Parrot 3.7.0 "Wanda" | parrot.org | Log: irclog.perlgeek.de/parrot/today | #parrotsketch meeting Tuesday 19:30 UTC
Set by moderator on 17 August 2011.
whiteknight i can't imagine how we would restrict them to just invokables 00:02
plobsing well for one, most things won't have a slot for them 00:03
but why would we want to restrict it?
if you ask for something stupid, parrot should serve it up on a silver platter 00:04
I used the :descriptor(string) syntax for something human readable, but I expect these will be compiled down into an efficient machine representation IMCC (or equivalent) 00:14
00:16 darbelo joined 00:17 darbelo joined 00:39 soh_cah_toa joined 01:12 rdesfo joined 01:30 nopaste joined 01:36 jsut joined
dalek rrot/tar_shasum: 9d842b8 | jkeenan++ | config/ (3 files):
Convert 'tar' and 'sha256sum' into Makefile variables. Allow for 'shasum -a 256' on Darwin.
 02:12
rrot/tar_shasum: 7a273f6 | jkeenan++ | config/init/hints/darwin.pm:
Correct Darwin version of sha256sum.
 02:34
Coke ooh, sweet.
kid51++
02:45 rfw joined 03:12 rdesfo joined 03:33 theory joined 03:40 rdesfo joined
cotto ~~ 04:12
04:26 particle1 joined 04:59 thowe joined
thowe Hi. So, I am trying to build rakudo along with parrot and I get this: scsys.co.uk:8002/133702 04:59
I asked the Perl6 guys about it and they said I should ask in here...
This is the latest rakudo tar ball. 05:00
05:51 rohit_nsit08 joined 05:54 fperrad joined 05:55 pathfinder joined 05:59 cotto joined 06:32 rohit_nsit08 joined 06:46 cotto joined 06:57 rfwalrus joined
dalek rrot/whiteknight/frontend_parrot2: a2bbc00 | plobsing++ | / (5 files):
avoid depending on interp's current context for backtrace printing

the top of a backtrace is the top, not wherever current context happens to be pointing now (as opposed to when the continuation being backtraced was captured)		 06:59
07:00 cotto joined 07:11 he joined 07:17 rohit_nsit08 joined 07:23 mj41 joined 07:31 contingencyplan joined 07:34 perlite_ joined 08:04 lucian joined
pmichaud gist.github.com/1153639 # lexpad iteration fail 08:15
lucian blog.bossylobster.com/2011/08/lesso...other.html 08:29
08:36 cottoo joined 08:44 not_gerd joined
not_gerd good morning, #parrot 08:44
08:45 cotto joined
dalek TT #2181 created by pmichaud++: Bug in HashIteratorKey .value method when used with LexPad PMC 08:55
TT #2181: trac.parrot.org/parrot/ticket/2181
10:26 rohit_nsit08 joined
dalek rrot/tar_shasum: 1f9923a | jkeenan++ | config/init/ (4 files):
Provide inline comments to document choices for 'tar' and 'sha256sum'. For
 10:58
11:02 kid51 joined
kid51 msg thowe Please try to build Parrot itself from source, i.e., from a release tarball. If you can do that, then that suggests the error is somewhere in the Rakudo Star build process. 11:04
aloha OK. I'll deliver the message.
11:09 fperrad_ joined 11:15 jsut_ joined 11:31 cosimo joined 11:43 rohit_nsit08 joined
kid51 posted on parrot-dev re 'sha256 checksum during make release' problem. If someone can supplement this with solution on Win32, please do so. 11:47
moritz kid51: aren't there perl core modules that implement sha256? 11:51
Digest.pm is a core module since perl 5.7.3 11:52
and at least the current Digest.pm supports sha256
not sure if older ones do too 11:53
its 'Changes' file says for release 1.03 (shipped with perl 5.8.3): 'Digest->new("SHA-256") and similar should work now 11:54
'
I see that lucian++ had the same idea 11:56
kid51 moritz: Thanks. I will look into that, as it will likely provide a more across-the-board solution.
