Camelia, the Perl 6 bug

IRC log for #parrot, 2009-12-22

Parrot | source cross referenced

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

All times shown according to UTC.

Time Nick Message
00:00 japhb cotto_work, I want to do substr (inout STR, in INT, in INT, in STR), but I get substr(out STR, inout STR, in INT, in INT, in STR) by default ... which I could use anyway, except the inout param doesn't seem to be respected.
00:00 cotto_work nopaste?
00:00 purl rumour has it nopaste is at http://nopaste.snit.ch/ (ask TonyC for new channels)  or http://paste.scsys.co.uk/ or App::Nopaste or tools/dev/nopaste.pl or at http://www.extpaste.com/ or http://paste.scsys.co.uk (for #catalyst, #dbix-class, #moose  and others) or http://gist.github.com/ or paste or gtfo or tools/dev/nopaste.pl or https://trac.parrot.org/parrot/br​owser/trunk/tools/dev/nopaste.pl
00:02 nopaste "japhb" at 208.201.228.107 pasted "Need a way to have updated; this doesn't work" (1 line) at http://nopaste.snit.ch/19159
00:02 japhb ouch, that got MANGLED
00:03 nopaste "japhb" at 208.201.228.107 pasted "Need a way to have $copy updated; this doesn't work" (1 line) at http://nopaste.snit.ch/19160
00:04 japhb Sorry it's all on one line
00:04 cotto_work I was just going to snark about that.
00:05 japhb The bit where I try to say '$copy := pir::substr($copy, $m.from + $offset, $len, "xxx");' was just one attempt of many to figure out what NQP-rx wants
00:05 japhb cotto_work, when I'm thinking about a problem, I tend to iterate one-liners.  :-)
00:06 cotto_work sure
00:08 kid51 joined #parrot
00:12 kid51 Is there anyone on channel who can do a checkout on the tt473_remove_memcpy_aligned and then do 'make smoke'?
00:12 kid51 On Win32 Intel, in particular.
00:12 kid51 Any BSD also helpful.
00:12 cotto_work japhb, what should that code do?
00:12 kid51 Cf.:  http://trac.parrot.org/parrot/ticket/473
00:14 cotto_work japhb, alternately, does changing using $copy := pir::substr($m.from + $offset, $len, "xxx"); dtrt?
00:15 japhb cotto_work, I'm trying to turn this file: http://gitorious.org/parrot-plumage/parr​ot-plumage/blobs/master/src/lib/Glue.pir  into NQP, in particular the subst() function with relevant logic starting at line 120
00:15 dalek TT #81 closed by jkeenan++: PDD19: Documentation of    .HLL
00:15 dalek TT #413 closed by jkeenan++: [TODO] config/auto/alignptrs.pm:  inform user to use --64compat for parrot
00:15 dalek TT #637 closed by jkeenan++: smolder: server error on submission
00:16 japhb cotto_work, no, that change doesn't work; it does strangely enough the substr__ssii form, which really doesn't do the right thing
00:22 cotto_work I don't think I'll be able to be of much help atm.
00:24 japhb np
00:33 dalek parrot: r43188 | jkeenan++ | trunk/src/io (2 files):
00:33 dalek parrot: Applying
00:33 dalek parrot: �http://trac.parrot.org/parrot/attachme​nt/ticket/886/edited.tt_886_win.patch.
00:33 dalek parrot: JimmyZ++.
00:33 dalek parrot: review: http://trac.parrot.org/parrot/changeset/43188/
00:38 dalek winxed: r312 | julian.notfound++ | trunk/winxedst1.winxed:
00:38 dalek winxed: fixes in null handling and predef functions arguments in stage 1
00:38 dalek winxed: review: http://code.google.com/p/w​inxed/source/detail?r=312
00:49 * dukeleto found 2 bugs in the company source filter today. This is why I like hacking on Parrot. No Source Filters.
00:53 abqar joined #parrot
00:55 brrant joined #parrot
01:11 eternaleye joined #parrot
01:14 Zak joined #parrot
01:31 mikehh joined #parrot
01:46 Casan joined #parrot
01:48 Zak joined #parrot
01:58 kid51 joined #parrot
01:59 JimmyZ joined #parrot
02:06 kid51 msg Coke re earlier question:  You probably want perl Configure.pl --fatal or --fatal-step=some::step,some::other_step,...
02:06 purl Message for coke stored.
02:07 Zak joined #parrot
02:17 JimmyZ messages
02:19 cotto_work You have 461 new messages.
02:21 kid51 JimmyZ: ping
02:21 JimmyZ kid51: hello
02:21 purl what's up, JimmyZ.
02:22 kid51 JimmyZ:  Do you have a win32/x86 box on which you could smoke test a branch?
02:22 kid51 specifically:  tt473_remove_memcpy_aligned branch
02:22 JimmyZ kid51: yep
02:22 JimmyZ kid51: I will do it
02:22 kid51 Thanks.
02:26 dngor joined #parrot
02:44 JimmyZ kid51: http://smolder.plusthree.com/app/pu​blic_projects/report_details/31198
02:56 TonyC joined #parrot
02:56 kid51 JimmyZ:  Thanks.  That will enable me to close that ticket.
02:57 kid51 And, while you're at it, if you want to smoke trunk, it will enable us to close off the last of your tickets that I was handling.
