Camelia, the Perl 6 bug

IRC log for #parrot, 2010-08-31

Parrot | source cross referenced

| Channels | #parrot index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
00:00 dafrito the nice thing about that problem is it's easy for us to detect; false redirections would be much nastier
00:01 cotto_work Missing a handful of revisions doesn't bug me too much but 9000-ish is suspicious.
00:01 cotto_work 7000-ish, rather
00:01 dafrito yeah, it's alot, admittedly
00:01 dafrito but it's also 600 branches (on my old repo) to like, 32 branches
00:02 Psyche^ joined #parrot
00:04 cotto_work The nice thing is that if we charge ahead and later find out that we missed some commits, they'll be relatively easy to add to trac's db once we have them.
00:05 Paul_the_Greek I see that no PMC extends default. Does a PMC extend default if it doesn't specify an extends clause?
00:07 Patterner left #parrot
00:07 Psyche^ is now known as Patterner
00:09 Tene left #parrot
00:09 cotto_work I believe that's the case
00:10 cotto_work confirmed
00:11 davidfetter left #parrot
00:14 Andy left #parrot
00:22 luben hello everyone
00:22 luben I'm back from vacation
00:22 cotto_work hi luben
00:23 seatek left #parrot
00:25 luben I have seen almost all of the patches from the ticket are applied. TT #1741 could be closed - all functionality now is in trunc.
00:26 cotto_work What about the ones that weren't applied?
00:27 luben they are not so important - mostly cleanups and fixes for bugs introduced in the cleanup. I will rework them for the current state
00:28 cotto_work great
00:28 luben thanks all for applying and thanks nwellnhof for spotting and fixing a bug I have introduced in hash_destroy
00:29 cotto_work opbots, names
00:29 Topic for #parrot is now Parrot 2.7.0 "Australian King" Released! | parrot.org Log:  irclog.perlgeek.de/parrot/today | Nopaste: nopaste.snit.ch:8001 | close 20 tickets (-10 to go), merge outstanding branches; parrot-members: vote by 2PM PT Tues Aug 31
00:38 Paul_the_Greek Do we have a tool that shows where each vtable method comes from for a given PMC?
00:38 particle joined #parrot
00:39 dalek TT #1741 closed by cotto++: Macro abstraction for hash iterations
00:39 dalek TT #1741: http://trac.parrot.org/parrot/ticket/1741
00:39 cotto_work It used to be the case that you could see that by looking at the generated C for a PMC, but we made that dynamic to avoid the cost of all the exported symbols of the VTABLE functions.
00:40 cotto_work You could break out gdb and look at the vtable struct for a given pmc.
00:40 Paul_the_Greek Yes, I could do that.
00:41 Paul_the_Greek chromatic suggested that I look at making Boolean not extend Integer.
00:41 cotto_work +1 fwiw
00:42 Paul_the_Greek So the question is: Which vtable methods are considered part of the definition of Boolean, and which are spurious because they are simply inherited?
00:42 Paul_the_Greek In other words, what is the official definition of Boolean (or any other PMC)?
00:42 cotto_work because of the difference between methods and VTABLE functions as implemented in Parrot, we try to avoid the term "vtable method"
00:43 Paul_the_Greek Oh yes, I saw that. Good idea. I'll watch my vocabulary.
00:43 cotto_work If you saw it somewhere, feel free to point it out or excise it yourself.
00:43 Paul_the_Greek The list of vtable functions in the documentation is not the official definition.
00:44 cotto_work src/vtable.tbl is the most official list
00:44 Paul_the_Greek That's the list, but it doesn't define the Boolean PMC.
00:45 Paul_the_Greek For example, Boolean modulus ... meaningful or nonsense?
00:45 whiteknight I read that file when I want to make myself angry
00:45 Paul_the_Greek Ha ha.
00:45 cotto_work are you asking about src/pmc/boolean.pmc ?
00:45 Paul_the_Greek Now, yes. But the philosophical question pertains to many PMCs.
00:45 Paul_the_Greek Where is the official definition of the semantics of Boolean?
00:46 cotto_work Mu
00:46 Paul_the_Greek The question doesn't make sense so there is no point in answering? :D
00:46 cotto_work It makes a faulty assumption.
00:47 Paul_the_Greek The assumption that there should be an official semantics of Boolean?
00:47 cotto_work that's the one
00:47 Paul_the_Greek Uh, how can we support a PMC if there is no official semantics? For example, can I blow off Boolean modulus without concern?
00:48 cotto_work the test suite kinda shows the expected semantics
00:48 x3nU_ left #parrot
00:48 x3nU joined #parrot
00:48 Paul_the_Greek In some sense, yes. But we could simply forget to test something that people rely on.
00:49 sri left #parrot
00:49 sri joined #parrot
00:49 cotto_work In that case, we add a test.  No, it's not optimal.
00:50 Paul_the_Greek So if I were to detach Boolean from Integer, I would have to be very careful to add vtable functions to Boolean for each Integer function that makes any sense at all.
