Camelia, the Perl 6 bug

IRC log for #parrot, 2009-06-18

Parrot | source cross referenced

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

All times shown according to UTC.

Time Nick Message
00:07 Theory joined #parrot
00:17 chromatic joined #parrot
00:19 Whiteknight joined #parrot
00:25 Whiteknight howdy everybody!
00:27 kid51 Whiteknight:  Isn't it just about your bedtime?  ;-)
00:28 kid51 and BTW, Whiteknight++ on early release
00:28 Whiteknight thanks!
00:28 Whiteknight yeah, it is getting close to bed time, and I'm sick so that time is earlier
00:34 Infinoid get well soon
00:34 Infinoid (before YAPC even!)
00:35 cotto chromatic, whatever happened to the L1 writeup you were going to write?
00:35 chromatic Life happened.
00:35 chromatic I might be able to write one on a plane, but I had a book to edit, two talks to write, three talks to give, and family visiting before YAPC.
00:36 chromatic Oh yeah, two books to edit.
00:36 cotto sounds a bit busy
00:36 chromatic I managed to get in a few minutes of Mario Kart the other day.
00:42 Coke I just got Bioshock. Bad Coke.
00:53 dukeleto joined #parrot
00:54 whoppix joined #parrot
00:58 Whiteknight chromatic: I've been busy with my own writeups, so if it's easier for you to just write a snappy criticism of that, that's fine too
00:59 Whiteknight assuming that my ideas aren't so far out of the ballpark that you wouldn't need to write more then a whole writeup to bash it
01:10 cotto Cool.  We made Hacker News again.
01:10 sekimura joined #parrot
01:11 cotto http://news.ycombinator.com/item?id=662479
01:15 cotto Oh.  Allison wrote that.  allison++
01:22 amuck joined #parrot
01:23 eternaleye joined #parrot
01:28 dukeleto joined #parrot
01:34 dalek parrot: r39636 | jkeenan++ | branches/cindent/idealpmc.txt:
01:34 dalek parrot: Delete unneeded comment.
01:34 dalek parrot: review: https://trac.parrot.org/parrot/changeset/39636/
01:43 Whiteknight I've never even heard of hacker news before tonight
01:44 dalek parrot: r39637 | jkeenan++ | branches/cindent/src/io/unix.c:
01:44 dalek parrot: Indent label to match following block (temporary change pending further analysis of c_indent coding standard).
01:44 dalek parrot: review: https://trac.parrot.org/parrot/changeset/39637/
01:47 dalek parrot: r39638 | jkeenan++ | branches/cindent/t/codingstd/c_indent.t:
01:47 dalek parrot: Partial fix of coding standard problem.
01:47 dalek parrot: review: https://trac.parrot.org/parrot/changeset/39638/
01:50 Theory joined #parrot
01:51 dalek parrot: r39639 | jkeenan++ | branches/cindent/t/codingstd/c_indent.t:
01:51 dalek parrot: Revert one block to its pre-debugging state.
01:51 dalek parrot: review: https://trac.parrot.org/parrot/changeset/39639/
01:54 dalek parrot: r39640 | jkeenan++ | branches/cindent/t/codingstd/c_indent.t:
01:54 dalek parrot: Remove module import no longer needed.
01:54 dalek parrot: review: https://trac.parrot.org/parrot/changeset/39640/
01:54 dalek parrot: r39641 | jkeenan++ | branches/cindent (3 files):
01:54 dalek parrot: Remove two dummy files used in testing and update MANIFEST.
01:54 dalek parrot: review: https://trac.parrot.org/parrot/changeset/39641/
01:55 pmichaud (Mario Kart)++
01:58 GeJ Which version?
01:58 purl Which version is that?
02:00 pmichaud I have Mario Kart Wii
02:02 GeJ So do I, but for some reason I couldn't explain I'm having more fun with the DS.
02:02 GeJ Maybe because it's closer to the original Mario Kart game, not sure.
02:03 chromatic joined #parrot
02:03 TonyC joined #parrot
02:04 dalek parrot: r39642 | jkeenan++ | trunk (3 files):
02:04 dalek parrot: Merge cindent branch into trunk.  This implements a partial solution to problem described in https://trac.parrot.org/parrot/ticket/764.
02:05 dalek parrot: review: https://trac.parrot.org/parrot/changeset/39642/
02:05 dalek parrot: r39643 | jkeenan++ | branches/cindent:
02:05 dalek parrot: Branch merged into trunk and no longer needed at head.
02:05 dalek parrot: review: https://trac.parrot.org/parrot/changeset/39643/
02:10 chromatic I have the N64, GC, DS, and Wii versions.
02:13 GeJ I started with the SNES and then jumped to the GC, DS and Wii.
02:15 chromatic You missed out.
02:15 chromatic Four player versus was the best on the N64.
02:16 kid51 I can't recall whether I saw this link posted here already or not:  http://www.linux-mag.com/id/7373
02:26 ZeroForce joined #parrot
02:26 amuck joined #parrot
02:27 dukeleto joined #parrot
03:02 buildbot joined #parrot
03:11 dukeleto joined #parrot
03:20 donaldh joined #parrot
04:17 Andy joined #parrot
04:27 brbrooks joined #parrot
04:33 cotto How do I find the length of a string in nqp?
04:33 brbrooks joined #parrot
04:33 japhb joined #parrot
04:33 cotto +$str doesn't dtrt
04:46 nopaste "cotto" at 74.61.2.46 pasted "someone please tell me this isn't the easiest way" (10 lines) at http://nopaste.snit.ch/16956
04:47 uniejo joined #parrot
05:10 uniejo joined #parrot
05:40 chromatic joined #parrot
06:20 uniejo joined #parrot
06:31 canhope joined #parrot
06:35 bacek joined #parrot
06:38 bacek cotto: src/builtins.pir :)
06:38 bacek compilers/pmcc/src/builtins.pir
06:41 moritz cotto: in Perl 6 it's .chars or .bytes, but that doesn't seem to work in NQP
06:44 clinton joined #parrot
06:46 cotto bacek, that looks like a sane place for it
06:47 bacek cotto: I've added specially for this purpose :)
06:47 bacek s/ / it /
06:53 iblechbot joined #parrot
07:10 dngor joined #parrot
07:20 donaldh joined #parrot
07:43 mj41 msg allison The article ( http://www.linux-mag.com/id/7373 ) for Linux Magazine is great. But there isn't link to poarrot.org.
07:43 purl Message for allison stored.
07:53 dalek parrot: r39644 | cotto++ | trunk/lib/Parrot/Pmc2c/PMCEmitter.pm:
07:53 dalek parrot: [pmc2c] remove a mysterious space in front of dynpmcs' provides string, misc cleanup
07:53 dalek parrot: review: https://trac.parrot.org/parrot/changeset/39644/
07:57 dalek parrot: r39645 | cotto++ | branches/pmc_pct/compilers/pmcc/src (4 files):
07:57 dalek parrot: [pmcc] generate the first part of the first pass of class_init
07:57 dalek parrot: review: https://trac.parrot.org/parrot/changeset/39645/
08:00 dalek parrot: r39646 | cotto++ | branches/pmc_pct/compilers/pmcc/src (2 files):
08:00 dalek parrot: [pmcc] unfreeze -> thaw
08:00 dalek parrot: review: https://trac.parrot.org/parrot/changeset/39646/
08:00 masak joined #parrot
08:05 dalek parrot: r39647 | cotto++ | branches/pmc_pct/compilers/pmcc/src (2 files):
08:05 dalek parrot: [pmcc] make a freeze sub, remove unnecessary commenting from inline POD
08:05 dalek parrot: review: https://trac.parrot.org/parrot/changeset/39647/
08:09 workbench joined #parrot
08:16 dukeleto joined #parrot
08:35 bacek cotto++ # Hooray! pmc_pct ftw!
08:48 cotto a good part of the work involves getting rid of (or not using) state from a parent's PMC, since pmcc won't have most of that information.
08:48 cotto plus the normal wrestling with nqp and occasional facepalms after an hour of debugging
08:49 szbalint :)
08:51 cotto apparently $<$identifier> isn't quite the same as $<identifier>
08:54 moritz right
08:54 moritz it would be $/{$identifier}
08:54 moritz $/<...> always quotes the ...