03:01 nopaste joined #parrot
03:01 eternaleye joined #parrot
03:02 JimmyZ kid51: please hold on
03:03 plobsing can someone with gc knowledge look at http://nopaste.snit.ch/19162 ?
03:03 kid51 You mean, we actually *have* someone with gc knowledge?  ;-)
03:05 plobsing well you don't have to have gc knowledge to look, but it might not make much sense
03:05 Zak joined #parrot
03:05 kid51 src/io/win32.c:  Is that whitespace change only?
03:06 plobsing yeah, to get fulltest to pass
03:06 plobsing I seem to be the only one running that these days
03:06 kid51 I'll run make codetest once I do this commit.
03:08 kid51 Grrr.  something amiss with t/library/test_more.t
03:09 JimmyZ this patch seems that it fixed some wrong mark
03:14 JimmyZ kid51: trunk are passed
03:15 kid51 Thanks.
03:19 TonyC joined #parrot
03:27 nopaste joined #parrot
03:28 nopaste "kid51" at 70.85.31.226 pasted "c_header_guards.t codingstd failure: platform_interface.h" (24 lines) at http://nopaste.snit.ch/19163
03:28 nopaste "kid51" at 70.85.31.226 pasted "c_header_guards.t codingstd failure: platform_interface.h" (24 lines) at http://nopaste.snit.ch/19164
03:28 kid51 (2nd paste is dupe)
03:31 dalek parrot: r43189 | jkeenan++ | trunk/src/io/win32.c:
03:31 dalek parrot: No trailing whitespace. plobsing++ for noticing this.
03:31 dalek parrot: review: http://trac.parrot.org/parrot/changeset/43189/
03:47 dalek parrot: r43190 | plobsing++ | trunk/src/gc/alloc_resources.c:
03:47 dalek parrot: fix typo in merge_pools
03:47 dalek parrot: review: http://trac.parrot.org/parrot/changeset/43190/
03:47 dalek parrot: r43191 | cotto++ | trunk/include/parrot/call.h:
03:47 dalek parrot: [pcc] rerun headerizer to avoid warnings about an ignored return value
03:47 dalek parrot: review: http://trac.parrot.org/parrot/changeset/43191/
03:50 japhb joined #parrot
03:50 JimmyZ anyone to take a look at TT #1375?
03:51 nopaste joined #parrot
03:55 cotto That's a big patch there.
04:00 cotto I'm going to see if I can narrow down the bug that makes mandelbrot-color.pl segfault before my head explodes.  I'll try to look at your patch tonight after that.
04:01 * kid51 must sleep
04:01 purl $kid51->sleep(8 * 3600);
04:01 cotto Actually, I wonder if the recent fix by plobsing++ will have any effect.
04:01 cotto better try that first
04:04 TonyC joined #parrot
04:04 JimmyZ it looks well, but I don't know how to find effects.
04:05 kid51 joined #parrot
04:11 cotto JimmyZ, did make fulltest pass
04:12 cotto (or at least make test)
04:12 JimmyZ yes
04:12 JimmyZ make smoke
04:12 purl make smoke is Smoke Testing: Run the test suite and send smoke.html to http://smoke.parrotcode.org/
04:16 bubaflub joined #parrot
04:17 mikehh joined #parrot
04:17 mikehh All tests PASS (pre/post-config, make corevm/make coretest, smoke (#31202), fulltest) at r43191 - Ubuntu 9.10 i386 (gcc with --optimize)
04:20 bubaflub ping Coke
04:25 TonyC joined #parrot
04:27 eternaleye joined #parrot
04:30 bubaflub msg Coke thanks, i already got an invite from dukeleto
04:30 purl Message for coke stored.
04:31 mikehh joined #parrot
04:31 JimmyZ cotto: make fulltest failed, t/pmc/eval.t Failed tests:  8-9, 13
04:31 nopaste joined #parrot
04:33 cotto does the same thing happen without the patch?
04:34 JimmyZ testing.
04:34 purl it has been said that testing is the best thing ever ZOMG
04:35 JimmyZ but prove t/pmc/eval.t passed.
04:39 Coke msg kid51 no, I want the step to /always/ be fatal.
04:39 purl Message for kid51 stored.
04:47 JimmyZ cotto: t/pmc/eval.t Failed test:  12
04:48 JimmyZ I don't why if failed, but I saw it's tested at least 3 times when making fulltest.
04:48 plobsing JimmyZ: the multiple test passes are with different runcores
04:49 JimmyZ plobsing: how do I change the runcores?
04:49 JimmyZ manually
04:50 plobsing ./parrot -R <runcore>
04:50 plobsing ./parrot --help for a list of runcores
04:51 JimmyZ thanks
04:52 dalek parrot-plumage: 228d3f6 | japhb++ |  (16 files):
04:52 dalek parrot-plumage: Convert to native NQP-rx regex; move last bits of Glue functionality to Util; kill all references to Glue; various related fixes
04:52 dalek parrot-plumage: review: http://gitorious.org/parrot-plumage/parrot-plumag​e/commit/228d3f6915d398e533618fb0d3acd07bf13ac581
04:52 japhb msg fperrad Plumage rev 228d3f6 was a massive change; there's a decent chance I got something wrong for Win32, since I can't test there.  Most likely place is in quoting of regexen in Configure.nqp.  Let me know if you have any troubles, or send a patch!
