»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!
Set by sorear on 4 February 2011.
00:06 stopbit left 00:19 cognominal left 00:21 buubot_backup joined 00:22 DarthGandalf joined 00:25 cognominal joined 00:30 Chillance left 00:40 buubot_backup left 00:41 buubot_backup joined 00:44 tokuhiro_ joined 00:49 tokuhiro_ left 00:52 anuby joined, am0c joined 00:53 whiteknight left 00:54 benabik left 01:00 immortal joined, immortal left, immortal joined 01:01 erkan left 01:02 atrodo left 01:03 kurahaupo1 joined 01:08 hypolin joined 01:10 benabik joined 01:11 immortal left 01:14 erkan joined, erkan left, erkan joined 01:18 ifim left 01:19 erkan left 01:28 erkan joined 01:30 bowtie left 01:45 kurahaupo1 left 01:48 erkan left
[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.
[Coke]: 11 Nov 10:59Z <isBEKaml> tell [Coke] ok, nvm - I'm just about to give up on pugs+windows. :)
[Coke] phenny: tell isBEKaml Nevermind. ;) 01:53
phenny [Coke]: I'll pass that on when isBEKaml is around.
01:53 fgomez left, erkan joined, erkan left, erkan joined, fgomez joined 01:56 fgomez_ joined 01:57 fgomez left 02:00 kurahaupo joined 02:09 ifim joined 02:21 DarthGandalf left 02:22 DarthGandalf joined 02:24 clkao_ joined 02:32 cognominal left
diakopter [Coke]: queue up replies 'til done logbacking 02:35
02:36 eternaleye left
diakopter later transactions may obviate earlier ones 02:40
02:45 alienswamp joined, alienswamp left 02:48 fgomez_ left 02:49 fgomez joined
dalek rl6-roast-data: 207bbe8 | coke++ | / (3 files):
today (automated commit)
 02:51
rl6-roast-data: 906b9f5 | coke++ | / (3 files):
today (automated commit)
rl6-roast-data: a98362d | coke++ | / (3 files):
today (automated commit)
02:53 cognominal joined
[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.
sorear bah, I should fix that someday if I ever figure out how
phenny sorear: 02:55Z <[Coke]> tell sorear - niecza failed all tests today with "Unhandled exception: System.MissingMethodException: Method not found: 'Niecza.Kernel.CreateHash'."
[Coke] sorear: github.com/coke/perl6-roast-data/b...ummary.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
rm -r ~/.local/share/NieczaModuleCache
dalek rl6-roast-data: 1198e62 | coke++ | bin/niecza.sh:
Remove stale file on every build.
 03:05
[Coke] sorear++ - 03:09
sorear what? %FOO ||= built_foo; doesn't work in p5? 03:12
03:21 DarthGandalf left
doy sorear: sure it does 03:21
sorear doy: Can't modify private hash in logical or assignment (||=) at -e line 1, 03:22
might be because my perl is ancient
03:22 DarthGandalf joined
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
sorear eval: $^V
buubot_backup sorear: bless( {original => "v5.16.0",qv => 1,version => [5,16,0]}, 'version' )
doy sounds like a bug to me 03:24
hmmm 03:25
maybe not
well
at the very least, that's a terrible error message
i think the problem is that || supplies scalar context 03:26
because you can't evaluate the truth value of a list 03:27
and that just causes all kinds of confusion because now you're treating the scalarified temporary version of that hash as an lvalue 03:28
or something 03:29
03:32 Entonian joined
[Coke] sorear: gist.github.com/4063778 - current niecza failures 03:33
03:36 Entonian left 03:43 BaggioKwok joined
colomon has been slacking off on checking niecza spectests 03:44
03:47 BaggioKwok left 03:56 orafu left 03:57 orafu joined 04:07 gootle joined 04:09 isBEKaml joined
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.
isBEKaml: 01:48Z <[Coke]> tell isBEKaml to ping rurban about cygwin parrot - cygwin was his main platform at some point.
isBEKaml: 01:53Z <[Coke]> tell isBEKaml Nevermind. ;)
04:12 fgomez left 04:18 fgomez joined 04:20 tokuhiro_ joined 04:22 DarthGandalf left 04:25 tokuhiro_ left 04:26 DarthGandalf joined 04:28 isBEKaml left 04:42 cognominal left 04:43 sftp left 04:46 cognominal joined 04:49 marloshouse left 04:52 marloshouse joined 04:55 Guest28559 left 05:04 xinming_ joined 05:07 erkan left, xinming left 05:08 marloshouse left, am0c left 05:17 am0c joined, telex left 05:22 DarthGandalf left 05:23 telex joined 05:24 Patterner left 05:28 DarthGandalf joined 05:33 Psyche^ joined, Psyche^ is now known as Patterner 05:41 scottp joined 05:52 erkan joined, erkan left, erkan joined 05:59 daniel-s joined 06:08 birdwindupbird joined 06:13 xinming_ left 06:19 xinming joined 06:26 cognominal left 06:32 cognominal joined 06:36 ifim left 06:48 SamuraiJack joined 06:50 SamuraiJack left 06:53 SamuraiJack joined 06:54 cognominal left 07:00 cognominal joined 07:16 daniel-s_ joined 07:20 topo left, topo joined, daniel-s left 07:21 jlaire left 07:25 kurahaupo left 07:26 jlaire joined 07:35 domidumont joined 07:47 domidumont left 07:56 sorenso joined, tokuhiro_ joined
moritz \o 07:58
08:01 am0c left, tokuhiro_ left
sorear o/ 08:01
08:01 am0c joined 08:03 domidumont joined 08:12 domidumont left 08:13 domidumont joined 08:21 cognominal left, kst left 08:22 kst joined 08:24 fhelmberger joined 08:29 flightrecorder joined 08:34 FROGGS joined 08:36 obra left, clkao_ left 08:38 kresike joined
kresike hello all you happy perl6 people 08:38
FROGGS morning 08:42
sorear heya 08:44
tadzik hello hello
09:05 domidumont left, domidumont joined 09:11 domidumont left 09:15 domidumont joined
dalek ecza: 8864bf6 | sorear++ | lib/ (6 files):
Top removal: SubInfo.ctor, STable.ctor
 09:24
09:25 xenoterracide joined
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' 09:29
tadzik come join the fun :) 09:30
bonsaikitten xenoterracide: so basically python, but more fun? ;)
sorear welcome to #perl6, xenoterracide
xenoterracide bonsaikitten: no I think I hate python more... the (ramble) was about python 09:32
bonsaikitten amusing
tadzik to me python just isn't exciting. It's like writing Java a bit
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...
or he's just being a dick
either way I hate whitespace based... harder to read imo
sorear xenoterracide: is it possible he's an elderly LISP programmer?
xenoterracide tadzik: no
sorear (foo 09:34
(bar
(baz)))
tadzik oh I hate this style :)
xenoterracide sorear: though I think I saw that he does lisp
bonsaikitten tadzik: I prefer boring ;) exciting means someone calls me saturday morning at 6am because the pritner is on fire
xenoterracide so yes
sorear is standard LISP whitespace style
xenoterracide a young lisp programmer
bonsaikitten whitespace is mandatory anyway
so why not make it do something useful
xenoterracide I corrected it with judicious use of perltidy
dalek ecza: 3db2fe1 | sorear++ | lib/ (3 files):
Top removal: inferior runloop system
 09:35
xenoterracide readability is useful
unfortunately I have not seen that python accomplished that
huf hey, python's pretty readable. it's just that almost everything is pretty readable. 09:36
aint nothing special
moritz OTRS source code is not pretty readable.
huf OTRS?
xenoterracide OTR's? 09:37
moritz a bug tracker
sorear I have never used python so I can't say I have an opinion on it
xenoterracide I've only dabbled
moritz it's written in perl 5 with no particularly advanced features
huf sorear: i can read it about as well as any other language i dont know :D
xenoterracide but all the python I've seen is not very good
tadzik I'm doing my uni project in python now, it's okay
huf sorear: not as well as a language i know. surprising :)
sorear huf: yeah, all ALGOL clones read the same
moritz but it's a big codebase, and uses pseudo OO designed by people who do't understand OO
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
moritz and no choice of language would make such code readable 09:38
tadzik mumbles something about scoping
sorear huf: how are you with Haskell or the APL family? :p
arnsholt There are some scoping-related things I don't like, but that's pretty much it
tadzik: Yeah, 's what happens when you don't explicitly declare variables =)
moritz and I can't get my head around join being a string method, not a list method
sorear J is the only language I've touched this year that was non-trivial to learn 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 eternaleye joined
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
bonsaikitten moritz: and since it works for their internal uses why change it 09:40
sorear xenoterracide: p5 p6 is worse
xenoterracide sorear: yeah but I'm not sure I think of them as the same "language" anymore
tadzik I have a feeling that py2k py3k is more like p5.10 p5.12 than p5 p6
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
tadzik I don't think python takes backwards compat seriously
We'll invent Python 3 so we can do backwards-incompatibile changes!