08:56 viklund_ joined #parrot
09:23 sparkymat joined #parrot
09:40 dalek parrot: r39648 | cotto++ | trunk/lib/Parrot/Pmc2c/PMCEmitter.pm:
09:40 dalek parrot: [pmc2c] make mro and isa_hash generation happen at runtime
09:40 dalek parrot: pmcc will be taking this approach and pmc2c makes a good guinea pig (all tests pass, of course)
09:40 dalek parrot: review: https://trac.parrot.org/parrot/changeset/39648/
09:40 bacek_ joined #parrot
10:04 dalek parrot: r39649 | cotto++ | branches/pmc_pct/compilers​/pmcc/src/emitter/pmc.pm:
10:04 dalek parrot: [pmcc] add stub get_mro_func and get_isa_func methods
10:04 dalek parrot: review: https://trac.parrot.org/parrot/changeset/39649/
10:34 bacek_ joined #parrot
10:45 Zak joined #parrot
10:52 muixirt joined #parrot
10:53 NotFound joined #parrot
10:53 mj41_ joined #parrot
10:53 ttbot joined #parrot
10:53 Hunger joined #parrot
10:54 jq joined #parrot
10:57 baest joined #parrot
10:58 GeJ joined #parrot
11:00 slavorg joined #parrot
11:04 muixirt does the Configure.pl generated by mk_language_shell.pl assume an installed parrot?
11:04 Infinoid bacek_: if I have time this afternoon, I can send the openbsd/hppa guy some C-only floating point tests... if they fail too, we might as well just TODO the tests
11:05 bacek_ Infinoid: ok, thanks.
11:07 Maddingu1 joined #parrot
11:07 mj41_ joined #parrot
11:08 bacek_ Infinoid: btw, floor(-5.0, 3.0) is good starting point...
11:09 muixirt or is it preferable to use the --gen-parrot option to roll your own parrot based language?
11:12 burmas joined #parrot
11:21 GeJ_ joined #parrot
11:21 donaldh joined #parrot
11:21 baest_ joined #parrot
11:21 mj41_ joined #parrot
11:42 * muixirt tries with an installed parrot-1.3.0
11:51 muixirt doesn't work :-(
11:55 muixirt Can't open perl script "/usr/local/lib/parrot/1.3.0​/tools/dev/gen_makefile.pl"
11:56 nopaste "mikehh" at 90.209.155.162 pasted "PATCH to fix TODO PASSes in t/examples/pod.t at r39649 (supercedes http://nopaste.snit.ch/16946)" (87 lines) at http://nopaste.snit.ch/16958
11:57 * muixirt forgot make install-dev
12:01 nopaste "mikehh" at 90.209.155.162 pasted "Summary results for: make -k fulltest - Kubuntu 9.04 Amd64 at r39649" (107 lines) at http://nopaste.snit.ch/16959
12:03 mikehh codetest, distro_tests FAIL, All others PASS - TODO PASSES in examples_tests
12:04 mikehh codetest and distro_tests FAIL the same test - should it be run twice?
12:05 mikehh t/distro/file_metadata.t - Failed tests:  3-4
12:06 mikehh all related to the new docs/book/pir
12:18 mikehh applying the PATCH http://nopaste.snit.ch/16958 fixes the TODOs and does not change the other test results
12:24 ruoso joined #parrot
12:47 away_from_coke src/thread.c:501: warning: variable 'sub_pmc' might be clobbered by 'longjmp' or 'vfork'
12:53 skids joined #parrot
12:56 dalek parrot: r39650 | coke++ | trunk/docs/book/pir (3 files):
12:56 dalek parrot: un-TODO some passing PIR examples
12:56 dalek parrot: Patch courtesy mikehh++
12:56 dalek parrot: (plus minor update to one example to move code from a comment into PIR to
12:56 dalek parrot: make sure it is also tested.)
12:56 purl okay, dalek.
12:56 dalek parrot: review: https://trac.parrot.org/parrot/changeset/39650/
12:59 dalek parrot: r39651 | coke++ | trunk/docs/book/pir (9 files):
12:59 dalek parrot: pass file_metadata.t
12:59 dalek parrot: Courtesy mikehh++
12:59 dalek parrot: review: https://trac.parrot.org/parrot/changeset/39651/
13:21 sparkymat joined #parrot
13:23 Andy joined #parrot
13:26 gryphon joined #parrot
13:44 mikehh so far builds ok - pre-config/post-config tests - smoke tests PASS - distro_tests PASS, codetest PASS, examples_tests PASS (no more TODOs)
14:00 sparkymat joined #parrot
14:00 iblechbot joined #parrot
14:01 ascent_ joined #parrot
14:17 chromatic joined #parrot
14:31 mikehh make fulltest PASSes at r39651 - Kubuntu 9.04 Amd64
14:36 dalek tpfwiki: Guru Murure | First Post in Academia Weblog
14:36 dalek tpfwiki: http://www.perlfoundation.org/parrot/i​ndex.cgi?first_post_in_academia_weblog
14:36 dalek tpfwiki: Guru Murure | Navigation for: Academia Weblog
14:36 dalek tpfwiki: http://www.perlfoundation.org/parrot/in​dex.cgi?navigation_for_academia_weblog
14:39 ilia joined #parrot
14:42 jan joined #parrot
14:43 dalek tpfwiki: Guru Murure | First Post in Academia Weblog
14:43 dalek tpfwiki: http://www.perlfoundation.org/parrot/i​ndex.cgi?first_post_in_academia_weblog
14:45 mj41 hmm, r39648 fixed FreeBSD http://tt.ro.vutbr.cz/report/pr-Parr​ot/do?trun-4903=on&amp;trun-4892=on
14:45 Coke hey, is that tpfwiki blog spam?
14:45 mj41 cotto++ :-)
14:46 Andy joined #parrot
14:48 dalek cardinal: f600405 | fperrad++ | src/classes/Integer.pir:
14:48 dalek cardinal: The opcode 'gcd' was not removed, just num variant.
14:48 dalek cardinal: So, use integer variant.
14:48 dalek cardinal: Signed-off-by: cardinal <tene+cardinal@allalone.org>
14:48 dalek cardinal: review: http://github.com/cardinal/cardinal/commit​/f6004056c99a9258dd9255e55381f8c6cc27201b
14:58 Coke RT's remaining: 299
15:10 dalek TT #768 created by Geero++: Writing my custom essay
15:14 Coke Geero--
15:14 Coke Geero--
15:14 Coke RT's remaining: 299
15:14 Coke RT's remaining: 299
15:14 Coke RT's remaining: 299
15:14 Coke RT's remaining: 299
15:15 Coke RT's remaining: 299
15:15 Coke Geero--
15:15 Coke Geero--
15:15 Coke Geero--
15:15 burmas left #parrot
15:16 Whiteknight joined #parrot
15:16 szbalint spammer?
15:16 purl Spammer, spanner. Spanner, spammer. "Ow!"
15:20 dalek TT #768 closed by coke++: Writing my custom essay
15:21 donaldh joined #parrot
15:26 Theory joined #parrot
15:39 amuck joined #parrot
15:41 NotFound Mmm... my patch for #767 had a serious problem, it only catched the duplicates on the same type, which are the less problematic X-)
15:42 NotFound And here is a problem: runtime/parrot/library/PGE/Perl6Grammar.pir, method 'compile': use a local symbol namespace as pmc and as string.
15:47 Whiteknight it's amazing to me that it works at all
15:50 Whiteknight joined #parrot
15:54 dukeleto joined #parrot
15:57 NotFound pmichaud: "(I don't expect it to arise frequently)" frequent enough to be unable to build parrot
15:57 pmichaud NotFound: ???
15:57 NotFound pmichaud: your comment about duplicate local and param
15:58 pmichaud oh, there's code that redeclares params?
15:58 NotFound pmichaud: at least two, and I didn't even finished the build
15:58 pmichaud I meant "I don't expect it to arise frequently for people who are generating code via PCT"
15:59 pmichaud (which was the context of my original concern about making all symbol redeclarations an error)
15:59 pmichaud 15:42 <NotFound> And here is a problem: runtime/parrot/library/PGE/Perl6Grammar.pir, method 'compile': use a local symbol namespace as pmc and as string.
15:59 pmichaud what problem there?
15:59 purl problem there is htat you need to invoke an alternate runloop, though that's doable.
16:00 Whiteknight purl forget problem there
16:00 purl Whiteknight: I forgot problem there
16:00 NotFound pmichaud: as long as we don't check anything, any problem
16:00 pmichaud oh, I see.
16:00 pmichaud "namespace" gets redeclared as a string.  That's a bug.
16:01 NotFound Yeah. And this is the reason that makes desirable to add the checks.
16:01 pmichaud Yes, I'd like very much for imcc to catch those.
16:01 pmichaud But only when one tries to redeclare a symbol with a different type.