04:52 purl Message for fperrad stored.
04:53 japhb Glue.pir is dead, long live Util.nqp!
04:59 eternaleye joined #parrot
05:01 mikehh joined #parrot
05:01 japhb Ah, time to take a break, I think
05:02 TonyC joined #parrot
05:09 bubaflub can i use make -j 4 fulltest?
05:10 cotto you have to set TEST_JOBS
05:11 nopaste joined #parrot
05:13 bubaflub thanks Coke
05:13 bubaflub whoops
05:13 bubaflub thanks cotto
05:15 Tene wtf?
05:15 Tene Makefile:29: *** missing separator.  Stop.
05:15 Tene when building...
05:15 Tene in src/dynpmc
05:15 Tene oh, looks lke a linewrapping issues.
05:15 Tene lines 28 and 29 are:
05:15 Tene LINKARGS        := $(LDFLAGS) $(LD_LOAD_FLAGS) $(LIBPARROT)  -licuuc -licudata -lpthread -lm   -lresolv -lnsl -ldl -lm -lcrypt -lutil -lpthread -lrt -lgmp -lreadline
05:15 Tene ack, my IRC client wrapped.
05:16 Tene LINKARGS        := $(LDFLAGS) $(LD_LOAD_FLAGS) $(LIBPARROT)  -licuuc -licudata -lpthread -lm
05:16 Tene -lresolv -lnsl -ldl -lm -lcrypt -lutil -lpthread -lrt -lgmp -lreadline
05:16 Tene If I join those two lines... it doesn't help. :)
05:19 Tene re-configure, and the same problem.
05:19 JimmyZ cotto: class.pmc.patch doesn't break make fulltest. with/without it the result is the same: t/pmc/eval.t Failed test: 12
05:20 cotto ok.
05:23 Tene if I do a completely clean checkout, I have the same problem.
05:27 Tene Okay, I have the same problem on a revision that I know worked previously, so the problem is something that broke on my system.
05:33 Tene Oh, it was that problem.
05:36 nopaste joined #parrot
05:41 dalek parrot: r43192 | bubaflub++ | trunk/t/dynpmc/subproxy.t:
05:41 dalek parrot: converted t/dynpmc/subproxy.t to PIR
05:41 dalek parrot: review: http://trac.parrot.org/parrot/changeset/43192/
05:41 dalek parrot: r43193 | bubaflub++ | trunk/t/compilers/tge/parser.t:
05:41 dalek parrot: convert t/compilers/tge/parser.t to PIR
05:41 purl I don't know how to convert t/compilers/tge/parser.t to PIR.
05:41 dalek parrot: review: http://trac.parrot.org/parrot/changeset/43193/
05:41 dalek parrot: r43194 | bubaflub++ | trunk/t/compilers/tge/basic.t:
05:41 dalek parrot: convert t/compilers/tge/basic.t to PIR
05:41 purl I don't know how to convert t/compilers/tge/basic.t to PIR.
05:41 dalek parrot: review: http://trac.parrot.org/parrot/changeset/43194/
05:41 dalek parrot: r43195 | bubaflub++ | trunk/t/compilers/pge/pge-hs.t:
05:41 dalek parrot: convert t/compilers/pge/pge-hs.t to PIR
05:41 purl I don't know how to convert t/compilers/pge/pge-hs.t to PIR.
05:41 dalek parrot: review: http://trac.parrot.org/parrot/changeset/43195/
05:41 dalek parrot: r43196 | bubaflub++ | trunk/t/compilers/pge/03-optable.t:
05:41 dalek parrot: convert t/compilers/pge/03-optable.t to PIR
05:41 purl I don't know how to convert t/compilers/pge/03-optable.t to PIR.
05:42 dalek parrot: review: http://trac.parrot.org/parrot/changeset/43196/
05:42 dalek parrot: r43197 | bubaflub++ | trunk/t/compilers/imcc/syn/labels.t:
05:42 dalek parrot: convert t/compilers/imcc/syn/labels.t to PIR
05:42 purl I don't know how to convert t/compilers/imcc/syn/labels.t to PIR.
05:42 dalek parrot: review: http://trac.parrot.org/parrot/changeset/43197/
05:44 cotto purl, go do something useful
05:44 purl cotto: huh?
05:44 bubaflub i guess i'll stick to past tense in my commit messages
05:44 cotto don't worry about it
05:45 cotto hating on the bot is something everyone can do
05:45 bubaflub msg Coke if you have any comments on r43192-r43197 i'd love to hear em
05:45 purl Message for coke stored.
05:45 cotto ;)
05:45 Coke did you mean coke or cotto?
05:46 bubaflub Coke: i thought i meant you.
05:47 cotto istr saying something about reviewing commits
05:48 bubaflub whoops.  yeah, cotto
05:48 bubaflub stupid tab complete on my IRC client...
05:51 bubaflub so ignore that message Coke.  sorry.
05:51 bubaflub cotto: same deal - if anything is sticks out on the last few commits lemme know.
05:51 nopaste joined #parrot
05:52 cotto I'll make sure to look at them.
05:53 bubaflub cool.
