Camelia, the Perl 6 bug

IRC log for #parrot, 2010-01-28

Parrot | source cross referenced

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

All times shown according to UTC.

Time Nick Message
00:06 Whiteknight ash_: yes
00:08 Whiteknight cotto_work, of course not. I would like to see her speak in front of a room full of parrot and perl people
00:11 dalek plparrot: 5f7bb98 | (David E. Wheeler)++ |  (9 files):
00:11 dalek plparrot: Update installer to use pgxs.
00:11 dalek plparrot: Moved `t/handler/Makefile` to the root and updated it to support `make installecheck` and an installer script. Other relevant changes.
00:11 dalek plparrot: * Moved the test from `t/test/sql` to `t/sql/test.sql` and added a symlink from `sql` to `t/sql`. This is because `pg_regress` is hard-coded to look for test scripts in `sql`.
00:11 dalek plparrot: * Added `expected/test.out`. This is for `pg_regress` to compare test results.
00:11 dalek plparrot: * Added `test` target using `pg_prove`. You need to install pgTAP to get pg_prove. I'll look at embedding it here soon. Either way, you'll need a recent Test::Harness to use it. It's best to use it for development, though.
00:11 dalek plparrot: * Replace `HOWTO` with `README.plparrot`. It's copied from the first 100 lines or so of pgTAP's `README.pgtap`, and contains build instructions with and without pgxs, as well as testing instructions.
00:11 dalek plparrot: * Added `plparrot.sql.in`. This is the script that can be used to install PL/Parrot. Inserting into `pg_pltemplate` does not appear to be the supported way to add a non-core PL.
00:11 dalek plparrot: review: http://github.com/leto/plparrot/commit/5​f7bb98c6b60de44b2bd5f7dc50bd05469442e64
00:11 theory That bot sure is noisy.
00:13 chromatic In theory, we could quiet it down.
00:13 theory I could poop it out!
00:15 * cotto_work completely misparsed that.
00:15 chromatic Don't be so sure.
00:16 * cotto_work hopes he completely misparsed that.
00:16 theory lol
00:21 slavorg joined #parrot
00:22 abqar joined #parrot
00:29 Whiteknight talk about pooping? I didn't know we were on efnet or #perl
00:31 payload joined #parrot
00:37 cotto_work Nah.  We're just talking about exhaustion and language processing.
01:33 Coke do we have any xml parsers?
01:34 Coke need to be able to parse: http://svn.cfeclipse.org/org.cfecl​ipse.cfml/trunk/dictionary/cf8.xml
01:49 bacek_at_work Coke, fperrad do have xml parser
01:50 bacek_at_work http://github.com/fperrad/xml
01:50 bacek_at_work Whiteknight, any luck with gc_encapsulate?
01:50 kid51 joined #parrot
01:50 Whiteknight bacek_at_work, no luck. I haven't had a lot of time
01:51 bacek_at_work Whiteknight, no worries.
03:04 JimmyZ joined #parrot
03:28 dalek plparrot: 38f8108 | (Joshua Tolley)++ | libsrc/handler/.placeholder:
03:28 dalek plparrot: Add libsrc/handler directory so the new Makefile stuff works
03:28 dalek plparrot: review: http://github.com/leto/plparrot/commit/3​8f81089844d0d8e26dc45e4a0e978b419ae4429
04:02 hicx174 joined #parrot
04:16 plobsing joined #parrot
06:35 abqar joined #parrot
07:01 chromatic joined #parrot
07:03 cotto hio chromatic
07:04 chromatic evening
07:08 JimmyZ joined #parrot
07:13 cotto In what way is "pbc_merge -o out.pbc a.pbc b.pbc c.pbc" different from compiling out.pir, which includes a.pir, et al, directly?
07:13 chromatic There should be no difference.
07:14 cotto That's what I thought.  lemme nopaste something that breaks pct
07:15 nopaste "cotto" at 96.26.227.153 pasted "attempt to break up pct makefile dependencies, breaks pct" (48 lines) at http://nopaste.snit.ch/19397
07:16 chromatic I'm not saying that pbc_merge works correctly.
07:17 cotto Are you confident enough in that diff that you suspect the problem lies in pbc_merge?
07:17 cotto (I don't see any reason why not, but many eyes ...)
07:21 cotto It does have some tests but they're all on small bits of code.
07:24 chromatic Does pbc_merge need to work on $(LIBRARY_DIR)/PCT/HLLCompiler.pbc too?
07:25 chromatic Hm, maybe not.  I'm misreading.
07:26 chromatic Nothing in the Makefile looks wrong to me.
