Camelia, the Perl 6 bug

IRC log for #parrot, 2011-06-15

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 kid51 is now known as kid51_at_dinner
00:00 TonyC joined #parrot
00:04 dalek parrot/soh-cah-toa/hbdb: a4eaefd | soh_cah_toa++ | frontend/hbdb/main.c:
00:04 dalek parrot/soh-cah-toa/hbdb: Grouped calls to Parrot_api_make_interpreter() and Parrot_api_set_executable_name() into one expression since it makes sense to call the latter immediately after the former
00:04 dalek parrot/soh-cah-toa/hbdb: review: https://github.com/parrot/parrot/commit/a4eaefdc5b
00:04 dalek parrot/soh-cah-toa/hbdb: dd11f54 | soh_cah_toa++ | src/hbdb.c:
00:04 dalek parrot/soh-cah-toa/hbdb: Renamed hbdb_run_code() to hbdb_runloop() because the former is a little misleading
00:04 dalek parrot/soh-cah-toa/hbdb: review: https://github.com/parrot/parrot/commit/dd11f540a8
00:07 nopaste joined #parrot
00:10 dalek parrot/soh-cah-toa/hbdb: 6c0de1f | soh_cah_toa++ | / (4 files):
00:10 dalek parrot/soh-cah-toa/hbdb: Fixed build errors started by last commit because I forgot to run headerizer and create the API wrapper for hbdb_runloop()
00:10 dalek parrot/soh-cah-toa/hbdb: review: https://github.com/parrot/parrot/commit/6c0de1ff54
00:13 dmalcolm left #parrot
00:40 bubaflub ping whiteknight
00:40 whiteknight pong
00:40 bubaflub whiteknight: i'd like to provide a default to_string type method for GMP Integer - do i override that in Winxed with just a function like:  function get_string[get_string]() { ... }
00:41 whiteknight function get_string[vtable]() { ... }
00:41 bubaflub whiteknight: great.  thanks.
00:41 whiteknight or, there's a more verbose syntax: function foo[vtable("get_string")]() { ... }
00:42 bubaflub get_string works for me
00:42 whiteknight and if you want it to be made available as a named method too, use something like "get_string[vtable,method]"
00:42 bubaflub can i also override things like addition and multiplication?
00:42 whiteknight yeah, but that's a little more complicated because there are several vtables for each operation
00:42 whiteknight take a look at src/vtable.tbl in the parrot repo
00:42 whiteknight that's a complete list of all vtables
00:43 whiteknight you'll see that there is quite a selection
00:43 whiteknight I would personally like to narrow that list down a little, but that's not something to be concerned with now
00:43 bubaflub whiteknight: ok.  it's not on the immediate horizon, but definitely convenient if you could do something like var x = new GMP.Integer(30); var y = new GMP.Integer(32); var z = x + y;
00:44 bubaflub whiteknight: yeah, all of those overrides must incur a penalty to lookup
00:45 sorear vtables are very fast, unless they're overriden, then they're very slow
00:46 bubaflub sorear: so in your opinion should i override addition and multiplication and the like for GMP.Integer?  does the convenience outweigh the speed penalty?
00:47 sorear bubaflub: there's no cost to having the overrides, and it will aid interop
00:47 bubaflub sorear: ok, so the penalty is only when it's used, not just existing.
00:47 sorear correct
00:48 sorear (unless you count the time needed to load the bytecode, of course)
00:50 bubaflub sorear: any idea if Rakudo folks would like to use GMP.Integer?