05:57 dalek parrot: r43198 | bubaflub++ | trunk/t/compilers/imcc/syn/objects.t:
05:58 dalek parrot: converted t/compilers/imcc/syn/objects.t to PIR
05:58 dalek parrot: review: http://trac.parrot.org/parrot/changeset/43198/
05:58 dalek parrot: r43199 | bubaflub++ | trunk/t/compilers/imcc/syn/op.t:
05:58 dalek parrot: converted t/compilers/imcc/syn/op.t
05:58 dalek parrot: review: http://trac.parrot.org/parrot/changeset/43199/
05:58 dalek parrot: r43200 | bubaflub++ | trunk/t/compilers/imcc/syn/scope.t:
05:58 dalek parrot: converted t/compilers/imcc/syn/scope.t to PIR
05:58 dalek parrot: review: http://trac.parrot.org/parrot/changeset/43200/
05:58 dalek parrot: r43201 | bubaflub++ | trunk/t/compilers/imcc/syn/veracity.t:
05:58 dalek parrot: converted t/compilers/imcc/syn/veracity.t to PIR
05:58 dalek parrot: review: http://trac.parrot.org/parrot/changeset/43201/
05:58 cotto how many more of those do you have?
05:59 TonyC joined #parrot
06:00 bubaflub cotto: that's pretty much it
06:00 eternaleye joined #parrot
06:00 bubaflub cotto: most are pretty stupid conversions
06:01 cotto good news for me
06:01 bubaflub some of those files only have one test, etc.
06:01 bubaflub just some low hanging fruit.
06:02 cotto like pears?
06:02 bubaflub sure.  if you like pears.
06:02 bubaflub i'm more of an avocado man.
06:05 bacek joined #parrot
06:06 bubaflub cotto: the only complex one was r43196
06:08 cotto I'm looking over all of them.
06:09 cotto I'm also wondering how long it'll be before we can safely kill TGE, but that's an separate matter.
06:11 bubaflub yeah, some of those test files were pretty ancient
06:14 cotto minor nit: in r43195 if the test fails, it'll run the ok(0,...) code and the ok(1,...) code.  It'll still successfully indicate that something's wrong, but it's a little odd.
06:20 cotto r43197 has 2 ok(1,...) lines in the goto_2 test, one of which looks out of place
06:26 bubaflub for r43195 i can just add a .return() after the ok(0, ...) so that doesn't happen
06:27 cotto yup
06:27 bubaflub and in r43197 i think the original had it jumping around a bit that hit both ok()s
06:27 bubaflub let me check
06:37 cotto I didn't know we had something called ".begin_return".
06:40 bubaflub yeah, i found that out too whilst reading that test
06:41 bubaflub i have a feeling the syntax has changed since then
06:42 cotto Parrot is like an onion.
06:43 bubaflub if only parrot were like The Onion
06:44 Coke I think .begin_return is still valid.
06:44 Coke it's the long form.
06:44 cotto Coke, yes.
06:46 dalek parrot: r43202 | bubaflub++ | trunk/t/compilers/pge/pge-hs.t:
06:46 dalek parrot: fix t/compilers/pge/pge-hs.t to not double test on failure
06:46 dalek parrot: review: http://trac.parrot.org/parrot/changeset/43202/
06:46 dalek parrot: r43203 | coke++ | trunk/config/gen/makefiles/root.in:
06:46 dalek parrot: Improve dependency status; make tools/dev/checkdepend.pl pass.
06:46 cotto not sure why anyone would want to do that, but it still works and we have a test to make sure it keeps working
06:46 dalek parrot: review: http://trac.parrot.org/parrot/changeset/43203/
06:46 dalek parrot: r43204 | coke++ | trunk/tools/dev/checkdepend.pl:
06:46 dalek parrot: Properly fail a test if no deps were defined in the Makefile for this C file.
06:46 dalek parrot: review: http://trac.parrot.org/parrot/changeset/43204/
06:47 cotto several, in fact
06:52 Coke anyone interested in switching to pirc would be very helpful in 1) adding all the deps to the various PIRC files in the makefile, and 2) fixing up checkdepend.pl to actually notice you'd done it right.
06:52 Coke (And then enabling pirc to always be built)
06:55 cotto bubaflub, I'm seeing a failure on make test on t/compilers/pge/pge_examples.t, but only when running make test.
06:55 bubaflub cotto: yeah, that's strange.
06:55 bubaflub let me try make test
06:56 bubaflub cotto: i don't think i've touched that test.  let me check the logs.
06:57 bacek joined #parrot
06:59 cotto I wouldn't be surprised.
06:59 bubaflub cotto: nah, i didn't.  just wanted to make sure i didn't break the build.  i'm running make test now
07:00 cotto I love these mysterious failures.
07:01 bubaflub haha.  it passed on my machine
07:01 bubaflub nqp tests don't pass, though
07:04 cotto maybe it's related to the context merge
07:04 cotto at least it's kinda reproducible
07:05 cotto btw, did you look into the issue before unconditionally unTODOing the FLOAT_IS_ZERO test in veracity.t?
07:07 bubaflub the todo was already commented out by the time i got ther
07:07 bubaflub e
07:08 cotto oh.  I missed that.
07:10 cotto I just need to look over the optable test, but the others look good.
