»ö« Welcome to Perl 6! | perl6.org/ | evalbot usage: 'perl6: say 3;' or rakudo:, niecza:, std:, or /msg camelia perl6: ... | irclog: irc.perl6.org | UTF-8 is our friend! Set by sorear on 25 June 2013. |
|||
00:01
abnorman joined
00:29
btyler left
00:39
btyler joined
00:40
berekuk left
00:44
btyler left
00:45
sidus left
00:50
berekuk joined
01:00
jnap joined
01:02
benabik joined
01:05
berekuk left
01:06
stevan_ joined
01:10
jaldhar joined
01:14
berekuk joined
01:22
Faxmachinen joined,
FROGGS_ joined
|
|||
Faxmachinen | Why, auto-flattening, why? Aaarghhh | 01:23 | |
(1..3) Xx (1..3) # Useful | |||
(1..3) Xxx (1..3) # Useless | |||
01:24
FROGGS left
01:25
Faxmachinen left
01:33
jnap left
01:35
berekuk left
02:05
thou joined
02:07
abnorman left
02:27
skinkitten left
02:32
BenGoldberg left
02:34
sunnavy left,
sunnavy joined,
moritz joined
|
|||
TimToady | I'd call it a bug | 03:07 | |
03:19
preflex_ joined,
ChanServ sets mode: +v preflex_
03:20
preflex left,
preflex_ is now known as preflex
|
|||
[Coke] spends a ton of time trying to do a bisect of rakudo before realizing it's an nqp issue. | 03:32 | ||
TimToady | nqp is also nqr | ||
03:35
ayu joined
03:37
ayu left
|
|||
[Coke] | pmurias - github.com/perl6/nqp/issues/114 - breaks rakudo.parrot | 04:01 | |
.to pmurias - github.com/perl6/nqp/issues/114 - breaks rakudo.parrot | |||
yoleaux | [Coke]: I'll pass your message to pmurias. | ||
04:06
Khisanth joined
04:24
kaare__ joined
04:41
kaare__ is now known as kaare_
|
|||
moritz | \o | 05:23 | |
05:36
SamuraiJack joined
05:56
Targen left,
Targen_ joined
06:05
lizmat joined
06:14
woolfy joined
06:29
berekuk joined
06:37
_ilbot left,
_ilbot joined
06:41
berekuk left
06:45
_ilbot left,
_ilbot joined
06:47
_ilbot left,
_ilbot joined
06:52
_ilbot left,
_ilbot joined
|
|||
TimToady | ilbot seems to be flapping | 06:53 | |
moritz | TimToady: I'm trying to debug why it doesn't log in #bioperl | 06:55 | |
TimToady | okay, just wanted to make sure you knew | 06:56 | |
06:58
Targen_ left
07:04
woolfy left
07:06
abnorman joined
07:07
lizmat left
07:25
FROGGS_ is now known as FROGGS
|
|||
FROGGS | [Coke]: I know what is from with the .fmt tests... | 07:25 | |
07:26
dwarring joined
|
|||
FROGGS | hold on... | 07:27 | |
it is working now O.o | |||
07:32
iSlug joined
|
|||
FROGGS | .... partially | 07:36 | |
07:39
pdcawley joined
07:41
denis_boyun joined
07:49
lizmat joined,
sidus joined
07:54
donaldh joined
|
|||
donaldh | [Coke]++ for dissecting S02-magicals/env.rakudo.parrot 14 - ENV members persist to child processes | 07:58 | |
Saved me that pain | 08:00 | ||
Will investigate what is up with it on JVM | 08:01 | ||
08:02
spider-mario joined
|
|||
donaldh will ^^ | 08:03 | ||
08:13
donaldh left
|
|||
lizmat | r: use MONKEY_TYPING; augment class Any { method foo { say "foo" } }; Cool.foo # is there any way to fix this ? | 08:18 | |
camelia | rakudo cacdd1: OUTPUT«No such method 'foo' for invocant of type 'Cool' in block at /tmp/JRB5GOYlzT:1» | ||
lizmat | amy I mossing something? | ||
missing? | |||
lizmat needs more tea | |||
FROGGS | r: use MONKEY_TYPING; augment class Any { method foo { say "foo" } }; augment class Cool { }; Cool.foo | ||
camelia | rakudo cacdd1: OUTPUT«foo» | 08:19 | |
FROGGS | lizmat: it is about method caches or so | ||
lizmat | yes, so is there a way to rebuild method caches? | ||
FROGGS | I guess so, since augment seem to do that | ||
lizmat | should the close of augment class Any not rebuild method caches? | ||
FROGGS | well, it does it for Any, but it needs to do it for every class that inherits from Any | 08:20 | |
lizmat | yup | ||
so this qualifies as a rakudo bug ? | |||
FROGGS | I think so | 08:21 | |
and there already might be a ticket | |||
I am sure there is at least one | |||
lizmat | ok :-) then I can mention that in my talk at FROSCON | ||
FROGGS | :o) | 08:22 | |
have fun there | |||
lizmat wishes the seats were better | |||
they're killing me | |||
FROGGS | ó.ò | 08:23 | |
lizmat | definitely intended for small(ler), young(er) people | ||
08:29
xinming left
08:30
xinming joined
08:34
xenoterracide left
|
|||
jnthn | Yes, the augment bug is already in RT | 08:42 | |
lizmat | jnthn: any pointers on fixing this? | 08:43 | |
something with Mu.WALK ? | |||
jnthn | lizmat: It needs design work. | ||
lizmat: Not just implementation. | |||
lizmat | :-( | ||
jnthn | No, nothing like that. | ||
Essentially, it needs a publication mechanism, but to make sure we can still collect anonymous subclasses it needs to have weak semantics. | 08:44 | ||
lizmat | ah, yes, of course | ||
lizmat had forgotten about anonymous subclasses | |||
jnthn | :) | 08:45 | |
08:46
SamuraiJack left,
teslos joined
08:50
teslos_ joined
08:51
pdcawley left
|
|||
timotimo | thingsfittingperfectlyintothings.tumblr.com/ - look what's the newest post :3 | 08:52 | |
08:53
teslos left,
teslos_ is now known as teslos
|
|||
moritz | timotimo++ | 08:54 | |
timotimo is spreading the "perl is not dead" meme | 08:55 | ||
08:55
eternaleye left
|
|||
jnthn | :D | 08:55 | |
lizmat | .oO( I always thought tuits were for empowering ) |
08:56 | |
FROGGS | timotimo++ | 08:57 | |
09:04
eternaleye joined
09:11
iSlug left
09:12
Rotwang joined
09:18
dmol joined
|
|||
dalek | p: 63d7244 | (David Warring)++ | src/vm/jvm/runtime/org/perl6/nqp/runtime/Ops.java: thinko on unicode surrogate range |
09:20 | |
p: 3c2b6e3 | jonathan++ | src/vm/jvm/runtime/org/perl6/nqp/runtime/Ops.java: Merge pull request #115 from dwarring/master thinko on unicode surrogate range |
|||
09:27
ksh joined
|
|||
masak | good antenoon, #perl6 | 09:29 | |
jnthn | ante', masak | 09:30 | |