00:55 kid51_at_dinner is now known as kid51
01:16 whiteknight left #parrot
01:20 woosley joined #parrot
01:51 ingy left #parrot
02:02 ingy joined #parrot
02:03 ingy left #parrot
02:03 Tene_ left #parrot
02:08 bubaflub incoming
02:08 dalek parrot-gmp: 72d7b0d | bubaflub++ | t/integer/init/01-init.t:
02:08 dalek parrot-gmp: fix formatting
02:08 dalek parrot-gmp: review: https://github.com/bubaflub/​parrot-gmp/commit/72d7b0d33d
02:08 dalek parrot-gmp: 19c1c4d | bubaflub++ | t/integer/set/01-set.t:
02:08 dalek parrot-gmp: tests for mpz_set_ui, mpz_set_si, mpz_set_d, and mpz_set
02:08 dalek parrot-gmp: review: https://github.com/bubaflub/​parrot-gmp/commit/19c1c4d928
02:08 dalek parrot-gmp: 0cf3a4a | bubaflub++ | t/integer/set/02-set-str.t:
02:08 dalek parrot-gmp: test mpz_set_str in a few bases
02:08 dalek parrot-gmp: review: https://github.com/bubaflub/​parrot-gmp/commit/0cf3a4a204
02:08 dalek parrot-gmp: 277df91 | bubaflub++ | t/integer/get/0 (2 files):
02:08 dalek parrot-gmp: tests for mpz_get_str, mpz_get_u, mpz_get_si, mpz_get_d, mpz_get_d_2exp
02:08 dalek parrot-gmp:
02:08 dalek parrot-gmp: mpz_get_d_2exp currently segfaults on my machine
02:08 dalek parrot-gmp: review: https://github.com/bubaflub/​parrot-gmp/commit/277df91976
02:08 dalek parrot-gmp: d6826c4 | bubaflub++ | t/integer/swap.t:
02:08 dalek parrot-gmp: add test for mpz_swap
02:08 dalek parrot-gmp:
02:08 dalek parrot-gmp: currently fails; only swaps one instead of both
02:08 dalek parrot-gmp: review: https://github.com/bubaflub/​parrot-gmp/commit/d6826c4a66
02:08 dalek parrot-gmp: ba972a3 | bubaflub++ | t/integer/add.t:
02:08 dalek parrot-gmp: add tests for mpz_add and mpz_add_ui
02:08 dalek parrot-gmp:
02:08 dalek parrot-gmp: currently mpz_add fails
02:09 dalek parrot-gmp: review: https://github.com/bubaflub/​parrot-gmp/commit/ba972a39e0
02:09 dalek parrot-gmp: 9967a90 | bubaflub++ | t/integer/sub.t:
02:09 dalek parrot-gmp: add tests for mpz_sub, mpz_sub_ui, and mpz_ui_sub
02:09 dalek parrot-gmp:
02:09 dalek parrot-gmp: currently mpz_sub fails
02:09 dalek parrot-gmp: review: https://github.com/bubaflub/​parrot-gmp/commit/9967a903cc
02:09 dalek parrot-gmp: 0fac193 | bubaflub++ | t/integer/mul.t:
02:09 dalek parrot-gmp: add tests for mpz_mul, mpz_mul_ui, mpz_mul_si
02:09 dalek parrot-gmp:
02:09 dalek parrot-gmp: currently mpz_mul fails
02:09 dalek parrot-gmp: review: https://github.com/bubaflub/​parrot-gmp/commit/0fac193585
02:09 dalek parrot-gmp: 9debf01 | bubaflub++ | t/integer/addmul.t:
02:09 dalek parrot-gmp: add tests for mpz_addmul and mpz_addmul_ui
02:09 dalek parrot-gmp:
02:09 dalek parrot-gmp: both tests currently fail
02:09 dalek parrot-gmp: review: https://github.com/bubaflub/​parrot-gmp/commit/9debf0183b
02:09 dalek parrot-gmp: 6dbbad2 | bubaflub++ | t/integer/submul.t:
02:09 dalek parrot-gmp: add tests for mpz_submul and mpz_submul_ui
02:09 dalek parrot-gmp:
02:09 dalek parrot-gmp: both of these currently fail
02:09 dalek parrot-gmp: review: https://github.com/bubaflub/​parrot-gmp/commit/6dbbad2fc2
02:09 dalek parrot-gmp: 4a69349 | bubaflub++ | t/integer/mul.t:
02:09 dalek parrot-gmp: add test for mpz_mul_2exp
02:09 dalek parrot-gmp: review: https://github.com/bubaflub/​parrot-gmp/commit/4a69349629
02:09 dalek parrot-gmp: a6f2742 | bubaflub++ | t/integer/neg.t:
02:09 dalek parrot-gmp: tests for mpz_neg
02:09 dalek parrot-gmp: review: https://github.com/bubaflub/​parrot-gmp/commit/a6f2742a38
02:09 dalek parrot-gmp: c7fc315 | bubaflub++ | t/integer/abs.t:
02:09 dalek parrot-gmp: add test for mpz_abs
02:09 dalek parrot-gmp: review: https://github.com/bubaflub/​parrot-gmp/commit/c7fc315872
02:14 ingy joined #parrot
02:20 kid51 left #parrot
02:21 bubaflub msg whiteknight how do i contribute a patch to the Winxed docs that you are hosting on your github pages?
02:21 aloha OK. I'll deliver the message.
02:27 soh_cah_toa bubaflub: you written some more docs for winxed?
02:27 bubaflub soh_cah_toa: i was going to help
02:28 soh_cah_toa bubaflub: that's great. i love the way winxed is transforming :)
02:28 bubaflub soh_cah_toa: i've kinda been putzing along - but the two sections that i could add to are currently blank (installing Winxed and try / catch statements)
02:28 bubaflub soh_cah_toa: yeah, Winxed definitely hides much of the ugliness of PIR.