And then they roll out 2.7 which breaks 2.6 code 09:42
xenoterracide lol
bonsaikitten moritz: yes, and it shows the usual lack of discipline of commercial projects
tadzik: py3 is academic wankery, about 2/3rds of all packages are still 2.x only
it's like declaring that the new version of ruby is perl5, and you better like it!
tadzik bonsaikitten: my use case is purely academic wanking, so I'm okay with using py3 ;)
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...
and I didn't want to be gpl
bonsaikitten still all the "big" libs aren't migrated
xenoterracide oh... and then I wanted to checkout sql alchemy
2 only
tadzik bonsaikitten: I just need to write an mbox parser from scratch, I'm not likely to use any libs at all
xenoterracide but the default python on arch is 3? 09:44
bonsaikitten xenoterracide: yes, and you better like it
tadzik well, arch is... adventurous
bonsaikitten xenoterracide: they don't care about "working" anymore
tadzik did they ever?
bonsaikitten yes
xenoterracide actually I generally have no problem with working
outside of this damn network driver
bonsaikitten that's how they got big, now the devs are just ... wtf ... users are not idiots you can yell at for fun
xenoterracide which is entirely lkml
tadzik I think they got big by the "simplicity" marketing, they never had really mature and responsible maintainers 09:45
not in the last 5(?) years at least
arnsholt So speaking of Arch, what's a good distro these days? 09:46
tadzik debian's okay I guess
xenoterracide it's better than what I have to deal with when I've used fedora or ubuntu
I'd rather arch any day
tadzik I don't like russian rulette on updates
arnsholt I suppose I can live with Debian. I never warmed to apt and friends though
xenoterracide though I haven't reinstalled since they went all manual
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
tadzik arnsholt: well, you don't use apt to often, do you :)
bonsaikitten FROGGS: I disagree
tadzik you install stuff with it and forget about it, not that you use it every 10 minutes
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
xenoterracide I don't like the idea of doing a painful upgrade every 6 months
so I stay away from release distro's 09:48
bonsaikitten s/upgrade/reinstall/
upgrade only works on "new" unmodified installs, half the time
tadzik ah, the "rolling release" hype
bonsaikitten tadzik: it's the only way to avoid stupid reinstalling
arnsholt tadzik: Before trying apt I was used to portage, which on the whole was a lot simpler to work with
moritz "trolling release"
xenoterracide hey it works for me
arnsholt (Even if compiling stuff from source takes forever)
tadzik bonsaikitten: I don't remember ever reainstalling a debian installation
xenoterracide arnsholt: um... arch is not binary
err..
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
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.
FROGGS: I've never seen plymouth work properly
arnsholt xenoterracide: Portage is Gentoo
tadzik I prefer working on my OS, not working "on my os", if you know what I mean
xenoterracide arnsholt: I know
bonsaikitten I call it "segfault generator"
arnsholt: and that's an advantage! ;)
xenoterracide arnsholt: I used it for more than half a decade
bonsaikitten tadzik: I hate compiling, that's why I use Gentoo 09:50
sorear what is plymouth?
tadzik I don't see the logic, but I don't think that matters :)
xenoterracide idk, can't stand "release" and "repository" distro's that's my whole thing... give me a rolling distro with centralized repositories
bonsaikitten sorear: an attempt at a fancy bootsplash
tadzik my point is: the distro is my tool to work with, not something I need to work on and take care of
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
sorear xenoterracide: can you stand cpanminus?
tadzik bonsaikitten: never happened to me
sorear oh, misread
xenoterracide sorear: that's what I use
bonsaikitten tadzik: well, their nginx is broken, bash used to be crippled, etc.
tadzik I don't mind old (kernel|bash|whatever), I'm not a (kernel|bash|whatever) developer
bonsaikitten I dislike segfaults and scripts not working 09:52
tadzik ok, maybe. I never had any problems
moritz I hate bashs too old to prog1 |& less
tadzik that's what everyone says though, about systems they use
bonsaikitten and when upstream tells me "stop using that old shit, man" then, well, I have to fix things
tadzik do you happen to be using Debian Stale?
xenoterracide is there any other?
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
bonsaikitten such a severe lack of any idea of QA :(
xenoterracide and cent
tadzik xenoterracide: there are 2 other
xenoterracide: I'm running testing, and I'm getting 10+ updates a day or so
xenoterracide sounds like fun
tadzik there's also unstable for the adventurers
labeled as "yes, this may break" 09:54
bonsaikitten too old ;) I'm using KDE 4.9.80 at the moment
huf debian sid doesnt break all that often actually
once every handful of months at most, in my experience
xenoterracide oh I'm only running 4.9.3
yeah I've not had many problems with arch breaking, and usually when they do it they tell me 09:55
sorear 10!
dalek ecza: a2d37f7 | sorear++ | lib/ (2 files):
Misc top removal 2
xenoterracide very rarely have I had a pants down moent
moment*
bonsaikitten I have sacrificial machines that are there to catch errors 09:56
makes life a lot easier :)
xenoterracide I don't have that, but I have lots of backups 09:57
tadzik I can't stand updates breaking vim
xenoterracide the worst pants down in the last year was a failing hard disk
and that's not the OS's fault
I haven't had updates break vim? 09:58
tadzik I did, back in Arch days
but then again, not much point in rambling about it now :)
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
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?
tadzik xenoterracide: No idea. I only cared since it broke me dev environment 10:01
as a user I don't even want to know what they're doing and why
xenoterracide I remember when I started using arch vim was pretty fucked as to how they implemented it differently from everyone else
sorear how does one implement vim differently?
xenoterracide so I spanked some people until they stopped using an example config as the default
sorear it's off the shelf software... 10:02
xenoterracide sorear: you use the "example" config as a default system config
which vim states do not do
moritz that's a matter of distribution, not implementation
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
like vim from cent to suse to debian is slightly different
10:03 hypolin left
xenoterracide but back then in arch it was so different it confused even experienced vim users 10:04
the trick was stop using a sample config and let vim have its own defaults
at some point they also switched vi to an actual vi implentation
no idea why
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
damn bootsplashes
xenoterracide sorear: well there's that
sorear rc.d must always give a process log
bonsaikitten sorear: you sound like a gentoo user ;)
xenoterracide I prefer it 10:06
the process log
sorear bonsaikitten: I have never used gentoo. my first linux was Debian woody, and it is an improvement on everything I have used since
xenoterracide though I liked gentoo's process log on pretty background best
bonsaikitten sorear: hmm, I think you'd complain about it
but it doesn't force silly defaults on you - 30MB mem use after boot on a "default" install 10:07
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
still that's pretty good for what's available now
10:08 mtymula joined
sorear I will have to give it a try next time I'm on Linux 10:08
10:08 am0c left
bonsaikitten sorear: kernel needs about 10MB already 10:08
mtymula HELLo!! 10:09
bonsaikitten 4MB doesn't sounds realistic for anything modern, unless you totally strip it down and customize it to hell
moritz hello mtymula
jnthn morning, #perl6
moritz bonsaikitten: debian woody is anything but modern :-)
\o jnthn
mtymula mortiz do you have any opinion about perl6maven.com/tutorial/perl6-introduction 10:10
bonsaikitten moritz: I prefer things where I can use upstream bugtrackers
mtymula i am trying to lern perl6...
