|
Parrot 4.1.0 "Black-headed Parrot" | parrot.org | Log: irclog.perlgeek.de/parrot | #parrotsketch meeting Tuesday 19:30 UTC Set by moderator on 28 February 2012. |
|||
| dukeleto | whiteknight: hello | 00:01 | |
| whiteknight | goodevening, duke | ||
| dukeleto: do I have authority to create a new list? | 00:02 | ||
| dukeleto | whiteknight: i hope so. I remember last time I had to ask #osuosl to do it, iirc | ||
| whiteknight: not really sure | |||
| whiteknight | give me a minute | ||
| dukeleto | whiteknight: but if you create parrot-ci, i will make travis spam that instead, so bacek++ is happy | 00:03 | |
| whiteknight | we have a parrot-vm-nyc list | 00:11 | |
| we also have a parrot-commits list, which I think might be suitable for things like ci reports | |||
| and parrot-tickets | |||
| dukeleto | whiteknight: ok. i don't care much. Do you want to mix together CI report and commits? | 00:15 | |
| whiteknight: i know some people do subscribe to -commits and might be bothered by lots of non-commit noise | |||
| whiteknight | I don't know. Probably not a call I should make by myself | ||
| dukeleto | whiteknight: but I really don't know. | ||
| whiteknight: i think it is safe to create a parrot-ci list. That won't bother anybody | |||
| whiteknight: and then i can spam parrot-ci instead of parrot-dev, which will increase happiness | |||
| whiteknight | Error: You are not authorized to create new mailing lists | 00:16 | |
| no dice | |||
| dukeleto | blarg | 00:22 | |
| whiteknight: i will send an email to support@osuosl and cc you | |||
| whiteknight | thanks | ||
| dukeleto | whiteknight: done | 00:23 | |
| dalek | sella: add8dcb | Whiteknight++ | src/ (2 files): [Core] Fix a few typos in my updates last night to get tests passing again |
00:31 | |
|
00:37
kid51 joined
00:48
benabik joined
|
|||
| benabik | ~~ | 01:04 | |
|
01:28
contingencyplan joined,
plobsing joined,
jsut_ joined,
preflex joined,
Patterner joined,
Coke joined,
brambles joined,
sri joined,
mtk joined,
jjore_ joined,
cosimo joined,
myhrlin joined,
eternaleye__ joined,
nopaste joined,
arnsholt joined,
hernan joined,
atrodo joined,
allison joined,
cotto joined,
cxreg joined,
jtpalmer joined
01:35
kid51 joined,
schmooster joined,
ascent joined,
rblackwe joined,
TonyC joined,
knewt joined
|
|||
| dalek | sella: ea835d8 | Whiteknight++ | src/ (4 files): [Container] update Container |
02:01 | |
| sella: d222619 | Whiteknight++ | src/ (2 files): [Core] Add in a new autoget_type_formatter routine similar to autoget_global. Use it to clean some code in Date |
|||
| sella: dbf9f29 | Whiteknight++ | src/random/ (3 files): [Random] Several cleanups |
|||
| sella: 46fa174 | Whiteknight++ | / (2 files): [t] test fixes |
|||
|
02:01
slavorg joined
02:54
preflex_ joined
|
|||
| dukeleto | msg nine your thesis PDF is awesome! Probably the best docs about Parrot around | 03:39 | |
| aloha | OK. I'll deliver the message. | ||
| dukeleto | nine++ | ||
| benabik | dukeleto: Did I miss something? | 03:42 | |
| dukeleto | benabik: from #ps today | 03:48 | |
| benabik | Oh. #ps! D'oh! | ||
| dukeleto | msg nine s/would have been to complex | 03:49 | |
| aloha | OK. I'll deliver the message. | ||
| dukeleto | darn. | ||
| nine: ping? | |||
| benabik: this write up of parrot is pretty awesome | |||
| benabik | Ack! I'm in it! ;-) | 03:50 | |
| dukeleto | Since cross threaded writes are already forbidden and all read access to | 03:52 | |
| other thread’s data goes through the very narrow channel of proxy objects, | |||
| forcing a complete separation of the thread’s memory domains was only a | |||
