Camelia, the Perl 6 bug

IRC log for #perl6, 2012-11-13

Perl 6 | Reference Documentation | Rakudo | Niecza | Specs

| Channels | #perl6 index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
00:21 buubot_backup joined #perl6
00:22 DarthGandalf joined #perl6
00:25 cognominal joined #perl6
00:41 buubot_backup joined #perl6
00:44 tokuhiro_ joined #perl6
00:52 anuby joined #perl6
00:52 am0c joined #perl6
01:00 immortal joined #perl6
01:00 immortal joined #perl6
01:03 kurahaupo1 joined #perl6
01:08 hypolin joined #perl6
01:10 benabik joined #perl6
01:14 erkan joined #perl6
01:14 erkan joined #perl6
01:27 erkan joined #perl6
01:48 [Coke] phenny: tell isBEKaml to ping rurban about cygwin parrot - cygwin was his main platform at some point.
01:48 phenny [Coke]: I'll pass that on when isBEKaml is around.
01:48 phenny [Coke]: 11 Nov 10:59Z <isBEKaml> tell [Coke] ok, nvm - I'm just about to give up on pugs+windows. :)
01:53 [Coke] phenny: tell isBEKaml Nevermind. ;)
01:53 phenny [Coke]: I'll pass that on when isBEKaml is around.
01:53 erkan joined #perl6
01:53 erkan joined #perl6
01:53 fgomez joined #perl6
01:56 fgomez_ joined #perl6
02:00 kurahaupo joined #perl6
02:09 ifim joined #perl6
02:22 DarthGandalf joined #perl6
02:24 clkao_ joined #perl6
02:35 diakopter [Coke]: queue up replies 'til done logbacking
02:40 diakopter later transactions may obviate earlier ones
02:45 alienswamp joined #perl6
02:49 fgomez joined #perl6
02:50 dalek perl6-roast-data: 207bbe8 | coke++ | / (3 files):
02:51 dalek perl6-roast-data: today (automated commit)
02:51 dalek perl6-roast-data: review: https://github.com/coke/perl6​-roast-data/commit/207bbe8160
02:51 dalek perl6-roast-data: 906b9f5 | coke++ | / (3 files):
02:51 dalek perl6-roast-data: today (automated commit)
02:51 dalek perl6-roast-data: review: https://github.com/coke/perl6​-roast-data/commit/906b9f53bd
02:51 dalek perl6-roast-data: a98362d | coke++ | / (3 files):
02:51 dalek perl6-roast-data: today (automated commit)
02:51 dalek perl6-roast-data: review: https://github.com/coke/perl6​-roast-data/commit/a98362dbd5
02:53 cognominal joined #perl6
02:55 [Coke] phenny: tell sorear - niecza failed all tests today with "Unhandled exception: System.MissingMethodException: Method not found: 'Niecza.Kernel.CreateHash'."
02:55 phenny [Coke]: I'll pass that on when sorear is around.
02:55 sorear bah, I should fix that someday if I ever figure out how
02:55 phenny sorear: 02:55Z <[Coke]> tell sorear - niecza failed all tests today with "Unhandled exception: System.MissingMethodException: Method not found: 'Niecza.Kernel.CreateHash'."
02:58 [Coke] sorear: https://github.com/coke/perl6-roast-​data/blob/master/niecza_summary.out has a stacktrace if that helps.
02:58 sorear [Coke]: I've seen this problem before, it's a stale compiled file slipping through the cracks
02:58 sorear rm -r ~/.local/share/NieczaModuleCache
03:05 dalek perl6-roast-data: 1198e62 | coke++ | bin/niecza.sh:
03:05 dalek perl6-roast-data: Remove stale file on every build.
03:05 dalek perl6-roast-data: review: https://github.com/coke/perl6​-roast-data/commit/1198e62eca
03:09 [Coke] sorear++ -
03:12 sorear what?  %FOO ||= built_foo; doesn't work in p5?
03:21 doy sorear: sure it does
03:22 sorear doy: Can't modify private hash in logical or assignment (||=) at -e line 1,
03:22 sorear might be because my perl is ancient
03:22 DarthGandalf joined #perl6
03:23 sorear eval: my %FOO; sub bar { 1,2 }; %FOO ||= bar
03:23 buubot_backup sorear: ERROR: Can't modify private hash in logical or assignment (||=) at (eval 20) line 2, at EOF
03:23 sorear eval: $^V
03:23 buubot_backup sorear: bless( {original => "v5.16.0",qv => 1,version => [5,16,0]}, 'version' )
03:24 doy sounds like a bug to me
03:25 doy hmmm
03:25 doy maybe not
03:25 doy well
03:25 doy at the very least, that's a terrible error message
03:26 doy i think the problem is that || supplies scalar context
03:27 doy because you can't evaluate the truth value of a list
03:28 doy and that just causes all kinds of confusion because now you're treating the scalarified temporary version of that hash as an lvalue
03:29 doy or something
03:32 Entonian joined #perl6
03:33 [Coke] sorear: https://gist.github.com/4063778 - current niecza failures
03:43 BaggioKwok joined #perl6
03:44 * colomon has been slacking off on checking niecza spectests
03:57 orafu joined #perl6
04:07 gootle joined #perl6
04:09 isBEKaml joined #perl6
04:11 isBEKaml phenny: tell rurban that I can now build parrot+rakudo successfully on cygwin. it was a line ending issue from msysgit, installing cygwin's git and a fresh checkout resolved them all.
04:11 phenny isBEKaml: I'll pass that on when rurban is around.
04:11 phenny isBEKaml: 01:48Z <[Coke]> tell isBEKaml to ping rurban about cygwin parrot - cygwin was his main platform at some point.
04:11 phenny isBEKaml: 01:53Z <[Coke]> tell isBEKaml Nevermind. ;)
04:18 fgomez joined #perl6
04:20 tokuhiro_ joined #perl6
04:26 DarthGandalf joined #perl6
04:46 cognominal joined #perl6
04:52 marloshouse joined #perl6
05:04 xinming_ joined #perl6
05:17 am0c joined #perl6
05:23 telex joined #perl6
05:28 DarthGandalf joined #perl6
05:33 Psyche^ joined #perl6
05:41 scottp joined #perl6
05:52 erkan joined #perl6
05:52 erkan joined #perl6
05:59 daniel-s joined #perl6
06:08 birdwindupbird joined #perl6
06:19 xinming joined #perl6
06:32 cognominal joined #perl6
06:48 SamuraiJack joined #perl6
06:53 SamuraiJack joined #perl6
07:00 cognominal joined #perl6
07:16 daniel-s_ joined #perl6
07:20 topo joined #perl6
07:26 jlaire joined #perl6
07:35 domidumont joined #perl6
07:56 sorenso joined #perl6
07:56 tokuhiro_ joined #perl6
07:58 moritz \o
08:01 sorear o/
08:01 am0c joined #perl6
08:03 domidumont joined #perl6
08:13 domidumont joined #perl6
08:22 kst joined #perl6
08:24 fhelmberger joined #perl6
08:29 flightrecorder joined #perl6
08:34 FROGGS joined #perl6
08:38 kresike joined #perl6
08:38 kresike hello all you happy perl6 people
08:42 FROGGS morning
08:44 sorear heya
08:44 tadzik hello hello
09:05 domidumont joined #perl6
09:15 domidumont joined #perl6
09:24 dalek niecza: 8864bf6 | sorear++ | lib/ (6 files):
09:24 dalek niecza: Top removal: SubInfo.ctor, STable.ctor
09:24 dalek niecza: review: https://github.com/sorear/niecza/commit/8864bf6117
09:25 xenoterracide joined #perl6
09:29 xenoterracide so today I was thinking I wonder what my ideal language would look like... I mean I'd kinda like ruby except for all of those ugly 'end' statements, (ramble) and then I thought... oh that's right the language I wish I was using looks like Perl 6. So i thought I'd pop in here and say ... keep on truckin'
09:30 tadzik come join the fun :)
09:30 bonsaikitten xenoterracide: so basically python, but more fun? ;)
09:30 sorear welcome to #perl6, xenoterracide
09:32 xenoterracide bonsaikitten: no I think I hate python more... the (ramble) was about python
09:32 bonsaikitten amusing
09:32 tadzik to me python just isn't exciting. It's like writing Java a bit
09:33 tadzik anyway, are you writing some Perl 6 code, xenoterracide?
09:33 xenoterracide about this time it also occurred to me that one of our developers either really likes python... because he's trying to make his closing braces as condensed as possible...
09:33 xenoterracide or he's just being a dick
09:33 xenoterracide either way I hate whitespace based... harder to read imo
09:33 sorear xenoterracide: is it possible he's an elderly LISP programmer?
09:33 xenoterracide tadzik: no
09:34 sorear (foo
09:34 sorear (bar
09:34 sorear (baz)))
09:34 tadzik oh I hate this style :)
09:34 xenoterracide sorear: though I think I saw that he does lisp
09:34 bonsaikitten tadzik: I prefer boring ;) exciting means someone calls me saturday morning at 6am because the pritner is on fire
09:34 xenoterracide so yes
09:34 sorear is standard LISP whitespace style
09:34 xenoterracide a young lisp programmer
09:34 bonsaikitten whitespace is mandatory anyway
09:34 bonsaikitten so why not make it do something useful
09:34 xenoterracide I corrected it with judicious use of perltidy
09:35 dalek niecza: 3db2fe1 | sorear++ | lib/ (3 files):
09:35 dalek niecza: Top removal: inferior runloop system
09:35 dalek niecza: review: https://github.com/sorear/niecza/commit/3db2fe1808
09:35 xenoterracide readability is useful
09:35 xenoterracide unfortunately I have not seen that python accomplished that
09:36 huf hey, python's pretty readable. it's just that almost everything is pretty readable.
09:36 huf aint nothing special
09:36 moritz OTRS source code is not pretty readable.
09:36 huf OTRS?
09:37 xenoterracide OTR's?
09:37 moritz a bug tracker
09:37 sorear I have never used python so I can't say I have an opinion on it
09:37 xenoterracide I've only dabbled
09:37 huf sorear: i can read it about as well as any other language i dont know :D
09:37 moritz it's written in perl 5 with no particularly advanced features
09:37 xenoterracide but all the python I've seen is not very good
09:37 tadzik I'm doing my uni project in python now, it's okay
09:37 huf sorear: not as well as a language i know. surprising :)
09:37 sorear huf: yeah, all ALGOL clones read the same
09:37 moritz but it's a big codebase, and uses pseudo OO designed by people who do't understand OO
09:37 arnsholt Python is OK. I'm used to Perl so it annoys me a bit when I can't do things exactly as I would in Python, but as a language it's pretty good
09:38 moritz and no choice of language would make such code readable
09:38 * tadzik mumbles something about scoping
09:38 sorear huf: how are you with Haskell or the APL family? :p
09:38 arnsholt There are some scoping-related things I don't like, but that's pretty much it
09:38 arnsholt tadzik: Yeah, 's what happens when you don't explicitly declare variables =)
09:38 moritz and I can't get my head around join being a string method, not a list method
09:39 sorear J is the only language I've touched this year that was non-trivial to learn
09:39 eternaleye joined #perl6
09:39 xenoterracide digresses that I haven't used it enough to really make a fair judgment about it, and now I've been perling for a while
09:39 bonsaikitten moritz: OTRS stuff grew over the years- mostly customer-demand-driven
09:39 xenoterracide though I'm fairly frustrated with the py2k py3k thing
09:40 bonsaikitten moritz: and since it works for their internal uses why change it
09:40 sorear xenoterracide: p5 p6 is worse
09:40 xenoterracide sorear: yeah but I'm not sure I think of them as the same "language" anymore
09:40 tadzik I have a feeling that py2k py3k is more like p5.10 p5.12 than p5 p6
09:41 xenoterracide tadzik: yeah, with use strict enabled implicitly and it changed each time
09:41 moritz bonsaikitten: I'm not saying that it should be rewritten; what I wanted to express is that readability is usually more a matter of the architecture of the application, less a matter of the programming language
09:41 tadzik I don't think python takes backwards compat seriously
09:41 tadzik We'll invent Python 3 so we can do backwards-incompatibile changes!
09:42 tadzik And then they roll out 2.7 which breaks 2.6 code
09:42 xenoterracide lol
09:42 bonsaikitten moritz: yes, and it shows the usual lack of discipline of commercial projects
09:42 bonsaikitten tadzik: py3 is academic wankery, about 2/3rds of all packages are still 2.x only
09:42 bonsaikitten it's like declaring that the new version of ruby is perl5, and you better like it!