moritz mtymula: it's good
tadzik hello mtymula
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 :)
10:11 xinming left, fgomez left, gootle left, mikemol left, daxim left, araujo left, nyuszika7h left, flussence left 10:12 _ilbot left, c1sung left, sjohnson left, avar left, huf left, cibs left, mtymula left, p6eval left, [Coke] left, sizz left, bapa left, lue left, BillySeth left, pmichaud left, odoacre left, rjbs left
tadzik not if you leave :/ 10:12
sorear tadzik: it's called a netsplit
it's not eir fault
tadzik ah, netsplit
sorear 02:00 [freenode] -mrmist(~mrmist@freenode/staff/amazing.groupcat)- [GLOBAL NOTICE] - In a few minutes there will be some network disruption whilst we sort out some internal routing issues. It shouldn't take too long, but could be a bit noisy. Thank you for your patience.
tadzik indeed
10:15 mtymula joined, xinming joined, fgomez joined, gootle joined, mikemol joined, daxim joined, sizz joined, araujo joined, nyuszika7h joined, bapa joined, lue joined, flussence joined, huf joined, avar joined, sjohnson joined, c1sung joined, _ilbot joined, cibs joined, BillySeth joined, p6eval joined, [Coke] joined, pmichaud joined, odoacre joined, mj41 joined, prammer joined, rjbs joined, Grrrr joined, Yappocall_ joined, crazedpsyc joined, groky joined, rajaniemi.freenode.net sets mode: +v p6eval, flightrecorder left 10:16 flightrecorder joined
tadzik mtymula: perl6.org/documentation/ covers about all the best things we have 10:16
xenoterracide reads over this tutorial and thinks this is very strange multiline comment
sorear #`()?
10:17 nyuszika7h left
bonsaikitten tadzik: bzImage is compressed, it'll still eat quite a bit of RAM when it actually boots 10:17
sadly on amd64 it appears impossible to get below 3MB without losing functionality
sorear also, that's just the text 10:18
xenoterracide sorear: yeah, I find it a bit odd
the example give is actualy #`(())
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?
sorear yes, it's spelled panda
modules.perl6.org
xenoterracide ah
10:20 PerlJam left, SamuraiJack left, buubot_backup left, ironcamel left, Gothmog_ left, hugme left, Util left, dakkar left, eternaleye left, anuby left 10:22 dakkar joined, eternaleye joined, SamuraiJack joined, anuby joined, buubot_backup joined, ironcamel joined, PerlJam joined, Gothmog_ joined, hugme joined, Util joined, larks joined, doy joined, `patch` joined, charsbar_____ joined, rhr joined, Tene joined, jfried joined, Ulti joined, lestrrat joined, pjcj joined, perigrin joined, rking joined, rajaniemi.freenode.net sets mode: +v hugme, flightrecorder left, flightrecorder joined
sorear mtymula: hi! 10:22
10:23 aloha left, _jaldhar_ left, [particle]1 left, Vlavv left, rurban left 10:24 aloha joined, _jaldhar_ joined, [particle]1 joined, Vlavv joined, rurban joined, twinshadow joined, rummik joined 10:25 nyuszika7h joined, dakkar left, eternaleye left, anuby left, jfried left, lestrrat left, Ulti left
tadzik bonsaikitten: well, do you really need it below 3MB on amd64? 10:25
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
bonsaikitten tadzik: "need" - no, my smallest boot devices are still >>32MB, but the constant increase in needs is annoying
10:26 SamuraiJack left, buubot_backup left, ironcamel left, Gothmog_ left, hugme left, Util left, larks left, doy left, pjcj left, Tene left, rhr left, charsbar_____ left, `patch` left, dakkar joined, eternaleye joined, anuby joined, jfried joined, Ulti joined, lestrrat joined 10:27 SamuraiJack joined, buubot_backup joined, ironcamel joined, Gothmog_ joined, hugme joined, Util joined, larks joined, doy joined, pjcj joined, Tene joined, rhr joined, charsbar_____ joined, `patch` joined, rajaniemi.freenode.net sets mode: +v hugme, flightrecorder left, flightrecorder joined 10:28 ggoebel left, sorear left, tadzik left, dalek left, SmokeMachine left
xenoterracide gods the nqp repo is huge... 10:28
10:28 ggoebel joined, sorear joined, tadzik joined, dalek joined, SmokeMachine joined, Hor|zon joined, quietfanatic joined, szbalint joined, shachaf joined, smash joined, BooK joined, rajaniemi.freenode.net sets mode: +v dalek, flightrecorder left 10:29 flightrecorder joined
xenoterracide wow parrots impressively even bigger 10:29
10:30 broquaint left, snarkyboojum left, Krunch left, Krunch joined, broquaint joined, snarkyboojum joined, steven joined
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
;(
*crushes
sorear dwimperl, was that Gabor's?
moritz yes 10:34
see szabgab.com/contact.html for contact options
dalek ecza: f2586f9 | sorear++ | lib/ (2 files):
Separate responsibilities of NewTypedScalar / NewAnyMuScalar
mtymula are there any other options or only dwimperl ??
jnthn github.com/rakudo/star/downloads has an MSI
Includes Rakudo Perl 6 compiler, some modules, etc. 10:35
Debugger :)
mtymula in a fancy nice GUI style or just command line?? 10:36
moritz fancy nice command line :-) 10:37
mtymula i would be great to have sth like eclipse... :p
10:38 domidumont left
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...
moritz xenoterracide: great for development, horrible for production
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
because I haven't gotten around to trying node
sorear nodejs.org/api/modules.html#modules_all_together 10:40
documentation says the rumor is false
(node's documentation is not trustworthy)
mtymula ok, need to restart the system... will be back to add some more stupid questions;p
dalek ecza: 3ffe64c | sorear++ | lib/ (2 files):
Top removal: NewMuAnyScalar
 10:41
xenoterracide sorear: maybe I didn't understand what they were talking about... or they didn't... 10:43
xenoterracide stress tests my system by compiling and checking for dupes at the same time 10:44
sorear sleep&
xenoterracide pretends UI should be responsive... yes?
10:49 domidumont joined
xenoterracide make && make test ran ok 10:56
but where would make install have put this stuff
it didn't end up in my path 10:57
oh maybe it just put it in ./install 10:58
10:59 mtymula left 11:07 gootle left, rindolf joined 11:08 domidumont left, gootle joined, domidumont joined 11:10 gootle left 11:17 marloshouse joined 11:18 anuby left 11:31 cognominal joined 11:32 cognominal left 11:33 cognominal joined, tokuhiro_ joined, mtymula joined
masak g'day, #perl6 11:33
masak has a Perl 6 day today o/ 11:34
11:34 marloshouse left
mtymula good for you:) 11:34
moritz still haz no internez at $HOME :(
masak moritz: :/ 11:35
mtymula here you have a guestion... why my *.pl file with utf-8 coding does not fire when i make perl6 *.pl 11:36
moritz well, if you write perl6 *.pl and there are multiple of them, it will only launch the first 11:37
11:37 tokuhiro_ left
moritz and take the other files names are command line arguments to the first 11:37
mtymula no... :) i just wrote it here
let's say it is test.pl 11:38
test.pl has utf-8 coding
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. :/
mtymula i get ===SORRY!=== Confused at line 1, near "\ufeffuse v6;\r\n" 11:39
jnthn hah, yes.
\ufeff
mtymula what shall i do??
jnthn Get rid of the \ufeff at the start of the file. 11:40
mtymula come on... there is no such thing
the whole file is: 11:41
moritz yes there is there
mtymula use v6; my $liczba = prompt "Podaj liczbe od 41 do 49 włacznie: ";
moritz your editor just doesn't show you
jnthn It's in the output you pasted...and what moritz said.
moritz use a hex editor to confirm, if you want
and a better text editor
mtymula well i use notepad ++
jnthn mtymula: Me too 11:42
Oh...