02:29 soh_cah_toa bubaflub: and lack of documentation was definitely one of the biggest thing that (initially) turned me away from it
02:29 bubaflub soh_cah_toa: same.  i'm glad Whiteknight++ made a huge contribution.  i'm also happy that both Whiteknight and NotFound are usually around so i can pester them with questions.
02:30 soh_cah_toa bubaflub: haha. yeah, those two are great
02:33 bubaflub soh_cah_toa: eventually i'll be using my Winxed class from Rakudo or NQP so i'll have some examples of that to put online
02:33 bubaflub soh_cah_toa: perhaps it'll be straightforward, but either way it's always nice to have something up
02:34 soh_cah_toa bubaflub: oh neat. i didn't realize you could do that
02:34 bubaflub soh_cah_toa: well, i certainly hope i can...
02:34 bubaflub soh_cah_toa: but in all seriousness, it should "just work" - i'll compile my stuff down to a .pbc and pray for the best
02:35 soh_cah_toa bubaflub: hmm...well yeah, you're right. it all gets compiled down to pbc anyway so i don't see how it couldn't
02:48 theory joined #parrot
02:55 nopaste left #parrot
03:02 nopaste joined #parrot
03:06 davidfetter left #parrot
03:19 theory left #parrot
03:20 lateau joined #parrot
03:38 JimmyZ joined #parrot
03:39 JimmyZ dukeleto: hello, dukeleto , your blog 'Real-Time Embedded Parrots' seems broken, at least in planet, both perl6 and parrot
03:59 hudnix left #parrot
04:01 bubaflub left #parrot
04:18 lateau left #parrot
04:20 cotto joined #parrot
04:21 cotto ohai
04:22 cotto left #parrot
04:22 cotto joined #parrot
04:23 dalek parrot/soh-cah-toa/hbdb: 74c37a3 | soh_cah_toa++ | src/hbdb.c:
04:23 dalek parrot/soh-cah-toa/hbdb: Defined hbdb_command_line() as a starting point for the CLI
04:23 dalek parrot/soh-cah-toa/hbdb: review: https://github.com/parrot/parrot/commit/74c37a3663
04:23 dalek parrot/soh-cah-toa/hbdb: 16f90db | soh_cah_toa++ | src/hbdb.c:
04:23 dalek parrot/soh-cah-toa/hbdb: Made a few minor changes to the perldoc
04:23 dalek parrot/soh-cah-toa/hbdb: review: https://github.com/parrot/parrot/commit/16f90db32f
04:23 dalek parrot/soh-cah-toa/hbdb: 5edf4b7 | soh_cah_toa++ | / (2 files):
04:23 dalek parrot/soh-cah-toa/hbdb: Defined hbdb_destroy() to free memory when the debugger exits
04:23 dalek parrot/soh-cah-toa/hbdb: review: https://github.com/parrot/parrot/commit/5edf4b718e
04:23 dalek parrot/soh-cah-toa/hbdb: 80ff3d4 | soh_cah_toa++ | src/embed/hbdb.c:
04:23 dalek parrot/soh-cah-toa/hbdb: Corrected the perldoc for Parrot_api_runloop() which reflected its old name
04:23 dalek parrot/soh-cah-toa/hbdb: review: https://github.com/parrot/parrot/commit/80ff3d4a7c
04:28 soh_cah_toa cotto: is there a string function to convert a STRING * to a native char *?
04:29 soh_cah_toa i'm looking through src/string/api.c now but maybe i'm not looking in the right place or something
04:29 cotto Parrot_str_to_cstring
04:30 soh_cah_toa got it
04:30 cotto If you're just messing around you can use ->strstart,  but I'll revert it if I see it in code that gets pushed. ;)
04:30 soh_cah_toa what's strstart?
04:31 cotto a member of the parrot_string_t struct
04:32 cotto If you're poking around with gdb, it's easier to use that.
04:33 soh_cah_toa eh...i probably won't need it
04:34 cotto that's what you think now
04:36 soh_cah_toa actually, whatever strstart is for, it looks like it's been deprecated anyway. according to pdd28, that is
04:37 cotto "deprecated" can mean a lot of things
04:37 cotto Don't depend on it though except possibly for debugging.
04:39 soh_cah_toa debugging a debugger. i really hope i don't ever have to do that
04:41 cotto profiling a profiler is fun too
04:42 * soh_cah_toa shudders
04:46 cjh left #parrot
04:46 cjh joined #parrot
04:54 dalek parrot/soh-cah-toa/hbdb: 0713faf | soh_cah_toa++ | src/hbdb.c:
04:54 dalek parrot/soh-cah-toa/hbdb: Made some further refinements to hbdb_get_command()
04:54 dalek parrot/soh-cah-toa/hbdb: review: https://github.com/parrot/parrot/commit/0713faf4d3
04:58 soh_cah_toa it's quitin' time
04:58 soh_cah_toa left #parrot
05:04 JimmyZ left #parrot
05:25 cotto dukeleto, ping
05:25 cotto seen dukeleto
05:25 aloha dukeleto was last seen in #parrot 10 hours ago saying "But I *would* like to finish my grant to I can move on to hacking on more interesting stuff.".
