Parrot 2.5.0 release time is on 15th at 12:00 UTC | parrot.org | Log: irclog.perlgeek.de/parrot/today | Nopaste: nopaste.snit.ch:8001 | GSOC Students: trac.parrot.org/parrot/wiki/GSoCersStartHere | Priorities: fix io_ops mess in corevm/coretest, review and update documentation before release
Set by moderator on 9 June 2010.
Coke msg NotFound does 'finalize' resolve the issues in TT #1151? 00:09
purl Message for notfound stored.
00:18 snarkyboojum joined
dalek TT #1301 closed by coke++: Update RT references in repository/wiki 00:21
TT #1301: trac.parrot.org/parrot/ticket/1301
00:37 bubaflub joined, hudnix joined
bubaflub can someone help me diagnose what's wrong with this? i'm trying to build Parrot and it throws this error: gist.github.com/421722/ 00:38
00:41 dduncan joined 00:44 dduncan left
sorear bubaflub: looks like your compiler doesn't provide limits.h, or else Configure didn't detect it 00:47
bubaflub sorear: yeah, i'm messing with the configuration settings
00:47 whiteknight joined
bubaflub to get cross-compilation working, i'll look into finding limits.h 00:47
thaks
sorear Parrot has helpfully noticed this, and, mediated by the non-definition of PARROT_HAS_HEADER_LIMITS, not tried to include it
00:50 kid51 joined
bubaflub sorear: it looks like it can't find execinfo.h and dlfcn.h - perhaps i've overwritten an include directory? 00:53
sorear dlfcn is not an ANSI C header file 00:55
it's part of the POSIX dynamic linking and loading optional extension
it's entirely reasonable for it to be missing if you're compiling for a very minimal OS
bubaflub yes, and for parts i have to manually override the config file 00:56
sorear oh wow fail 00:57
feature.h line 51
we're unconditionally including dlfcn
ah, not quite unconditionally
only if PARROT_HAS_BACKTRACE
bubaflub ok 00:58
sorear since supporting PARROT_HAS_BACKTRACE requires dlfcn.h and execinfo.h, you should change Configure to only define HAS_BACKTRACE if all three are available
or maybe change features.h to test all three
bubaflub i can turn off backtrace for now
sorear are you compiling against rtems or are you just fiddling config settings of a normal build? 00:59
bubaflub compiling against RTEMS
(it's my GSoC project)
((with RTEMS, not PaFo) 01:00
)
sorear yeah, but didn't expect you to have gotten that far yet 01:02
bubaflub ah, i didn't know anyone (besides my mentors) cared / was tracking my project 01:03
so i've got my dev system setup on Fedora Core 12 in Virtual Box, and have the full RTEMS stuff installed in there
the last two weeks i worked on getting a script (actually a Makefile, but that's temporary) to pull as much info as I could about the installed RTEMS to make a Configure.pl file and a hints file 01:04
now i'm just ironing everything out so parrot will build clean
dalek rrot: r47581 | coke++ | trunk/tools/dev/vgp:
My copy of valgrind only goes to 50.
 01:10
rrot: r47582 | coke++ | trunk/tools/dev/mk_manifest_and_skip.pl:
We're moving towards .git; no point in soliciting git-svn in the meantime.
rrot: r47583 | coke++ | trunk (6 files):
remove PARROT_NET_DEVEL-
TT #534 closed by coke++: Get rid of PARROT_NET_DEVEL 01:11
TT #534: trac.parrot.org/parrot/ticket/534
TT #989 closed by jkeenan++: t/compilers/tge/grammar.t: FAIL under 'make test', but passes with ... 01:28
TT #989: trac.parrot.org/parrot/ticket/989
01:35 Andy joined
dalek rrot: r47584 | coke++ | trunk/docs/pdds/draft/pdd06_pasm.pod:
Document the 100 register limit in PASM.
