Camelia, the Perl 6 bug

IRC log for #parrot, 2011-10-01

Parrot | source cross referenced

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

All times shown according to UTC.

Time Nick Message
00:05 whiteknight damnit
00:05 whiteknight diakopter: can you nopaste the errors?
00:05 diakopter I submitted http://trac.parrot.org/parrot/ticket/2196
00:05 whiteknight I thought we were cygwin-clean
00:05 whiteknight thanks! diakopter++
00:06 diakopter it is very unapparent from trac.parrot.org how to submit a bug report; I had to ask [Coke], who suggested the tickets email address.
00:06 diakopter and *now* I find the explanation on http://www.parrot.org/dev/get-involved
00:08 diakopter hm, maybe this STATUS_ACCESS_VIOLATION is a Windows 7/cygwin problem.
00:12 dalek TT #2196 created by Matthew Wilson <diakopter@…>++: cygwin build fails
00:12 dalek TT #2196: http://trac.parrot.org/parrot/ticket/2196
00:14 * diakopter did a rebaseall and is now trying again
00:22 diakopter didn't help :(
00:43 whiteknight bleh
00:47 cotto ~~
00:52 diakopter left #parrot
00:52 diakopter joined #parrot
00:52 diakopter I knew slavorg op'd me like 2 years ago
00:56 Coke diakopter++ for testing this platform.
00:57 sorear there's an easier way to kick slavorg
00:57 sorear opbots, names
01:16 whiteknight joined #parrot
01:19 diakopter same failure on git HEAD
01:20 soh_cah_toa_ what does "slavorg" mean anyway?
01:21 soh_cah_toa_ is that like a portmanteau of "slave" and "cyborg"?
01:24 whiteknight I would have never thought that
01:24 diakopter his was my assumption too
01:25 soh_cah_toa_ slavorg: are you a slave cyborg?
01:26 soh_cah_toa_ slavorg: say nothing if you're a slave cyborg
01:26 soh_cah_toa_ :O
01:26 soh_cah_toa_ it's true
01:28 sorear http://movieos.org/code/slavorg/
01:29 soh_cah_toa_ oh yeah, i remember seeing that last time i tried to find out what it meant
01:30 * soh_cah_toa_ sometimes thinks about porting our irc bots to perl 6
01:31 soh_cah_toa_ that's why i wrote IRC::Utils originally
02:22 contingencyplan joined #parrot
02:29 cotto I can feel the damage spreading.  If I ever say that PHP's not so bad, somebody please shoot me.
03:01 plobsing cotto: half of what makes it bad is that it has thousands of inconsistently-named functions built right into the interpreter. Parrot is way better... oh wait!
03:04 benabik I'm sure there are worse languages…  I can't think of one at the moment.  ;-)
03:41 soh_cah_toa_ plobsing: ha! it drives me mad!!! >:O
04:35 TiMBuS * soh_cah_toa_ sometimes thinks about porting our irc bots to perl 6
04:35 TiMBuS please do!
04:35 TiMBuS itd be nice if someone used my module
05:47 alvis joined #parrot
06:05 * nine understands runops now. Scared of himself.
06:05 cotto joined #parrot
06:19 fperrad joined #parrot
07:11 perlite_ joined #parrot
09:22 dalek rakudo/nom: fb48457 | moritz++ | src/core/ (2 files):
09:22 dalek rakudo/nom: restrain IO.print and say to defined invocants; add a Mu.print
09:22 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/fb4845759f
10:24 nbrown joined #parrot
10:38 whiteknight joined #parrot
11:11 whiteknight good morning, #parrot
11:18 tadzik good morning whiteknight
11:19 whiteknight hello tadzik, how are you today?
11:19 tadzik pretty well, how about you?
11:21 whiteknight doing alright. Trying to get some computer time in before the rest of the house wakes up
11:22 tadzik what time is it up there?
11:23 whiteknight 7:20AM
11:27 tadzik ouch
11:27 tadzik it's been months since I was awake on such hour, but the semester is starting, so I should probably start getting used to it
11:28 whiteknight I wake up every day at 5:30 whether I want to or not. My body is programmed to wake up then
11:28 nine good morning whiteknight
11:29 whiteknight hello nine
11:33 nine whiteknight: got it to resume preempted tasks :) Still segfaulting later in the build process though
11:34 whiteknight wow, that's a good milestone.
11:34 tadzik so it works, but it doesn't build?
11:34 tadzik that's a new one :)
11:34 whiteknight do a "make corevm" and then a "make coretest". Some of the tests might be able to help you narrow down what is failing
11:35 nine tadzik: it can build a ./parrot and use that during the build process. At least till ./parrot -o runtime/parrot/library/PGE/Util.pbc runtime/parrot/library/PGE/Util.pir which segfaults
11:35 whiteknight it's much easier than trying to pick through the parrot-nqp build steps looking for failures
11:35 * nine has to remove some debug output to find anything :)
