00:00 rgrau_ left, franek left 00:04 rdesfo left 00:10 Juerd_ joined
dalek ecza: 11bf899 | sorear++ | / (3 files):
Revert use of collectable assemblies for eval

It turns out they've only been supported in Mono since Feb 09 2010. This does mean eval will leak memory. Try not to overuse it.
00:14
00:15 whiteknight joined 00:22 orafu joined 00:29 ponbiki left
tadzik neutro did not age gently 00:31
*, I'm asleep now
00:33 pigdude left 00:35 p6eval left, p6eval joined, ChanServ sets mode: +v p6eval
sorear niecza: eval "say 2 " ~ "+ 2" 00:36
p6eval niecza v3-15-g11bf899: OUTPUT«4␤»
dalek albot: d582aa5 | sorear++ | / (2 files):
Fix niecza build and execution
00:38
00:40 pmichaud left, pmichaud joined, Juerd_ left, hatseflats left 00:42 hatseflats joined 00:44 stifynsemons left 00:47 hatseflats left, hatseflats joined 00:50 qiyong joined 01:01 thorat joined
sorear moritz__: I don't get any FormatExceptions with t/spec/S02/bool 01:03
01:04 woosley joined 01:10 Juerd_ joined 01:18 cotto left, _jaldhar left 01:19 _jaldhar joined
donri sorear: wait what eval returns stdout? 01:19
oh wait nevermind :D 01:20
parsefail
dalek ecza: 762eae4 | sorear++ | / (2 files):
Note 2.6 breakage in README
01:25
ecza: a7846ce | sorear++ | / (3 files):
Turn panics and sorries into catchable exceptions
01:46 masonkramer_ joined, masonkramer left, masonkramer_ is now known as masonkramer 01:57 arlinius left
dalek ecza: 980923f | sorear++ | lib/Test.pm6:
Implement eval_dies_ok, eval_lives_ok
01:58
02:01 noganex_ joined
sorear niecza: say -1 02:01
p6eval niecza v3-17-ga7846ce: OUTPUT«-1␤»
sorear HEY! Someone swapped the arguments to skip! 02:02
dalek ecza: 0711b24 | sorear++ | lib/Test.pm6:
Track &skip API change
02:03
02:04 noganex left 02:09 jferrero left 02:10 agentzh joined, thorat left 02:19 Juerd_ left 02:20 pmichaud left, hatseflats left 02:21 stifynsemons joined, Juerd_ joined, hatseflats joined 02:22 pmichaud joined 02:25 whiteknight left 02:28 masonkramer left
qiyong does perl6 have a native implementation? 02:31
not on vm
sorear What does that question mean? 02:34
qiyong does perl6 have a native implementation? like the perl5, not the rakudo 02:35
plobsing perl5 walks a tree of ppcodes, which is very much what I would define as a VM 02:36
sorear perl 5's internal VM is slower than Parrot at a lot of stuff 02:38
just because it doesn't have a name doesn't mean it doesn't exist
02:39 leprevost left
donri maybe they mean interpreter, or wonder if parrot can be bundled into rakudo 02:40
02:43 cdarroch left 02:52 arlinius joined 02:56 JimmyZ joined 03:02 _twitch joined 03:06 leprevost joined 03:07 donri left 03:10 GinoMan_ left 03:19 GinoMan joined 03:27 lopaway is now known as lopnor 03:43 leprevost left 03:46 felipe__ joined 03:47 felipe__ left 03:48 leprevost joined
qiyong sorear, is there a perl5 parrot implement? 03:50
rakudo is a perl6 compiler target at parrot? 03:51
JimmyZ qiyong, are you chinese?
qiyong JimmyZ, why 03:53
JimmyZ looks like you're chinese. 03:54
I'm chinese
qiyong you can see my face?