 01:43
rrot: r47585 | coke++ | trunk/t/compilers/imcc/syn/regressions.t:
PASM does have a hardcoded limit.
rrot: r47586 | coke++ | trunk/docs/pdds/draft/pdd06_pasm.pod:
Remove some cruft from PDD 06 - if we want a list of opcodes, we can use
01:44 eternaleye joined
dalek TT #1025 closed by coke++: PASM Registers limited to 2 digits. 01:44
TT #1025: trac.parrot.org/parrot/ticket/1025
TT #949 closed by coke++: Make tools/dev/mk_manifest_and_skip.pl work with git svn
TT #949: trac.parrot.org/parrot/ticket/949
bubaflub sorear: i've cleared those errors by setting the right configuration file settings, any ideas about this one? gist.github.com/421722/ looks to be related to IO subsystem 01:45
sorear PIO_STDIN and friends are not defined anywhere 01:54
try changing to _PIO_STDIN etc
bubaflub is there a way to flag that from the config options? 01:56
or a hints file?
purl hmmm... a hints file is busted
bubaflub purl, forget a hints file
purl bubaflub: I forgot hints file
Coke msg kid51 re: tt #1068 - optimizing for tab expansion is only a nice to have. that said, if you want to move that test file, go ahead. if you dont' care any more, please close the ticket. thanks. 01:57
purl Message for kid51 stored.
kid51 Coke: AAMOF, I was looking at that ticket just 20 minutes ago. 01:59
Waiting for a 'make test' to finish, then will do that move. 02:00
bubaflub sorear: i think i figured it out, i'm doing a realclean and a build to check 02:01
dalek TT #1668 closed by coke++: NQP: Problem with arrays when unshifting/pushing
TT #1668: trac.parrot.org/parrot/ticket/1668
TT #741 closed by coke++: PIR line numbers in backtraces are substantially wrong.
TT #741: trac.parrot.org/parrot/ticket/741
sorear bubaflub: the code looks completely wrong. You're probably the first person to test Parrot on a computer that's not _WIN32 and doesn't have unistd.h
unless, of course, you *have* unistd.h, in which case yes you can flag that by changing the options 02:02
bubaflub sorear: yeah, i flagged as having that 02:03
now i'm getting a different error
src/io/socket_api.c:49: error: 'AF_LOCAL' undeclared here (not in a function) 02:05
which is strange because i'm not seeing an AF_LOCAL *anywhere*
02:05 szabgabx_ joined
bubaflub PF_LOCAL all over the place, but not AF_LOCAL 02:05
you can tell i've rarely ventured into Parrot guts
tcurtis msg chromatic I have PAST::Pattern seemingly finished. I'm in the process of writing documentation. github.com/ekiru/nqp-rx/ is the repo I've begun writing and testing optimizations in. 02:07
purl Message for chromatic stored.
Coke AAMOF? 02:12
purl AAMOF is as a matter of fact
Coke ahhh
serendipitious.
dalek rrot: r47587 | jkeenan++ | trunk (1 files):
Reposition test files objects.t under t/oo/ rather than t/pmc/, per
 02:16
TT #1068 closed by jkeenan++: Rename test files whose names are too similar 02:18
TT #1068: trac.parrot.org/parrot/ticket/1068
Coke down to 638 tickets... 02:19
02:36 janus joined
dalek rrot: r47588 | coke++ | trunk/config/gen/config_pm.pm:
Not every bit of config information is ASCII-safe.
 02:49
rrot: r47589 | coke++ | trunk/.gitignore:
Regen file
TT #888 closed by coke++: [BUG]parrot make failed with chinese TEMP PATH 02:50
TT #888: trac.parrot.org/parrot/ticket/888
TT #1678 created by jkeenan++: t/pir/macro.t: test 'basic macro with argument' fails on Darwin/PPC 02:51
TT #1678: trac.parrot.org/parrot/ticket/1678
Coke rant: make html hides everything inside a perl script instead of using make. 02:57
dalek rrot: r47590 | coke++ | trunk/lib/Parrot/Docs/Section/Parrot.pm:
remove commented out code and SEE ALSO section that duplicates the use list.