00:50 cotto_work It does help that Parrot has inherited Perl's testing culture, so we do have a good chunk of tests lying around.
00:51 Paul_the_Greek Yes, we do. I'm quite impressed with the test suite. And with the deprecation policy.
00:51 Paul_the_Greek I'm thinking I could make up my own semantics for --bool :D
00:52 Paul_the_Greek False becomes true; true stays true.
00:52 Paul_the_Greek Oh crap, never mind. I can see programmers deciding that's useful.
00:52 cotto_work Yes.  Anyone who relies on negating a bool rather than inverting it is asking for problems.
00:53 cotto_work I'd probably just throw an exception.
00:53 Paul_the_Greek I think we'd be doing the world a favor if we threw exceptions for all that whacky stuff.
00:53 Paul_the_Greek We are both geniuses.
00:53 cotto_work or return a bit indicating whether it's leap year
00:53 Paul_the_Greek Eliminate a whole class of inscrutable bugs.
00:54 Paul_the_Greek -bool: This operation results in a boolean indicating whether the current year is a leap year.
00:54 Paul_the_Greek ++bool: This operations results in a boolean indicating whether cotto is working.
00:55 Paul_the_Greek Is this appropriate to chat about on #ps tomorrow?
00:55 whiteknight yes
00:55 cotto_work yes, unless you're prepasting your report
00:56 whiteknight and use cursewords
00:56 Paul_the_Greek Should I start out this way: Why oh why did anyone propagate the stoopid C convention of representing booleans with arbitrary integers? Wail and moan!
00:56 Paul_the_Greek Or might that intimidate people?
00:57 cotto_work It'll ingratiate you with some.
00:57 Paul_the_Greek Excellent, then I'll start that way.
00:57 Paul_the_Greek I can mention leading 0s indicating octal, too.
00:58 Paul_the_Greek Okay folks, thanks for all your help, as always. See you tomorrow.
01:08 dngor_ left #parrot
01:08 dngor joined #parrot
01:17 moritz left #parrot
01:17 moritz joined #parrot
01:21 Myhrlin joined #parrot
01:25 Paul_the_Greek left #parrot
01:29 cotto ~~
01:33 whiteknight left #parrot
01:47 singalong joined #parrot
01:53 kid51 joined #parrot
01:59 Topic for #parrot is now Parrot 2.7.0 "Australian King" Released! | parrot.org Log:  irclog.perlgeek.de/parrot/today | Nopaste: nopaste.snit.ch:8001 | close 20 tickets (-12 to go), merge outstanding branches; parrot-members: vote by 2PM PT Tues Aug 31
02:04 dalek TT #677 closed by jkeenan++: Which tools/*/*.pl programs belong in which tools/* directory?
02:04 dalek TT #677: http://trac.parrot.org/parrot/ticket/677
02:25 singalong left #parrot
02:34 Myhrlin left #parrot
02:35 janus left #parrot
02:35 janus joined #parrot
02:38 kid51 left #parrot
02:39 singalong joined #parrot
02:48 Andy joined #parrot
03:24 cotto seen paul_the_greek
03:24 purl paul_the_greek was last seen on #parrot 2 hours, 26 minutes and 34 seconds ago, saying: Okay folks, thanks for all your help, as always. See you tomorrow.
03:24 aloha paul_the_greek was last seen in #parrot 2 hours 26 mins ago saying "Okay folks, thanks for all your help, as always. See you tomorrow.".
03:43 dukeleto wow, 32 tickets closed this week!
04:25 Topic for #parrot is now seeParrot 2.7.0 "Australian King" Released! | parrot.org Log:  irclog.perlgeek.de/parrot/today | Nopaste: nopaste.snit.ch:8001 | close 20 tickets (-13 to go), merge outstanding branches; parrot-members: vote by 2PM PT Tues Aug 31
04:25 * cotto tosses another on the pile
04:28 cotto and another if that slacker paul_the_greek can get back to me before #ps
04:32 cotto anyone else on XP by chance?