09:42 tadzik bonsaikitten: my use case is purely academic wanking, so I'm okay with using py3 ;)
09:43 bonsaikitten tadzik: as long as you don't rely on many external libs that's usually ok
09:43 xenoterracide I wanted to use py3, but pyside is 2 only...
09:43 xenoterracide and I didn't want to be gpl
09:43 bonsaikitten still all the "big" libs aren't migrated
09:43 xenoterracide oh... and then I wanted to checkout sql alchemy
09:43 xenoterracide 2 only
09:43 tadzik bonsaikitten: I just need to write an mbox parser from scratch, I'm not likely to use any libs at all
09:44 xenoterracide but the default python on arch is 3?
09:44 bonsaikitten xenoterracide: yes, and you better like it
09:44 tadzik well, arch is... adventurous
09:44 bonsaikitten xenoterracide: they don't care about "working" anymore
09:44 tadzik did they ever?
09:44 bonsaikitten yes
09:44 xenoterracide actually I generally have no problem with working
09:44 xenoterracide outside of this damn network driver
09:44 bonsaikitten that's how they got big, now the devs are just ... wtf ... users are not idiots you can yell at for fun
09:44 xenoterracide which is entirely lkml
09:45 tadzik I think they got big by the "simplicity" marketing, they never had really mature and responsible maintainers
09:45 tadzik not in the last 5(?) years at least
09:46 arnsholt So speaking of Arch, what's a good distro these days?
09:46 tadzik debian's okay I guess
09:46 xenoterracide it's better than what I have to deal with when I've used fedora or ubuntu
09:46 xenoterracide I'd rather arch any day
09:46 tadzik I don't like russian rulette on updates
09:46 arnsholt I suppose I can live with Debian. I never warmed to apt and friends though
09:46 xenoterracide though I haven't reinstalled since they went all manual
09:47 moritz arnsholt: the key is to only use them on the command line :-)
09:47 FROGGS ubuntu LTS is quiet good, but the 12.10 is just too new
09:47 tadzik arnsholt: well, you don't use apt to often, do you :)
09:47 bonsaikitten FROGGS: I disagree
09:47 tadzik you install stuff with it and forget about it, not that you use it every 10 minutes
09:47 arnsholt My last experience with Ubuntu was terrible. For some reason it's the one Linux I've worked with in recent memory that required the most wizardry to get working
09:47 xenoterracide I don't like the idea of doing a painful upgrade every 6 months
09:48 xenoterracide so I stay away from release distro's
09:48 bonsaikitten s/upgrade/reinstall/
09:48 bonsaikitten upgrade only works on "new" unmodified installs, half the time
09:48 tadzik ah, the "rolling release" hype
09:48 bonsaikitten tadzik: it's the only way to avoid stupid reinstalling
09:48 arnsholt tadzik: Before trying apt I was used to portage, which on the whole was a lot simpler to work with
09:48 moritz "trolling release"
09:48 xenoterracide hey it works for me
09:48 arnsholt (Even if compiling stuff from source takes forever)
09:48 tadzik bonsaikitten: I don't remember ever reainstalling a debian installation
09:48 xenoterracide arnsholt: um... arch is not binary
09:48 xenoterracide err..
09:49 xenoterracide arch is* binary
09:49 FROGGS bonsaikitten: I have 12.10 x86_64 on a "brand new" dell laptop (from july or so), I have problems almost every day with Xorg and plymouth for example
09:49 bonsaikitten thanks to "stable enterprisey release" I'm stuck with things like bash 3.2 with custom patches, kernel 2.6.18+idiocy, etc.etc.
09:49 bonsaikitten FROGGS: I've never seen plymouth work properly
09:49 arnsholt xenoterracide: Portage is Gentoo
09:49 tadzik I prefer working on my OS, not working "on my os", if you know what I mean
09:49 xenoterracide arnsholt: I know
09:49 bonsaikitten I call it "segfault generator"
09:49 bonsaikitten arnsholt: and that's an advantage! ;)
09:49 xenoterracide arnsholt: I used it for more than half a decade
09:50 bonsaikitten tadzik: I hate compiling, that's why I use Gentoo
09:50 sorear what is plymouth?
09:50 tadzik I don't see the logic, but I don't think that matters :)
09:50 xenoterracide idk, can't stand "release" and "repository" distro's that's my whole thing... give me a rolling distro with centralized repositories
09:50 bonsaikitten sorear: an attempt at a fancy bootsplash
09:50 tadzik my point is: the distro is my tool to work with, not something I need to work on and take care of
09:51 xenoterracide probably also why I like CPAN
09:51 bonsaikitten tadzik: on debian I *always* end up manually compiling things, which is just tedious, so I upgrade to gentoo
09:51 sorear xenoterracide: can you stand cpanminus?
09:51 tadzik bonsaikitten: never happened to me
09:51 sorear oh, misread
09:51 xenoterracide sorear: that's what I use
09:51 bonsaikitten tadzik: well, their nginx is broken, bash used to be crippled, etc.
09:51 tadzik I don't mind old (kernel|bash|whatever), I'm not a (kernel|bash|whatever) developer
09:52 bonsaikitten I dislike segfaults and scripts not working
09:52 tadzik ok, maybe. I never had any problems
09:52 moritz I hate bashs too old to prog1 |& less
09:52 tadzik that's what everyone says though, about systems they use
09:52 bonsaikitten and when upstream tells me "stop using that old shit, man" then, well, I have to fix things
09:52 tadzik do you happen to be using Debian Stale?
09:52 xenoterracide is there any other?
09:53 bonsaikitten no, the only debianish things I need to coerce into cooperation at the moment are ubuntu, and that's ... augh
09:53 xenoterracide the only thing older than debian is rhel
09:53 bonsaikitten such a severe lack of any idea of QA :(
09:53 xenoterracide and cent
09:53 tadzik xenoterracide: there are 2 other
09:53 tadzik xenoterracide: I'm running testing, and I'm getting 10+ updates a day or so
09:53 xenoterracide sounds like fun
09:53 tadzik there's also unstable for the adventurers
09:54 tadzik labeled as "yes, this may break"
09:54 bonsaikitten too old ;) I'm using KDE 4.9.80 at the moment
09:54 huf debian sid doesnt break all that often actually
09:54 huf once every handful of months at most, in my experience
09:54 xenoterracide oh I'm only running 4.9.3
09:55 xenoterracide yeah I've not had many problems with arch breaking, and usually when they do it they tell me
09:55 sorear 10!
09:55 dalek niecza: a2d37f7 | sorear++ | lib/ (2 files):
09:55 dalek niecza: Misc top removal 2
09:55 dalek niecza: review: https://github.com/sorear/niecza/commit/a2d37f793d
09:55 xenoterracide very rarely have I had a pants down moent
09:55 xenoterracide moment*
09:56 bonsaikitten I have sacrificial machines that are there to catch errors
09:56 bonsaikitten makes life a lot easier :)
09:57 xenoterracide I don't have that, but I have lots of backups
09:57 tadzik I can't stand updates breaking vim
09:57 xenoterracide the worst pants down in the last year was a failing hard disk
09:57 xenoterracide and that's not the OS's fault
09:58 xenoterracide I haven't had updates break vim?
09:58 tadzik I did, back in Arch days
09:58 tadzik but then again, not much point in rambling about it now :)
09:59 sorear I had a hilarious update failure BITD that broke, of all things, df
09:59 bonsaikitten my solution to that is to become distro maintainer of packages I care about ;)
10:00 dakkar joined #perl6
10:00 sorear the free block counts were replaced with pseudo-random int64_t values (yes, signed)
10:00 xenoterracide tadzik: oh was that when they were unfucking there implementation of vim?
10:01 tadzik xenoterracide: No idea. I only cared since it broke me dev environment
10:01 tadzik as a user I don't even want to know what they're doing and why
10:01 xenoterracide I remember when I started using arch vim was pretty fucked as to how they implemented it differently from everyone else
10:01 sorear how does one implement vim differently?
10:01 xenoterracide so I spanked some people until they stopped using an example config as the default
10:02 sorear it's off the shelf software...
10:02 xenoterracide sorear: you use the "example" config as a default system config
10:02 xenoterracide which vim states do not do
10:02 moritz that's a matter of distribution, not implementation
10:03 sorear I *expect* different distributions to have different system defaults for any moderately complex piece of software
10:03 xenoterracide well there's slightly different and crazy does not work how you expect at all different
10:03 xenoterracide like vim from cent to suse to debian is slightly different
10:04 xenoterracide but back then in arch it was so different it confused even experienced vim users
10:04 xenoterracide the trick was stop using a sample config and let vim have its own defaults
10:04 xenoterracide at some point they also switched vi to an actual vi implentation
10:04 xenoterracide no idea why
10:05 xenoterracide so vi isn't vim
10:05 sorear well, most distro software doesn't work at all how I expect period because all the major distros have decided that software should mimic Windows by default
10:05 sorear damn bootsplashes
10:05 xenoterracide sorear: well there's that
10:05 sorear rc.d must always give a process log
10:05 bonsaikitten sorear: you sound like a gentoo user ;)
10:06 xenoterracide I prefer it
10:06 xenoterracide the process log
10:06 sorear bonsaikitten: I have never used gentoo.  my first linux was Debian woody, and it is an improvement on everything I have used since
10:06 xenoterracide though I liked gentoo's process log on pretty background best
10:06 bonsaikitten sorear: hmm, I think you'd complain about it
10:07 bonsaikitten but it doesn't force silly defaults on you - 30MB mem use after boot on a "default" install
10:08 sorear bonsaikitten: a fresh install of Debian woody will boot in 4MB, you need 8 if you want to run any programs after booting
10:08 sorear still that's pretty good for what's available now
10:08 mtymula joined #perl6
10:08 sorear I will have to give it a try next time I'm on Linux
10:08 bonsaikitten sorear: kernel needs about 10MB already
10:09 mtymula HELLo!!
10:09 bonsaikitten 4MB doesn't sounds realistic for anything modern, unless you totally strip it down and customize it to hell
10:09 moritz hello mtymula
10:09 jnthn morning, #perl6
10:09 moritz bonsaikitten: debian woody is anything but modern :-)
10:09 moritz \o jnthn
10:10 mtymula mortiz do you have any opinion about http://perl6maven.com/tutorial/perl6-introduction
10:10 bonsaikitten moritz: I prefer things where I can use upstream bugtrackers
10:10 mtymula i am trying to lern perl6...
10:10 moritz mtymula: it's good
10:10 tadzik hello mtymula
10:11 mtymula ok thanks, any more tutorials that you would recomend?
10:11 tadzik bonsaikitten: back on gentoo days I built a kernel which was below 1.44 MB and covered everything I needed for my laptop :)
10:14 mtymula or some well described projects to learn from?
10:14 mtymula mortiz, tadzik ??