07:11 bubaflub cool
07:11 bubaflub the optable test took me some time to convert
07:11 nopaste joined #parrot
07:14 carlin joined #parrot
07:15 cotto I imagine that they did.
07:16 clubs joined #parrot
07:29 clubs On OpenSolaris Parrot doesn't build with readline support, because readline needs tgetnum et al. from Solaris' curses lib. The configure tries to fall back to ncurses but on Solaris it is just curses. %s/ncurses/curses in config/auto/readline.pm makes it work ... would it be worth trying to get a more permanent fix for that?
07:39 cotto clubs, yes.  If you can't figure it out, please file a tt.
07:40 dalek partcl-nqp: 10d7f3c | duff++ | src/Glob/ (2 files):
07:40 dalek partcl-nqp: Remove regex-like alternation from glob
07:40 dalek partcl-nqp: review: http://github.com/partcl/partcl-nqp/commit​/10d7f3cf21432300f87f861c51d9afba28e5505f
07:45 iblechbot joined #parrot
07:49 cotto night
07:50 bubaflub night
07:51 eternaleye joined #parrot
07:51 nopaste joined #parrot
07:58 bacek joined #parrot
08:11 Coke PerlJam++
08:12 xenoterracide joined #parrot
08:15 JimmyZ versatile man
08:26 nopaste joined #parrot
08:54 JimmyZ_ joined #parrot
09:11 fperrad joined #parrot
09:12 eternaleye joined #parrot
09:14 eternaleye joined #parrot
09:21 nopaste joined #parrot
09:22 TonyC joined #parrot
09:27 bacek joined #parrot
09:39 eternaleye joined #parrot
09:53 Zak joined #parrot
09:59 TiMBuS joined #parrot
09:59 fperrad_ joined #parrot
10:16 eternaleye joined #parrot
10:35 bacek joined #parrot
10:41 fperrad msg allison the file Lib/test/bootstrap/class-type.py seems missing in Pynie repository
10:41 purl Message for allison stored.
11:29 mikehh joined #parrot
11:37 JimmyZ joined #parrot
11:39 bluescreen joined #parrot
11:48 eternaleye joined #parrot
12:09 bluescreen joined #parrot
12:27 bacek joined #parrot
12:36 integral joined #parrot
13:57 patspam joined #parrot
14:15 whiteknight joined #parrot
14:19 plobsing joined #parrot
14:36 dalek parrot: r43205 | coke++ | trunk/tools/dev/checkdepend.pl:
14:36 dalek parrot: Sort test output.
14:36 dalek parrot: review: http://trac.parrot.org/parrot/changeset/43205/
14:38 bubaflub joined #parrot
14:38 mikehh fat/compilers/imcc/syn/labels.t  FAILs in make corevm/make coretest - passes in make test (like doesn't work with corevm)
14:39 mikehh dunno how I got that s/fat/t/
14:40 bubaflub mikehh: that's my fault.  i can fix that real quick.
14:40 bubaflub instead of using a throws_like i'll change it to a dies_ok()
14:40 mikehh bubaflub - was going to msg you - I'll leave it to you
14:42 mikehh bubaflub - "load_bytecode" couldn't find file 'PGE.pbc'
14:42 bubaflub ok, i just pushed the fix to the trunk
14:42 bubaflub should be good to go
14:43 bubaflub mikehh: i'm going to run a make corevm && make coretest real quick just to be sure
14:43 Coke just to note for the record that that test is now much less specific.
14:43 Coke I am not sure in this case that that matters.
14:44 bubaflub Coke:  yep.  now if that codes dies *for any reason* the test will pass... not necessarily just the one we are looking for.
14:50 Coke dalek?
14:50 purl dalek is #parrot's spammy little rss bot or (see: dalek plugins)
14:51 dalek parrot: r43206 | coke++ | trunk/compilers/pirc (2 files):
14:51 dalek parrot: don't #include 2x, or try to include a .h that doesn't exist.
14:51 dalek parrot: review: http://trac.parrot.org/parrot/changeset/43206/
14:51 dalek parrot: r43207 | coke++ | trunk/config/gen/makefiles/root.in:
14:51 dalek parrot: add a small # of missing dependencies.
14:52 dalek parrot: review: http://trac.parrot.org/parrot/changeset/43207/
14:52 dalek parrot: r43208 | bubaflub++ | trunk/t/compilers/imcc/syn/labels.t:
14:52 dalek parrot: can't use throws_like() for corevm tests; using dies_ok() instead
14:52 dalek parrot: review: http://trac.parrot.org/parrot/changeset/43208/
15:06 Psyche^ joined #parrot
15:11 PacoLinux joined #parrot
15:11 JimmyZ joined #parrot
15:27 mikehh All tests PASS (pre/post-config, make corevm/make coretest, smoke (#31236), fulltest) at r43208 - Ubuntu 9.10 amd64 (g++ with --optimize)
15:30 davidfetter joined #parrot
15:44 iblechbot joined #parrot
16:04 pmichaud joined #parrot
16:09 Coke pmichaud: PPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPmichaud!
16:09 pmichaud good morning.
16:14 whiteknight good morning pmichaud
16:29 dalek parrot: r43209 | coke++ | trunk/tools/dev/checkdepend.pl:
16:29 dalek parrot: ignore ops files for now.
