Priorities for this week: irclog.perlgeek.de/parrotsketch/201...#i_3126985 | 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/parrotsketch/today
Set by moderator on 16 January 2011.
01:15 PerlJam left, PerlJam joined 03:19 cotto_work left 03:20 contingencyplan joined 03:21 cotto_work joined 06:38 Tene left 06:43 Tene joined 08:30 particle left 11:56 contingencyplan left 13:22 atrodo joined 14:19 particle joined
particle identify particle 14:29
ww, and i always forget that nickserv registration syntax.... 14:30
15:27 sorear left 15:33 sorear joined, TimToady left, TimToady joined 16:11 mikehh joined 17:33 tadzik joined 18:10 bluescreen joined 18:14 contingencyplan joined 18:51 bluescreen left 19:03 kid51 joined
kid51 kid51's report 19:03
DONE
* SELF
** trac.parrot.org/parrot/ticket/1957: Configure.pl ignores argument given to --optimize: solved; closed.
** trac.parrot.org/parrot/ticket/1925: remove config step auto::jit: implemented;closed.
19:04 NotFound joined
kid51 ** trac.parrot.org/parrot/ticket/1922: review 'make help': some patches already applied, more to come. 19:04
* CAGE cleaning and/or work with Andy Dougherty led to closing of these tickets:
** trac.parrot.org/parrot/ticket/1953
** trac.parrot.org/parrot/ticket/1950
** trac.parrot.org/parrot/ticket/1132
** trac.parrot.org/parrot/ticket/395
** trac.parrot.org/parrot/ticket/735
** trac.parrot.org/parrot/ticket/1289
** trac.parrot.org/parrot/ticket/97
19:04 bluescreen joined
kid51 * OTHERS 19:04
** Reported several TTs that were largely fixed by others: TT #1942 (nwellnhof++); TT #1963 (NotFound++);
WILL DO/MERGE 19:05
* trac.parrot.org/parrot/ticket/1934: implement coding_std (copyright only) for Python
* trac.parrot.org/parrot/ticket/540: installed versions of dynext/*.so still link to -lparrot in build directory: will close if no objection
* trac.parrot.org/parrot/ticket/1954: 'make docs' target is currently superfluous
ALSO
* Email-ed several Parrot developers in Asia/Australia re good start-time for Parrot Developer Summit Jan 29-30 ... but got no feedback ... bacek? 19:06
EOR
NotFound What I did: 19:22
-parrot
* Added and fixed some tests
* Allowed using a code string in Test;More throws_type
* Minor fixes
-winxed
* Added named parameter 'target' to compile method in compreg'ed object.
What I will do:
No plan
EOR
19:47 plobsing joined 19:49 tcurtis joined
plobsing What I Did 19:50
- pbc-backref
- tt-triage for rgrjr++ tickets
What I Plan 19:51
- no plan ATM
EOR
tcurtis What I did: nothing. What I will do: try to fix bugs and otherwise improve tree-optimization. 19:58
mikehh What I did since my last report: 20:15
* building and testing parrot on amd64/i386, with gcc/g++
* testing and some fixes for release 3.0.0
What I intend to do in the next week:
* testing and fixing
* set up html_cleanup_2 to remove the cruft left over
* (thought I would wait until after the release for this)
* start preparations for next release
.eor
Util # Done: 20:19
* Misc Perl 6 work
# Plan to do:
* Finish .dmg for Rakudo Star (usable by Parrot)
# 7-day ticket report:
31 closed: fixed
10 closed: wontfix 20:20
1 closed: worksforme
34 new
2 reopened
.end
20:23 allison joined 20:25 whiteknight joined
allison - Parrot 2.9.1 packages finally accepted for Debian (unstable) this week, just in time to turn around and submit the 3.0 packages 20:25
EOR
bacek Report: start rewriting PCT to NQP. Will try to squeeze more time to finish it. 20:26
20:26 Eclesia joined
whiteknight WHAT I DID: 20:27
* Tested.
* Worked on imcc_cleanups. Got to a stable point in the interface rework and merged
* Been preparing the next few steps in the process.
* merged exception_backtrace
* Back to blogging
WHAT I WILL DO:
* merge gci_orderedhash_coverage, after some testing
* Get PLA up to date, get it testing again, and release something to use with 3.0
* Create a new branch to continue IMCC work.