moritz | \o masak, jnthn, * | 09:31 | |
moritz still actively waits for t3 reviews | |||
jnthn | .oO( moritz does busy waiting? he's as bad as the select impl... ) |
09:32 | |
timotimo | i hope you still have some cores free for breathing, eating, sleeping... | ||
moritz | if not, you must send me an interrupt now and again :-) | 09:35 | |
09:37
eternaleye left,
eternaleye joined
|
|||
lizmat has fetched a better seat from the hall area | 09:37 | ||
timotimo | oh, huh? the bless change is already in the released rakudo? | 09:38 | |
oh | |||
moritz | I doubt that | ||
timotimo | that's just the specs | ||
reading comprehension fail :) | |||
09:39
iSlug joined
09:40
rindolf joined
|
|||
jnthn | airport & | 09:40 | |
09:40
bruges left
09:41
bruges joined
09:46
donaldh joined
|
|||
moritz spectests a patch that removes the * from Mu.bless | 09:46 | ||
dalek | p: b563f22 | (Tobias Leich)++ | src/HLL/sprintf.nqp: fix sprintfdirectives, which returns an int |
09:48 | |
09:48
jnthn_ joined
09:49
odoacre joined
|
|||
dalek | kudo/nom: 83180ce | (Tobias Leich)++ | / (2 files): properly box sprintfdirective's return value |
09:49 | |
09:50
iSlug left,
zakharyas joined
|
|||
FROGGS | [Coke]: t/spec/S02-names-vars/fmt.t is clean now again under rakudo@jvm | 09:51 | |
dalek | p: 430d89b | (Donald Hunter)++ | tools/build/Makefile-JVM.in: Add rule for running individual test files. |
||
p: 448aafa | (Donald Hunter)++ | src/HLL/sprintf.nqp: Merge branch 'master' of github.com:perl6/nqp |
|||
donaldh | Did I just do something wrong there? | 09:53 | |
why the double commit message? | |||
09:53
berekuk joined
|
|||
FROGGS | donaldh: you just merged my changes that I commited a few seconds before into yours | 09:54 | |
not to worry | |||
donaldh | Safely ignorable? | ||
FROGGS | int this case yeah, this would happen with all commits if you dont pull before btw | 09:55 | |
so, best thing is you pull directly before adding+committing+pushing things, and don't wait too long to push | 09:56 | ||
dalek | p: 44190db | (Donald Hunter)++ | src/vm/jvm/runtime/org/perl6/nqp/io/ (2 files): Add generic type to remove compiler warning. |
09:57 | |
donaldh | FROGGS: I did pull before. How do I update my branch (with commits) before pushing? | 09:58 | |
FROGGS: to avoid this message in future ? | |||
FROGGS | donaldh: I think this does it: git pull --rebase origin master | ||
donaldh | Ah, okay, thanks. | 09:59 | |
lizmat | git pull --rebase | ||
is generally enough | |||
FROGGS | donaldh: the problem with this is that you might discard changes of others if there are intersections | ||
donaldh wishes git had saner default behaviour | |||
FROGGS | it is sane, because it does not merge automatically | 10:00 | |
donaldh | okay, so the rebase ensures that it is not mistaken for code I merged before pushing. | ||
FROGGS | donaldh: it basically rewinds before your pending commit, applies the remote commit, then it applies yours | 10:01 | |
lizmat | also, a git pull --rebase may result in conflicts that you want to resolve before pushing | ||
donaldh | agreed. The saner default behaviour I'd like is for pull to prefer rebase for a remote tracking branch. | 10:02 | |
FROGGS | lizmat: yeah, which can be nasty if you are not used to git | ||
donaldh | I'm used to ClearCase where the branching / merging model is very similar. Can do this in my sleep in ClearCase. git is still catching me out occasionally. | ||
arnsholt | pull --rebase is nice when it works automatically, but can cause all kinds of grief when something goes wrong with the rebase | 10:03 | |
donaldh | (similar if you ignore the massive difference between files and change sets, that is) | ||
10:04
denis_boyun left,
denisboyun joined
|
|||
arnsholt | Someone I know receommended to always pull --rebase to a git beginner I know, which eventually resulted in tears of course, since he had no idea what rebase did (and didn't read the error messages from git >.<) | 10:04 | |
In the end I got to scold both of them for being idiots =) | 10:05 | ||
FROGGS | *g* | ||
masak | an interesting take on the XY problem: weblogs.asp.net/alex_papadimoulis/a...08925.aspx | 10:07 | |
arnsholt: there you go, that's why --rebase is not the default for pull. | |||
(even though it's loads more convenient once you know how to use it) | |||
10:08
ksh left,
ksh joined
|
|||
arnsholt | Exactly | 10:09 | |
I was the most annoyed with the guy who just told the beginner to always pull --rebase, to be honest | 10:10 | ||
You can always trust beginners to not read the error messages, but more advanced users should know better than to recommend blindly doing dangerous things | |||
masak | every person needs two teachers. one who tells the lie-to-children, and one who tells the detailed truth. | 10:11 | |
lizmat | only beginners don't read error messages ? | ||
masak | and preferably in that order, with some delay to digest things :P | ||
FROGGS | lizmat: that is a good definition to draw the line, yes | 10:12 | |
arnsholt | You can probably estimate proficiency by the probability P of the user reading the error messages. As P approaches 1, skill level approaches wizard level =) | 10:13 | |
masak | necessary but not sufficient. :) | ||
arnsholt | Most likely, yeah | ||
10:13
virtualsue joined
10:19
teslos left
|
|||
timotimo | fwiw, i usually do a git fetch before i want to push, so that i can see what's going on and then i can decide to rebase or merge | 10:20 | |
it's now two steps, but it feels righter tom e | |||
moritz | IMHO the biggest problem with git pull --rebase is that it removes merge commits you made locally | ||