07:26 chromatic I think it's pbc_merge.
07:27 cotto ok.  I'll play with it on that assumption.
07:30 dalek parrot: r43632 | cotto++ | trunk (2 files):
07:30 dalek parrot: [build] change PBCMERGE to PBC_MERGE for naming consistency
07:30 dalek parrot: review: http://trac.parrot.org/parrot/changeset/43632/
07:35 dalek TT #1419 created by cotto++: pbc_merge doesn't correctly merge
07:36 cotto I want to hack at that some more, but I really need to sleep.
07:36 cotto night
07:51 bacek joined #parrot
07:56 fperrad joined #parrot
07:57 iblechbot joined #parrot
09:03 mikehh joined #parrot
09:05 bacek joined #parrot
09:08 mikehh joined #parrot
09:32 hicx174 joined #parrot
09:40 dalek parrot: r43633 | bacek++ | failed to fetch changeset:
09:40 dalek parrot: Sync branch with trunk
09:40 dalek parrot: review: http://trac.parrot.org/parrot/changeset/43633/
09:41 moritz
09:41 moritz
09:41 treed ?
09:41 moritz sorry, paste failure
09:41 treed Heh.
09:44 * bacek suspect some conspiracy
09:45 moritz bacek_at_work: that's always a good guess
09:46 bacek clock?
09:46 purl bacek: LAX: Thu 1:46am PST / CHI: Thu 3:46am CST / NYC: Thu 4:46am EST / LON: Thu 9:46am GMT / BER: Thu 10:46am CET / IND: Thu 3:16pm IST / TOK: Thu 6:46pm JST / SYD: Thu 8:46pm EST /
09:46 bacek at_work???
09:46 bacek NO WAY!
09:46 * moritz wonders why his tab completion picked bacek_at_work over bacek
09:48 bacek moritz, MATRIX HAS YOU!
09:53 hicx174 joined #parrot
10:32 rblasch joined #parrot
10:33 cotto joined #parrot
10:37 gaz joined #parrot
11:18 cotto joined #parrot
11:56 cotto joined #parrot
12:00 bluescreen joined #parrot
12:18 payload joined #parrot
12:25 ruoso joined #parrot
12:49 Coke pbc_merge might be meant to be the same, but it's a completely different way to get a similar effect.
12:50 cghene joined #parrot
13:01 dalek TT #1420 created by masak++: Failed assertion when running 240th test in 01-regex.t in GGE
13:17 whiteknight joined #parrot
13:18 dalek TT #1412 closed by jkeenan++: Remove 'running-make-test'
13:18 dalek TT #1372 closed by jkeenan++: Change usages of 'Getopt::Obj' to ['Getopt';'Obj']
13:24 cognominal joined #parrot
13:39 bluescreen joined #parrot
13:43 * Coke -> offline
13:43 dalek parrot: r43634 | coke++ | trunk/DEPRECATED.pod:
13:43 dalek parrot: getopt has already been updated.
13:43 dalek parrot: We don't need to deprecate things in "examples". Those are meant to reflect
13:43 dalek parrot: functioning code, they aren't part of the standard library or API.
13:43 dalek parrot: review: http://trac.parrot.org/parrot/changeset/43634/
13:57 payload joined #parrot
13:57 mikehh joined #parrot
13:59 mikehh All tests PASS (pre/post-config, make corevm/make coretest, smoke (#31970), fulltest) at r43633 - Ubuntu 9.10 i386 (g++ with --optimize)
14:04 silug joined #parrot
14:19 plobsing joined #parrot
14:22 plobsing is there a reason F[^P]A won't let me allocate a zero length array?
14:30 bluescreen joined #parrot
14:31 mikehh All tests PASS (pre/post-config, make corevm/make coretest, smoke (#31975), fulltest) at r43634 - Ubuntu 9.10 amd64 (g++ with --optimize)
14:35 whiteknight plobsing: what do you mean?
14:53 JimmyZ joined #parrot
14:57 plobsing whiteknight: $P0 = new ['FixedIntegerArray'] <newln> $P0 = $I0
14:58 plobsing where $I0 is non-negative (but not always non-zero)
15:06 PerlJam joined #parrot
15:19 NotFound plobsing: looking at the code, the reason seems to be that zero is used to indicate that it hasn't be sized yet.
15:23 iblechbot joined #parrot
15:37 cognominal joined #parrot
15:38 Psyche^ joined #parrot
15:50 rurban joined #parrot
15:50 rurban left #parrot
16:05 davidfetter joined #parrot
16:27 plobsing NotFound: would it be reasonable to change them to use a flag to indicate that?
16:29 NotFound plobsing: there is some good reason to justify that change?
16:29 Andy joined #parrot
16:30 NotFound The current way has the virtue of simplicity, I don't think people like to change that without a good reason.
16:30 plobsing laziness - why check that I have zero elements to put in the array?
16:30 plobsing consistency - FPA allows this