EOR
20:28 chromatic joined
kid51 Good afternoon/evening/morning 20:30
Util Hello
allison hi
mikehh Hi there
chromatic afternoon
Coke ~~
tcurtis Hiello.
s/i//
whiteknight hello
NotFound Hola 20:31
bacek ~~
whiteknight okay, let's get started. Thanks to cotto for the 3.0 release 20:33
cotto_work late report
*did:
- reviewed rfcs, approved a handful
- fixed install to update libparrot symlink, fix didn't go into 3.0.0
- kicked PDD10 (embedding) out of draft
+ whiteknight++ for all of his continuing work there
+ whiteknight++ suggested that we do RCs to shake out such bugs in the future
*will do:
- more rfc review
- PDS prep
*blockers:
- none
*eor 20:34
and hello
20:34 dukeleto joined
dukeleto hola 20:34
whiteknight did we have any goals for last week? I didn't write them down and they aren't in the channel topic
cotto_work quite a few
GOAL 1: Write a GettingStartedWithPlumage wiki page
GOAL 2: Fix the setlocale bug (plobsing)
GOAL 3: Review the documentation produced by make html (all)
GOAL 4: test HLLs, especially Rakudo and nqp-rx after the feature freeze (all)
GOAL 5: review all deprecation tickets, file proper notices (cotto, whiteknight, bacek)
whiteknight Okay, I see a few in there that got done 20:35
dukeleto did #2 get done?
cotto_work dukeleto: yes
plobsing++ took care of it later that day
Coke q1q 20:36
cotto_work tadzik++ did #1
dukeleto cotto_work++ # nice work on 3.0.0
kid51 goal 1 appears met: trac.parrot.org/parrot/wiki/Getting...ithPlumage -- but needs to be linked to from main wiki page 20:37
tadzik si. I'd like to receive some feedback on it: more, different, just right?
dukeleto so it seems that we hit all those goals. nice work peeps!
cotto_work We got some good testing against Rakudo, both from the Parrot side and from moritz++
NotFound winxed was tested 20:38
cotto_work q1q
mikehh q2q 20:39
NotFound lua doesn't install from plumage for me, a test fail because it expects a native lua available, or something like that,
kid51 q1q
cotto_work thoughts on the release? 20:40
problems? 20:41
NotFound No problem
cotto_work goals for the week?
kid51 There are quite a few branches to be merged and patches to be applied; let's try to get them done before the weekend. 20:42
Also, people need to start posting on parrot-dev in prep for Parrot Dev Summit
tadzik cotto_work: I'll take care about Deprecations As Data in the weekend 20:43
cotto_work tadzik: I'm very happy to hear that.
tadzik will surely be consulting the ideas and decisions once I get to it
cotto_work that'll make life much more pleasant for users as it gets implemented
whiteknight okay, we ready for questions? 20:44
cotto_work whiteknight: did you have any thoughts on doing an RC?
whiteknight cotto_work: nothing yet
cotto_work ok
Coke: you had a question.
Coke? 20:46
whiteknight ENOCOKE 20:47
kid51 cotto_work: You're next
cotto_work alright. I had one
We need volunteers release managers. 3.3, 3.4 and 3.5 are open. Any volunteers?
whiteknight I was just about to send this question to the lsit 20:48
I'm happy to take 3.3
mikehh let me see how 3.1 goes and I'll let you know if I can handle another one
cotto_work mikehh: no pressure
mikehh: you had a couple questions 20:49
mikehh smolder: I am finding it difficult to get reliable info from smolder these days, there are too many 20:50
different tests being run by the compile farn for me to distill reliable information on pass/fail
We need a new report/interface so we can see recent problems and so on
whiteknight I was worried that reports would get drowned out by the sheer volume
cotto_work dukeleto mentioned something to that effect that earlier this week too. 20:51
mikehh I am not objecting to more tests, but need to be able to distill the info better
cotto_work mikehh: yes
kid51 Need something queryable
cotto_work we need to deal gracefully with a large volume of test submissions
whiteknight what do we want to see, just the failures? 20:52
kid51 Perhaps we should start a wiki page "wishlist: features desired in smoke reporting"
cotto_work kid51: something not entirely unlike that is at trac.parrot.org/parrot/wiki/AutomatedHLLTesting 20:53
mikehh a more condensed report perhaps, showing passes/fails etc. 20:54
cotto_work Please add requests to that wiki page so they won't be lost. 20:55
dukeleto yes, i will implement stuff on that page as I have time 20:56
mikehh btw, how many reports does smolder retain? 20:57
dukeleto it is configurable
mikehh previously it used to delete the older ones
dukeleto i think it keeps the last 1000 for parrot, and less for all the other projects
it keeps statistics for all time, but only keeps full TAP output of the last 1000 parrot smoke reports 20:58
whiteknight We probably do not need that many
especially if disk space is an issue on our VM
dukeleto so what action items do we need?
dukeleto will volunteer for some kind of parrot release in the future, needs to look at his calendar 20:59
cotto_work For now, I think that we should work on the wishlist and figure out what all we'll need.
dukeleto we only have 2 smokers that only test 16 combinations across 2 compilers
I could easily increase that number and seriously DOS smolder 21:00
Smolder can't handle the GCC compile farm
cotto_work We want something that can.
dukeleto we need a spec of what EXACTLY needs to be "supported"
and then we test that
whiteknight It would be very nice if smolder only held the list of failures. We don't need to send the complete TAP output for a passing test suite 21:02
dukeleto will brb
cotto_work mikehh: does that come close to addressing your firstquestion?
s/tq/t q/ 21:03
mikehh something to work on
Coke I'm back.