03:54 _jaldhar left
JimmyZ from your nick, if you're curious 03:56
03:59 anonymousarray joined 04:00 jaldhar joined 04:02 lopnor is now known as lopaway 04:03 cotto joined 04:05 satyavvd joined 04:23 Su-Shee_ joined 04:25 Su-Shee left 04:28 am0c joined 04:34 estrabd left, estrabd joined 04:35 tewk left, tewk joined 04:45 coldhead left 04:49 woldrich_ is now known as woldrich, woldrich left, woldrich joined 04:54 anonymousarray left
mberends qiyong: SMOP is a non virtual machine runtime for Perl 6, but it is incomplete and has not been worked on for about two years :-( www.perlfoundation.org/perl6/index.cgi?smop 04:55
qiyong: Ponie was a project more than 2 years ago to implement Perl 5 on the same runtime as Perl 6, but that was abandoned because the developers could not make it work reasonably. 05:02
05:03 GinoMan left 05:27 envi left 05:31 VXZ left
sorear In what sense do you call SMOP "non virtual macine"? 05:45
05:45 bacek left
sorear JimmyZ: out of curiousity, how did you come to be called "Jimmy"? 05:45
05:48 lue left, __rnddim__ joined
woosley qiyong looks like a chinese name 05:49
sorear see also: www.ip2location.com/60.23.248.89 05:50
tadzik what an investigation 05:52
05:53 satyavvd left
mberends sorear: to quote Daniel Ruoso from the perlfoundation page, "SMOP is not a Virtual Machine. SMOP is simply a runtime engine that happens to have a interpreter run loop." AFAIK that means compilers targeting SMOP emit C code that makes calls into the SMOP runtime system. 05:53
woosley What is std.pm6 different from rakudo and other implements? I see they all can run perl6 code 05:54
05:57 kaare_ joined
mberends woosley: STD.pm6 is a reference grammar (with a working parser) only. Niecza can compiler and execute a subset of what STD.pm6 describes. Rakudo has a slightly different grammar and can also execute much more of your code. 05:57
*compile
woosley mberends: So STD.pm6 can not execute perl6 code 05:59
mberends woosley: correct
sorear std: say 2 + 2 06:01
p6eval std 4608239: OUTPUT«ok 00:01 120m␤»
sorear it just does semantic checking
woosley perl6: 2+2 06:02
p6eval pugs, rakudo 74dcce, niecza v3-19-g0711b24: ( no output )
sorear mberends: I cry a little inside when people describe (any piece of software which defines a specialized programming environment) as "not a virtual machine" 06:04
tadzik std: say "ok 00:01 {150.rand.Int}m" # my favourite joke
p6eval std 4608239: OUTPUT«ok 00:01 119m␤»
mberends sorear: you're too sentimental ;) 06:05
Tene sorear: +1
mberends anything which interprets is a virtual machine by some definition, I suppose. 06:06
JimmyZ sorear: my employer asked for my english name, not chinese;) 06:08
06:22 nymacro joined 06:30 quester_ joined
mberends wonders about rescuing SMOP from the Pugs catacombs. Where to find some more tuits? 06:34
06:34 cjk101011 joined, cjk101011 left
diakopter mberends: and even round ones! 06:34
mberends heh
sorear mberends: maybe ask a VB6 or ActiveX developer 06:39
mberends lol
sorear SMOP is, coincedentally, extremely similar to MS COM at the ABI level
it would probably take less than a month to make it completely compatible
and then you could embed perl 6 in IE6 06:40
diakopter :)
using a NaCl plugin, mono (and thus niecza (and rakudo/mono)) will be in IE, Chrome, Firefox, Opera, etc... 06:41
06:41 cjk101010 left
diakopter (someday) 06:41
oh yeah, Safari 06:42
sorear NaCl better than Moonlight?