10:18 dakkar joined #perl6
10:18 eternaleye joined #perl6
10:18 xenoterracide joined #perl6
10:18 domidumont joined #perl6
10:18 kresike joined #perl6
10:18 FROGGS joined #perl6
10:18 flightrecorder joined #perl6
10:18 fhelmberger joined #perl6
10:18 kst joined #perl6
10:18 sorenso joined #perl6
10:18 jlaire joined #perl6
10:18 topo joined #perl6
10:18 daniel-s_ joined #perl6
10:18 SamuraiJack joined #perl6
10:18 birdwindupbird joined #perl6
10:18 erkan joined #perl6
10:18 scottp joined #perl6
10:18 Patterner joined #perl6
10:18 DarthGandalf joined #perl6
10:18 telex joined #perl6
10:18 orafu joined #perl6
10:18 benabik joined #perl6
10:18 anuby joined #perl6
10:18 buubot_backup joined #perl6
10:18 ironcamel joined #perl6
10:18 LordVorp joined #perl6
10:18 zb joined #perl6
10:18 revdiablo joined #perl6
10:18 baest joined #perl6
10:18 ruoso joined #perl6
10:18 mtk joined #perl6
10:18 circlepuller joined #perl6
10:18 xillver joined #perl6
10:18 alester joined #perl6
10:18 Gryllida joined #perl6
10:18 bbkr_ joined #perl6
10:18 imarcusthis joined #perl6
10:18 colomon joined #perl6
10:18 lumi_ joined #perl6
10:18 aloha joined #perl6
10:18 aindilis joined #perl6
10:18 bruges joined #perl6
10:18 _jaldhar_ joined #perl6
10:18 sunnavy joined #perl6
10:18 arnsholt joined #perl6
10:18 robinsmidsrod joined #perl6
10:18 lorn joined #perl6
10:18 PerlJam joined #perl6
10:18 bloonix joined #perl6
10:18 [particle]1 joined #perl6
10:18 Woodi joined #perl6
10:18 integral joined #perl6
10:18 HarryS joined #perl6
10:18 eiro joined #perl6
10:18 Gothmog_ joined #perl6
10:18 bbkr joined #perl6
10:18 sivoais joined #perl6
10:18 tomaw joined #perl6
10:18 Pleiades` joined #perl6
10:18 SHODAN joined #perl6
10:18 chee joined #perl6
10:18 jerome joined #perl6
10:18 awwaiid joined #perl6
10:18 wtw joined #perl6
10:18 geekosaur joined #perl6
10:18 Timbus joined #perl6
10:18 Vlavv joined #perl6
10:18 masak joined #perl6
10:18 felher joined #perl6
10:18 drbean joined #perl6
10:18 Khisanth joined #perl6
10:18 hugme joined #perl6
10:18 Krunch joined #perl6
10:18 jrockway joined #perl6
10:18 ggoebel joined #perl6
10:18 broquaint joined #perl6
10:18 sorear joined #perl6
10:18 tadzik joined #perl6
10:18 Util joined #perl6
10:18 sergot joined #perl6
10:18 Juerd joined #perl6
10:18 dalek joined #perl6
10:18 snarkyboojum joined #perl6
10:18 rurban joined #perl6
10:18 SmokeMachine joined #perl6
10:18 au joined #perl6
10:18 risou joined #perl6
10:18 ilogger2 joined #perl6
10:18 MikeyG joined #perl6
10:18 ponbiki joined #perl6
10:18 PZt joined #perl6
10:18 hoelzro|away joined #perl6
10:18 Rix joined #perl6
10:18 mhasch joined #perl6
10:18 bonsaikitten joined #perl6
10:18 kcwu joined #perl6
10:18 ChoHag joined #perl6
10:18 krunen_ joined #perl6
10:18 spanner_ joined #perl6
10:18 mikec_ joined #perl6
10:18 nebuchadnezzar joined #perl6
10:18 mst joined #perl6
10:18 twinshadow joined #perl6
10:18 larks joined #perl6
10:18 gfldex joined #perl6
10:18 gugod joined #perl6
10:18 ashleydev joined #perl6
10:18 colo-work joined #perl6
10:18 doy joined #perl6
10:18 ingy joined #perl6
10:18 patspam joined #perl6
10:18 infosophy joined #perl6
10:18 timotimo joined #perl6
10:18 yeltzooo joined #perl6
10:18 ivan`` joined #perl6
10:18 cotto joined #perl6
10:18 s1n joined #perl6
10:18 breinbaas joined #perl6
10:18 steven joined #perl6
10:18 felipe joined #perl6
10:18 Celelibi joined #perl6
10:18 BooK joined #perl6
10:18 smash joined #perl6
10:18 shachaf joined #perl6
10:18 szbalint joined #perl6
10:18 quietfanatic joined #perl6
10:18 Hor|zon joined #perl6
10:18 rummik joined #perl6
10:18 rking joined #perl6
10:18 perigrin joined #perl6
10:18 pjcj joined #perl6
10:18 lestrrat joined #perl6
10:18 Ulti joined #perl6
10:18 jfried joined #perl6
10:18 Tene joined #perl6
10:18 rhr joined #perl6
10:18 charsbar_____ joined #perl6
10:18 `patch` joined #perl6
10:18 literal joined #perl6
10:18 rafl joined #perl6
10:18 sweet_kid joined #perl6
10:18 mathw joined #perl6
10:18 Exodist joined #perl6
10:18 cosimo joined #perl6
10:18 bakedb_ joined #perl6
10:18 phenny joined #perl6
10:18 japhb joined #perl6
10:18 LoRe joined #perl6
10:18 Bucciarati joined #perl6
10:18 simcop2387 joined #perl6
10:18 Pompel joined #perl6
10:18 nwc10 joined #perl6
10:18 ServerSage joined #perl6
10:18 renormalist joined #perl6
10:18 zamolxes joined #perl6
10:18 iemelianov joined #perl6
10:18 sjn joined #perl6
10:18 Maddingue joined #perl6
10:18 jnthn joined #perl6
10:18 diakopter joined #perl6
10:18 yves__ joined #perl6
10:18 TimToady joined #perl6
10:18 takesako_ joined #perl6
10:18 betterwo1ld joined #perl6
10:18 __sri joined #perl6
10:18 spacebat_ joined #perl6
10:18 moritz joined #perl6
10:18 cxreg joined #perl6
10:18 pnu_ joined #perl6
10:18 BinGOs joined #perl6
10:18 nuba joined #perl6
10:18 jtpalmer joined #perl6
10:18 flightrecorder joined #perl6
10:18 tadzik mtymula: http://perl6.org/documentation/ covers about all the best things we have
10:18 * xenoterracide reads over this tutorial and thinks this is very strange multiline comment
10:18 sorear #`()?
10:18 bonsaikitten tadzik: bzImage is compressed, it'll still eat quite a bit of RAM when it actually boots
10:18 bonsaikitten sadly on amd64 it appears impossible to get below 3MB without losing functionality
10:18 sorear also, that's just the text
10:18 xenoterracide sorear: yeah, I find it a bit odd
10:18 xenoterracide the example give is actualy #`(())
10:19 sorear bss (does the kernel use this?) and dynamic allocation (which the kernel definitely uses heavily) is not accounted for in the image
10:19 xenoterracide is there a p6pan yet?
10:19 sorear yes, it's spelled panda
10:19 sorear modules.perl6.org
10:19 xenoterracide ah
10:22 dakkar joined #perl6
10:22 eternaleye joined #perl6
10:22 SamuraiJack joined #perl6
10:22 anuby joined #perl6
10:22 buubot_backup joined #perl6
10:22 ironcamel joined #perl6
10:22 PerlJam joined #perl6
10:22 Gothmog_ joined #perl6
10:22 hugme joined #perl6
10:22 Util joined #perl6
10:22 larks joined #perl6
10:22 doy joined #perl6
10:22 `patch` joined #perl6
10:22 charsbar_____ joined #perl6
10:22 rhr joined #perl6
10:22 Tene joined #perl6
10:22 jfried joined #perl6
10:22 Ulti joined #perl6
10:22 lestrrat joined #perl6
10:22 pjcj joined #perl6
10:22 perigrin joined #perl6
10:22 rking joined #perl6
10:22 flightrecorder joined #perl6
10:22 sorear mtymula: hi!
10:24 aloha joined #perl6
10:24 _jaldhar_ joined #perl6
10:24 [particle]1 joined #perl6
10:24 Vlavv joined #perl6
10:24 rurban joined #perl6
10:24 twinshadow joined #perl6
10:24 rummik joined #perl6
10:25 nyuszika7h joined #perl6
10:25 tadzik bonsaikitten: well, do you really need it below 3MB on amd64?
10:26 xenoterracide well maybe I should try playing with Perl 6
10:26 sorear you need it below 1.44 to make a netboot floppy that can be read with standard hardware
10:26 bonsaikitten tadzik: "need" - no, my smallest boot devices are still >>32MB, but the constant increase in needs is annoying
10:26 dakkar joined #perl6
10:26 eternaleye joined #perl6
10:26 anuby joined #perl6
10:26 jfried joined #perl6
10:26 Ulti joined #perl6
10:26 lestrrat joined #perl6
10:27 SamuraiJack joined #perl6
10:27 buubot_backup joined #perl6
10:27 ironcamel joined #perl6
10:27 Gothmog_ joined #perl6
10:27 hugme joined #perl6
10:27 Util joined #perl6
10:27 larks joined #perl6
10:27 doy joined #perl6
10:27 pjcj joined #perl6
10:27 Tene joined #perl6
10:27 rhr joined #perl6
10:27 charsbar_____ joined #perl6
10:27 `patch` joined #perl6
10:27 flightrecorder joined #perl6
10:28 xenoterracide gods the nqp repo is huge...
10:28 ggoebel joined #perl6
10:28 sorear joined #perl6
10:28 tadzik joined #perl6
10:28 dalek joined #perl6
10:28 SmokeMachine joined #perl6
10:28 Hor|zon joined #perl6
10:28 quietfanatic joined #perl6
10:28 szbalint joined #perl6
10:28 shachaf joined #perl6
10:28 smash joined #perl6
10:28 BooK joined #perl6
10:29 flightrecorder joined #perl6
10:29 xenoterracide wow parrots impressively even bigger
10:30 Krunch joined #perl6
10:30 broquaint joined #perl6
10:30 snarkyboojum joined #perl6
10:30 steven joined #perl6
10:33 mtymula who can I ask about development kit for perl6 :) on windows?? i've installed dwimperl-5.14.2.1-v7-32bit.exe on my XP32 and added some features for perl6... and then it cruches...
10:33 mtymula ;(
10:33 mtymula *crushes
10:33 sorear dwimperl, was that Gabor's?
10:34 moritz yes
10:34 moritz see http://szabgab.com/contact.html for contact options
10:34 dalek niecza: f2586f9 | sorear++ | lib/ (2 files):
10:34 dalek niecza: Separate responsibilities of NewTypedScalar / NewAnyMuScalar
10:34 dalek niecza: review: https://github.com/sorear/niecza/commit/f2586f9497
10:34 mtymula are there any other options or only dwimperl ??
10:34 jnthn https://github.com/rakudo/star/downloads has an MSI
10:35 jnthn Includes Rakudo Perl 6 compiler, some modules, etc.
10:35 jnthn Debugger :)
10:36 mtymula in a fancy nice GUI style or just command line??
10:37 moritz fancy nice command line :-)
10:37 mtymula i would be great to have sth like eclipse... :p
10:38 xenoterracide curious I heard a rumor of a feature in node... of if you `require`ed a module you didn't have, it would simply download and install it for you at require time. Some people seem to like this... bad idea?
10:38 mtymula but this dwimperl gives me lots of problems...
10:38 moritz xenoterracide: great for development, horrible for production
10:39 sorear I am using node and I have not encountered that
10:39 xenoterracide it is a rumor, perhaps I do not know what I'm talking about
10:39 xenoterracide because I haven't gotten around to trying node
10:40 sorear http://nodejs.org/api/module​s.html#modules_all_together
10:40 sorear documentation says the rumor is false
10:40 sorear (node's documentation is not trustworthy)
10:40 mtymula ok, need to restart the system... will be back to add some more stupid questions;p
10:41 dalek niecza: 3ffe64c | sorear++ | lib/ (2 files):
10:41 dalek niecza: Top removal: NewMuAnyScalar
10:41 dalek niecza: review: https://github.com/sorear/niecza/commit/3ffe64c5af
10:43 xenoterracide sorear: maybe I didn't understand what they were talking about... or they didn't...
10:44 * xenoterracide stress tests my system by compiling and checking for dupes at the same time
10:44 sorear sleep&
10:44 * xenoterracide pretends UI should be responsive... yes?
10:49 domidumont joined #perl6
10:56 xenoterracide make && make test ran ok
10:56 xenoterracide but where would make install have put this stuff
10:57 xenoterracide it didn't end up in my path
10:58 xenoterracide oh maybe it just put it in ./install
11:07 rindolf joined #perl6
11:08 gootle joined #perl6
11:08 domidumont joined #perl6
11:17 marloshouse joined #perl6
11:31 cognominal joined #perl6
11:33 cognominal joined #perl6
11:33 tokuhiro_ joined #perl6
11:33 mtymula joined #perl6
11:33 masak g'day, #perl6
11:34 * masak has a Perl 6 day today o/
11:34 mtymula good for you:)
11:34 * moritz still haz no internez at $HOME :(
11:35 masak moritz: :/
11:36 mtymula here you have a guestion... why my *.pl file with utf-8 coding does not fire when i make perl6 *.pl
11:37 moritz well, if you write  perl6 *.pl and there are multiple of them, it will only launch the first
11:37 moritz and take the other files names are command line arguments to the first
11:37 mtymula no... :) i just wrote it here
11:38 mtymula let's say it is test.pl
11:38 mtymula test.pl has utf-8 coding
11:38 jnthn mtymula: Perhaps because a handful of Windows programs seem to think utf-8 needs a byte order mark at the start of the file. :/
11:39 mtymula i get ===SORRY!=== Confused at line 1, near "\ufeffuse v6;\r\n"
11:39 jnthn hah, yes.