(list of failures) use the RSS feed.
mikehh the smoke reports are about 150/160K so should not overwhelm a server
cotto_work mikehh: it's more an issue of presenting the data, though a very large number of reports could fill up the server if nobody keeps an eye on them. 21:05
Coke (e.g. smolder.parrot.org/app/projects/feed/1/failed)
the server is already set to only keep N reports.
cotto_work right 21:06
21:06 plobsing left
cotto_work Coke: you had a question? 21:06
Coke related to mikehh's - we should try to get the failed feed to be green so any NEW failures aren't lost in that noise. 21:07
bacek have to go. Will backlog
Coke this is probably mostly skipping tests that shouldn't be run when config'd with certain options.
that's it. 21:08
cotto_work Avoiding false test failures sounds like a good goal. There are a few on Windows that I'd like to see go away. 21:09
mikehh: how about your other question? 21:10
mikehh make html: I haven't had any feedback on any problems encountered there, I would like to include
more documents/pod - for example projects folder - release-manager.pod etc. any ideas on that. 21:11
NotFound The problem with skipping tests is that they usually remain in that state for years until no one remebers why.
Util Marking tests as "skip" is supposed to be ties to a ticket, with the TT# in the skip description 21:12
mikehh NotFound: suitable comments should help, but we should review that
Util s/ties/tied/
cotto_work Util: yes
NotFound mikehh: there are hundreds of skipped/todoed without comment or with uninfotmative ones.
mikehh NotFound: yes, we need to start a project on that 21:13
cotto_work NotFound: We can work on cleaning it out, but there's no need to build it up.
mikehh I started to review skipped tests a year or so ago, but got distracted 21:14
21:14 Eclesia left
cotto_work sounds like another good goal. Any volunteers to focus on that this week? 21:15
mikehh I think it is a longer term goal
cotto_work mikehh: sure, but we can get a start this week.
mikehh it is a goal like closing TT's - continuing 21:16
cotto_work Getting back to your question, I did a brief review of the html docs but didn't get far. It would be nice if the internal docs were included in the generated html. 21:18
mikehh like the native_pbc tests TT#357
cotto_work e.g. the release manager guide
NotFound Also, be careful of not using skip when really should be todo.
cotto_work NotFound: that's an important point.
kid51: did you have a question? 21:21
mikehh regarding make html, bear in mind it pulls the title from the generated pod (or directly from .pod files)
we need to follow up on the codetests related to that 21:22
dukeleto is back
kid51 cotto_work: Thanks.
bacek: Did you get my email re scheduling start-time for PDS weekend after next?
(In previous ps, we said we would try to accommodate devs in different time zones.) 21:23
(This time around, trying Asia/Australia).
cotto_work I saw a message. It seemed to me that the next step would be that you'd work with folks to figure out the time.
kid51 Yes, but I didn't get feedback from the 3 devs I tried to contact.
cotto_work kid51: have you set up a doodle poll? That worked well last time. 21:24
kid51 If, say, we were to meet at what would be 6:00 pm NYC time, that would be 10:00 am next day Sydney
cotto_work: The problem with that poll is that given our geographic distribution, polling will always be biased in favor of North American developers 21:25
cotto_work kid51: that's not a problem. Just give priority to certain developers.
There's nothing about the poll that forces any particular outcome. 21:26
kid51 I won't have time for several days to set up a polling mechanism.
cotto_work It's a 5 minute job. I can do it if you don't have the tuits. 21:27
I have plans for the weekend and I would like to know how the PDS will affect them.
bb5m 21:28
kid51 Did anyone have questions queued after mine? 21:29
whiteknight I think that was it for queued questions
any other questions?
cotto_work back 21:32
whiteknight I don't think there are any other questions
wrap up?
cotto_work none queued
wfm
oh goals
dukeleto Deprecations as Data is important 21:33
cotto_work GOAL 1: Deprecations as Data (tadzik)
GOAL 2: Merge branches, deal with fallout (all)
GOAL 3: Test aggregation and CI wishlist wiki page - AutomatedHLLTesting (all)
GOAL 4: avoid false test failures
21:33 plobsing_ joined
mikehh oh ok 21:33
dukeleto cotto_work: sounds good to me
cotto_work we also need to get the PDS scheduled. kid51, should I set up the poll?
whiteknight yes 21:34
mikehh cotto_work: go for it, can't do any harm
kid51 cotto_work: Yes
cotto_work alright
kid51 has to focus on $job right now
cotto_work I guess that's a wrap.
whiteknight goodbye everybody 21:35
mikehh cu
tadzik o/
21:36 atrodo left, whiteknight left 21:39 kid51 left 21:40 dukeleto left 21:45 NotFound left 21:54 PacoLinux left 21:59 plobsing_ left 23:31 bluescreen 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!