diakopter I'd think so 06:47
*imagine
all the cool interpreters are porting to NaCl 06:48
06:52 JimmyZ left
sorear do you think nacl will ever have significant market penetration? 06:53
dalek ecza: c01053e | sorear++ | t/spectest.data:
Remove test which never should have worked
ecza: 31427c7 | sorear++ | lib/SAFE.setting:
Fix $str ~~ m// always being true
ecza: 12393a7 | sorear++ | src/niecza:
Fix compiler crash on [+]
ecza: 1237dcf | sorear++ | t/spectest.data:
Remove another file that only passed by accident
07:09
ast: b8687da | sorear++ | S0 (4 files):
Refudge a few evals for niecza
07:11 cosimo left
moritz__ good morning 07:14
07:15 Mowah joined
sorear good night 07:16
moritz__ fwiw the FormatExceptions seem to be gone now 07:22
07:26 Su-Shee_ is now known as Su-Shee, coldhead joined
diakopter phenny: tell sorear "tough to see". NaCl has a lot of potential, especially if Adobe teams up with Google (like it seems might occur) 07:34
phenny diakopter: I'll pass that on when sorear is around.
07:38 Rotwang joined 07:41 wtw joined 07:42 broquain1 is now known as broquaint 07:47 stifynsemons left 07:50 fhelmberger joined 07:53 Rotwang left 07:54 shi joined
dalek ast: 9944532 | moritz++ | S04-statements/return.t:
fudge return.t for niecza
08:01
08:23 noganex_ is now known as noganex 08:27 quester_ left
jnthn morning, #perl6 08:34
mberends morning, jnthn 08:35
colomon \o
mberends jnthn: which implementation of 6model do you suggest I overview for my NLPW talk? 08:36
08:38 justatheory left, justatheory joined
TimToady finish up the JVM one, and then talk about that :P 08:38
colomon TimToady! \o/ 08:39
TimToady really has no opinion...
08:39 rhr left
mberends o noes, that means inventing a goto in Java ! 08:40
TimToady well, was supposed to be on an airplane right now, but somehow I swapped my passport with my mother-in-law's while we were on safari, and didn't notice till we got to the airport...
mberends TimToady: your is the best answer though ;-)
*yours
TimToady they didn't think I was pretty enough to pass for her... 08:41
so we get a 20 hour extention to our stay here, and an award of -$1k or so for changing tickeets :( 08:42
colomon :(
jnthn mberends: The .Net one is probably easiest in that it's a ground-up implementation. It's missing a few bits, but it is cleaner since it doesn't have to interact with Parrot-y bits. 08:43
TimToady: Ouch. :(
08:43 justatheory left
mberends jnthn: and which repo? the 6model one or one of the NQP's (I lost track of where you're working now). 08:44
jnthn 6model one is active repo for CLR and JVM implemenations. Only Parrot one is currently in nqp repo. 08:47
mberends ok thanks, that's a good starting point for me :) 08:48
08:48 rhr joined 09:05 Juerd_ left 09:07 domidumont1 left
jnthn mberends: OK, great. 09:09
jnthn should work on his NLPW talks very soon too :)
09:13 domidumont joined
moritz__ jnthn: fwiw I have built a separate checkout of the nqp/match-nom branch under a different user ID, so uninfluenced by any installed parrots/rakudos/nqps/*. I still get that weird, uninformative parse error. So I fear I can't really do anything productive on that branch :( 09:14
jnthn moritz__: OK, that's...confusing 09:19
jnthn will try it on a different machine
09:23 dakkar joined 09:31 daxim joined 09:32 sftp left 09:49 envi joined 09:50 bacek joined 09:51 nymacro left 09:52 nymacro joined 09:56 bacek left 09:59 JimmyZ joined 10:02 JimmyZ left 10:05 woosley left 10:08 mberends left 10:11 leprevost left 10:17 moritz__ is now known as moritz_ 10:27 tzhs joined 10:34 hcchien left 10:35 domidumont left 10:38 coldhead left 10:39 woosley joined 10:40 domidumont joined 10:46 nsh- left 10:47 nsh_ joined 11:05 am0c left 11:06 jferrero joined 11:16 gdey joined 11:17 JimmyZ joined, am0c joined, gdey left 11:21 charsbar left, gdey joined 11:26 charsbar joined 11:29 Gothmog_ left, donri joined 11:30 charsbar left 11:33 charsbar joined 11:35 pmurias joined 11:38 Khisanth left 11:39 orafu left 11:40 orafu joined, necrodearia left 11:41 agentzh left 11:43 necrodearia joined 11:45 awoodland joined 11:46 orafu left, orafu joined 11:54 MayDaniel joined, MayDaniel left 12:12 am0c left 12:17 Gothmog_ joined, Gothmog_ left, Gothmog_ joined 12:21 bacek joined 12:25 am0c joined 12:30 awoodland left 12:31 mfranc joined 12:43 pmurias left, mtk joined 12:46 amkrankruleuen left, amkrankruleuen joined
donri What other languages do you good folk find interesting, from a non-pragmatic perspective? 12:47
colomon Factor! 12:48
moritz_ donri: haskell and ook 12:49
donri lol
colomon ook?