| small step. By having separate memory areas for each interpreter, it became | |||
| possible to have each interpreter to run its own garbage collector. This way, | |||
| the known to work, single threaded GC implementation can be used nearly | |||
| unchanged. | |||
| benabik | msg nine Thanks for the thanks. | 04:01 | |
| aloha | OK. I'll deliver the message. | ||
| sorear | odd dukeleto is still gone | 04:05 | |
| benabik | sorear: Maybe you hurt his feelings? | 04:09 | |
| bacek_at_work | Unlikely. He started work in finance recently. They have no feelings! | 04:35 | |
|
06:00
davidfetter joined
06:40
fperrad joined
07:17
johbar_work joined
07:19
preflex_ joined
07:43
jsut joined
08:33
wagle joined
08:47
mj41 joined
|
|||
| dalek | kudo/nom: bb09ec6 | jonathan++ | src/core/Block.pm: NEXT phasers run in reverse order. |
08:47 | |
| kudo/nom: 7e2b6e6 | jonathan++ | src/Perl6/Actions.pm: Implement 'also' syntax. |
|||
| kudo/nom: 870f800 | jonathan++ | src/ops/perl6.ops: Convey result of a block to the leave phaser runner, so it can consider whether to run KEEP and UNDO blocks. |
|||
| kudo/nom: c77b6f8 | jonathan++ | src/ops/perl6.ops: Quick but test-passing implementation of KEEP and UNDO. Based on mls++ patch, with some changes from me. |
|||
| kudo/nom: 288fdb2 | jonathan++ | t/spectest.data: Run S04-phasers/keep-undo.t. |
|||
| kudo/nom: df250d8 | jonathan++ | docs/ChangeLog: Add some ChangeLog entries. |
|||
| kudo/macros3: e6f26fc | masak++ | src/Perl6/Actions.pm: lookups of undeclared & vars turn into Nil |
09:12 | ||
|
09:16
lucian joined
09:25
mj41 joined
|
|||
| dalek | kudo/nom: e29b2f1 | masak++ | / (7 files): implemented quasi quotes and macros This works: - Macro declarations - Calling a macro (using `macro()` and `macro` and operators) - Quasi quotes - Variable lookup from within the quasi quote This doesn't, yet: - Variable lookup from within a macro parameter |
09:34 | |
| kudo/nom: fdf3306 | masak++ | t/spectest.data: [t/spectest.data] added S06-macros/macros-d1.t |
09:44 | ||
| kudo/nom: 182b2f9 | jonathan++ | src/Perl6/World.pm: Toss unused methods. |
09:49 | ||
|
10:38
johbar_work joined
11:45
mj41 joined
12:29
bluescreen joined
|
|||
| dalek | kudo/name-cleanup: 2da46c4 | jonathan++ | src/Perl6/World.pm: Start to sketch out an attempt at neater name handling. |
12:36 | |
| kudo/name-cleanup: 04b9e4e | jonathan++ | src/Perl6/World.pm: Add method for turning a longname into the name component. |
|||
| kudo/name-cleanup: 316db7b | jonathan++ | src/Perl6/Grammar.pm: Start using the new longname handling in the type-declaration related parts of the grammar. |
|||
| kudo/name-cleanup: 50daf80 | jonathan++ | src/Perl6/Actions.pm: Update enum and subset actions for using new longname handling. |
|||
| kudo/name-cleanup: effeb6f | jonathan++ | src/Perl6/World.pm: Toss install_package_longname, which we no longer use. |
|||
|
13:33
benabik joined
|
|||
| Coke | bacek_at_work: hey, I work in finance, sort of, and I... wait, you're right. | 13:51 | |
| . o O (corporate tax information systems, close enough.) | |||
| dalek | kudo/name-cleanup: 95d31f8 | jonathan++ | src/Perl6/ (3 files): Rename for clarity. |
14:23 | |
| kudo/name-cleanup: 76fb938 | jonathan++ | src/Perl6/World.pm: Start recognizing pseudo-package names and complaining about invalid use of some of them in a package declaration. |
|||
| kudo/name-cleanup: c6747d2 | jonathan++ | src/Perl6/ (2 files): Eliminate canonical_type_longname. |
|||
|
14:35
benabik joined
|
|||
| benabik | Good morning, #parrot | 14:36 | |
|
14:39
PacoAir joined
|
|||
| nine | Good morning, benabik | 14:58 | |
