»ö« Welcome to Perl 6! | perl6.org/ | evalbot usage: 'perl6: say 3;' or rakudo:, niecza:, std:, or /msg p6eval perl6: ... | irclog: irc.perl6.org/ | UTF-8 is our friend! | Rakudo Star Released!
Set by diakopter on 6 September 2010.
jnthn sleep & 00:00
00:00 flussence left 00:01 flussence joined 00:02 Fuad left 00:11 mkramer1 joined, [hudnix] joined, ashleyde1 joined 00:12 mkramer left, hudnix left, ashleydev left, araujo joined 00:16 xinming_ left 00:27 zby_home joined 00:32 sftp left 00:46 zby_home left, QinGW joined
dalek tpfwiki: (Herbert Breunung)++ | www.perlfoundation.org/perl6/index....kup_tablet 00:53
00:55 risou left
dalek odel: 40a0b88 | diakopter++ | dotnet/compiler/ (3 files):
[dotnet] tons of codegen api refactoring/cleanup. for generation of efficient code in the regex engine:

NOT, AND, OR, XOR, BAND, BOR, BXOR, and short-circuiting OR and AND. More porting of the ginormous regex prologue (ported from Compiler-Regex.pir)		 01:11
01:12 ashleyde1 is now known as ashleydev 01:14 ch3ck joined
dalek odel: 7f92cdc | diakopter++ | dotnet/compiler/PAST2DNSTCompiler.pm:
[dotnet] remove debug/testing code.. (oops, free karma) diakopter--
 01:16
01:23 __david__ left 01:47 Holy_Cow joined 01:51 bluescreen joined, bluescreen is now known as Guest30561 02:09 redicaps joined 02:10 envi joined 02:20 felipe left 02:29 redicaps left 02:35 ch3ck3r joined 02:39 ch3ck left 02:46 _twitch joined 02:56 kanishka joined 02:59 perigrin left, edenc left, perigrin joined 03:00 edenc joined, Holy_Cow left 03:03 f00li5h left 03:10 whiteknight left 03:18 f00li5h joined 03:28 felipe joined 03:38 QinGW left, QinGW joined 03:39 QinGW left 03:54 kanishka left 04:05 Guest30561 left 04:07 _twitch left 04:09 satyavvd joined 04:23 _twitch joined 04:28 _twitch left 04:33 _twitch joined 04:38 zibri left 04:46 zibri joined 04:50 _twitch left 04:52 simcop2387-lap left 04:53 PZt left 04:56 PZt joined 04:57 nymacro joined 04:58 QinGW joined 05:08 simcop2387 joined 05:09 `26 joined 05:15 am0c left 05:18 zibri left 05:21 kanishka joined 05:22 zibri joined 05:32 frew_ joined 05:39 am0c^ joined 05:44 redicaps joined 06:00 sftp joined 06:04 yahooooo left 06:05 _kaare joined, yahooooo joined 06:07 sftp left 06:11 Patterner left, Psyche^ joined, Psyche^ is now known as Patterner 06:20 yves_ joined 06:31 jhuni joined 06:43 kanishka left 06:45 masak joined
masak morning, #perl6 06:45
tadzik hey masak 06:47
good to see a friendly soul
masak I... fell asleep before November blogging yesterday. seems I needed to catch up on sleep a bit. 06:48
I'll make the best of it and blog a bit now.
tadzik :)
neutro wasn't building for Star? crap 06:51
06:53 wtw joined
tadzik what are star modules built with? ufo? 06:56
masak think so. 07:01
tadzik I thought it works well for bin/ only modules
sorear good * #perl6! 07:02
moritz_ good mornin' 07:03
sorear TimToady: Why does STD use temp $*STUB = ... instead of my $stub = ... ? 07:05
lue HP&TMoR is a great book!