e.g. you merge a pull request, somebody was faster and pushed to master (or nom :-) first, your push is rejected | 10:21 | ||
if you now do a 'git pull --rebase', it flattens out the commits from the pull request, instead of making it a merge | |||
ss/making it/keeping it as/ | 10:24 | ||
10:24
hummeleBop joined
10:26
ksh left
|
|||
timotimo | moritz: so will invocations of .bless with * as first argument still work, but warn about deprecation? or will these fail? | 10:26 | |
moritz | timotimo: I'm afraid they will misbehave in weird ways, because bless in rakudo also has a positional slurpy argument | 10:27 | |
maybe I should add a heuristic to that one | 10:28 | ||
timotimo | it would be hard to handle the progression throughout the ecosystem if we turn it around 100% between releases i fear | ||
10:32
berekuk left,
virtualsue left,
Faxmachinen joined
10:34
berekuk joined
|
|||
Faxmachinen | Anyone know how to get around auto-flattening when grepping? | 10:34 | |
10:34
celesta joined
|
|||
celesta | ahoi #perl6 | 10:35 | |
Faxmachinen | Hi celesta | ||
celesta | r: class Actions { method TOP($/) { for $/ { my $bar := $_.ast; say "found $bar"; } }; method foo($/) { make ~$/; } }; my grammar G { rule TOP { ^ <foo>* $ }; rule foo { \w } }; my $actions = Actions.new; say G.parse( "a b c", :actions( $actions ) ); | 10:37 | |
camelia | rakudo 83180c: OUTPUT«use of uninitialized value of type Any in string context in method TOP at /tmp/mKzfAZZIYo:1found 「a b c」 foo => 「a 」 foo => 「b 」 foo => 「c」» | ||
celesta | i dont understand, why is $bar not initialized? | ||
FROGGS | r: class Actions { method TOP($/) { for $<foo> { my $bar := $_.ast; say "found $bar"; } }; method foo($/) { make ~$/; } }; my grammar G { rule TOP { ^ <foo>* $ }; rule foo { \w } }; my $actions = Actions.new; say G.parse( "a b c", :actions( $actions ) ); | 10:38 | |
camelia | rakudo 83180c: OUTPUT«found a found b found c「a b c」 foo => 「a 」 foo => 「b 」 foo => 「c」» | ||
FROGGS | celesta: in TOP the <foo> is array-ish, not $/ | 10:39 | |
celesta | ahhh, thank you very much | ||
10:44
SamuraiJack joined
|
|||
moritz | ./perl6 -e 'Mu.bless(*)' | 10:47 | |
Passing an object candidate to Mu.bless is deprecated in block at -e:1 | |||
10:49
lizmat left
|
|||
dalek | ecs: 41296cb | moritz++ | S12-objects.pod: [S12]: bless is a method, not a function |
10:50 | |
10:53
berekuk left
10:54
grondilu joined,
abnorman left
10:55
berekuk joined
|
|||
dalek | kudo/nom: 5caf928 | moritz++ | src/ (13 files): remove the candidate argument to bless warn when it is still passed along. |
10:56 | |
10:57
rindolf left
|
|||
Faxmachinen | rakudo: say ((1..3) Xxx (1..3)).perl; # Any way to get a lol out of this? | 11:01 | |
camelia | rakudo 83180c: OUTPUT«(1, 1, 1, 1, 1, 1, 2, 2, 2, 2, 2, 2, 3, 3, 3, 3, 3, 3).list» | ||
11:03
donaldh left
11:06
Rotwang left
11:16
xinming left
11:20
xinming joined
11:22
dmol1 joined
11:23
dmol left
11:24
zakharyas left
11:26
dmol joined,
dmol1 left
11:38
dmol1 joined,
dmol1 left
11:41
dmol left
11:43
xinming left
|
|||
timotimo | r: say ((1..3) X, (1..3)).lol.perl.say | 11:45 | |
camelia | rakudo 5caf92: OUTPUT«LoL.new(ListIter.new())True» | ||
11:45
xinming joined
|
|||
jercos | r: ((1..3) X, (1..3)).lol.perl.say | 11:48 | |
camelia | rakudo 5caf92: OUTPUT«LoL.new(ListIter.new())» | ||
11:49
ksh joined,
berekuk left
11:50
berekuk joined,
rindolf joined,
pecastro left
|
|||
moritz | r: say ((1..3) X, (1..3)).tree.perl | 11:53 | |
camelia | rakudo 5caf92: OUTPUT«([1, 1], [1, 2], [1, 3], [2, 1], [2, 2], [2, 3], [3, 1], [3, 2], [3, 3]).list» | ||
moritz | Faxmachinen: ^^ | ||
rindolf | moritz: what does .tree do? | 11:58 | |
11:59
rindolf left
12:03
berekuk left
|
|||
Faxmachinen | moritz: Interesting, but what I want is something like what the single x operator does | 12:05 | |
grondilu | r: say ((1..3) X, (1..3)).perl | ||
camelia | rakudo 5caf92: OUTPUT«((1, 1), (1, 2), (1, 3), (2, 1), (2, 2), (2, 3), (3, 1), (3, 2), (3, 3)).list» | ||
12:05
berekuk joined
|
|||
Faxmachinen | rakudo: ((1..3) Xx (1..3)).perl | 12:05 | |
camelia | rakudo 5caf92: ( no output ) | ||
Faxmachinen | say rakudo: ((1..3) Xx (1..3)).perl | ||
Damn it | 12:06 | ||
grondilu | r: say * | ||
camelia | rakudo 5caf92: OUTPUT«*» | ||
Faxmachinen | rakudo: say ((1..3) Xx (1..3)).perl | ||
camelia | rakudo 5caf92: OUTPUT«("1", "11", "111", "2", "22", "222", "3", "33", "333").list» | ||
Faxmachinen | Like that buy as lol instead of los | ||
*but | |||
12:09
rindolf joined
|
|||
timotimo | r: say ((1..3) Xvv (1..3)).tree.perl | 12:10 | |
camelia | rakudo 5caf92: OUTPUT«===SORRY!=== Error while compiling /tmp/UsfK4Lr7fCBogus statementat /tmp/UsfK4Lr7fC:1------> say ((1..3) X⏏vv (1..3)).tree.perl expecting any of: whitespace» | ||
timotimo | r: say ((1..3) Xxx (1..3)).tree.perl | ||
camelia | rakudo 5caf92: OUTPUT«(1, 1, 1, 1, 1, 1, 2, 2, 2, 2, 2, 2, 3, 3, 3, 3, 3, 3).list» | ||
timotimo | righto | ||
r: say ((1..3) X (1..3)).map({ ($^a xx $^b).item }).perl | |||
camelia | rakudo 5caf92: OUTPUT«((1,).list.item, (1, 1).list.item, (1, 1, 1).list.item, (2,).list.item, (2, 2).list.item, (2, 2, 2).list.item, (3,).list.item, (3, 3).list.item, (3, 3, 3).list.item).list» | ||
timotimo | :( | ||
Faxmachinen | That looks good. Thanks! | 12:11 | |
timotimo | i don't think it looks good | 12:12 | |
it's a workaround at best | |||
Faxmachinen | True | ||
timotimo | but it'll do at least | ||
12:17
lizmat joined,
kaare_ left
|
|||
bbkr | r: my $b = Buf.new(0x03); say $b ~> 1 | 12:22 | |
camelia | rakudo 5caf92: OUTPUT«===SORRY!=== Error while compiling /tmp/QRj3ul_x8yPreceding context expects a term, but found infix > insteadat /tmp/QRj3ul_x8y:1------> my $b = Buf.new(0x03); say $b ~>⏏ 1» | ||