 03:05
03:10 patspam joined
Coke ugh. make html is byzantine. 03:13
kid51: are the lib/Parrot/Docs/* classes used for anything other than "make html" ?
kid51 Coke: I agree with your last two comments. 03:16
I've been looking at Parrot::Docs for 4 years and my head still hurts 03:17
Answer to your question: I don't think so, but I can't conclusively prove my belief
03:19 bluescreen joined
kid51 likes the Delete button in Trac. Enabled me to delete a completely misleading post (by me) 03:20
Coke kid51: I am tempted to rip that all out and start over. (make html) 03:33
but that's a lot of work for soemthing I hate.
cotto: ping 03:34
cotto: deping 03:39
bubaflub Coke: i'm currently in shoulder-deep in Configure.pl and Makefiles, do you want me to take a swing at make html? 03:45
kid51 bubaflub: what are you attempting there? 03:47
bubaflub kid51: for my GSoC project i'm trying to get parrot to configure/build/run on RTEMS
i have to play with Configure.pl and hints files 03:48
kid51 Well, I think that 'make html' would just be a big time-suck for you.
It merely takes the docs that appear in one format and rewrites them in another. 03:49
IIRC, very little of it is in the config system per se, because it invokes a perl script
bubaflub kid51: ah, ok. 03:50
kid51 So it's a very different area from what you're working on. You probably are one of the few people who could work on cross-build things.
We could get any Perl 5 person (with too much time on his hands) to work on make html. 03:51
There was a time a couple of years back when I considered working on it ....
... but it seemed like it would be a lot of work for little value added ...
... so I became the maintainer of Configure instead ;-) 03:52
kid51 must sleep
purl $kid51->sleep(8 * 3600);
dalek rrot: r47591 | coke++ | trunk/tools/dev/pprof2cg.pl:
Make script give usage when invoked incorrectly instead of exploding.
 03:55
Coke is sad that the profiler is still borked. :P
bubaflub: you can take TT#1516, though. =-) 03:59
bubaflub Coke: if time allows that'll be part of my GSoC work 04:01
dalek rrot: r47592 | coke++ | trunk/DEPRECATED.pod:
This appears to be complete. Closing ticket.
 04:11
TT #1260 closed by coke++: bitwise ops and VTABLE functions 04:13
TT #1260: trac.parrot.org/parrot/ticket/1260
04:29 LoganLK joined 04:39 JimmyZ joined
JimmyZ the news Improved implemented PAST::Pattern, is this about gsoc_past_optimization branch? 04:41
05:06 plobsing joined
Coke if so, it should be backed out of NEWS. 05:06
NEWS == trunk
JimmyZ commit, reviews welcome 05:07
*committed
05:10 theory joined
dalek rrot: r47593 | jimmy++ | trunk/NEWS:
cleaned up NEWS which is advancing or old or not about parrot.
 05:17
rrot: r47594 | jimmy++ | trunk/NEWS:
re-ordered NEWS
05:46 particle joined 07:12 mmcleric joined 07:42 JimmyZ joined 07:50 snarkyboojum joined
cotto anyone know what "take care by packing" in NEWS means? 08:10
JimmyZ I don't know, something like making rakudo use op2c? 08:12
cotto I don't like putting items with "almost" in NEWS. 08:19
JimmyZ I don't think it's a parrot news. 08:22
it's RTEMS news? 08:23
cotto We need to figure out a standard for what makes interesting NEWS items. 08:37
cotto wishes he had something in mind. 08:38
mikehh needs to re-boot 08:41
dalek rrot: r47595 | cotto++ | trunk/NEWS:
[NEWS] Make news more relevant to people who aren't deeply into Parrot but are curious.
