Camelia, the Perl 6 bug

IRC log for #parrot, 2010-09-24

Parrot | source cross referenced

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

All times shown according to UTC.

Time Nick Message
00:03 chromatic GB allocation sizes seem big.
00:14 mikehh All tests PASS (pre/post-config, make corevm/make coretest, smoke (#204) fulltest) at r49279 - Ubuntu 10.10 beta amd64 (g++-4.5 with --optimize)
00:14 aloha joined #parrot
00:15 bacek_at_work aloha, insult bacek
00:15 bacek_at_work aloha, insult bacek
00:16 bacek_at_work aloha, karma aloha
00:16 aloha bacek_at_work: aloha has karma of 0.
00:16 bacek_at_work aloha++
00:16 aloha bacek_at_work: Thanks!
00:16 bacek_at_work aloha, karma aloha
00:16 aloha bacek_at_work: aloha has karma of 1.
00:16 bacek_at_work ok, it works
00:16 bacek_at_work I moved aloha to my box at $work
00:16 bacek_at_work It should be little bit more reliable
00:23 mikehh bacek_at_work: you all should set it up on parrot.org
00:23 bacek_at_work mikehh, probably. But I don't have access to feather
00:24 mikehh bacek_at_work: you need to get the latest karma details from moritz
00:25 bacek_at_work mikehh, yes. But first I want to rewrite Karma module from Bot::BasicBot.
00:25 kid51 is now known as kid51_at_dinnner
00:25 sorear bacek_at_work: Juerd hands out feather accounts like there's no tomorrow
00:26 sorear you can even ask for a sudoers bit and you'll get it
00:26 bacek_at_work sorear, :)
00:26 mikehh bacek_at_work: 'k, you got it on github or anything?
00:26 bacek_at_work mikehh, nope. It's on cpan
00:26 bacek_at_work but aloha on github.
00:27 sorear (however, Juerd says he doesn't want anyone else using sudo adduser for paperwork's sake)
00:30 patspam left #parrot
00:32 whiteknight eventually I would like to set up a server like feather for parrot
00:33 whiteknight I dont think the parrot.org VM has enough muscle
00:34 dngor_ joined #parrot
00:35 theory left #parrot
00:35 dngor left #parrot
00:59 dngor_ is now known as dngor
01:00 theory joined #parrot
01:03 whiteknight left #parrot
01:08 mikehh hmmn - the last email I got from parrot was 12 hours ago - r49275 (there have been at least 4 commits since then and I am sure other mail)
01:09 chromatic Same here.
01:15 dalek TT #1801 created by bacek++: Properly merge pools in GC MS2 when child interpreter destroyed.
01:15 dalek TT #1801: http://trac.parrot.org/parrot/ticket/1801
01:27 dngor left #parrot
01:27 dngor joined #parrot
01:31 dalek parrot: r49280 | bacek++ | trunk/src/gc/gc_ms2.c:
01:31 dalek parrot: Uncomment soon-to-be-used code in gc_ms2_finalize
01:31 dalek parrot: review: http://trac.parrot.org/parrot/changeset/49280/
01:31 dalek parrot: r49281 | bacek++ | trunk/t/op/gc-non-recursive.t:
01:31 dalek parrot: Add test for TT#1723
01:31 dalek parrot: review: http://trac.parrot.org/parrot/changeset/49281/
01:32 dalek TT #1723 closed by bacek++: [bug] linked lists result in segfaults
01:32 dalek TT #1723: http://trac.parrot.org/parrot/ticket/1723
01:32 dalek TT #1753 closed by bacek++: PMC attribute blocks not aligned
01:32 dalek TT #1753: http://trac.parrot.org/parrot/ticket/1753
01:32 Topic for #parrot is now Parrot 2.8.0 released | parrot.org Log: irclog.perlgeek.de/parrot/today | Nopaste: nopaste.snit.ch:8001 | close 25 tickets; remove deprecated items (especially CodeString);
01:43 kid51_at_dinnner msg Coke how do we file bug tickets with osuosl.org re mail not functioning?
01:43 aloha OK. I'll deliver the message.
01:51 kid51_at_dinnner is now known as kid51
01:51 kid51 I filed ticket at support@osuosl.org
02:20 kid51 left #parrot
02:35 janus left #parrot
02:41 janus joined #parrot
02:53 jmcadams left #parrot
02:53 theory left #parrot
03:29 plobsing why does PCC handle Key PMC arguments specially?
03:31 chromatic Where?
03:31 plobsing clone_key_arg
03:32 plobsing src/call/args.c:389
03:32 plobsing passing an argument of type key clones the key (which replaces the register components of the clone with values from the caller's context)
03:33 plobsing seems like a nifty shortcut, but makes working with keys harder (eg: compilers)
03:36 ash_ left #parrot
03:40 bacek_at_work plobsing, Key part can reference something like "PMC register". After call we'll have different CallContext. So, this "PMC register" have to be propagated to new CallContext
03:40 bacek_at_work from original one
03:41 plobsing yes, but a Key accessible from a PMC register isn't associated with a context. it is a object, nothing more.