bbkr | n: my $b = Buf.new(0x03); say $b ~> 1 | ||
camelia | niecza v24-88-g1f87209: OUTPUT«Unhandled exception: Cannot call new; none of these signatures match: Mu, *%_ at /tmp/FBZwenjlfv line 1 (mainline @ 3)  at /home/p6eval/niecza/lib/CORE.setting line 4579 (ANON @ 3)  at /home/p6eval/niecza/lib/CORE.setting line 4580 (module-CORE @… | 12:23 | |
bbkr | ^ bug or buffer bitwise shifts are NYI ? | ||
timotimo | i cannot find ~> in the specs :o | 12:26 | |
oh | 12:27 | ||
there it is | |||
bbkr | perlcabal.org/syn/S03.html#Multipli...precedence | ||
timotimo | ah, yes. NYI. | 12:28 | |
at least not in Buf.pm | |||
timotimo also doesn't really know how to properly implement that | 12:29 | ||
bbkr | ok, thanks. I'll convert to Int and use +> instead | 12:30 | |
timotimo | huge bigint? :) | 12:31 | |
bbkr | I'm working on IEE754 double precision converter, so 8 bytes max. | ||
timotimo | ah, that's good then | 12:32 | |
12:34
iSlug joined
|
|||
bbkr | I need it for BSON format (to push MongoDB development forward) and pack/unpack does not provide it yet like in P5 | 12:35 | |
lizmat | BSON? | 12:36 | |
bbkr | bsonspec.org/#/specification - it's like binary JSON, allows much after parallel encoding/decoding | 12:37 | |
lizmat | you mean e BER compressed integer? | ||
12:37
Moukeddar joined
|
|||
lizmat | ah | 12:37 | |
why not Sereal ? | |||
bbkr | because MongoDB uses BSON only | 12:38 | |
lizmat | :-( | ||
so you're saying pack() in Perl 5 supports BSON ? | 12:39 | ||
how? | |||
bbkr | no, I'm saying it supports IEE754 format | ||
lizmat | (just curious) | ||
bbkr | which is used by BSON | ||
12:40
teslos joined
|
|||
lizmat | can't find anything related to pack() and IEE754 | 12:40 | |
bbkr | print unpack("H*", pack("d*", 5.05)) | 12:42 | |
lizmat | ah, IEEE | ||
bbkr | indeed | 12:43 | |
I haven't found any smart way to do the same in P6 (which will use internal IEEE representation without implementing whole conversion algorithm) | 12:45 | ||
like "my num64 $x = 5.05; $x.get-raw-p6-object.get-body-bytes" | 12:48 | ||
12:50
rindolf left
|
|||
lizmat is stumped a bit as well | 12:59 | ||
13:03
PacoAir joined
13:06
berekuk left
|
|||
timotimo | python has this "buffer interface" idea where you can ask a couple different objects for its bytes | 13:08 | |
lizmat | I naively came up with someting like | ||
my num64 $x = 5.05; printf( "%x", my Buf64 $ := $x ) | 13:09 | ||
13:10
berekuk joined
13:13
dmol joined
13:15
berekuk left,
Moukeddar left
13:18
berekuk joined
13:21
berekuk left
13:27
mtk left,
bakedb_ joined
13:28
awwaiid_ joined
13:29
awwaiid left
13:30
mtk joined
13:33
PerlJam joined,
mtk left
13:34
lizmat left,
bonsaikitten joined,
lizmat joined,
dmol left,
dmol joined
13:36
mtk joined
13:40
prammer joined
13:46
dmol left
13:47
dmol joined
13:52
ajr joined,
ajr is now known as Guest2086
|
|||
lizmat | spectesting HEAD fails with t/spec/S02-magicals/env.rakudo.parrot (Wstat: 0 Tests: 16 Failed: 1) | 13:53 | |
Failed test: 14 | |||
is this known? | 13:54 | ||
JimmyZ | yes | 13:57 | |
13:59
denisboyun left
14:01
berekuk joined
|
|||
moritz | yes | 14:07 | |
14:07
berekuk left
|
|||
moritz | [Coke]++ tracked it down to an NQP commit by pmurias++ | 14:08 | |
www.youtube.com/watch?v=roNyfDHjtnc doesn't even get the Perl 6 method call syntax right | 14:18 | ||
tadzik | I don't think I can be disappointed by that... ;) | 14:20 | |
moritz stops watching | 14:21 | ||
diakopter | what is it? | ||
moritz | comparing perl 5 with unidiomatic and wrong perl 5 without any kind of actual argument just isn't interesting | ||
diakopter: one of the p6 "haters" talk from YAPC::EU | 14:22 | ||
14:24
berekuk joined
14:27
denis_boyun joined
|
|||
masak | moritz: wow -- all I got from people's reports was that those talks were "confused". now that I watch the talk, I realize how apt a description that was. | 14:32 | |
I'm now at the 4-minute mark, and all that's been clear so far is that I really should look into hieroglyphics. | 14:33 | ||
he seems to be speaking in patterns of the shape "I could say X, but that's wrong, so I'm not saying that". | 14:35 | ||
and so he's taking up time and space but not really getting somewhere. | 14:36 | ||
lizmat | decommute& | 14:38 | |
14:38
lizmat left
14:40
Psyche^_ joined
|
|||
masak | now I learned that as much as he likes memes, he does not like zombies. | 14:40 | |
14:40
Psyche^ left
14:45
berekuk left
14:46
berekuk joined
|
|||
masak | what is clear, is that the Perl 6 code on these slides is written by someone who doesn't write Perl 6 code. | 14:46 | |
moritz | and it's not run/tested either | 14:48 | |
tadzik | of course. If only there was a perl 6 release he'd have ran it | 14:49 | |
masak | we should get around to releasing something runnable one of these days | 14:51 | |
grondilu | rn: say ([1, 4, 5], [2, 1], []).grep(+*).min(:by(+*[*-1])); | 14:53 | |
camelia | niecza v24-88-g1f87209: OUTPUT«Unhandled exception: Excess arguments to List.min, unused named by at /home/p6eval/niecza/lib/CORE.setting line 0 (List.min @ 1)  at /tmp/I2PsJdkbFz line 1 (mainline @ 4)  at /home/p6eval/niecza/lib/CORE.setting line 4579 (ANON @ 3)  at /home/p6eva… | 14:54 | |
..rakudo 5caf92: OUTPUT«1 4 5» | |||
grondilu | r: say min :by(+*[*-1]), grep +*, [1, 4, 5], [2, 1], [] | ||
camelia | rakudo 5caf92: OUTPUT«2 1» | ||
grondilu | n: say ([1, 4, 5], [2, 1], []).grep(+*).min(:by(+*[*-1])); | ||