donri Esoteric 12:50
flussence YES! I finally convinced $dayjob we actually need tests :D
colomon \o/
moritz_ \o/
flussence (and all it took was years of breakages...)
donri "I don't always write tests, but when I do, I write them first." 12:51
12:55 pi_benjamin joined, pi_benjamin left, woosley left 13:09 Khisanth joined 13:16 stifynsemons joined 13:19 MayDaniel joined 13:22 shi left 13:29 awoodland joined
ggoebel is there something equivalent to blizkost for niecza? I.e. a way to access perl5 modules? 13:29
13:29 mtk left, ab5tract joined 13:33 woosley joined 13:34 mtk joined
[Coke] tadzik++ 13:38
tadzik [Coke]: what do I get my karma from?
[Coke] std: say "ok 00:01 {150.rand.Int}m" # my favourite joke 13:39
p6eval std 4608239: OUTPUT«ok 00:01 119m␤»
13:39 starcoder left
tadzik heh 13:40
13:40 ddima left, ddima joined
moritz_ patches to make that just 'parse OK' or so are welcome 13:40
13:41 starcoder joined 13:50 plainhao joined
woosley rakudo: for <a b c> { say } 13:53
p6eval rakudo 74dcce: OUTPUT«␤␤␤»
moritz_ std: for <a b c> { say }
p6eval std 4608239: OUTPUT«Potential difficulties:␤ Unsupported use of bare 'say'; in Perl 6 please use .say if you meant $_, or use an explicit invocant or argument at /tmp/xgqdmDFctS line 1:␤------> [32mfor <a b c> { say[33m⏏[31m }[0m␤ok 00:01 119m␤»
moritz_ that's what rakudo should say too 13:54
woosley rakudo: for <a b c> {.say} 13:55
donri Make it so, number one.
p6eval rakudo 74dcce: OUTPUT«a␤b␤c␤»
moritz_ patches welcome :-)
should be mostly copy&pasting from STD
donri rakudo: for <a b c> *.say
p6eval rakudo 74dcce: OUTPUT«===SORRY!===␤Missing block at line 22, near ""␤» 13:56
donri Why doesn't this work
moritz_ didn't we discuss this before?
'for' is syntactic, and needs a block.
donri yes but this is syntax
why a block and not simply a code though?
rakudo: for <a b c> &say
p6eval rakudo 74dcce: OUTPUT«===SORRY!===␤Missing block at line 22, near ""␤»
moritz_ donri: because it breaks the "no two terms in a row" rule 13:57
and such breaks can only be tolerated for a very limited syntax 13:58
donri why is it not a function then
for <a b c>: { ... }
because it's syntactic elsewhere?