lucian depending on perl's digest module, it might still be better to use the shell tools on unix 12:00
moritz if better = faster, yes
though it's just factors slower, not orders of magnitude 12:01
lucian so there could be several sha256 platforms: darwin, linux, *bsd, perl5digest
moritz lucian: but maybe it's just not worth the effort, and one always uses perl 12:02
moritz benchmarkes for comparison
lucian moritz: if the shell ones are already written and work, it doesn't seem that much effort to maintain them 12:04
moritz lucian: well, one must maintain the code that decides which to use 12:05
lucian sure, and if the perl one's fast enough, it's ok 12:06
i'd expect them to use the same libs anyway 12:07
moritz just timed it with a 50M file: 0.519s perl version, 0.506s command line utility
lucian moritz: on linux? 12:10
moritz lucian: yes
lucian i guess perl'll be using the same libs everywhere
moritz lucian: yes, it's a C/XS module 12:12
12:13 whiteknight joined 12:15 whiteknight joined
Coke thowe: you hit the channel at a dead time, sorry about that; if you open a ticket, then the bug repot won't get lost. 12:21
(trac.parrot.org)
kid51 I like moritz's idea, as in the branch I'm already up to 3 different shell commands. 12:22
Won't have time to try it out until tonight at earliest, so if anyone else wants to create a patch, that would be cool. 12:23
kid51 to dayjob
12:24 ligne joined
dalek nxed: 2a5ac47 | NotFound++ | / (2 files):
optimize a bit handling of the is_multi flag
 12:27
nxed: ba8694a | NotFound++ | pir/winxed_compiler.pir:
update installable compiler
 12:28
TT #2182 created by ligne++: argument assertion failure when quitting parrot_debugger
TT #2182: trac.parrot.org/parrot/ticket/2182
12:48 JimmyZ joined 12:57 bluescreen joined
whiteknight good morning, #parrot 13:06
lucian waves 13:07
whiteknight hello lucian 13:10
msg benabik lets get together sometime for a meeting 13:19
aloha OK. I'll deliver the message.
13:19 bluescreen joined
cotto nom gets rid of rebless_subclass! <3 13:30
13:32 JimmyZ joined
dalek rrot: a0dbe1f | NotFound++ | / (4 files):
add function Parrot_str_new_init and replace common parts of to_string methods with a call to it
 13:33
NotFound Ups... wrong function name
plobsing whiteknight: in your most recent blog post you maligned some comments I wrote. Do you have any advice on how I could have written them better? 13:36
whiteknight plobsing: I maligned your comments? 13:37
plobsing "I’ve removed comments for clarity (No, that’s not some kind of a joke. There were comments in this snippet, and they are not helpful for understandability)." 13:38
those were *my* comments. what did I do wrong?
whiteknight oh, those comments were XXX type comments, not helpful for the readability of the blog post
I'm sorry, I didn't intend that to be a swipe at the comment author
plobsing oh. what I got from that was that my comments were not useful. 13:39
plobsing needs to read more closely
*carefully
whiteknight Yeah, I'm sorry. I definitely did not intend to insult, I didn't think the comments would help the blog post, and I removed them
but the phrase "I've removed comments for clarity" seems like some kind of weird, deadpan joke, and that's not how I wanted it to be read 13:40
I should have been more clear
plobsing ok, glad we got that cleared up 13:41
whiteknight They were like XXX and TT #... comments, right?
whiteknight has to go look at the code again 13:42
plobsing both XXX
no TT's involved IIRC
13:42 contingencyplan joined
dalek rrot: 865c881 | NotFound++ | src/embed.c:
allow NULL in ignored parameter, TT #2182, ligne++
 13:47
TT #2182 closed by NotFound++: argument assertion failure when quitting parrot_debugger 13:53
TT #2182: trac.parrot.org/parrot/ticket/2182
ttbot Parrot a0dbe1f7 MSWin32-x86-multi-thread make error tt.taptinder.org/cmdinfo/44848 13:55
whiteknight win32-- 13:58
We should tell ttbot to change the name of the platform to "win32--"
atrodo whiteknight++ 13:59
whiteknight so we don't need to manually deduct karma on every failure
moritz believes that C++ wasn't meant to be a real programming language, just a means for giving C more karma
whiteknight in the early days of mediawiki, you couldn't have "+" in an article title. So people were setting up article names like "C-/- -/-" and "C_plus_plus" and other terrible things 14:03
-/- looks like a + if you squint your eyes and ignore the fact that it doesn't 14:04
cotto + looks like a + too 14:05
plobsing -/- looks like someone squiting their eyes trying to make -/- look like +
dalek kudo/nom: 48a44ea | pmichaud++ | src/core/ (8 files):
Move &infix:<eqv> and &infix:<===> candidates into their appropriate files.