camelia | niecza v24-88-g1f87209: OUTPUT«Unhandled exception: Excess arguments to List.min, unused named by at /home/p6eval/niecza/lib/CORE.setting line 0 (List.min @ 1)  at /tmp/kqDzTzGWnq line 1 (mainline @ 4)  at /home/p6eval/niecza/lib/CORE.setting line 4579 (ANON @ 3)  at /home/p6eva… | ||
grondilu | r: say ([1, 4, 5], [2, 1], []).grep(+*).min(:by(+*[*-1])); | ||
camelia | rakudo 5caf92: OUTPUT«1 4 5» | ||
grondilu | r: say min :by(+*[*-1]), grep +*, [1, 4, 5], [2, 1], [] | ||
camelia | rakudo 5caf92: OUTPUT«2 1» | ||
grondilu expected the same result | |||
r: say min :by(+*[*-1]), [1, 4, 5], [2, 1] | 14:55 | ||
camelia | rakudo 5caf92: OUTPUT«2 1» | ||
grondilu | r: say ([1, 4, 5], [2, 1]).min(:by(+*[*-1])); | ||
camelia | rakudo 5caf92: OUTPUT«1 4 5» | ||
masak | finished watching the talk. I feel dumber than 20 minutes ago. | ||
14:58
Faxmachinen left
14:59
kaare_ joined
|
|||
FROGGS | I'd like to add comments... | 14:59 | |
grondilu | you can post it on reddit and talk about it there if you don't like youtube comments | 15:00 | |
celesta | bye #perl6 | 15:08 | |
15:08
celesta left
|
|||
Ulti | from inside a sub call is there a way to get the name of the variable passed as a parameter from where the sub was invoked? so: sub stuff (Int $thing) { return $thing;} my Int $other_thing = 1; stuff($other_thing); | 15:08 | |
I want to be able to know '$other_thing' was passed when inside the stuff sub | 15:09 | ||
tadzik | why would you ever want to do that? :) | ||
Ulti | funny everyone always asks that ;D gist.github.com/MattOates/6195743 | ||
I wrote some horrible Python to do it | |||
was wondering how less evil it is in Perl 6 if possible | 15:10 | ||
tadzik: basically I have a function that dumps variables to sqlite files and by default it uses the name of the variable passed in to create the table names | |||
15:10
rindolf joined
|
|||
moritz | r: sub varname(\a) { a.VAR.name }; say varname my $x = 42 | 15:10 | |
camelia | rakudo 5caf92: OUTPUT«$x» | ||
moritz | Ulti: ^^ | 15:11 | |
Ulti | dump($variable) is nicer than dump($variable,variable='variable') | ||
cool! | |||
tadzik | you can always do dump(:$variable) for less magic | ||
moritz | Ulti: Perl 6 also lets you write dump(:$variable) | ||
which is more reliable and just one character more | |||
though you don't know if @variable or $variable was passed in if it's an array | 15:12 | ||
Ulti | for what I wanted that doesn't really matter | ||
its funny how many people see that and just throw up in my face and start raging... I really don't see where the evil lies | 15:13 | ||
unless you have to do evil unmaintainable things to get the result, like my Python | |||
15:17
Rotwang joined
|
|||
grondilu | r: say ([1, 4, 5], [2, 1]).min(:by(+*[0])); | 15:20 | |
camelia | rakudo 5caf92: OUTPUT«1 4 5» | ||
grondilu | r: say ([1, 4, 5], [2, 1]).min(:by(+*[*-1])); | 15:21 | |
camelia | rakudo 5caf92: OUTPUT«1 4 5» | ||
grondilu | r: say ([1, 4, 5], [2, 1]).min(:by({ +.[*-1]})); | ||
camelia | rakudo 5caf92: OUTPUT«1 4 5» | ||
Ulti | ahh now I see Method 'name' not found for invocant of class 'Array' | ||
moritz | r: sub varname(\a) { a.VAR.name }; say varname my @a | 15:22 | |
camelia | rakudo 5caf92: OUTPUT«No such method 'name' for invocant of type 'Array' in sub varname at /tmp/XnlFcYkRl1:1 in block at /tmp/XnlFcYkRl1:1» | ||
moritz | r: sub varname(\a) { a.VAR.?name }; say varname my @a | ||
camelia | rakudo 5caf92: OUTPUT«Nil» | ||
moritz | Ulti: ah, will most likely only work for scalars | ||
r: sub varname(\a) { a.VAR.?name }; say varname &sin | |||
camelia | rakudo 5caf92: OUTPUT«sin» | ||
moritz | ... and callables | 15:23 | |
15:23
teslos left
|
|||
grondilu | r: say ([1, 4, 5], [-1, 1]).min(:by({ +.[*-1]})); | 15:24 | |
camelia | rakudo 5caf92: OUTPUT«-1 1» | ||
15:24
teslos joined
|
|||
grondilu | r: say ([-1, 4, 5], [-1, 1]).min(:by({ +.[*-1]})); | 15:25 | |
camelia | rakudo 5caf92: OUTPUT«-1 1» | ||
grondilu | r: say ([-2, 4, 5], [-1, 1]).min(:by({ +.[*-1]})); | ||
camelia | rakudo 5caf92: OUTPUT«-1 1» | ||
grondilu | r: say ([2, 4, 5], [1, 1]).min(:by({ +.[*-1]})); | ||
camelia | rakudo 5caf92: OUTPUT«1 1» | ||
grondilu | r: say ([1, 4, 5], [2, 1]).min(:by({ +.[*-1]})); | ||
camelia | rakudo 5caf92: OUTPUT«1 4 5» | ||
Ulti | is there a reason why hashes or arrays don't have a name? | ||
or is none of this specced and its really all evil even though its easy to do | 15:26 | ||
masak | my guess is that the same "scalar container" level-of-indirection isn't there for hashes and arrays. | ||
that doesn't seem to me to be reason enough for it not to work, however. | |||
Ulti | Parcels are ok | 15:27 | |
dmol | Hello! I have created test for bug Rakudo bug rt.perl.org/rt3/Public/Bug/Display...l?id=76896 . Here is its: bpaste.net/show/126089/ . Into which repository i should create an pull request? And where the test file should be stored? | 15:29 | |
Ulti | r: sub varname(\a) { a.VAR.?name }; say varname 1 | ||
camelia | rakudo 5caf92: OUTPUT«Nil» | ||
Ulti | r: sub varname(\a) { a.VAR.name }; say varname 1 | ||
camelia | rakudo 5caf92: OUTPUT«No such method 'name' for invocant of type 'Int' in sub varname at /tmp/Qdicwwbfz_:1 in block at /tmp/Qdicwwbfz_:1» | ||
Ulti | yeah so it's not the best plan ;) | ||
15:29
rindolf left
|
|||
Ulti | I guess you can catch that and then use the value | 15:30 | |
pmichaud | dmol: tests go into the "roast" repository. github.com/perl6/roast/ | ||
15:30
benabik left
|
|||
moritz | dmol: it should go into the perl6/roast repository (the one checked out as t/spec/ in your rakudo directory after you did 'make testable') | 15:30 | |
pmichaud | I'll let moritz++ take it from here. :) | 15:31 | |
moritz | dmol: and I can give you commit access if you tell me your github user name | ||