.say for <a b c>
woosley perl6: my @a = <a b c>; say "@a"; my $a = "f"; say "$a"; 13:59
p6eval niecza v3-23-g1237dcf: OUTPUT«a b c␤f␤»
..pugs, rakudo 74dcce: OUTPUT«@a␤f␤»
moritz_ donri: there's a function version, and it's called map
woosley std: my @a = <a b c>; say "@a";
p6eval std 4608239: OUTPUT«ok 00:01 122m␤»
moritz_ woosley: rakudo and pugs are correct here 14:00
colomon I assume that's a known niecza bug?
moritz_ (std just does syntax checking)
colomon: not to me, but I guess sorear++ knows :-)
(and if not, the spectests will tell him)
donri moritz_: true :) 14:01
woosley Why say "@a" is not working like perl5?
donri sorta wish map function semantics was inverted
map @a: {} vs current map {}, @a
jnthn woosley: use say "@a[]" for those
woosley: Too many accidental say "mail me at [email@hidden.address] OH NOES interpolation!!!" 14:02
moritz_ woosley: to avoid unintiontial interpolation of things that are meant to be email addresses
woosley Oh, my bad^^
colomon donri: I know I prefer @a.map({}). My impression is that is a fairly common thing among the zebras.. 14:03
donri but that's method not function :) 14:04
colomon donri: that's my point precisely 14:05
donri also if the function was an export of the method the list should come first too right?
method map is export ...
map @instance, args...
moritz_ well, it's not
donri why not! 14:06
moritz_ just like join() has the separator first
donri: to adher to common idioms when used a function
jnthn wonders how much of that is Perl 5 history
donri idioms that fit less nicely in perl6 :(
moritz_ *all* functional languages have map CODE LIST
donri i know
jnthn Ah, true.
donri pretend it means "map over" then 14:07
map-over @list, &code
14:07 kaare_ left
moritz_ and join @list, 'separator' 14:07
moritz_ shudders
donri is that how it's now? 14:08
moritz_ donri: map, join and the likes can't be a simple 'is export' anyway
no
donri because that looks better too
"join this @list with this 'separator'"
moritz_ because it doesn't just want an array, but a slurpy arrray in the signature
colomon moritz_: exactly
moritz_ and you can't have required positionals after variadic positionals 14:09
colomon rakudo: say ~map(* + 10, 2, 3, 15)
p6eval rakudo 74dcce: OUTPUT«12 13 25␤»
moritz_ and when you have lazy lists, it just makes much more sense to treat the *first* argument special, in contrast to the last, which might not even exist
14:10 pmurias joined
moritz_ just think of map (1 ... *), *.sqrt 14:10
it'll never get to the closure
pmurias ggoebel: not yet 14:11
woosley rakudo: class KKK{ has Bool $.att is rw}; my $class = KKK.new; $class.att = "string"
p6eval rakudo 74dcce: ( no output )
woosley rakudo: class KKK{ has Bool $.att is rw}; my $class = KKK.new; $class.att = "string"; say $class.att;
p6eval rakudo 74dcce: OUTPUT«string␤»
moritz_ woosley: attributes aren't typed yet. ETA: maybe 3 months
jnthn Will happen with 6model merge as they're needed for natively typed attrs. 14:14
14:14 plobsing left
moritz_ looks forward to it 14:15
and as always it will break quite some code
moritz_ remembers all that code with my Array of Int @a; etc.
colomon but that's the best sort of breakage, as it gives us direction and focus and more power when it is fixed!