 08:51
rrot: r47596 | jimmy++ | trunk/NEWS:
correct typo
08:51 mikehh joined
cotto Jimmy++ for catching that. I had to read the diff 3 times before I spotted the goof. 08:53
08:54 mmcleric joined
JimmyZ ^_^ 08:55
mikehh me too 09:03
09:20 lucian joined
dalek kudo: b931990 | moritz++ | (2 files):
try to make callframe().my.<$x> read-only

you can do to the caller's lexpad Also run new callframe.t		 09:24
09:28 snarkyboojum joined
bacek aloha, humans 09:49
JimmyZ aloha, bacek 09:50
bacek aloha, JimmyZ
mikehh aloha, bacek 09:55
nopaste "mikehh" at 192.168.1.3 pasted "t/compilers/imcc/syn/regressions.t - Failed test: 18 in testr" (37 lines) at nopaste.snit.ch/21212 10:10
mikehh t/compilers/imcc/syn/regressions.t - Failed test: 18 in testr - see nopaste.snit.ch/21212 10:13
all other tests PASS (pre/post-config, make corevm/make coretest, smoke (#34340), fulltest) at r47596 - Ubuntu 10.04 i386 (g++)
t/op/exit.t - TODO passed: 6 in testf
also note that the TODO pass does not happen on amd64
10:17 dmagnus__ joined 10:21 whiteknight joined
whiteknight good morning, #parrot 10:32
mikehh aloha, whiteknight 10:33
dalek kudo: 197b4ba | (Patrick Abi Salloum)++ | src/core/MAIN.pm:
output default usage message to stderr unless --help is specified
10:47 lucian_ joined
mikehh All tests PASS (pre/post-config, make corevm/make coretest, smoke (#34341), fulltest) at r47596 - Ubuntu 10.04 i386 (gcc with --optimize) 10:53
t/op/exit.t - TODO passed: 6 in testf
whiteknight Good morning, mikehh 10:54
whiteknight thinks there should be a setting where small ticket changes, like changing the milestone, don't send emails to the list
dalek p-rx: 1a63388 | bacek++ | (2 files):
Actually provide slurp in setting library.
 10:55
bacek msg pmichaud iwbn, if perl6/nqp will catch '=end' without identified and warn user about it... 11:00
purl Message for pmichaud stored.
dalek rrot: r47597 | bacek++ | trunk/ext/nqp-rx/src/stage0/nqp-setting.nqp:
Bring new nqp setting
 11:02
11:15 mmcleric joined
dalek r: 419157a | bacek++ | t/ (2 files):
Add last test from alpha's t/data/assign
 11:39
r: 6597565 | bacek++ | t/parse.nqp:
Add simple grammar to move tests into datafiles.
r: 84b382f | bacek++ | t/ (2 files):
Convert old call test
r: 64aa2d9 | bacek++ | t/ (2 files):
Move almost all logic of running tests from datafile into t/parse.nqp
r: 19c2016 | bacek++ | t/data/call.pir:
Remove old testfile
r: 6f6a78e | bacek++ | t/data/call.txt:
Split big test into smaller chunks.
r: a0baf89 | bacek++ | TODO:
Add TODO item.
mikehh sometimes I hate the world or whatever 11:43
i have two different parrot directories and i usually test with g++ in on and gcc in the other 11:44
dalek r: 5befded | bacek++ | t/parse.nqp:
Consmetic changes in data parser.
mikehh got an error in the one directory this morning with g++ and not the other with gcc, but when I tested with g++ in the second directory, no error 11:45
11:46 mmcleric joined
mikehh i have just done clean checkouts and do not get the error 11:46
i nearly always do a make realclean but i still don't see what went wrong 11:47
mikehh need a break - bbl 11:48
12:07 cognominal joined
dalek r: 2503052 | bacek++ | src/PIR/Compiler/Grammar.pm:
Parse fatarrow in params.
 12:07
r: b884606 | bacek++ | t/data/call.txt:
Add more tests.
r: 987d8eb | bacek++ | src/PIR/Compiler/Grammar.pm:
Fix # comments parsing
r: 7c4a328 | bacek++ | src/PIR/Compiler/Grammar.pm:
And more fixes to # comments.
 12:13
whiteknight mikehh: weird 12:16
12:52 Util_ joined 13:12 kid51 joined 13:20 JimmyZ joined 14:21 tetragon joined 14:26 mikehh joined 14:44 dmagnus__ joined, particle joined, eternaleye joined, aukjan joined, mj41 joined, dukeleto joined, dalek joined, he joined, baest joined, NotFound joined, arnsholt joined, athomason joined 15:06 fperrad joined
dalek rrot: r47599 | darbelo++ | branches/gsoc_nfg (2 files):
Add a failing test for NFG 'literals'.