04:33 dalek parrot: r48731 | cotto++ | trunk/t/run/options.t:
04:33 dalek parrot: add a test to ensure that VERSION match the version of the built parrot for TT #1721
04:33 dalek parrot: review: http://trac.parrot.org/parrot/changeset/48731/
04:38 dalek TT #1721 closed by cotto++: parrot version numbers in oplib
04:38 dalek TT #1721: http://trac.parrot.org/parrot/ticket/1721
05:01 singalong_ joined #parrot
05:01 singalong left #parrot
05:02 singalong_ is now known as singalong
05:12 fperrad joined #parrot
05:17 aloha left #parrot
05:17 bacek left #parrot
05:26 Andy left #parrot
05:32 Util left #parrot
05:32 Util joined #parrot
05:32 dukeleto left #parrot
05:32 dukeleto joined #parrot
05:32 pmichaud left #parrot
05:37 pmichaud joined #parrot
05:37 Andy joined #parrot
05:38 Andy left #parrot
05:55 dalek github-trac: 0ef5fd9 | cotto++ | github/hook.py:
05:55 dalek github-trac: add tooltip support to the github hook
05:55 dalek github-trac: review: http://github.com/cotto/github-trac/commit​/0ef5fd929052ac1080030e5e048b042bf15a91ca
05:56 Topic for #parrot is now Parrot 2.7.0 "Australian King" Released! | parrot.org Log:  irclog.perlgeek.de/parrot/today | Nopaste: nopaste.snit.ch:8001 | close 20 tickets (-13 to go), merge outstanding branches; parrot-members: vote by 2PM PT Tues Aug 31
06:06 uniejo joined #parrot
06:07 theory left #parrot
06:27 jan left #parrot
06:49 sorear seen coke
06:49 purl coke was last seen on #parrot 9 hours, 38 minutes and 16 seconds ago, saying: http://thingsrbetterwithaparrott.typepad.com/
07:05 tcurtis left #parrot
07:40 dalek parrot: r48732 | NotFound++ | trunk/src/pmc/exception.pmc:
07:40 dalek parrot: missing break that gave wrong result from getattribute exit_code
07:40 dalek parrot: review: http://trac.parrot.org/parrot/changeset/48732/
07:52 moritz http://moritz.faui2k3.org/tmp/parrot_karma karma extacted from #parrot logs
07:53 moritz *extracted
07:56 moritz I don't quite trust it yet
07:58 moritz in particular it seems to record negative karma for words before ' -- '
07:59 * moritz fixes extraction regex
08:00 moritz ah, much better
08:00 moritz link updated
08:01 moritz it also contains some aliasing code, so for example karma from 'allisonrandal' is transferred to 'allison'
08:13 dalek parrot: r48733 | NotFound++ | trunk/t/pmc/exception.t:
08:13 dalek parrot: refactor Exception attributes tests and add some more
08:13 dalek parrot: review: http://trac.parrot.org/parrot/changeset/48733/
08:16 bacek joined #parrot
08:23 aloha joined #parrot
08:33 dalek winxed: r614 | NotFound++ | trunk/winxedst1.winxed:
08:33 dalek winxed: use new keyed instead of new 'literal' for parrot core types in stage 1 compiler
08:33 dalek winxed: review: http://code.google.com/p/w​inxed/source/detail?r=614
08:38 dalek winxed: r615 | NotFound++ | trunk/pir/winxed_compiler.pir:
08:39 dalek winxed: update installable compiler
08:39 dalek winxed: review: http://code.google.com/p/w​inxed/source/detail?r=615
08:43 jsut_ joined #parrot
08:45 dip joined #parrot
08:48 jsut left #parrot
08:52 NotFound purl: whiteknight?
08:52 purl i guess whiteknight is mailto:wknight8111@gmail.com or the grand master funk or http://wknight8111.blogspot.com/
08:54 NotFound msg whiteknight ./plumage install parrot-linear-algebra fails, looks like parrot_setup should be nqp_setup in the instructions section of its metadata
08:54 purl Message for whiteknight stored.
08:54 aloha OK. I'll deliver the message.
09:19 bluescreen left #parrot
09:20 bluescreen joined #parrot
09:24 jan joined #parrot
10:04 singalong left #parrot
10:17 dalek rakudo: f8e959c | nwellnhof++ | src/ops/perl6.ops:
10:17 dalek rakudo: Fix memory handling in rebless_subclass
10:17 dalek rakudo:
10:17 dalek rakudo: Se tt #1746 for details
10:17 dalek rakudo: review: http://github.com/rakudo/rakudo/commit/f​8e959c1f416a4615a72ce1e21ad2dd2535f4f47
10:22 aloha left #parrot
10:23 bacek left #parrot
12:26 ilbot2 joined #parrot
12:26 Topic for #parrot is now Parrot 2.7.0 "Australian King" Released! | parrot.org Log:  irclog.perlgeek.de/parrot/today | Nopaste: nopaste.snit.ch:8001 | close 20 tickets (-13 to go), merge outstanding branches; parrot-members: vote by 2PM PT Tues Aug 31
12:29 moritz joined #parrot
12:31 smash hello everyone
12:32 whiteknight good morning, smash
12:33 smash whiteknight: mornin'
12:40 whiteknight smash: the whole election thing is running very smoothly. You're doing a fantastic job with it
12:43 ash_ left #parrot
12:44 smash whiteknight: yes, and it ends today, thank you
12:47 whiteknight when do results get announced?
12:53 smash today i hope
12:54 smash (tomorrow the latest)
13:05 JimmyZ left #parrot
13:06 Coke ... why is the annual meeting overlapping with the #ps meeting?
13:06 Coke (was that intentional?)
13:17 smash Coke: yes it, so it gave a change to reming everyone on #ps of elections start and deadline
13:22 Coke ok. sadly that's a crappy time for me, so we'll see. ;)
13:23 robin-gvx joined #parrot
13:23 Coke is there anything to do outside of announce the vote tally?
13:28 uniejo left #parrot
13:35 whiteknight Coke: always a crappy time for me as well.