moritz_ aye
anyway, I think I've learned my lesson from last time and don't put type constraints in that aren't enforced 14:16
14:19 amkrankruleuen left 14:20 amkrankruleuen joined 14:23 risou joined 14:27 amkrankruleuen left 14:29 bluescreen joined 14:32 bluescreen is now known as bluescreen10 14:33 amkrankruleuen joined 14:34 qiyong left, qiyong joined 14:36 bluescreen100 joined, bluescreen10 left, bluescreen100 left, bluescreen10 joined 14:46 amkrankruleuen left 14:47 amkrankruleuen joined 14:48 pigdude joined 14:49 mberends joined, stifynsemons left 14:50 mkramer joined 14:57 amkrankruleuen left, sftp joined, amkrankruleuen joined 14:58 bluescreen10 left 14:59 stifynsemons joined 15:03 amkrankruleuen left 15:04 amkrankruleuen joined 15:05 bluescreen10 joined, bluescreen10 left 15:06 bluescreen10 joined 15:07 kaare_ joined 15:09 tyatpi_ joined, bluescreen100 joined 15:10 bluescreen10 left 15:12 bluescreen100 is now known as bluescreen10 15:13 PacoLinux joined 15:16 alester joined, stifynsemons left 15:18 hanekomu joined
sorear good * #perl6 15:18
phenny sorear: 07:34Z <diakopter> tell sorear "tough to see". NaCl has a lot of potential, especially if Adobe teams up with Google (like it seems might occur)
sorear moritz_: I did not know about the "@a" issue
moritz_ sorear: now you do :-) 15:19
sorear wonders where the p6a mail went
moritz_ sorear: I have a patch at home that adds more passing files to spectesst... will you give me a commit bit, or should I pull request? 15:20
(can't push right now, don't have access to that machine)
sorear oh! sorry, meant to do that after the release. 15:21
moritz_ sorear: btw rakudo stringifies parcels just like it stringifies arrays and lists; I guess changing that would make quite a few tests pass too
perl6: say ~(1, 2, 3) 15:22
p6eval pugs, rakudo 74dcce: OUTPUT«1 2 3␤»
..niecza v3-23-g1237dcf: OUTPUT«Parcel()<instance>␤»
sorear just moritz, right?
moritz_ sorear: right
sorear done.
moritz_ thank you
15:25 Juerd left, Juerd joined
pmurias sorear: hi 15:25
dalek ecza: 8a2c6e7 | pmurias++ | / (2 files):
[cl-backend] added support for hint_hack
15:26
15:26 PerlJam left 15:27 PerlJam joined
sorear pmurias: hi 15:28
15:31 MayDaniel left
moritz_ perl6: say ~(1, 2, a => 1, b => 2, c => 3) 15:31
pmurias sorear: what does letscope do?
p6eval pugs, rakudo 74dcce: OUTPUT«1 2 a 1 b 2 c 3␤»
..niecza v3-24-g8a2c6e7: OUTPUT«Parcel()<instance>␤»
takadonet morning all 15:32
moritz_ o/
tadzik o/ 15:35
15:38 MayDaniel joined
dalek ecza: f05ff59 | sorear++ | docs/nam.pod:
Document letscope
15:38
15:38 MayDaniel left 15:41 plobsing joined
sorear pmurias: how's that? 15:41
pmurias the description seems clear 15:42
i don't like having flags in opcodes
sorear: or did you refer to something other then the commit? 15:45
is redefining infix:<=> allowed in Perl 6? 15:50
sorear I was referring to the commit, in reply to 09:31 < pmurias> sorear: what does letscope do?
I'm pretty sure it is
there should be a test for that in test.pl
PerlJam "allowed" and "Perl 6" are words that don't seem to belong in the same sentence 15:51
unless it's something like "You're allowed to do almost anything in Perl 6"
sorear out
moritz_ PerlJam: am I allowed to use Perl 6 at work?
PerlJam moritz_: yes!
pmurias moritz_: are you forbidden from using Perl 6 at work? 15:55
moritz_ that was more rhetorical :-) 15:56
15:56 stifynsemons joined
moritz_ pmurias: not at all, they only care if I publish papers in the end -- but currently no Perl 6 implementation is suited for what I do (symbolic calculations, number crunching, generating pretty plots) 15:57
15:57 Guest87704 joined 15:59 woosley left 16:00 mfranc left
pmurias moritz_: what do you use currently? 16:06
moritz_ pmurias: Mathematica, C, matlab, gnuplot, ...