On the encoding menu select "UTF-8 without BOM"
jnthn thought that was the default...
mtymula <hahaha> 11:43
ok
it fiered
but
in the output my polish charaters are displayed incorrectly 11:44
ł
instead of them i see some kind of forest....
moritz maybe your console/shell doesn't understand UTF-8?
r: say "włacznie" 11:45
p6eval rakudo 72a27f: OUTPUT«włacznie␤»
tadzik włącznie
moritz phenny: "włacznie"?
phenny moritz: "inclusive" (pl to en, translate.google.com)
tadzik that's incorrect
it's włącznie, not włacznie
mtymula :)
tadzik how do you know that?? 11:46
;p
tadzik mtymula: z doświadczenia życiowego ;)
mtymula pozdrowienia, jesteśmy wszędzie:)
tadzik yep, there's at least few of us here
Woodi, Gli<TAB><TAB>
sergot too 11:47
mtymula so oh... nice:)
11:48 brrt joined
mtymula are you located in poland or somewhere else?? 11:48
11:48 marloshouse joined
jnthn cmd.exe does pretty awfully at utf-8 11:49
tadzik mtymula: Warsaw
jnthn chcp 65001 # sorta helps
mtymula cmd.exe it is 11:50
tadzik: me Rzeszow
tadzik nice
mtymula i was here several times (meaning on this IRC) 11:51
and every time i came accross something that blows my mind:) 11:52
tadzik hehe
mtymula i really start loving this perl community:)
tadzik yeah, it's awesome
masak utf-8 files with a BOM has to be one of the stupidest things ever. 11:53
mtymula loving << not quite English...
masak whatever you say... i have no idea
11:54 grondilu joined
mtymula to be honest i am new at this all stuff especially perl6 11:54
grondilu r: say <foo bar longstring>.max( *.chars <=> *.chars );'
p6eval rakudo 72a27f: OUTPUT«===SORRY!===␤Unable to parse expression in quote:sym<apos>; couldn't find final "'" at line 2, near ""␤»
FROGGS masak: can you determine the endianess withouth the bom?
grondilu r: say <foo bar longstring>.max( *.chars <=> *.chars );
p6eval rakudo 72a27f: OUTPUT«longstring␤»
11:54 pmurias joined
grondilu whishes he could just write: say <foo bar longstring>.max( *.chars );' 11:55
masak FROGGS: yes. see en.wikipedia.org/wiki/Byte_order_mark#UTF-8
11:55 tokuhiro_ joined
FROGGS ahh I see 11:55
masak I wonder why we shouldn't simply strip the BOM in Rakudo if we find it.
FROGGS stupid thing it is then
masak would make for fewer surprised first-time users.
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
masak that's what I'm suggesting.
sorry, I have an FP background; "strip" doesn't mean "modify original" to me :P 11:57
jnthn Yes, we could diddle the grammar.
mtymula thx for help:)
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.
just not "Confused" :(
masak submits rakudobug
jnthn masak: Confused is how I feel about utf-8 having a BOM :P
masak :P
jnthn I suggest we make things just work.
masak me too. 11:59
mtymula ;)
FROGGS "===SORRY!=== Please remove that stupid BOM at line 1 at ..." ?
DWIM++
jnthn Should be a one-line patch, but I'm currently writing slides for a talk this evening :P
mtymula ===SORRY!=== Rakudo is confused please don't do that..."
FROGGS btw, I made a pull request ages ago about me beeing a rakudo release-manager... I guess I still want to do it 12:00
masak weighing in on the discussion about "readable Python". any language can be written in an obfuscated manner. p-nand-q.com/python/obfuscated_python.html -- most encouragement for "writing things well" must come from the community. 12:03
12:04 pmurias left
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
i.e., no cpan module comes without a test suite
or very few, maybe
hardly any python modules come with decent tests, or documentation 12:05
masak FROGGS: cool!
FROGGS: for November?
FROGGS right
12:05 pmurias joined
masak tadzik: is it OK if I move you to February? 12:06
tadzik masak: sure
masak makes it so
moritz \o/ new release managers
mtymula tadzik i see that pmurias i also... ONE OF US...;pp
masak yay, instead of 0 planned release slots, we now have 4 :)
tadzik right :)
FROGGS k, so I'll try to bundle everything these days to be prepared on 22nd
there is a msi-howto these days?
masak mtymula: ja tez mowie po polsku :) 12:07
(hm, that's probably not exactly right...)
mtymula prove it ;p 12:08
dalek kudo/nom: 0aca028 | masak++ | docs/release_guide.pod:
[docs/release_guide.pod] FROGGS++ November release

Bumped tadzik forward to February.		 12:09
mtymula masak: udowodnij;p
tadzik last time I checked masak distinguished tą and tę correctly :)
mtymula oh!!
masak oh, 's easy.
masak says "OONNGG", "EENNGG" 12:10
oh, you mean grammatically?
tadzik right
mtymula so read this: stół z powyłamywanymi nogami
tadzik also, "last time I checked" was what, 1.5 year ago? ;)
mtymula wtf... i see I don't know perl6 more than i suspected... 12:13
i wrote my $kwadrat_roznicy = $robocza^2;
where $kwadrat_roznicy had value of 8
masak tadzik: yeah, almost.
jnthn Raise to power is **
mtymula and i surely did't suspected that i t will be one(8, 2) 12:14
masak mtymula: ^ is a junctive operator.
mtymula what does it mean... never mind i will find out
12:15 pmurias left
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
masak mtymula: you can do things like this: 'if $answer == 2 | 3 | 5'
mtymula masak: try chsząszcz brzmi w trzcinie
masak those are junctions. they allow you to do several similar computations at the same time.
mtymula: did you mean "chrząszcz"? :) 12:16
mtymula masak: thats nice
masak: yea you got it
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 ??
masak "quest", "look for"? 12:18
mtymula google translate, huh?
;p
masak oh yes.
what tipped you off?
masak is curious
arnsholt That's kind of similar to "seek".
mtymula hm... 12:19
arnsholt wonders what sounds changes happened in Slavic
masak arnsholt: I was thinking the same.
mtymula a deeeeeelay and two meanings;p
arnsholt To the textbooks!
masak mtymula: the delay will be taken care of once people find a way to implant Google Translate into my brain.
mtymula and any one of you know what does it mean i n Czech republic??
masak mtymula: something naughty, I'll bet.
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
gooood:) you won pride
Woodi afternoon ppls
masak .oO( she said she'll go look for him, and she's been gone for a long time... )
mtymula ^^ 12:21
masak Woodi! \o/
Woodi: we're having a Polish moment here.
join us.
12:21 pmurias joined
Woodi :) 12:21
mtymula indeed
Woodi I am 3 more screens from here... sec...
mtymula and it is my foult 12:22
masak hey, don't go taking all the blame there! :P
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. :)
masak likes it what the BOM is spelled "feff" -- that summarizes his feelings for it
mtymula ok, I will not
jnthn masak: *lol*...never noticed that :)
huf mtymula: the channel is logged publicly on the web :)
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)
mtymula good one
huf mtymula: when the revolution comes, you wont be able to deny being polish :)
arnsholt Might be a German loan
mtymula what revolution ?? 12:24
huf mtymula: oh you know, the next one, whatever it will be about.
masak Perl 6: a predominantly Polish project... 12:25
huf why, do you want to hide your polishicity?
mtymula beside that why would i like to deny being polish
huf "Perl 6: Polished"
could be a tv series
masak arnsholt: cool.
mtymula and with two meanings 12:26
nice
ok since we came into poland tell me who want's to come to Rzeszów, dring some beer;p 12:27
tadzik mumbles something about Polish Perl Workshop 12:28
jnthn ooh, beer!
12:29 pmurias left, brrt left
masak Polish beer! \o/ 12:29
jnthn I'd go to Poland and drink beer :)
Even with the flight factored in it's probably cheaper than drinking here in Sweden :P
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... :)
moritz jnthn: for your next Perl 6 introduction talk, write a script that calculates how much... what masak said
huf why calculate? experiment!
jnthn hah!
moritz masak: don't be so creepy and think the same things I do :-)
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
moritz nothing wrong with experiments; science is based on them 12:32
huf yes, "science" mythbusters style :)
mtymula indeed
huf in this case
moritz did what the scientists call "computer experiments"
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
tadzik jnthn: yeah, we'll think about it ;) 12:33
mtymula or drink random beer at random places??
moritz masak: given that 99% of our total communication volume is not on IRC, that should happen automatically
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
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...
Woodi ehm... this was few centuries ago... do good beer stil left here ? :) 12:36
mtymula tuff << nqe
yes Woodi, there are lots of it
aspeically in when someone is making it for himself;p 12:37
by himself
or herself
whatever 12:38
Woodi heh :) /me just do nalewki :)
mtymula księżycówki??
you can lite it up wuith fire??