masak Haskell is more easy to pick up after Perl 6. Haskell makes one grateful for certain things in Perl 6. comichunter.net/nowhere/opinions/pe...skell.html 07:07
07:08 frew_ left
lue
.oO(I completely forgot about the timey-wimey that was briefly introduced in Harry Potter)
 07:12
masak lue: how's Pod going? 07:16
Greenpea, I mean. 07:17
07:22 justatheory left
lue Good. Although I already spy a refactor when I want blocks-in-blocks working, and I haven't created one test :/ 07:23
(I can't wait for that House style epiphany) 07:26
07:28 am0c joined
masak lue: I'd like you to meet this idea. lue, red-green-refactor. red-green-refactor, lue. agileinaflash.blogspot.com/2009/02/...actor.html 07:28
lue: TDD has been among the most revolutionary ideas as I've learned to code better in the past few years. it keeps informing the way I approach a coding task. 07:30
lue If I understood correctly, the process is: write tests -> fail tests -> write code -> pass tests -> refactor (change/add/remove code, moar tests, etc.) -> fail tests ... 07:34
masak sounds about right.
and they embody that in the three-step cycle red, green, refactor. 07:35
07:36 am0c^ left
lue hm, I feel temporarily ignoring the fact I've written any code is a Very Good Idea™. It also helps that after all of the interesting errors I talked about were fixed, I'm "bored" and can't figure out what to do next. 07:37
This could be the reason why just about every coding project I've ever started is unfinished!!! O.o [ thank you masak, I knew a House-style epiphany would come! ] 07:38
masak my only remaining quibble with the idea of TDD is that sometimes I don't want to switch contexts between writing tests and writing code. so sometimes I cheat a bit and write lots of tests up front. it spoils the idea a bit, but I must confess it sometimes feels more "right". 07:39
lue: finishing things is hard. however, I find my success rate increases if there are people downstream using the thing. that seems to be the biggest thing determining whether programming projects I start get completed. 07:40
sorear needs to find downstream users 07:41
dalek ecza: 2390300 | sorear++ | lib/SAFE.setting:
Fix build errors
 07:42
masak sorear: downstream users were the success of Pugs and Rakudo. I believe a little implementation like niecza could really soar given a few dedicated downstream users, too. 07:43
07:48 dual left
sorear masak: Who used Pugs? 07:48
07:49 wamba joined
masak sorear: I did, and hundreds of others. 07:50
08:03 palik joined 08:04 Holy_Cow joined, Holy_Cow left
dalek ecza: 34db8bc | sorear++ | src/Optimizer/Simplifier.pm:
Implement optimized code for %h<foo>:exists and :delete
 08:10
lue Wow, I've been having many midnights during my vacation. It's fun though, I wish I could do it more often :) 08:23
08:25 araujo left
sorear lue: when I was your age, I thought the same thing. 08:25
now I'm 20 and I cannot fall asleep any earlier than 3AM
it's like I'm living on Melbourne time ... and I need to take classes in San Diego 08:26
lue I'll be sure to remember that.
masak blog post! I'm 29, and 08:31
blog post! strangelyconsistent.org/blog/novemb...e-innocent
that unfinished phrase was part of "I'm 29, and my sleep isn't describably by a simple rule anymore :/"
s/bly/ble/
but hopefully it'll get better now that I've turned it right to get my early mornings back. 08:32
08:32 EXABTOPN joined
sorear wonders if that nick is trying to be cyrillic 08:37
masak what would 5 be? not to mention T? 08:38
08:40 Woodi left
lue masak: about your blog post, I think you meant obligations instead of onligations (first paragraph after history bit) 08:40
08:40 Woodi joined
masak lue: thank you. lue++. fixing now. 08:41
(maybe "onligations" are an onerous kind of obligations?) :)
lue I wonder if onligations qualify as a malamanteau... 08:46
masak fixed. new, better URL: strangelyconsistent.org/blog/novemb...e-innocent
lue I'll have fun with tests tomorrow! 'night o/ 08:47
masak night, lue. 08:51
Infinoid That's a little odd 09:03
t/00-parrot/07-op-string.t (Wstat: 11 Tests: 12 Failed: 0)
Non-zero wait status: 11
Parse errors: Bad plan. You planned 18 tests but ran 12.
but I can't reproduce it
moritz_ Infinoid: that's the parrot GC going wild 09:04
09:05 timbunce joined
Infinoid sounds painful 09:07
moritz_ trac.parrot.org/parrot/ticket/1871 09:08
Infinoid Thanks
09:12 Woodi left
sorear 11? not 137? 09:12
09:17 masonkramer left 09:19 Woodi joined 09:26 dakkar joined
Tene masak: Heh; I wondered if that might be the case (re: blog post) 09:28
09:35 Woodi left, Woodi joined
masak Tene: I should've suspected something, too. 09:40
double-checking sure would've helped in that instance.
