrotsketch/201...#i_3798168 | Post closed tickets in your report. | Note: This channel is for our Tuesday status meetings (at 20:30 UTC); you probably want #parrot instead. | irclog: irclog.perlgeek.de/ 21:18
21:27 bluescreen left 21:37 bubaflub left 21:42 benabik left 21:52 NotFound left
tcurtis late post-report since I managed to doze off both before and after #ps: 22:55
What I did: 22:56
Created repository for my project: github.com/ekiru/lalrskate
Wrote a LALR::Grammar class and a prettyprinter tool for same. 22:57
Got my build system working.
What I will do:
Blog.
More reading.
Tests for the grammar class. 22:59
Sketch out a more detailed schedule for the next few weeks. 23:00
Hopefully get started on same.
EOR
23:17 lucian left 23:20 soh_cah_toa left, mikehh left 23:41 whiteknight joined 23:44 darbelo 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!