13:36 whiteknight I'll probably make the end of the normal #ps meeting, like always
13:48 smash Coke: not sure, but don't think so
13:50 nwellnhof joined #parrot
14:10 JimmyZ joined #parrot
14:19 fperrad left #parrot
14:23 davidfetter joined #parrot
14:24 luben_work joined #parrot
14:28 JimmyZ left #parrot
14:33 patspam joined #parrot
14:39 Paul_the_Greek joined #parrot
14:39 Paul_the_Greek G'day folks.
14:40 whiteknight good morning Paul_the_Greek
14:40 Paul_the_Greek Hey there, whiteknight
14:43 whiteknight Paul_the_Greek: excellent first few commits!
14:44 theory joined #parrot
14:46 Paul_the_Greek Thanks, whiteknight. I've got one more to go and then it's time to find something else to do. I have a couple of things to talk about during the #ps chat this afternoon.
14:46 whiteknight nice.
14:46 whiteknight hopefully you get the answers and direction that you need
14:47 Paul_the_Greek One has to do with the Boolean PMC, which might turn into a quasi-religious war. :D
14:48 davidfetter left #parrot
14:48 Paul_the_Greek I love religious wars, but I've learned to step carefully.
14:49 tcurtis joined #parrot
14:53 particle resizable boolean array packs booleans into bits, dunnit?
14:54 Paul_the_Greek I'm not sure. Let me check ...
14:54 particle wonder if that breaks any 'isa Integer' semantics
14:55 Paul_the_Greek fixedbooleanarray does.
14:55 patspam left #parrot
14:56 Paul_the_Greek So does resizeable.
14:56 Paul_the_Greek My argument is that providing things like add, subtract, increment, neg, modulus, etc., is just silly.
14:57 Paul_the_Greek But some HLLs (Perl?) may rely on it.
14:58 Paul_the_Greek chromatic's question is: If we stopped inheriting from Integer, could we make Boolean faster?
14:58 Coke while they may in practice, hll-specific autoboxing means they don't have to rely on the core for that.
15:00 Paul_the_Greek Does Perl use Boolean, or does it just use Integer?
15:01 Coke easy enough to ack the Rakudo source.
15:02 patspam joined #parrot
15:02 Paul_the_Greek The other thing is to check the Perl spec to see if any of those whacky operators can be applied to a boolean.
15:02 Coke 23:    boolproto = p6meta.'new_class'('Bool', 'parent'=>'parrot;Boolean Cool')
15:02 Coke Paul_the_Greek: no, that's rakudo's problem.
15:03 Paul_the_Greek Providing the whacky operators?
15:03 Coke I'm not sure what checking their operators means for the core parrot Boolean.
15:03 Coke 7
15:04 Coke that is, what problem are you trying to fix?
15:04 Coke if you want to find out if your changes break rakudo, there's an easy way to determine that.
15:06 Paul_the_Greek We're just wondering if Booleans could be more efficient. I have to take a look first. I'll bring this up this afternoon.
15:21 Paul_the_Greek left #parrot
15:22 ash_ joined #parrot
15:25 Coke ... then looking at the spec is probably the wrong place to look.
15:25 Coke if you want /parrot's/ Boolean to be fast, come up with a benchmark (in PIR) and callgrind it.
15:26 Coke the Perl 6 spec, I mean.
15:31 whiteknight I'm not convinced that we even need a boolean type, especially if it's just an inherited integer type
15:32 whiteknight I would be surprised if Boolean was used too often
15:32 whiteknight it's a terrible waste of memory space in any case: allocating an entire PMC header just to hold a single bit
15:33 moritz ... assuming that the caller knows that a boolean is returned, and not an int
15:33 Coke Mainly useful for HLLs, IMO.
15:33 moritz The db is set to default encoding LATIN1
15:33 moritz sorry, pasting fail
15:34 szbalint ouch
15:34 ash_ is there any way to hide the boolean in something else? like an INTVAL or a pointer? i mean you only need 1 bit to tell if its true or false and a pointer gives you at least 32 bits
15:34 szbalint regardless of paste failure, that's "ouch" moritz :)
15:35 whiteknight ash_: of course you can use an INTVAL. Most things that return a "boolean" value actually return INTVAL
15:35 whiteknight this is just the particular case of the Boolean PMC, which I suggest is unnecessary
15:37 whiteknight ResizableBooleanArray is actually pretty nice, since it compacts the bits together in a reasonable way
15:38 moritz szbalint: I don't see the "ouch". a function can rightfully return either an Int or a Bool, and mean different things
15:39 szbalint moritz: the latin1 thing...
15:39 szbalint ;-)
15:39 NotFound whiteknight: Have you seen my msg?
15:39 moritz szbalint: ah that. Yes...
15:39 moritz szbalint: http://www.perlmonks.org/?node_id=858107
15:40 moritz that's even more "ouch" :-)
15:40 whiteknight NotFound: yes. PLA has been through some major infrastructural changes. I'll update the plumage metadata when I get closer to a real release
15:40 NotFound whiteknight: ok
15:41 szbalint moritz: oh yeah.