09:42 satyavvd left
Tene I considered asking you, but then I thought "Nah, I'm certain he would have checked that. It wouldn't be worth asking." 09:42
masak Tene: in other words, both of us overestimated me yesterday. :) 09:43
Tene :) 09:44
masak diakopter: do I understand correctly that one of your latest commits contains an optimization turning /foo/ into .index('foo')? nice. 09:45
09:54 tzhs joined 10:01 Woodi left, Woodi joined
jnthn o/ 10:06
diakopter++ # mad hacking 10:08
masak jnthn! \o/
sorear wants love
dalek ecza: ccbb9b8 | sorear++ | / (4 files):
Store contextuals in lexical arrays
 10:09
masak sorear: ♥
moritz_ hugme: hug sorear
hugme hugs sorear
10:19 xinming joined
frettled hugme: hug sorear again 10:20
hugme hugs sorear
masak hugme: hug sorear and blush 10:22
hugme hugs sorear
jnthn I guess hugme only blushes spontaneously. 10:24
masak just like people! :)
10:24 masak left 10:28 jhuni left 10:37 lambdabot joined 10:38 HarryS left 10:43 HarryS joined 10:52 scottp joined, scottp left 10:58 araujo joined 11:00 gimix left 11:07 QinGW left 11:13 satyavvd joined 12:03 satyavvd left 12:07 bluescreen joined 12:08 bluescreen is now known as Guest64114
flussence whoa, something knocked half an hour off my nightly spectest run 12:16
oh, probably the 45 new failures. 12:17
12:22 Fuad joined
Fuad Hello 12:22
12:22 daxim joined 12:25 ilogger2 joined, ChanServ sets mode: +v ilogger2 12:29 orafu joined
Teratogen is there a CPAN for Perl 6? 12:32
jnthn Teratogen: Not as such but see modules.perl6.org/ 12:33
12:35 masak joined
ch3ck3r hi masak \o 12:53
masak o/ 12:54
13:11 whiteknight joined, broquaint joined 13:15 takadonet joined 13:16 nymacro joined 13:19 MayDaniel joined, nadim joined 13:25 Psyche^ joined, sftp joined, Psyche^ is now known as Patterner 13:31 timbunce joined 13:38 MayDaniel left 13:52 alc joined 13:55 broquaint left 13:57 broquaint joined
tadzik masak: happy birthday is today? 14:12
masak tadzik: I'm sorry?
moritz_ thought today was sad turkey day
tadzik masak | blog post! I'm 29 14:13
masak tadzik: that was a mis-paste.
tadzik oh, ok
jnthn It confused me too :)
masak sorry for the confusion.
tadzik does anyone have an idea why didn't neutro build for star?
jnthn "How many birthdays does this guy get a year?"
moritz_ tadzik: how did it not build? 14:14
masak jnthn: no more than two, generally.
tadzik moritz_: pmichaud commitmessaged so
masak jnthn: one regular birthday, and one "rogue" birthday.
tadzik rogue? Like on February 29th? 14:15
masak February 29th is always worth celebrating. 14:16
moritz_ tadzik: then ask pmichaud
masak I mean, it's an entire extra day.
tadzik msg pmichaud how didn't neutro build for Star, what was the reason?
moritz_ ... and I don't get paid for it :-) 14:17
masak tadzik: ENOPURL
moritz_ phenny exists 14:18
tadzik phenny: tell pmichaud how didn't neutro build for Star, what was the reason? 14:19
phenny tadzik: I'll pass that on when pmichaud is around.
tadzik thanks masak 14:20
14:24 Lorn joined 14:38 MindosCheng joined
masak only one comment on the ycombinator thread so far, but a positive one: news.ycombinator.com/item?id=1939968 15:01
MindosCheng interesting article. 15:10
15:15 buu left 15:18 XaRDaX joined, MayDaniel joined 15:31 whiteknight left 15:53 fhelmberger joined 15:55 MayDaniel left
masak a few failures in the spectests. 15:56
"bad plan" in t/spec/S02-literals/quoting-unicode.t -- planned 72 but ran 55.
