Camelia, the Perl 6 bug

IRC log for #parrot, 2011-07-17

Parrot | source cross referenced

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

All times shown according to UTC.

Time Nick Message
00:11 rurban_ joined #parrot
00:15 rurban left #parrot
00:15 rurban_ is now known as rurban
00:27 Ademan approximately how many opcodes does parrot have implemented? and do certain projects (say pynie) use all of them or is it likely that a given project only uses a small subset?
00:27 Ademan (since it looks like there a bunch of sets of opcodes)
00:28 cotto Ademan, <1100
00:28 dalek parrot: 743b6b8 | cotto++ | docs/project/release_manager_guide.pod:
00:28 dalek parrot: add coke++ as the 3.7.0 release manager
00:28 dalek parrot: review: https://github.com/parrot/parrot/commit/743b6b8baf
00:28 Coke what, did I forget to push? bah.
00:28 cotto you'll probably needs a subset of them
00:28 Coke ah, got distracted.
00:29 cotto you can add more via dynops
00:29 dukeleto Ademan: 1088 according to src/ops/core_ops.c
00:30 dukeleto Ademan: there are also dynamically loadable ops too
00:46 jmueller joined #parrot
00:53 dafrito left #parrot
01:03 whiteknight joined #parrot
01:19 bubaflub good evening whiteknight
01:20 whiteknight hello bubaflub
01:20 bubaflub whiteknight: how goes it?
01:21 whiteknight well enough. I'm wiped tonight because I woke up too early and had a very busy day. I'm trying to work up the motivation to rub some code together
01:22 whiteknight you?
01:22 bubaflub busy day as well - getting ready to move.  trying to write some docs and do a blog post.
01:22 bubaflub wife is busy playing castle crashers so that's a bit distracting
01:23 whiteknight heh. We're still not sure where we are moving. We are only sure that it's happening in two weeks
01:23 whiteknight There are two problems with buying a house
01:23 whiteknight 1) The asshole sellers
01:23 whiteknight 2) The asshole realtor
01:23 whiteknight besides those things, it wouldn't be so bad
01:24 bubaflub we're in a similar situation - we have to move by next friday but aren't sure we'll have our new place in time
01:24 sorear Does seling houses involve asshole buyers?
01:24 whiteknight We're trying to buy a house, and we agreed on a price after a lot of negotiation. The sellers insist the house is worth like 10k more than it is
01:25 whiteknight and then BANG, the apprasal came back and proved that the house is worth less than the sellers were willing to go. Now we can't get a mortgage for the agreed price
01:25 whiteknight mortgage company won't go higher than the appraisal
01:26 whiteknight so we have to either get the appraisal re-done to get it up to the agreed price, or we have to get the sellers to agree to come down
01:26 whiteknight and our #&@#$!@ realtor hasn't answered his phone in 3 days
01:26 whiteknight So I may have to go over to his house eventually and perform an invasive medical procedure with a rake
01:27 whiteknight anyway, if it isn't obvious, this is very stressful and I'll take any opportunity to vent about it :)
01:29 whiteknight how are you doing?
01:31 bubaflub pretty great compared to the nonsense you have to deal with
01:31 whiteknight maybe I'm blowing it out of proportion. I might be one of those asshole buyers that sorear mentioned
01:32 Felipe hehe
01:33 whiteknight Hello Felipe
01:34 Felipe Hello Andrew
01:34 cotto ~~
01:38 whiteknight hello cotto
01:38 whiteknight What Ubuntu package do I need to install in order to build Parrot with zlib support?
01:38 jsut_ joined #parrot
01:39 cotto possibly zlib1g-dev
01:39 whiteknight libz1g-dev?
01:39 whiteknight oh, okay
01:39 cotto where do we use that?
01:39 whiteknight looks like Configure picked that up
01:40 whiteknight We use it in the tar dynpmc, or whatever. The one that distutils uses to upload smolder reports
01:40 sorear whiteknight: money brings out the worst in people
01:41 whiteknight sorear: yeah
01:41 cotto sorear, not with gsoc students, so far. ;)
01:41 whiteknight of course, having no money would bring out the worst in my wife
01:43 jsut left #parrot
01:46 whiteknight parrot passes all tests here. Plumage is passing tests. Rosella passing all tests. PLA passing all tests
01:46 whiteknight all a good sign before a supported release
01:47 cotto whiteknight, on windows?
01:47 whiteknight no, linux. I only have windows at work
01:47 Felipe someone uses windows here?
01:48 cotto whiteknight, same here.
01:48 cotto Felipe, not enough people.
01:50 cotto highly uninformative article: http://developerloremipsum.com/
01:51 whiteknight :)
01:51 Felipe haha
01:52 whiteknight On windows, Parrot is going to link against a vanilla msvcrt.dll, I assume. At least, I hope it does
01:52 whiteknight if so, I think I can fix the nciutils.t failure there
01:52 whiteknight first try to GetModuleHandle("libparrot"), and if that bears no fruit, do GetModuleHandle("msvcrt")
01:53 whiteknight Maybe we want a configure probe to figure out which version of libc is being loaded on windows. There are potentially many of them
01:56 whiteknight Anyway, nothing I can do about it until monday. For now, bed.