03:41 plobsing keys associated with contexts only occur as arguments to special keyed ops
03:42 plobsing what I'm running into: .sub 'crash'; $P0 = new ['Key']; $P0.'set_register'(1, 4); sub_call($P0) .end
03:42 bacek_at_work Key.get_foo can access PMC within particular CallContext
03:43 plobsing OK, then how are you supposed to work with keys as generic objects?
03:43 bacek_at_work they are not "generic" enough.
03:44 bacek_at_work Personally, I hate them
03:45 bacek_at_work http://trac.parrot.org/parrot/ticket/903
03:45 plobsing giving them carte blanche to access the callframe screws up any optimization you could hope to acheive.
03:45 plobsing *at the pir lel
03:45 plobsing s/lel/level/
03:48 bacek_at_work agreed
03:50 plobsing bacek_at_work: I'm with you on getting rid of them outright. But short of that, I'd like them to at least be a lot less magical.
03:50 bacek_at_work plobsing, heh. I did a lot of cleanups making Keys less magical already. Last step is to kill them all.
03:52 chromatic If we do anything with Keys, can we also consider having *one* way to navigate nested NameSpaces?
03:52 chromatic If we're going to use polymorphism in Parrot, we should pretend we understand it.
03:57 plobsing chromatic: how many ways do we have to do it now?
03:59 chromatic Keys, RSAs, FSAs, STRINGs, String PMCs, and I think that's it.
04:02 plobsing but then how would I use a different way per weekday?
04:03 dalek parrot: r49282 | plobsing++ | trunk (3 files):
04:03 dalek parrot: modify Key.get_repr appropriately and use it to de-dup keys in IMCC
04:03 dalek parrot: this acheives the PBC size reduction that should have been had with typesafe_consttable
04:03 dalek parrot: review: http://trac.parrot.org/parrot/changeset/49282/
04:07 chromatic Let's see how much smaller and what that does to Rakudo.
04:08 plobsing (30465456 - 30272480)/30465456
04:08 aloha 0.00633425608334896
04:08 plobsing 0.6% smaller
04:09 plobsing not much, but then I still have a bunch of stuff that should fall out from that branch
04:11 chromatic How much smaller can it go>
04:11 bacek_at_work 0 bytes
04:12 plobsing I'm seeing if I can rip out the fixup table entirely ATM
04:12 bacek_at_work best PBC doesn't exists, but everyone is happy
04:13 chromatic bacek_at_work, what's your next project?
04:19 chromatic Hah, want to segfault Parrot?  Call Parrot_pmc_new() with a really big base_type.
04:19 chromatic or a negative one
04:20 plobsing chromatic: how did you find that one?
04:23 bacek_at_work chromatic, GenGC
04:24 bacek_at_work based on SPJ's "vtable override" idea
04:34 Andy joined #parrot
04:42 bacek_at_work left #parrot
04:45 bacek_at_work joined #parrot
05:09 janus left #parrot
05:09 jan left #parrot
05:09 Khisanth left #parrot
05:09 slavorgn left #parrot
05:09 betterworld left #parrot
05:09 cotto left #parrot
05:09 szbalint left #parrot
05:09 TiMBuS_ left #parrot
05:09 sjn_ left #parrot
05:09 p6eval left #parrot
05:09 slavorg left #parrot
05:09 edenc left #parrot
05:09 dngor left #parrot
05:09 mikehh left #parrot
05:09 silug left #parrot
05:09 plobsing left #parrot
05:09 sjn left #parrot
05:09 hatseflats left #parrot
05:09 AzureStone left #parrot
05:09 estrabd left #parrot
05:09 dalek left #parrot
05:09 cognominal left #parrot
05:09 cosimo_ left #parrot
05:09 athomaso1 left #parrot
05:09 krunen left #parrot
05:09 PerlJam left #parrot
05:09 dukeleto left #parrot
05:09 sorear left #parrot
05:09 pmichaud left #parrot
05:09 rblackwe left #parrot
05:09 chromatic left #parrot
05:09 pjcj left #parrot
05:09 PacoLinux left #parrot
05:09 esskar left #parrot
05:09 preflex left #parrot
05:09 x3nU left #parrot
05:09 gottreu left #parrot
05:09 NotFound left #parrot
05:09 ashleyde1 left #parrot
05:09 cxreg left #parrot
05:09 GeJ left #parrot
05:09 baest left #parrot
05:09 jhelwig left #parrot
05:09 Util left #parrot
05:09 treed left #parrot
