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 5 August 2011.
00:05 bluescreen joined 01:34 contingencyplan joined 08:01 lucian joined 09:24 lucian joined 10:15 particle joined 11:37 lucian joined 12:20 dafrito joined 13:23 bluescreen joined 13:55 contingencyplan joined 15:01 bluescreen joined 16:33 darbelo joined 17:39 lucian joined 17:52 darbelo_ joined 18:25 kid51 joined
kid51 kid51's report 18:25
* DONE 18:26
** Took and worked on trac.parrot.org/parrot/ticket/2007 (links in 'make html'). If no objection, will merge it tonight.
** Applied a partial fix for trac.parrot.org/parrot/ticket/2167 (Storable). Long-term approach still needed.
** Commented and/or posted test results in other tickets.
** Overall count (Trac report #10): 513 (it was 511 one week ago).
** A lot of testing and debugging for HBDB and other branches.
* WILL BE DOING (same as last week)
** Planning Parrot Foundation annual meeting and election.
** Hope to find some time to think about the benchmarking goal and open TT for Win32 focus goal. 18:27
** Taking additional look at docs/project/release_manager_guide.pod.
* EOR
Note: PDS is past; topic needs to be changed.
18:35 NotFound joined
Util # Done: 18:37
* Discussed Parrot 3.6 LTS release (and its Win32 issues) with 18:38
interested people at the Atlanta.pm after-meeting dinner.
* Talked with colomon about trig functions for Complex (TT#1891)
# Plan to do:
* Nothing this week
# Blockers
* $WORK
# 7-day ticket report:
1 closed: done
2 closed: fixed
1 closed: wontfix
5 new
.end
NotFound What I did: 18:39
-parrot
* Implemented the as_string method for NCI usages.
-winxed
* Added builtin load_packfile.
* Assume unresolved qualified symbols are functions.
* Updated Mysql examples for better handling of encodings.
What I will do:
No plan
EOR
19:19 mikehh joined
kid51 kid51's report ADDENDUM 19:19
* As pasted two days ago (nopaste.snit.ch/68743), am concerned that pace of blogging by GSOC students has dropped off 19:20
* This may be inauspicious for completion of all 6 projects.
* IIRC, GSOC participants are getting paid -- but if they don't carry out assignments, pay could be affected, could it not? 19:21
kid51 defers to project GSOC leaders and project mentors on this, but thinks it should be discussed
mikehh been very involved with a new project, unfortunately had very little time for parrot in the last couple of weeks 19:25
19:27 soh_cah_toa joined
dukeleto will start #ps today, but has to dip out early, when cotto++ shows up 19:29
kid51 hello 19:30
mikehh hi there
soh_cah_toa hey
19:31 darbelo joined
tcurtis LATE REPORT: DID: * finished LR(0) DPDA construction, fix bugs, blogged, revised schedule at dukeleto's request (see blog); WILL DO: * Hopefully get SLR(k) grammars ready and things actually parsing. Blog again. EOR 19:31
Hi. 19:32
dukeleto howdy
19:32 atrodo joined
dukeleto how did we do on goals this week? 19:32
tcurtis: glad to hear from you
19:33 pmichaud joined
NotFound Hola 19:35
kid51 duke: Are you leading today? 19:36
19:36 bubaflub joined
dukeleto kid51: would you mind taking over? 19:36
kid51 Will do -- at least until $job interferes.
dukeleto kid51: until cotto++ shows up? He said he would be a few mins late. Today is turning into a $job horrorshow for me.
kid51 So, it appears we don't have too many people here today, and we don't have too many pre-reports.
Net increase of 2 in open Trac tickets (count via Report #10) 19:37
cotto_work ~~ 19:39
kid51 We are 7 days out from our August 3.7 developers release.
Any concerns about that?
Anything people really want to get into master before then?
tadzik o/
how about the --prefix doubts?
NotFound I already merged nci_as_string, which was my bigger interest for the release. 19:40
kid51 tadzik: Has that been documented in Trac? on parrot-dev?
kid51 can't remember where he saw that discussion
lucian is late. sorry
dukeleto kid51: there is a branch on github
kid51: all other discussion was on #parrot
tadzik yep 19:41
kid51 Was that a branch I started?
So, is there a policy decision you/someone wants us to make?
That would be the tadzik/whine-on-relative-prefix branch, correct? 19:43
tadzik kid51: no, I started, you pushed
yes
kid51 k
tadzik I just found the fact of broken relative prefix case disturbing
kid51 I believe I remarked that it would be good to test this on Windows ... if only because the term 'path' is involved.
NotFound I'm against adding things for Configure that serves no purpose other than filtering mistakes at the cost of a lot of platform-dependant checks. 19:44
tadzik I don't like things that pretend to work, but don't. That's all
dukeleto I think it sucks that if you pass in a relative path to --prefix, it silently goes along and then stuff doesn't work much later on. 19:45
tadzik it's LTA when make install finishes and everything seems alright, but libparrot does not load for some reason
cotto_work I see it as an implementation of the principle of failing early.
dukeleto as does tadzik++
kid51 So you think we should require an absolute path to --prefix, correct?
tadzik we do
dukeleto Configure.pl should just fail loudly if --prefix is not an absolute path
tadzik at the moment
dukeleto kid51: non-absolute paths to --prefix makes stuff bork much later on 19:46
lucian dukeleto: me was bitten by that
NotFound And how do you check a valid path? Platform specific string parsing?
kid51 dukeleto: I don't disagree.
tadzik that sucks, but that sucks less than a suprisingly broken parrot, imho
kid51 NotFound: See that branch. There's nothing platform-specific there, really.
is there someone who could volunteer to do a code review on that branch? 19:47
tadzik kid51: does it work on windows?