15:57 masonkramer joined, MayDaniel joined
masak non-zero wait status in t/spec/S12-introspection/methods.t and t/spec/S32-num/exp.t 15:57
moritz_ masak: if you repeat your spectest run, I bet other tests will fail instead 15:59
masak moritz_: ah. known Parrot bug? 16:00
16:02 risou joined
moritz_ masak: yes 16:03
known and reported
masak moritz_++
16:03 rhr joined 16:11 nymacro left 16:17 masak` joined 16:19 masak left, masak` is now known as masak 16:28 MayDaniel left 16:35 rgrau_ joined 16:52 mila__ joined 17:12 risou_ joined 17:13 risou left 17:20 Guest64114 joined, justatheory joined 17:21 alc left 17:33 buu joined 17:35 lichtkind joined, buubot joined
lichtkind karma Herbert Breunung 17:36
aloha Herbert Breunung has karma of 69.
lichtkind aloha: thanks
jnthn: thanks 17:39
17:40 buubot left
takadonet lichtkind: good job man 17:41
lichtkind takadonet: thanks but what you mean especially? 17:42
17:42 buubot joined
takadonet lichtkind: you are doing the perl 6 tablets stuff right? 17:42
lichtkind yes 17:43
but its still in the beginning
but good to see someone reads my blog :)
takadonet: want to participate? 17:45
17:46 buubot left
takadonet lichtkind: writing is not my strong suit. However I do read the blog and use it all the time 17:47
lichtkind takadonet: also the kephra and soterik stuff i wrote a while ago? :) 17:48
takadonet lichtkind: It makes it easier when I'm porting p5 module to p6
lichtkind: kephra? soterik?
lichtkind i ment esoteric
kephra the perl editor , my other big project 17:49
takadonet urls? 17:51
lichtkind kephra.sourceforge.net/site/en/home_news.shtml 17:52
17:57 buubot joined, buubot left
takadonet nice 17:58
no i have not seen it before
18:00 buubot joined
lichtkind takadonet: its nice and stable 18:00
18:02 M_o_C joined 18:03 jummy joined
diakopter masak: to answer your question, not precisely; each literal in a regex pattern compiles to that 18:08
masak oh; even better.
diakopter (no smartness/optimization is going on yet)
18:11 risou joined 18:13 risou_ left, jummy left
diakopter nqpnet: say("zzzzabracadabra"~~/zzzzabra/) 18:16
p6eval nqpnet: OUTPUT«zzzzabra␤»
diakopter nqpnet: say("zzzzabracadabra"~~/kkkzzzzabra/)
p6eval nqpnet: OUTPUT«␤»
diakopter nqp: say("zzzzabracadabra"~~/kkkzzzzabra/) 18:17
p6eval nqp: OUTPUT«␤»
diakopter nqp: say("zzzzabracadabra"~~/zzzzabra/)
p6eval nqp: OUTPUT«zzzzabra␤»
18:24 isBEKaml joined 18:32 Chillance joined
tadzik ~~ 18:33
masak .ACCEPT 18:36
dalek tpfwiki: (Herbert Breunung)++ | www.perlfoundation.org/perl6/index....ble_tablet
diakopter S
masak oh, indeed. 18:39
isBEKaml Perl6 tablets? (Sorry, I haven't been here for some time. Btw, hi folks!)
masak isBEKaml! \o/
isBEKaml masak: How's things? :) 18:40
masak is happy to see so many twitterpeople RT-ing twitter.com/rakudoperl/status/7850505862647808
isBEKaml: things are good. almost done with November :)
tadzik oh hi isBEKaml
masak (the month)
tadzik seen alexm?
aloha Sorry, I haven't seen alexm.
masak seen sense
aloha Sorry, I haven't seen sense.
masak titters
tadzik seen the light?
aloha Sorry, I haven't seen the light.
diakopter hee
isBEKaml tadzik: :) 18:41
aloha: aloha
aloha
masak seen you in ages?
aloha Sorry, I haven't seen you in ages.
isBEKaml oh, it isn't that intelligent to answer back. :|
seen the WHAT?
aloha Sorry, I haven't seen the WHAT.
tadzik alexm was giving my R* presentation in July, and I haven't spoken to him ever since 18:42
isBEKaml masak: so RT stands for re-tweet. unTweeTer here.
masak isBEKaml: oh, yes. sorry. :) 18:43
it's far too easy to soak up ingroup lingo.
isBEKaml so it is. When in Rome....