*with
Woodi hmm, no, probably... just fruits and spirit... 12:39
12:39 Patterner left, Psyche^ is now known as Patterner
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
Woodi ok, speeking about method modifiers eg. before - I have no idea how callsame replace it in deterministic way... 12:41
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␤»
masak like that.
Woodi before and after are clear but method call dispatch and find right method. then callsame find next same method...
jnthn same here means "same arguments"
masak right.
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 ?
naive view of before and after is they are wrappers. inheritance not needed... 12:44
12:47 masak left, masak joined 12:48 cosimo joined
Woodi anybody here ? 12:54
mtymula ofc
tadzik yes
rindolf Woodi: hi.
Woodi just probably me not grasping things again :) 12:55
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
Woodi: in the case of multi dispatch, it goes to the next-looser candidate
Woodi moritz: ok. that start my problem with it :) 12:57
12:57 lumi_ left
Woodi it find first. what it if finds other one and later that "first" ? it is not clear comparing to "before" etc 12:57
12:57 lumi_ joined
moritz doesn't quite understand the question 12:58
Woodi dispach is like magic box
moritz ah, it's not :-)
Woodi with inheritance it is clear
moritz I think S06 has something on how dispatch works
basically it sorts the signatures by narrowness 12:59
Woodi k :)
but signature is same..
moritz in which example?
Woodi < jnthn> same here means "same arguments"
moritz arguments != signature
sub f(Int $x) { say $x}; f 5; 13:00
(Int $x) here is a signature
and 5 is an argument
jnthn The way to understand it is that whenever you make a call, there are a set of possible things that could be called for those arguments.
By default you just call the best one
The [next|call][with|same] just let you pass contorl to the next best one.
It's like iterating through a list of possible things.
Woodi it's non-quite-determistic for me so far.. 13:01
moritz iterating a list is quite deterministic
Woodi: if S06 doesn't help you, come up with some examples, and we can explain why candidates are sorted the way they are 13:04
jnthn The list doesn't mutate with time, if that helps.
Woodi k, will read in sec. but "callsame" means call another with same signature. but defining function with same signature (with absence of inheritance) is error for me 13:06
13:06 sftp joined
moritz no 13:06
Woodi is shocked here :)
moritz callsame means call another subroutine with the same PARAMATERS
not with the same signature
you need to clearly distinguish signatures and paramaters, otherwise you're lost 13:07
Woodi ahh... Int -> Numeric -> Any... something like this ?
jnthn class A { method m($a) { say "in A with $a" } }; class B is A { method m($x) { say "in B with $x"; nextsame; } }; B.m
r: class A { method m($a) { say "in A with $a" } }; class B is A { method m($x) { say "in B with $x"; nextsame; } }; B.m
p6eval rakudo 72a27f: OUTPUT«Not enough positional parameters passed; got 1 but expected 2␤ in method m at /tmp/kyFlgCdmSa:1␤ in block at /tmp/kyFlgCdmSa:1␤␤»
jnthn r: class A { method m($a) { say "in A with $a" } }; class B is A { method m($x) { say "in B with $x"; nextsame; } }; B.m(42)
p6eval rakudo 72a27f: OUTPUT«in B with 42␤in A with 42␤»
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 :)
13:08 mikemol left
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 rindolf left
p6eval rakudo 0aca02: OUTPUT«in f(Int) with 42␤In f(Cool) with 42␤» 13:09
jnthn Woodi: Yes, and nextsame took you do the next thing up the inheritance tree. 13:10
*to
Woodi ok, so reading and learning is essential :)
masak news at 11.
13:13 krunen joined 13:17 avar left, avar joined, avar left, avar joined 13:19 timotimo joined, colomon left 13:21 groky left 13:23 groky joined, bbkr left, bbkr__ joined
Woodi just found "proto" which looks like begin&after :) 13:25
moritz wants to send a present to whoever invented git reflog 13:36
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 :-)
most of the time, at least 13:39
it's so good to see that even some apparently destructive operations can be undone in git 13:40
13:41 colomon joined
shachaf git reflog is the future 13:42
(Well, the past. Whatever.)
jnthn It's a window back to the past's alternative futures. :) 13:43
shachaf Traversals are the future, though! 13:44
13:45 Gothmog_ left
moritz looks into the future and sees... errors 13:48
mtymula :) 13:57
13:57 tokuhiro_ left 13:58 tokuhiro_ joined 13:59 benabik left, yves__ left, benabik joined, tokuhiro_ left 14:00 tokuhiro_ joined 14:04 tokuhiro_ left 14:05 PacoAir joined 14:06 tokuhiro_ joined, tokuhiro_ left, tokuhiro_ joined 14:08 atrodo joined 14:11 tokuhiro_ left, FROGGS left, flightrecorder left 14:13 flightrecorder joined 14:20 benabik left 14:24 FROGGS joined 14:30 yves_ joined 14:32 bowtie joined 14:33 GlitchMr joined 14:34 isBEKaml joined
dalek ast: 7adc0af | masak++ | S02-lexical-conventions/bom.t:
[S02-lexical-conventions/bom.t] new .t file

Tests that an implementation can handle/ignore a BOM at the start of the file.		 14:37
ast: 42e1808 | masak++ | .gitignore:
[.gitignore] ignore S32-io/server-ready-flag
isBEKaml we lost a bom! 14:39
dalek kudo/nom: fc349a9 | masak++ | / (3 files):
ignore a BOM at the beginning of a file

Now when somebody sets up us it, all their base will be belong to Rakudo.
masak that surely was easy. 14:40
masak goes on to close the RT ticket
FROGGS masak: but it just ignores it when it is at the beginning of a file, right? 14:41
geekosaur bom be bomp
masak FROGGS: right. 14:42
FROGGS good, thanks ;o)
masak beginning of a compilation unit, even. so you can put it at the start of your &eval strings, too.
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␤»
..niecza v22-29-g3ffe64c: OUTPUT«["foo", "foo", "foo", "foo", "foo"].list␤foo foo foo foo foo␤»
isBEKaml Oh, something.perl now gives the type info too? 14:50
14:51 isBEKaml left 14:52 rindolf joined
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
erm, meant isBEKaml, sorry
Woodi r: sub foo() { say "Foo!" }; foo 14:58
p6eval rakudo 0aca02: OUTPUT«Foo!␤»
Woodi f: sub detect-foo() { syslog("WARNING! Foo fires at " ~ Time::HiRes::now() ); }
f: sub-detect-detect-foo() { ... }
how do this with (next|call)same 14:59
?
moritz you don't; there's no call to a function of the same name in your example
Woodi f sub detect-foo() before foo() { ... } 15:00
geekosaur wrap()?
mtymula is it something in the air or programming in perl6 is so fkng gooooooood:)
??
Woodi mtymula: no idea :)
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 )
mtymula but to be honest it is easier to start writing code than in c# or java... at least for me:)
moritz Woodi: something like that
Woodi ok :) so no idea why callsame & co are mentioned in before|after context 15:03
15:04 stopbit joined
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 :) 15:07
and when you add stable|testing|sid difference it is like debian was realy general-something 15:08
15:08 sorenso left
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 15:09
now, when back on debian I quite miss 1 or 2 erratas by year :) 15:10
15:11 kaare_ joined 15:14 obra joined
mtymula someone said before that uses notepad++ 15:27
do anyone has a knowladge about plugin to make syntax of perl6 look nice in this editor??
Woodi just saw wrap example use nextsame... 15:28
mtymula: perl6 syntax is hard thing
tadzik mtymula: one '?' is usually sufficient to make a question :)
mtymula: something for Perl 5 may work okay
personally I prefer to have no syntax hilighting, esp. for Perl 6 15:29
mtymula ?? < work related. It is just how i do it. I just cannot resist to make it double... 15:30
tadzik: why?