05:26 contingencyplan left #parrot
05:39 cotto msg soh_cah_toa Don't forget to add hbdb_destroy to Parrot_runcore_hbdb_init.
05:39 aloha OK. I'll deliver the message.
05:48 cotto msg soh_cah_toa For any internal purpose, you should be using STRING*.  Is there any reason for hbdb_t to use char* instead of STRING*?  (note that "parrot_debugger does it" isn't necessarily a good reason)
05:48 aloha OK. I'll deliver the message.
05:51 fperrad joined #parrot
06:01 ingy left #parrot
06:01 ingy joined #parrot
06:13 atrodo left #parrot
06:32 cotto msg dukeleto I updated https://gist.github.com/1019986 with a list of the Perl FFI modules I could find and why they're likely not a good choice for the M0 prototype.  I may try to go ahead with FFI, but it may also end up being necessary to put off ffi until the C implementation.
06:33 aloha OK. I'll deliver the message.
06:45 SHODAN joined #parrot
07:14 mj41 joined #parrot
07:15 atrodo joined #parrot
07:30 mikehh All tests PASS (pre/post-config, make corevm/make coretest, make world/make test, fulltest) at 3_4_0-270-g6e4dc98
07:30 mikehh Kubuntu 11.04 amd64 (g++)
07:41 mikehh left #parrot
08:01 mikehh joined #parrot
08:44 lbr joined #parrot
09:12 contingencyplan joined #parrot
09:54 ambs joined #parrot
10:03 woosley left #parrot
10:32 lichtkind joined #parrot
11:24 dalek nqp: 00a7fe4 | jonathan++ | src/HLL/Actions.pm:
11:24 dalek nqp: Fix broken error reporting in string_to_int (daxim++).
11:24 dalek nqp: review: https://github.com/perl6/nqp/commit/00a7fe4a2c
11:28 lbr left #parrot
11:35 moritz http://www.pathscale.com/ekop​ath4-open-source-announcement - anybody want to test parrot on top of it?
11:38 dalek left #parrot
11:38 d4l3k_ joined #parrot
11:41 d4l3k_ is now known as dalek
11:41 daxim joined #parrot
11:41 woosley joined #parrot
11:42 daxim two website bugs (and I don't want to sign up for parrot trac):
11:42 daxim 1. on http://parrot.org/download =~ s/Mandirva/Mandriva/
11:44 daxim 2. on <http://parrot.org/download/Parrot_RPMs>, replace the last sentence with a link to <http://s.opensu.se/obs/parrot>.  this is a search result across distributions, not just the latest
11:50 ambs left #parrot
11:59 janus left #parrot
12:31 Coke daxim: fixed #1.
12:33 Coke daxim;for #2, do you mean, replace the link in the sentence?
12:34 daxim no, replace the whole sentence.  why restrict to Factory only?  it makes no sense.  people already know which distro they're running and can pick the appropriate package with the highest version number
12:35 Coke daxim; ok. remove the last sentence, replace with a link to that URL ... what's the text for the link?
12:36 daxim Parrot packages on <acronym title="Open Build Service">OBS</acronym>
12:38 hudnix joined #parrot
12:39 Coke ok. done with very small modification to your suggestion. Thanks!
12:39 daxim thank you, Coke++
12:42 bubaflub joined #parrot
12:45 whiteknight joined #parrot
12:48 mtk joined #parrot
12:48 whiteknight good morning, #parrot
12:48 atrodo =~
12:48 whiteknight msg bubaflub: Create a fork of Rosella on github and make your changes. I can pull them in to my repo from there
12:48 aloha OK. I'll deliver the message.
12:49 bubaflub whiteknight: ok.
12:54 whiteknight bubaflub: I really need to do some updates to that myself. I've been lazy with it
12:56 bubaflub whiteknight: i figure i can pitch in with what i know; pull requests will be pretty easy to merge with the big ole button
12:57 whiteknight yes
12:57 whiteknight I might have to merge manually since it's not on the master branch, but it's still easy to do
14:01 whiteknight http://mendel.soup.io/post/138357929/​If-a-programming-language-was-a-boat
14:19 estrabd joined #parrot
14:20 plobsing left #parrot
14:25 bluescreen joined #parrot
14:34 dmalcolm joined #parrot
14:37 bluescreen left #parrot
14:51 mtk left #parrot
14:52 bluescreen joined #parrot
14:56 lucian joined #parrot
14:56 mtk joined #parrot
15:01 dalek nqp: ecc33ba | jonathan++ | src/pmc/sixmodelobject.pmc:
15:01 dalek nqp: Ensure error messages for 'method not found' and 'vtable not implemented' include the class name, not just SixModelObject.