05:09 spinclad left #parrot
05:09 knewt left #parrot
05:20 Andy left #parrot
05:55 krunen joined #parrot
06:01 ttbot left #parrot
06:01 ttbot joined #parrot
06:01 janus joined #parrot
06:01 dngor joined #parrot
06:01 mikehh joined #parrot
06:01 jan joined #parrot
06:01 silug joined #parrot
06:01 Khisanth joined #parrot
06:01 sjn joined #parrot
06:01 hatseflats joined #parrot
06:01 pjcj joined #parrot
06:01 PacoLinux joined #parrot
06:01 esskar joined #parrot
06:01 AzureStone joined #parrot
06:01 slavorgn joined #parrot
06:01 preflex joined #parrot
06:01 estrabd joined #parrot
06:01 dalek joined #parrot
06:01 betterworld joined #parrot
06:01 x3nU joined #parrot
06:01 cotto joined #parrot
06:01 ashleyde1 joined #parrot
06:01 gottreu joined #parrot
06:01 NotFound joined #parrot
06:01 cognominal joined #parrot
06:01 cosimo_ joined #parrot
06:01 cxreg joined #parrot
06:01 GeJ joined #parrot
06:01 athomaso1 joined #parrot
06:01 szbalint joined #parrot
06:01 TiMBuS_ joined #parrot
06:01 sjn_ joined #parrot
06:01 baest joined #parrot
06:01 p6eval joined #parrot
06:01 slavorg joined #parrot
06:01 edenc joined #parrot
06:01 PerlJam joined #parrot
06:01 dukeleto joined #parrot
06:01 sorear joined #parrot
06:01 pmichaud joined #parrot
06:01 jhelwig joined #parrot
06:01 Util joined #parrot
06:01 treed joined #parrot
06:01 rblackwe joined #parrot
06:01 knewt joined #parrot
06:01 spinclad joined #parrot
06:02 dalek parrot: r49283 | bacek++ | trunk/src/gc/fixed_allocator.c:
06:02 dalek parrot: Update POD little bit.
06:02 dalek parrot: review: http://trac.parrot.org/parrot/changeset/49283/
06:16 ttbot Parrot trunk/ r49283 MSWin32-x86-multi-thread make error http://tt.taptinder.org/file/cmdout/402378.txt ( http://tt.taptinder.org//bui​ldstatus/pr-Parrot/rp-trunk/ )
06:20 jsut_ joined #parrot
06:21 luben_ joined #parrot
06:21 luben_ left #parrot
06:25 jsut left #parrot
06:29 jan left #parrot
06:32 dalek roast: 885149f | moritz++ | S02-builtin_data_types/bool.t:
06:32 dalek roast: [bool.t] test stringification more precisely
06:32 dalek roast: review: http://github.com/perl6/roast/commit/88​5149f5b0de839f939b557c488db6ccd409908c
06:35 dalek rakudo: 5e7b432 | moritz++ | build/PARROT_REVISION:
06:35 dalek rakudo: bump PARROT_REVISION for testing
06:35 dalek rakudo: review: http://github.com/rakudo/rakudo/commit/5​e7b4321d00eab5d7ca852075286ad731d256feb
07:01 contingencyplan left #parrot
07:05 luben_work joined #parrot
07:07 mikehh Still no emails from parrot
07:18 PacoLinux left #parrot
07:20 M_o_C joined #parrot
07:26 p6eval left #parrot
07:27 p6eval joined #parrot
07:27 cotto left #parrot
07:28 cotto joined #parrot
07:37 M_o_C left #parrot
07:38 AzureStone left #parrot
07:39 AzureStone joined #parrot
07:42 dngor left #parrot
07:42 dngor joined #parrot
07:42 gottreu left #parrot
07:42 gottreu joined #parrot
07:52 bacek_at_work msg plobsing Looks like you broke win32 build in r49276. Check http://tt.taptinder.org/bui​ldstatus/pr-Parrot/rp-trunk
07:52 aloha OK. I'll deliver the message.
08:01 fperrad joined #parrot
08:06 cognominal left #parrot
08:26 bacek joined #parrot
08:34 dalek parrot: r49284 | mikehh++ | trunk/MANIFEST:
08:34 dalek parrot: re-generate MANIFEST
08:34 dalek parrot: review: http://trac.parrot.org/parrot/changeset/49284/
08:39 cognominal joined #parrot
08:51 dalek parrot: r49285 | mikehh++ | trunk/t/op/gc-non-recursive.t:
08:51 dalek parrot: add svn properties and pir coda
08:51 dalek parrot: review: http://trac.parrot.org/parrot/changeset/49285/
08:53 tadzik joined #parrot
09:02 bacek mikehh, ping
09:03 bacek msg mikehh Cheap karma for removing old "gc" branches :)
09:03 aloha OK. I'll deliver the message.
09:05 mikehh bacek: which ones are old now?
09:05 moritz all? :-)
09:06 sorear bacek: 04:06 < Juerd> sorear: And feathers goals specifically include Parrot as well as Perl 6.