11:36 Psyche^ joined #parrot
11:37 tadzik oh, PGE is still there
11:41 dalek nqp: a430d8d | tadzik++ | / (10 files):
11:41 dalek nqp: Merge branch 'kill-useless-defaults'
11:41 dalek nqp:
11:42 dalek nqp: Conflicts:
11:42 dalek nqp: src/PAST/NQP.pir
11:42 dalek nqp: tools/build/PARROT_REVISION
11:42 dalek nqp: review: https://github.com/perl6/nqp/commit/a430d8d8be
11:43 dalek rakudo: d4cb879 | tadzik++ | tools/build/NQP_REVISION:
11:43 dalek rakudo: Bump NQP_REVISION
11:43 dalek rakudo: review: https://github.com/rakudo/rakudo/commit/d4cb8793c5
11:44 dalek rakudo/nom: 4880cea | tadzik++ | tools/build/NQP_REVISION:
11:44 dalek rakudo/nom: Bump NQP_REVISION
11:44 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/4880cea9fe
12:07 dalek jaesop: 5747abe | Whiteknight++ | / (3 files):
12:07 dalek jaesop: Add in switch/case statements and a basic test file for them
12:07 dalek jaesop: review: https://github.com/Whiteknig​ht/jaesop/commit/5747abea1e
12:16 redicaps joined #parrot
12:16 redicaps left #parrot
12:37 dalek rakudo/nom: 5a38367 | moritz++ | / (2 files):
12:37 dalek rakudo/nom: implement Grammar.parsefile, run a test for it
12:37 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/5a38367c26
13:08 preflex_ joined #parrot
13:44 dalek jaesop: ee73942 | Whiteknight++ | stage0/ (3 files):
13:44 dalek jaesop: Add in a constructor for String and a few methods on it.
13:44 dalek jaesop: review: https://github.com/Whiteknig​ht/jaesop/commit/ee739420b9
14:09 alvis joined #parrot
15:02 benabik o/ #parrot
15:13 nine longjmp makes debugging....interesting
15:14 benabik :-D
15:16 nine free_runloops_until (interp=0x62d3b8, id=1048576)... how on earth can the runloop id be exactly 1024*1024?
15:23 nine especially as only one runloop ever got created when running the program
15:39 nine Well, I'd say, exception handling is broken. This will become interesting
16:06 nine It's pretty obvious that the stack is corrupted. So after the longjmp everything breaks.
16:06 benabik longjmp shouldn't do anything to the stack.
16:06 benabik (Well, other than rewind it._
16:14 dalek rakudo/nom: 464fb9f | jnthn++ | src/Perl6/Actions.pm:
16:14 dalek rakudo/nom: Complain about declaration of placeholders in the mainline.
16:14 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/464fb9f25e
16:16 nine At least before the longjmp everything seems to be fine and after the jump, everything is screwed. So either my stack or my jmp_buf or both are corrupt.
16:45 whiteknight joined #parrot
16:47 gbacon joined #parrot
17:01 cotto ~~
17:04 whiteknight hello cotto
17:11 cotto 'morning, whiteknight
17:22 whiteknight I'm trying to add pcre bindings to Jaesop today
17:41 cotto whiteknight, just to test how easy it is to add bindings?
18:01 soh_cah_toa joined #parrot
18:05 cotto looks like the rakudo release is out
18:42 whiteknight no, we need regexes in stage0 to support a parser in stage1
18:43 moritz cotto: aye, rakudo was released yesterday
18:49 Kulag joined #parrot
18:53 Kulag joined #parrot
19:04 dalek nqp: a31d3f0 | moritz++ | src/PAST/NQP.pir:
19:04 dalek nqp: generate get_id instead of get_addr
19:04 dalek nqp: review: https://github.com/perl6/nqp/commit/a31d3f0e65
19:11 dalek rakudo/nom: 2772e64 | jnthn++ | src/binder/bind.c:
19:11 dalek rakudo/nom: Make sure we type check default values. Note that we should really promote the one for literals to compile time.
19:11 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/2772e6415f
19:32 dalek rakudo/nom: d655c14 | moritz++ | tools/build/NQP_REVISION:
19:32 dalek rakudo/nom: bump NQP_REVISION to something not so broken
19:32 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/d655c14ece
19:36 contingencyplan joined #parrot
19:42 whiteknight ...and Jaesop has regexes
19:48 dalek jaesop: 588c3fe | Whiteknight++ | stage0/runtime/jsobject.winxed:
19:48 dalek jaesop: First draft, untested, of pcre bindings
19:48 dalek jaesop: review: https://github.com/Whiteknig​ht/jaesop/commit/588c3fe33e
19:48 dalek jaesop: e6fee0e | Whiteknight++ | t/stage0/regexp.t:
19:48 dalek jaesop: Add a quick test file for RegExp
19:48 dalek jaesop: review: https://github.com/Whiteknig​ht/jaesop/commit/e6fee0ec02
20:56 cotto woot
20:57 cotto whiteknight++
20:58 cotto whiteknight, ping
21:08 dalek rakudo/nom: e32b7c2 | jnthn++ | src/core/List (2 files):
21:08 dalek rakudo/nom: Couple of small optimizations in lists.
21:08 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/e32b7c244b