PerlJam s/matlab/PDL/ and you get some of what I do too
16:10 JimmyZ left 16:11 qiyong left 16:16 bluescreen10 left, Patterner left 16:18 Psyche^ joined, Psyche^ is now known as Patterner
donri yay lua is faster than PASM ;) 16:19
TimToady quick, rewrite p6 in lua
moritz_ or rewrite lua in p6, to make it slower :-) 16:20
donri there's lua on parrot :)
16:21 shi joined, tyatpi_ left
donri just testing startup time for hello world in different languages 16:21
lua: 4ms, pir/pasm: 12ms, python: 27ms, rakudo: 490ms :D
moritz_ concluding that "lua is faster" without further qualification seems oh so wrong 16:22
sbp rakudo just adds a pause for emphasis
donri moritz_: not really being serious here
moritz_ donri: good
donri perl5: 6ms 16:23
take that python ;)
moritz_ p5 is amazingly fast for a non-jitted scripting language 16:24
it pays dearly for it... you'll see what I mean when you try to fix a bug in the regex engine :-)
or in the lexer
donri does it even do vm/bytecode? 16:25
PerlJam donri: what's the startup time of the empty program?
moritz_ donri: it compiles to an optree that's executed in memory 16:26
PerlJam donri: also, check out the memory footprint of each of those programs too
16:28 ymasory_ joined
plobsing parrot starts up slow. this is a known issue. but there are many priorities above it. such as parrot runs slow, parrot is memory hungry, etc. 16:29
plobsing sometimes wonders how hard it would be to flatten perl 5's op trees to threaded code. they already flatten it to a linked list, so it could work. 16:32
probably very hard though 16:33
moritz_ rurban was working on JIT-compiling the optree, last I talked to him
16:39 cxreg2 is now known as cxreg 16:45 justatheory joined 16:52 tzhs left 16:58 GinoMan joined 17:01 coela joined 17:04 shi left 17:05 Bzek left 17:12 awoodland left 17:14 GinoMan left 17:22 bluescreen10 joined
colomon #phasers in 97 minutes? 17:23
tadzik alredy? /o\ 17:24
if everything goes right, I'll be late :)
17:24 mtk left
mberends -> amsterdam.pm o/ 17:26
17:31 mberends left
sjohnson after reading the camel book for a while 17:31
im kind of curious if p6 is going to be as integrated with a system as p5 is..
by that i mean, all the pipe stuff, syswrite stuff, and a lot of the "Special Names" from the camel look that seem pretty advanced 17:32
17:33 mtk joined
moritz_ sjohnson: I guess it will be available, but via an ordinary API, not special syntax 17:35
dalek ecza: 77a9821 | moritz++ | t/spectest.data:
more passing test files
17:38
17:39 Hien joined
Hien perl6: say 3; 17:39
p6eval pugs, rakudo 74dcce, niecza v3-25-gf05ff59: OUTPUT«3␤»
Hien perl6: say 'Hello'; 17:40
p6eval pugs, rakudo 74dcce, niecza v3-25-gf05ff59: OUTPUT«Hello␤»
moritz_ hello Hien
Hien @moritz_ hello
sjohnson moritz_: more so just wondering if the design goal is to be just as powerful systemwise for some easy-to-write highlevel tasks 17:41
some tasks are making me realize that p5 is pretty refined and comprehensive
17:42 _twitch left 17:44 LumberCartel joined, LumberCartel left 17:46 daxim left, masak joined
masak \o/, zebras 17:46
jnthn masak! \o/
masak \o/!
jnthn wonders how to factorial a o/ 17:47
17:47 pmurias left
moritz_ well, by [*] 1..\o/ 17:48
masak duh :)
jnthn :P
moritz_ #phasers in 71 17:49
jnthn will be about for it.