16:29 dalek parrot: review: http://trac.parrot.org/parrot/changeset/43209/
16:29 dalek parrot: r43210 | coke++ | trunk/tools/dev/checkdepend.pl:
16:29 dalek parrot: also skip dynoplibs, and skip earlier.
16:29 dalek parrot: review: http://trac.parrot.org/parrot/changeset/43210/
16:29 dalek parrot: r43211 | mikehh++ | trunk/src/ops/pmc.ops:
16:29 dalek parrot: restore partial reversion as it passes now (TT #1368 resolved after context_unify3_simple branch merge)
16:29 dalek parrot: review: http://trac.parrot.org/parrot/changeset/43211/
16:30 jaffa8 joined #parrot
16:31 jaffa8 hi
16:31 purl hola, jaffa8.
16:31 jaffa8 how do I specify the dearch path for parrot?
16:31 jaffa8 it cannot find some pbc
16:31 whiteknight that's a good question
16:34 Coke ./parrot --help
16:34 Coke look for "path"
16:35 jaffa8 pbc
16:35 jaffa8 is it related to include path or library path?
16:35 Coke include is probably ".include". library is probably "load_bytecode,load_library", etc.
16:35 dalek TT #1368 closed by mikehh++: some failures with test 31 of t/op/exceptions.t
16:36 Coke I would try -L.
16:41 jaffa8 How would you generate pbc file from a pir file?
16:42 ash_ joined #parrot
16:44 Coke ./parrot -o foo.pbc foo.pir
16:44 Coke (also see http://docs.parrot.org/parrot/la​test/html/docs/running.pod.html for lots more options)
16:45 dalek parrot: r43212 | coke++ | trunk/tools/dev/checkdepend.pl:
16:45 dalek parrot: dynpmc are not gen'd from main makefile yet.
16:45 dalek parrot: review: http://trac.parrot.org/parrot/changeset/43212/
16:45 dalek parrot: r43213 | coke++ | trunk/config/gen/makefiles/root.in:
16:45 dalek parrot: Add/fix build deps.
16:45 dalek parrot: tools/dev/checkdepend.pl currently passing...
16:45 dalek parrot: review: http://trac.parrot.org/parrot/changeset/43213/
16:48 Coke (huh. never used -v before.)
16:50 jaffa8 Coke, what do you know pbc to exe?
16:50 dalek winxed: r313 | julian.notfound++ | trunk/winxedst0.cpp:
16:50 dalek winxed: cosmetic changes in stage 0 pir output
16:50 dalek winxed: review: http://code.google.com/p/w​inxed/source/detail?r=313
16:50 jaffa8 Is it reliable?
16:54 dalek winxed: r314 | julian.notfound++ | trunk/Makefile:
16:54 dalek winxed: add stage2 make target for testing
16:54 dalek winxed: review: http://code.google.com/p/w​inxed/source/detail?r=314
16:54 dalek winxed: r315 | julian.notfound++ | trunk/winxed.winxed:
16:54 dalek winxed: add stage 2 option to driver
16:54 dalek winxed: review: http://code.google.com/p/w​inxed/source/detail?r=315
16:57 Coke jaffa8: the closest is 'pbc_to_exe'
16:58 Coke (you end up with an exe that loads the parrot library and has a copy of the pbc.)
16:58 Coke (it doesn't actually compile it to machine asm.)
16:59 jaffa8 no
16:59 jaffa8 ?
16:59 jaffa8 it does not compile to?
17:00 jaffa8 shame.
17:01 dalek parrot: r43214 | plobsing++ | trunk/src/gc (2 files):
17:01 dalek parrot: make guaranteed_reclaimable and possibly_reclaimable disjoint
17:01 dalek parrot: review: http://trac.parrot.org/parrot/changeset/43214/
17:04 jaffa8 Coke, are you sure?
17:09 mikehh All tests PASS (pre/post-config, make corevm/make coretest, smoke (#31245), fulltest) at r43213 - Ubuntu 9.10 amd64 (gcc with --optimize)
17:11 jaffa8 Hello Coke, are you here?
17:14 jaffa8 left #parrot
17:18 dalek winxed: r316 | julian.notfound++ | trunk/winxedst1.winxed:
17:18 dalek winxed: refactor and fix return and yield in stage 1
17:18 dalek winxed: review: http://code.google.com/p/w​inxed/source/detail?r=316
17:23 dalek winxed: r317 | julian.notfound++ | trunk/winxedst1.winxed:
17:23 dalek winxed: fix string initialization from pmc in stage 1
17:23 dalek winxed: review: http://code.google.com/p/w​inxed/source/detail?r=317
17:24 theory joined #parrot
17:33 dalek winxed: r318 | julian.notfound++ | trunk/winxedst1.winxed:
17:33 dalek winxed: fix for annotations in yield and return in stage 1 forgotten in r316
17:33 dalek winxed: review: http://code.google.com/p/w​inxed/source/detail?r=318
17:47 dalek winxed: r319 | julian.notfound++ | trunk/winxedst1.winxed:
17:47 dalek winxed: fix += in stage 1
17:47 dalek winxed: review: http://code.google.com/p/w​inxed/source/detail?r=319
17:54 mikehh All tests PASS (pre/post-config, make corevm/make coretest, smoke (#31246), fulltest) at r43214 - Ubuntu 9.10 amd64 (g++)
17:56 dalek winxed: r320 | julian.notfound++ | trunk/winxedst1.winxed:
17:56 dalek winxed: minor fixes in stage 1
17:56 dalek winxed: review: http://code.google.com/p/w​inxed/source/detail?r=320
18:06 dalek winxed: r321 | paco.linux++ | trunk/Makefile:
18:06 dalek winxed: We have to call the driver from the current dir.