16:01 Whiteknight I'm very surprised that IMCC doesn't freak out about redeclarations when building it's symbol table
16:03 dukeleto joined #parrot
16:04 NotFound pmichaud: Can you fix that? I'm not familirized enough with PGE
16:04 pmichaud sure, I'll fix it.
16:04 pmichaud actually, enter a ticket for it and assign me as owner.  I'm a bit swamped today and won't have a good chance to test.
16:05 NotFound Ok
16:05 pmichaud (rakudo release today, preparing for pvwm, preparing for travel)
16:06 pmichaud and lunch would be good here.
16:06 NotFound I'm thinking about emitting a warning instead of failing, in the meantime.
16:06 pmichaud I still want the option of no warning on redeclared symbol, though.
16:07 pmichaud (redeclared with same type)
16:07 NotFound I'll leave that for later, now checking only for the different type case.
16:07 pmichaud okay, that would be great.
16:07 pmichaud NotFound++
16:07 dalek rakudo: 6765def | pmichaud++ | docs/spectest-progress.csv:
16:07 dalek rakudo: spectest-progress.csv update: 404 files, 11536 passing, 0 failing
16:07 dalek rakudo: review: http://github.com/rakudo/rakudo/commit/6​765def9568a106f029a8f39d44b739242d20fa0
16:08 pmichaud on http://www.parrot.org/download, the link to "Current developer release" still points to 1.2.0
16:09 pmichaud also, the link to "Current supported release" goes to a place that has "stable" in the url.  I'd like to see s/stable/supported/
16:09 Coke I think both of those are on the ftp site. ticket's probably best, allison or jerry can resolve it later.
16:10 pmichaud okay.
16:10 * pmichaud files tickets.
16:10 Coke danke.
16:10 pmichaud I really wish we had an email interface for filing tickets.
16:10 pmichaud logging into trac is a pain.
16:12 darbelo joined #parrot
16:15 Whiteknight oh, hmmm. I must not have updated the download link on www.parrot.org
16:15 pmichaud I'd do it, but I don't see how to edit the page.
16:16 Whiteknight I'm doing it now
16:16 dalek TT #769 created by pmichaud++: downloads link on parrot.org points to wrong tarball
16:16 Whiteknight (at least, I think I am doing it)
16:17 Psyche^ joined #parrot
16:19 Whiteknight okay, I was looking at the right thing. I don't think I have administrator rights on www.parrot.org
16:20 dalek TT #770 created by pmichaud++: urls for "supported releases" say "stable"
16:21 NotFound Fortunately, it looks like there were no more pir files with wrong duplicates other than those two :)
16:26 dalek parrot: r39652 | NotFound++ | trunk/compilers/imcc/symreg.c:
16:26 dalek parrot: [imcc] emit a warning on duplicate .local or .param identifiers with different type, TT #767
16:26 dalek parrot: review: https://trac.parrot.org/parrot/changeset/39652/
16:27 HG` joined #parrot
16:36 Coke NotFound: Duplicated IDENTIFIER 'value'
16:37 Coke getting those in .tcl; any chance the warning can say /where/ this is occuring?
16:37 dalek TT #771 created by NotFound++: [bug] Duplicate identifier in Test;More.is_deeply
16:38 Coke (happens when I build runtime/tcllib.pir, but value isn't duplicated there, I think.
16:40 NotFound Coke: I was wondering the same, is a shame that IMCC_warning does not show line number.
16:40 dalek TT #772 created by NotFound++: [BUG] Duplicate identifier in PGE;Perl6Grammar.compile
16:40 Coke NotFound: without that, the warning is prety much just an annoyance for me. =-()
16:40 Coke er, =-)
16:41 Coke Whiteknight: the link on teh page is to http://www.parrot.org/release/developer
16:41 cotto mj41, I have no idea how r39648 could cause *any* user-visible changes, even good ones.
16:41 cotto I wasn't even thinking about that problem with that commit.
16:41 Whiteknight Coke: Yeah, I don't have rights to change that link
16:41 Whiteknight or, change where it points to
16:41 purl Whiteknight: that doesn't look right
16:41 cotto purl, it?
16:41 purl it is, like, in the universe repo which is not enabled by default
16:41 Coke right. we don't want to change the download page.
16:41 NotFound Coke: well, annoting people is sometimes a way to make them fixing things }:)
16:42 Coke NotFound: or annoy them.
16:42 NotFound Urgh
16:42 NotFound I'll take a look at imcc/debug.c
16:43 mj41 cotto: I know. These are weird bugs.
16:43 Coke NotFound: thank you. =-)
16:43 pmichaud afk, lunch
16:44 Coke Whiteknight: did you do the steps in the release manager guide about mkdir on the ftp server?
16:44 Whiteknight Coke, the release is on the ftp server, yes
16:44 Coke no.
16:44 Whiteknight ftp://ftp.parrot.org/pub/pa​rrot/releases/devel/1.3.0/
16:44 Coke line 194.
16:44 Coke of docs/project/release*
16:44 Coke did you do that?
16:44 * Whiteknight has to look it up
16:45 Coke re-reading I think you did.
16:45 Whiteknight yes, that's the link I just gave you
16:45 Coke I wonder what trigger-parrot does.
16:46 Whiteknight it propagates the new file to all the FTP mirrors
16:46 Whiteknight The place where I can't continue is at line 270
16:46 Coke I meant under the hood.
16:47 Coke what's your drupal id?
16:48 Coke ah, there you are.
16:49 Coke Whiteknight: try now.
16:50 Coke (if that works, add "request the admin role" instead of "story creating powers"
16:50 Coke thank you for you patience.
16:50 Whiteknight done!
16:50 Whiteknight Coke++
16:51 Coke you need to be an admin to change the links.
16:51 Coke whee.
17:03 Whiteknight it's my sneaky way of getting admin privileges on a bunch of websites
17:05 cotto Coke, does partcl work with r39648?
17:05 cotto That commit may break PMC multiple inheritance.
17:06 NotFound Coke: a little problem: if I make IMCC_warning show file and line number infomation a lot of test that expect the warning message fails.
17:07 NotFound (test that check minutiae)--
17:08 dalek parrot: r39653 | cotto++ | branches/pmc_pct/compilers/pmcc/src/builtins.pir:
17:08 dalek parrot: [pmcc] pir simplification from coke++
17:08 dalek parrot: review: https://trac.parrot.org/parrot/changeset/39653/
17:11 sekimura joined #parrot
17:11 particle notfound: create a branch, apply there, and ask for help from the cage cleaners
17:13 NotFound particle: I think I'll take the 'special case' route instead.
17:19 Coke cotto: checking.
17:19 purl somebody said checking was just different
17:20 cotto I have a fix in mind if it breaks partcl.
17:20 Coke cotto: partcl no longer users PMC multiple inheritance. Does that obviate your request?
17:20 Coke s/users/uses/
17:23 HG` joined #parrot
17:23 Coke cotto: seems to work.
17:23 cotto Coke, very yes.
17:23 Coke but that may be because TclObject is gone.
17:23 Coke I was trying to avoid leaving anything around for you to break! =-)
17:23 mikehh_ joined #parrot
17:24 cotto actually, I still need to support it.  It's an explicit feature that shouldn't be broken without proper discussion and a deprecation notice.
17:25 cotto I kinda wish something in Parrot used (or at least exercised) that feature.
17:25 Coke I won't be your whipping boy. =-)
17:26 Coke write a test?
17:26 Coke or deprecate it for post-1.4?
17:28 NotFound Did we need a dprecation cycle for a feature that never worked well?
17:29 Coke it worked for tcl.
17:29 Coke (presuming we're talking about multiple inheritance still)
17:30 cotto It behaved consistently enough that HLL devs could rely on it.
17:30 Coke so, yah. :|
17:30 Coke but now's a great time to deprecate stuff, just before the next stable. =-)
17:30 cotto I would like to get rid of it.
17:31 dukeleto joined #parrot
17:33 Coke it just helps with composition at compile time, no? I think we can deprecate it; for any real class-level multiple inheritance, that's a different thing.
17:33 NotFound Ah, sorry then. I've heard someone saying that it never worked.
17:34 cotto If bacek and I aren't very careful, it could be a rocky transition between pmc2c and pmcc.
17:35 cotto I'll file a tt on that.
17:39 dalek TT #773 created by coke++: Can't override vtables of a PMC from PIR
17:41 Coke NotFound, cotto: tcl only ever used it to avoid redefining vtables; turns out there were only 3 vtables, each needed in 2 different pmcs. all of them about 3 lines long. I just copied the code. =-)
17:41 cotto Hmmm.  I'm having trouble coming up with a compelling reason other than "I don't want to support it."