15:01 dalek nqp: review: https://github.com/perl6/nqp/commit/ecc33baa7a
15:02 SHODAN left #parrot
15:16 jsut joined #parrot
15:16 jsut_ left #parrot
15:30 [hudnix] joined #parrot
15:32 hudnix left #parrot
15:32 lucian left #parrot
15:46 theory joined #parrot
15:57 ambs joined #parrot
16:03 cotto_work ~~
16:05 whiteknight hello cotto_work
16:11 woosley left #parrot
16:13 cotto_work hio whiteknight
16:26 bluescreen left #parrot
16:26 bluescreen joined #parrot
16:36 davidfetter joined #parrot
16:41 whiteknight slow day today. Where are all the busy students at today?
16:42 benabik Busy?
16:42 benabik Good morning, BTW.  :-D
16:42 moritz having exams?
16:44 cotto_work buying expensive cars with all their $$$$ from Google?
16:49 dodathome joined #parrot
16:53 klavs joined #parrot
17:02 dalek dm0: b5c886d | klavs++ | MAKEFILE:
17:02 dalek dm0: first commit
17:02 dalek dm0: review: https://github.com/klavs/dm0/commit/b5c886dc47
17:03 benabik Gah.  If anyone can spare a set of eyes, this commit is causing parrot-nqp to go into infinite loops compiling JSON and I can't figure out why.  https://github.com/Benabik/parrot/commit/31b289f
17:04 dalek dm0: b220796 | klavs++ | item.c:
17:04 dalek dm0: first commit
17:04 dalek dm0: review: https://github.com/klavs/dm0/commit/b2207968ad
17:05 mj41 left #parrot
17:08 benabik Hm.  Okay, I think the issue is that it doesn't compile itself correctly.  make->bootstrap->make works, but after another bootstrap make fails.
17:26 whiteknight oh fun
17:27 moritz benabik: can you break up the commit in independent parts, and test them separately?
17:31 whiteknight benabik: how do you know it's an infinite loop?
17:31 whiteknight Do you eventually get to a recursion depth exceeded exception?
17:33 daxim left #parrot
17:36 dmalcolm left #parrot
17:40 benabik whiteknight: No I don't, it just keeps running and eating memory.
17:42 whiteknight benabik: Insert some debugging statements in there, and run it. See if you can find the loop that way
17:42 whiteknight I don't see anything obvious on that diff
17:53 mj41 joined #parrot
17:58 darbelo joined #parrot
18:02 bubaflub left #parrot
18:03 benabik moritz: Hm.  Narrowing it down to the single most complex change makes it work.  One of the small changes is the problem.  Thanks for the suggestion.
18:04 moritz benabik: you're welcome
18:06 whiteknight benabik: post the new diff, with just the small change?
18:07 benabik https://github.com/Benabik/parrot/commit/1559ed4
18:07 benabik whiteknight: ^^ that one works.
18:09 whiteknight oh, so the changes in is_vivipost?
18:09 benabik whiteknight: It might be one of the register changes in as_vivipost or vivify.
18:09 whiteknight ok
18:11 benabik I had thought it was the more complicated change in lexical.  Bad assumptions make debugging harer.
18:11 benabik *harder
18:16 klavs left #parrot
18:16 mj41 left #parrot
18:21 bubaflub joined #parrot
18:35 benabik Survey says: as_vivipost.  Huh.
18:36 atrodo i demand a recount
18:37 whiteknight thats...weird
18:37 benabik whiteknight: No kidding.
18:41 cotto_work seen klavs
18:41 aloha klavs was last seen in #parrot 24 mins 45 seconds ago leaving the channel.
18:45 cotto_work seen dukeleto
18:45 aloha dukeleto was last seen in #parrot 23 hours 20 mins ago saying "But I *would* like to finish my grant to I can move on to hacking on more interesting stuff.".
18:46 cotto_work seen cotto_work
18:46 aloha cotto_work was last seen in #parrot 0 seconds ago saying "seen cotto_work".
18:46 cotto_work seen aloha
18:46 aloha aloha was last seen in #perl6 3 hours 25 mins ago joining the channel.
18:46 benabik The offending change:
18:46 benabik https://github.com/Benabik/parrot/commit/e03dbb9
18:47 benabik That's supposed to be the same as the first hunk of this commit: https://github.com/parrot/parrot/commit/efc90318
18:47 dmalcolm joined #parrot
18:50 bubaflub whiteknight: can you explain the difference between 'p' and 'P' NCI types?