18:06 dalek winxed: review: http://code.google.com/p/w​inxed/source/detail?r=321
18:06 dalek winxed: r322 | julian.notfound++ | trunk/winxedst1.winxed:
18:06 dalek winxed: fix string with pmc comparaisons in stage 1
18:06 dalek winxed: review: http://code.google.com/p/w​inxed/source/detail?r=322
18:16 dalek winxed: r323 | julian.notfound++ | trunk/winxedst1.winxed:
18:16 dalek winxed: fix a variable scope in stage 1
18:16 dalek winxed: review: http://code.google.com/p/w​inxed/source/detail?r=323
18:17 payload joined #parrot
18:25 dalek winxed: r324 | julian.notfound++ | trunk/winxedst1.winxed:
18:25 dalek winxed: cosmetic chnages in pir output in stage 1
18:25 dalek winxed: review: http://code.google.com/p/w​inxed/source/detail?r=324
18:26 cotto_work ps in 3
18:27 NotFound Uh, I've completely forgot that was tuesday.
18:28 cotto_work I thought that was why you were committing so much.
18:30 NotFound Sill 29 in my clock
18:35 dalek winxed: r325 | julian.notfound++ | trunk/winxedst1.winxed:
18:35 dalek winxed: fix instanceof in stage 1
18:35 dalek winxed: review: http://code.google.com/p/w​inxed/source/detail?r=325
18:36 Coke msg jaffa8 (pbc 2 exe reliable) It's used by rakudo and partcl all the time. haven't had any problems.
18:36 purl Message for jaffa8 stored.
18:44 dalek winxed: r326 | julian.notfound++ | trunk/winxedst1.winxed:
18:44 dalek winxed: optimize step for MemberExpr in stage 1
18:44 dalek winxed: review: http://code.google.com/p/w​inxed/source/detail?r=326
18:55 dalek parrot: r43215 | coke++ | trunk/config/gen/makefiles/root.in:
18:55 dalek parrot: Build ./pirc by default, but don't use it.
18:55 dalek parrot: (all deps for it should be resolved now; this will insure it at least compiles everywhere) - smoke reports welcome.
18:55 dalek parrot: review: http://trac.parrot.org/parrot/changeset/43215/
18:59 dalek winxed: r327 | julian.notfound++ | trunk/winxedst1.winxed:
18:59 dalek winxed: fix hash initialization in stage 1
18:59 dalek winxed: review: http://code.google.com/p/w​inxed/source/detail?r=327
19:00 Coke I meant that it will /eventually/ insure.
19:08 dalek winxed: r328 | julian.notfound++ | trunk/winxedst1.winxed:
19:08 dalek winxed: predef getinterp in stage 1
19:08 dalek winxed: review: http://code.google.com/p/w​inxed/source/detail?r=328
19:13 dalek winxed: r329 | julian.notfound++ | trunk/winxedst1.winxed:
19:13 dalek winxed: fix return and yield with literal null in stage 1
19:13 dalek winxed: review: http://code.google.com/p/w​inxed/source/detail?r=329
19:16 ttbot Parrot trunk/ r43215 cygwin-thread-multi-64int make error http://tt.ro.vutbr.cz/file/cmdout/176991.txt ( http://tt.ro.vutbr.cz//buil​dstatus/pr-Parrot/rp-trunk/ )
19:57 iblechbot joined #parrot
20:06 dalek winxed: r330 | julian.notfound++ | trunk/ (4 files):
20:06 dalek winxed: start restructuring tests
20:06 dalek winxed: review: http://code.google.com/p/w​inxed/source/detail?r=330
20:11 bluescreen joined #parrot
20:22 ash_ joined #parrot
20:25 dalek winxed: r331 | julian.notfound++ | trunk/t/basic/03functions.t:
20:25 dalek winxed: more basic tests
20:25 dalek winxed: review: http://code.google.com/p/w​inxed/source/detail?r=331
20:39 payload joined #parrot
20:51 bacek joined #parrot
20:52 plobsing joined #parrot
21:03 NotFound How the fuck is one supposed to compile the .y sources of pirc? By hand?
21:06 cotto_work NotFound, it looks like the generated files are properly checked in.
21:07 cotto_work or were you asking about regeneration
21:07 NotFound cotto_work: I'm talking about being able to fix it
21:10 cotto_work Good point.  --maintainer doesn't dtrt for pirc.
21:10 NotFound Some people wants to build it from the main Makefile, but that breaks the c++ build. If we can't fix it it must not be built by default.
21:11 cotto_work sure
21:12 cotto_work seen kj
21:12 purl kj was last seen on #parrot 37 days, 7 hours, 36 minutes and 35 seconds ago, saying: for the update :-)  [Nov 15 13:34:01 2009]
21:12 cotto_work seen kjs
21:12 purl kjs was last seen on #perl 1 years, 182 days, 5 hours, 37 minutes and 8 seconds ago, saying: yo  [Jun 23 15:33:30 2008]
21:13 cotto_work email is probably a better bet if you can't (or don't care to) figure it out
21:14 bacek joined #parrot
21:15 NotFound Maybe I can, but I don't think that something in that poor stage must be built by default.
