#parrot Parrot 0.6.4 "St. Vincent Amazon" Released | parrotcode.org/ | 15 new, 648 open, 80 stalled tix | logged in irclog.perlgeek.de/parrot/today
Set by moderator on 28 July 2008.
00:00 alvar joined, Coleoid_afk joined 00:10 AndyA joined
dalek r29923 | coke++ | trunk: 00:17
: RT #57312 - delete languages/pugs
diff: www.parrotvm.org/svn/parrot/revision?rev=29923
r29924 | coke++ | trunk: 00:33
: Missed a pugs reference; make test pass again.
diff: www.parrotvm.org/svn/parrot/revision?rev=29924
r29925 | coke++ | trunk: 00:37
: RT #46909 - We don't need to cope with escaped quotes here, they're already
: dealt with earlier in the build process.
diff: www.parrotvm.org/svn/parrot/revision?rev=29925
neopallium I am trying to debug a problem with the Lua source compiler and need a way to dump the PIR/PASM that is generated. 00:53
The Lua bytecode -> PIR translator is working and doesn't have the same problem. 00:54
00:57 Andy joined
cotto_work neopallium, if you're stuck you should email fperrad. He's the main Lua dev iirc and doesn't visit #parrot 01:00
fperrad? 01:01
purl, fperrad is francois.perrad@gadz.org
purl OK, cotto_work.
cotto_work fperrad?
purl fperrad is, like, mailto:francois.perrad@gadz.org
neopallium thanks I will e-mail him.
cotto_work np 01:02
neopallium but part of my question is just about dumping the PIR/PASM for a compiled function
is that possible?
cotto_work it should work as ../../parrot lua.pbc --target=pir script.lua, but it doesn't 01:07
you'll probably have to ask him about that
actually, you can add -dXX after parrot, where XX is some hex number corresponding to debug flags 01:09
-dff will give you all imcc's debugging output, which will end in the pir it spits out 01:10
../../parrot -dff lua.pbc hello.lua
not quite what you wanted, but close
neopallium I tried the trace flag for opcodes
but it is hard to read that output, since I don't know what parts where generated from the lua source 01:11
I will try the imcc's flags and see if that helps. Thanks 01:12
cotto_work you're welcome 01:14
neopallium this command worked for dumping the broken PIR code: ../../parrot luap.pir --target=pir script.lua 01:15
it was luap.pir instead of lua.pir that supports the --target=pir option 01:16
seneca Wow... make is much faster on my new hardware. I just got a build done in about the same duration as the old box took for Configure.pl, and I didn't even remember to add in some -j value
seneca really has to switch to a properly working client at some point, as she's really tetragon 01:17
cotto_work (tetragon's new hardware)++
tetragon (tetragon's current IRC client)-- 01:18
Tene Ew, I've heard bad things about colloquy. 01:19
tetragon It's really annoying
The only reason I'm trying it is that my past usual IRC client (X-Chat Aqua) is pretty much unmaintained, and the OS X users I spoke to in the past seemed to like it 01:20
dalek r29926 | jkeenan++ | darwin_fix_hints: 01:25
: Creating darwin_fix_hints in svn.perl.org/parrot/branches
diff: www.parrotvm.org/svn/parrot/revision?rev=29926
r29927 | jkeenan++ | darwin_fix_hints-29925:
: Tagging trunk at r29925 so that the darwin_fix_hints can later be synched to it.
diff: www.parrotvm.org/svn/parrot/revision?rev=29927
kid51 tetragon: I am working on a patch for rt.perl.org/rt3/Ticket/Display.html?id=41508, and when it's ready for submission I would like you to test it out. 01:27
Should be ready sometime tomorrow.
tetragon kid51: Hopefully it doesn't require a PPC I no longer have
And when I last played with GMP and Configure.pl, it was to get the thing to recognise my framework build (didn't get too far into that). 01:30
kid51 No, I don't think it's PPC-specific.
Though I mourn the loss of your PPC box, because you knew more about those PPC vs. Intel issues than practically anybody. 01:31
I still have uncommitted tickets awaiting feedback from people who attended YAPC buildfest.
tetragon Bleh, still getting the readline/libedit conflict
And I'm currently looking for a used PPC (and considering asking at my office if I could use the spare cycles in my PPC grid over lunches) 01:32
kid51 Plus ļæ½a change ...
dalek r29928 | jkeenan++ | darwin_fix_hints: 01:33
: Refactor, and write tests for, config/init/hints/darwin.pm. Partially complete.