21:08 dalek rakudo/nom: 1c27dea | jnthn++ | src/Perl6/Metamodel/ (2 files):
21:08 dalek rakudo/nom: Make boolification cheaper in cases where things don't override .Bool and instead just rely on the definedness check.
21:08 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/1c27dea077
21:08 dalek rakudo/nom: c857a18 | jnthn++ | src/core/Parcel.pm:
21:08 dalek rakudo/nom: Make parcel assignmnet a little cheaper.
21:08 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/c857a1804e
21:09 benabik Why is dalek faster in #perl6?
21:09 tadzik wow, something is faster in Perl 6!
21:09 contingencyplan joined #parrot
21:14 cotto a bot can't be everywhere at once
21:21 cotto seen dukeleto
21:21 aloha dukeleto was last seen in #perl6 1 days 12 hours ago joining the channel.
21:32 dalek rakudo/nom: dd1307e | jnthn++ | src/core/Str.pm:
21:32 dalek rakudo/nom: Optimize RANGEPOS a little, which improves Str.succ and Str.pred, though still not really enough.
21:32 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/dd1307e447
22:16 whiteknight cotto: pong
22:25 cotto whiteknight, could you look at the remaining blocked message to parrot-dev?
22:25 whiteknight sure, what's wrong with it?
22:26 whiteknight T3H SPAMZ
22:26 whiteknight as far as I am concerned, if they are going to send us T3H SPAMZ, they could at least include some of T3H PR0NZ
22:26 whiteknight so that's just rude
22:27 cotto lulz
22:27 cotto maybe if you send them your cc info
22:29 cotto in pennance, I present you with NONONONO cat: http://boingboing.net/2011/09/30/nononono-cat.html
22:29 tadzik . o O ( please send this message to all your friends, and they will send it to their friends, and eventually the spammer will get this and will be annoyed too )
22:37 dalek jaesop: b993cbc | Whiteknight++ | / (3 files):
22:37 dalek jaesop: Add in some simple definitions of a Process and a FileStream object. Now we can use commandline arguments, standard handles, and other files for basic IO operations. The API is loosly based on the Process and Stream objects from the node.js runtime, with changes because we don't have asynchronicity.
22:37 dalek jaesop: review: https://github.com/Whiteknig​ht/jaesop/commit/b993cbcafa
22:37 whiteknight you can now write a basic "Hello world" program with Jaesop stage 0
22:38 whiteknight it's about damn time
22:40 tadzik :)
22:40 tadzik whiteknight++
22:40 particle joined #parrot
22:42 dalek rakudo/macros: ae1f0a3 | masak++ | / (5 files):
22:42 dalek rakudo/macros: implemented quasi quotes and macros
22:42 dalek rakudo/macros:
22:42 dalek rakudo/macros: This works:
22:42 dalek rakudo/macros:
22:42 dalek rakudo/macros: - Macro declarations
22:42 dalek rakudo/macros: - Calling a macro (using `macro()` and `macro` and operators)
22:42 dalek rakudo/macros: - Quasi quotes
22:42 dalek rakudo/macros: review: https://github.com/rakudo/rakudo/commit/ae1f0a3884
22:47 dalek rakudo/nom: f1eba49 | jnthn++ | src/ (2 files):
22:47 dalek rakudo/nom: Add a couple of ops for checking if we have a VM list/hash.
22:47 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/f1eba49a37
22:47 dalek rakudo/nom: 27090e7 | jnthn++ | src/core/List.pm:
22:47 dalek rakudo/nom: Make List.shift cheaper in the (common) case that we have an element. This in turn makes list assignment cheaper, wich does a gimme to ensure there is an element, followed by a shift.
22:47 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/27090e7edb
23:03 dalek rakudo/nom: 698e301 | jnthn++ | src/core/Array.pm:
23:03 dalek rakudo/nom: Optimize array access a little.
23:03 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/698e30178a
23:17 rfw joined #parrot
23:22 dalek rakudo/nom: cbe898f | jnthn++ | src/core/List.pm:
23:22 dalek rakudo/nom: Use nqp::islist to better handle vivifying list storage.
23:22 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/cbe898f41b
23:28 dalek jaesop: 4a461f2 | Whiteknight++ | stage0/js/ (2 files):
23:28 dalek jaesop: Add in the /regexp/ syntax
23:28 dalek jaesop: review: https://github.com/Whiteknig​ht/jaesop/commit/4a461f2efe
23:31 PacoLinux joined #parrot
23:37 dalek jaesop: 333cb44 | Whiteknight++ | stage0/js/wast.js:
23:37 dalek jaesop: Fix MemberExpr codegen to generate the correct sequence. We want to use .* instead of . in almost all cases that I can imagine.
23:37 dalek jaesop: review: https://github.com/Whiteknig​ht/jaesop/commit/333cb44af3
23:40 preflex_ joined #parrot

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

Parrot | source cross referenced