01:56 Felipe good night
01:57 cotto 'night
01:57 whiteknight left #parrot
02:07 dalek parrot/soh-cah-toa/hbdb: 5542137 | soh_cah_toa++ | / (2 files):
02:07 dalek parrot/soh-cah-toa/hbdb: Defined free_file() to free memory allocated for storing source file. Also defined hbdb_label_t since it's necessary for free_file() and I will need it soon anyway for the "disassemble" command.
02:07 dalek parrot/soh-cah-toa/hbdb: review: https://github.com/parrot/parrot/commit/5542137a00
02:07 dalek parrot/soh-cah-toa/hbdb: f239702 | soh_cah_toa++ | / (2 files):
02:07 dalek parrot/soh-cah-toa/hbdb: Added "disassemble" command (kinda sorta). This is only the beginning. I want to work on something else but don't want to abandon this just yet.
02:07 dalek parrot/soh-cah-toa/hbdb: review: https://github.com/parrot/parrot/commit/f239702168
02:07 dalek parrot/soh-cah-toa/hbdb: 9a2cd46 | soh_cah_toa++ | / (3 files):
02:07 dalek parrot/soh-cah-toa/hbdb: Commands successfully repeat when nothing is entered. Renamed HBDB_ENTERED to HBDB_STARTED since the meaning of the later is more clear. Snuck in "step" command by accident.
02:07 dalek parrot/soh-cah-toa/hbdb: review: https://github.com/parrot/parrot/commit/9a2cd467a8
02:18 GodFather left #parrot
02:21 jmueller left #parrot
02:22 kid51 joined #parrot
02:56 dalek parrot-gmp: de29766 | bubaflub++ | docs/tutorial.md:
02:56 dalek parrot-gmp: added tutorial docs
02:56 dalek parrot-gmp: review: https://github.com/bubaflub/​parrot-gmp/commit/de29766d36
02:56 jmueller joined #parrot
02:56 bubaflub jay - latest commit (de29766) on parrot-gmp has some basic docs about generating NCI thunks
03:03 dalek parrot/soh-cah-toa/hbdb: efc7ebe | soh_cah_toa++ | / (2 files):
03:03 dalek parrot/soh-cah-toa/hbdb: CLI now starts regardless of whether a file was given on the command-line or not.
03:03 dalek parrot/soh-cah-toa/hbdb: review: https://github.com/parrot/parrot/commit/efc7ebe4a9
03:03 dalek parrot/soh-cah-toa/hbdb: 760d290 | soh_cah_toa++ | / (2 files):
03:03 dalek parrot/soh-cah-toa/hbdb: Moved definition of welcome() to src/hbdb.c and called it from hbdb_runloop(). This way the banner is displayed right before the runloop is entered.
03:03 dalek parrot/soh-cah-toa/hbdb: review: https://github.com/parrot/parrot/commit/760d290873
03:03 dalek parrot/soh-cah-toa/hbdb: 4580000 | soh_cah_toa++ | / (2 files):
03:03 dalek parrot/soh-cah-toa/hbdb: Removed hbdb_t variable in frontend/hbdb/main.c since the frontend should not have access to this because the debugger isn't even started yet anyway. As such, INTERP_ATTR becomes useless so I removed that as well.
03:03 dalek parrot/soh-cah-toa/hbdb: review: https://github.com/parrot/parrot/commit/4580000569
03:06 dalek website: bubaflub++ | Parrot-GMP: NCI Thunk generation complete
03:06 dalek website: http://www.parrot.org/content/parro​t-gmp-nci-thunk-generation-complete
03:08 jay congrats, bubaflub!
03:11 dalek NQR: 15160ce | (Jay Emerson)++ | / (6 files):
03:11 dalek NQR: Fixed up while and not-quite-R for, tightened rep()
03:11 dalek NQR: review: https://github.com/NQRCore/NQR/commit/15160ceb13
03:14 bubaflub left #parrot
03:21 soh_cah_toa jay: ping
03:25 awwaiid left #parrot
03:28 jmueller left #parrot
03:38 kid51 left #parrot
03:39 cotto this should make release managers happy
03:39 dalek parrot: 1f90d57 | cotto++ | tools/release/update_version.pl:
03:39 dalek parrot: add more laziness to update_version.pl
03:39 dalek parrot: review: https://github.com/parrot/parrot/commit/1f90d57fa5
03:39 dalek parrot: e0acdf6 | cotto++ | / (2 files):
03:39 dalek parrot: update release manager guide for better laziness, remove debugging print statements
03:39 dalek parrot: review: https://github.com/parrot/parrot/commit/e0acdf6b7b
03:40 cotto that script wasn't nearly lazy enough
03:45 cotto msg kid51 I updated your update_version script and the release manager guide.  It should be a bit nicer now, but please review the change to make sure it's clear what you're supposed to do.
03:45 aloha OK. I'll deliver the message.
03:46 dukeleto cotto++ # making future release managers happy
03:50 cotto dukeleto, that's bugged me every time I've been on the hook.