15:31
benabik joined
|
|||
moritz | dmol: and the best file to put it into is probably S06-other/anon-hashes-vs-blocks.t | 15:31 | |
dmol | moritz: you already provided it to me. Could you review the test? Is it ok? | 15:32 | |
moritz | dmol: looks fine | ||
FROGGS | I commented: www.youtube.com/watch?v=roNyfDHjtnc | 15:37 | |
I would like to takes his slides to the next conference and talk about how one would write Perl 6 instead | |||
dalek | ast: 5765af9 | (Steve Mynott)++ | S02-magicals/ (2 files): import of spectest for RT #78258 |
15:40 | |
roast: 5ab54fb | (Steve Mynott)++ | S05-transliteration/79778.t: | |||
roast: import of spectest for RT #79778 | |||
15:40
dalek left
|
|||
moritz | rakudo's ROADMAP still has this line: | 15:41 | |
2 ? re-organize Rakudo, build process, etc. for multi-backend | |||
FROGGS | done I'd say :o) | ||
15:41
dalek joined,
ChanServ sets mode: +v dalek
|
|||
FROGGS | I guess the fperez talk is worth looking? www.youtube.com/watch?v=pcJ6P8fzF80 | 15:45 | |
err, nperez | |||
ohh, I guess I'm mixing him up with sommebody else | 15:46 | ||
dalek | kudo/nom: c311139 | moritz++ | docs/ROADMAP: [ROADME] remove an item that has been done also downgrade typed exceptions, since it is mostly done |
15:48 | |
15:53
Guest2086 left,
ajr joined
15:54
ajr is now known as Guest32799
15:55
Guest32799 is now known as ajr_
|
|||
jnthn_ | evening, #perl6 | 15:59 | |
15:59
jnthn_ is now known as jnthn
|
|||
nwc10 | what is this UGT heresy? :-) | 16:00 | |
16:05
denis_boyun left
|
|||
masak | GOOD CENTRAL EUROPEAN LATE AFTERNOON TO YOU TOO, JNTHN | 16:06 | |
FROGGS | hi jnthn | 16:14 | |
moritz: that is done too, right? github.com/rakudo/rakudo/blob/c311...OADMAP#L47 | 16:15 | ||
16:15
iSlug left
|
|||
moritz | FROGGS: looks like it, yes | 16:20 | |
r: say 'a' ~~ /(a)?/' | 16:21 | ||
camelia | rakudo c31113: OUTPUT«===SORRY!=== Error while compiling /tmp/eq7uI1lKhlTwo terms in a rowat /tmp/eq7uI1lKhl:1------> say 'a' ~~ /(a)?/⏏' expecting any of: postfix infix stopper infix or meta-infix statement end… | ||
moritz | r: say 'a' ~~ /(a)?/ | ||
camelia | rakudo c31113: OUTPUT«「a」 0 => 「a」» | ||
FROGGS | hmmm | ||
moritz | r: 'a' ~~ /(a)?/ and say $0.^name | ||
camelia | rakudo c31113: OUTPUT«Match» | ||
moritz | yes, not an Array | ||
r: '' ~~ /(a)?/ and say $0.^name | |||
camelia | rakudo c31113: OUTPUT«Any» | ||
moritz | looks like Any instead of Nil, but I'm not quite sure about that these days :-) | 16:22 | |
FROGGS | r: say 'a' ~~ /a?/; say 'b' ~~ /b**0..1/ | ||
camelia | rakudo c31113: OUTPUT«「a」「b」» | ||
16:22
domidumont joined
|
|||
moritz | r: 'b' ~~ /b**0..1/ and say $0.WHAT | 16:22 | |
camelia | rakudo c31113: OUTPUT«(Any)» | ||
moritz | r: 'b' ~~ /(b)**0..1/ and say $0.^name | ||
camelia | rakudo c31113: OUTPUT«Parcel» | ||
FROGGS | k | ||
moritz | FROGGS: will you remove it from the ROADMAP, or should I? | 16:23 | |
FROGGS | I don't care :o) | ||
you can change it if it is still open in your editor | |||
moritz | it is | 16:24 | |
dalek | kudo/nom: 38a0f9a | moritz++ | docs/ROADMAP: [ROADMAP] remove another done item. FROGGS++ |
||
16:25
donaldh joined
|
|||
FROGGS | moritz: thank you | 16:25 | |
moritz | jnthn: am I right in assuming that the problem with making Buf mutable is the same as with ++ for native ints? | 16:33 | |
dmol | r: my @t; for < a b > { @t.push( sub { $OUTER::_ } ) }; say $_() for @t | 16:36 | |
camelia | rakudo c31113: OUTPUT«ab» | ||
16:37
Rotwang left
16:38
rindolf joined
|
|||
masak | yep, closures work on $_ too :) | 16:39 | |
16:45
Rotwang joined
|
|||
jnthn | moritz: Yes, same underlying problem. | 16:47 | |
moritz: I have an idea for how to fix it, just need time to work on fleshing it out | |||
dmol | does that bpaste.net/show/126117/ covers the problem rt.perl.org/rt3/Ticket/Display.html?id=74076 ? | 16:50 | |
16:54
pecastro joined
16:58
denis_boyun joined
17:03
denis_boyun left
|
|||
moritz | dmol: it covers half of the problem | 17:05 | |
rakudo: { say $OUTER::_ } | |||
camelia | rakudo c31113: OUTPUT«Nil» | ||
moritz | this one isn't covered by your tests | ||
dmol | moritz: thanks. I'll think what I can do. | 17:06 | |
moritz | r: my $x; { $x := $OUTER::_ }; use Test; ok $x === Nil, '$OUTER::_ works in mainline'; | 17:11 | |
camelia | rakudo 38a0f9: OUTPUT«ok 1 - $OUTER::_ works in mainline» | ||
dmol | r: say $OUTER::_ | 17:12 | |
camelia | rakudo 38a0f9: OUTPUT«(Any)» | ||
FROGGS | r: say [*.defined] (0, 2, 3) | ||
camelia | rakudo 38a0f9: OUTPUT«===SORRY!=== Error while compiling /tmp/GwyoSIWgY9Two terms in a rowat /tmp/GwyoSIWgY9:1------> say [*.defined] ⏏(0, 2, 3) expecting any of: postfix infix stopper infix or meta-infix statem… | ||
dmol | moritz: why it is nil for been called from inner block ? | ||
17:16
ajr_ left
|
|||
dmol | moritz: where should it be committed ( bpaste.net/show/126124/ ) ? Under t/spec/S04-blocks-and-statements/outer.t would be ok ? | 17:17 | |
17:23
denis_boyun joined
17:40
rindolf left
17:41
rindolf joined
17:46
rindolf left
17:47
rindolf joined
17:50
teslos left
17:51
donaldh left
18:01
denis_boyun left
18:25
Seganku left,
denis_boyun joined,
skulls joined
18:28
ksh left
|
|||
FROGGS | new videos of YAPC::NA: www.youtube.com/user/yapcna/videos?...mp;sort=dd | 18:29 | |
18:31
rindolf left
|
|||
masak | ooh | 18:38 | |
18:48
donaldh joined
|
|||
jnthn | ahaha...that really simple talk I gave on reversing grammars is there :P | 18:50 | |
masak | now more people can enjoy applied duality! \o/ | 18:58 | |
dalek | p: d3cfdd6 | (Donald Hunter)++ | tools/build/Makefile-Parrot.in: Add rule for running individual test files to parrot Makefile. |