15:49 cotto d'oh.  wrong tt number in this next commit
15:50 cotto fortunately I can just fix that before anyone pulls.
15:50 cotto svn--
15:51 dalek parrot: r48736 | cotto++ | trunk/compilers/opsc (2 files):
15:51 dalek parrot: make opsc less whitespace-sensitive when reading VERSION, fixing TT #1721 (plus a minor docs fix)
15:51 dalek parrot: review: http://trac.parrot.org/parrot/changeset/48736/
15:52 Topic for #parrot is now Parrot 2.7.0 "Australian King" Released! | parrot.org Log:  irclog.perlgeek.de/parrot/today | Nopaste: nopaste.snit.ch:8001 | close 20 tickets (-14 to go), merge outstanding branches; parrot-members: vote by 2PM PT Tues Aug 31
15:52 particle what's wrong with putting http:// in the title? i don't see that anymore
15:54 cotto It might have gotten garbled out.  xchat does that on occasion.
15:55 * cotto goes to wrk
16:00 dalek TT #1760 closed by cotto++: Spurious newline in version string
16:00 dalek TT #1760: http://trac.parrot.org/parrot/ticket/1760
16:01 ash_ left #parrot
16:02 ash_ joined #parrot
16:02 moritz cotto: maybe space safing
16:03 moritz *saving
16:07 ash__ joined #parrot
16:12 ash_ left #parrot
16:12 ash__ is now known as ash_
16:12 cotto_work moritz: it'll be shorter than yesterday ;]
16:35 Andy joined #parrot
16:36 janus left #parrot
16:42 nwellnhof left #parrot
16:43 janus joined #parrot
16:59 ruoso joined #parrot
17:15 sorear Coke: ping
17:16 Coke yes?
17:16 sorear you have a top running on feather1
17:16 sorear 132 minutes CPU time eaten
17:17 Coke thanks, killed.
17:25 Coke someone should test our build with 5.12.2-RC1
17:30 whiteknight 5.12.2-RC1?
17:31 Coke of perl, yes.
17:31 whiteknight oh
17:34 szbalint yeah, still hot on CPAN :)
17:35 szbalint or actually
17:35 szbalint Show me how you would provide a backwards compatibility layer for a
17:35 szbalint feature, and show me how code would be written so it worked fine in a
17:35 szbalint new perl and an old perl.
17:35 szbalint err
17:35 szbalint wrong paste sorry :)
17:35 szbalint http://fsck.com/~jesse/tmp/2010-08-30/bce29e12-3d​24-471a-af3a-c206c2fe56ee/perl-5.12.2-RC1.tar.bz2
17:37 Tene joined #parrot
18:06 mikehh_ joined #parrot
18:06 NotFound left #parrot
18:07 mikehh left #parrot
18:09 NotFound joined #parrot
18:29 nwellnhof joined #parrot
18:42 silug left #parrot
18:44 robin-gvx left #parrot
18:50 mikehh_ is now known as mikehh
18:50 mikehh opbots, names
18:57 cotto_work #ps in 93
19:20 allison joined #parrot
19:25 dalek TT #1746 closed by moritz++: A way to get Rakudo to segfault
19:25 dalek TT #1746: http://trac.parrot.org/parrot/ticket/1746
19:27 Topic for #parrot is now Parrot 2.7.0 "Australian King" Released! | parrot.org Log:  irclog.perlgeek.de/parrot/today | Nopaste: nopaste.snit.ch:8001 | close 20 tickets (-15 to go), merge outstanding branches; parrot-members: vote by 2PM PT Tues Aug 31
19:27 Coke I added masak as a dev on trac (but not submitter) so he can close tix.
19:27 Coke more ticket devs yay.
19:29 Coke odd. I voted when the polls opened, but I am allowed to vote again.
19:30 Coke but now when I vote again, it' says "cancel your vote", not the choices.
19:30 cotto_work Where's the poll?  I see nothing on parrot.org
19:31 Coke I have the links from the emal smash sent.
19:31 particle coke: i think as an admin you can modify your vote.  check drupal poll settings
19:31 particle actually, it'd be in admin | permissions
19:32 Coke particle: note that when I revote now, the option changes.
19:32 Coke I go to the vote: I'm asked to vote. I vote. I try to vote it agian, now it says 'you can ccancel your vote'...
19:32 Coke also, as an admin, if I click on votes, I see 2 votes by name, but the summary shows 12 votes.
19:33 Coke so, everyone, double check yer vote.
19:34 cotto_work I found it, though by a circuitous route.
19:35 * Coke hopes he didn't double vote.
19:35 sorear did we want to summon hachi to #parrot?
19:35 Coke I'm surprised there's only about a dozen votes.
19:35 Coke sorear: he's usually here lurking.
19:35 sorear not now
19:36 Coke you could check the logs. We were talking about booting purl in favor of aloha.
19:36 Coke but last I heard we weren't ready.
19:36 cotto_work getting there, but not yet
19:43 cotto_work dukeleto: ping
19:50 dukeleto cotto_work: pong
19:50 cotto_work dukeleto: barring other factors, how close to ready do you think the github clone of Parrot is for use as Parrot's official repo?