18:50 whiteknight one is a raw C-level pointer, I think, and the other is a PMC
18:51 whiteknight I think P is the PMC, while p is VTABLE_get_pointer(interp, pmc)
18:51 bubaflub whiteknight: ok.  i have some interesting failures on my test suite, and even on segfault.  i think what i need to do is either switch to P or implement get_pointer vtable on my Winxed class to return the var ptr that my class wraps.
18:51 bubaflub whiteknight: i'm leaning towards the second option.
18:52 whiteknight that probably is best, if that's what p really does
18:52 whiteknight I'm not entirely sure
18:52 bubaflub whiteknight: i hope so, i'm looking at the draft of pdd16.
18:52 whiteknight take a look at src/nci/extra_thunks.nci. I think there are some explanatory comments in there
18:52 bubaflub whiteknight: also, what does it take to get a PDD out of draft and into the main doc area?
18:52 bubaflub whiteknight: also, this draft pdd still has the 't' type
18:52 whiteknight bubaflub: sacrafice a goat. Some latin words I can't remember
18:53 bubaflub whiteknight: sounds like we need bruce campbell
18:53 whiteknight okay, the PDD is probably way out of date
18:53 bubaflub seen plobsing
18:53 aloha plobsing was last seen in #parrot 21 hours 23 mins ago joining the channel.
18:54 atrodo I agree.  We need more Cowbell, err, Bruce Campbell
18:56 bubaflub whiteknight: well, i have an interest in solidifying all of this NCI stuff and getting good docs for people to follow.  i'll bring it up at the next #ps
19:01 benabik It's bad to get distracted when testing for a bug that eats all available memory.  Computer takes a while to get going again.
19:16 preflex left #parrot
19:17 moritz benabik: ulimit exists
19:18 benabik moritz: That would be smart.
19:19 preflex joined #parrot
19:21 bluescreen left #parrot
19:22 benabik Ah.  This isn't good: `find_lex $P1756, "@arglist" ; find_lex $P1756, "$inline"`
19:22 NotFound It's like people not being able to remember to take his memory improving pills.
19:26 bubaflub left #parrot
19:28 skids joined #parrot
19:31 skids dukeleto: Yes!
19:36 skids dukeleto: Unfortunately "Yes" is all the feedback I can offer on a blogpost title that just hyperlinks back to  planet.parrotcode.org :-)
19:37 skids left #parrot
19:38 NotFound A post about recursivity by example?
19:40 whiteknight for more details, click <a href="#">here</a>
19:42 NotFound http://www.arrakis.es/~ninsesabe/recursi/index.htm
19:42 NotFound This is a toy I made years ago.
19:43 benabik NotFound: It makes a pretty pattern.
19:44 NotFound Unfortunately, some brwosers give up about 10 levels.
19:49 tadzik I clicked so long the link is not visible anymore
19:51 cotto_work aloha: recursion?
19:51 aloha cotto_work: Dunno.
19:51 cotto_work aloha: recursion is see recursion helper
19:51 aloha cotto_work: Okay.
19:51 cotto_work aloha: recursion helper is see recursion helper
19:51 aloha cotto_work: Okay.
19:52 cotto_work NotFound++
19:52 benabik I don't recall which book, but one of my favorite index entries is: "Infinite recursion, see: Recursion, infinite.  Recursion, infinite, see: Infinite recursion"
19:56 davidfetter left #parrot
19:57 atrodo cotto_work> reminds me of a joke in Monkey Island 2
20:01 NotFound Also classic: To understand recursion you must first understand recursion.
20:01 tadzik this joke never gets old
20:28 whiteknight left #parrot
20:32 bubaflub joined #parrot
20:35 davidfetter joined #parrot
20:43 soh_cah_toa joined #parrot
20:48 mj41 joined #parrot
20:51 dodathome left #parrot
21:09 dalek parrot/soh-cah-toa/hbdb: 228b605 | soh_cah_toa++ | src/hbdb.c:
21:09 dalek parrot/soh-cah-toa/hbdb: Added stub to echo commands after they're entered
21:09 dalek parrot/soh-cah-toa/hbdb: review: https://github.com/parrot/parrot/commit/228b605cda
21:12 soh_cah_toa cotto_work: ping
21:18 ambs left #parrot
21:27 cotto_work soh_cah_toa: poing
21:29 [hudnix] left #parrot
21:30 dalek parrot/soh-cah-toa/hbdb: a3d1e87 | soh_cah_toa++ | frontend/hbdb/main.c:
21:30 dalek parrot/soh-cah-toa/hbdb: Commented out call to Parrot_api_hbdb_get_command() to test if new control flow works
21:30 dalek parrot/soh-cah-toa/hbdb: review: https://github.com/parrot/parrot/commit/a3d1e87807
21:31 soh_cah_toa cotto_work: i'm having a problem and i have no idea what's causing it. if run hbdb w/ a .pbc file (i usually use parrot-prove.pbc) and you get "ERROR: "load_bytecode" couldn't find file 'Getopt/Obj.pbc'"
21:32 cotto_work soh_cah_toa: how did you run Configure.pl?