19:01 | |
19:13
cognominal left
19:14
cognominal joined
|
|||
donaldh | nqp: nqp::istrue(nqp::null()) | 19:17 | |
camelia | nqp: OUTPUT«Null PMC access in get_bool()current instr.: '' pc 42 ((file unknown):40) (/tmp/_UNMsjTs7W:1)» | ||
jnthn | Oh noes a Null...oh, wait... :P | ||
donaldh | it causes an nqp test to fail on parrot | 19:18 | |
ok( nqp::istrue(nqp::null()) == 0, 'nqp::istrue on null'); | 19:19 | ||
jnthn | Hmm... | ||
I'd expect that to explode. | |||
donaldh | t/nqp/59-nqpops.t | ||
19:21
SamuraiJack left,
SamuraiJack joined
|
|||
timotimo | moritz: "ROADME"? :) | 19:21 | |
donaldh | > nqp::istrue(nqp::null()) | 19:22 | |
java.lang.NullPointerException | |||
moritz | timotimo: in sovjet ukraine, ROAD YOU! | ||
donaldh | nuke the test ? | ||
moritz | +1 | ||
jnthn | +1 | 19:25 | |
timotimo | :3 | ||
tadzik | haha, nuke it | ||
jnthn | I dunno what backend thinks it should do anything but die :) | ||
19:27
domidumont left
|
|||
donaldh | protect itself from dying maybe | 19:29 | |
donaldh needs a faster computer | 19:30 | ||
dalek | p: ee3f55a | (Donald Hunter)++ | / (7 files): First tests for file ops, plus fixes on JVM. |
19:46 | |
FROGGS | donaldh: why dont you add your tests to t/nqp/73 ? | 19:49 | |
donaldh | oh | 19:50 | |
sorry, why? | |||
FROGGS | why test readlinefh in two separate files? | 19:52 | |
donaldh | I have t/nqp/73-delegation.t - no test for readlinefh | 19:53 | |
FROGGS | donaldh: sorry | 19:57 | |
it is only here: github.com/MoarVM/MoarVM/tree/mast...p-cc/t/nqp | |||
then we merge them when moarvm goes into nqp... | |||
donaldh | The tests I have added check the interaction of various ops. We should merge and keep them all. I'll not write any more. I'd end up duplicating. | 20:00 | |
20:00
sidus left
|
|||
FROGGS | k | 20:01 | |
my chdir tests need to be skipped on jvm though | |||
until we have chidr emulation | 20:02 | ||
chdir* | |||
20:02
Mouq joined
|
|||
donaldh | hmm. chdir emulation is in rakudo, not nqp. | 20:02 | |
FROGGS | ahh, k | 20:04 | |
20:04
GranKain joined
|
|||
dalek | p: a49aa93 | (Donald Hunter)++ | tools/build/Makefile-Parrot.in: Fix typo in last Makefile commit. |
20:05 | |
Mouq | std: macro mac () { quasi :COMPILING { "aster" } } | ||
camelia | std 06636e4: OUTPUT«ok 00:01 50m» | ||
Mouq | r: macro hi () { quasi :COMPILING { "hello $s" } } | ||
camelia | rakudo 38a0f9: OUTPUT«===SORRY!=== Error while compiling /tmp/LLoo_gXNMUMissing blockat /tmp/LLoo_gXNMU:1------> macro hi () { quasi ⏏:COMPILING { "hello $s" } } expecting any of: statement list prefix or term prefix or… | ||
masak | Mouq: yeah. haven't gotten that far. :) | ||
it's up next, though. | 20:06 | ||
Mouq | :) | ||
20:06
kaare_ left
|
|||
Mouq | std: my $why = macro dumb () { quasi { "aster" } } | 20:07 | |
camelia | std 06636e4: OUTPUT«ok 00:01 51m» | ||
Mouq | r: my $why = macro dumb () { quasi { "aster" } } | ||
camelia | rakudo 38a0f9: OUTPUT«===SORRY!===Error while compiling block (source text: "my $why = macro dumb () { quasi { \"aster\" } }"): Error while compiling op call: Error while compiling block : Error while compiling op p6store (source text: "$why = macro dumb () { quasi { \"aster\" } }"): … | ||
20:07
SamuraiJack left
20:08
GranKain left
|
|||
jnthn | huh...how can just assigning it go wrong... | 20:08 | |
masak submits rakudobug | 20:09 | ||
moritz | it, like, doesn't return a proper code object? | ||
r: say (macro d() { quasi { 42 }}).perl | 20:10 | ||
camelia | rakudo 38a0f9: OUTPUT«()» | ||
moritz | r: say (macro d() { quasi { 42 }}).^name | ||
camelia | rakudo 38a0f9: OUTPUT«Parcel» | ||
moritz | WAT | ||
masak | indeed. | ||
r: say Macro | 20:11 | ||
camelia | rakudo 38a0f9: OUTPUT«(Macro)» | ||
jnthn | r: Macro.^set_name('Marco'); say Macro | ||
camelia | rakudo 38a0f9: OUTPUT«(Marco)» | ||
FROGGS | -.- | ||
masak takes aim at jnthn and fires | 20:12 | ||
masak pulls up his crossbow and blows on it | 20:13 | ||
jnthn takes a monad too the knee | |||
masak: You aren't meant to use *arrows*! | |||
*were | |||
d'oh :) | 20:14 | ||
diakopter | bolts | ||
crossbows fire bolts | |||
FROGGS | what weapon fires dots? | 20:15 | |
geekosaur | peashooter? :p | 20:16 | |
masak | FROGGS: video game spaceships. | ||
FROGGS | true | ||
dalek | : 39b25dd | (Tobias Leich)++ | / (2 files): skip unpack directives wit too less arguments |
||
FROGGS | that reminds me of rewriting my asteroids clone in perl 6 | 20:17 | |
masak | jnthn: I may fire bolts, but I don't really have a quarrel. :P | ||
Mouq | r: $masak.weapon = macro infix:<< -> >> ($a, $b) { quasi { say {{{$b}}}, " has been vanquished" } } | ||
camelia | rakudo 38a0f9: OUTPUT«===SORRY!=== Error while compiling /tmp/74_RjvvQICVariable '$masak' is not declaredat /tmp/74_RjvvQIC:1------> $masak.weapon⏏ = macro infix:<< -> >> ($a, $b) { quasi expecting any of: method arguments post… | ||
FROGGS submits Mouqbug | 20:18 | ||
:o) | |||
Mouq | "Tests needed" tag added | 20:22 | |
masak | FROGGS: s/too less/too few/ | ||
FROGGS | masak: ahh, thanks | 20:24 | |
masak: I watched my v5 talk today, my english is awful :o( | 20:25 | ||
Mouq | Talks are up? | ||
FROGGS | Mouq: yapcna talks and some yapceu talks | 20:26 | |
masak | FROGGS: fwiw, I didn't have any problems with your English in your v5 talk. | 20:27 | |
FROGGS | well yes, one should be able to understand what I said, but hearing myself was quite a bit surprising | 20:28 | |
jnthn hates hearing himself in videos | 20:29 | ||
I think it must be a common thing | |||
masak | hearing oneself is always... unsettling. | ||
tadzik | yep | ||