:)
masak seen the Romans in a while?
aloha Sorry, I haven't seen the Romans in a while.
isBEKaml seen Pompeii? 18:44
aloha Sorry, I haven't seen Pompeii.
isBEKaml wunderbar!
masak aloha: you need to travel more.
diakopter seen my toes in ages?
isBEKaml seen aloha
aloha Sorry, I haven't seen my toes in ages.
aloha was last seen in #parrot 8 days 14 hours ago joining the channel.
masak huh? aloha hasn't seen herself in 8 days?
isBEKaml ejjactly
masak aloha: you're a strange little bot. 18:45
isBEKaml she never answers back too.
diakopter seen the inside of my eyelids ever?
aloha Sorry, I haven't seen the inside of my eyelids ever.
18:47 dual joined
isBEKaml the HN post about haskell and perl6 strikes a good chord. Haskell certainly feels nice to program in, especially after looking at Perl6. :) 18:49
I have been going over haskell tutorials the past couple of weeks. 18:50
The odd thing here is that, I didn't quite think of Perl6 as a functional programming language. It just made more sense. 18:53
Any haskellers here? If so, it'd be good to see some blog post comparing haskell and perl6 (even as it's too early to do that) :) 18:55
masak too early? it's eight o'clock in the evening. 18:56
isBEKaml and it's half past midnight. ;)
(here)
masak: you programmed in haskell? 18:57
masak a bit. 18:58
not nearly enough.
isBEKaml job or fun?
masak fun.
isBEKaml nice. 18:59
masak I felt Haskell to be very mathematical in nature. and I mean that in a nice way. 19:00
edenc haskell is awesome
masak refactoring felt like rewriting math formulas in better ways.
isBEKaml well, yeah. It's like those math function stuff from high school. 19:01
edenc haha, high school? 19:02
isBEKaml well, would elementary school be right? :D
edenc there's a lot more than high school math in haskell
but yeah, it's nice to be able to write the theoretical formulae directly 19:03
straight from the textbook 19:04
no imperative munging
isBEKaml no disagreement here. your last point was what I meant.
edenc isBEKaml: you know that operators in haskell are just functions right?
isBEKaml yes. 19:05
tadzik familiar 19:06
isBEKaml edenc: more than that, I liked how the type specification in functions is just more straightforward.
and, that functions can take only one param at a time. It's basically chained all the way for the rest of the params. 19:07
edenc well yes, it uses the same notation as set theory/functions
isBEKaml set theory was so long ago. I don't remember them at all. :/
19:12 risou left, risou joined
TimToady <sorear> TimToady: Why does STD use temp $*STUB = ... instead of my $stub = ... ? 19:21
probably because STD warns on variables that aren't used 19:22
diakopter ajoy 19:24
19:24 fhelmberger left
frettled pmichaud++ - congrats on getting 2010.11 out the door! 19:26
19:28 `26 joined, `26 left, `26 joined 19:29 rgrau_ left 19:30 yahooooo joined 19:31 rgrau_ joined 19:38 MayDaniel joined 19:39 colomon joined
masak rakudo: say "this sentence no verb".trans( / \s+ / => " " ) 19:46
p6eval rakudo : OUTPUT«this senten e n ver ␤»
masak :(
masak submits rakudobug
colomon I thought you were fixing that? ;)
masak yeah. me too. 19:47
TimToady masak++ is astrobugging
isBEKaml that takes out a cob? ;)
masak TimToady: oh, I wish I had the secret backing of a powerful organization :) 19:48
I'm submitting the ticket, and then I'm *fixing* it.