 14:09
kudo/nom: c69d4ba | pmichaud++ | src/core/ (11 files):
Refactor core class stubs a bit.
atrodo -|- looks more like a +
14:10 ambs joined
dalek kudo/nom: 66675a3 | pmichaud++ | tools/build/Makefile.in:
Need update to Makefile ordering for stubs refactor.
 14:12
whiteknight atrodo: | runs afoul of the same HTML encoding restrictions as + does 14:17
ttbot Parrot 865c8814 MSWin32-x86-multi-thread make error tt.taptinder.org/cmdinfo/44864
14:23 JimmyZ joined
dalek kudo/nom: 39f444a | masak++ | / (2 files):
Revert "implement p5chomp, p5chop (scalar case only)"

This reverts commit 6cac7f1d1b44ba80b0ce799ae9774dd27004393a.		 14:28
kudo: a553460 | masak++ | t/spectest.data:
[t/spectest.data] remove p5chop.t and p5chomp.t
 14:36
14:39 rohit_nsit08 joined
dalek website: rohit_nsit08++ | corellaScript : Arrays 14:42
website: www.parrot.org/content/corellascript-arrays
kudo/nom: 3a2e045 | masak++ | t/spectest.data:
[t/spectest.data] remove p5chop.t and p5chomp.t
14:44 JimmyZ joined 14:53 logie joined
dalek kudo/nom: 85e004f | pmichaud++ | src/core/metaops.pm:
Fix "undefined value" warning in hyperoperators noted by jnthn++ .
 14:58
15:08 rohit_nsit08 joined 15:16 cotto joined 15:28 rdesfo joined 15:50 SHODAN joined 16:06 rohit_nsit08 joined 16:15 theory joined 16:16 rdesfo joined, rdesfo left 16:20 dmalcolm joined 16:30 cotto joined 16:37 preflex_ joined 16:47 rohitnsit08 joined 16:50 lucian joined 16:58 ambs_ joined 17:09 bluescreen joined 17:33 rohit_nsit08 joined 17:36 Eclesia joined
Eclesia hi 17:36
17:39 fperrad joined
whiteknight hello Eclesia 17:41
dalek kudo/nom: f147bf8 | tadzik++ | lib/Pod/To/Text.pm:
[Pod::To::Text] Align table columns
 17:50
kudo/nom: d664591 | tadzik++ | / (2 files):
Fix a bug in tables with trailing newlines, add test
kudo/nom: 195fef4 | tadzik++ | t/pod/08-formattingcodes.t:
Add formattingcodes tests
cotto ~~ 17:53
17:54 rohit_nsit08 joined 17:59 mj41 joined
whiteknight hello cotto 17:59
cotto hi whiteknight 18:12
I spent the day attending a Perl6 class lead by jnthn++. Turns out that Rakudo is substantially awesomer than I suspected. 18:13
He also showed me through 6model a bit. 18:14
tadzik I should have gone to that one 18:15
cotto tadzik, you could have joined masak and pmichaud for heckling. 18:16
tadzik heckling?
cotto making sarcastic and/or distracting remarks 18:17
tadzik :)
cotto sadly, they didn't get very much heckling done.
Coke wishes he could see jnthn++ 18:25
s talks.
18:38 Kulag joined
bubaflub was it recorded? 18:51
cotto jnthn's talk wasn't. It was several hours long.
tadzik you mean the training, right? So all the other talks were? 18:52
cotto tadzik, correct
tadzik cool
cotto they should be on yapc.tv at some point
plobsing yapc.tv doesn't appear to have anything from the last 2 years 18:57
18:58 Drossel joined
cotto I noticed that just now. Videos from yapcs past sometimes disappear into the ether. 18:58
19:03 rdesfo joined 19:04 rdesfo left 19:13 rohit_nsit08 joined 19:15 Caldrin joined
Caldrin Hi! 19:15
whiteknight hello Caldrin
Eclesia ~~
damn, switching everyday from java to winxed, makes me write silly things : final var string str 19:19
whiteknight :) 19:22
Caldrin I try to work to the Perl6 book which is referenced on perl6.org. There is an example in it, that does not work for me. I pasted my version to perlpunks.de/paste/show/4e4d663b.6c46.76
This brings the following error message:
No such attribute '$!salary' in class 'Employee' in 'GeekCook::new' at line 42:/home/hmai/Projekte/Practice/Perl6/Book/class2.p6 in main program body at line 46:/home/hmai/Projekte/Practice/Perl6/Book/class2.p6 19:23
atrodo Eclesia> I had a similar experience this week going form Delphi to JavaScript. For some reason, that switch was hard
Caldrin Can someone please tell me what the issue in my example code is? As far as I understood, the $!salary should exist and if I subclass Employee without overwriting new is actually does.