;p
tadzik I tried it for a week and now syntax hilighted code looks awful for me :) 15:31
are you working for Gadu-Gadu perhaps? ;)
Woodi tadzik: maybe other "theme" can help :)
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
Woodi mtymula: it's not secure :)
mtymula what?? GG?? 15:33
;p
Woodi or some encryption is now implemented ?
however emails have same lvl of security :)
mtymula no, it is not secure.
but great for fast contact
Woodi right
tadzik emails are usually ssl'd between you and the mail server 15:34
mtymula because, and i hate that fact, everyone has it
tadzik you don't go with a phone/laptop to a shopping mall, open wireshark and read people's emails
*cough*, yeah
Woodi but server <-> server it is clear text
tadzik no. It's not guaranteed to be something more than clear text.
same with xmpp
15:38 bapa left, bapa joined 15:39 brrt joined
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 left, brrt joined
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) 15:42
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
p6eval rakudo fc349a: OUTPUT«bar␤foo␤»
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 )
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␤»
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␤»
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...
arnsholt You need to explicitly state where to invoke the wrapped sub
Since the compiler can't automatically infer what goes before and what goes after 15:49
Woodi yes, but something like "callwrapped" would be natural. but here is used more general mechanism probably 15:50
masak "callwrapped" seems a weird break on the current nomenclature. 15:55
given that the other two variants are "-same" and "-with".
16:09 grondilu left 16:17 benabik joined
kresike bye folks 16:17
16:17 kresike left, benabik left, benabik joined
PerlJam Woodi: then use "callsame" instead of "nextsame" (but I see what you're saying wrt feeling weird) 16:21
16:22 MayDaniel joined 16:29 spanner joined 16:32 fgomez left 16:38 sisar joined
sisar o/ 16:38
masak sisar! \o/ 16:39
16:39 mikemol joined
masak mikemol! \o/ 16:40
mikemol o/
PerlJam r: sub foo { say "foo" }; &foo.wrap(sub { say "bar" });
p6eval rakudo fc349a: ( no output )
PerlJam Hmm 16:41
TimToady welcome back shortci^W<window closed>
masak :P
sisar my @a = 1, 2, 3, 4; say sqrt(@a); #how do I make sqrt apply to all the elements of @a?
rn: my @a = 1, 2, 3, 4; say sqrt(@a); #how do I make sqrt apply to all the elements of @a?
masak sisar: a for loop? :)
p6eval rakudo fc349a, niecza v22-29-g3ffe64c: OUTPUT«2␤»
masak rn: my @a = 1..4; say @a>>.sqrt
p6eval rakudo fc349a: OUTPUT«1 1.4142135623731 1.73205080756888 2␤» 16:42
..niecza v22-29-g3ffe64c: OUTPUT«1 1.4142135623730951 1.7320508075688772 2␤»
masak rn: my @a = 1..4; say (sqrt $_ for @a)
p6eval rakudo fc349a: OUTPUT«1 1.4142135623731 1.73205080756888 2␤»
..niecza v22-29-g3ffe64c: OUTPUT«1 1.4142135623730951 1.7320508075688772 2␤»
sisar masak: yeah, ithink i was looking for the >> thingy
masak people always look for the >> thingy.
sisar what is it called ?
masak hyper method call.
sisar masak: thanks. 16:43
16:43 fgomez joined 16:45 grondilu joined
grondilu Check out my version of the evolutionary algorithm: rosettacode.org/wiki/Evolutionary_a...thm#Perl_6 16:46
it's much shorter than in other languages
masak grondilu: nice! 16:47
grondilu++
masak always enjoys seeing max :by 16:48
grondilu is glad you like it
TimToady it would be easier to follow in the log if you didn't check in *quite* so often :)
masak sorry...
grondilu: nice implicit loop with 'mutate($parent) xx C', too. 16:49
TimToady I guess I can just do a manual diff on the two versions though...
grondilu TimToady: sorry, I just happen to have little ideas coming one at a time
TimToady happens to me sometimes too, just something to bear in mind
grondilu masak: I quite hesitated to use 'xx' but it seems ok
masak yeah. the semantics to re-eval the lhs is relatively new. 16:50
16:50 sorenso joined
masak (and an improvement, if you ask me) 16:50
grondilu I wonder how it knows whether the lhs is deterministic or not, though
masak it doesn't. 16:51
at least not current implementations. 16:52
they just re-eval the thunk regardless.
rn: say (2.rand.Int) xx 5
grondilu ah ok
p6eval rakudo fc349a: OUTPUT«1 0 1 1 0␤»
..niecza v22-29-g3ffe64c: OUTPUT«1 1 1 0 1␤»
masak oops, superstitious parens. 16:53
rn: say (2.rand.Int xx 5).join
p6eval niecza v22-29-g3ffe64c: OUTPUT«01000␤»
..rakudo fc349a: OUTPUT«01001␤»
16:53 sqirrel joined
masak rn: say Bool.roll xx 5 16:53
p6eval rakudo fc349a: OUTPUT«False False True True True␤»
..niecza v22-29-g3ffe64c: OUTPUT«False True False False True␤»
masak (if that were an iterated prisoner's dilemma, Niecza would come out on top) :P 16:54
PerlJam r: gist.github.com/4066950
p6eval rakudo fc349a: OUTPUT«foo␤No such method 'wrap' for invocant of type 'Bool'␤ in block at /tmp/zvbNpHDRp5:5␤␤»
PerlJam there we go
IS that a known bug? 16:55
16:55 tipdbmp joined
masak PerlJam: yes. 16:55
try a semicolon at the end of the first sub declaration.
PerlJam crazy 16:56
masak it's in RT. I'm looking.
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?
TimToady I guess my first question would be, "mathematical expressions" in what language? 16:57
masak tipdbmp: sounds like a nice exercise, yes.
PerlJam tipdbmp: Do you mean grammars in general, or just Perl 6 grammars? 16:58
TimToady real "mathematical expressions" are not strings
since mathematicians tend to use 2-dimensional notations
PerlJam (Perl 6 will hide some details and make things easier for you than you'd get from other tools like yacc or bison)
tipdbmp Simple math exprs: 1 + 2 * (3 / 4), for now. 16:59
TimToady have a look at rosettacode.org/wiki/Arithmetic_eva...ion#Perl_6 17:00
PerlJam tipdbmp: the "best" way to learn about grammars is to try to write a parser from scratch ;)
masak tipdbmp: I wrote one for this post: strangelyconsistent.org/blog/what-y...nt-without 17:01
it's linked from it.
TimToady (though that example may have bitrotted since it was put there)
masak tipdbmp: 17:02
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")
p6eval rakudo fc349a: OUTPUT«｢4 + 2 ｣␤ add => ｢4 + 2 ｣␤ mul => ｢4 ｣␤ term => ｢4｣␤ addop => ｢+｣␤ mul => ｢ 2 ｣␤ term => ｢2｣␤␤»
masak oops. :/ 17:03
interesting. I should run this through jnthn's debugger... :)
grondilu instead of 'max :by( *.&fitness <=> *.&fitness )', I whish I could have written just: 'max :by(*.&fitness)' 17:04
17:05 domidumont left
masak er. you can. 17:06
:by(&fitness), even.
grondilu r: say max :by(*.chars), <foo bar long>;
p6eval rakudo fc349a: OUTPUT«long␤»
grondilu I thought I had tried that. Don't know why it failed. 17:07
masak r: say max :by(*.&chars), <foo bar long> 17:08
p6eval rakudo fc349a: OUTPUT«long␤»
masak r: say max :by(&chars), <foo bar long>
p6eval rakudo fc349a: OUTPUT«long␤»
PerlJam grondilu: writing something that utilizes an evolutionary algorithm? 17:09
FROGGS [17:46 ]<grondilu> Check out my version of the evolutionary algorithm: rosettacode.org/wiki/Evolutionary_a...thm#Perl_6
PerlJam FROGGS++ thanks :)
grondilu PerlJam: No. Just having fun writting stuff in RosettaCode 17:10
benabik finds the *.foo version clearer than *.&foo or &foo
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.
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
rn: say (3 => 'foo', 3 => 'bar', 4 => 'long').max 17:12
p6eval rakudo fc349a, niecza v22-29-g3ffe64c: OUTPUT«4 => "long"␤»
TimToady well, maybe it doesn't matter much inside max 17:13
it's not like sort, where it's doing a lot of repeated comparisons 17:14
17:14 kurahaupo joined
TimToady (which does a schwartzian transform anyway, so nevermind) 17:14
TimToady looks around for more caffeine...