21:32 soh_cah_toa cotto_work: what do you mean? just "perl Configure.pl"
21:36 cotto_work soh_cah_toa: ok.  That indicates that Parrot doesn't know where to look for libraries.
21:38 janus joined #parrot
21:39 soh_cah_toa cotto_work: alright, what does that mean for me though?
21:41 PurityControl joined #parrot
21:41 cotto_work soh_cah_toa: there might be some search path code that hbdb needs to initialize
21:42 soh_cah_toa cotto_work: search path code?
21:42 cotto_work soh_cah_toa: I'm not sure what the right fix is without digging for a bit.
21:43 soh_cah_toa alright
21:43 cotto_work I'd look at what the parrot frontend does.  src/library.c is where the code that initializes paths lives.
21:44 cotto_work parrot_init_library_paths
21:44 soh_cah_toa alright, i'll take a look. if not, i'll just live w/ it for now and try to find something else to work on
21:45 cotto_work ok
21:45 cotto_work I'll have more free time than usual tonight.
21:45 soh_cah_toa ok
21:46 Patterner left #parrot
21:46 PurityControl is now known as Patterner
21:52 soh_cah_toa that's odd. i thought i removed the cwd from the search path months ago yet the code is still here in src/library.c. eh, whatever...
21:56 sorear you removed it, but only in a branch, which nobody ever bothered to merge
21:56 soh_cah_toa sorear: oh yeah, i was working on a fork. that's right
21:56 sorear this is fairly normal for bugfixes
21:57 soh_cah_toa cotto_work: i think it's kinda weird that i (as in hbdb) have to perform search path initialization. isn't that libparrot's job?
22:01 cotto_work soh_cah_toa: that's just my guess.  For now, I need to focus mostly on $dayjob.
22:09 dmalcolm left #parrot
22:13 soh_cah_toa yeah, i know. that's fine
22:19 darbelo left #parrot
22:23 mj41 left #parrot
22:24 whiteknight joined #parrot
22:26 cotto_work whiteknight might know
22:27 whiteknight that's not something people say very often
22:27 soh_cah_toa whiteknight knows everything
22:27 whiteknight ...and that's a first on that sentence
22:27 soh_cah_toa :)
22:31 soh_cah_toa let's see if that's true...ever seen anything about "couldn't find file Getopt/Obj.pbc"? i've got some serious breakage over here
22:33 fperrad left #parrot
22:54 whiteknight soh_cah_toa: where are you seeing that error?
22:54 whiteknight I have seen that before, winxed maybe
22:54 soh_cah_toa hbdb
22:58 whiteknight when I've seen that error before, it's because I had an old installed version of Parrot around
22:58 whiteknight like, I had a parrot 3.3.0 installed, but a parrot 3.4.0 binary
22:58 whiteknight so it was searching for /lib/whatever/parrot/3.4.0/*, but I had a 3.3.0
22:58 soh_cah_toa ehhh...i get that A LOT
22:59 whiteknight uninstall everything. realclean, reconfig, make, install, etc
22:59 soh_cah_toa will do
23:06 bubaflub left #parrot
23:09 Tene joined #parrot
23:10 kid51 joined #parrot
23:23 cotto_work soh_cah_toa: did you see my aloha-o-grams from last night?
23:25 soh_cah_toa cotto_work: yeah, i tried to add hbdb_destroy to the interp's destroy field but it caused the build to fail. something completely unrelated as usual. something about miniparrot
23:25 bubaflub joined #parrot
23:27 cotto_work orly?
23:27 soh_cah_toa yeah, it's stupid
23:31 cotto_work src/embed/hbdb.c:73: warning: passing argument 2 of �hbdb_get_line_number� from incompatible pointer type
23:31 soh_cah_toa eh, i'm not even using that function
23:31 cotto_work If you see one of those, it's very likely bug
23:32 soh_cah_toa how did you get that? it builds fine for me
23:32 cotto_work If you'll never use it, delete it.  If you think (or know) you might, fix it.
23:33 cotto_work I just happened to notice it.  This machine is slow enough that I can occasionally see build warnings.
23:33 cotto_work ;]
23:33 soh_cah_toa damn, it's still borked even after running a clean install. "make realclean," "perl Configure.pl," "make," "make test," "make install," the whole works
23:33 soh_cah_toa oh yeah, and now some tests fail. makes no sense, i don't even have any hbdb tests yet
23:33 * soh_cah_toa bangs head on desk
23:34 cotto_work make test runs all tests
23:35 cotto_work or most of them
23:35 soh_cah_toa yeah, it usually passes
23:35 soh_cah_toa this is the first time i've ever had "FAILED"
23:36 whiteknight it's a bummer seeing that, isn't it?