My rakudo version is from git clones on Tuesday. 19:25
nine cotto: read a lot of Parrot code on the way back home from YAPC. Is there more information available about the current state? Seems like basic threading support is there and it even works according to 901 lines of test code. From what I've seen, theres no GC as long as more than one thread is running? 19:26
whiteknight Caldrin: good question
Caldrin Thank you.
whiteknight Caldrin: the code looks good. You might need to talk to a perl6 expert 19:27
Eclesia atrodo: the worse thing is loosing the auto-completion
Caldrin whiteknight: Actually I thought I could find them here. Can you please give me a pointer where to look for one?
whiteknight moritz: ping
Caldrin: The main IRC channel for perl6 is irc.freenode.net/#perl6 19:28
19:29 Kulag joined
Caldrin IC. 19:29
Thank you.
sorear Caldrin: need help? 19:38
Caldrin sorear: Indeed I do. 19:41
19:44 Drossel joined 19:52 rohit_nsit08 joined 19:53 ambs joined
moritz whiteknight: pong 20:14
whiteknight moritz: Caldrin was looking for a perl6 expert, but I think sorear caught up first 20:15
Caldrin whiteknight: No he did not.
moritz: Query
whiteknight oh, I was mistaken 20:16
20:26 thowe joined
thowe Hi. I'm trying to setup Rakudo with Parrot and it is crashing when it gets to ./parrot-nqp --target=pir compilers/data_json/JSON.nqp > runtime/parrot/languages/JSON/JSON.pir 20:47
OK, so I tried just building parrot on its own and it crashes here: ./parrot-nqp --target=pir --output=compilers/opsc/gen/Ops/Compiler/Actions.pir compilers/opsc/src/Ops/Compiler/Actions.pm 20:55
benabik thowe: Platform? 20:56
thowe OpenBSD 4.9 i386
I could also try on a 5.0 beta machine in a few minutes... 20:57
I have successfully used past versions.
benabik Sounds like it's a problem with something in/used by NQP. 20:58
thowe I could paste more data somewhere if it helps...
I take it this isn't a known issue...
benabik Not known by me. 20:59
3.6 worked and 3.7 doesn't?
thowe this is 3.6
20:59 Kulag joined
thowe I was trying same version that failed with rakudo 20:59
also, it's the latest I see here: ftp.parrot.org/releases/supported/ 21:00
benabik Ah. 3.7 was released yesterday. 21:01
I'm looking around to see if there's anything that stands out… Not 100% sure how to debug the issue. If it's a segfault, you might try running the failed command via GDB. 21:02
thowe yeah, it's a segfault
benabik I'd lay decent odds on a GC issue then. I wonder if there's a missing write barrier somewhere. 21:03
thowe I'll have to look up how to do that. I've only used gdb a little some time ago
gc?
benabik Garbage Collection
21:04 rohit_nsit08 joined 21:05 rohit_nsit08 joined, rohitnsit08 joined
thowe yeah, I don't know how to make gdb run that whole command. 21:07
benabik gdb ./parrot-nqp ; set args --target=pir --output=compilers/opsc/gen/Ops/Compiler/Actions.pir compilers/opsc/src/Ops/Compiler/Actions.pm ; run ; # wait for crash ; bt
Where ; == return
thowe cool, running 21:08
huh, said executed normally.. must be something after that that I don't see that is crashing... 21:09
benabik Do you have parrot installed? It might be using the old libraries and such. Not sure how to fix that.
thowe no, no parrot 21:10
benabik Oh. Then it's probably a GC issue. Those are fun because sometimes observing them causes them to change. :-( Will need someone more versed in the deep magic to help.