03:54 jay soh_cah_toa: pong
03:55 soh_cah_toa jay: i sent you a message on github about nqr. did you see it?
03:56 jay Yes... already changed it, thanks!
03:56 soh_cah_toa ok good
03:57 jay Slow and steady... making progress.  Did some benchmarks, though, that weren't terrific...
03:59 soh_cah_toa i think nqr is doing great so far
04:00 jay Thanks.  I still feel like I'm juggling too many languages...
04:01 cotto dukeleto, we're at 42 M0 ops now.  I'm tempted to keep it there.
04:01 jay I actually appreciate := now, because then I *know* it's NQP.
04:01 soh_cah_toa yeah, i can understand that
04:11 preflex left #parrot
04:13 daniel-s joined #parrot
04:15 preflex joined #parrot
04:20 jay Just posted... http://www.parrot.org/conten​t/funny-behavior-squaak-loop
04:21 jay Bedtime for me...  if you know the answers to either question, please leave a comment on the blog.
04:21 jay left #parrot
04:24 dalek website: jayemerson++ | Funny behavior in a squaak 'for' loop
04:24 dalek website: http://www.parrot.org/conten​t/funny-behavior-squaak-loop
04:38 daniel-s left #parrot
04:58 dafrito joined #parrot
04:59 soh_cah_toa left #parrot
05:18 benabik joined #parrot
05:22 Coke jay - that looks like a binding issue - every value in a got the same version of i. I bet if you unbox the i, it'll give you the answer you expect (the array index implicitly unboxes, I bet.)
05:31 theory left #parrot
06:24 fperrad joined #parrot
07:40 dalek parrot: 173b634 | fperrad++ | t/harness.pir:
07:40 dalek parrot: [harness.pir] add Winxed
07:40 dalek parrot: review: https://github.com/parrot/parrot/commit/173b6344a5
07:42 preflex left #parrot
07:44 preflex joined #parrot
07:52 daniel-s joined #parrot
08:09 jsut joined #parrot
08:09 jsut_ left #parrot
08:13 rurban__ joined #parrot
08:14 dukeleto interesting: http://www.oscon.com/oscon2011​/public/schedule/detail/20848
08:16 rurban left #parrot
08:16 rurban__ is now known as rurban
09:03 Eclesia joined #parrot
09:03 Eclesia hi
09:26 mj41 joined #parrot
09:31 dalek parrot-libgit2: 04ac7bd | dukeleto++ | / (3 files):
09:31 dalek parrot-libgit2: Remove the pbc step of the Makefile, because setup.winxed does it correctly and load the correcty bytecode. Passing tests!
09:31 dalek parrot-libgit2: review: https://github.com/letolabs/pa​rrot-libgit2/commit/04ac7bd7f4
09:31 dalek parrot-libgit2: a4dd73b | dukeleto++ | / (4 files):
09:31 dalek parrot-libgit2: Add a test for creating a Git2.Repository object in winxed
09:31 dalek parrot-libgit2: review: https://github.com/letolabs/pa​rrot-libgit2/commit/a4dd73b6d1
09:45 dalek parrot-libgit2: 486796a | dukeleto++ | / (2 files):
09:45 dalek parrot-libgit2: Add a failing test for calling git_repository_open in Winxed
09:45 dalek parrot-libgit2: review: https://github.com/letolabs/pa​rrot-libgit2/commit/486796a017
09:46 rurban left #parrot
09:47 rurban joined #parrot
10:16 whiteknight joined #parrot
10:17 mj41 left #parrot
10:23 mj41 joined #parrot
10:35 redicaps joined #parrot
10:37 dalek rakudo/nom: 2ad1cba | moritz++ | src/core/Any.pm:
10:37 dalek rakudo/nom: add &sort sub; only works without a comparison block for now
10:37 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/2ad1cba315
10:47 JimmyZ joined #parrot
10:55 whiteknight good morning, #parrot
10:55 Eclesia hi whiteknight
10:56 whiteknight good morning, Eclesia
10:57 mj41 left #parrot
11:03 user_8212 joined #parrot
11:03 user_8212 left #parrot
11:12 preflex left #parrot
11:15 preflex joined #parrot
11:20 dalek parrot/whiteknight/pbc_pbc: 081df99 | Whiteknight++ | src/pmc/packfile.pmc:
11:20 dalek parrot/whiteknight/pbc_pbc: Add a .view method to Packfile PMC, to return a PackfileView. This is the form most useful for execution
11:20 dalek parrot/whiteknight/pbc_pbc: review: https://github.com/parrot/parrot/commit/081df99307
11:21 whiteknight msg benabik I just added a .view() method to Packfile PMC in the whiteknight/pbc_pbc branch. Take a look. I'll try to get that branch merged shortly after 3.6 if all goes well
11:21 aloha OK. I'll deliver the message.