11:39 jnthn \ufeff
11:39 mtymula what shall i do??
11:40 jnthn Get rid of the \ufeff at the start of the file.
11:40 mtymula come on... there is no such thing
11:41 mtymula the whole file is:
11:41 moritz yes there is there
11:41 mtymula use v6;  my $liczba = prompt "Podaj liczbe od 41 do 49 włacznie: ";
11:41 moritz your editor just doesn't show you
11:41 jnthn It's in the output you pasted...and what moritz said.
11:41 moritz use a hex editor to confirm, if you want
11:41 moritz and a better text editor
11:41 mtymula well i use notepad ++
11:42 jnthn mtymula: Me too
11:42 jnthn Oh...
11:42 jnthn On the encoding menu select "UTF-8 without BOM"
11:42 * jnthn thought that was the default...
11:43 mtymula <hahaha>
11:43 mtymula ok
11:43 mtymula it fiered
11:43 mtymula but
11:44 mtymula in the output my polish charaters are displayed incorrectly
11:44 mtymula ł
11:44 mtymula instead of them i see some kind of forest....
11:44 moritz maybe your console/shell doesn't understand UTF-8?
11:45 moritz r: say "włacznie"
11:45 p6eval rakudo 72a27f: OUTPUT«włacznie␤»
11:45 tadzik włącznie
11:45 moritz phenny: "włacznie"?
11:45 phenny moritz: "inclusive" (pl to en, translate.google.com)
11:45 tadzik that's incorrect
11:45 tadzik it's włącznie, not włacznie
11:45 mtymula :)
11:46 mtymula tadzik how do you know that??
11:46 mtymula ;p
11:46 tadzik mtymula: z doświadczenia życiowego ;)
11:46 mtymula pozdrowienia, jesteśmy wszędzie:)
11:46 tadzik yep, there's at least few of us here
11:46 tadzik Woodi, Gli<TAB><TAB>
11:47 tadzik sergot too
11:47 mtymula so oh... nice:)
11:48 brrt joined #perl6
11:48 mtymula are you located in poland or somewhere else??
11:48 marloshouse joined #perl6
11:49 jnthn cmd.exe does pretty awfully at utf-8
11:49 tadzik mtymula: Warsaw
11:49 jnthn chcp 65001 # sorta helps
11:50 mtymula cmd.exe it is
11:50 mtymula tadzik:  me Rzeszow
11:50 tadzik nice
11:51 mtymula i was here several times (meaning on this IRC)
11:51 mtymula and every time i came accross something that blows my mind:)
11:52 tadzik hehe
11:52 mtymula i really start loving this perl community:)
11:52 tadzik yeah, it's awesome
11:53 masak utf-8 files with a BOM has to be one of the stupidest things ever.
11:53 mtymula loving << not quite English...
11:53 mtymula masak whatever you say... i have no idea
11:54 grondilu joined #perl6
11:54 mtymula to be honest i am new at this all stuff especially perl6
11:54 grondilu r: say <foo bar longstring>.max( *.chars <=> *.chars );'
11:54 p6eval rakudo 72a27f: OUTPUT«===SORRY!===␤Unable to parse expression in quote:sym<apos>; couldn't find final "'" at line 2, near ""␤»
11:54 FROGGS masak: can you determine the endianess withouth the bom?
11:54 grondilu r: say <foo bar longstring>.max( *.chars <=> *.chars );
11:54 p6eval rakudo 72a27f: OUTPUT«longstring␤»
11:54 pmurias joined #perl6
11:55 * grondilu whishes he could just write:  say <foo bar longstring>.max( *.chars );'
11:55 masak FROGGS: yes. see https://en.wikipedia.org/w​iki/Byte_order_mark#UTF-8
11:55 tokuhiro_ joined #perl6
11:55 FROGGS ahh I see
11:55 masak I wonder why we shouldn't simply strip the BOM in Rakudo if we find it.
11:55 FROGGS stupid thing it is then
11:55 masak would make for fewer surprised first-time users.
11:56 masak we could even put it in the grammar, and just ignore it.
11:56 FROGGS I'd say that skipping is better than actually modifying some code
11:56 masak that's what I'm suggesting.
11:57 masak sorry, I have an FP background; "strip" doesn't mean "modify original" to me :P
11:57 jnthn Yes, we could diddle the grammar.
11:57 mtymula thx for help:)
11:58 mtymula i will give up writing in polish it is to damn complicated anyway:)
11:58 masak *or* we could flag it up with a dedicated error message, if we were so inclined.
11:58 masak just not "Confused" :(
11:58 * masak submits rakudobug
11:58 jnthn masak: Confused is how I feel about utf-8 having a BOM :P
11:58 masak :P
11:58 jnthn I suggest we make things just work.
11:59 masak me too.
11:59 mtymula ;)
11:59 FROGGS "===SORRY!=== Please remove that stupid BOM at line 1 at ..." ?
11:59 FROGGS DWIM++
11:59 jnthn Should be a one-line patch, but I'm currently writing slides for a talk this evening :P
11:59 mtymula ===SORRY!=== Rakudo is confused please don't do that..."
12:00 FROGGS btw, I made a pull request ages ago about me beeing a rakudo release-manager... I guess I still want to do it
12:03 masak weighing in on the discussion about "readable Python". any language can be written in an obfuscated manner. http://p-nand-q.com/python/obfuscated_python.html -- most encouragement for "writing things well" must come from the community.
12:04 brrt chiming in a bit on python, i find that the standards of code for perl (5) are much higher than that on python
12:04 brrt i.e., no cpan module comes without a test suite
12:04 brrt or very few, maybe
12:05 brrt hardly any python modules come with decent tests, or documentation
12:05 masak FROGGS: cool!
12:05 masak FROGGS: for November?
12:05 FROGGS right
12:05 pmurias joined #perl6
12:05 masak tadzik: is it OK if I move you to February?
12:06 tadzik masak: sure
12:06 * masak makes it so
12:06 moritz \o/ new release managers
12:06 mtymula tadzik i see that pmurias i also... ONE OF US...;pp
12:06 masak yay, instead of 0 planned release slots, we now have 4 :)
12:06 tadzik right :)
12:06 FROGGS k, so I'll try to bundle everything these days to be prepared on 22nd
12:06 FROGGS there is a msi-howto these days?
12:07 masak mtymula: ja tez mowie po polsku :)
12:07 masak (hm, that's probably not exactly right...)
12:08 mtymula prove it ;p
12:09 dalek rakudo/nom: 0aca028 | masak++ | docs/release_guide.pod:
12:09 dalek rakudo/nom: [docs/release_guide.pod] FROGGS++ November release
12:09 dalek rakudo/nom:
12:09 dalek rakudo/nom: Bumped tadzik forward to February.
12:09 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/0aca028ff0
12:09 mtymula masak: udowodnij;p
12:09 tadzik last time I checked masak distinguished tą and tę correctly :)
12:09 mtymula oh!!
12:09 masak oh, 's easy.
12:10 * masak says "OONNGG", "EENNGG"
12:10 masak oh, you mean grammatically?
12:10 tadzik right
12:10 mtymula so read this: stół z powyłamywanymi nogami
12:10 tadzik also, "last time I checked" was what, 1.5 year ago? ;)
12:13 mtymula wtf... i see I don't know perl6 more than i suspected...
12:13 mtymula i wrote my $kwadrat_roznicy = $robocza^2;
12:13 mtymula where $kwadrat_roznicy had value of 8
12:13 masak tadzik: yeah, almost.
12:13 jnthn Raise to power is **
12:14 mtymula and i surely did't suspected that i t will be  one(8, 2)
12:14 masak mtymula: ^ is a junctive operator.
12:14 mtymula what does it mean... never mind i will find out
12:15 masak mtymula: I can say "powyłamywanymi" aloud, but only very slowly. probably 1/5 the speed you can.
12:15 mtymula but after that i see that this will be a loooooooong journey
12:15 masak mtymula: you can do things like this: 'if $answer == 2 | 3 | 5'
12:15 mtymula masak:  try  chsząszcz brzmi w trzcinie
12:15 masak those are junctions. they allow you to do several similar computations at the same time.
12:16 masak mtymula: did you mean "chrząszcz"? :)
12:16 mtymula masak: thats nice
12:16 mtymula masak: yea you got it
12:17 masak mtymula: "rz" doesn't scare me much. I've been to the Czech republic, too.
12:17 mtymula masak: and do you know what does it mean szukać ??
12:18 masak "quest", "look for"?
12:18 mtymula google translate, huh?
12:18 mtymula ;p
12:18 masak oh yes.
12:18 masak what tipped you off?
12:18 * masak is curious
12:18 arnsholt That's kind of similar to "seek".
12:19 mtymula hm...
12:19 * arnsholt wonders what sounds changes happened in Slavic
12:19 masak arnsholt: I was thinking the same.
12:19 mtymula a deeeeeelay and two meanings;p
12:19 arnsholt To the textbooks!
12:19 masak mtymula: the delay will be taken care of once people find a way to implant Google Translate into my brain.
12:19 mtymula and any one of you know what does it mean i n Czech republic??
12:19 masak mtymula: something naughty, I'll bet.
12:20 jnthn mtymula: I think in Slovak it sounds like the word "to fuck" at least.
12:20 mtymula masak; i am also looking forward to it
12:20 mtymula gooood:) you won pride
12:20 Woodi afternoon ppls
12:20 * masak .oO( she said she'll go look for him, and she's been gone for a long time... )
12:20 mtymula ^^
12:21 masak Woodi! \o/
12:21 masak Woodi: we're having a Polish moment here.
12:21 masak join us.
12:21 pmurias joined #perl6
12:21 Woodi :)
12:21 mtymula indeed
12:21 Woodi I am 3 more screens from here... sec...
12:22 mtymula and it is my foult
12:22 masak hey, don't go taking all the blame there! :P
12:23 mtymula i am worried it will go to some kind of a repository of posts
12:23 jnthn And hey, many folks here like natural languages as well as programming languages. :)
12:23 * masak likes it what the BOM is spelled "feff" -- that summarizes his feelings for it
12:23 mtymula ok, I will not
12:23 jnthn masak: *lol*...never noticed that :)
12:23 huf mtymula: the channel is logged publicly on the web :)
12:23 arnsholt masak: At least in the transition PIE->Balto-Slavic g would be preserved as g, not g->k as in Germanic, so it's not an inherited word (unless g->k happens later in the development of Polish of course)
12:23 mtymula good one
12:23 huf mtymula: when the revolution comes, you wont be able to deny being polish :)
12:23 arnsholt Might be a German loan
12:24 mtymula what revolution ??
12:24 huf mtymula: oh you know, the next one, whatever it will be about.
12:25 * masak Perl 6: a predominantly Polish project...
12:25 huf why, do you want to hide your polishicity?
12:25 mtymula beside that why would i like to deny being polish
12:25 huf "Perl 6: Polished"
12:25 huf could be a tv series
12:25 masak arnsholt: cool.
12:26 mtymula and with two meanings
12:26 mtymula nice
12:27 mtymula ok since we came into poland tell me who want's to come to Rzeszów, dring some beer;p
12:28 * tadzik mumbles something about Polish Perl Workshop
12:28 jnthn ooh, beer!
12:29 masak Polish beer! \o/
12:29 jnthn I'd go to Poland and drink beer :)
12:29 jnthn Even with the flight factored in it's probably cheaper than drinking here in Sweden :P
12:30 * jnthn pokes tadzik to get the Polish Perl Workshop planned :D
12:30 masak jnthn: well, you can calculate how much beer you'd have to drink... :)
12:30 moritz jnthn: for your next Perl 6 introduction talk, write a script that calculates how much... what masak said
12:30 huf why calculate? experiment!
12:30 jnthn hah!
12:30 moritz masak: don't be so creepy and think the same things I do :-)
12:31 jnthn I can't remember whether the cheapest beer I ever had was in a village just over the Czech side of the Czech/Slovak border, or in Shanghai. I do know which of these two were better :P
12:31 mtymula experiment?? it is called an practical observation of chemical influence over sate of ones mind and financial status
12:32 moritz nothing wrong with experiments; science is based on them
12:32 huf yes, "science" mythbusters style :)
12:32 mtymula indeed
12:32 huf in this case
12:32 * moritz did what the scientists call "computer experiments"
12:32 masak moritz: whenever you end up in an entangled state like that, the clue is to make lots of random measurements to disentangle. like, turn on the radio on a random channel :P
12:33 tadzik jnthn: yeah, we'll think about it ;)
12:33 mtymula or drink random beer at random places??