 15:10
15:12 fperrad_ joined
dalek r: 6ba8b94 | bacek++ | t/data/compunit.txt:
Update testdata to modern PIR.
 15:12
r: 964cc12 | bacek++ | (2 files):
Fix filepaths after shuffling files.
r: 6a5f93c | bacek++ | src/PIR/Grammar.pm:
Implement top-level .line directive
r: f5a5956 | bacek++ | t/data/compunit.txt:
Remove TODO from test.
r: ad1049e | bacek++ | t/ (3 files):
More tests from alpha.
r: 13396e2 | bacek++ | t/ (2 files):
Get more tests from alpha.
r: c5037f2 | bacek++ | src/PIR/Grammar.pm:
Update for sub pragmas.
r: f8f6d11 | bacek++ | src/PIR/Grammar.pm:
More ws fixes.
r: 91f09ff | bacek++ | t/ (4 files):
More tests from alpha
r: f820287 | bacek++ | t/data/sub.txt:
Forgotten sub.txt
rrot: r47600 | coke++ | trunk/DEPRECATED.pod:
2.6 is a supported release - can't remove until after that goes out.
 15:26
rrot: r47601 | whiteknight++ | branches/oo_func_cleanup:
Branch is unused, will start a new branch when it's time to do this work
fperrad seen dami 15:37
purl dami was last seen on #dbix-class 1 years, 73 days, 19 hours, 49 minutes and 35 seconds ago, saying: arcanez: or keep an internal stack of $logic values within the object [Mar 31 19:47:25 2009]
pmichaud msg bacek currently, =end without identifier is perfectly legal in Perl 6. 15:40
purl Message for bacek stored.
bacek pmichaud, erm... =end without identifier doesn't end pod block. 15:43
pmichaud it ends a pod block that starts with =begin and no identifier.
as I said, it's currently legal in Perl 6. 15:44
bacek But inside =begin foo it's ignored...
And my vim syntax highlighter isn't smart enough to mark whole block as pod. 15:45
pmichaud right, because pod blocks can nest.
I'll bring it up with the Perl 6 folks.
bacek github.com/perl6/nqp-rx/commit/1a63...16c3a3e36d
pmichaud yes, I knew that one was there. 15:46
bacek I spent about 15 minutes trying to find "slurp".
It actually wasn't.
pmichaud yes, I know exactly the problem you're talking about -- it cost me 20 minutes once.
bacek :)
pmichaud I'm just saying that the current spec doesn't see anything wrong with a =end inside of another pod block. 15:47
bacek Yes, I understand.
pmichaud so, I'll bring it up with the language folks and see if we can get a spec change.
bacek ok
sleep time 15:48
Good night.
15:49 khairul joined 16:05 kjeldahl joined, cognominal joined 16:20 theory joined 16:24 ambs joined 16:37 kid51 joined
dalek rrot: r47602 | jkeenan++ | trunk/config/gen/makefiles/root.in:
'examples_tests' lacked 'test_prep' prerequisite. Consequently, calling 'make examples_tests' directly after configuration failed miserably.
 16:48
rrot: r47603 | jkeenan++ | branches/cfunctionsdocs/config/gen/makefiles/root.in:
'examples_tests' lacked 'test_prep' prerequisite. Consequently, calling 'make examples_tests' directly after configuration failed miserably.
rrot: r47604 | fperrad++ | trunk/tools/dev/mk_inno.pl:
[windows installer] mingw/gcc 4.4.3 introduces a new DLL dependency. So, ship it.
16:51 szabgab joined
dalek TT #1679 created by jkeenan++: examples/compilers/japhc.c: Misleading documentation 16:54
TT #1679: trac.parrot.org/parrot/ticket/1679
rrot: r47605 | jkeenan++ | branches/cfunctionsdocs (2 files):
[codingstd] Provide documentation for C functions.