17:41 dalek parrot: r39654 | NotFound++ | trunk/compilers/imcc/symreg.c:
17:41 dalek parrot: [cage] add file/line info to duplicate local warning, TT #767
17:41 dalek parrot: review: https://trac.parrot.org/parrot/changeset/39654/
17:42 Coke cotto: "we don't need it."
17:42 cotto sold
17:42 mj41 \msg NotFound r39652 http://tt.ro.vutbr.cz/report/pr-Parr​ot/do?trun-4935=on&amp;trun-4915=on on all my machines http://tt.ro.vutbr.cz/report/pr-Parrot/rp-trunk Sorry if I am annoying.
17:42 Coke someone may disagree with that...
17:42 cotto That's why I'm filing a tt.
17:43 NotFound mj41: I'm fixing right now a failure in a test
17:43 NotFound Yes, that test
17:43 purl that test is all set, the rest is all you
17:44 NotFound mj41: check r39655
17:45 Coke msg mikehh Have you considered submitted a CLA?
17:45 purl Message for mikehh stored.
17:45 dalek parrot: r39655 | NotFound++ | trunk/t/compilers/pge/perl6regex/context.t:
17:45 dalek parrot: [cage] fix a duplicate local in a perl6regex test
17:45 dalek parrot: review: https://trac.parrot.org/parrot/changeset/39655/
17:45 Coke msg mikehh s/ted/ting/
17:45 purl Message for mikehh stored.
17:46 Coke NotFound++
17:48 NotFound Coke: Catched the problem?
17:51 dalek partcl: r507 | coke++ | trunk/runtime/ (2 files):
17:51 dalek partcl: Remove duplicate variable declarations.
17:51 dalek partcl: Notfound++ for the parrot warning to track these down.
17:51 dalek partcl: review: http://code.google.com/p/p​artcl/source/detail?r=507
17:53 mj41 NotFound: tapir2 is the bastest one ... http://tt.ro.vutbr.cz/report/pr-Parrot/rp-trunk
17:53 mj41 fastest
17:54 NotFound mj41: in cases like this, I'm faster than robots ;)
17:54 Coke NotFound: there you go. yes, found the duplicates and removed them.
17:55 NotFound Coke: I suppose we have demonstrated that the problem was worth fixing
17:55 Whiteknight Coke: at runtime, I think get_class will return a Class if it's a PIR-defined object, and a PMCProxy if it's builtin
17:55 Whiteknight I can't test that right now though
17:56 pmichaud Whiteknight: that sounds correct.
17:56 Coke Whiteknight: so I can call typeof on the result of get_class?
17:56 Whiteknight Coke, as is my understanding
17:56 pmichaud typeof on get_class will always return the Class PMC
17:56 mikehh Coke: what's the procedure
17:56 purl Procedure must always be followed. It's procedure.
17:56 pmichaud which is not the same as your PIR class
17:57 Coke cla?
17:57 purl cla is Contributor License Agreement or http://www.perlfoundation.org/​contributor_license_agreement or http://www.parrot.org/foundation/legal or http://www.parrot.org/files/parrot_cla.pdf
17:57 Coke the parrot one
17:59 Whiteknight is mikehh becoming a committer?
18:00 Coke I'm willing to sponsor him.
18:00 Whiteknight mikehh++
18:00 Coke (if for no other reason than I don't want to be his patch monkey. =-)
18:01 Whiteknight better to be a patch monkey then to not e receiving any patches
18:02 mikehh hey I've had barney, kid51, whiteknight and moritz, maybe others :-}
18:04 NotFound mikehh: "That's good. You have taken your first step into a larger world."
18:05 Whiteknight i'd be happy to sponsor him too
18:05 Whiteknight karma mikehh
18:05 purl mikehh has karma of 32
18:06 Tene Man, I remember back when I was submitting patches...
18:06 Tene Well, okay, I actually don't...
18:07 Coke Whiteknight: great, you can have him. =-)
18:07 * Coke wonders if he's violating a foundation procedure or something.
18:07 Coke You'd think the VP would know.
18:08 Coke let's ask him.
18:08 Coke "to be eligible, a person must be nominated by two current members..." and must sign a CLA. woot.
18:09 * Whiteknight seconds the motion
18:09 Whiteknight w00t indeed
18:10 Whiteknight I got fasttracked into committership because I was a GSOC student
18:10 Whiteknight never would have made it in this new world of "rules"
18:11 dukeleto joined #parrot
18:12 mikehh I hope this works out ok - great care to be taken - or something like that :-}
18:13 mikehh how does one get the CLA signature to you all
18:13 Whiteknight postal service
18:13 purl i guess postal service is too slow.
18:14 MoC joined #parrot
18:14 Whiteknight purl no postal service is very slow but necessary for CLAs
18:14 purl OK, Whiteknight.
18:14 Whiteknight actually, you might be able to fax
18:14 Whiteknight postal service?
18:14 purl i heard postal service was too slow.
18:14 Whiteknight damnit purl!
18:14 purl i guess damnit i am a bot!!!
18:15 mikehh you mean i gotta use snail mail - horrors
18:16 Coke it's on the site, I think. moment.
18:16 cotto I thought a scanned copy sent by email also worked.
18:16 Coke mikehh: are you going to YAPC?
18:17 Coke you get extra bonus points if you just hand it in in person. =-)
18:17 Theory joined #parrot
18:17 mikehh unfortunately I am in Aberdeen, Scotland at the moment
18:17 Coke I'm pretty sure there's a fax number. I think a scan of one via email might be sufficient.
18:17 Coke you can always try.
18:18 Coke (send it to me, I can forward it to the board.)
18:18 Whiteknight GeJ is on an island in the middle of the freaking pacific, and he sent in a CLA
18:19 mikehh he he - ok - I'll print/scan it and send it both ways
18:20 NotFound Whiteknight: from spain it took a month or so
18:20 Whiteknight you're not supposed to use actual "snails" for the mail
18:21 Whiteknight send it with a person
18:21 Coke ^_-
18:21 dukeleto joined #parrot
18:22 NotFound Whiteknight: Spain's postal service is... my english knowledge isn't enough to express that.
18:23 pmichaud Coke: (re #773)  I generally test to see if the class isa PMCProxy
18:24 dalek TT #774 created by cotto++: deprecate PMC multiple inheritance
18:29 Whiteknight that's just for C-based PMCs, right?
18:29 Coke there are no non-c-based pmcs. =-)
18:30 Whiteknight whatevs
18:30 * Coke supposes that's is both technically true and technically not true, depending.)
18:31 Whiteknight Coke: We only have two more release managers on the schedule, July (cotto) and August (rgrjr)
18:31 Coke do .pmc files have to have the same name as the thing they contain?
18:31 Whiteknight Coke: yes
18:32 Coke Whiteknight: excellent point; do you want to coordinate getting us more releasors up through 2.0 ?
18:32 Coke howto: (solicit for volunteers, pick dates, make sure everyone's happy, update the doc.)
18:32 Whiteknight Yeah, I can do that. I'll shoot an email to the list and see what pops up
18:33 * Coke gets rid of his last remaining parrot task.
18:36 mikehh I am heading back to my Ubuntu 9.04 i386 setup - got to reboot - cu in a bit
18:36 Whiteknight sent
18:40 Coke whee. ok. now to transition from C to PIR, I have a dynpmc that extends a core pmc and then a class that subclasses that.
18:40 dngor joined #parrot
18:42 workbench joined #parrot
18:42 Whiteknight yay complexity!
18:42 Whiteknight what's the purpose of the dynpmc at all, why not just have it in pure PIR?
18:43 pmichaud there once was a time when pure PIR classes weren't a viable option.
18:43 bacek_ joined #parrot
18:43 Whiteknight pmichaud: with a few notable bugs and exceptions, that's not really the case anymore
18:43 Whiteknight mostly
18:43 pmichaud agreed, I'm just saying that partcl comes from that time.
18:43 pmichaud s/partcl/much of partcl/
18:44 pmichaud that said, Rakudo also has a dynpmc that extends a core pmc and then a class that subclasses it.  :-)
18:44 pmichaud and the reason is that there's not an easy way to add C-based methods to a PIR class
18:44 pmichaud (and we have an operation that really wants to be done in C, not PIR)
18:44 Whiteknight pmichaud: that's actually a very interesting problem
18:45 Coke I think the first version of partcl was committed to the parrot repo back in 2001.
18:45 pmichaud I've tried a number of different workarounds with no real success.  I'd really like to get rid of the intermediate dynpmc, but haven't found a way to do it.
18:45 pmichaud Objects weren't really viable until October 2007.