19:51 dukeleto cotto_work: it is pretty close. Do you mean official as in it will not have to mirror the svn repo any more?
19:52 cotto_work Yes.
19:52 dukeleto cotto_work: also, we still need to figure out how we want to setup to Parrot github org. i.e. does everyone with a CLA get to be part of the parrot github org (i.e. get a commit bit) ?
19:53 cotto_work Sure.  There are other questions.
19:53 dukeleto cotto_work: the parrot github clone is ready, it is mostly how we want to interact with it that is not totally defined
19:53 whiteknight dukeleto: have you figured out how to add people to the organization?
19:54 dukeleto whiteknight: i am pretty sure only pmichaud can do that, unless he gives me more permissions
19:54 dukeleto whiteknight: if he added me as an admin of the parrot org, then I could manage adding/removing users
19:57 kid51 joined #parrot
19:58 kid51 gsoc_threads branch now PASS make fulltest r48736 linux/i386
19:59 luben joined #parrot
19:59 luben_work left #parrot
20:00 cotto_work dukeleto: can you bring that up during #ps?  I have a pesky meeting at 2:00 that has a good chance of preventing me from asking it.
20:02 smash FYI, PaFo elections closing in 1 hour
20:02 smash if you haven't voted yet this is the time
20:05 whiteknight dukeleto: ah, I didn't realize you weren't the creator of the group
20:05 whiteknight dukeleto: it probably is a good idea if adminish parroteers are admins over there too
20:06 cotto_work +1 for whenever pmichaud's around next
20:07 Coke smash: even if you already voted, apparently now is also the time.
20:07 Coke I'm happy to not be an admin everywhere. ;)
20:09 GeJ Bonjour everyone.
20:10 whiteknight hello GeJ
20:10 smash Coke: :-)
20:10 dukeleto whiteknight: yes, i will ask pmichaud if he will make me an admin and then I can take care of that stuff
20:10 whiteknight dukeleto++
20:12 cotto_work particle++ for the spam cleanup
20:12 chromatic joined #parrot
20:13 whiteknight left #parrot
20:14 dukeleto msg pmichaud could you add me to the admin group of the parrot github repo, so that I can add other contributors? thanks!
20:14 purl Message for pmichaud stored.
20:14 Coke MEGO?
20:14 Coke purl, MEGO?
20:14 purl coke: i don't know
20:14 Coke util, MEGO?
20:15 Util "My eyes glaze over"
20:17 Util http://catb.org/jargon/html/M/MEGO.html
20:19 Coke MEGO is \my eyes glaze over
20:20 Coke purl?
20:20 purl Coke?
20:20 chromatic Where's the voting link?
20:21 Coke in email. moment.
20:21 Coke FOR last minute voting, see:
20:21 Coke Jerry Gay: http://www.parrot.org/content/elect​ion-jerry-gay-2010-2011-board-term
20:21 Coke Jim Keenan: http://www.parrot.org/content/elect​ion-jim-keenan-2010-2011-board-term
20:21 Coke Jonathan Leto: http://www.parrot.org/content/electio​n-jonathan-leto-2010-2011-board-term
20:21 Coke Andrew Whitworth:
20:21 Coke http://www.parrot.org/content/election​-andrew-whitworth-2010-2011-board-term
20:23 chromatic Thanks.
20:23 Util Log in to parrot.org first. 38 minutes to vote.
20:23 Paul_the_Greek joined #parrot
20:27 ash_ left #parrot
20:28 chromatic #ps in 2
20:31 sri left #parrot
20:35 Topic for #parrot is now Parrot 2.7.0 "Australian King" Released! | parrot.org Log:  irclog.perlgeek.de/parrot/today | Nopaste: nopaste.snit.ch:8001 | close 25 tickets (25 to go), merge outstanding branches; parrot-members: vote by 2PM PT Tues Aug 31
20:41 hercynium joined #parrot
20:42 patspam left #parrot
20:42 Coke it occurs to me that the "annual meeting" email doesnt' say where the meeting is.
20:42 Coke (and no one asked.)
20:46 cotto_work seen khairul
20:46 purl khairul was last seen on #parrot 8 days, 16 hours, 24 minutes and 30 seconds ago, saying: hio cotto. i'm slowly moving the code over to here, http://github.com/khairulsyamil/parrot-instrument  [Aug 23 04:21:36 2010]
20:46 aloha Sorry, I haven't seen khairul.
20:49 sorear seen darbelo
20:49 purl darbelo was last seen on #parrot 1 days, 20 hours, 57 minutes and 11 seconds ago, saying: whiteknight: Is svn tied to Trac or drupal's auth?  [Aug 29 23:52:30 2010]
20:49 aloha darbelo was last seen in #parrot 1 days 20 hours ago saying "whiteknight: Is svn tied to Trac or drupal's auth?".