12:33 moritz masak: given that 99% of our total communication volume is not on IRC, that should happen automatically
12:34 Woodi quest: intro: one of the popes on his dying-bed wanted "piwa di Polonia"... (pl: piwo = beer) They who hear this was probably consternated and they started talking: "O, Santa Piva di Polonia!"... Quest-todo: find polish beer which is such good you want it on dying-bed :)
12:34 mtymula oh, you've made my day :)
12:35 Psyche^ joined #perl6
12:35 * Woodi confirms "stół z powyłamywanymi nogami" is important test in polish-speed-talking :)
12:35 mtymula "find polish beer which is such good you want it on dying-bed" now this is not so tuff task...
12:36 Woodi ehm... this was few centuries ago... do good beer stil left here ? :)
12:36 mtymula tuff  << nqe
12:36 mtymula yes Woodi, there are lots of it
12:37 mtymula aspeically in when someone is making it for himself;p
12:37 mtymula by himself
12:37 mtymula or herself
12:37 mtymula whatever
12:38 Woodi heh :) /me just do nalewki :)
12:38 mtymula księżycówki??
12:38 mtymula you can lite it up wuith fire??
12:38 mtymula *with
12:39 Woodi hmm, no, probably... just fruits and spirit...
12:39 mtymula it is this amount of a ghost (spirit) that i am worried ...
12:39 masak moritz: in that case, there's a disproportionate amount of entanglement going on between our brains. cause for worry, I'd say. :P
12:41 Woodi ok, speeking about method modifiers eg. before - I have no idea how callsame replace it in deterministic way...
12:42 masak r: class C { method foo { say "C" } }; class D is C { method foo { say "before"; callsame; say "after" } }; D.new.foo
12:42 p6eval rakudo 72a27f: OUTPUT«before␤C␤after␤»
12:42 masak like that.
12:42 Woodi before and after are clear but method call dispatch and find right method. then callsame find next same method...
12:42 jnthn same here means "same arguments"
12:42 masak right.
12:43 jnthn contrasted with nextwith where you get to diddle the arguments
12:43 Woodi but call same means: go up in hierarchy and find next like it ?
12:44 Woodi naive view of before and after is they are wrappers. inheritance not needed...
12:47 masak joined #perl6
12:48 cosimo joined #perl6
12:54 Woodi anybody here ?
12:54 mtymula ofc
12:54 tadzik yes
12:54 rindolf Woodi: hi.
12:55 Woodi just probably me not grasping things again :)
12:56 moritz Woodi: (I hope that I'm not telling you bullshit here) in the case of single dispatch, nextsame just goes to the next method of the same name by walking the MRO
12:56 moritz Woodi: in the case of multi dispatch, it goes to the next-looser candidate
12:57 Woodi moritz: ok. that start my problem with it :)
12:57 Woodi it find first. what it if finds other one and later that "first" ? it is not clear comparing to "before" etc
12:57 lumi_ joined #perl6
12:58 * moritz doesn't quite understand the question
12:58 Woodi dispach is like magic box
12:58 moritz ah, it's not :-)
12:58 Woodi with inheritance it is clear
12:58 moritz I think S06 has something on how dispatch works
12:59 moritz basically it sorts the signatures by narrowness
12:59 Woodi k :)
12:59 Woodi but signature is same..
12:59 moritz in which example?
12:59 Woodi < jnthn> same here means "same arguments"
12:59 moritz arguments != signature
13:00 moritz sub f(Int $x) { say $x}; f 5;
13:00 moritz (Int $x)  here is a signature
13:00 moritz and 5 is an argument
13:00 jnthn The way to understand it is that whenever you make a call, there are a set of possible things that could be called for those arguments.
13:00 jnthn By default you just call the best one
13:00 jnthn The [next|call][with|same] just let you pass contorl to the next best one.
13:00 jnthn It's like iterating through a list of possible things.
13:01 Woodi it's non-quite-determistic for me so far..
13:01 moritz iterating a list is quite deterministic
13:04 moritz Woodi: if S06 doesn't help you, come up with some examples, and we can explain why candidates are sorted the way they are
13:04 jnthn The list doesn't mutate with time, if that helps.
13:06 Woodi k, will read in sec. but "callsame" means call another with same signature. but defining function with same signature (with absence of inheritance) is error for me
13:06 sftp joined #perl6
13:06 moritz no
13:06 * Woodi is shocked here :)
13:06 moritz callsame means call another subroutine with the same PARAMATERS
13:06 moritz not with the same signature
13:07 moritz you need to clearly distinguish signatures and paramaters, otherwise you're lost
13:07 Woodi ahh... Int -> Numeric -> Any... something like this ?
13:07 jnthn class A { method m($a) { say "in A with $a" } }; class B is A { method m($x) { say "in B with $x"; nextsame; } }; B.m
13:07 jnthn r: class A { method m($a) { say "in A with $a" } }; class B is A { method m($x) { say "in B with $x"; nextsame; } }; B.m
13:07 p6eval rakudo 72a27f: OUTPUT«Not enough positional parameters passed; got 1 but expected 2␤  in method m at /tmp/kyFlgCdmSa:1␤  in block  at /tmp/kyFlgCdmSa:1␤␤»
13:07 jnthn r: class A { method m($a) { say "in A with $a" } }; class B is A { method m($x) { say "in B with $x"; nextsame; } }; B.m(42)
13:07 p6eval rakudo 72a27f: OUTPUT«in B with 42␤in A with 42␤»
13:08 moritz note that those two methods m don't have the same signature, because they have an implicit invocant argument
13:08 Woodi jnthn: you used inheritance here :)
13:09 Woodi call same is like call SUPER or call-Any-after-calling-int ?
13:09 moritz r: multi f(Cool $x) { say "In f(Cool) with $x"; }; multi f(Int $x) { say "in f(Int) with $x"; nextsame }; f 42
13:09 p6eval rakudo 0aca02: OUTPUT«in f(Int) with 42␤In f(Cool) with 42␤»
13:10 jnthn Woodi: Yes, and nextsame took you do the next thing up the inheritance tree.
13:10 jnthn *to
13:10 Woodi ok, so reading and learning is essential :)
13:10 masak news at 11.
13:13 krunen joined #perl6
13:17 avar joined #perl6
13:17 avar joined #perl6
13:19 timotimo joined #perl6
13:23 groky joined #perl6
13:23 bbkr__ joined #perl6
13:25 * Woodi just found "proto" which looks like begin&after :)
13:36 * moritz wants to send a present to whoever invented git reflog
13:38 masak git reflog ejects some git concerns orthogonally from the ordinary git usage. it's like the AOP of git. :)
13:38 moritz it's like an implicit try/catch around git rebase -i to me :-)
13:39 moritz most of the time, at least
13:40 moritz it's so good to see that even some apparently destructive operations can be undone in git
13:41 colomon joined #perl6
13:42 shachaf git reflog is the future
13:42 shachaf (Well, the past. Whatever.)
13:43 jnthn It's a window back to the past's alternative futures. :)
13:44 shachaf Traversals are the future, though!
13:48 * moritz looks into the future and sees... errors
13:57 mtymula :)
13:58 tokuhiro_ joined #perl6
13:59 benabik joined #perl6
14:00 tokuhiro_ joined #perl6
14:05 PacoAir joined #perl6
14:06 tokuhiro_ joined #perl6
14:06 tokuhiro_ joined #perl6
14:08 atrodo joined #perl6
14:13 flightrecorder joined #perl6
14:24 FROGGS joined #perl6
14:30 yves_ joined #perl6
14:32 bowtie joined #perl6
14:33 GlitchMr joined #perl6
14:34 isBEKaml joined #perl6
14:37 dalek roast: 7adc0af | masak++ | S02-lexical-conventions/bom.t:
14:37 dalek roast: [S02-lexical-conventions/bom.t] new .t file
14:37 dalek roast:
14:37 dalek roast: Tests that an implementation can handle/ignore a BOM
14:37 dalek roast: at the start of the file.
14:37 dalek roast: review: https://github.com/perl6/roast/commit/7adc0afa72
14:37 dalek roast: 42e1808 | masak++ | .gitignore:
14:37 dalek roast: [.gitignore] ignore S32-io/server-ready-flag
14:37 dalek roast: review: https://github.com/perl6/roast/commit/42e18080b5
14:39 isBEKaml we lost a bom!
14:39 dalek rakudo/nom: fc349a9 | masak++ | / (3 files):
14:39 dalek rakudo/nom: ignore a BOM at the beginning of a file
14:39 dalek rakudo/nom:
14:39 dalek rakudo/nom: Now when somebody sets up us it, all their base
14:39 dalek rakudo/nom: will be belong to Rakudo.
14:39 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/fc349a9dd6
14:40 masak that surely was easy.
14:40 * masak goes on to close the RT ticket
14:41 FROGGS masak: but it just ignores it when it is at the beginning of a file, right?
14:41 geekosaur bom be bomp
14:42 masak FROGGS: right.
14:42 FROGGS good, thanks ;o)
14:42 masak beginning of a compilation unit, even. so you can put it at the start of your &eval strings, too.
14:49 isBEKaml rn: my @foo = "foo" xx 5; @foo.perl.say; @foo.say;
14:49 p6eval rakudo 0aca02: OUTPUT«Array.new("foo", "foo", "foo", "foo", "foo")␤foo foo foo foo foo␤»
14:49 p6eval ..niecza v22-29-g3ffe64c: OUTPUT«["foo", "foo", "foo", "foo", "foo"].list␤foo foo foo foo foo␤»
14:50 isBEKaml Oh, something.perl now gives the type info too?
14:52 rindolf joined #perl6
14:56 moritz ironcamel: well, it's supposed to spit out something that, when eval()ed, produces an object of the same type as before
14:56 moritz erm, meant isBEKaml, sorry
14:58 Woodi r: sub foo() { say "Foo!" }; foo
14:58 p6eval rakudo 0aca02: OUTPUT«Foo!␤»
14:58 Woodi f: sub detect-foo() { syslog("WARNING! Foo fires at " ~ Time::HiRes::now() ); }
14:58 Woodi f: sub-detect-detect-foo() { ... }
14:59 Woodi how do this with (next|call)same
14:59 Woodi ?
14:59 moritz you don't; there's no call to a function of the same name in your example
15:00 Woodi f sub detect-foo() before foo() { ... }
15:00 geekosaur wrap()?
15:00 mtymula is it something in the air or programming in perl6 is so fkng gooooooood:)
15:00 mtymula ??
15:00 Woodi mtymula: no idea :)
15:01 Woodi but all that after|before and call same discusion is about not doing after|before becouse callsame & co do the same...
15:01 moritz r: sub trait_mod:before(&c) { &c.wrap(sub {"WARNING: {&c.name} fires at " ~ now; nextsame }) };
15:02 p6eval rakudo 0aca02:  ( no output )
15:02 mtymula but to be honest it is easier to start writing code than in c# or java... at least for me:)
15:02 moritz Woodi: something like that
15:03 Woodi ok :) so no idea why callsame & co are mentioned in before|after context
15:04 stopbit joined #perl6
15:07 Woodi and about that Debian offtopic: one I was working in firm where Debian was used (good!). but admin there was using dselect! it so distracting as he would be using fedora tools... I like apt-get :) and ppls like aptitude and it is quite religious difference :)
15:08 Woodi and when you add stable|testing|sid difference it is like debian was realy general-something
15:09 Woodi btw. I am addicted to daily apt-get update ; apt-get upgrade ; apt-get dist-upgrade to the level I was missing it using OpenBSD
15:10 Woodi now, when back on debian I quite miss 1 or 2 erratas by year :)
15:11 kaare_ joined #perl6
15:14 obra joined #perl6
15:27 mtymula someone said before that uses notepad++
15:27 mtymula do anyone has a knowladge about plugin to make syntax of perl6 look nice in this editor??
15:28 * Woodi just saw wrap example use nextsame...
15:28 Woodi mtymula: perl6 syntax is hard thing
15:28 tadzik mtymula: one '?' is usually sufficient to make a question :)
15:28 tadzik mtymula: something for Perl 5 may work okay
15:29 tadzik personally I prefer to have no syntax hilighting, esp. for Perl 6
15:30 mtymula ?? < work related. It is just how i do it. I just cannot resist to make it double...
15:30 mtymula tadzik: why?