18:45 Whiteknight wow, that's recent
18:46 * Whiteknight doesn't know a lot about Parrot history
18:46 NotFound Coke: At a HAL factory in Illinois?
18:46 pmichaud there was some support for Objects prior to that, but being able to reliably overload :vtable methods and handle nested namespaces and a variety of other things weren't really available until 2007.
18:47 pmichaud It's not coincidental that that's when PCT and Rakudo really started to take off.
18:53 chromatic joined #parrot
18:57 Coke hurm. should hll_map() work with classes instead of PMCs?
18:57 pmichaud it's supposed to, yes.
19:01 dalek parrot: r39656 | NotFound++ | trunk/t/compilers/imcc/syn/errors.t:
19:01 dalek parrot: [cage] put todo'ed test up to date with TT #767 changes
19:01 athomason joined #parrot
19:01 dalek parrot: review: https://trac.parrot.org/parrot/changeset/39656/
19:01 Andy joined #parrot
19:02 MoC hi, I'm currently trying to build parrot on windows and would be glad if someone could tell my what is wrong with my makefile:
19:02 nopaste "MoC" at 91.11.26.226 pasted "Invalid Makefile on Win32" (3355 lines) at http://nopaste.snit.ch/16964
19:04 Coke I always use ming32-make on windows, myself; quite possible that there are some non-dmake-friendly items in the generated file.
19:05 MoC ok, I'll try it, thanks
19:05 Coke what did Configure.pl tell you to use?
19:05 MoC "You can now use `dmake' to build your Parrot."
19:06 MoC which means that it probably would be good if it got fixed anyway...
19:06 Coke yup. the O redefinition there must be conflicting with an internal dmake rule.
19:06 Coke $(O) of .o on windows seems wrong though.
19:06 NotFound MoC: we can't fix any possible predefined values for any default of config of any variant of make in the world.
19:07 Coke NotFound: if we don't support dmake, we shouldn't claim to.
19:08 Coke MoC: if you can open a ticket (attach the generated makefile instead of inlining it), we can see about not redefining those items if we know we're using dmake.
19:08 MoC at parrot's trac?
19:10 Coke <nod>
19:10 NotFound Coke: That depends, if the value comes from some predefined rule of the installation we can't prevent it.
19:10 Coke NotFound: huh?
19:10 Coke you mean, if someone has customized their dmake?
19:10 NotFound Coke: yeah
19:10 Coke MoC: have you customized your dmake?
19:11 amuck joined #parrot
19:11 MoC no. I'm using the one from cpan, if that makes a difference?
19:11 NotFound I don't think any sane variant of make predefine common names as A and O
19:12 pmichaud afk for a while
19:13 Coke NotFound: let me turn this around.
19:13 Coke How would you suggest we fix moc's problem?
19:13 NotFound Let's diagnose it before fixing ;)
19:14 Coke Sure. what would you suggest?
19:15 NotFound I'll try to install dmake on my windows machine and see what happens
19:20 donaldh joined #parrot
19:23 NotFound Uh, lots of modules to update, will take a while
19:24 Whiteknight not exactly a flood of responses to my email yet
19:24 mikehh joined #parrot
19:25 Whiteknight NotFound: have you ever done a release?
19:25 NotFound Whiteknight: I almost never knows in advance when I have enough free time
19:28 Whiteknight is Darbelo a committer?
19:31 Whiteknight I've got such a terrible memory for that kind of stuff
19:31 Coke darbelo is the new GSOC student.
19:31 Whiteknight right, so was he made into a committer for that?
19:31 Coke let him concentrate on his work. =-)
19:31 Whiteknight He could manage a release after the summer is over
19:33 particle Whiteknight: sign me up, it's been too long
19:34 Whiteknight Coke: How do we go about making a new committer anyway? Do we raise the issue at #ps or on the list or what?
19:35 particle i'll take september, for now.
19:35 Whiteknight September it is
19:35 particle Whiteknight: we vote at #ps
19:35 Coke Whiteknight: asking on #ps is fine.
19:35 Whiteknight particle: I didn't know if the process changed now that we have a fancy foundation with bylaws
19:35 Coke particle: not according to the bylaws. =-)
19:35 * Whiteknight never read all the bylaws
19:36 Coke but voting at PS covers the bylaws.
19:36 particle bylaws schmylaws.
19:37 dalek parrot: r39657 | whiteknight++ | trunk/docs/project/release_manager_guide.pod:
19:37 dalek parrot: [release_managers] particle++ has september
19:37 dalek parrot: review: https://trac.parrot.org/parrot/changeset/39657/
19:37 particle Whiteknight: while you're in there, can you correct the sep release date to the 15th?
19:39 brbrooks joined #parrot
19:42 solarion joined #parrot
19:43 Whiteknight done
19:44 dalek parrot: r39658 | whiteknight++ | trunk/docs/project/release_manager_guide.pod:
19:44 dalek parrot: [release_managers] the september release is actually on the 15th.
19:44 dalek parrot: review: https://trac.parrot.org/parrot/changeset/39658/
19:46 * darbelo is not a commiter.
19:47 Whiteknight damnit
19:47 athomason hi all, who can fix parrot.org problems?
19:47 Whiteknight athomason: what problem?
19:47 purl problem is getting an installed parrot to recognize dynops/dynpmcs
19:47 Whiteknight purl forget problem
19:47 purl Whiteknight: I forgot problem
19:47 dalek parrot: r39659 | whiteknight++ | trunk/docs/project/release_manager_guide.pod:
19:47 dalek parrot: [release_managers] fix a few more dates too
19:47 dalek parrot: review: https://trac.parrot.org/parrot/changeset/39659/
19:47 athomason http://www.parrot.org/cage-cleaners/ points to the old svn repo
19:47 Whiteknight ah, good one
19:48 Whiteknight fixed
19:48 Whiteknight thanks!
19:48 athomason np :)
19:48 * athomason 's first parrot contribution in about 6 years
19:48 NotFound Did you asked for darbelo? "I went down with a case of acute bronchitis"
19:49 darbelo I *think* I submitted a CLA, I think I did it when GSoC started, but then it turned out to be unnecessary so I forgot about the whole thing.
19:49 NotFound Oh, you are here X-)
19:49 darbelo NotFound: But I came back up!
19:49 NotFound Good :)
19:50 cotto athomason, wb!  You'll find a very different bird.
19:50 athomason cotto: thanks, no doubt
19:53 darbelo Ha! I kew I had signed a CLA, I've just found the bounced mail in my spam folder.
19:54 darbelo The date on the pdf is "30 Apr 2009" is that still good for a resend?
19:56 NotFound darbelo: I think a hardcopy is required
19:57 brbrooks Is there anything about CPS in Parrot?
19:58 Whiteknight brbrooks: what do you mean?
19:58 brbrooks Any written documentation/explanation
19:59 PerlJam docs/pdds/pdd03_calling_conventions.pod perhaps
19:59 PerlJam or book/ch03_pir.pod
20:02 darbelo The CLA says "Please complete and send a PDF of the signed CLA to legal@parrot.org, or fax a copy of the signed CLA to (number), or send an original signed CLA to :"
20:03 darbelo it even puts the "or" in boldface.
20:03 mikehh_ joined #parrot
20:04 Whiteknight brbrooks: There probably isn't any good documentation of the low level operations of it
20:06 eternaleye joined #parrot
20:06 Whiteknight at least none that I am aware, although I would be happy to write some eventually
20:09 brbrooks So I understand the gist of L1... Are there any other "smaller" projects regarding the core VM that would be a good place for a new developer to start?
20:10 Whiteknight brbrooks: yeah, L1 is really in the distant future, but there are lots of other things to work on now
20:10 Whiteknight any particular subsystems you're interested in?
20:10 brbrooks As low as you can go
20:11 Whiteknight Ah, an on-the-metal man. I like that
20:11 Whiteknight I'm planning some work on the GC and the IO system in the very near future, and the Calling Conventions system too, if you're interested in any of that
20:11 Whiteknight always need more help in those areas
20:12 brbrooks I love PL theory, don't get me wrong.  But right now I'm concerned about core VM, lower level operations, performance, where the VM hits the metal, how to make things scream, etc...
20:12 Whiteknight gotchar
20:12 Whiteknight gotcha*
20:12 particle gc! gc!
20:13 Whiteknight the GC is a huge slowdown, so if you're looking to optimize something that's a good place to start
20:13 particle we have a student working on the jit subsystem this summer
20:13 PerlJam It would be awesome if parrot could be made to scream (especially in those operations that rakudo or other languages depend heavily upon)
20:13 particle so ignore that for now.
20:13 Whiteknight particle: how is tewk doing? I haven't heard a single peep about his project
20:13 cotto I was just going to ask the same thing about tewk.