diff: www.parrotvm.org/svn/parrot/revision?rev=29928
tetragon ponders PearPC 01:36
kid51 must go grocery shopping 01:56
dalek r29929 | jkeenan++ | darwin_fix_hints: 01:57
: Add one more test for init::hints::darwin::_get_adjusted_user_options().
diff: www.parrotvm.org/svn/parrot/revision?rev=29929
r29930 | chromatic++ | pdd25cx: 02:11
: [src] Fixed C++ compilation by casting to the proper type.
diff: www.parrotvm.org/svn/parrot/revision?rev=29930
r29931 | chromatic++ | pdd25cx: 02:15
: [IMCC] Fixed compilation by prepending EXCEPTION_ to all exception enums.
diff: www.parrotvm.org/svn/parrot/revision?rev=29931
02:37 apeiron joined
s1n is it possible to have the harness (lib/Parrot/Test.pm) not redirect std(out|err) so the individual test results could be monitored? 02:49
cotto_home s1n, you can run them with perl instead of prove 03:12
s1n cotto_home: i'm really looking at a Test/Harness.pm change so it can be a "make localtest" with verbosity=>1 set 03:14
03:17 kid51 joined
dalek r29932 | chromatic++ | trunk: 04:17
: [editor] Added :lexid to Vim syntax highlighting.
diff: www.parrotvm.org/svn/parrot/revision?rev=29932
s1n to answer my own question, i submitted a patch: rt.perl.org/rt3/Public/Bug/Display.html?id=57522
04:19 Andy joined
dalek r29933 | chromatic++ | pdd25cx: 04:20
: [PCT] Changed generated code which sets the type of Exceptions to match PDD 23
: style (s/_type/type/). This fixes Rakudo, and probably everything else which
: throws exceptions ultimately generated by PCT.
diff: www.parrotvm.org/svn/parrot/revision?rev=29933
04:25 Lorn joined 04:28 rurban_ joined 04:38 tetragon joined 05:45 Psyche^ joined 06:06 cotto_home joined
pmichaud Aloha! 06:13
Tene pmichaud! 06:14
purl i think pmichaud is www.pmichaud.com/ or "Patrick R. Michaud" <mailto:pmichaud@pobox.com>
Tene How's the vacation? 06:15
cotto_home you're not supposed to be here on vacation. I suppose next we're going to see a long series of commits from you too. ;) 06:16
Tene What are you talking about, cotto? That's the best kind of vacation! 06:17
cotto_home or if you're audreyt, the best kind of shower. 06:18
pmichaud the vacation is going nicely. Currently I'm sitting in our (rented) house overlooking the Pacific ocean, about 50 ft away from the surf. :) 06:19
Tene nice
pmichaud with wifi. :-)
commits will probably come tomorrow. Or maybe Sunday. 06:20
This place shuts down on Sunday, so the kids will undoubtedly want to play in the surf and so there won't be much else for me to do but sit here on this nice porch and hack. :-P 06:21
(This place == Laie, Hawaii == everything is closed on Sunday.) 06:22
cotto_home which part of Laie? 06:24
pmichaud south... just across from the Polynesian Cultural Center
cotto_home and you've got 2 weeks there? 06:25
pmichaud maps.google.com/maps?f=q&hl=en&...iwloc=addr
no, only until Monday
on Monday we head to Hilo and the Big Island 06:26
so, 3 days here
cotto_home looks like a very nice spot
enjoy yourself! 06:27
pmichaud it is, we will
cotto_home "This article is about the Haskell interpreter. For the affectionate physical gesture, see Hug." 06:35
Wikipedia makes me happy sometimes.
07:29 alvar joined
dalek r29934 | allison++ | pdd25cx: 08:19
: [pdd25cx] Fixing Lua's exception handling to work with the new model, by
: significantly lightening the execution of 'docall'.
diff: www.parrotvm.org/svn/parrot/revision?rev=29934
08:41 Ron joined
dalek r29935 | allison++ | pdd25cx: 09:13
: [pdd25cx] Eliminating 'tcl_error_sc' opcode from PIR files, as it doesn't exist
: anymore.
diff: www.parrotvm.org/svn/parrot/revision?rev=29935
09:35 alvar joined 09:52 ambs joined 09:58 pjcj joined
dalek r29936 | petergibbs++ | trunk: 10:01
: [jako] Fix factorial test, and allow for different answers for 32 vs 64 bits
diff: www.parrotvm.org/svn/parrot/revision?rev=29936
10:10 masak joined 11:58 kid51 joined 11:59 cognominal joined
rurban kid51: the freebsd ticket is ok 12:12
kid51: other tickets - [perl #39742] [BUG] installed parrot conflicts with dev parrot this works fine on mingw + cygwin. 12:14
-L/usr/local/lib from perl5 ldflags doesn't pull in the installed dll
kid51: all freebsd 7 tests passed. 12:22
12:28 rurban_ joined 12:31 kid51 joined 12:32 Senaka joined, Casan joined
Senaka nopaste: test 12:33
purl: nopaste 12:35
purl nopaste is at nopaste.snit.ch/ (ask TonyC for new channels) or rafb.net/paste or poundperl.pastebin.com/ or paste.scsys.co.uk/ or App::Nopaste or tools/dev/nopaste.pl
Senaka is puzzled... does nopaste work?