11:28 mj41 joined #parrot
11:36 redicaps left #parrot
11:43 Felipe good morning
11:58 whiteknight hello Felipe
11:59 jay joined #parrot
12:02 Felipe hello
12:06 jay morning Felipe
12:14 kid51 joined #parrot
12:17 jmueller joined #parrot
12:30 dalek NQR: 4a4f899 | (Jay Emerson)++ | / (2 files):
12:30 dalek NQR: Updated progress and plans
12:30 dalek NQR: review: https://github.com/NQRCore/NQR/commit/4a4f89946f
12:31 janus left #parrot
12:31 janus joined #parrot
12:36 Eclesia whiteknight: quick question : how can I get the values which are declared as .annotation ?
12:46 JimmyZ left #parrot
12:49 autark joined #parrot
12:59 Eclesia left #parrot
13:06 ambs joined #parrot
13:07 fperrad left #parrot
13:09 fperrad joined #parrot
13:13 redicaps joined #parrot
13:21 Felipe function addition should be requested (with patch) as a RFC ticket?
13:21 kid51 Yes.
13:21 Felipe okay
13:22 kid51 Felipe:  We have a quarterly support release this Tuesday, which means that in two weekends, we have our quarterly online Parrot Developers Summit.
13:23 kid51 That's the place where we set our roadmap goals for the next supported release, and form teams to work on those goals.
13:23 kid51 You will probably enjoy attending that.
13:23 Felipe nice
13:23 kid51 At this week's #parrotsketch (Tuesday 2030 UTC), we will begin discussing that summit.
13:29 dalek nqp-rx: 232e5cf | pmichaud++ | README:
13:29 dalek nqp-rx: Update NQP -> NQP-rx .
13:29 dalek nqp-rx: review: https://github.com/perl6/nqp-rx/commit/232e5cf2ba
13:36 mj41 left #parrot
13:42 lucian joined #parrot
13:43 whiteknight Eclesia: Here's a code example for you:https://github.com/Whiteknight/Rosella/​blob/master/src/core/Parrot.winxed#L34
13:43 whiteknight Eclesia: That code takes an exception, pulls the Context information from it, and then looks up annotations in the context
13:44 whiteknight Eclesia: Look at the get_backtrace_strings() function for how to get current annotations without an exception
13:46 redicaps left #parrot
13:48 JimmyZ joined #parrot
13:49 woosley joined #parrot
13:49 moritz http://pastebin.com/CDadwnaH a fellow trying to build parrot on macos - any idea what's wrong?
13:51 jay http://www.parrot.org/conten​t/funny-behavior-squaak-loop : Updated to provide better context, explaining why it wasn't a post to parrot-dev.
13:52 jay motirz++ Once again, thank you for your help yesterday on the :pirop<set__iQi> thingy and getting going with the custom infix operators.  Huge progress yesterday because of that help.
13:54 jay < moritz++ > sorry, wrong place for a typo.  Need coffee.
13:58 pmichaud_ jay/moritz:  The difference between uppercase and lowercase signature letters is whether the opcode can accept a constant or requires a register
13:58 pmichaud_ thus    opcode__vsii   would accept string and integer constant operands, while   opcode__vSII would first convert each operand into a register before invoking the opcode.
13:59 pmichaud_ (some Parrot opcodes will accept constant arguments, some will not)
14:02 whiteknight I wasn't aware of any opcodes which will not accept a constant
14:03 moritz anything that modifies stuff
14:03 moritz upcase__ss wouldn't make sense
14:04 moritz or setattribute with a constant object
14:10 pmichaud_ there are some opcodes that take in operands that don't accept a constant
14:10 pmichaud_ I just ran into one this last week.. just a sec
14:11 kid51 afk
14:13 dalek rakudo/nom: cf0da7d | moritz++ | src/core/Str.pm:
14:13 dalek rakudo/nom: make +Str die with an unhelpful error message, instead of segfaulting
14:13 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/cf0da7d5fb
14:20 pmichaud_ "escape" is one   (but I don't think that's the one I ran into)
14:20 whiteknight hmmm
14:23 pmichaud_ actually, that may be the one I ran into.
14:23 pmichaud_ some of the comparison opcodes require registers as well.