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 :)
one place we should optimize, for instance, is when you write *+*, it just probably replace it with &infix:<+>. 17:16
and *.&foo should just turn into &foo as appropriate 17:17
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
benabik Does &foo automatically look for methods? 17:19
17:19 birdwindupbird left
TimToady no, but neither does *.&foo 17:20
17:20 domidumont joined 17:22 brrt left
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 17:23
masak agreed. 17:24
TimToady [+] foo Zeq bar is obviously putting the Z into an eager context, for instance
benabik grondilu: You keep using & in IRC, but it's not there in the code since you're calling methods.
17:24 atrodo left
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
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
TimToady note that you can't simply turn a Zeq into a »eq« though, since the semantics are different 17:28
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.
TimToady and then it can call into »eq« efficiently for the comparable prefixes, if that's going to be faster 17:30
17:30 cognominal left
TimToady or into whatever does fast memory-to-memory comparisons... 17:30
that is presumably also used by »eq« in those cases where it can be fast 17:31
especially when comparing two compact arrays (also an optimization that is not done yet) 17:32
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.
diakopter both++
masak can you actually reduce infix:<eq> to a memory comparison? what about combining characters and NFG and stuff?
diakopter if they're already NFG then the comparison just works 17:35
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++
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.
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.
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.
TimToady p5-p6 integration is important, but will always impose an overhead that may outweigh the performance benefits of using p5 17:40
17:40 grondilu left, arlinius joined, rurban left
doy honestly, a lot of things on cpan could really benefit from the opportunity for a rewrite too 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
TimToady there's also cultural overhead :) 17:43
doy yeah, similar to why swig always produces such terrible results
tipdbmp What's the $^y variable? 17:44
PerlJam tipdbmp: it's a self-declared variable in the current block
tipdbmp: unless you're asking about perl5, then I have no idea :)
tipdbmp =)
17:46 sorenso left 17:47 kurahaupo left
sorear good * #perl6 17:48
TimToady o/
masak sorear! \o/ 17:49
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. 17:58
17:58 dakkar left
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. 17:59
TimToady yes, but it's the opposite of Einstein's idea; it's a wireless with an actual cat transmitting the signals 18:00
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
nwc10 masak: my assumption (not verified) is that much of the useful stuff on CPAN ends up depending indirectly on something XS on CPAN.
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
nwc10 the sort of anti Inline::Rakudo 18:03
well, inverse
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
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.
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)
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.
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)
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 18:08
doy guessing between Str and Buf will almost certainly lead to a whole lot of pain 18:09
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
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) 18:12
18:13 _jaldhar_ left
diakopter (same configure options as the perl available when configure of p6 is run, so it can use the same installed libraries) 18:13
18:15 FROGGS left 18:16 flightrecorder left
TimToady keeps wondering whether we should steal ^ from exclusive-or and use something else for that 18:22
how do mathematician's write exclusive-or? 18:23
*cians
⊻ or ⊕ it would appear 18:24
sorear .u ⊕
phenny U+2295 CIRCLED PLUS (⊕)
sorear I've never seen the former 18:25
mostly it's either ⊕ or +
TimToady infix:<o> # "one of" ;)
sorear not (Texas) functional composition?
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.
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.
TimToady or make it do what people expect :)
18:30 sisar left
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 18:31
18:32 _jaldhar_ joined
masak looks at the symbols on his keyboard and draws a blank 18:34
18:34 spider-mario joined
TimToady well, we certainly can't use the +, so I went with the o 18:35
sorear meh, xor is just + on GF(2)
TimToady GF(2) can overload + however it likes, but it's not Perl(6) :) 18:36
benabik .u ⚪
phenny U+26AA MEDIUM WHITE CIRCLE (⚪)
benabik That has funny shading on my computer that makes it look like a sphere. :-/ 18:37
huf it's a square here :) 18:39
fitting.
diakopter are you seriously considering making xor solely use a non-ascii character? 18:40
PerlJam that's odd, it's a dodecahedron here.
;-P 18:41
doy 'o' doesn't look very non-ascii to me
diakopter doy: I guess I wasn't addressing only TimToady...
18:42 sqirrel left, snearch joined
sorear wait, I think there's something wrong with my display, all my characters are two-dimensional 18:44
no spheres or dodecahedrons, or rhododendrons
diakopter .u 2331 18:45
phenny U+2331 DIMENSION ORIGIN (⌱)
diakopter (I have no idea what that looks like) 18:46
oh.
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
18:53 sorenso joined, sjohnson left
TimToady +o+ or o+o+ just don't work too well 18:54
18:56 SamuraiJack left
masak I think I've used +^ in Perl 6 code once. 18:58
I've never used ^ or ^^ -- and I don't think I will.
18:59 benabik left, ifim left
PerlJam er, why would we take back ^ in the first place? What's the motivation there? 19:02
19:03 benabik joined, rindolf left, sjohnson joined
dalek rl6-roast-data: 17f2d1e | coke++ | / (4 files):
today (automated commit)
 19:06
19:08 risou left, risou joined
[Coke] rakudo is choking on t/spec/S03-sequence/basic.t 19:08
TimToady the discussion in en.wikipedia.org/wiki/Exclusive_or#...ve_symbols is interesting 19:14
PerlJam: it produces very confusing results if the user expects exponentiation, and is badly huffmanized from the standpoint of frequency of use 19:16
19:18 Chillance joined
TimToady also, infix ^ has no relation to prefix ^ to preserve 19:20
so an alphabetic operator becomes a possibility, unlike operators that must also work as prefixes 19:22
PerlJam okay.
TimToady which is why I was thinking about o 19:23
oh, another benefit of such a change:
reduces the ambiguity with ~^ and such 19:24
when used as prefix
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…
..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␤␤»
TimToady heh, well, potential ambiguity
rn: say ^20; say ~ ^20;
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␤»
TimToady well, actual ambiguity, with a LTA result, I guess 19:26
19:26 snearch left
GlitchMr Shouldn't it be contamination of say function and 0..19 range? 19:27
(lol wrong word)
concatenation*
TimToady no, because the ~ is where a prefix is expected, not an infix 19:28
19:29 sisar joined
PerlJam (context is still king in Perl :-) 19:29
19:30 daniel-s__ joined 19:33 Chillance left 19:34 daniel-s_ left, Chillance joined
jnthn back 19:39
sorear yay jnthn 19:40
jnthn shesh, how much backlog while I went to teach folks git... :P
*sheesh
jnthn sees what he missed 19:41
19:41 brrt joined
[Coke] Since jnthn is in the room, is S03-sequence/basic.t failing for any other rakudoers? 19:45
19:45 cognominal joined
jnthn Looks good to be (at ca9c30, anyway) 19:47
19:52 rindolf joined
jnthn [Coke]: Just built latest and it's also seeming happy. 19:52
19:59 kurahaupo joined
jnthn oh noes, operator changes /o\ 20:01
Though thankfully to one probably rarely used :) 20:02
masak jnthn: nothing's changed yet :) 20:04
jnthn :)
I kinda like the symmetry of &, | and ^ all being one-char non-\w operators. 20:05
masak aye.
jnthn Though there's precedent for not having ^ in so far as there's nothing for none
masak Perl 5 does fine without a symbolic (boolean) xor operator. 20:06
20:12 kst left 20:13 kst joined 20:21 domidumont left
TimToady one must have the bitwise ops though, somehow 20:22
PerlJam If we go by huffman, the bitwise ops are way too short for the amount of use they get 20:23
TimToady we already doubled them from p5 :) 20:24
PerlJam still :)
20:25 kaare_ left 20:26 ifim joined
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 20:28
jnthn otoh, I doubt we can do anything more sensible with those chars :) 20:29
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? 20:30
[Coke] r: class MyClass {}; my $c = MyClass.new; say $c.WHAT; 20:31
p6eval rakudo fc349a: OUTPUT«MyClass()␤»
[Coke] r: class MyClass {}; my $c = MyClass.new; say $c.WHAT; say MyClass.WHAT; 20:32
p6eval rakudo fc349a: OUTPUT«MyClass()␤MyClass()␤»
jnthn r: class MyClass { }; my $c = MyClass.new; say $c ~~ MyClass; my $a = 42; say $a ~~ MyClass;
p6eval rakudo fc349a: OUTPUT«True␤False␤»
jnthn Use ~~ for type checking
.WHAT will tell you what it is, if you're curious. But for type checking, use ~~
(question coulda meant either :))
tipdbmp tnx =)
20:33 Moukeddar joined
sorear o/ Moukeddar 20:34
Moukeddar \o/ sorear :) 20:35
how are you doing, Sir!
sorear very good 20:36
rindolf Moukeddar: hi.