15:30 mtymula ;p
15:31 tadzik I tried it for a week and now syntax hilighted code looks awful for me :)
15:31 tadzik are you working for Gadu-Gadu perhaps? ;)
15:31 Woodi tadzik: maybe other "theme" can help :)
15:32 tadzik I like my "everything is yellow" theme :)
15:32 mtymula i contact my friends at work via GG, my clients and my friends... so... ;pp
15:32 Woodi mtymula: it's not secure :)
15:33 mtymula what?? GG??
15:33 mtymula ;p
15:33 Woodi or some encryption is now implemented ?
15:33 Woodi however emails have same lvl of security :)
15:33 mtymula no, it is not secure.
15:33 mtymula but great for fast contact
15:33 Woodi right
15:34 tadzik emails are usually ssl'd between you and the mail server
15:34 mtymula because, and i hate that fact, everyone has it
15:34 tadzik you don't go with a phone/laptop to a shopping mall, open wireshark and read people's emails
15:34 tadzik *cough*, yeah
15:34 Woodi but server <-> server it is clear text
15:34 tadzik no. It's not guaranteed to be something more than clear text.
15:34 tadzik same with xmpp
15:38 bapa joined #perl6
15:39 brrt joined #perl6
15:39 Woodi r: sub foo() { say "abc foo" }; sub trait_mod:before(&c) { &c.wrap(sub {"WARNING: {&c.name} fires at " ~ now; nextsame }) }; before(foo); foo
15:39 p6eval rakudo fc349a: OUTPUT«===SORRY!===␤CHECK FAILED:␤Undefined routine '&before' called (line 1)␤»
15:41 brrt joined #perl6
15:42 moritz if you want to call it like that, don't make it a trait (but a normal sub), and call it as   before(&foo)
15:44 Woodi moritz: for now I have no idea how to use trait so just used call :)
15:44 arnsholt r: sub foo { say "foo" }; sub trait_mod:<is>(Routine $wrapper, :$before!) { $before.wrap($wrapper) }; sub bar is before(&foo) { say "bar"; nextsame; }; foo # Another option
15:44 p6eval rakudo fc349a: OUTPUT«bar␤foo␤»
15:45 Woodi r: sub foo { say "foo" }; sub trait_mod:<is>(Routine $wrapper, :$before!) { $before.wrap($wrapper) }; sub bar is  before(&foo) { say "bar"; };
15:45 p6eval rakudo fc349a:  ( no output )
15:46 Woodi r: sub foo { say "foo" }; sub trait_mod:<is>(Routine $wrapper, :$before!) { $before.wrap($wrapper) }; sub bar is  before(&foo) { say "bar"; }; foo
15:46 p6eval rakudo fc349a: OUTPUT«bar␤»
15:47 tadzik r: sub foo { say "foo" }; sub trait_mod:<is>(Routine $wrapper, :$before!) { $before.wrap(sub { $wrapper(); nextsame }) }; sub bar is before(&foo) { say "bar"; }; foo
15:47 p6eval rakudo fc349a: OUTPUT«bar␤foo␤»
15:48 Woodi if I use wrap then nextsame feels strange for me. this is "wrap" not another like one before...
15:48 geekosaur dunno, it makes sense to me...
15:48 arnsholt You need to explicitly state where to invoke the wrapped sub
15:49 arnsholt Since the compiler can't automatically infer what goes before and what goes after
15:50 Woodi yes, but something like "callwrapped" would be natural. but here is used more general mechanism probably
15:55 masak "callwrapped" seems a weird break on the current nomenclature.
15:55 masak given that the other two variants are "-same" and "-with".
16:17 benabik joined #perl6
16:17 kresike bye folks
16:17 benabik joined #perl6
16:21 PerlJam Woodi: then use "callsame" instead of "nextsame"  (but I see what you're saying wrt feeling weird)
16:22 MayDaniel joined #perl6
16:29 spanner joined #perl6
16:38 sisar joined #perl6
16:38 sisar o/
16:39 masak sisar! \o/
16:39 mikemol joined #perl6
16:40 masak mikemol! \o/
16:40 mikemol o/
16:40 PerlJam r: sub foo { say "foo" }; &foo.wrap(sub { say "bar" });
16:40 p6eval rakudo fc349a:  ( no output )
16:41 PerlJam Hmm
16:41 TimToady welcome back shortci^W<window closed>
16:41 masak :P
16:41 sisar my @a = 1, 2, 3, 4; say sqrt(@a); #how do I make sqrt apply to all the elements of @a?
16:41 sisar rn: my @a = 1, 2, 3, 4; say sqrt(@a); #how do I make sqrt apply to all the elements of @a?
16:41 masak sisar: a for loop? :)
16:41 p6eval rakudo fc349a, niecza v22-29-g3ffe64c: OUTPUT«2␤»
16:41 masak rn: my @a = 1..4; say @a>>.sqrt
16:42 p6eval rakudo fc349a: OUTPUT«1 1.4142135623731 1.73205080756888 2␤»
16:42 p6eval ..niecza v22-29-g3ffe64c: OUTPUT«1 1.4142135623730951 1.7320508075688772 2␤»
16:42 masak rn: my @a = 1..4; say (sqrt $_ for @a)
16:42 p6eval rakudo fc349a: OUTPUT«1 1.4142135623731 1.73205080756888 2␤»
16:42 p6eval ..niecza v22-29-g3ffe64c: OUTPUT«1 1.4142135623730951 1.7320508075688772 2␤»
16:42 sisar masak: yeah, ithink i was looking for the >> thingy
16:42 masak people always look for the >> thingy.
16:42 sisar what is it called ?
16:42 masak hyper method call.
16:43 sisar masak: thanks.
16:43 fgomez joined #perl6
16:45 grondilu joined #perl6
16:46 grondilu Check out my version of the evolutionary algorithm:  http://rosettacode.org/wiki/E​volutionary_algorithm#Perl_6
16:46 grondilu it's much shorter than in other languages
16:47 masak grondilu: nice!
16:47 masak grondilu++
16:48 * masak always enjoys seeing max :by
16:48 * grondilu is glad you like it
16:48 TimToady it would be easier to follow in the log if you didn't check in *quite* so often :)
16:48 masak sorry...
16:49 masak grondilu: nice implicit loop with 'mutate($parent) xx C', too.
16:49 TimToady I guess I can just do a manual diff on the two versions though...
16:49 grondilu TimToady: sorry, I just happen to have little ideas coming one at a time
16:49 TimToady happens to me sometimes too, just something to bear in mind
16:49 grondilu masak: I quite hesitated to use 'xx' but it seems ok
16:50 masak yeah. the semantics to re-eval the lhs is relatively new.
16:50 sorenso joined #perl6
16:50 masak (and an improvement, if you ask me)
16:50 grondilu I wonder how it knows whether the lhs is deterministic or not, though
16:51 masak it doesn't.
16:52 masak at least not current implementations.
16:52 masak they just re-eval the thunk regardless.
16:52 grondilu ah ok
16:52 masak rn: say (2.rand.Int) xx 5
16:52 p6eval rakudo fc349a: OUTPUT«1 0 1 1 0␤»
16:52 p6eval ..niecza v22-29-g3ffe64c: OUTPUT«1 1 1 0 1␤»
16:52 masak oops, superstitious parens.
16:53 masak rn: say (2.rand.Int xx 5).join
16:53 p6eval niecza v22-29-g3ffe64c: OUTPUT«01000␤»
16:53 p6eval ..rakudo fc349a: OUTPUT«01001␤»
16:53 sqirrel joined #perl6
16:53 masak rn: say Bool.roll xx 5
16:53 p6eval rakudo fc349a: OUTPUT«False False True True True␤»
16:53 p6eval ..niecza v22-29-g3ffe64c: OUTPUT«False True False False True␤»
16:54 masak (if that were an iterated prisoner's dilemma, Niecza would come out on top) :P
16:54 PerlJam r: https://gist.github.com/4066950
16:54 p6eval rakudo fc349a: OUTPUT«foo␤No such method 'wrap' for invocant of type 'Bool'␤  in block  at /tmp/zvbNpHDRp5:5␤␤»
16:54 PerlJam there we go
16:55 PerlJam IS that a known bug?
16:55 tipdbmp joined #perl6
16:55 masak PerlJam: yes.
16:55 masak try a semicolon at the end of the first sub declaration.
16:56 PerlJam crazy
16:56 masak it's in RT. I'm looking.
16:56 tipdbmp Where can I find a grammar that can parse mathematical expressions? I would like to learn a bit about grammars and thought that Perl 6 would be a nice way to do it. Should I start with math exprs in the first place?
16:57 TimToady I guess my first question would be, "mathematical expressions" in what language?
16:57 masak tipdbmp: sounds like a nice exercise, yes.
16:58 PerlJam tipdbmp: Do you mean grammars in general, or just Perl 6 grammars?
16:58 TimToady real "mathematical expressions" are not strings
16:58 TimToady since mathematicians tend to use 2-dimensional notations
16:58 PerlJam (Perl 6 will hide some details and make things easier for you than you'd get from other tools like yacc or bison)
16:59 tipdbmp Simple math exprs: 1 + 2 * (3 / 4), for now.
17:00 TimToady have a look at http://rosettacode.org/wiki/​Arithmetic_evaluation#Perl_6
17:00 PerlJam tipdbmp: the "best" way to learn about grammars is to try to write a parser from scratch  ;)
17:01 masak tipdbmp: I wrote one for this post: http://strangelyconsistent.org/blog/what-you-c​an-do-with-ggeoptable-that-you-couldnt-without
17:01 masak it's linked from it.
17:01 TimToady (though that example may have bitrotted since it was put there)
17:02 masak tipdbmp:
17:02 masak r: grammar Math { rule TOP { <add> }; rule add { <mul>+ % <addop> }; rule mul { <term>+ % <mulop> }; token addop { <[+-]> }; token mulop { <[*/]> }; token term { \-? \d+ } }; say Math.parse("4 + 2 * 5")
17:02 p6eval rakudo fc349a: OUTPUT«「4 + 2 」␤ add => 「4 + 2 」␤  mul => 「4 」␤   term => 「4」␤  addop => 「+」␤  mul => 「 2 」␤   term => 「2」␤␤»
17:03 masak oops. :/
17:03 masak interesting. I should run this through jnthn's debugger... :)
17:04 grondilu instead of 'max :by( *.&fitness <=> *.&fitness )', I whish I could have written just:  'max :by(*.&fitness)'
17:06 masak er. you can.
17:06 masak :by(&fitness), even.
17:06 grondilu r: say max :by(*.chars), <foo bar long>;
17:06 p6eval rakudo fc349a: OUTPUT«long␤»
17:07 grondilu I thought I had tried that.  Don't know why it failed.
17:08 masak r: say max :by(*.&chars), <foo bar long>
17:08 p6eval rakudo fc349a: OUTPUT«long␤»
17:08 masak r: say max :by(&chars), <foo bar long>
17:08 p6eval rakudo fc349a: OUTPUT«long␤»
17:09 PerlJam grondilu: writing something that utilizes an evolutionary algorithm?
17:09 FROGGS [17:46 ]<grondilu> Check out my version of the evolutionary algorithm:  http://rosettacode.org/wiki/E​volutionary_algorithm#Perl_6
17:09 PerlJam FROGGS++ thanks :)
17:10 * benabik finds the *.foo version clearer than *.&foo or &foo
17:10 grondilu PerlJam: No.  Just having fun writting stuff in RosettaCode
17:11 TimToady this seems inefficient to me, insofar as it's going to recalculate the parent's fitness each time
17:11 grondilu TimToady: yeah I considered using a cache.  But I'm not sure it worths it.
17:11 TimToady I'd put them into pairs, where the key is the fitness, and the value the individual, and the max should work straight on it
17:12 TimToady rn: say (3 => 'foo', 3 => 'bar', 4 => 'long').max
17:12 p6eval rakudo fc349a, niecza v22-29-g3ffe64c: OUTPUT«4 => "long"␤»
17:13 TimToady well, maybe it doesn't matter much inside max
17:14 TimToady it's not like sort, where it's doing a lot of repeated comparisons
17:14 kurahaupo joined #perl6
17:14 TimToady (which does a schwartzian transform anyway, so nevermind)
17:14 * TimToady looks around for more caffeine...
17:15 grondilu it's true that the code needs optimization.  It might be short but it's terribly slow.
17:15 TimToady actually, it's the optimizer that needs the optimization :)
17:16 TimToady one place we should optimize, for instance, is when you write *+*, it just probably replace it with &infix:<+>.