09:06 sorear bacek: go ask him for the aloha acct
09:08 dalek parrot: r49286 | bacek++ | branches/generational_gc:
09:08 dalek parrot: Branch for implementing Generational GC
09:08 dalek parrot: review: http://trac.parrot.org/parrot/changeset/49286/
09:08 dalek parrot: r49287 | mikehh++ | trunk/compilers/imcc/pbc.c:
09:08 dalek parrot: add a cast to get g++ to build
09:08 dalek parrot: review: http://trac.parrot.org/parrot/changeset/49287/
09:08 mikehh bacek we have gc_massacre, gc_massacre_no_constant and I take it generational_gc is new?
09:09 mikehh **** still NOT getting emails from parrot
09:22 fperrad left #parrot
09:28 tadzik left #parrot
09:32 mikehh All tests PASS (pre/post-config, make corevm/make coretest, smoke (#208) fulltest) at r49287 - Ubuntu 10.10 beta amd64 (g++-4.5 with --optimize)
09:40 bacek_mobile joined #parrot
09:40 bacek_mobile mikehh, yeah, gengc is new :)
09:41 bacek_mobile sorear, will di
09:41 bacek_mobile Do
09:47 lucian joined #parrot
09:49 lucian left #parrot
09:58 dalek parrot: r49288 | mikehh++ | branches/gc_massacre:
09:58 dalek parrot: remove gc_massacre branch (now in trunk)
09:58 dalek parrot: review: http://trac.parrot.org/parrot/changeset/49288/
09:58 dalek parrot: r49289 | mikehh++ | branches/gc_massacre_no_constant:
09:58 dalek parrot: remove gc_massacre_no_constant branch (old)
09:59 dalek parrot: review: http://trac.parrot.org/parrot/changeset/49289/
10:11 fperrad joined #parrot
10:15 bacek_mobile mikehh, string_gc* branches too :)
10:16 mikehh bacek_mobile: 'k
10:30 mikehh particle: ping
10:32 dalek parrot: r49290 | mikehh++ | branches/string_gc_encapsulate:
10:32 dalek parrot: remove string_gc_encapsulate branch
10:32 dalek parrot: review: http://trac.parrot.org/parrot/changeset/49290/
10:32 dalek parrot: r49291 | mikehh++ | branches/string_gc_split:
10:32 dalek parrot: remove string_gc_split branch
10:32 dalek parrot: review: http://trac.parrot.org/parrot/changeset/49291/
10:33 tadzik joined #parrot
11:17 whiteknight joined #parrot
11:23 dalek parrot: r49292 | nwellnhof++ | branches/string_macros/src/string/encoding/null.c:
11:23 dalek parrot: [str] Fix ASSERT_ARGS and non-optimized build
11:23 dalek parrot: review: http://trac.parrot.org/parrot/changeset/49292/
11:23 dalek parrot: r49293 | nwellnhof++ | branches/string_macros/src/string/api.c:
11:23 dalek parrot: [str] Change check for NULL keys in string API
11:23 dalek parrot: review: http://trac.parrot.org/parrot/changeset/49293/
11:23 dalek parrot: r49294 | nwellnhof++ | branches/string_macros/src/hash.c:
11:23 dalek parrot: [src] Make key_hash_STRING parameter const
11:23 dalek parrot: review: http://trac.parrot.org/parrot/changeset/49294/
11:27 whiteknight good morning, #parrot
11:35 moritz good morning, monochromatic knight!
11:36 mikehh hi whiteknight
11:38 mikehh whiteknight: BTW don't seem to by getting emails from partrot (for about 24 hours)
11:38 mikehh whiteknight: I emailed particle but not sure what else to do
11:41 dalek parrot: r49295 | mikehh++ | failed to fetch changeset:
11:41 dalek parrot: [html_cleanup] merge from trunk at r49291
11:41 dalek parrot: review: http://trac.parrot.org/parrot/changeset/49295/
11:50 contingencyplan joined #parrot
12:03 whiteknight mikehh: Yeah, kid51 already send an email to the OSU folks
12:03 whiteknight no reply yet
12:04 moritz maybe their mail setup doesn't work either :-)
12:09 Coke left #parrot
12:16 patspam joined #parrot
12:21 bacek left #parrot
12:22 ruoso joined #parrot
12:28 bluescreen joined #parrot
13:07 dalek parrot: r49296 | mikehh++ | failed to fetch changeset:
13:07 dalek parrot: html_cleanup - some files did not get deleted properly in merge (and property changes)
13:07 dalek parrot: review: http://trac.parrot.org/parrot/changeset/49296/
13:19 Patterner left #parrot
13:20 contingencyplan left #parrot
13:22 contingencyplan joined #parrot
13:24 dalek parrot: r49297 | mikehh++ | branches/html_cleanup/MANIFEST:
13:24 dalek parrot: [html_cleanup] re-generate MANIFEST
13:24 dalek parrot: review: http://trac.parrot.org/parrot/changeset/49297/
13:29 Psyche^ joined #parrot
13:29 Psyche^ is now known as Patterner
13:30 PacoLinux joined #parrot
13:55 PacoLinux left #parrot
13:55 ash_ joined #parrot
14:06 frodwith joined #parrot
14:14 PacoLinux joined #parrot
14:23 arnsholt left #parrot
14:46 theory joined #parrot
15:05 bacek_mobile left #parrot
15:05 jan joined #parrot
15:41 patspam left #parrot
15:42 patspam joined #parrot
16:05 cotto left #parrot
16:05 patspam1 joined #parrot
16:08 patspam left #parrot
16:13 Andy joined #parrot
16:18 chromatic joined #parrot
16:30 cotto joined #parrot
16:43 dip left #parrot
16:44 dalek parrot: r49298 | nwellnhof++ | trunk/src/io/buffer.c:
16:44 dalek parrot: [io] Fix GC-related bug in Parrot_io_read_buffer
16:44 dalek parrot: If Parrot_str_new_init triggers a GC run, out_buf might become invalid.