not least because I either caused it myself or didn't make it go away. 19:49
TimToady bug-driven design
masak that makes it sound better than it actually is... :) 19:50
TimToady cobs are bugs too
masak thought they were birds 19:52
TimToady though I daresay that most English speakers don't know cob === spider even while they use the term "cobwebs"
masak oh!
isBEKaml :) 19:53
TimToady Tolkien of course knew this, which is why Bilbo uses Attercop! as a term of derision to the big spiders
moritz_ is an unEnglish speaker
TimToady I always associated that dialect with Transylvania... 19:54
colomon Attercop versus Attercob?
colomon has not read the Hobbit in 25 years...
moritz_ Tolkien was also from Frankonia, where the distinction between b and p is severly blurred :-) 19:55
TimToady easy to lose voicing on word-final voiced consonants in indo-european
flussence waitaminute...
rakudo: say 'hi';
p6eval rakudo : OUTPUT«hi␤»
flussence where's the version number?!
moritz_ gone with the wind, it seems 19:56
TimToady maybe we need to unwind it
colomon masak++ # usually I just fix it without submitting. :)
TimToady usually I just leave breadcrumbs for masak++ to pick up and submit :) 19:57
masak colomon: consider it a safety net. maybe I'll be hit by a meteor between reporting and fixing it. :)
TimToady++ # you're the witch to my Hans and Gretl
TimToady you're much more likely to die from an asteroid than from a meteor
19:58 MayDaniel left
colomon yes, but if he's hit by an asteroid, probably the rest of us won't be worried about fixing the bug 19:58
isBEKaml cos that's a cursed bug? :D
colomon cos we'll be wondering where Sweden went... 20:02
TimToady wonders how fast we could build nuclear plants if we decided we needed them Right Now, If Not Sooner
sorear good * #perl6 20:04
TimToady perhaps we shouldn't place all our bets on solar
sjohnson Right Now(tm) 20:07
masak the saying "we are stardust" could as well be "we are nuclear waste". 20:08
blog post! strangelyconsistent.org/blog/novemb...th-emerges 20:10
sorear isBEKaml: I know Haskell well
20:11 wallberg joined
isBEKaml sorear: then a blog post? :) 20:11
masak sleep &
20:11 masak left 20:12 XaRDaX left 20:13 XaRDaX joined 20:16 isBEKaml left 20:17 M_o_C left
dalek tpfwiki: (Herbert Breunung)++ | www.perlfoundation.org/perl6/index....ble_tablet 20:19
sorear IIUC there are a fair number of nuclear plants sitting idle due to environmental panic 20:21
interestingly, 80-90% of the *calls* to my LTM engine come from |, but 98% of the *time* is spent from protoregexes 20:24
TimToady that probably makes sense, given | tends to be for low-level decisions, while protos are more for high-level decisions 20:33
sorear I suspect the issue is superlinearity somewhere, and the profile being skewed by the 7000+ node 'term' NFA
hmm
TimToady not trying to discourage you from optimizing protos, of course :) 20:34
are you counting the setup of lexers as part of the 98%? 20:35
sorear yes, but if I didn't it would still be like 97% 20:36
265784 time units for the main protoregex entry point, 48742 for GetProtoregexLexer 20:37
the Mono profiler only reports cumulative inclusive times, which is... interesting
TimToady maybe small lexers are easier to keep in CPU caches at various levels 20:38
20:39 Guest64114 left
TimToady and high-level things like term still have to recurse down through many levels despite knowing their "fate" in advance 20:39
or are you doing those with continuations these days? 20:40
in which case it might be continuation overhead
maybe we need to port NYTProf to that VM :) 20:42
sorear I got rid of fates because they seem to be a pessimization now
[a | b | c] [d | e | f] generates a much bigger NFA with fates enabled 20:43
timbunce was summoned by a magic word
sorear because the tail has to be duplicated
20:43 Guest64114 joined, kthakore joined
TimToady then you're redoing a lot of work 20:43
kthakore hello
sorear also I think I've found my problem
TimToady on recursion, not on seqs
kthakore I can't seem to get a successfull build of rakudo on linux
sorear kthakore: RAM? 20:44
kthakore sorear: not a problem I have 8GB. May I mention my problem first?