14:28 jay < back... copying these comments >  thanks, pmichaud++
14:30 dalek parrot/whiteknight/pbc_pbc: 44bd07b | Whiteknight++ | / (2 files):
14:30 dalek parrot/whiteknight/pbc_pbc: PackFile_Header_validate doesn't appear to be used from anywhere, so is made static
14:30 dalek parrot/whiteknight/pbc_pbc: review: https://github.com/parrot/parrot/commit/44bd07b655
14:30 dalek parrot/whiteknight/pbc_pbc: c4b06f5 | Whiteknight++ | / (2 files):
14:30 dalek parrot/whiteknight/pbc_pbc: Fix some headerizer warnings. Add deprecation notices and suggested upgrade paths for various functions. Misc Cleanups
14:30 dalek parrot/whiteknight/pbc_pbc: review: https://github.com/parrot/parrot/commit/c4b06f53cb
14:30 dalek parrot/whiteknight/pbc_pbc: bdebe72 | Whiteknight++ | src/pmc/packfile.pmc:
14:30 dalek parrot/whiteknight/pbc_pbc: nit
14:30 dalek parrot/whiteknight/pbc_pbc: review: https://github.com/parrot/parrot/commit/bdebe724f2
14:30 dalek parrot/whiteknight/pbc_pbc: 24b1e91 | Whiteknight++ | / (4 files):
14:30 dalek parrot/whiteknight/pbc_pbc: track down remaining uses of PackFile_fixup_subs. Replace it where possible, add TODO notes otherwise
14:30 dalek parrot/whiteknight/pbc_pbc: review: https://github.com/parrot/parrot/commit/24b1e91236
14:30 dalek parrot/whiteknight/pbc_pbc: b86c00a | Whiteknight++ | compilers/imcc/pbc.c:
14:30 dalek parrot/whiteknight/pbc_pbc: Cut out some code which appears to not be needed. Will comment about it more on TT #1324
14:30 dalek parrot/whiteknight/pbc_pbc: review: https://github.com/parrot/parrot/commit/b86c00ae46
14:46 dalek TT #2154 created by Felipe++: Add mem_sys_strndup() function
14:46 dalek TT #2154: http://trac.parrot.org/parrot/ticket/2154
14:50 whiteknight left #parrot
14:54 jay pmichaud: to clarify your point above, writing a custom infix<+> in NQP, I have
14:54 jay pir::add__III($a[0], $b[$i]);
14:55 jay where a and b are ResizableIntegerArray objects.  Capital I or lower-case i: does it matter, and if so, how?
14:56 kid51 left #parrot
14:59 plobsing joined #parrot
15:01 pmichaud_ jay: in this case it won't matter, since neither $a[0] nor $b[$i] are constants.
15:01 mj41 joined #parrot
15:01 pmichaud_ where it might make a difference is if you write something like:
15:01 pmichaud_ pir::add__III($a, 3)
15:02 pmichaud_ in this case, PAST will first store the 3 into an I register before invoking the add opcode
15:02 pmichaud_ whereas
15:02 pmichaud_ pir::add__Iii($a, 3)
15:03 pmichaud_ would generate an add opcode that has a constant '3' directly in the opcode.
15:03 jay pmichaud: ok.  Subtle.  BOth would work?  The second would be faster I assume?
15:04 pmichaud_ in the case of add__Iii, the second would be faster, yes.
15:04 pmichaud_ Using constants in opcodes is not always faster.
15:04 pmichaud_ (it's often much slower)
15:05 jay Ok.  I should probably focus on "works" versus "doesn't work" first, then think about performance.  !-)
15:05 pmichaud_ absolutely
15:06 jay Many thanks
15:06 theory joined #parrot
15:16 woosley left #parrot
15:22 mj41 left #parrot
15:23 whiteknight joined #parrot
15:24 dalek parrot/whiteknight/pbc_pbc: ebaa1b3 | Whiteknight++ | src/packfile/api.c:
15:24 dalek parrot/whiteknight/pbc_pbc: Reduce nesting, misc cleanups
15:24 dalek parrot/whiteknight/pbc_pbc: review: https://github.com/parrot/parrot/commit/ebaa1b3f25
15:24 dalek parrot/whiteknight/pbc_pbc: f847664 | Whiteknight++ | src/packfile/api.c:
15:24 dalek parrot/whiteknight/pbc_pbc: Remove unnecessary #includes from src/packfile/api.c
15:25 dalek parrot/whiteknight/pbc_pbc: review: https://github.com/parrot/parrot/commit/f847664886
15:25 dalek parrot/whiteknight/pbc_pbc: 51fe42e | Whiteknight++ | src/packfile/api.c:
15:25 dalek parrot/whiteknight/pbc_pbc: fix up some docs, add caveats about Parrot_pf_destroy
15:25 dalek parrot/whiteknight/pbc_pbc: review: https://github.com/parrot/parrot/commit/51fe42e8f7
15:28 Coke left #parrot
15:28 Coke joined #parrot
15:33 dalek parrot/whiteknight/pbc_pbc: 5d3a343 | Whiteknight++ | config/gen/makefiles/root.in:
15:33 dalek parrot/whiteknight/pbc_pbc: Remove unnecessary .h files from dependencies list for packfile/api.c
15:33 dalek parrot/whiteknight/pbc_pbc: review: https://github.com/parrot/parrot/commit/5d3a343b44
15:50 PacoLinux left #parrot
15:51 PacoLinux joined #parrot
15:51 JimmyZ left #parrot
15:58 bluescreen joined #parrot
16:08 daniel-s left #parrot
16:12 theory left #parrot
16:13 rurban_ joined #parrot
16:15 rurban left #parrot
16:16 rurban_ is now known as rurban
16:20 jsut left #parrot
17:35 cotto ~~
17:37 Kulag left #parrot
17:37 Felipe ~~
17:38 tadzik ~~
17:39 fivetonsflax "~~"?
17:39 jay If $arg is Integer or Float, coerce to a string:   pir::set__sN($arg)  ?
17:39 cotto waving hello
17:40 * fivetonsflax thanks cotto
17:41 jay I mean, it seems to work on the one example I tried.  Anything I should be aware of, though?
17:41 cotto jay, in nqp you just prefix ~
17:42 jay I'm sorry, I don't follow.    my $ans := ~$arg;       ? produces a string ?