20:13 particle Whiteknight: you know as much as i do.  i've asked allison to bother him.
20:14 Whiteknight fair enough
20:14 particle i'm not happy.
20:15 Whiteknight yeah, luckily we're not irretreivably far into the summer, he still could bang it out
20:16 Whiteknight darbelo++ is kicking ass though
20:17 particle great! i haven't tracked his progress to closely, just reviewing commits and browsing blog entries
20:17 particle looks quite promising, but i haven't asked whether it's on track or not.  cotto?
20:17 cotto I know nothing.
20:18 Whiteknight as far as I can see, it definitely is on track
20:18 Whiteknight is darbelo still here?
20:18 moritz same here
20:18 cotto darbelo is still around.  He was just out of commission for a while.
20:18 cotto He's back (and blogging) now.
20:19 NotFound I think I've found the problem with dmake (and maybe other make incarnations)
20:19 whoppix joined #parrot
20:20 NotFound It has problems with commands that start with a '['
20:20 NotFound The solution is to use test instead
20:20 particle the make engine, or the shell has problems?
20:21 NotFound particle: don't know
20:21 muixirt apropos llvm, moritz what were the steps of building parrot with clang?
20:21 Whiteknight I can't even get the DecNumber library built on my machine, I need darbelo's help with that so I can test his stuff
20:21 Whiteknight moritz: yes, I would love to build Parrot with clang
20:21 moritz muixirt, Whiteknight: last time I tried it failed somewhere during linking phase
20:22 NotFound particle: I suppose the make engine, never heard of mingw problems with that
20:22 moritz the step were simply 1) build clang + llvm from source 2) perl Configure --cc='path/to/clang/
20:22 muixirt moritz, what version did you use?
20:22 moritz muixirt: about two months ago from svn
20:23 moritz it did work with llvm-gcc at some point
20:23 muixirt no Configure.pl --link parameter?
20:23 moritz maybe it needs that.
20:24 NotFound Some objection to replace the [ ] usage with test in config/gen/makefiles/root.in ?
20:24 moritz as I said, I wasn't entirely successful
20:24 muixirt out of the build directory of llvm/clang?
20:24 moritz muixirt: aye
20:24 brbrooks GC might be a good place for me to start.  I am becoming verrrrrry familiar with Tofte/Talpin's region-based mm for my thesis
20:25 Whiteknight I read that paper recently, very interesting
20:25 cotto Whiteknight, a co from darbelo's svn repo is necessary and sufficient.  It comes with the DecNumber code.
20:25 Whiteknight what's your thesis?
20:25 purl well, your thesis is on what, again?  And have you defended yet?
20:25 moritz muixirt: there used to be a script named 'ccc' that wrapped clang to accept gcc options - I'm not sure it's still needed
20:26 clunker3 joined #parrot
20:26 muixirt moritz, oh I was under the impression that clang itself is the wrapper...
20:27 brbrooks 100% type safety without GC or runtime checks... very ambitious, but taking an alias/linear type sys approach rather than region-based
20:27 moritz muixirt: no, clang is an entirely new C compiler
20:27 purl okay, moritz.
20:28 moritz I've just svn up'ped llvm, and rebuilding now
20:29 Whiteknight cotto: thanks, I'll try that tonight
20:30 NotFound More problems: looks like dmake is not using a shell to execute commands, but tries to execute it directly. So for example 'if' fails
20:30 muixirt moritz, fine
20:30 NotFound No idea of how to fix that.
20:30 particle NotFound: as it, it's using cmd.exe for a 'shell'?
20:31 NotFound particle: no, the error message will be different
20:31 particle ok
20:31 NotFound It tries to execute a program called 'if'
20:31 particle in general, these things are done in perl, not in the shell
20:31 particle that  way they're portable
20:32 NotFound Looks like parrot is not general. Maybe commander ;)
20:32 particle well, there are no 'if' calls in the windows-generated makefile
20:33 NotFound particle: I'm almost sure I'm running Windows XP Home
20:33 Theory joined #parrot
20:39 NotFound #IF(win32):     if exist $@.manifest ..... and several lines like that
20:39 pmichaud iiuc, those get pre-processed out of the Makefile
20:39 pmichaud during parrot's Configure.pl stage.
20:40 NotFound pmichaud: looks like not
20:40 pmichaud oh, then I guess they're to be treated as comments.
20:41 pmichaud But they don't show up in any of my Makefiles, afaict
20:42 kgilmer joined #parrot
20:42 pmichaud I'm pretty sure they're supposed to be filtered out by the Configure.pl sequence, though.
20:42 pmichaud If they're not, that sounds like a bug somewhere.
20:43 NotFound Well, looks like they are filtered... except for windows
20:43 pmichaud 20:32 <particle> well, there are no 'if' calls in the windows-generated makefile
20:43 pmichaud so, something else must be amiss here.
20:44 pmichaud because it's being filtered for particle.
20:44 NotFound There are 'if' calls in my windows-generated makefile
20:45 MoC In mine, too (Thanks for taking a look NotFound :) )
20:45 pmichaud I'm not claiming you're lying, I'm simply saying it's not just windows that is causing the problem.
20:45 pmichaud it has to be something else.
20:45 NotFound MoC: thanks to Coke, he convinced me ;)
20:45 MoC ok, thank you, too, Coke :)
20:46 NotFound pmichaud: I don't have special interest in blaming Windows, lots of people in the world already does.
20:47 pmichaud the code that is supposed to be handling the #IF(...) directives is lib/Parrot/Configure/Compiler.pm:442
20:48 pmichaud ugh, and it decides to define its own syntax.
20:48 NotFound And this is not supposed to insert it when (win32) ?
20:48 pmichaud it's supposed to add it without the #IF(win32):  part if win32 is true.
20:49 NotFound Then I don't find strange to have it in windows
20:49 pmichaud regardless, the "#IF(win32):" is not supposed to appear in the Makefile
20:49 pmichaud either the line is completely removed, or just the "#IF(win32)" part is.
20:49 NotFound pmichaud: I think we are mixing 'if' %-)
20:49 pmichaud oh.
20:49 pmichaud you mean the 'if'
20:49 pmichaud my apologies.
20:50 NotFound The line gets expanded to: (tab)if ....
20:50 pmichaud and dmake complains about that?
20:50 pmichaud because it doesn't have a shell it can call?
20:51 NotFound No, dmake just tries to execute it... and fails.
20:51 NotFound Looks like it doesn't try to use a shell, rather it directly executes the command.
20:51 pmichaud what is that line trying to do, anyway?
20:52 NotFound mt.exe something
20:52 pmichaud http://msdn.microsoft.com/e​n-us/library/aa375649.aspx
20:53 NotFound Resource compiling or linking, maybe?
20:53 NotFound if exist $@.manifest mt.exe -nologo -manifest $@.manifest -outputresource:$@;1
20:53 particle oh!
20:53 particle i'm not sure windows xp home has 'if'
20:54 particle on vista:
20:54 particle c:\Users\particle\dev\rakudo>if /?
20:54 particle Performs conditional processing in batch programs.
20:54 particle ...
20:54 NotFound particle: In cmd.exe? Sure it has.
20:54 particle ok, then it should work.
20:54 MoC well, I think it should, because "if" is a command used for batch scripting
20:54 MoC yep
20:54 NotFound Even command.com has it
20:54 pmichaud if dmake isn't going through cmd.exe, then yes, 'if' would fail.
20:55 particle dmake must internally set SHELL=cmd.exe
20:55 particle i've not used dmake, so i can't be sure
20:55 particle but the make engine passes that off to a shell, it doesn't do it itself.
20:56 NotFound I've never used dmake until today, I'm just looking at the error messages
20:58 NotFound But if the Makefile is trying to use the cmd incarnation of if as if it where a sh-like shell, no wonder it fails.
20:58 eternaleye joined #parrot
20:58 NotFound Mmmm.... no, if exist looks windows'ish enough
20:59 MoC yes, afaik if exist should be valid syntax
21:00 particle yep.
21:00 particle can you run mt -nologo and see what it outputs?
21:00 particle i get the help for mt.
21:01 MoC I don't have any mt.exe. I suppose it's a linux util...
21:02 particle no, it's a msvc utis
21:02 particle *util
21:02 particle but, you need a new enough version of msvc, that's the problem
21:02 NotFound Possible quick solution: put 'cmd /c ' before
21:02 Whiteknight probably need to add the msvc bin directory to your path
21:02 particle you only need that command with msvc>13 or so
21:02 particle are you running 6.0 or something?