16:32 NotFound Mmmm... FPA just ignore it.
16:32 NotFound if (!size) \n      return;
16:33 plobsing yes. allow/ignore is the same for me in that it doesn't die when I try to do it
16:34 NotFound plobsing: but that is different of setting a fixed size of 0.
16:35 plobsing understood, and I would like be able to permanently set the size to 0, but ignoring is good enough for my current objectives
16:36 NotFound plobsing: I'll better open a RFC ticket about it.
16:37 NotFound In order to have some sort of documented rationale, at least.
16:43 dalek parrot-plumage: cb3d38b | japhb++ | metadata/pynie.json:
16:43 dalek parrot-plumage: [METADATA] Added pynie.json, fperrad++
16:43 dalek parrot-plumage: review: http://gitorious.org/parrot-plumage/parrot-plumag​e/commit/cb3d38bcc6004b3beb4cb5e7f5dacc48258cc194
16:49 dalek plparrot: 55b195e | dukeleto++ | HACKING:
16:49 dalek plparrot: Add a HACKING doc that goes over git workflow
16:49 dalek plparrot: review: http://github.com/leto/plparrot/commit/5​5b195ed426c6ff5579f74551bb18686098b235f
16:59 theory joined #parrot
17:01 riffraff joined #parrot
17:06 silug joined #parrot
17:17 dukeleto joined #parrot
17:38 Andy joined #parrot
17:41 dalek TT #1421 created by plobsing++: [RFC] zero length FxA behaviour
17:42 ZeroForce joined #parrot
17:43 ZeroForce joined #parrot
17:45 ZeroForce joined #parrot
17:46 ZeroForce joined #parrot
17:47 ZeroForce joined #parrot
17:48 ZeroForce joined #parrot
17:53 plobsing joined #parrot
18:17 Tene wagle: A while back, leto said that you wanted to talk with me about scheme on Parrot.  afaik, that never happened.
18:21 wagle Tene..  i wrote a parser in haskell tuesday evening, and i'm currently pondering a similar session in a day or two for the backend (scheme -> pir for a factorial function)..  when I have that going, i'll want to improve it, and then have questions
18:22 wagle i think i know what code to generate, just have to generate it
18:22 Tene Okay, great.  You know where to find me.  :)
18:22 Tene I'm glad to answer any questions.
18:22 wagle heh..  sure..  thanks!
18:49 ruoso joined #parrot
18:52 bluescreen joined #parrot
19:04 bacek joined #parrot
19:43 silug joined #parrot
20:23 joeri joined #parrot
20:38 software joined #parrot
21:14 coke joined #parrot
21:24 cghene joined #parrot
21:30 japhb joined #parrot
21:36 plobsing joined #parrot
21:41 mikehh joined #parrot
22:08 Whiteknight joined #parrot
22:11 patspam joined #parrot
22:30 * coke
22:33 Whiteknight hello
22:34 cotto_work hello
22:56 cotto_work It looks like pbc_merge naively concatenates the const tables from the pbc files it merges.  Is it possible/likey that the problem in TT #1419 occurs as a consequence of that behavior?
22:59 NotFound cotto_work: I guess yes
23:01 particle joined #parrot
23:05 cotto_work I can see why the current code took the lazy route.
23:07 cotto_work This is definitely suspicious.  The merged pbc has two otherwise identical constants with the value of "language", which is the method name that parrot complains about not seeing.
23:09 Coke I got a 1928 2-dollar bill from the bank today.
23:09 Coke whee!
23:09 cotto_work That'll bring in more than $2.
23:10 Coke cotto_work: when partcl originally adopted pbc_merge, we were warned that it wasn't 100%.
23:10 cotto_work Is it still even legal tender?
23:10 Coke (we don't even use it anymore, I don't think.)
23:10 cotto_work and so it isn't
23:10 cotto_work I don't mind hacking on it.
23:10 Coke it /should/ be, though, so ... thanks!
23:10 cotto_work and no, we don't currently use it.
23:11 Coke and yes, I think this bill is still legal tender. the wording is slightly different on that note than on the '76 $2 I have next to it.  and it has red ink. and says "will pay to the bearer on demand".
23:13 Tene I've seen and used $2 bills plenty of times.
23:14 cotto_work Sure, but one from 1928 would be unusual.
23:14 Tene Yeah.
23:20 payload joined #parrot
23:32 jsut_ joined #parrot
23:52 chromatic joined #parrot
23:58 elmex joined #parrot

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

Parrot | source cross referenced