17:42 Kulag joined #parrot
17:43 moritz yes, ~  converts to string
17:43 jay Very cool.  No performance surprises or anything?  Very convenient!
17:44 jay Wow.  Nice.
17:45 jay cotto++
17:46 Felipe is there an exec() function available for calling in a PIR?
17:48 tadzik I think you'll have to use the OS pmc
17:48 Felipe okay
17:48 Felipe thanks
17:56 lucian hmm http://skillsmatter.com/podcast/​agile-testing/bobs-last-language
17:57 fivetonsflax of course, let's use my favorite.
17:57 fivetonsflax wait, it's not the same as your favorite?  oh well, it was a nice idea.
18:05 dalek parrot-libgit2: 9be35b2 | dukeleto++ | / (3 files):
18:05 dalek parrot-libgit2: Add some tests for creating C strings
18:05 dalek parrot-libgit2: review: https://github.com/letolabs/pa​rrot-libgit2/commit/9be35b20b4
18:05 dukeleto fivetonsflax: ~~ is the "do what I mean operator" and is used to say "hello"
18:05 fivetonsflax unless you mean goodbye, in which case it does that?
18:06 dukeleto fivetonsflax: it means whatever you want it to mean, because it is the DWIM operator :)
18:06 dukeleto fivetonsflax: it also triggers the message bots to privmsg and waiting messages
18:06 dukeleto s/and/any/
18:06 fivetonsflax ~~ ~~ (~~, ~~ ~~).
18:07 fivetonsflax if you see what I mean
18:08 dukeleto lucian: wazzup
18:08 lucian dukeleto: hi
18:13 dukeleto lucian: how is puffin?
18:13 lucian dukeleto: sleeping
18:13 dukeleto lucian: anything I can help with?
18:13 lucian dukeleto: not in particular, i've just been busy working
18:13 lucian from now on, i'll have at least 3 days a week free for puffn
18:14 dalek rakudo: 1c91ecb | (Rakudo Perl)++ | src/ops/perl6.ops:
18:14 dalek rakudo: Merge pull request #26 from atrodo/master
18:14 dalek rakudo:
18:14 dalek rakudo: The new GC needs write barriers for certain ops
18:14 dalek rakudo: review: https://github.com/rakudo/rakudo/commit/1c91ecb6ea
18:14 lucian so i'll wake it up gently
18:15 dukeleto lucian: what things are you working on next?
18:15 lucian dukeleto: improving the compiler first
18:15 lucian then, fill in missing bits of the object system
18:16 Felipe https://gist.github.com/1087886
18:18 dalek NQR: 09973e2 | (Jay Emerson)++ | / (6 files):
18:18 dalek NQR: Added ! prefix operator; created paste(); started testing script
18:18 dalek NQR: review: https://github.com/NQRCore/NQR/commit/09973e2e4b
18:19 jay left #parrot
18:22 dukeleto msg bubaflub any idea what is up here https://github.com/letolab​s/parrot-libgit2/issues/2 ?
18:22 aloha OK. I'll deliver the message.
18:28 dukeleto lucian: what is the simplest next possible improvement to the puffin compiler?
18:29 dalek NQR: fa4148c | (Jay Emerson)++ | STATUS:
18:29 dalek NQR: Updated comments for Sunday on progress# Please enter the commit message for your changes. Lines starting
18:29 dalek NQR: review: https://github.com/NQRCore/NQR/commit/fa4148c532
18:37 lucian dukeleto: implement functions
18:39 lucian dukeleto: (sorry for the delay). next step after that would be scoping
18:43 dalek Rosella: d0f8bb1 | dukeleto++ | README.md:
18:43 dalek Rosella: Add a helpful note to the README about cleaning a recently updated source tree
18:43 dalek Rosella: review: https://github.com/Whiteknig​ht/Rosella/commit/d0f8bb1804
18:44 dukeleto lucian: what is blocking you on implementing functions?
18:44 lucian dukeleto: time
18:44 lucian which i should have more of now
18:45 * dukeleto rejoices
18:46 dalek rakudo: f63b825 | pmichaud++ | build/PARROT_REVISION:
18:46 dalek rakudo: Bump PARROT_REVISION.
18:46 dalek rakudo: review: https://github.com/rakudo/rakudo/commit/f63b825892
19:00 whiteknight lucian: you were approved in the midterm evaluations, by the way. That should mean a new check heading your way
19:02 lucian whiteknight: yeah, i found out
19:08 ambs left #parrot
19:40 Coke left #parrot
19:40 Coke joined #parrot
19:42 benabik left #parrot
19:53 Coke left #parrot
19:53 Coke joined #parrot
19:59 Coke_ joined #parrot
19:59 Coke left #parrot
20:08 Coke_ left #parrot
20:09 Coke joined #parrot
20:15 Coke left #parrot
20:16 Coke joined #parrot
20:17 PacoLinux left #parrot
20:45 dalek Rosella/gh-pages: 4ddaaa8 | dukeleto++ | libraries/test.md:
20:45 dalek Rosella/gh-pages: Add some Winxed examples to the Test library docs
20:45 dalek Rosella/gh-pages: review: https://github.com/Whiteknig​ht/Rosella/commit/4ddaaa8ed1
20:48 whiteknight dukeleto++
20:51 cotto ~~
20:51 awwaiid joined #parrot
20:53 particle joined #parrot
20:53 particle1 left #parrot
20:55 dukeleto whiteknight: i am finding the backtraces from rosella assertions not so helpful
20:56 whiteknight dukeleto: Really? What's wrong with them?