| benabik | nine: Haven't gotten a chance to read your paper in detail, but at first look it looks very nice. | 14:59 | |
| nine | benabik: thanks | ||
| msg dukeleto thanks, fixed | 15:00 | ||
| aloha | OK. I'll deliver the message. | ||
| nine | benabik: the description of the implementation really needs some flow diagram... | 15:01 | |
| benabik | Diagrams are often very useful, although I'm not always very good at them. :-D | 15:03 | |
| nine | Me neither. That's the reason why there are no in the paper yet. Takes me ages to make them | 15:09 | |
| benabik | What are you using to make the paper? LaTeX? | ||
| I find Dot and TikZ/PGF useful for diagrams. | 15:10 | ||
| Also Inkscape, but PGF has the bonus of just being part of my LaTeX docs. | |||
| nine | Yes, it's LaTeX | 15:12 | |
|
15:27
bluescreen joined
15:45
Psyche^ joined
16:06
alester joined
17:00
Patterner joined
|
|||
| moritz | btw inkscape has a TexText plugin | 17:35 | |
| so you can use latex glyphs and formulas to label your inkscape drawings | 17:36 | ||
| aloha | (parrot/parrot) Issues opened : 731 (Add apt-get update to .travis.yml) by leto : github.com/parrot/parrot/issues/731 | 18:36 | |
| cotto | ~~ | 18:37 | |
|
18:52
contingencyplan_ joined,
contingencyplan joined
18:57
dukeleto joined
|
|||
| dukeleto waves | 19:00 | ||
| benabik | o/ dukeleto | ||
|
19:09
contingencyplan joined
19:31
mj41 joined
|
|||
| dukeleto | benabik: how goes it? | 19:39 | |
| benabik | dukeleto: Goes alright. I'm on my spring break, which means "Working on my thesis all the time" | ||
|
19:48
davidfetter joined,
davidfetter left
|
|||
| nine | Can we please remove all of Parrot's OO support? | 20:14 | |
| tadzik | :) | 20:17 | |
|
20:17
johbar_work joined
|
|||
| dukeleto | nine: ? | 20:21 | |
| johbar_work: welcome | |||
| benabik: good luck! | |||
| johbar_work | hi | ||
| nine | I've come full circle. I'm back at full cloning of namespaces and classes when creating a new thread. But the GC still finds PMCs belonging to the main thread. And anyway, cloning cannot be the answer. | 20:22 | |
| Tene | nine: removing Parrot's OO support is already planned, so yes, certainly. | 20:56 | |
| nine | This for me begs the question if threading support would be useful which would only work as long as Parrot's OO stuff is not used? | 20:57 | |
|
21:11
lucian joined
21:22
colomon joined
|
|||
| colomon | hi all. I'm having issues getting parrot to build on my Windows box. gist.github.com/1996282 | 21:22 | |
| dalek | nxed: ee3bbc1 | Whiteknight++ | winxed_installed.winxed: Fix installed driver after update to use PackfileView PMC in .compile |
21:32 | |
| nxed: 53b455f | NotFound++ | winxed_installed.winxed: Merge pull request #19 from Whiteknight/fix_installed_driver Fix installed driver after update to use PackfileView PMC in .compile |
|||
| dukeleto | colomon: howdy | 21:41 | |
| colomon | dukeleto: hello! | 21:42 | |
| dukeleto | colomon: still having issues building parrot on win? | ||
| colomon: that is a baffling error | 21:43 | ||
| colomon | still having issues | 21:44 | |
| this is 64-bit vista | 21:45 | ||
| and gcc / perl from the strawberry perl installation | |||
| dukeleto | colomon: try dtruss on the command that fails | 21:52 | |
| colomon: blarg, no dtruss on windows | |||
| colomon: do you have access to strace? ktrace? | |||
| colomon: what is the output of ".\\miniparrot.exe -Iruntime/parrot/include config_lib.pir" ? | |||
| colomon | what are strace and ktrace (or rather, where do they come from)? | 21:53 | |
| dukeleto | nine: it kind of sounds like you need the features of 6model to finish thread support | ||