lucian NotFound: perl's gotta have something for that. python does
kid51 tadzik: As noted above, that's what I'd like to know ... but I have no box to test that on.
tadzik I see
kid51 Code review of branch + No problems running branch on Windows => okay to merge in my book 19:48
NotFound lucian: In configure?
kid51 NotFound: Just check the most recent commit(s) in that branch. There are only 1 or 2 of them. 19:49
lucian NotFound: in perl in general. python has os.path to check if paths exist, and are absolute. i don't know perl, but i'd expect it to have an equivalent
NotFound lucian: we are talking about Configure.pl
lucian NotFound: right. and it's a perl program, isn't it?
tadzik NotFound: we're using File::Spec::is_absolute() atm iirc 19:50
kid51 Volunteers for code review, ladies and gentlemen?
NotFound Ah, ok. That's a lort better.
kid51 Volunteers to test it on Windows?
tadzik note: it's 3 LOC, you get karma, go for it! :P 19:51
kid51 Well, let's hope somebody volunteers before the end of #ps. Let's move on. 19:52
cotto_work I'll test on windows.
kid51 Other pre-3.7 concerns.
cotto_work: Thank you very much.
pre-3.7 concerns?
Going once! 19:53
cotto_work Coke: anything from your end?
kid51 Actually, I have a pre-release concern.
There's a pir program called 'crow' that is used in one step in the release process, mainly to compose the release announcement. 19:54
I couldn't get it to work for 3.6 ...
... but since I had already manually composed the announcement in both .txt and .html, I was able to sidestep it.
That was probably my noobie ignorance, but I did want to mention it as something that might not be working correctly. 19:55
Coke++ has released many times, so I doubt it will be an obstacle.
NotFound kid51: A perl program? 19:56
kid51 No. PIR
tools/release/crow.pir
NotFound I'll take a look at it.
kid51 Thx. 19:57
Coke particle wrote it ages ago.
NotFound++
kid51 Other pre-3.7 concerns?
Coke It is only marginally helpful in cutting the release.
cotto_work It's nice but not vital.
It does seem to break more often than not. Tests would be good. 19:58
Coke cotto_work: easier to rip it out than test it, IMO. But yes, if we're going to keep it, it needs to just work.
NotFound Anyway, we should fix it, or delete it if it doesn't pay the effort. 19:59
cotto_work I'm fine either way.
but yes, we should pick one
kid51 cotto_work: Are you ready to assume the helm?
Coke Regarding GSOC projects, I've been approached offlist by whiteknight & dukeleto about my student, so there is discussion happening.
cotto_work kid51: sure
kid51 q1q 20:00
cotto_work I don't see any other questions, so go ahead.
kid51 Thx. As pre-noted, though I'm not administratively involved in GSOC, I'm concerned that pace of student blogging has fallen off sharply. 20:01
For the participants, this is a summer job, is it not?
dukeleto kid51: yes, as am I. I specifically told students that blogging once a week was needed to Pass.
kid51 So we have to have performance expectations.
dukeleto kid51: yes, they get paid twice, once after they pass midterms and again after passing finals. 20:02
So not blogging means they might not pass and might not get paid.
lucian kid51: don't worry, if any of us fail, we don't get the last payment (which is about half)
lucian might very well fail
kid51: of course, that's not entirely ideal 20:03
kid51 okay, as I said, I defer to the admins and mentors on this ... but I do like to read those blog posts ... and I've been drawn in to 1 of the projects largely due to the student's persistence in committing and blogging 20:04
cotto_work kid51: and your help is appreciated
dukeleto GSoC students: even if you are stuck, blog!
Coke I will note that mentioning that sort of thing only in #parrotsketch is just putting it into an echo chamber.
cotto_work especially if you're stuck
20:04 benabik joined
pmichaud weren't the gsocers supposed to regularly attend #parrotsketch? or is that a fossil from previous years? 20:05
dukeleto pmichaud: that was recommended as well 20:06
pmichaud: and of course, they might not be able to make every one, but gsoc students that rarely attend, that does make it seem like they are putting in a good faith effort
benabik was doing an errand and forgot #ps had moved. 20:07
Coke but, with the pencils down date so close, I'm not sure any behavioral changes at this point will help.
<- bad mentor.
lucian <- bad student :) 20:08
tcurtis <- also bad student
soh_cah_toa i'm always here. the conversatins usually don't involve me though :)
dukeleto It is obvious which students are on IRC and asking questions on IRC and parrot-dev. The others, I worry about. 20:09
Coke: have you heard from rohit in the last 2-3 days? He showed up on #parrot a few days ago, made some promises, and then went *poof* again 20:10
Coke I didn't see him on the channel (my name-dar didn't go off), but he just asked me a PIR question today on gmail chat, yes. 20:12
I've encouraged him MANY times in the past to use #parrot for that sort of thing.
dukeleto i see 3 commits by him in the last 3 weeks: github.com/rohit-nsit08/corellaScr...its/master 20:13
Coke none of which is that big.
And he was a week behind before half time.
So it's not looking good.
dukeleto no, it isn't. Ok, this can be discussed outside of #ps 20:14
Coke Aye.
cotto_work any further questions or comments? 20:15
kid51 is satisfied with discussion 20:17
dukeleto has to dip out 20:21
cotto_work Let's call it a wrap.
20:24 soh_cah_toa left 20:25 benabik left 20:27 NotFound left 20:41 darbelo_ joined 22:07 whiteknight joined

Powered by Cro and the Raku® Programming Language.

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

Thank you!