Parrot 3.6.0 "P�jaros del Caribe" released | parrot.org | Log: irclog.perlgeek.de/parrot/today | #parrotsketch meeting Tuesday 19:30 UTC
Set by moderator on 31 July 2011.
cotto_work It's nice how Apple users can be identified by their cursing of autocorrect. 00:04
benabik I find it mostly useful. 00:05
whiteknight I've seen some horrible and hilarious examples of it 00:18
benabik It's spread from the phone to the desktop now, which is interesting for chatting on a technical IRC channel. 00:20
00:21 PacoLinux__ left 00:31 fperrad left
cotto whiteknight, no response so far 00:34
00:43 rdesfo joined 00:46 silug left 00:47 theory left
whiteknight urg 00:48
I'm primed and ready to help out with his project, I just don't know exactly what he needs
00:56 rdesfo left
dafrito Is running ./installable_luap --target=parse on a 400-line file expected to take ~10 seconds? 00:58
I guess I'm asking if other HLLs are similarly slow, or if I should look inside the HLL to find the bottleneck
dafrito isn't trolling, I promise! 01:01
whiteknight I don't know how the lua parser is written. I don't know what the performance is like for it 01:02
it might be a bottleneck
dafrito whiteknight, okay - thanks anyway 01:05
I know it's a pretty ambiguous question 01:06
01:10 kid51 left
dalek sella/cleanup_core: 908afc6 | Whiteknight++ | / (4 files):
Redo version numbering. Load the versions hash lazily on request. Don't do it on load anymore
 01:11
sella/cleanup_core: 8c41b85 | Whiteknight++ | / (2 files):
Some cleanups to Rosella.winxed, and add lots more test coverage for the functions in it
sella/cleanup_core: f7957e0 | Whiteknight++ | t/core/ (6 files):
Add extra tests for Error and Version, and add in stub test files for the rest of the Rosella core namespaces
cotto whiteknight, soh_cah_toa says he's on vacation and will be back in action tomorrow. 01:18
so that's good
whiteknight okay, I won't sit around waiting for him like a lonely puppy 01:19
cotto lolz
01:20 preflex_ is now known as preflex 01:22 Kulag left, Kulag joined
dalek kudo/nom: 0756691 | Coke++ | t/spectest.data:
run some fudged tests
 01:26
kudo/nom: 57db83c | Coke++ | t/spectest.data:
track failure modes
01:26 dafrito left
01:27 dafrito joined 01:28 Kulag left
dalek sella/gh-pages: a1cd850 | Whiteknight++ | libraries/template.md:
+stub docs for template
 01:32
01:32 whiteknight left, Kulag joined 01:44 woosley joined 01:53 theory joined 01:59 mikehh left 02:03 Kulag left, Kulag joined 02:06 davidfetter joined 02:15 mikehh joined 02:18 Kulag left 02:22 kurahaupo joined 02:24 Kulag joined 02:30 contingencyplan left 02:35 kid51 joined 02:40 silug joined
dalek rrot/soh-cah-toa/hbdb: 23c2ede | jkeenan++ | / (9 files):
[codingstd] Corrections for c_arg_assert, c_parens, copyright, checkisxx, linelength, trailingspace, etc.
 02:40
kid51 msg soh_cah_toa Did many codingstd corrections in your branch. Please check github.com/parrot/parrot/commit/23c2ede0c3 to make sure I didn't screw things up in .c and .h files. 02:41
aloha OK. I'll deliver the message.
02:47 theory left 02:50 theory joined, rdesfo joined
dalek rrot: c2f2543 | jkeenan++ | src/pmc/packfileview.pmc:
[codingstd] Add ASSERT_ARGS to two functions.
 02:55
kid51 msg whiteknight The two functions to which I added ASSERT_ARGS in c2f2543400 still need POD. 02:56
aloha OK. I'll deliver the message.
02:56 kid51 left, rdesfo left 03:00 theory left
dafrito The lua parser takes about 10 seconds to generate a parse tree from a perl6 grammar, but I don't know how to profile it. Any suggestions? 03:06
Running -Rprofiling did get me a hefty chunk of data; I guess learning/using callgrind is the next step? 03:07
03:20 rdesfo joined 03:25 luben left 03:27 rdesfo left 03:34 theory joined 04:07 rdesfo joined 04:08 rdesfo left
cotto dafrito, once you've post-processed it with tools/dev/pprof2cg.pl, you'll in theory get a callgrind-compatible profile. In practice, there may be bugs. 04:11
dafrito cotto, awesome, thanks. pprof2cg worked, so I'll investigate it further tomorrow :) 05:06
05:08 dafrito left 05:21 theory left 05:26 daniel-s joined
dukeleto Anyone ever heard of Gosu? gosu-lang.org/ 06:08
cotto That second-to-last feature is pretty cool. 06:18
looks less annoying than Java 06:19
06:24 contingencyplan joined
benabik dukeleto: I heard about it not long ago from some guy on IRC. 06:24
dukeleto I seem to hear about a new JVM language every week. 06:28
benabik The JVM really isn't that bad to target. 06:29
dukeleto benabik: until Oracle starts charging
benabik dukeleto: If charging comes with a decent framework to target the VM with, I might be okay with that. The strange backward visitor pattern took some getting used to. 06:30
benabik can't remember the name of that library...