Moukeddar: what does "Moukeddar" mean?
20:37 wamba joined
Moukeddar Hola rindolf, it's in arabic, it's my last name, it doesn't have a specific meaning actually ;) 20:37
rindolf Moukeddar: ah, OK. 20:38
Moukeddar it's pretty unique, therefor i use it as a nickname
rindolf Moukeddar: ah.
Moukeddar: where are you from?
Moukeddar Morocco, you?
rindolf Moukeddar: unfortunately with this respect my last name is "Fish".
Moukeddar: I'm from Israel.
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 www.shlomifish.org/ - but I'm using a Fish emblem here.
And elsewhere.
sorear PerlJam: transpose? 20:40
Moukeddar PerlJam, even people here have problem with my name, i guess it's a curse lol
20:40 brrt left
sorear rindolf: does fish (he) mean the same as fish (en) ? 20:40
20:41 tipdbmp left
Moukeddar rindolf, nice website ;) , you too people have problems with hebrew support 20:41
rindolf sorear: no, there is no word "Fish" in Hebrew. 20:42
Moukeddar: Hebrew support is usually not a problem.
Moukeddar rindolf, talking about RTL ;)
rindolf Moukeddar: but I prefer that most of my home site will be in English.
Moukeddar: yes, I know.
Moukeddar: see www.shlomifish.org/meta/FAQ/#website_in_english . 20:43
sorear: my last name is not Hebrew. It is German in origin or something.
sorear: maybe English.
sorear: my grandfather came from Poland. 20:44
tadzik ha, fellow :)
Moukeddar rindolf, good reasons ;) i could borrow them ;)
TimToady my ancestors lived in Poland about 5 generations ago
on the way from Belgium to Ukraine...
PerlJam < rindolf> sorear: no, there is no word "Fish" in Hebrew. 20:45
tadzik wow
PerlJam Do they not eat fish? Do fish not exist in hebrew?
geekosaur I think this is use/mention
sorear PerlJam: rindolf's last name is F-I-S-H 20:46
PerlJam: I wondered if those letters meant something in Hebrew
benabik PerlJam: Don't feel bad, that was my first thought on reading that sentence too.
Moukeddar Watch out, #perl6 is turning into Language analytics :) 20:47
geekosaur wasn't it always?
PerlJam wonders if he should use more emoticons in his writing on IRC
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
geekosaur "dag" 20:47
TimToady
.oO(and Jonah was swallowed by a great <mumble>)
 20:48
PerlJam "large legless water-dwelling animal bigger than an elephant" 20:49
Moukeddar thinks PerlJam could use some more emotions
PerlJam Moukeddar: my wife tells me that all the time :)
sorear rindolf: interesting favicon you have there 20:50
rindolf PerlJam: the Hebrew name for Fish is "דג" - "Dag". 20:51
[Coke] Yo, Dag. 20:52
PerlJam heh
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
rindolf PerlJam: heh.
sorear Hai dag
GlitchMr Why not both?
sorear </injoke>
[Coke] TimToady: Sounds Fishy.
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
rindolf sorear: en.wikipedia.org/wiki/Pe_%28letter%29 20:54
geekosaur dag nabbit? 20:55
sorear rindolf: I am not sure what you are replying to
Moukeddar you guys are hilarious
rindolf sorear: to the Fish discussion.
sorear: in Hebrew, you cannot start a word with "f".
sorear: it must start with a "p". 20:56
GlitchMr gist.github.com/4068343
Python is too limited for that...
But perhaps with complex hacks... 20:57
sorear rindolf: do Hebrew first-languagers call you /piʃ/ ?
TimToady and is that really [pʰiʃ] as an English speaker would pronounce it? 21:00
rindolf sorear: no, they don't.
sorear dunno, my skillz don't extend to the ʰ level
TimToady rindolf: then do they call you "דג"? 21:01
21:02 kst left
rindolf TimToady: no, they call me "Fish" 21:02
diakopter doubling-- bitwise-- again--
rindolf Well, my first name is Shlomi.
21:02 kst joined 21:03 _jaldhar_ left, jaldhar joined
geekosaur grumbles, having suddenly noticed that his irc client didn;t rtl properly 21:04
21:05 sisar left 21:07 GlitchMr left 21:09 jaldhar left 21:11 fhelmberger left, jaldhar joined 21:12 nebuchadnezzar joined 21:23 jlaire left 21:29 Moukeddar left, rindolf left 21:35 jlaire joined, gaussblurinc_ joined 21:44 benabik left
tipdbmp Where does one place the regex modifiers? if $str ~~ m/:i something / { "doesn't seem to work" } 21:45
masak rn: say "aBc" ~~ m:i/ AbC / 21:46
p6eval rakudo fc349a: OUTPUT«｢aBc｣␤␤»
..niecza v22-29-g3ffe64c: OUTPUT«#<match from(0) to(3) text(aBc) pos([].list) named({}.hash)>␤»
masak rn: say "aBc" ~~ m/ :i AbC /
p6eval rakudo fc349a: OUTPUT«｢aBc｣␤␤»
..niecza v22-29-g3ffe64c: OUTPUT«#<match from(0) to(3) text(aBc) pos([].list) named({}.hash)>␤»
masak either outside or inside works.
tipdbmp Oh, ok. 21:47
21:53 kst left, kst joined 21:59 ifim left 22:13 japhb_ joined
japhb_ grondilu++ # Very nice work on rosettacode.org/wiki/Evolutionary_a...thm#Perl_6 22:15
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
ETOOMANYTASKS
22:17 cognominal left
jnthn wonders which task is most intstresting... :) 22:17
masak -Ointeresting 22:18
japhb_ TimToady, re: 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 22:23 cognominal joined 22:24 spider-mario left, sorenso left 22:25 lawrmcfa joined, lawrmcfa left 22:26 lawrmcfa joined
tipdbmp How can one subtract character classes? Say: <[a..z]> - <[aeiou]> 22:28
jnthn inside the angles 22:29
<[a..z]-[aeiou]>
tipdbmp no space between ]-[?
jnthn Up to you
tipdbmp Ah, ok. tnx =)
masak can also do + between the [], and a + or - before the first one. 22:30
22:31 telex left
sorear but space between < and [ is likely to confuse things 22:32
22:32 telex joined
jnthn Yes, that's not optional. :) 22:33
Otherwise it'll get taken as quote words I guess...
sorear n: say grep /<[0-z]^:Letter>/, <0 a ð …> 22:34
p6eval niecza v22-29-g3ffe64c: OUTPUT«===SORRY!===␤␤Unsupported use of - as character range; in Perl 6 please use .. at /tmp/JTfZjDs1sn line 1:␤------> say grep /<[0-z⏏]^:Letter>/, <0 a ð …>␤␤Parse failed␤␤»
sorear n: say grep /<[0..z]^:Letter>/, <0 a ð …>
p6eval niecza v22-29-g3ffe64c: OUTPUT«0 ð␤»
sorear \o/
22:37 gaussblurinc_ left
masak nice. 22:37
[Coke] can you deparse that for me? 22:40
sorear < [0..z] ^ :Letter > ? 22:41
[Coke] (ah, there's a nice module to have. regex::deparse)
sorear regex::deparse sounds like it converts a Regex object back to text 22:42
[Coke] so, that's "Any thing that's a letter or in the range 0..z, but not both" ? 22:43
(looks like.) 22:47
flussence xor. 22:48
[Coke] the spec calls it by a more set-like name, but yah. 22:49
masak "symmetric set difference"? 22:51
[Coke] Aye.
masak just a stab in the dark. :P
22:52 wamba left 22:58 cognominal left 23:01 lawrmcfa left 23:02 lawrmcfa joined 23:05 alester left 23:06 wamba joined 23:14 mtk left 23:19 am0c joined 23:24 MayDaniel left 23:29 am0c left 23:36 jlaire left 23:37 wamba left 23:43 fgomez left 23:44 jlaire joined 23:48 Chillance left 23:49 lawrmcfa left 23:54 sjohnson left, sjohnson joined 23:55 nebuchadnezzar left 23:56 am0c joined

Powered by Cro and the Raku® Programming Language.

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

Thank you!