16:44 dalek parrot: review: http://trac.parrot.org/parrot/changeset/49298/
16:50 * dukeleto waves hello
16:51 * moritz particles hello
16:51 * dukeleto corpuscles hello
16:52 * dukeleto prepares the eigenmodes of the NLS
16:53 * moritz prepares the boundary conditions
16:53 * moritz looks at the emerging solitions
16:56 dalek roast: ef86bef | moritz++ | S03-se (10 files):
16:56 dalek roast: [sequence] is the new series
16:56 dalek roast: review: http://github.com/perl6/roast/commit/ef​86bef352d19689ee4eee980ffc976fbedca9ba
16:57 kid51 joined #parrot
16:57 kid51 Update on mail situation
16:58 kid51 Oregon State U Open Source Lab has responded to my RT filing but does not yet have a diagnosis of the problem (as of 1647 UTC Sep 24).
16:59 kid51 They're CC-ing whiteknight on the RTs, so he can also provide details as they become available.
16:59 cotto whiteknight++ for the blog post about bringing parrot to maturity
17:01 cotto chromatic++ for the typically pragmatic response
17:02 mikehh just started to get some emails now
17:02 * cotto too
17:03 kid51 Yes.  Google Groups just picked up a test message I posted 15+ hours ago.
17:04 dukeleto moritz: did you mean solutions or solitons? mmmmm, solitons.
17:05 moritz dukeleto: solitions
17:05 moritz erm
17:05 moritz solitons
17:05 * moritz type-challenged :-)
17:05 moritz the shape-preseving wave packets
17:05 kid51 is that why we need type-checking?
17:05 moritz yes
17:06 dukeleto moritz: i studied solitons in grad skool, I know them well :)
17:07 kid51 It also looks like lists.parrot.org archives are starting to pick up messages again (have checked parrot-dev and parrot-tickets)
17:07 moritz dukeleto: I encountered them while reading stuff for my PHD; haven't really worked with them though
17:07 dukeleto moritz: http://leto.net/math/soliton/ <-- old, but has pretty pictures
17:07 dukeleto moritz: if you ever want to nerd out about solutions to nonlinear wave equations, let me know :)
17:08 moritz dukeleto: :-)
17:08 sjn left #parrot
17:08 kid51 is now known as kid51_to_lunch
17:09 sjn joined #parrot
17:09 dalek rakudo: c94bfe1 | moritz++ | src/Perl6/Grammar.pm:
17:09 dalek rakudo: parse require
17:09 dalek rakudo: review: http://github.com/rakudo/rakudo/commit/c​94bfe1e40619461b738830735cc59dc7d41e005
17:09 dalek rakudo: b3efa7a | moritz++ | src/Perl6/Actions.pm:
17:09 dalek rakudo: implement argumentless form of require
17:09 dalek rakudo: review: http://github.com/rakudo/rakudo/commit/b​3efa7a74647512f7ed9b648fd76cfea43b25771
17:09 dalek rakudo: 7b21bf1 | moritz++ | src/Perl6/ (2 files):
17:10 dalek rakudo: implement require EXPR;
17:10 dalek rakudo: review: http://github.com/rakudo/rakudo/commit/7​b21bf137f8c1592f0edd98fdd4df4e91bd73162
17:10 dalek rakudo: 82ebb54 | moritz++ | t/spectest.data:
17:10 dalek rakudo: [t/spectest.data] track file name changes in roast
17:10 dalek rakudo: review: http://github.com/rakudo/rakudo/commit/8​2ebb545e88c8b754b35f31b59002efdd4d2d7fd
17:11 necrolyte joined #parrot
17:14 tadzik yay
17:16 davidfetter joined #parrot
17:16 dalek parrot: r49299 | fperrad++ | trunk/compilers/opsc/src/Ops/Emitter.pm:
17:16 dalek parrot: typo
17:16 dalek parrot: review: http://trac.parrot.org/parrot/changeset/49299/
17:18 whiteknight solitons? I haven't heard that word since grad school
17:19 whiteknight and good riddance!