23:36 whiteknight I thought about changing it to something less depressing
23:36 whiteknight "SPECIAL"
23:36 whiteknight "HAS POTENTIAL"
23:37 soh_cah_toa ha!
23:38 cjh left #parrot
23:42 dalek parrot/soh-cah-toa/hbdb: 5766617 | soh_cah_toa++ | src/runcore/cores.c:
23:42 dalek parrot/soh-cah-toa/hbdb: Moved call to Parrot_pcc_set_pc() to inside the while loop because that's what cotto said ;)
23:42 dalek parrot/soh-cah-toa/hbdb: review: https://github.com/parrot/parrot/commit/5766617d86
23:42 dalek parrot/soh-cah-toa/hbdb: 2178eb0 | soh_cah_toa++ | src/ (2 files):
23:42 dalek parrot/soh-cah-toa/hbdb: Removed hbdb_get_line_number() and its API wrapper because thanks to IMCC, I'll probably never use it
23:42 dalek parrot/soh-cah-toa/hbdb: review: https://github.com/parrot/parrot/commit/2178eb0ebe
23:44 dalek parrot/soh-cah-toa/hbdb: 9ae8e60 | soh_cah_toa++ | include/parrot/ (2 files):
23:44 dalek parrot/soh-cah-toa/hbdb: As usual, I forgot to run headerizer for the previous commit
23:44 dalek parrot/soh-cah-toa/hbdb: review: https://github.com/parrot/parrot/commit/9ae8e6030e
23:46 cotto left #parrot
23:46 elmex left #parrot
23:46 athomason left #parrot
23:46 TimToady left #parrot
23:46 GeJ left #parrot
23:46 PerlJam left #parrot
23:46 aloha left #parrot
23:46 autark left #parrot
23:46 bubaflub left #parrot
23:46 kid51 left #parrot
23:46 Patterner left #parrot
23:46 preflex left #parrot
23:46 lichtkind left #parrot
23:46 contingencyplan left #parrot
23:46 pmichaud left #parrot
23:46 dod left #parrot
23:46 simcop2387 left #parrot
23:46 ascent left #parrot
23:46 KaeseEs left #parrot
23:46 frodwith left #parrot
23:46 arnsholt left #parrot
23:46 TiMBuS left #parrot
23:46 baest left #parrot
23:46 tcurtis left #parrot
23:46 NotFound left #parrot
23:46 rblackwe left #parrot
23:46 allison left #parrot
23:46 Maddingue left #parrot
23:46 soh_cah_toa what the heck?
23:48 mikehh soh_cah_toa: happens every so often, they come back together again
23:48 pjcj left #parrot
23:49 soh_cah_toa that's odd...
23:50 mikehh see Netsplit on Wikipedia
23:52 soh_cah_toa interesting. can't believe i've never heard of that before
23:53 mikehh caugt me out a time or two, until I enquired (a while back though)
23:54 mikehh caught
23:55 soh_cah_toa hmm...doesn't really say what causes it though. it looks like its usually either the result of some fatal error in one of the servers or an attack
23:56 sorear soh_cah_toa: network fluctuations
23:56 cotto_work network glitches are bound to happen
23:56 mtk left #parrot
23:56 sorear soh_cah_toa: server/server connections are done using the same protocols as server/client connections
23:57 sorear and sometimes servers ping out
23:57 soh_cah_toa yeah
23:58 cjh joined #parrot
23:58 bubaflub joined #parrot
23:58 kid51 joined #parrot
23:58 Patterner joined #parrot
23:58 preflex joined #parrot
23:58 lichtkind joined #parrot
23:58 contingencyplan joined #parrot
23:58 cotto joined #parrot
23:58 elmex joined #parrot
23:58 athomason joined #parrot
23:58 pmichaud joined #parrot
23:58 TimToady joined #parrot
23:58 GeJ joined #parrot
23:58 dod joined #parrot
23:58 simcop2387 joined #parrot
23:58 PerlJam joined #parrot
23:58 aloha joined #parrot
23:58 ascent joined #parrot
23:58 KaeseEs joined #parrot
23:58 frodwith joined #parrot
23:58 arnsholt joined #parrot
23:58 TiMBuS joined #parrot
23:58 autark joined #parrot
23:58 baest joined #parrot
23:58 tcurtis joined #parrot
23:58 allison joined #parrot
23:58 NotFound joined #parrot
23:58 rblackwe joined #parrot
23:58 Maddingue joined #parrot
23:58 mikehh there you go ...
23:59 soh_cah_toa ah, all better

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

Parrot | source cross referenced