17:17 TimToady and *.&foo should just turn into &foo as appropriate
17:18 grondilu I couldn't use &foo directly because it was inside a binary (it was: *.&foo op *.&bar)
17:18 TimToady perhaps more importantly, we can optimize out all the lazy apparatus when we know the data values don't need it
17:19 benabik Does &foo automatically look for methods?
17:20 TimToady no, but neither does *.&foo
17:20 domidumont joined #perl6
17:23 TimToady re optimizing away laziness, we often speak disparagingly of Perl 5's model of propagating context down the syntax tree, but this is primarily because we don't want the user to worry about it.  It is still, however, very much in the province of the optimizer to propagate such information downward, since vast simplifications are possible when you know your lazy expression is in one of sink, item, eager, hyper, or race context
17:24 masak agreed.
17:24 TimToady [+] foo Zeq bar is obviously putting the Z into an eager context, for instance
17:24 benabik grondilu: You keep using & in IRC, but it's not there in the code since you're calling methods.
17:25 masak I guess the difference is that whereas Perl 5 propagates a kind of context inwards which changes the final return value of the inner thing, what we're talking about here is propagating optimization information.
17:25 grondilu benabik: are you sure you're looking at the most recent version?
17:26 atrodo joined #perl6
17:26 masak which is how most optimizations work, I'd bet: (1) look at the surrounding context, and if it allows some optimization, (2) do it.
17:26 benabik grondilu: I wasn't aware you were refactoring it that aggressively after posting it.  :-D
17:26 ifim joined #perl6
17:28 TimToady note that you can't simply turn a Zeq into a »eq« though, since the semantics are different
17:29 TimToady what you can do with Zeq, however, is have a version that is allowed to examine the lengths of each side first, when you know it's eager
17:29 masak right. eageer things are allowed to look ahead any amount they want.
17:30 TimToady and then it can call into »eq« efficiently for the comparable prefixes, if that's going to be faster
17:30 TimToady or into whatever does fast memory-to-memory comparisons...
17:31 TimToady that is presumably also used by »eq« in those cases where it can be fast
17:32 TimToady especially when comparing two compact arrays (also an optimization that is not done yet)
17:34 TimToady amid all the other important work that is going on, I just don't want to lose sight of the fact that performance is still the main blocker to widespread adoption of Perl 6
17:34 masak and CPAN/Perl 5 integration.
17:34 diakopter both++
17:34 masak can you actually reduce infix:<eq> to a memory comparison? what about combining characters and NFG and stuff?
17:35 diakopter if they're already NFG then the comparison just works
17:36 TimToady if Perl 6 were fast enough, we couldn't stop people from translating Perl 5 CPAN into Perl 6 CPAN in droves
17:36 colomon TimToady++
17:38 grondilu I tried to use Perl6 on rosalind.info.  I couldn't.  It was too slow for most problems.  I had to use Perl5.  So yeah I think performance improvement would be awesome.
17:38 masak I dunno. there's something to be said for just using a CPAN module (well-tested, developed for years) across the p5-p6 chasm rather than having to port it.
17:38 masak if there were excellent p5-p6 integration, so that the p5 half ran reasonably fast, maybe the slowness of p6 would be less of an issue.
17:39 TimToady masak: if two NFG strings are eq, they will have an identical sequence of integers.  if the integers are unequal, and any of them are negative, then you have to do more work.  if none of the integers are negative, you have fast falsification too
17:39 masak TimToady: yeah, that's what I thought.
17:40 TimToady p5-p6 integration is important, but will always impose an overhead that may outweigh the performance benefits of using p5
17:40 arlinius joined #perl6
17:41 doy honestly, a lot of things on cpan could really benefit from the opportunity for a rewrite too
17:41 doy (:
17:41 PerlJam for long running programs, that don't need to shuttle data across the p5/p6 barrier very much, the overhead would be worth it
17:43 TimToady there's also cultural overhead :)
17:43 doy yeah, similar to why swig always produces such terrible results
17:44 tipdbmp What's the $^y variable?
17:44 PerlJam tipdbmp: it's a self-declared variable in the current block
17:44 PerlJam tipdbmp: unless you're asking about perl5, then I have no idea :)
17:44 tipdbmp =)
17:48 sorear good * #perl6
17:48 TimToady o/
17:49 masak sorear! \o/
17:58 masak Perl 5 is great to a large extent because of CPAN. in some sense, it doesn't *matter* if the CPAN module I'm using is up for a rewrite. someone solved the problem, I can download the module and make use of that prior work.
17:59 masak if by $magic-wand Perl 6 would have access to all that today, Perl 6 would be a lot more attractive to a lot of people, I think.
18:00 TimToady yes, but it's the opposite of Einstein's idea; it's a wireless with an actual cat transmitting the signals
18:01 * TimToady thinks in his more arrogant moments that Perl 5 is great because it was designed to allow a CPAN to develop...
18:01 colomon o/, sorear
18:01 nwc10 masak: my assumption (not verified) is that much of the useful stuff on CPAN ends up depending indirectly on something XS on CPAN.
18:02 nwc10 masak: if so, I'm then not sure whether it's going to be easier to rewrite those XS dependencies in Perl 6 than to write a general XS thunking layer that's "good enough" for a sufficiently polished and reliable value of "good enough"
18:02 doy presumably, if it were a p5 bridge (rather than a rewrite of p5 in p6), xs stuff would continue to work
18:03 nwc10 the sort of anti Inline::Rakudo
18:03 nwc10 well, inverse
18:04 nwc10 yes. but still "who knows" on the complexity/cost trade off
18:04 TimToady in addition to transmission overhead, and maybe marshalling overhead, there's the difficulty of reconciling the type systems
18:05 TimToady part of the "cultural overhead" I was referring to earlier
18:05 masak nwc10: that assumption sounds very testable. maybe I should write a blog post.
18:06 TimToady every string would have to be translated between utf8 and nfg
18:06 diakopter (but possibly only once if it can be cached/versioned)
18:06 masak TimToady: there will always be translation. what Tene's work on Parrot interop showed (or what I learned, at least), is that translation is something that the host language has to solve in-situ when talking to the embedded language.
18:07 TimToady assuming we can even guess which p5 strings should turn into Str, and which into Buf
18:07 masak (as opposed to it being a global one-size-fits-all solution)
18:08 TimToady so yeah, the exact API will have to be specified from the p6 side, which can make the type distinctions, and presumably is written by someone who can guess the intent of the p5 code in question
18:09 doy guessing between Str and Buf will almost certainly lead to a whole lot of pain
18:10 diakopter I believe that the bridge is the best option, perhaps with a custom/forked build of perl with the same configure options
18:10 doy probably better to just pick one and let the person writing the p6 side do the translation manually based on what's expected
18:12 TimToady oh, re Stringy in the backlog, the intent is not just that Cat be Stringy, but also Buf.  in fact, Stringy was invented with Buf in mind more than Cat (at least in one mind, senile though it is getting)
18:13 diakopter (same configure options as the perl available when configure of p6 is run, so it can use the same installed libraries)
18:22 * TimToady keeps wondering whether we should steal ^ from exclusive-or and use something else for that
18:23 TimToady how do mathematician's write exclusive-or?
18:23 TimToady *cians
18:24 TimToady ⊻ or ⊕ it would appear
18:24 sorear .u ⊕
18:24 phenny U+2295 CIRCLED PLUS (⊕)
18:25 sorear I've never seen the former
18:25 sorear mostly it's either ⊕ or +
18:25 TimToady infix:<o>   # "one of"  ;)
18:25 sorear not (Texas) functional composition?
18:26 sorear sub infix:<o>($f,$g) { -> \x { f(g(x)) } }
18:26 masak I don't think there's a need for infix:<^>. I hardly ever use &one.
18:27 TimToady well, we don't build in infix:<⚪> either...
18:27 masak maybe just leave infix:<^> unclaimed for people who want to define a nice short operator.
18:27 TimToady or make it do what people expect :)
18:31 TimToady but whatever xor turns into would need a single char form like o, so we can have the related forms like oo and +o
18:32 _jaldhar_ joined #perl6
18:34 * masak looks at the symbols on his keyboard and draws a blank
18:34 spider-mario joined #perl6
18:35 TimToady well, we certainly can't use the +, so I went with the o
18:35 sorear meh, xor is just + on GF(2)
18:36 TimToady GF(2) can overload + however it likes, but it's not Perl(6)  :)
18:36 benabik .u ⚪
18:36 phenny U+26AA MEDIUM WHITE CIRCLE (⚪)
18:37 benabik That has funny shading on my computer that makes it look like a sphere.  :-/
18:39 huf it's a square here :)
18:39 huf fitting.
18:40 diakopter are you seriously considering making xor solely use a non-ascii character?
18:40 PerlJam that's odd, it's a dodecahedron here.
18:41 PerlJam ;-P
18:41 doy 'o' doesn't look very non-ascii to me
18:41 diakopter doy: I guess I wasn't addressing only TimToady...
18:42 snearch joined #perl6
18:44 sorear wait, I think there's something wrong with my display, all my characters are two-dimensional
18:44 sorear no spheres or dodecahedrons, or rhododendrons
18:45 diakopter .u 2331
18:45 phenny U+2331 DIMENSION ORIGIN (⌱)
18:46 diakopter (I have no idea what that looks like)
18:46 diakopter oh.
18:53 TimToady well, as with hypers, we could have ⊕ be the preferred form, and define a Texas version of that; I was thinking of o as the Texas form of ⊕ becuase of things like +⊕, ~⊕, and ⊕⊕ would come out really strange if we used a digraph
18:53 sorenso joined #perl6
18:54 TimToady +o+ or o+o+ just don't work too well
18:58 masak I think I've used +^ in Perl 6 code once.
18:58 masak I've never used ^ or ^^ -- and I don't think I will.
19:02 PerlJam er, why would we take back ^ in the first place?  What's the motivation there?
19:03 benabik joined #perl6
19:03 sjohnson joined #perl6
19:06 dalek perl6-roast-data: 17f2d1e | coke++ | / (4 files):
19:06 dalek perl6-roast-data: today (automated commit)
19:06 dalek perl6-roast-data: review: https://github.com/coke/perl6​-roast-data/commit/17f2d1e289
19:08 risou joined #perl6
19:08 [Coke] rakudo is choking on t/spec/S03-sequence/basic.t
19:14 TimToady the discussion in http://en.wikipedia.org/wiki/Ex​clusive_or#Alternative_symbols is interesting
19:16 TimToady PerlJam: it produces very confusing results if the user expects exponentiation, and is badly huffmanized from the standpoint of frequency of use
19:18 Chillance joined #perl6
19:20 TimToady also, infix ^ has no relation to prefix ^ to preserve
19:22 TimToady so an alphabetic operator becomes a possibility, unlike operators that must also work as prefixes
19:22 PerlJam okay.
19:23 TimToady which is why I was thinking about o
19:23 TimToady oh, another benefit of such a change:
19:24 TimToady reduces the ambiguity with ~^ and such
19:24 TimToady when used as prefix
19:25 TimToady rn: say ^20; say ~^20; say ~ ^20;
19:25 p6eval niecza v22-29-g3ffe64c: OUTPUT«0..^20␤Unhandled exception: Buffer bitops NYI␤  at /home/p6eval/niecza/lib/CORE.setting line 1443 (die @ 5) ␤  at /home/p6eval/niecza/lib/CORE.setting line 3350 (prefix:<~^> @ 4) ␤  at /tmp/fnoDLBIh4P line 1 (mainline @ 5) ␤  at /home/p6eval/niecza/lib/COR…
19:25 p6eval ..rakudo fc349a: OUTPUT«0..^20␤prefix:<~^> NYI␤  in method gist at src/gen/CORE.setting:10149␤  in sub say at src/gen/CORE.setting:7460␤  in block  at /tmp/KnLqQXn0xH:1␤␤»
19:25 TimToady heh, well, potential ambiguity
19:25 TimToady rn: say ^20; say ~ ^20;
19:25 p6eval rakudo fc349a, niecza v22-29-g3ffe64c: OUTPUT«0..^20␤0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19␤»
19:26 TimToady well, actual ambiguity, with a LTA result, I guess
19:27 GlitchMr Shouldn't it be contamination of say function and 0..19 range?