17:22 dukeleto whiteknight: hey now! They are fun. ::pouts::
17:23 whiteknight I can't remember all the specifics about them. I think we were doing fourier decompositions of various things into solitons and wavelets, and all sorts of other stuff
17:23 whiteknight it's great if you're a professor trying to decrease the average grade of the class :)
17:27 davidfetter left #parrot
17:29 moritz well, in the field where I "work" they actually use optical solitons in the wild
17:29 moritz for transmitting data
17:33 cotto dukeleto, ping
17:34 cotto msg dukeleto Can you add http://www.postbin.org/12dx3gx as a post-receive url for parrot/parrot on github?
17:34 aloha OK. I'll deliver the message.
17:34 cotto seen purl
17:34 aloha Sorry, I haven't seen purl.
17:34 cotto karma svn
17:34 aloha svn has karma of 0.
17:34 cotto that's lta
17:35 atrodo way too high
17:35 patspam joined #parrot
17:35 cotto yeah
17:38 necrolyte left #parrot
17:39 patspam1 left #parrot
17:40 whiteknight Parrot mailing lists should be back up now, and OSUOSL has put in a check to nagios to prevent it from happening again
17:40 cotto what happened?
17:40 whiteknight somebody did a restart on mailman, which brought it down but didn't bring it back up
17:41 whiteknight added a check that mailman is actually running in nagios
17:42 dalek roast: e9774db | moritz++ | S11-modules/require.t:
17:42 dalek roast: basic tests for require
17:42 dalek roast: review: http://github.com/perl6/roast/commit/e9​774db45b4d2dd9ecab08a5ba97c374b935a8cf
17:43 dukeleto cotto: pong
17:43 dukeleto svn--
17:43 dukeleto karma svn
17:43 aloha svn has karma of -1.
17:43 cotto dukeleto, see alohaogram
17:44 dukeleto cotto: postbin looks cool
17:44 cotto that's for seeing what data github will throw at the post-receive url
17:44 cotto I found it on github's wiki: http://help.github.com/testing-webhooks/
17:45 cotto It's funny how things sometimes become much easier when you put a tiny bit of effort into them.
17:46 dukeleto cotto: look at the link now, i did a test payload
17:46 cotto wheee
17:46 dalek rakudo: 107cc16 | moritz++ | t/spectest.data:
17:46 dalek rakudo: run tests for require
17:46 dalek rakudo: review: http://github.com/rakudo/rakudo/commit/1​07cc160ce9f4863cce87a4c0fc99336bc3a302c
17:46 dalek rakudo: 13c8a23 | moritz++ | docs/ChangeLog:
17:46 dalek rakudo: [docs] update Changelog with require and Bool stringification
17:46 dalek rakudo: review: http://github.com/rakudo/rakudo/commit/1​3c8a23c3ab39dfa97d6c03fa2e51d203934c684
17:46 dukeleto cotto: to get full commit diffs, we will use the github api for the diff of each sha1 we see in the post-receive hook
17:47 moritz dukeleto: dalek already has code for that
17:47 cotto That appears to be the best approach, sort of getting github to change their commit emails.
17:47 cotto s/sort/short/
17:47 moritz or at least for getting the file names... not sure
17:48 dukeleto moritz: we need the full commit diffs to send to parrot-commits
17:49 dukeleto moritz: does any perl6 repo currently send commit emails anywhere, that include full commit diffs?
17:49 cotto I'm glad you're committed to making that happen.
17:49 dukeleto cotto: curl http://github.com/api/v2/json/co​mmits/show/parrot/parrot/master
17:50 dukeleto cotto: we basically just ask for a URL, and we get the diff inside some JSON
17:50 moritz dukeleto: no; we'd like to have that perl6/specs, but we don't yet
17:51 dukeleto moritz: ok, i am pretty sure the code that we write will be very general, so you can add it to all the perl6 repos if you like
17:51 dukeleto "we" meaning "me"
17:51 dukeleto cotto: our post-receive hook needs to live on parrot.org, right?
17:51 moritz dukeleto: that would be awesome
17:52 dukeleto cotto: i guess it could live on feather, but parrot.org seems better
17:52 cotto dukeleto, they need to live on the trac server
17:52 cotto they're part of the github trac plugin
17:53 cotto unless I'm misunderstanding you
17:56 dukeleto cotto: the post-receive script can live anywhere. it is the URL that gets hit by github, that sends an email. What relation does it have to trac?
17:56 dukeleto cotto: does it have to notify trac somehow as well?
17:56 luben_work left #parrot
18:02 cotto dukeleto, the github trac plugin also needs to have commit data sent to it via a post-receive url.
18:02 cotto That's why I care about getting the POST data, so I can test that the plugin responds properly.
18:03 cotto Nice.  bacekbot seems to have set his sights on a generational gc.