| colomon: standard unix/linux utils. | |||
| colomon: there is also ltrace | |||
| colomon | dukeleto: output appears to be nothing (or maybe just newline) | 21:54 | |
| dukeleto: trying to get stracent working, but I think it's not savvy to more recent versions of Windows. | 22:00 | ||
|
22:02
not_gerd joined
|
|||
| dalek | itor: ee9e5cd | NotFound++ | src/Guitor.winxed: typo in Menu |
22:02 | |
| not_gerd | colomon: did you pass --intval="long long" to parrot's Configure.pl? | ||
| colomon | nope. is that needed? | ||
| not_gerd | yes - I've got a branch which fixes that, but no one has yet shown any interest in getting it merged :( | 22:03 | |
|
22:10
benabik joined
|
|||
| colomon | that gets me the following message: | 22:14 | |
| During configuration the following steps failed: | |||
| 24: auto::sizes | |||
| 25: auto::byteorder | |||
| 27: auto::format | |||
| You should diagnose and fix these errors before calling 'gmake' | |||
| dukeleto | colomon: interesting | 22:17 | |
| not_gerd: i am interesting in helping merge that branch, but not sure that I can help test it much | |||
| colomon: you can still type "make" | |||
| colomon: but Configure.pl is telling you that some stuff is odd | |||
| colomon: does "make" work for you now? | |||
| not_gerd: which branch fixes this issue? Is there a pull request for it? | 22:18 | ||
| not_gerd | dukeleto, colomon: make will probably fail | ||
| I think I know what's up with at leastsome of these failures... | |||
| dukeleto: github.com/parrot/parrot/pull/704 - fixing yuto::sizes was the main reason I made the branch | 22:19 | ||
| dukeleto: but I started adding fixes for all other problems I found | 22:20 | ||
| ^auto::sizes | |||
| alester | NEW TOY! blogs.perl.org/users/rurban/2012/03...und-2.html wheee! | 22:22 | |
| cotto | alester: nice | 22:24 | |
|
22:25
jsut_ joined
|
|||
| alester | I LOVE NEW STATIC TOYS. | 22:25 | |
| cotto | I love what you get excited about. | 22:26 | |
| colomon | is aloha a channel bot? seems to be sending me old messages from bubaflub | 22:27 | |
| not_gerd | colomon: I can't reproduce your problem using strawberry-perl-5.14.2.1-64bit-portable on Win7 | ||
| colomon | I'm on perl 5, version 14, subversion 2 (v5.14.2) built for MSWin32-x64-multi-thread | 22:29 | |
| gotta run nom | |||
| not_gerd | colomon: same here; parrot master compiles fine here using `perl Configure.pl --optimize --intval="long long"` | 22:31 | |
| dukeleto | colomon: yes, aloha sends you a message when you say something in #parrot | 22:33 | |
| colomon: you probably had really old messages waiting for you | |||
| colomon: i.e. aloha is not triggered to send you your messages until you say something in here | |||
| alester: asan looks spiffy | 22:39 | ||
| colomon | not_gerd: oooo, if I use just that configure line of yours, it seems to work fine. | 23:01 | |
| now making parrot... | |||
|
23:16
preflex_ joined
|
|||
| colomon | oooo, new error: | 23:21 | |
| .\\miniparrot.exe -Iruntime/parrot/include config_lib.pir > runtime/parrot/include/config.fpmc | |||
| error:imcc:Constant 'PARROT_SHA1' value must be a number, stringliteral or register | 23:22 | ||
| in file 'parrot_version.pir' line 1 | |||
| included from 'config_lib.pir' line 1 | |||
| not_gerd | colomon: and what does your runtime/parrot/include/parrot_version.pir look like? | 23:27 | |
| colomon | oooo, this one's my fault -- somehow it includes quotes from my quote file in it. mucked up shell called, I suspect. I'll try to straighten it out once my little boy is asleep | 23:31 | |