19:27 GlitchMr (lol wrong word)
19:27 GlitchMr concatenation*
19:28 TimToady no, because the ~ is where a prefix is expected, not an infix
19:29 sisar joined #perl6
19:29 PerlJam (context is still king in Perl :-)
19:30 daniel-s__ joined #perl6
19:34 Chillance joined #perl6
19:39 * jnthn back
19:40 sorear yay jnthn
19:40 jnthn shesh, how much backlog while I went to teach folks git... :P
19:40 jnthn *sheesh
19:41 * jnthn sees what he missed
19:41 brrt joined #perl6
19:45 [Coke] Since jnthn is in the room, is S03-sequence/basic.t failing for any other rakudoers?
19:45 cognominal joined #perl6
19:47 jnthn Looks good to be (at ca9c30, anyway)
19:52 rindolf joined #perl6
19:52 jnthn [Coke]: Just built latest and it's also seeming happy.
19:59 kurahaupo joined #perl6
20:01 jnthn oh noes, operator changes /o\
20:02 jnthn Though thankfully to one probably rarely used :)
20:04 masak jnthn: nothing's changed yet :)
20:04 jnthn :)
20:05 jnthn I kinda like the symmetry of &, | and ^ all being one-char non-\w operators.
20:05 masak aye.
20:05 jnthn Though there's precedent for not having ^ in so far as there's nothing for none
20:06 masak Perl 5 does fine without a symbolic (boolean) xor operator.
20:12 kst joined #perl6
20:22 TimToady one must have the bitwise ops though, somehow
20:23 PerlJam If we go by huffman, the bitwise ops are way too short for the amount of use they get
20:24 TimToady we already doubled them from p5 :)
20:24 PerlJam still :)
20:26 ifim joined #perl6
20:28 PerlJam I'd be perfectly happy with awkward-to-type or Texas-sized bitwise ops as the default.  If someone happens to need bitwise ops frequently, they can always use a slang
20:29 jnthn otoh, I doubt we can do anything more sensible with those chars :)
20:30 tipdbmp How can I check what type an object is? class MyClass { } my $c = MyClass.new; # how can I checknow if $c's class is really MyClass?
20:31 [Coke] r: class MyClass {}; my $c = MyClass.new; say $c.WHAT;
20:31 p6eval rakudo fc349a: OUTPUT«MyClass()␤»
20:32 [Coke] r: class MyClass {}; my $c = MyClass.new; say $c.WHAT; say MyClass.WHAT;
20:32 p6eval rakudo fc349a: OUTPUT«MyClass()␤MyClass()␤»
20:32 jnthn r: class MyClass { }; my $c = MyClass.new; say $c ~~ MyClass; my $a = 42; say $a ~~ MyClass;
20:32 p6eval rakudo fc349a: OUTPUT«True␤False␤»
20:32 jnthn Use ~~ for type checking
20:32 jnthn .WHAT will tell you what it is, if you're curious. But for type checking, use ~~
20:32 jnthn (question coulda meant either :))
20:32 tipdbmp tnx =)
20:33 Moukeddar joined #perl6
20:34 sorear o/ Moukeddar
20:35 Moukeddar \o/ sorear :)
20:35 Moukeddar how are you doing, Sir!
20:36 sorear very good
20:36 rindolf Moukeddar: hi.
20:36 rindolf Moukeddar: what does "Moukeddar" mean?
20:37 wamba joined #perl6
20:37 Moukeddar Hola rindolf, it's in arabic, it's my last name, it doesn't have a specific meaning actually ;)
20:38 rindolf Moukeddar: ah, OK.
20:38 Moukeddar it's pretty unique, therefor i use it as a nickname
20:38 rindolf Moukeddar: ah.
20:38 rindolf Moukeddar: where are you from?
20:38 Moukeddar Morocco, you?
20:38 rindolf Moukeddar: unfortunately with this respect my last name is "Fish".
20:38 rindolf Moukeddar: I'm from Israel.
20:39 PerlJam Moukeddar: you know, every time I see your nick, I transliterate the k and the d in my head for some reason.
20:39 rindolf http://www.shlomifish.org/ - but I'm using a Fish emblem here.
20:39 rindolf And elsewhere.
20:40 sorear PerlJam: transpose?
20:40 Moukeddar PerlJam, even people here have problem with my name, i guess it's a curse lol
20:40 sorear rindolf: does fish (he) mean the same as fish (en) ?
20:41 Moukeddar rindolf, nice website ;) , you too people have problems with hebrew support
20:42 rindolf sorear: no, there is no word "Fish" in Hebrew.
20:42 rindolf Moukeddar: Hebrew support is usually not a problem.
20:42 Moukeddar rindolf, talking about RTL ;)
20:42 rindolf Moukeddar: but I prefer that most of my home site will be in English.
20:42 rindolf Moukeddar: yes, I know.
20:43 rindolf Moukeddar: see http://www.shlomifish.org/m​eta/FAQ/#website_in_english .
20:43 rindolf sorear: my last name is not Hebrew. It is German in origin or something.
20:43 rindolf sorear: maybe English.
20:44 rindolf sorear: my grandfather came from Poland.
20:44 tadzik ha, fellow :)
20:44 Moukeddar rindolf, good reasons ;) i could borrow them ;)
20:44 TimToady my ancestors lived in Poland about 5 generations ago
20:44 TimToady on the way from Belgium to Ukraine...
20:45 PerlJam < rindolf> sorear: no, there is no word "Fish" in Hebrew.
20:45 tadzik wow
20:45 PerlJam Do they not eat fish?  Do fish not exist in hebrew?
20:45 geekosaur I think this is use/mention
20:46 sorear PerlJam: rindolf's last name is F-I-S-H
20:46 sorear PerlJam: I wondered if those letters meant something in Hebrew
20:46 benabik PerlJam: Don't feel bad, that was my first thought on reading that sentence too.
20:47 Moukeddar Watch out, #perl6 is turning into Language analytics :)
20:47 geekosaur wasn't it always?
20:47 * PerlJam wonders if he should use more emoticons in his writing on IRC
20:47 sorear they need seafood words, if for no other reason than to explain the rules on what is and isn't edible :p
20:47 tipdbmp joined #perl6
20:47 geekosaur "dag"
20:48 TimToady .oO(and Jonah was swallowed by a great <mumble>)
20:49 PerlJam "large legless water-dwelling animal bigger than an elephant"
20:49 * Moukeddar thinks PerlJam could use some more emotions
20:49 PerlJam Moukeddar: my wife tells me that all the time  :)
20:50 sorear rindolf: interesting favicon you have there
20:51 rindolf PerlJam: the Hebrew name for Fish is "דג" - "Dag".
20:52 [Coke] Yo, Dag.
20:52 PerlJam heh
20:53 PerlJam rindolf: you must be pleased to be so important to the inner workings of git.  :)
20:53 TimToady if a Cat is a list pretending to be a string, maybe a Dog is a string pretending to be a list
20:53 rindolf PerlJam: heh.
20:53 sorear Hai dag
20:53 GlitchMr Why not both?
20:53 sorear </injoke>
20:53 [Coke] TimToady: Sounds Fishy.
20:54 PerlJam TimToady: the PDL folks use cat/dog as opposites too IIRC
20:54 rindolf sorear: in Hebrew F and P share the same letter, and when it is stressed it is "P". And it is stressed at the beginning of a word.
20:54 skids joined #perl6
20:54 rindolf sorear: http://en.wikipedia.org/wiki/Pe_%28letter%29
20:55 geekosaur dag nabbit?
20:55 sorear rindolf: I am not sure what you are replying to
20:55 Moukeddar you guys are hilarious
20:55 rindolf sorear: to the Fish discussion.
20:55 rindolf sorear: in Hebrew, you cannot start a word with "f".
20:56 rindolf sorear: it must start with a "p".
20:56 GlitchMr https://gist.github.com/4068343
20:56 GlitchMr Python is too limited for that...
20:57 GlitchMr But perhaps with complex hacks...
20:57 sorear rindolf: do Hebrew first-languagers call you /piʃ/ ?
21:00 TimToady and is that really [pʰiʃ] as an English speaker would pronounce it?
21:00 rindolf sorear: no, they don't.
21:00 sorear dunno, my skillz don't extend to the ʰ level
21:01 TimToady rindolf: then do they call you "דג"?
21:02 rindolf TimToady: no, they call me "Fish"
21:02 diakopter doubling-- bitwise-- again--
21:02 rindolf Well, my first name is Shlomi.
21:02 kst joined #perl6
21:03 jaldhar joined #perl6
21:04 * geekosaur grumbles, having suddenly noticed that his irc client didn;t rtl properly
21:11 jaldhar joined #perl6
21:12 nebuchadnezzar joined #perl6
21:34 jlaire joined #perl6
21:35 gaussblurinc_ joined #perl6
21:45 tipdbmp Where does one place the regex modifiers? if $str ~~ m/:i something / { "doesn't seem to work" }
21:46 masak rn: say "aBc" ~~ m:i/ AbC /
21:46 p6eval rakudo fc349a: OUTPUT«「aBc」␤␤»
21:46 p6eval ..niecza v22-29-g3ffe64c: OUTPUT«#<match from(0) to(3) text(aBc) pos([].list) named({}.hash)>␤»
21:46 masak rn: say "aBc" ~~ m/ :i AbC /
21:46 p6eval rakudo fc349a: OUTPUT«「aBc」␤␤»
21:46 p6eval ..niecza v22-29-g3ffe64c: OUTPUT«#<match from(0) to(3) text(aBc) pos([].list) named({}.hash)>␤»
21:46 masak either outside or inside works.
21:46 tipdbmp Oh, ok.
21:53 kst joined #perl6
22:13 japhb_ joined #perl6
22:15 japhb_ grondilu++ # Very nice work on http://rosettacode.org/wiki/E​volutionary_algorithm#Perl_6
22:16 japhb_ That feels like the kind of thing that should go in snippet rotation on www.perl6.org.  Um, when one of us makes that work ...
22:16 japhb_ ETOOMANYTASKS
22:17 * jnthn wonders which task is most intstresting... :)
22:18 masak -Ointeresting
22:18 japhb_ TimToady, re: http://irclog.perlgeek.de/​perl6/2012-11-13#i_6149996 , I couldn't agree more.  The top two blockers at my $day_job are definitely performance and breadth of ecosystem.
22:19 ifim joined #perl6
22:23 cognominal joined #perl6
22:25 lawrmcfa joined #perl6
22:26 lawrmcfa joined #perl6
22:28 tipdbmp How can one subtract character classes? Say: <[a..z]> - <[aeiou]>
22:29 jnthn inside the angles
22:29 jnthn <[a..z]-[aeiou]>
22:29 tipdbmp no space between ]-[?
22:29 jnthn Up to you
22:29 tipdbmp Ah, ok. tnx =)
22:30 masak can also do + between the [], and a + or - before the first one.
22:32 sorear but space between < and [ is likely to confuse things
22:32 telex joined #perl6
22:33 jnthn Yes, that's not optional. :)
22:33 jnthn Otherwise it'll get taken as quote words I guess...
22:34 sorear n: say grep /<[0-z]^:Letter>/, <0 a ð …>
22:34 p6eval niecza v22-29-g3ffe64c: OUTPUT«[31m===[0mSORRY![3​1m===[0mâ�¤â�¤Unsupported use of - as character range; in Perl 6 please use .. at /tmp/JTfZjDs1sn line 1:â�¤------> [32msay grep /<[0-z[33mâ��[31m]^:Letter>/, <0 a ð …>[0mâ�¤â�¤Parse failedâ�¤â�¤Â»
22:34 sorear n: say grep /<[0..z]^:Letter>/, <0 a ð …>
22:34 p6eval niecza v22-29-g3ffe64c: OUTPUT«0 ð␤»
22:34 sorear \o/
22:37 masak nice.
22:40 [Coke] can you deparse that for me?
22:41 sorear < [0..z] ^ :Letter > ?
22:41 [Coke] (ah, there's a nice module to have. regex::deparse)
22:42 sorear regex::deparse sounds like it converts a Regex object back to text
22:43 [Coke] so, that's "Any thing that's a letter or in the range 0..z, but not both" ?
22:47 [Coke] (looks like.)
22:48 flussence xor.
22:49 [Coke] the spec calls it by a more set-like name, but yah.
22:51 masak "symmetric set difference"?
22:51 [Coke] Aye.
22:51 masak just a stab in the dark. :P
23:02 lawrmcfa joined #perl6
23:06 wamba joined #perl6
23:19 am0c joined #perl6
23:44 jlaire joined #perl6
23:54 sjohnson joined #perl6
23:56 am0c joined #perl6

| Channels | #perl6 index | Today | | Search | Google Search | Plain-Text | summary

Perl 6 | Reference Documentation | Rakudo | Niecza | Specs