20:50 patspam joined #parrot
20:51 particle Coke:  no one asked, so i didn't offer.  it's wherever i am, since i'm the only one attending it in seattle
20:52 silug joined #parrot
20:53 davidfetter joined #parrot
20:53 mikehh particle: do we have the meeting logged on irc or anything like that?
20:53 Util Coke, the attached .pdf says it the meeting is in Seattle.
20:54 allison mikehh: it's generally a virtual meeting
20:54 particle mikehh: the meeting will be gn #parrotsketch
20:55 allison mikehh: it just needs an "official location" for the records
20:55 particle s/gn/in/
20:55 mikehh alreeady there
21:01 whiteknight joined #parrot
21:04 smash particle: did we get ay votes by mail or in legal@ ?
21:05 particle smash: none
21:05 smash particle: thank you
21:06 dukeleto http://code.google.com/p/goo​gle-summer-of-code-2010-tpf/ <-- Place for GSoC students to upload their code, due by 10th September
21:10 Topic for #parrot is now Parrot 2.7.0 "Australian King" Released! | parrot.org Log:  irclog.perlgeek.de/parrot/today | Nopaste: nopaste.snit.ch:8001 | close 25 tickets (25 to go), merge outstanding branches; profile your favorite PIR for memory leaks with valgrind
21:11 Coke I'm out. good luck to everyone running.
21:14 patspam left #parrot
21:14 patspam joined #parrot
21:15 dalek parrot: r48737 | NotFound++ | trunk/t/src/extend.t:
21:15 dalek parrot: rearrange some extend tests to make its coverage reports more useful
21:15 dalek parrot: review: http://trac.parrot.org/parrot/changeset/48737/
21:24 GeJ Is there someone at the foundation that I could contact to verify whether my CLA was received or not?
21:25 whiteknight geJ: I thouht that happened a LONG time ago?
21:25 dukeleto GeJ: particle usually is the one to receive CLAs
21:25 nwellnhof yeah, it was particle who confirmed that my CLA was received
21:26 nwellnhof has anyone ever created a new branch with git-svn?
21:26 GeJ I sent it sometime last year if memory serves. I haven't been contacted yet about it to the best of my knowledge.
21:27 GeJ It's OK though. $work and $life got seriously in the way since then and I haven't been able to work on Parrot as much as I did back then.
21:28 GeJ I was just curious about the fate of my CLA.
21:28 GeJ Sea turtles are not efficient courriers, it seems.
21:29 nwellnhof i faxed my CLA. maybe you should try that ;)
21:29 whiteknight GeJ: what is your trac username?
21:29 Paul_the_Greek I mailed mine and it was received.
21:30 rurban joined #parrot
21:30 GeJ whiteknight: that would be geraud
21:31 whiteknight GeJ: okay, you arent a committer
21:31 whiteknight I seriosly thought you were
21:34 GeJ nwellnhof: I'll wait for someone to suggest my name in a future #parrotsketch. As I said, I'm not sure I can commit enough time to the project to be a valuable asset. If someone wants to sponsor me, that's fine.
21:34 chromatic I thought you were a committer too.
21:35 whiteknight I do remember the issue being raised at #ps, sometime in the distant past
21:36 nwellnhof GeJ: was that related to my branch question?
21:37 GeJ nwellnhof: nope, it was about your suggestion to fax my CLA.
21:37 nwellnhof GeJ: ah, ok
21:37 Paul_the_Greek left #parrot
21:39 nwellnhof my branch question was completely unrelated, but i just repeat it: has anyone ever created a new branch with git-svn?
21:39 nwellnhof i tried it with the -n (dry run) flag and got some strange warnings.
21:40 chromatic I have.  It worked fine for me.
21:40 nwellnhof the warnings were about other branches. gsoc_threads, i think.
21:41 nwellnhof so i didn't dare to go on and run 'git svn branch' without the -n flag
21:41 GeJ parrotsketch of 2009-01-13, wow that's older than I thought.
21:42 chromatic git svn fetch; git co master; git svn rebase first and all should be good.
21:46 nwellnhof yeah, it seems to work when i'm on master
21:47 kid51 left #parrot
21:48 dalek parrot: r48738 | nwellnhof++ | branches/charset_massacre:
21:48 dalek parrot: Merge charsets into encodings
21:48 dalek parrot: review: http://trac.parrot.org/parrot/changeset/48738/
21:49 smash particle: you can revoke my admin privileges on parrot.org
21:49 particle smash++ for a fabulous job running this year's elections
21:50 whiteknight smash++
21:51 NotFound smash++
21:51 mikehh smash++
22:05 ruoso left #parrot
22:16 japhb left #parrot
22:16 cotto_work khairul's gsoc project is uploaded
22:17 cotto_work msg khairul I've uploaded a tarball of the pencils down state of your code to http://code.google.com/p/google-sum​mer-of-code-2010-tpf/downloads/list for gsoc.
22:17 purl Message for khairul stored.