21:02 NotFound I'm not using msvc
21:02 particle well, there you go. what compiler?
21:03 particle the #IF condition needs to check both win32 and msvcver>13 i think
21:04 NotFound Strawbeery perl
21:05 MoC interlude: pthreads.h should not be required to build parrot, right?
21:05 MoC *pthread.h
21:08 rg bacek: ping?
21:09 Theory joined #parrot
21:09 NotFound After prefixing all if with cmd /c another funny problem: it tries to execute 'rem'
21:10 particle ok, strawberry perl is gcc, so you won't be using manifest files or mt
21:13 NotFound particle: looks like it doesn't try to execute it because the file checked by the if exist doesn't exist
21:14 NotFound After fixing the 'rem' problem, it builds.
21:14 dukeleto joined #parrot
21:16 eternaleye joined #parrot
21:16 NotFound But I don't know from where that 'rem' comes
21:18 cotto going to yapc10?
21:18 purl going to yapc10 is probably kid51 or pmichaud or davidfetter or cotto or whiteknight
21:18 cotto particle, are you going?
21:18 NotFound 'ranlib' in parrot_config
21:18 Whiteknight oh, i didn't realize it was an "or"
21:29 particle going to yapc10 is also particle
21:29 purl okay, particle.
21:30 particle yep, i leave in ~16h
21:35 cotto When do you arrive?  I'd like to share a cab but I'm getting to PIT at 1025
21:36 cotto nm.  probably too late for me.
21:43 cotto PMC-level MI should be functional now.
21:44 ttbot cotto: Parrot trunk/ r39660 i386-linux-thread-multi make error http://tt.ro.vutbr.cz/file/cmdout/39114.txt
21:44 dalek parrot: r39660 | cotto++ | trunk/lib/Parrot/Pmc2c (2 files):
21:44 dalek parrot: [pmc2c] generate _get_mro and _get_isa in more MI-friendly way
21:44 dalek parrot: review: https://trac.parrot.org/parrot/changeset/39660/
21:44 particle cotto: pmichaud and i land within 3 minutes of each other, at ~6p
21:45 particle cotto: i think chromatic gets in at a similar time
21:45 particle btw there's a bus (28x) that goes from pit to cmu in 50m for $2.80
21:46 pmichaud particle: http://irclog.perlgeek.de/p​arrot/2009-06-17#i_1247396  (chromatic arrival)
21:46 cotto perfect.
21:46 purl well, perfect is the enemy of good enough.
21:46 cotto particle, the wiki says chromatic's getting in on Sunday.
21:46 particle oh, my mistake
21:46 pmichaud that's what chromatic said yesterday, too.  (see irclog link :-)
21:47 particle i didn't realize he was missing the workshop
21:48 pmichaud Is jhorwitz coming?
21:48 pmichaud (to the workshop -- I know he's at the conference)
21:51 NotFound MoC: I'm unable to find a solution that does not risk breaking the build in other windows environments. Try to just ignore the Configure.pl recommendation and use mingw32-make instead of dmake.
21:51 MoC ok
21:51 NotFound MoC: you can open a bug ticket, so others can take a look at the problem.
21:55 MoC will do
21:57 cotto ttbot?
21:57 cotto who owns that bot?
22:00 MoC concerning ticket creation: component would be "install"?
22:01 dalek parrot: r39661 | cotto++ | trunk/lib/Parrot/Pmc2c/PMCEmitter.pm:
22:01 dalek parrot: [pmc2c] unbreak the build
22:01 dalek parrot: review: https://trac.parrot.org/parrot/changeset/39661/
22:02 NotFound MoC: configure, I think, even if the error is not properly during configure is closely related to it.
22:03 MoC ok
22:03 dalek TT #775 created by MoC++: Building Parrot doesn't work with dmake
22:06 MoC created too early, now it's still marked as "component: install", if someone could fix it please?
22:13 dalek rakudo: 5948c11 | pmichaud++ |  (3 files):
22:13 dalek rakudo: Allow Parrot VERSION to be used in cases when the svn revision
22:13 dalek rakudo: information isn't available (e.g., Parrot built from tarball
22:13 dalek rakudo: or non-svn checkout).
22:13 dalek rakudo: review: http://github.com/rakudo/rakudo/commit/5​948c11826b66b8b31e2d0b8027b1c33c74f8100
22:13 dalek rakudo: 99cad48 | pmichaud++ | docs/ (2 files):
22:13 dalek rakudo: Documentation updates for release.
22:13 dalek rakudo: review: http://github.com/rakudo/rakudo/commit/9​9cad48b9f2c530e84a743e7c5fe665206c0b80e
22:13 dalek rakudo: 7c981fc | pmichaud++ | docs/announce/2009-06:
22:13 dalek rakudo: More announcement updates.
22:13 dalek rakudo: review: http://github.com/rakudo/rakudo/commit/7​c981fcfd414c9ec2575f44304a8b6b52b8e142d
22:13 dalek rakudo: c9d76a9 | pmichaud++ | README:
22:13 dalek rakudo: Formatting fix.
22:13 dalek rakudo: review: http://github.com/rakudo/rakudo/commit/c​9d76a9458a6ba29828f2421dc8e578e268f6768
22:18 dalek rakudo: 1b06df8 | pmichaud++ | docs/ (2 files):
22:18 dalek rakudo: More news.
22:18 dalek rakudo: review: http://github.com/rakudo/rakudo/commit/1​b06df80900096dab8a9f74467f504f73a6115dd
22:28 NotFound MoC: changed, and added a comment with my results
22:28 MoC thanks
22:29 Limbic_Region joined #parrot
22:30 dukeleto joined #parrot
22:34 rg1 joined #parrot
22:36 pjcj joined #parrot
22:41 MoC building (with mingw32-make) actually worked. :)
22:41 MoC (after some adjustments)
22:45 darbelo squawks of the parrot
22:46 darbelo sorry, wrong window.
22:52 Theory joined #parrot
22:55 kid51 joined #parrot
22:58 Whiteknight joined #parrot
22:59 Marti1 joined #parrot
23:00 Marti1 left #parrot
23:02 bacek_ joined #parrot
23:03 kurahaupo joined #parrot
23:04 kurahaupo New here. I'm in NZ (UTC+12) -- is anyone else awake?
23:08 * kid51 is getting sleepy, very sleepy
23:09 * kid51 has to pack for YAPC now
23:10 cotto kurahaupo, hi
23:13 kurahaupo Have just been reading the p6l thread about CPAN6 and the numerous imprecations to join #parrot, so here I am ... but it's deathly quiet here.
23:15 kurahaupo BTW, how many Kiwis & Aussies are here?
23:15 bobke joined #parrot
23:17 Infinoid At least 2 that I know of
23:17 kid51 Australia:  wayland alias
23:17 kid51 New Caledonia:  GeJ
23:18 Infinoid Australia: bacek too I think
23:18 kid51 Australia:  bacek
23:18 patspam joined #parrot
23:18 cotto kurahaupo, sometimes it's dead and sometimes you can hardly keep the conversations straight.
23:19 cotto btw, does anyone know if laundry will be available at cmu?
23:19 kid51 In Western Hemisphere, at least, we're getting ready for Workshop this weekend in Pittsburgh
23:19 kid51 cotto: Yes
23:20 brbrooks joined #parrot
23:20 kid51 College dorms => occasional need for students to change their clothes, stay fashionable  => laundry
23:20 donaldh joined #parrot
23:22 bacek_ good morning
23:22 purl Here I am, brain the size of a planet, and all they say is 'Good Morning'
23:22 Infinoid hai bacek_
23:22 bacek_ oh hai Infinoid
23:23 Whiteknight hello bacek_ and Infinoid
23:23 rg1 hi bacek
23:23 Infinoid ohai Whiteknight and rg
23:23 bacek_ rg: pong
23:24 Whiteknight KTHXBYE
23:24 bacek_ Whiteknight: good localtime()
23:24 Infinoid EHLO
23:24 Whiteknight good localtime() indeed
23:24 rg bacek: is it time to close tt#724? i didn't follow the commit messages close enough
23:24 Infinoid Whiteknight: I just reserved delaware license plates that say "KTHXBYE" :)
23:25 Whiteknight did you really?!?!
23:25 Infinoid yep!
23:25 bacek_ rg: yes.
23:25 Infinoid my CA ones say "KTNXBYE"; someone already had the other spelling. :(
23:25 Whiteknight that's awesome
23:25 Whiteknight mine just says GRV9692
23:26 Whiteknight (which doesn't stand for anything, it's just a boring old licenseplate)
23:27 Infinoid Generally Random Vehicle?
23:28 Infinoid Gosh, Really Virtual!