nopaste "Infinoid" at 75.5.242.0 pasted "Nopaste works." (1 line) at nopaste.snit.ch/13719
masak could someone explain to me a bit about rakudo and list assignment? 12:37
what's planned for implementing the latter in the former, I mean.
nopaste "kid51" at 71.247.43.105 pasted "kid51 tries nopaste" (1 line) at nopaste.snit.ch/13720
kid51 worksforme 12:38
Senaka kid51: yes I was expecting it to work the other way
like I paste and it posts on nopaste.snit.ch
kid51 Well, that's what it does. My paste is posted at nopaste.snit.ch/13720 12:39
Infinoid he's talking about pasting in an IRC msg window, I think 12:40
kid51 Ah! Yes, *that* won't work.
Senaka Infinoid: you got me
masak that's why it's called "nopaste", because it avoids pasting in IRC 12:41
Senaka masak: well yes... so the bot i'm talking about will paste a private conversation with it by user X for time N in site s 12:42
masak private conversation, that could work, yes 12:43
but it does sound error-prone, what with IRC message maxlengths and all
Senaka ļ»æmasak: does it clip off?
Infinoid yeah 12:44
there's a message size limit of 250 bytes or whatever it is.
(per line, I mean)
Senaka or else does it like get into a new line?
Infinoid no, it clips
Senaka Infinoid... oops 12:45
ok so if I need a nopaste like behaviour... i will have to do it like in the way perl does it? 12:46
Infinoid I don't understand... in what way does perl do it? 12:49
Senaka ok nopaste.snit.ch has perl channels only isn't it? or am I wrong? 12:50
Infinoid I think nopaste.snit.ch is only on the magnet irc network, so it's gonna be a bit biased that way, yes
13:12 Whiteknight joined
dalek r29937 | fperrad++ | trunk: 14:15
: [Lua] 64bits fix
: patch from Robert G. Jakabosky
: see rt.perl.org/rt3/Public/Bug/Display.html?id=57504
diff: www.parrotvm.org/svn/parrot/revision?rev=29937
14:19 Senaka left 14:28 Andy joined
dalek r29938 | petergibbs++ | trunk: 14:46
: [lua] Tweak Robert Jakabosky's patch for the os.time test so that it works
: for all time zones.
diff: www.parrotvm.org/svn/parrot/revision?rev=29938
r29939 | coke++ | pdd25cx: 14:49
: un-TODO passing test.
diff: www.parrotvm.org/svn/parrot/revision?rev=29939
15:07 ambs joined 15:08 davidfetter joined 15:19 gryphon__ joined
dalek r29940 | coke++ | pdd25cx: 15:21
: [tcl] The error message is actually in the 'message' slot now; having the wrong name here was causing segfaults (see RT #56786). All tests now pass in trunk.
diff: www.parrotvm.org/svn/parrot/revision?rev=29940
15:33 Theory joined 15:41 Andy joined 15:42 jhorwitz joined 16:12 tetragon joined
dalek r29941 | allison++ | pdd25cx: 16:19
: [pdd25cx] Change the Exception PMC to throw an exception when attempting to
: access a non-existent attribute.
diff: www.parrotvm.org/svn/parrot/revision?rev=29941
16:20 iblechbot joined
dalek r29942 | allison++ | pdd25cx: 16:31
: [pdd25cx] Add regression test for exception attribute access segfault.
diff: www.parrotvm.org/svn/parrot/revision?rev=29942
16:33 alvar joined 17:14 tetragon joined
tetragon Ah... my byteorder is '12345678' instead of '1234' 17:16
17:35 ambs joined
pmichaud Aloha 17:40
ambs Aloha, pmichaud!