Ah. ASM: asm.ow2.org/
dukeleto does anybody have recommendations for a good document describing C++ rules and regulations? All I can find is babysteps... 06:32
benabik The C++ Programming Language by Bjarne Stroustrup. 06:33
06:33 particle1 is now known as particle
benabik Big book. Excellent reference, decent introduction. 06:33
particle benabik++
benabik loves it to death.
sorear dukeleto: the C++ standard itself. let me find a link 06:34
benabik I wonder if there'll be a fourth edition with all the new stuff in 0xB
sorear www.open-std.org/jtc1/sc22/wg21/doc.../n3242.pdf 06:35
benabik sorear++
06:36 fperrad joined
benabik That does look like a draft of 0xB, so you may periodically run across NYI stuff. 06:37
(I call it OxB because it was referred to as C++0x for a long time, in the expectation it would be released last decade, for those who didn't catch it.) 06:38
benabik is apparently chatty late at night.
dukeleto benabik: indeed :) 06:43
sorear++
benabik I also recommend the boost libraries. Good interface to many useful things, some of which became part of 0xB. www.boost.org Used it to multi-thread some brute force searches in my Crypto class. 06:44
dukeleto what is the new_<Foo> syntax called? 06:46
dukeleto promises he won't have too many more questions. Maybe. 06:47
sorear looks like a template instantiation to me
new_ is an ordinary identifier
07:09 mj41 joined 07:15 SHODAN joined
NotFound A shorter book: "Essential C++", Stanley B. Lippman 07:47
07:57 fperrad_ joined 07:58 fperrad left, fperrad_ is now known as fperrad
dalek rrot: 098363a | fperrad++ | runtime/parrot/library/Archive/Zip.pir:
[Archive/Zip] revert some commits about I/O
 08:07
rrot: a39491c | fperrad++ | runtime/parrot/library/Archive/Zip.pir:
[Archive/Zip] some fix, see TT#2171
09:41 woosley left
moritz public service announcement: the server hosting the IRC logs will go down in a few hours for a software upgrade; all its services will be unavailable for some time 09:52
10:35 dafrito joined 11:46 GodFather joined 12:14 contingencyplan left 12:21 daniel-s left 12:24 bluescreen joined 12:40 whiteknight joined
whiteknight good morning, #parrot 12:43
www.cse.shirazu.ac.ir/~azimi/prog88...ograms.pdf
interesting read for people who program in C but don't know anything abou tit
about it
13:04 SHODAN left
NotFound I'll add another recommendation to it: "How ro write large programs in C: don't do it, use C++" ;) 13:07
whiteknight how are you doing today, NotFound? 13:16
13:17 GodFather left, GodFather joined 13:24 kurahaupo left
NotFound whiteknight: good 13:24
whiteknight I looked a little bit yesterday at the using statement, and trying to add a lookup operator, but it was beyond my abilities 13:38
so I need to study the code more
NotFound That parts of the compiler need some improvement. 13:41
whiteknight it's not messy. There's just a lot of code and it takes time to learn 13:45
bubaflub ~ 13:46
NotFound Is messy, the scope lookups are done in several different ways. 13:48
14:10 cotto left 14:25 ilbot2 joined
moderator Parrot 3.6.0 "P�jaros del Caribe" released | parrot.org | Log: irclog.perlgeek.de/parrot/today | #parrotsketch meeting Tuesday 19:30 UTC
14:33 PacoLinux_ joined 14:34 ilbot2 joined
moderator Parrot 3.6.0 "P�jaros del Caribe" released | parrot.org | Log: irclog.perlgeek.de/parrot/today | #parrotsketch meeting Tuesday 19:30 UTC
14:37 Tomc joined
Tomc hey guys, im developing a little functional language 14:37
it uses recursion pretty heavily
once i translate my syntax tree into PIR (or whatever) can parrot do tail call optimization or anything? 14:38
moritz if you tell it to use .tailcall instead of .return, yes 14:39
but you have to do the analysis and reordering (if necessary) yourself
14:43 bluescreen left, bluescreen joined
Tomc oh and one more thing 14:44
can i call C code from PIR?