20:56 whiteknight are they less helpful than the stock PIR backtraces?
20:56 dukeleto whiteknight: https://github.com/letolab​s/parrot-libgit2/issues/2
20:57 whiteknight okay, so what's wrong with that?
20:57 dukeleto whiteknight: the winxed part of the backtrace is awesome, but most of the parrot backtrace is missing
20:57 whiteknight oh, is the rest written in PIR?
20:57 whiteknight right now the backtraces use annotation information
20:58 dukeleto whiteknight: there is a bit of pir that dlfunc's all the appropriate functions in libgit2. it was generated by ncidef2pir.pl
20:59 whiteknight okay. So you need backtrace information for that
20:59 whiteknight I'll take a look at it, see what I can do. Parrot's utilities for providing backtraces without annotations is not awesome
21:00 dukeleto whiteknight: well, maybe we just need to make ncidef2pir generate annotations
21:00 whiteknight fixing rosella is probably easier. Give me a few minutes to stare at it
21:00 dukeleto whiteknight: hokey dokey
21:01 dukeleto whiteknight: also, how do dump the contents of an object in Winxed/Rosella ?
21:01 dukeleto whiteknight: for debugging, mostly
21:01 whiteknight dukeleto: Data::Dumper, I guess
21:02 dukeleto whiteknight: i keep forgettig I can use any pbc from winxed. awesomesauce
21:03 whiteknight I may add improved Dumper functionality to Rosella eventually. I haven't hit a level of frustration yet with the parrot-provided one
21:06 dukeleto whiteknight: a dumper that recursed into object would be delicious
21:12 jmueller left #parrot
21:14 soh_cah_toa joined #parrot
21:15 whiteknight hmm...Improving backtraces is going to be a bit ugly
21:15 whiteknight probably going to involve changes to Parrot
21:16 dukeleto whiteknight: like what?
21:17 whiteknight The Continuation.backtrace() method returns an array of hashes. Each hash represents a frame and has two keys: "sub" and "annotations"
21:17 whiteknight I don't know why all frames aren't appearing
21:19 lichtkind joined #parrot
21:20 whiteknight dukeleto: hmm... The backtrace should include frames for all CallContexts in the chain. Are you certain you are missing frames in that backtrace?
21:22 whiteknight in that test, is get_repository_open PMCNULL? A say() op there would be very instructive
21:23 whiteknight er, git_repository_open
21:24 dukeleto whiteknight: when i dump the repo object, it said something like PMC => ['Git2;Repository']
21:24 dukeleto whiteknight: i don't think it is a pmcnull, but I think I am not allocating the internal datastructures correctly
21:25 dukeleto whiteknight: so something deep is going boom
21:25 dukeleto whiteknight: i am not sure i am missing frames, but I am at least missing the line number of the PIR code that is causing that null pmc access
21:29 whiteknight dukeleto: it looks to me like the call to the function git_repository_open itself is null
21:29 whiteknight I suspect winxed cannot find it
21:32 benabik joined #parrot
21:39 perlite_ joined #parrot
21:42 perlite left #parrot
21:42 perlite_ is now known as perlite
21:43 Psyche^ joined #parrot
21:48 Patterner left #parrot
21:48 Psyche^ is now known as Patterner
22:01 dalek parrot-libgit2: 96af518 | dukeleto++ | / (2 files):
22:01 dalek parrot-libgit2: Load common bytecode in Git2.Repository and fiddle with the tests
22:01 dalek parrot-libgit2: review: https://github.com/letolabs/pa​rrot-libgit2/commit/96af5187ad
22:07 mj41 joined #parrot
22:10 lucian left #parrot
22:14 dalek parrot-libgit2: a672915 | dukeleto++ | / (3 files):
22:14 dalek parrot-libgit2: Put all the raw libgit2 functions in the Git::Raw namespace and fix #3
22:14 dalek parrot-libgit2: review: https://github.com/letolabs/pa​rrot-libgit2/commit/a672915f9f
22:14 dukeleto whiteknight: the latest commit changes the error to # invoke() not implemented in class 'Undef'
22:14 dukeleto whiteknight: so i suspect you are correct
22:16 whiteknight dukeleto: weird
22:22 whiteknight is that symbol actually a function in libgit2, or is it a macro?
22:23 jay joined #parrot
22:24 dalek NQR: 0e7a416 | (Jay Emerson)++ | src/nqr/Runtime.pm:
22:24 dalek NQR: Document a few things.