Coke could try overriding the default GC instead. 21:11
thowe this time it died elsewhere... last command I see is ./parrot-nqp --target=pir --output=compilers/opsc/gen/Ops/Compiler/Grammar.pir compilers/opsc/src/Ops/Compiler/Grammar.pm
Coke I assume thowe is more interesting just getting rakudo to work at this point.
thowe well, playing with Perl6 was my main goal...
plobsing thowe: you ran the crashing command from make in gdb and it ran fine? can you try rerunning make to see if it progresses further?
thowe no, I just ran the one command. I could run make instead. 21:12
running make in gdb doesn't seem to give more info... backtrace says "no stack" 21:16
the error code is 139, "Stop in /home/tim/parrot-3.6.0 (line 14 of compilers/opsc/Rules.mak)."
Emitter.pm stuff on that line 21:17
benabik If parrot can't run NQP properly, I'm not sure compiling Rakudo will go very well. 21:18
And I'm going to go AFK for a bit and rest my hand.
21:21 lucian joined
plobsing thowe: can you put the make output on a pastebin somewhere and give us a link? 21:21
thowe sure 21:22
fresh make, after a config run?
plobsing sure, sounds fine 21:23
thowe the thing is, if I keep running make it seems to get a little further every time. this seems to support the gc theory
plobsing are you running parallel make (eg: -j3)? perhaps it is failing for one dependancy, but making progress on independant targets 21:25
thowe scsys.co.uk:8002/134088 21:26
just "make"
plobsing alternatively, doesn't one of the bsds' makes randomize its orders? that would have a similar effect.
thowe don't know, randomizing of certain things is an OpenBSD thing 21:27
but I think it was Parrot 3.3.0 that was working for me last. Is this one much different? 21:28
scsys.co.uk:8002/134088
plobsing so does './parrot-nqp --target=pir --output=compilers/opsc/gen/Ops/Emitter.pir compilers/opsc/src/Ops/Emitter.pm' segfault for you then?
thowe that seems to be where it ended this time... If I type make a few more times I am sure I'll get further 21:29
could this be the same thing: permalink.gmane.org/gmane.comp.comp...kets/11344 21:30
this patch is not in my tarball 21:35
plobsing you could try patching and remaking 21:36
21:37 soh_cah_toa joined
benabik That commit is in 3.7 but not 3.6 21:37
(would have found it if the commit mentioned BSD, but it only has the TT#
thowe I just manually applied the changes (I hope), trying again. 21:38
21:43 rohitnsit08 joined 21:45 Psyche^ joined
thowe seems to build after that patch \\o/ 21:45
21:45 rohitnsit08 joined
thowe I should try 3.7 21:45
21:47 rohitnsit08 joined 21:48 rohitnsit08 joined 21:49 rohitnsit08 joined 21:52 rohitnsit08 joined 21:53 rohitnsit08 joined
thowe is there not a 3.7 tar ball? 21:55
21:55 rohitnsit08 joined 21:57 rohitnsit08 joined
benabik 3.7 is a developer release… IIRC, *.0 *.3, *.6, and *.9 are supported, others are dev. 21:59
ftp://ftp.parrot.org/pub/parrot/releases/devel/3.7.0
plobsing perhaps we need to cut a 3.6.1 release to support openbsd? 22:01
benabik And perhaps the msys fixes as well. 22:08
thowe 3.7 seems to compile OK. 22:21
on OpenBSD, that is 22:22
dalek Heuristic branch merge: pushed 31 commits to parrot/nqp_pct by Benabik 22:34
22:47 rfw joined, kid51 joined
Coke (3.6.1 bugfix release) depends on how proactive we wish to be, but yah. 23:28
I might set the bar as low as "if one more person comes to the channel..." ;)
dalek TT #2183 created by coke++: sha256 problem encountered during 'make release'
TT #2183: trac.parrot.org/parrot/ticket/2183
rrot/tar_shasum: 6989471 | jkeenan++ | / (7 files):
Use Perl program written by moritz++ as cross-platform way of creating sha256 checksum on release tarballs.
 23:30
23:44 whiteknight joined
whiteknight good evening, #parrot 23:47
soh_cah_toa whiteknight: hey 23:48
23:53 kid51_ joined
kid51_ burst of lightning broke my net connection 23:53
whiteknight kid51_: it's a sign 23:54
hello soh_cah_toa
bleh. Any git experts hereabouts? 23:58

Powered by Cro and the Raku® Programming Language.

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

Thank you!