sorear You didn't
of course
kthakore PackFile_unpack: This Parrot cannot read bytecode files with version 9.2. 20:45
I get that when I run ./perl6
sorear you have stale files somewhere
kthakore aww ...
sorear rm -rf parrot_install parrot and try again
kthakore ok
tadzik also consider make realclean in parrot dir
kthakore sorear: thanks .. sorry if I seemed rude. You are just really fast.
tadzik oh, there is also a parrot after rm. I try to avoid this, Parrot repo is HUGE 20:46
sorear TimToady: in cgexp:name [part of the Niecza syntax extensions], we have { <-[ ' " ( ) { } \[ \] \s ]>+ }
TimToady: this desugars to [ <![ . . . ]> . ]+, which generates LTM of .+ 20:47
this causes quadratic behavior because .+ has to be matched to the end of the file
TimToady yes, I think the LTMer must be aware of negated character classes more than the desugar would suggest 20:49
sets of characters are the proper domain of LTM, so you can't afford to hide that construct behind a desugar
20:53 Guest64114 left 20:55 takadonet left 20:58 wallberg left
TimToady and re [a | b | c] [d | e | f], that seems like a really good place for a :: to stop LTM explosion rather than an excuse to redo a lot of work in submatches 21:07
21:09 Guest64114 joined
TimToady 'course, as LTM gets faster, that becomes less of an issue, I'll grant 21:09
I'll also grant that an adequate profiler beats intuition most any day 21:11
(the jury still being out on whether your profiler is adequate)
21:12 florz joined
sorear timbunce: How would you react to someone trying to do a 3rd party nytprof.out emitter? 21:17
21:17 masonkramer left 21:18 masonkramer joined
timbunce sorear: I'd be delighted. That's one reason why Nicholas and I worked on factoring the writing code into FileHandle.xs 21:18
TimToady some yaks need more shaving than others... 21:19
21:21 jasonmay joined
timbunce sorear: Ideally, though, the standard format ought to be an sqlite db file. (I almost went crazy trying to 'edit' the internal data structure in the reporting code to implement eval merging.) I figure we'll end up with a data file to slqite converter at some point then the reporting tools would work from the db. 21:21
sorear hmm, I haven't seen FileHandle.xs, I guess it's been too long since I read through this :) 21:23
tadzik where are contextual vars speced?
moritz_ probably S02 or S04 21:24
sorear S02:2825 21:25
timbunce sorear: FileHandle.xs would need refactoring to split out the raw C parts from the perl5 dependencies so you'd end up with a libnytprofwrite.so you could link into other things.
dalek ecza: 85bf01e | sorear++ | src/Niecza/Actions.pm:
Expose complex character classes to LTM
 21:26
21:26 XaRDaX left
tadzik thank you sorear 21:27
sorear timbunce: would be a little tricky since I'm not working in C 21:28
timbunce sorear: ah, forgive my assumption. What's the use-case you're aiming for? 21:29
dalek tpfwiki: (Herbert Breunung)++ | www.perlfoundation.org/perl6/index....ble_tablet 21:30
sorear timbunce: I'm pondering a real Perl6-level profiler, and I would like to reuse somebody's pretty output generator
timbunce sorear: Probably not tooo hard. You'd be at the mercy of file format changes, but they don't happen often. 21:33
dalek tpfwiki: (Herbert Breunung)++ | www.perlfoundation.org/perl6/index....kup_tablet 21:35
sorear took 12 seconds off the niecza SAFE.setting parsing test 21:42
TimToady WSJ headline today: "Shunned Profiling Technology on the Verge of Comeback" 21:45
dalek tpfwiki: (Herbert Breunung)++ | www.perlfoundation.org/perl6/index....ble_tablet 21:46
sorear TimToady: which? 21:54
21:55 Guest64114 left