18:04 cotto bcaek++
18:04 cotto bacek++
18:21 dukeleto cotto: we can have multiple post-receive urls on github, so one of them will be for parrot-commits, the other for trac
18:21 cotto yup
18:23 dukeleto cotto: i guess those can live in tools/dev in the parrot repo. i will put them in the cotto/git_aware-tools branch, ok ?
18:23 kid51_to_lunch left #parrot
18:24 M_o_C joined #parrot
18:24 cotto wfm
18:24 fperrad left #parrot
18:30 ash_ left #parrot
18:32 dukeleto cotto: the post-receive hook gets hit *once per push* with N commits. how do we want to split that into emails? One email per commit, or one email per push?
18:32 dukeleto cotto: it is currently 1 per commit with svn
18:33 dukeleto cotto: i am leaning toward keeping it that way
18:35 cotto me too
18:39 Coke joined #parrot
18:40 tadzik left #parrot
18:41 tadzik joined #parrot
19:01 jsut joined #parrot
19:06 jsut_ left #parrot
19:22 whiteknight a big push turns into a barrage of emails though
19:30 atrodo And with the git way of thinking, you'd have more tiny commits leading up to a push
19:32 whiteknight right. I think the push is probably more along the lines of what we want to see in the email list
19:32 whiteknight we can link to diffs for individual commits, but the push diffs are probably more important
19:35 dukeleto whiteknight: are you talking about the full commit diff for a set of commits in one email?
19:42 pyrimidine joined #parrot
19:43 whiteknight that would be nice, wouldn't it? I don't know what github provides
19:44 whiteknight whatever it is, if we have a push with 100 small commits in it, I don't want to get 100 emails in my inbox
19:44 ttbot Parrot trunk/ r49299 MSWin32-x86-multi-thread make error http://tt.taptinder.org/file/cmdout/402551.txt ( http://tt.taptinder.org//bui​ldstatus/pr-Parrot/rp-trunk/ )
19:44 whiteknight we could threshold, so a big push only generates one email, but a small push could generate emails for all commits therein
19:46 whiteknight anyway, I'm heading home now. Later
19:46 whiteknight left #parrot
19:52 tadzik left #parrot
19:54 tadzik joined #parrot
20:03 tadzik left #parrot
20:20 bluescreen left #parrot
20:21 davidfetter joined #parrot
20:25 M_o_C left #parrot
20:31 bluescreen joined #parrot
20:49 whiteknight joined #parrot
20:54 whiteknight good afternoon, #parrot
20:54 plobsing joined #parrot
21:08 dukeleto whiteknight: welcome to the madness
21:09 whiteknight This is madness! This is SPARTA!
21:16 * plobsing plobsing sparta
21:17 plobsing was kicked by plobsing: sparta
21:17 plobsing joined #parrot
21:25 bluescreen left #parrot
21:25 patspam left #parrot
21:35 whiteknight :)
21:36 pyrimidine left #parrot
21:37 whiteknight left #parrot
21:38 patspam joined #parrot
21:49 ruoso left #parrot
21:52 mikehh left #parrot
21:57 plobsing urg. I don't understand the win32 make failure at all. src/pbc_dump.o doesn't even reference PMCNULL!
22:01 mikehh joined #parrot
22:03 bacek joined #parrot
22:12 fperrad joined #parrot
22:21 dalek parrot: r49300 | fperrad++ | trunk/src/ops (2 files):
22:21 dalek parrot: fix LuaThread after r49012.
22:21 dalek parrot: LuaThread vtable init throws an exception, an instantiation must be done by vtable init_pmc.
22:21 dalek parrot: So refactor not_p_p with clone instead of init/new.
22:21 dalek parrot: review: http://trac.parrot.org/parrot/changeset/49300/
22:21 dalek parrot: r49301 | plobsing++ | trunk/src/packdump.c:
22:21 dalek parrot: use safe accessors for keys
22:21 dalek parrot: review: http://trac.parrot.org/parrot/changeset/49301/
22:40 cotto plobsing, what kind of code needs to change after your recent typesafe consttable merge?
22:41 plobsing cotto: anything that tries to use the constant table directly. but extensions should *NOT* be doing that.
22:41 cotto I'm thinking about PIRATE.
22:42 plobsing cotto: I updated PackFileConstantTable and the associated tests. =
22:43 plobsing the only real change for PIRATE is that nums, strs, and pmcs now have independant indices. So adding a num constant doesn't increase the index of subsequent string constants.
22:43 cotto Great.  So I can just look at the diffs on those to get the gist.
22:43 cotto wfm
22:44 cotto What made you decide to work on that?
22:44 cotto (mostly ooc)
22:44 plobsing it makes other changes to the constant table easier (eg: linkage, backreferences among PMC constants)
22:45 plobsing also it has been nagging me for a while
22:48 cotto evercookie++
22:48 cotto raising awareness ftw
22:48 plobsing evercookie?
22:49 sorear aloha: msg p6eval Does msg work yet?
22:49 aloha sorear: OK. I'll deliver the message.