22:24 dalek NQR: review: https://github.com/NQRCore/NQR/commit/0e7a416dc5
22:37 NotFound joined #parrot
22:37 NotFound Hi
22:42 Felipe hey
22:43 kid51 joined #parrot
22:52 whiteknight NotFound: ping
22:53 NotFound whiteknight: pong
22:53 whiteknight NotFound: "Variable 'Winxed_Builtins' is not defined near Winxed_Builtins"
22:54 whiteknight var foo = Winxed_Builtins.pop(array);
22:55 NotFound whiteknight: there is no builtin pop
22:55 whiteknight didn't there used to be? I've been using it all over Rosella
22:55 Felipe whiteknight: what about a stringbuilder pmc to handle the imcc error message?
22:55 whiteknight Felipe: what do you mean?
22:56 NotFound whiteknight: I've checked the compiler just in case. All uses of pop are methods in PMCs.
22:58 kid51 whiteknight: ping
22:58 mj41 left #parrot
22:58 whiteknight kid51: pong
22:58 kid51 whiteknight:  Did you catch back-and-forth re Rakudo on Win32 on parrot-dev today?  Do we need to take any specific action?  Can we?
22:59 whiteknight I didn't see it, no
22:59 whiteknight I'll fix what I can tomorrow, if possible
22:59 whiteknight I don't even know what the breakage is
22:59 kid51 Could you review it please?  pmichaud and Gabor discuss.
23:00 whiteknight ok
23:00 NotFound whiteknight: the reason of the absence is that in push the bulitin can know the type of push desired (even if it really doesn't, passing the argument to pir is enough), but with push is not possible.
23:00 kid51 whiteknight: I think it's more a question of Rakudo not building on Win32, rather than Parrot not building per se
23:00 whiteknight okay
23:00 kid51 Also: This is the *best* Smolder report we've been getting on Win32:  http://smolder.parrot.org/app/​projects/report_details/17989
23:01 kid51 If we can address those 5 fails in 2 files, we'll have at least one Win32 box passing.
23:04 NotFound "Describe in one sentence the purpose of Winxed." ? I just need a word: general.ç
23:06 dalek TT #2155 created by soh_cah_toa++: Additional README Files in Top-Level Directories
23:06 dalek TT #2155: http://trac.parrot.org/parrot/ticket/2155
23:11 dalek Rosella: 2a83f5a | Whiteknight++ | src/include/Core.winxed:
23:11 dalek Rosella: Add some commented-out forward declarations to include/Core.winxed
23:11 dalek Rosella: review: https://github.com/Whiteknig​ht/Rosella/commit/2a83f5a9dd
23:11 dalek Rosella: b29f4c3 | Whiteknight++ | s (5 files):
23:11 dalek Rosella: Add FileSystem.Path to the build. Use it in Harness.TestFile.Factory to properly detect files with a .t extension. Add an experimental try_report function to Core.Parrot, for some standardized error reporting
23:11 dalek Rosella: review: https://github.com/Whiteknig​ht/Rosella/commit/b29f4c35bb
23:11 dalek Rosella: 1cb68f8 | Whiteknight++ | README.md:
23:11 dalek Rosella: Merge branch 'master' of github.com:Whiteknight/Rosella
23:11 dalek Rosella: review: https://github.com/Whiteknig​ht/Rosella/commit/1cb68f861e
23:11 NotFound If a word is not enough, I'll plagiarize Bjarne Stroustrup: "Winxed is a general purpose language designed to make (parrot) programming more enjoyable (than PIR) for the serious programmer"
23:11 whiteknight it's all about the enjoyment
23:11 cotto s/\s//g and it's one word
23:12 soh_cah_toa i think that sums it up well
23:17 NotFound Replied to the list with that description, then.
23:19 kid51 NotFound: http://winxed.org/ currently has no content.  Where did it go?
23:20 NotFound kid51: there is some problem with the server and I haven't contacted the admin yet.
23:20 kid51 k
23:21 NotFound I'll try to redirect it to other site tomorrow.
23:25 NotFound left #parrot
23:25 NotFound joined #parrot
23:35 dalek Rosella: 7d8cddd | Whiteknight++ | src/filesystem/Path.winxed:
23:35 dalek Rosella: Add method to FileSystem.Path to get the complete system path. Cleanups
23:35 dalek Rosella: review: https://github.com/Whiteknig​ht/Rosella/commit/7d8cddda3a
23:35 dalek Rosella: 0001e72 | Whiteknight++ | s (4 files):
23:35 dalek Rosella: Rename Path to FilePath, since it's only for files
23:35 dalek Rosella: review: https://github.com/Whiteknig​ht/Rosella/commit/0001e724b0
23:37 dafrito left #parrot
23:37 whiteknight okay, the test harness shouldn't be picking up vim files anymore
23:38 theory_ joined #parrot
23:38 theory_ is now known as theory
23:50 Drossel joined #parrot
23:51 Kulag left #parrot
23:51 bubaflub joined #parrot
23:51 bubaflub ~
23:52 dalek TT #2156 created by soh_cah_toa++: Relocate Python Scripts in "blib" Directory
23:52 dalek TT #2156: http://trac.parrot.org/parrot/ticket/2156

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

Parrot | source cross referenced