 17:06
kudo: bdbdfb2 | quester++ | src/core/MAIN.pm:
Pass False to MAIN for --/bool, not "True does False"
 17:15
17:28 japhb joined 17:29 mj41_ joined 17:37 patspam joined 17:57 aukjan1 joined 18:18 patspam1 joined 18:33 snarkyboojum joined 18:36 plobsing joined 18:45 radu_ joined 19:13 tcurtis joined 19:39 tcurtis joined 19:41 eternaleye joined 20:04 Psyche^ joined 20:43 mattp joined
dalek kudo: 33948ca | moritz++ | src/core/RangeIter.pm:
hide sub RangeIterCmp from userland
 21:01
kudo: 1345a33 | moritz++ | src/Perl6/Actions.pm:
variables should now where they come from
21:08 ambs joined 21:35 ruoso joined
dalek lambda: 8e5b2fc | fperrad++ | unl.pir:
refactor without dynops io
 21:37
zy-k: 0a8275c | fperrad++ | lazy.pir:
refactor without dynops io
 21:42
rrot: r47606 | darbelo++ | branches/gsoc_nfg/src/string (3 files):
Hack in some NFG-awareness into the unicode cclass functions.

It should work out most of the time.		 22:18
kudo: fca899b | moritz++ | docs/ChangeLog:
add MAIN to ChangeLog
 22:28
22:30 ruoso joined 22:32 jan joined 22:47 hercynium joined
dalek kudo: 7d80cd4 | masak++ | (4 files):
[Buf] switched from ByteView to ByteBuffer

figure out how to handle encodings besides UTF-8, though. One of the tests wants to decode using iso-8859-1.		 23:08
23:16 japhb joined
Coke msg gerd - can we close TT #904 ? 23:17
purl Message for gerd stored.
23:23 japhb joined 23:27 kid51 joined
mikehh All tests PASS (pre/post-config, make corevm/make coretest, smoke (#34348), fulltest) at r47606 - Ubuntu 10.04 i386 (g++ with --optimize) 23:37
t/op/exit.t - TODO passed: 6 in testf
23:37 snarkyboojum joined
mikehh note: the TODO pass in testf does not happen with amd64 23:37
kid51 mikehh: Do you have any guess about TT #1678? 23:39
(I realize it's on an OS you don't test.)
dalek rrot: r47607 | coke++ | tags/tt_1153:
Branch to work on �trac.parrot.org/parrot/ticket/1153
 23:40
rrot: r47608 | jkeenan++ | trunk/t/steps/auto/warnings-01.t:
Set test plan.
rrot: r47609 | coke++ | tags/tt_1153:
Branch, not tag.
rrot: r47610 | coke++ | branches/tt_1153:
BRANCH for working on �trac.parrot.org/parrot/ticket/1153
rrot: r47611 | coke++ | branches/tt_1153:
nevermind.
mikehh Coke: it works for me, I am not that familiar with darwin 23:45
kid51: sorry that was meant for you 23:46
kid51: it works for me, I am not that familiar with darwin
kid51 That's what I suspected. Does anyone know how to diagnose macro expansion problems?
Docs say: "The macro layer is completely implemented in the lexical analysis phase. The parser does not know anything about what happens in the lexical analysis phase." 23:47
plobsing That is to say, macro expansion occurs entirely within compilers/imcc/imcc.l 23:55
imcc.y never sees it
kid51 Do you have any clue as to why this would work on one OS and not another? 23:56
dalek rrot: r47612 | jkeenan++ | branches/gsoc_nfg/src/string (2 files):
[codingstd] No C++-style comments.
 23:57
plobsing give me a sec to check the relevant source
mikehh kid51: I ran it with ./parrot -t t/pir/macro.t - which gives loads of output - but you might pick up something
kid51 The -t option does ...? 23:58
mikehh trace
look for ok 1 then up to ok/not ok 2 will give that part 23:59
plobsing hmmm... I don't see anything in there that should be system dependant

Powered by Cro and the Raku® Programming Language.

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

Thank you!