22:25 Coke smash+=3
22:25 patspam left #parrot
22:29 patspam joined #parrot
22:29 davidfetter left #parrot
22:39 dalek parrot: r48739 | NotFound++ | trunk/t/pmc/stringiterator.t:
22:41 dalek parrot: some more StringIterator tests
22:41 dalek parrot: review: http://trac.parrot.org/parrot/changeset/48739/
22:41 dalek parrot: r48740 | nwellnhof++ | branches/charset_massacre (66 files):
22:41 dalek parrot: String refactoring, merge charsets into encodings
22:41 dalek parrot: review: http://trac.parrot.org/parrot/changeset/48740/
22:43 aloha left #parrot
22:43 nwellnhof chromatic: do you know how i can make an existing git branch follow an svn branch?
22:44 nwellnhof i tried 'git branch --set-upstream' but somehow it doesn't work
22:44 nwellnhof creating a new branch with --track works fine, though
22:45 aloha joined #parrot
22:46 nwellnhof and if anyone wants to have a look at my charset/encoding changes...
22:46 dalek rakudo: 2ac5feb | colomon++ | src/core/Cool-str.pm:
22:46 dalek rakudo: Have Cool.chomp return a proper Str.
22:47 dalek rakudo: review: http://github.com/rakudo/rakudo/commit/2​ac5feb2b9e6b8791a4e5e2e71a86536df24ad59
22:47 nwellnhof i just created a new branch: http://trac.parrot.org/parrot/br​owser/branches/charset_massacre
22:47 nwellnhof (name inspired by gc_massacre)
22:49 nwellnhof 74 files changed, 4596 insertions(+), 7378 deletions(-)
22:50 nwellnhof 70 files changed, 4027 insertions(+), 6809 deletions(-) (with git's -M option)
22:50 cotto_work Not bad for less than an hour's work.  Are you another magical coding robot?
22:52 chromatic nwellnhof, I'm sure there's a way, but I'd likely use --track and cherry pick anything important.
22:52 nwellnhof That's the work of last week.
22:52 nwellnhof chromatic: yeah, that's what i did
22:53 nwellnhof cotto: but i already have some interesting patches on top of that
22:55 nwellnhof cotto: the whole patch also looks bigger than it is. it was mostly moving existing code around.
22:55 chromatic What's the goal of this branch?
22:56 nwellnhof chromatic: http://lists.parrot.org/pipermail/​parrot-dev/2010-August/004633.html
22:57 chromatic Thanks.  Full speed ahead.
22:58 chromatic seen nothingmuch
22:58 purl nothingmuch was last seen on #moose 1 days, 4 hours, 35 minutes and 41 seconds ago, saying: gphat: https://sites.google.com/a/chromium.o​rg/dev/developers/design-documents/gp​u-accelerated-compositing-in-chrome  [Aug 30 18:22:26 2010]
22:59 aloha Sorry, I haven't seen nothingmuch.
23:03 cotto_work nwellnhof++ for tackling that
23:05 cotto_work We generally like branches that have "massacre" in their names.
23:06 nwellnhof massacre branches FTW
23:06 whiteknight though it's a bit of a shame that so many of our systems need to be massacred
23:06 cotto_work I'm looking forward to imcc_massacre.
23:08 nwellnhof though for my branch, "masscare" is a bit of an exaggeration
23:10 whiteknight yeah, it's too bad that charsets_and_encodings_smoosh_together doesn't flow off the tongue as well
23:12 patspam left #parrot
23:15 cotto_work How big a project is it?
23:17 smash left #parrot
23:17 kid51 joined #parrot
23:20 nwellnhof cotto: you mean the charset/encoding merging?
23:22 cotto_work yes
23:23 whiteknight purl messages
23:23 cotto_work You have 34200 messages waiting.
23:23 whiteknight NotFound: ping
23:23 whiteknight cotto_work: not a bad estimate!
23:23 nwellnhof cotto: it's all done in my new branch
23:24 nwellnhof it's not a WIP branch
23:24 nwellnhof mainly for review
23:24 nwellnhof there are a few user visible changes
23:25 nwellnhof i think it's best to write about them on parrot-dev
23:25 cotto_work +1
23:25 purl 1
23:26 nwellnhof the API of the bytebuffer PMC has changed a little
23:28 nwellnhof i also added backward compatibility for converting to the old "unicode" charset
23:28 nwellnhof but the resulting strings have a "utf8" charset
23:29 nwellnhof encoding actually, but charset and encoding are synonyms now
23:29 nwellnhof just have look at my changes to the test suite
23:35 rurban left #parrot
23:35 whiteknight purl msg NotFound: I've been looking around for examples, but I have no idea how to cange parrot_setup to nqp_setup in PLA. Any ideas?
23:35 purl Message for notfound stored.
23:37 cotto_work msg khairul I'd like to delete the gsoc_instrument branch if you're done with it.
23:37 purl Message for khairul stored.
23:38 nwellnhof chromatic: i think we can remove the gc_threshold_tuning branch
23:42 chromatic Agreed.
23:42 Andy left #parrot
23:47 jsut joined #parrot
23:51 jsut_ left #parrot

| Channels | #parrot index | Today | | Search | Google Search | Plain-Text | summary

Parrot | source cross referenced