sorear Lexer::Run now takes a much more reasonable 3699 time units, not ~240000 21:56
21:56 timbunce_ joined
dalek tpfwiki: (Herbert Breunung)++ | www.perlfoundation.org/perl6/index....dex_tablet 21:57
21:58 timbunce left, timbunce_ is now known as timbunce
dalek ecza: 914f55d | sorear++ | lib/Kernel.cs:
Str.Str was accidentally omitted for optimization
 22:07
tpfwiki: (Herbert Breunung)++ | www.perlfoundation.org/perl6/index....ign_tablet 22:19
tpfwiki: (Herbert Breunung)++ | www.perlfoundation.org/perl6/index....ory_tablet
tpfwiki: (Herbert Breunung)++ | www.perlfoundation.org/perl6/index....tro_tablet
tpfwiki: (Herbert Breunung)++ | www.perlfoundation.org/perl6/index....ble_tablet
tpfwiki: (Herbert Breunung)++ | www.perlfoundation.org/perl6/index....dex_tablet
lichtkind sorry reformat necesary stops soon 22:20
lue Is there some sort of reference to the Test module used in all the testing in perl6 stuff (roast at least)? I can learn enough from looking at other tests, but it'd be nice to have some sort of reference in front of me. 22:23
lichtkind TimToady: in s29 is written "each See .pairs() method, above." but there is no pairs above , yet another fossil? 22:24
dalek tpfwiki: (Herbert Breunung)++ | www.perlfoundation.org/perl6/index....nks_tablet
tpfwiki: (Herbert Breunung)++ | www.perlfoundation.org/perl6/index....l_6_tablet
tpfwiki: (Herbert Breunung)++ | www.perlfoundation.org/perl6/index....lta_tablet
tpfwiki: (Herbert Breunung)++ | www.perlfoundation.org/perl6/index....ook_tablet
tpfwiki: (Herbert Breunung)++ | www.perlfoundation.org/perl6/index....oop_tablet
tpfwiki: (Herbert Breunung)++ | www.perlfoundation.org/perl6/index....gex_tablet
tpfwiki: (Herbert Breunung)++ | www.perlfoundation.org/perl6/index....ine_tablet
tpfwiki: (Herbert Breunung)++ | www.perlfoundation.org/perl6/index....ock_tablet
tpfwiki: (Herbert Breunung)++ | www.perlfoundation.org/perl6/index...._io_tablet
tpfwiki: (Herbert Breunung)++ | www.perlfoundation.org/perl6/index....tor_tablet
22:25 lichtkind left 22:27 cotto joined
lue rakudo: grammar A { token B { abcd } }; say "abcd" ~~ A.B 22:34
p6eval rakudo : OUTPUT«Type objects are abstract and have no attributes, but you tried to access &!regex␤ in 'A::B' at line 22:/tmp/204fQIVc1D␤ in main program body at line 22:/tmp/204fQIVc1D␤»
lue rakudo: grammar A { token B { abcd } }; say "abcd" ~~ Regex.new(A.B) 22:35
p6eval rakudo : OUTPUT«Type objects are abstract and have no attributes, but you tried to access &!regex␤ in 'A::B' at line 22:/tmp/pqWA5j8lyy␤ in main program body at line 22:/tmp/pqWA5j8lyy␤»
lue how would you do that?
sorear you wouldn't 22:37
it makes no sense
A.^can('B') has an invocant of type A; it cannot be used on an invocant of type Cursor 22:38
lue how so?
sorear it's like trying to use Str.comb on a FileHandle
lue rakudo: grammar A { token B { abcd } }; say A.parse("abcd", :rule<B>); 22:39
p6eval rakudo : OUTPUT«abcd␤»
sorear please take a step back and explain what you're *actually* trying to do
lue I'm just writing tests. 22:41
22:46 timbunce left 23:13 risou_ joined
sjohnson has anyone ever accidently piped vi something.txt | less ? 23:14
Tene sjohnson: no, but you may be interested in vipe
23:16 risou left
sjohnson i make that mistake all the time 23:17
it hurts me 23:24
23:26 hanekomu joined
sjohnson seems that a decent hex editor doesn't exist, either 23:27
</rant>
23:48 b0m1ro joined 23:50 b0m1ro left 23:57 orafu left
ingy star: $_ = 'hello'; .say 23:59
p6eval star 2010.09: ( no output )

Powered by Cro and the Raku® Programming Language.

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

Thank you!