[particle] sjohnson: perl6:perl5::physics:biology - lots of rules and relationships, less to memorize
17:51 hanekomu left 17:54 dwhipp joined 17:56 risou left
dwhipp Looking at Numeric.pm I see a whole bunch of methods implemented as method foo() { note "foo needed"; fail "foo needed" } 18:01
I'm wondering if { fail :verbose, "..." } would be the right way to prevent this duplication 18:02
moritz_ dwhipp: rakudo sometimes silently eats fail() messages (and at the same time aborts execution of a statement) 18:03
18:03 bluescreen10 left
moritz_ dwhipp: so it's more of a rakudo problem than a general problem with fail 18:03
18:03 lue joined, __rnddim__ left
dwhipp I'd been thinking that implementing such an option would be a good way to learn PIR ... seems probably not needed 18:04
18:06 coela left
masak there are lots of other good reasons to learn PIR, most of them sitting in RT ;) 18:07
moritz_ btw dwhipp++ for the series patch 18:08
18:08 dsp_ left, chitragupt left 18:09 gdey_ joined
dwhipp Thanks -- I'll have to go trawling. 18:10
18:10 Hien left 18:12 jedai_ joined, jedai left 18:13 chitragupt joined 18:14 Axius joined 18:15 cdarroch joined, cdarroch left, cdarroch joined 18:16 bluescreen10 joined
masak dwhipp: it's a good idea to run whatever you find with the people here on #perl6. sometimes the status of tickets is partly encoded in people's heads. 18:23
for example, there's a large number of open tickets that will miraculously be fixed when nom merges. 18:24
18:24 dsp_ joined 18:26 ymasory_ left
moritz_ where nom = new object model 18:26
masak specifically, its representative branch in the rakudo repo. 18:30
18:32 lamstyle_ joined 18:35 GinoMan joined 18:37 hanekomu joined 18:39 pigdude left, hanekomu_ joined 18:41 pigdude joined, hanekomu left 18:45 envi left, zby_home joined 18:46 dsp_ left, noganex_ joined 18:49 noganex left 18:50 am0c left
pmichaud good afternoon, #perl6 19:00
19:02 bluescreen10 left 19:03 Axius left 19:05 kaare_ left, kaare__ joined 19:08 MayDaniel joined 19:10 dsp_ joined 19:12 fhelmberger left 19:15 mberends joined, bluescreen10 joined
moritz_ good afternoon pmichaud. #phasers is in progress 19:17
19:41 mberends left, hercynium joined 19:45 shi joined
pyrimidine takadonet++ #bioperl6 work 19:50
19:51 spq left
takadonet pyrimidine: thanks :) 19:57
cannot wait to get translate working completely!
19:58 nymacro left
pyrimidine takadonet: I think moritz_ or masak had a a nice perl6 post a year or so ago on translation 20:02
masak 20:03
it was just a one-liner.
takadonet pyrimidine: i remember something was posted on this channel a long time ago
ya
I was more thinking in the lines of unambiguous bp changes 20:04
i.e M.K,S etc
pyrimidine Bio::Tools::CodonTable had much of this stuff
which I see you're working on 20:05
takadonet most of still stub methods 20:06
20:08 bluescreen10 left
masak strangelyconsistent.org/blog/the-ge...-in-perl-6 20:11
20:11 zby_home left
pyrimidine masak: yep, still seems to work :) 20:11
masak yay 20:12
takadonet probably going to steal that.... 20:13
already have a working version but it's the p5 way :P
pyrimidine That could very easily be applied for other codon tables
masak please don't put the one-liner as-is in production code :P
20:13 kaare__ is now known as kaare_
pyrimidine oh, come on 20:14
:)
masak when one optimizes for readability, it's slightly longer.
takadonet we will see
:)
pyrimidine masak: very true. though it's fairly readable now, if you know how to read it.
takadonet ya
masak I disagree :)
pyrimidine What would you improve on? 20:15
masak I think I'd do the long string as a hash table, laying it out so that the codons stood out.
I'd definitely have things as several shorter statements, instead of one long.
pyrimidine ok 20:16
masak descriptive variable names.
stuff like that.
takadonet well atm I'm still in getting it to work stage
Once everything is ported over and all the tests are passing. Then it's time to improve it 20:17
pyrimidine nods 20:18
takadonet Since we have no real users, we can do whatever we want, when we want 20:19
masak I think I'd write the hash table as "AAA" => "K" etc for extra clarity. 20:24
one can lay it out in rows and columns so as to group things in the usual way.
20:24 bluescreen10 joined 20:34 dakkar left