for example i have a pattern matching function i would like to write in c/c++ 14:45
tadzik it's possible to use C from PIR, yes
14:50 cosimo joined 18:17 ilbot2 joined
moderator Parrot 3.6.0 "P�jaros del Caribe" released | parrot.org | Log: irclog.perlgeek.de/parrot/today | #parrotsketch meeting Tuesday 19:30 UTC
whiteknight Did anybody in here watch the old Dr Katz when it was on tv? 18:19
because that's what my monitor looks like right now
benabik All jittery? 18:21
whiteknight yeah
benabik boo
atrodo That stinks. 18:22
whiteknight i have a really bad adaptor on this beast, and occasionally they come loose 18:23
cotto_work ~~ 18:27
18:37 PacoLinux_ joined
whiteknight var x = 5; 18:39
say((->(y) x + y() ) (-> 4));
with my patch, you can do that in winxed
that's a slightly contrived example, but it does work 18:40
slightly nicer than the current syntax: say((function(y) { return x + y(); })(function(){ return 4; })); 18:41
NotFound Oh, God, the first case of winxed feature lobbying! ;) 18:46
whiteknight I'm playing with it whether it goes in to master or not. I'm having fun 18:48
it's your fault for making winxed so hackable :)
this most recent change was much harder than the last one 18:49
NotFound A few more hacks like that, and you'll have demonstrated that the winxed compiler basis is adaptable to completely different syntax styles. 18:54
18:54 rdesfo joined 18:56 Eclesia joined
Eclesia hi 18:56
atrodo This just in! Winxed has more line noise than perl! Story at 11 18:57
whiteknight oh come on, that's just a contrived example.
I want to put together a Y combinator with that new syntax. *that* will be messy
bubaflub Eclesia: i don't know if you saw it, but it looks like fperrad has made some fixes to Archive/Zip 18:59
Eclesia bubaflub: I'm just back from work, big day. haven't checked anything yet 19:00
19:15 rdesfo joined, rdesfo left
dalek kudo/nom: c52595a | tadzik++ | src/ (2 files):
Fix "is doc()" trait
 19:19
19:23 jay joined 19:26 evhan joined
dalek R: 689f4db | (Jay Emerson)++ | STATUS.pod:
Updated status; posted talk slides
 19:32
20:21 Eclesia left 20:38 contingencyplan joined 20:58 kurahaupo joined 21:05 mj41 joined 21:12 darbelo joined, PacoLinux_ joined 21:20 davidfetter joined 21:44 Psyche^ joined
dalek kudo/nom: 4b19d47 | tadzik++ | src/Perl6/SymbolTable.pm:
[SymbolTable] Add a :nocache option to add_constant
 22:26
kudo/nom: 2e38c54 | tadzik++ | src/Perl6/Actions.pm:
Don't attempt to document class attributes
kudo/nom: 71df76f | tadzik++ | src/ (2 files):
Push declarator blocks to $=POD as a Pod::Block::Declarator objects

The order is messed up unfortunately. Given a class with a method in it, the method declarator is added before the class declarator.
kudo/nom: 4472ab8 | tadzik++ | Pod/To/Text.pm:
Add declarator blocks support to Pod::To::Text
23:17 kid51 joined 23:19 bluescreen joined
dalek kudo/nom: 855106b | jonathan++ | src/core/Parcel.pm:
Fix Parcel.Capture.
 23:21
kudo/nom: 01c7286 | jonathan++ | src/core/Capture.pm:
Turns out that some tests relied on the Parcel.Capture returning Parcel hack, because certain things in Capture were not implemented. This adds .elems, .Str and .gist on Capture.
kudo/nom: 2b00770 | jonathan++ | src/Perl6/Actions.pm:
Apply patch from mls++ that forbids duplicate CATCH and CONTROL blocks, and saves adding a handler to a try block if it already has a CATCH.
kudo/nom: 3ee1e23 | jonathan++ | src/core/Capture.pm:
Some missing 'multi's.
kudo/nom: ef02d7d | jonathan++ | src/core/Capture.pm:
Oops, bit overzealous multi-ing.
 23:34
23:36 PacoLinux__ joined
kid51 Am I correct in thinking that irclog.perlgeek.de has been down most of the day? 23:36
jnthn__ kid51: Yes, it's undergoing hardware maint.
cotto_work so I guess it's a good thing nothing much is happening today. 23:55
kid51 whiles away his life reading jnthn__'s Beijing Perl Workshop talks 23:59

Powered by Cro and the Raku® Programming Language.

Please report any issues / comments / feature requests as an issue on App::Raku::Log.

Thank you!