23:29 Whiteknight yeah, that's probably it
23:29 Whiteknight :)
23:30 bacek_ license plates in NSW limited to 6 characters only...
23:30 Infinoid Time to move then :)
23:30 kurahaupo same in NZ
23:31 bacek_ rg++ # keep cage clean
23:31 bacek_ Infinoid: move where???
23:31 purl move is probably revision 2107.
23:32 Infinoid purl, forget move
23:32 purl Infinoid: I forgot move
23:32 Infinoid bacek_: somewhere you can get a KTHXBYE license plate :)
23:32 Infinoid though "SQUAWK" would be good too
23:32 Whiteknight PIXPLZ
23:32 Whiteknight or OMGBBQ
23:32 Infinoid WANT
23:33 Whiteknight wouldn't that be awesome?
23:33 Infinoid ZOMGBBQ is already taken in CA
23:33 Infinoid (of course)
23:33 dngor joined #parrot
23:34 workbench joined #parrot
23:34 bacek_ Cost:  $110.00 ordering fee plus annual fee of $440.00
23:34 dalek TT #724 closed by bacek++: [bug]  Parrot fails numeric conversion of ucs2 strings
23:34 bacek_ oh shi...
23:34 Infinoid eew
23:34 Infinoid that's for custom plates in NSW?
23:34 Whiteknight NSW?
23:34 purl NSW is lej, TonyC and XFire's state. Sydney is the capital of NSW. or New South Wales, Australia or full of mug gamblers stuck on the pokies
23:35 bacek_ yes.
23:35 Whiteknight ah, australia
23:35 purl ah, australia is big
23:35 Whiteknight purl forget ah, australia
23:35 purl Whiteknight: I forgot ah, australia
23:35 Infinoid purl++
23:35 Whiteknight karma purl
23:35 purl purl has karma of 8610
23:35 Infinoid silly mug gamblers, driving up the prices
23:35 * Whiteknight doesn't believe purl has been useful or even entertaining 8610 separate times
23:35 Infinoid (what's a mug gambler?)
23:36 Whiteknight purl-=8609
23:36 purl Whiteknight: huh?
23:36 Whiteknight damnit
23:36 rg whiteknight: purl is on a lot of channels
23:36 cotto left #parrot
23:36 Whiteknight still, that's a lot of times
23:36 cotto joined #parrot
23:36 Infinoid purl cheats.
23:36 purl cheats are fun.
23:36 cotto That was not what I meant to do.
23:36 Infinoid purl--
23:36 purl Infinoid: sorry...
23:36 Infinoid karma purl
23:36 purl purl has karma of 8610
23:36 Infinoid see?
23:37 cotto Whiteknight, write a bot.
23:37 bacek_ << purl-- >> # Cheating are bad
23:37 cotto an anti-bot bot
23:37 bacek_ karma purl
23:37 purl purl has karma of 8609
23:37 Whiteknight cotto: I wrote a bot once and it was lousy and I got bored
23:37 cotto It'd just have to connect and say purl-- 8600 times ;)
23:38 Whiteknight perl6: say "purl--";
23:38 Whiteknight damnit, not on this channel
23:38 bacek_ polyglotbot: say "purl--";
23:38 bacek_ erm...
23:38 rg rakudo: say "purl--"
23:38 bacek_ rakudo: say "purl--";
23:38 rg hmm or not
23:39 Whiteknight polyglotbot?
23:39 purl it has been said that polyglotbot is running an outdated version of rakudo.
23:39 kurahaupo Whiteknight: if you want OMGBBQ come to NZ -- it's available here.
23:39 Infinoid oh wow, polyglotbot is actually here
23:39 rg it's here, but doesn't look alive
23:39 Infinoid which means, feather3 is up
23:39 * Infinoid goes and tries to fix it.
23:39 kurahaupo ditto PIXPLZ
23:39 Whiteknight kurahaupo: that's far too long a trip for a license plate
23:39 kurahaupo http://www.plates.co.nz/search/
23:40 polyglotbot OUTPUT[error:imcc:syntax error, unexpected $end ('þ')␤   in file '/home/tene/parrot-build2/runt​ime/parrot/library/config.pbc' line 1␤purl--␤]
23:40 polyglotbot OUTPUT[error:imcc:syntax error, unexpected $end ('þ')␤   in file '/home/tene/parrot-build2/runt​ime/parrot/library/config.pbc' line 1␤purl--␤]
23:40 polyglotbot OUTPUT[error:imcc:syntax error, unexpected $end ('þ')␤   in file '/home/tene/parrot-build2/runt​ime/parrot/library/config.pbc' line 1␤purl--␤]
23:41 bacek_ time for $dayjob
23:41 bacek_ see you guys.
23:42 Infinoid happy friday bacek_
23:42 Whiteknight later
23:42 Whiteknight rakudo: say "purl-" . "-";
23:42 Infinoid Whiteknight: polyglotbot needs some updating, I'll let you know when it's ready
23:42 Whiteknight oh wait, P6 doesn't use . anymore for it's concats
23:42 particle Whiteknight: what language is that?
23:42 purl I'm not sure... but it doesn't sound like any kind of English I know.
23:43 particle ~
23:43 Whiteknight right
23:43 Infinoid wasn't it _ at some point?
23:43 polyglotbot OUTPUT[error:imcc:syntax error, unexpected $end ('þ')␤   in file '/home/tene/parrot-build2/runt​ime/parrot/library/config.pbc' line 1␤Statement not terminated properly at line 1, near ". \"-\";"␤␤current instr.: 'parrot;PGE;Util;die' pc 129 (runtime/parrot/library/PGE/Util.pir:83)␤called from Sub
23:43 polyglotbot ..'parrot;Perl6;Grammar;eat_terminator' pc 32164 (src/...
23:43 Whiteknight rakudo: say "purl-" ~ "-";
23:43 * cotto has found washington's license plate site and can apparently get zomgbbq
23:43 MoC joined #parrot
23:44 particle rakudo: print [~] <purl - ->
23:44 particle i'll have to wait :)
23:44 amuck joined #parrot
23:44 Tene polyglot is broken, isn't rebuilding rakudo properly, anyone interested let me know and I'll give you an account on feather3
23:44 Infinoid Tene: I'm looking at it now
23:45 Tene Okay.
23:45 Tene I'm heading home now.
23:45 Infinoid And updating some seriously old checkouts...
23:45 Infinoid oki
23:51 Whiteknight I'm hoping to do some serious hacking this weekend
23:52 Whiteknight I want to get IO cleaned up more, improve pipe support, and eventually rewrite Parrot's test harness in pure PIR
23:52 Infinoid cool!
23:52 Whiteknight tell me that wouldn't be a cool result of YAPC?
23:52 Infinoid and that's before YAPC actually even starts
23:53 Whiteknight well, who knows what will happen, I'll work on whatever other people are working on
23:53 Whiteknight I would love to get together with chromatic and look at the pcc_rewiring branch too
23:53 Whiteknight or if we could get some good minds together, get threads fixed up and implement AIO
23:54 cotto I'm looking forward to getting some good pmc_pct hacking done.
23:54 polyglotbot joined #parrot
23:54 Whiteknight cotto: yeah, that branch is awesome and I hope lots of work gets done on it
23:54 Infinoid rakudo: say 1
23:54 polyglotbot OUTPUT[1␤]
23:54 Infinoid ooh.
23:55 Infinoid lolcode: visible "OH HAI"
23:55 polyglotbot OUTPUT[Parrot VM: Can't stat languages/lolcode/lolcode.pbc, code 2.␤main: Packfile loading failed␤]
23:55 Whiteknight rakudo: say "purl-"~"-";
23:55 polyglotbot OUTPUT[purl--␤]
23:55 cotto Whiteknight, you beat me to it
23:55 Infinoid well, 1 language is better than 0 I suppose
23:55 * Infinoid works on lolcode next
23:55 Whiteknight I almost did a "for 1..8600" to kill purl
23:55 Whiteknight but, decided against it
23:56 Infinoid maybe in some other channel :)
23:58 Whiteknight rakudo: say <purl> X <-- ++>
23:58 polyglotbot OUTPUT[purl--purl++␤]
23:58 Whiteknight !!
23:59 Tene rakudo say 'purl' <<~<< <-- ++>
23:59 Tene rakudo: say 'purl' <<~<< <-- ++>
23:59 polyglotbot OUTPUT[purl--purl++␤]
23:59 Whiteknight TIMTOWTDI
23:59 purl There Is More Than One Way To Do It
23:59 Whiteknight WYSIWYG
23:59 Whiteknight ?

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

Parrot | source cross referenced