21:15 cotto_work If it's anything like imcc, the yacc and flex commands won't be very complex at all.
21:16 cotto_work It's a chicken and egg problem.  It'll stay in that state until we build it by default and start trying to use it.
21:17 NotFound cotto_work: the solution is to write a proper Makefile, or at least acknowledge the problem. Isn't even in the TODO.
21:20 joeri joined #parrot
21:22 cotto_work He probably didn't think to add it.  For a while pirc didn't even have a proper makefile.
21:24 Coke pirc isn't getting its own makefile. =-)
21:24 Coke put it in the big one.
21:24 Coke but feel free to remove it for now if it's killing your build.
21:24 cotto_work NotFound, but you're right that it's needed and it should be added.
21:24 NotFound The big gun? }:)
21:25 Coke I also plan on on fixing the --maintainer option. :|
21:25 Coke (not for pirc, but so that all it does is swap out 'touch' for 'real program')
21:26 Coke I am in the middle of killing GENERAL_H_FILES
21:30 Coke adding a ticket for --maintainer update...
21:34 szabgabx joined #parrot
21:35 NotFound I definitely can't fix that thing. Manipulating enums via macros as if they were plain int is a big NO in C++
21:36 NotFound I don't see the point of replacing imcc with a nigthmare of the same size, BTW
21:37 Coke me either, but I cannot comment on the C. I have been assured that it is an improvement.
21:37 NotFound Coke: imcc at least compiles
21:37 Coke if you could open a ticket about pirc is broken in C++, that would be helpful.
21:43 Casan joined #parrot
21:46 dalek TT #1387 created by coke++: need --maintainer option for pirc
21:48 NotFound Coke: done
21:48 cotto_work plobsing, ping
21:49 plobsing cotto_work: pong
21:49 cotto_work plobsing, how does your freeze cleanup branch effect http://trac.parrot.org/parrot/ticket/362 ?
21:49 cotto_work ("optimize freeze_size")
21:52 Coke NotFound++
21:52 plobsing cotto_work: not immediately.
21:52 cotto_work ok, so it's fine to leave the tt as-is?
21:52 plobsing I've killed VISIT_FREEZE_SIZE (unused, and I have something better)
21:53 plobsing But when we move to PMC visitors, we can make a visitor that can estimate the size.
21:53 dalek parrot: r43216 | NotFound++ | trunk/config/gen/makefiles/root.in:
21:53 dalek parrot: don't build pirc in the default target, it breaks c++ builds
21:53 dalek parrot: review: http://trac.parrot.org/parrot/changeset/43216/
21:55 cotto_work So it sounds like the ticket can be closed.
21:55 plobsing yes and no. It might still be a worthwhile optimization, but we'll do it differently
21:55 plobsing I'll leave a comment on the TT
21:56 Casan I've been out of the loop for the past 4 months, but now I'm back and I wonder if any progress was made on on hll and libraries for mysql,pgsql and such?
22:01 NotFound Casan: the Mysql that used to be in examples is now an external module.
22:02 dalek TT #1388 created by NotFound++: pirc does not build with C++
22:02 Casan NotFound: Thanks, is it operational?
22:03 NotFound See http://trac.parrot.org/parrot/wiki/Modules
22:04 NotFound Casan: it has just basic functionality, but works.
22:04 ash_ joined #parrot
22:06 Casan Thanks, looking forward to get back in..
22:12 NotFound Casan: feel free to open issues in the project.
22:13 kid51 joined #parrot
22:16 TiMBuS joined #parrot
22:35 eternaleye joined #parrot
22:50 dduncan joined #parrot
22:52 dduncan_ joined #parrot
22:53 dduncan_ test post
22:54 Tene hi dduncan
22:54 dduncan hm, so irc.parrot.org and irc.perl.org are the same server
22:54 Tene Yes.
22:54 * dduncan will release duplicate connection ...
22:55 dduncan and now I know
23:01 Coke MARKETING
23:01 purl it has been said that MARKETING is from the pit of hell. or C-R-A-P or (see: advertising) or NBI (no brains involved) or All Style, No Substance (thanks memepool) or product line management. or All Show, No Go or <reads marketing literature> a fucking load of bullshit or "window dressing" or like the retarded trust-fund-sibling of psychology but where psychology has been cut out of the parents will so has all the money
23:01 cotto_work purl may be biased
23:01 purl cotto_work: i'm not following you...
23:03 Casan marketing is abou customers, satisfy needs and wants and you'll have resources to do the tech (although not necessarily the tech you want)
23:03 TiMBuS joined #parrot
23:04 Zak joined #parrot
23:52 dalek parrot-plumage: dfc7b7b | japhb++ | src/lib/Util. (2 files):
23:52 dalek parrot-plumage: [LIB] Util: Fix do_run() Windows bug
23:52 dalek parrot-plumage: review: http://gitorious.org/parrot-plumage/parrot-plumag​e/commit/dfc7b7ba01ba796a9e4d41fb50f0400dc9241a28
23:58 payload joined #parrot

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

Parrot | source cross referenced