17:43 tetragon joined 17:45 ruoso joined
cognominal maintenant que OSCON est terminĆ©, sa principale organisatrice, Allison, se lache et produit patche aprĆØs patche pour parrot. 17:52
oops
tetragon Just waiting for patches now? 17:55
tetragon grumbles at the listed price for used PPC Macs 17:56
18:05 Senak1 joined
dalek r29943 | allison++ | pdd25cx: 18:10
: [pdd25cx] Bringing the pdd25cx branch up-to-date with trunk r29866.
diff: www.parrotvm.org/svn/parrot/revision?rev=29943
r29944 | allison++ | pdd25cx: 18:28
: [pdd25cx] Bringing the pdd25cx branch up-to-date with trunk r29867 (without
: updating the [R]eplaced test files).
diff: www.parrotvm.org/svn/parrot/revision?rev=29944
18:28 Senak1 left 18:35 Ron joined, Senak1 joined 18:36 Senak1 left
dalek r29945 | allison++ | pdd25cx: 18:41
: [pdd25cx] Bringing the pdd25cx branch up-to-date with trunk r29867 (manually
: updating [R]eplaced test files).
diff: www.parrotvm.org/svn/parrot/revision?rev=29945
r29946 | allison++ | pdd25cx: 18:51
: [pdd25cx] Bringing the pdd25cx branch up-to-date with trunk r29945.
diff: www.parrotvm.org/svn/parrot/revision?rev=29946
r29947 | chromatic++ | pdd25cx: 19:06
: [Rakudo] Fixed the exception-type setting code to match PDD 23 exceptions
: (s/_type/type/).
diff: www.parrotvm.org/svn/parrot/revision?rev=29947
r29948 | chromatic++ | pdd25cx: 19:08
: [IMCC] Updated IMCC lexer to compile from branch merge (renamed exception type
: constants to use the EXCEPTION_ prefix) and regenerated lexer.
diff: www.parrotvm.org/svn/parrot/revision?rev=29948
19:15 clunker3 joined
dalek r29949 | chromatic++ | pdd25cx: 19:16
: [t] Fixed t/steps/auto_pmc-01.t; it contains two hard-coded lists of PMC files,
: for some reason, and the ExceptionHandler PMC is now in
: src/pmc/exceptionhandler.pmc, not src/pmc/exception_handler.pmc.
: This doesn't answer the question of *why* this test needs to contain two
: hard-coded lists of PMC files.
diff: www.parrotvm.org/svn/parrot/revision?rev=29949
19:19 Lorn joined
tetragon Hrm... getting a few segfaults from make test on my 64-bit build (and one during Configure) 19:31
Ones where the crash reporter couldn't even get a stack trace
And no, the triangle no longer spins for me 19:34
Infinoid ohnoes 19:44
tetragon Gah, gdb isn't giving me a decent backtrace (but it is better than the crash reporter's one) 19:46
nopaste "tetragon" at 76.10.153.9 pasted "Ugly segfault running shootout test 19 with a 64-bit Intel OS X build" (62 lines) at nopaste.snit.ch/13722 19:49
tetragon Erm... included the wrong test's registers 19:51
nopaste "tetragon" at 76.10.153.9 pasted "The correct crash reporter output for that segfault" (46 lines) at nopaste.snit.ch/13723 19:52
japhb casts a bleary-eyed glance at IRC ... 19:53
tetragon, am I correct in assuming 'the triangle no longer spins for me' was just a way of saying that things were in general broken, and that you're not having an OpenGL-specific problem? 19:54
tetragon Everything seems to be working in my 32-bit build, but the 64-bit build is tad broken 19:55
japhb nod
nopaste "tetragon" at 76.10.153.9 pasted "Trying to run the OpenGL examples with the 64-bit build" (11 lines) at nopaste.snit.ch/13724 19:56
Whiteknight I'm dreading trying to update my branch to account for all the pdd25cx changes
tetragon My system even segfaults during Configure (the MMX check) 19:57
japhb ouch 19:58
20:18 japhb joined 20:28 rurban_ joined
dalek r29950 | pmichaud++ | trunk: 20:49
: [rakudo]: Update atan2 builtin with correct arguments.
diff: www.parrotvm.org/svn/parrot/revision?rev=29950
20:52 rurban joined
dalek r29951 | pmichaud++ | trunk: 21:00
: [rakudo]: spectest-progress.csv: 116 files, 1998 passing, 3 failing
diff: www.parrotvm.org/svn/parrot/revision?rev=29951
pmichaud rakudo now passing 2012 tests. :-) 21:13
(r29950)
afk, time for culture. :-) 21:16
tetragon Culture, isn't that what's growing in my fridge? 21:20
21:29 kid51 joined, ruoso joined
tetragon kid51: Prepare for more OS X tickets, I've prepared a crashy, atypical build of parrot 21:41
22:02 teknomunk joined
dalek r29952 | allison++ | trunk: 22:46
: [pdd17pmc] Merging the pdd25cx branch into trunk for r26555 to r29949.