22:49 sorear nqp: say(2)
22:49 p6eval nqp: OUTPUT«PackFile_unpack: This Parrot cannot read bytecode files with version 8.3.␤␤PackFile header failed during unpack␤»
22:49 sorear looks like a "no" :/
22:49 cotto evercookie is http://samy.pl/evercookie/
22:49 sorear seen kid5`
22:49 aloha Sorry, I haven't seen kid5`.
22:49 sorear seen kid51
22:49 cotto evercookie?
22:49 aloha kid51 was last seen in #parrot 5 hours 42 mins ago saying "It also looks like lists.parrot.org archives are starting to pick up messages again (have checked parrot-dev and parrot-tickets)".
22:49 cotto evercookie?
22:50 cotto aloha, evercookie?
22:50 aloha cotto: Sorry, I don't know.
22:50 cotto aloha--
22:50 aloha cotto: Pbbbbtt!
22:50 cotto < aloha-- >
22:50 cotto all the snark but none of the factoids or karma
22:53 plobsing aloha is like a lobotomized version of purl :(
22:54 sorear aloha, karma aloha
22:54 aloha sorear: aloha has karma of -1.
22:55 sorear bacek: ping
22:55 patspam1 joined #parrot
22:58 patspam left #parrot
23:04 bacek sorear, pong
23:04 bacek aloha, evercookie is http://samy.pl/evercookie/
23:04 aloha bacek: Okay.
23:04 bacek aloha, evercookie?
23:04 aloha bacek: evercookie is http://samy.pl/evercookie/
23:04 bacek cotto, aloha doesn't implicitly learn things
23:05 bacek sorear, aloha uses privmsg for delivery
23:07 cotto bacek, why not?  That's part of what made purl useful.
23:08 bacek cotto, it was pretty annoying. But I can switch it on.
23:08 cotto I liked it, but I don't speak for everyone who frequents #parrot.
23:10 bacek aloha, purl?
23:10 aloha bacek: I have no idea.
23:10 bacek purl is pretty annoying girl
23:11 bacek aloha, purl?
23:11 aloha bacek: purl is pretty annoying girl
23:11 bacek cotto, done
23:11 cotto thanks!
23:12 bacek cotto, I can also switch on passive_answer, but I rather not :)
23:12 cotto That's probably wise.
23:12 sorear I agree with bacek, fwiw
23:12 sorear I don't like chatbots
23:16 patspam1 left #parrot
23:19 sorear aloha: msg kid51 I got your test mail
23:19 aloha sorear: OK. I'll deliver the message.
23:19 patspam joined #parrot
23:21 bacek cotto, http://github.com/bacek/aloha
23:22 bacek cotto, feel free to hack Msg.pm if you want to change behaviour :)
23:22 cotto I feel free to do so.
23:22 cotto should the need arise
23:23 sorear aloha: msg whiteknight Juerd says feather.perl6.nl is for Parrot too; ask him if you want an account (re. wanting a community server)
23:23 aloha sorear: OK. I'll deliver the message.
23:33 kid51 joined #parrot
23:34 sorear bacek: this chain of thought just happened: generational_gc -> write barriers -> incremental GC -> real time GC -> bubaflub?
23:35 bacek world domination
23:35 sorear aloha, msg sorear testing whether messages are dispatched on join
23:35 aloha sorear: OK. I'll deliver the message.
23:35 sorear left #parrot
23:35 sorear joined #parrot
23:35 bacek sorear, nope. Messages aren't dispatched on join.
23:36 sorear makes sense
23:36 sorear lambdabot and purl didn't do that either
23:37 sorear (also, people with unstable connections and autoreconnect)
23:39 plobsing (988029519 - 984219908) / 988029519
23:39 aloha 0.00385576637817033
23:40 cotto Trac seems snappier than usual.
23:41 dalek tracwiki: v5 | cotto++ | GitHubTracPluginTests
23:41 dalek tracwiki: add better (or non-worse) instructions for testing the github hook
23:41 dalek tracwiki: http://trac.parrot.org/parrot/wiki/GitHub​TracPluginTests?version=5&amp;action=diff
23:42 plobsing (30272480 - 30061776) / 30272480
23:42 aloha 0.00696024904467688
23:43 cotto what are you measuring?
23:44 plobsing size and speed improvement on perl6 after removing fixup table.
23:44 plobsing not as big as I'd hoped
23:45 plobsing 0.4% better speed, 0.7% better size
23:46 ruoso joined #parrot
23:46 bacek plobsing, how many lines of did you remove?
23:49 plobsing patch is 1271 lines
23:50 plobsing 874 deletions, 60 insertions
23:52 plobsing that doesn't count the outright deletion of PackfileFixupTable.pmc and PackfileFixupEntry.pmc
23:53 fperrad left #parrot
23:57 bacek I like it.
23:58 plobsing ok. once I deal with a couple of test failures, I'll ship it.
23:59 perlite left #parrot

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

Parrot | source cross referenced