diff: www.parrotvm.org/svn/parrot/revision?rev=29952
jonathan allison++ 22:48
dalek r29953 | allison++ | pdd25cx: 22:52
: Removing concurrency development branch from the repository
diff: www.parrotvm.org/svn/parrot/revision?rev=29953
tetragon My 32-bit build broke 22:54
nopaste "tetragon" at 76.10.153.9 pasted "Broken 32-bit build on Intel OS X" (10 lines) at nopaste.snit.ch/13725 22:56
dalek r29954 | chromatic++ | trunk: 23:03
: [config] Fixed crypto PMC configuration when building without crypto libs
: (reported by Seneca Cunningham).
diff: www.parrotvm.org/svn/parrot/revision?rev=29954
tetragon Built this time 23:06
dalek r29955 | chromatic++ | trunk: 23:10
: [lib] Made Parrot::Configure::Compiler add read-only editor flag to generated
: files.
diff: www.parrotvm.org/svn/parrot/revision?rev=29955
r29956 | chromatic++ | trunk: 23:12
: [config] Fixed two uninitialized variable warnings.
diff: www.parrotvm.org/svn/parrot/revision?rev=29956
tetragon All the crash reports generated by make test save for one look to be the usual ones I encounter
tetragon grumbles that the new box runs through make test and the crash reporter too quickly to easily identify the offending set of tests 23:13
t/pmc/threads.t, test 13 23:15
kid51 looks at #parrot for first time in 2 hours 23:17
nopaste "tetragon" at 76.10.153.9 pasted "t/pmc/threads.t test 13 crash" (181 lines) at nopaste.snit.ch/13726 23:18
23:19 alvar joined
kid51 A box that runs 'make test' too fast -- now there's something to cry over ;-) 23:27
tetragon I grew used to the 20s delay while the crash reporter was grinding away 23:28
Whiteknight allison++ and chromatic++ 23:40
dalek r29957 | jkeenan++ | darwin_fix_hints: 23:44
: Integrate value for ccflags supplied by user on command-line with that located via init::defaults.
diff: www.parrotvm.org/svn/parrot/revision?rev=29957
r29958 | jkeenan++ | darwin_fix_hints: 23:47
: Eliminate use of values supplied by user on command-line. These will be
: handled by inter::progs. Refactor more code from
: init::hints::darwin::runstep() into internal subs, then test those internal
: subs in t/steps/init_hints-02.t.
diff: www.parrotvm.org/svn/parrot/revision?rev=29958
tetragon kid51: Will proper framework support be added? 23:48
kid51 tetragon: What do you mean?
tetragon For instance, my late iBook had a framework build of GMP
The only time parrot used it was when I modified the appropriate step to change the flags and locations 23:49
A number of the libraries parrot can use could be in either framework or Unix-style builds 23:50
kid51 Still not clear what you mean by framework. This is new vocabulary for me. 23:51
tetragon A decent number of libraries (such as OpenGL) on OS X boxes are prepared as framework builds
nopaste "kid51" at 70.107.14.221 pasted "svn diff between trunk and darwin_fix_hints branch" (512 lines) at nopaste.snit.ch/13727 23:52
kid51 In any event, this paste represents what I was talking about in my email this morning:
tetragon These have different structures and take different compiler and linker flags to work with than the better handled Unix-style libraries
Whiteknight starts the arduous update process 23:53
tetragon Instead of, say -lgmp, I would use -framework GMP
23:54 ruoso joined
kid51 Well, all I can say is: Proper framework support will be added when the people who want it and know how to do it have sufficient tuits. 23:54
Translation: It will be done when *you* do it. 23:55
tetragon I was looking at aspects of it when my iBook went away
And right now, I'm using command line flags to get my 64-bit build going 23:56
kid51 has a solution to all our Mac problems: Assemble Macs representing all combinations of 10.4 vs 10.5; ppc vs. intel; 32-bit vs 64-bit. Place in room. Lock Allison and tetragon in said room. Unlock when Mac problems solved. :-)
tetragon At work I deal with all all combinations on 64-bit save for 10.5 Intel, but my MacBook Pro fills that blank in what I handle on an everyday basis 23:58
kid51 The first step would be to file a [TODO] RT which outlines the issues. People like $self who merely use Macs are blissfully unaware of many of these issues (similar to editline vs. readline).
tetragon Which still causes a crash report on my box since I haven't downloaded my patch yet
kid51 In any case, the patch that I just pasted -- can you check it out to see that it doesn't damage your objective of stripping confusing -arch flags? 23:59

Powered by Cro and the RakuĀ® Programming Language.

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

Thank you!