masak | comes from being behind one's eyes most of the time, I guess. | ||
FROGGS | jnthn: your talks are awesome | ||
jnthn: I've seen the "what if grammars can...", it was cool! | |||
I've expected more bad puns, but it was okay :o) | 20:30 | ||
lue | .oO(jnthn, with less bad puns than expected? I can barely parse the notion.) |
20:31 | |
jnthn | FROGGS: Yes, but we overdosed on them in the debugger talk... :) | 20:33 | |
FROGGS | I liked the row of puns about jnhtn being a train very much :o) | ||
colomon | Well, if you expect every sentence to be a pun, you might be disappointed... | ||
FROGGS | colomon: true | ||
colomon | btw, as long as I'm talking here, the module smoke test failed overnight. | 20:34 | |
FROGGS | :/ | ||
colomon | ===SORRY!=== | ||
Could not find Shell::Command in any of: /home/smoker/rakudo/install/lib/parrot/5.5.0-devel/languages/perl6/site/lib, | |||
tadzik | aw | ||
colomon | etc…. for every module it tried, I think. | ||
tadzik | maybe the File::Tools split? | 20:35 | |
colomon | oh, did the parrot version bump? | ||
jnthn | colomon: If you stop talking here, will it stop failing? :P | ||
tadzik | dunno | ||
colomon | jnthn: :p | ||
FROGGS | nqp --version | ||
This is nqp version 2013.08-18-g44190db built on parrot 5.5.0 revision RELEASE_5_5_0 | |||
no, still 5.5.0 | |||
colomon should probably analyze these things before reporting them.... | 20:36 | ||
FROGGS .oO( parrot is pretty stable these days, only minor patches needed ) | |||
colomon should probably analyze these things before reporting them…. | |||
The first failure is in the pandabrew rebootstrap process. | 20:40 | ||
FROGGS | I can imagine that this is about the nqp::getenvhash patch | 20:41 | |
colomon | that sounds plausible. | ||
what did it do? | |||
FROGGS | github.com/perl6/nqp/commit/96bacd7 | 20:42 | |
colomon: what I had in v5 what breaks now is: nqp::atkey( nqp::getenvhash(), "V5DEBUG" ) | |||
this fails when V5DEBUG is unset | 20:43 | ||
I had to change it to: nqp::getenvhash()<V5DEBUG> | |||
20:44
spider-mario left
|
|||
donaldh | we talked about rakudo not mutating the process env. | 20:45 | |
nqp::getenvhash() is immutable on JVM so the special behaviour of modifying the process env is already removed on JVM. | |||
FROGGS | I'm just checking if V5DEBUG is set or not (to print debug messages) | ||
r: say nqp::getenvhash() | 20:46 | ||
camelia | rakudo 38a0f9: OUTPUT«("MANPATH" => "/home/p6eval/perl5/perlbrew/perls/perl-5.14.1/man:", "PERLBREW_VERSION" => "0.42", "PERLBREW_PERL" => "perl-5.14.1", "SHELL" => "/bin/bash", "TERM" => "screen", "USER" => "p6eval", "LS_COLORS" => "rs=0:di=01;34:ln=01;36:mh=00:pi=40;33:so=01;35:do=01;… | ||
FROGGS | r: say nqp::getenvhash()<SHELL> | ||
camelia | rakudo 38a0f9: OUTPUT«No such method 'postcircumfix:<{ }>' for invocant of type 'Hash' in block at /tmp/JZVrcrl4L8:1» | ||
FROGGS | nqp: say nqp::getenvhash()<SHELL> | ||
camelia | nqp: OUTPUT«Confused at line 2, near "say nqp::g"current instr.: 'panic' pc 14738 (src/stage2/gen/NQPHLL.pir:5238) (src/stage2/gen/NQPHLL.nqp:279)» | ||
FROGGS | nqp: say( nqp::getenvhash()<SHELL> ) | ||
camelia | nqp: OUTPUT«/bin/bash» | ||
FROGGS | nqp: say( nqp::atkey(nqp::getenvhash(), 'SHELL') ) | 20:47 | |
camelia | nqp: OUTPUT«/bin/bash» | ||
FROGGS | nqp: say( nqp::atkey(nqp::getenvhash(), 'TESTEST') ) | ||
camelia | nqp: OUTPUT«Null PMC access in get_string()current instr.: '' pc 768 (src/stage2/gen/NQPCORE.setting.pir:355) (src/stage2/NQPCORE.setting:665)» | ||
FROGGS | nqp: say( nqp::getenvhash()<TESTTEST> ) | ||
camelia | nqp: OUTPUT«» | ||
FROGGS | donaldh: you see the difference? | ||
donaldh | FROGGS: on parrot ? | 20:48 | |
FROGGS | yes | ||
timotimo | huh. emma watson just tweeted a link to things fitting perfectly into things, now a horde of peple saw the blog while my perl tuit in power socket picture was on top :3 | ||
FROGGS | O.o | ||
wow | |||
donaldh | FROGGS: yeah, that looks like it is down to pmurias commit. It's not been reverted, right? | 20:49 | |
20:51
pmurias joined
|
|||
tadzik | timotimo: haha, som much win :) | 20:51 | |
FROGGS | donaldh: no, I worked around it, and [Coke]++ made an issue | 20:52 | |
pmurias was not online yet I think | |||
timotimo | must you make a nissue out of everything? :) | ||
Mouq | Is infix:«~>» NYI? | 20:53 | |
timotimo | yes, it is | ||
Mouq | >~> | 20:54 | |
pmurias | re rakudo-parrot breakage with nqp I still need to implement nqp::shell fully to move parrot to the new semantics | ||
yoleaux | 04:01Z <[Coke]> pmurias: - github.com/perl6/nqp/issues/114 - breaks rakudo.parrot | ||
FROGGS | r: my $a = "abc"; say $a ~> 1 | ||
camelia | rakudo 38a0f9: OUTPUT«===SORRY!=== Error while compiling /tmp/_3IlinCSRAPreceding context expects a term, but found infix > insteadat /tmp/_3IlinCSRA:1------> my $a = "abc"; say $a ~>⏏ 1» | ||
pmurias | FROGGS: re v5 breaking with the changed nqp::getenvhash, it's now behaving like on the jvm | 20:55 | |
20:56
Khisanth left
|
|||
FROGGS | nqp-jvm: say( nqp::atkey(nqp::getenvhash(), 'TESTEST') ) | 20:56 | |
camelia | nqp-jvm: OUTPUT«No registered operation handler for 'getenvhash' in compile_op in as_jast in as_jast in <anon> in <anon> in <anon> in compile_op in as_jast in as_jast in <anon> in process_args in <anon> in <anon> in compile_op in as_jast in as_jast in <… | ||
FROGGS | gah, too old | ||
Mouq | 'prolly explain why perl can't parse "$x ~>= 8" in roast | 20:57 | |
timotimo | rakudo* | ||
Mouq | ^^ | ||
FROGGS | pmurias: true, but something must be different so the spectests fail | ||
pmurias | there are #ifdef's on parrot | 20:58 | |
FROGGS: so rakudo depends on crazy behavior of nqp::getenvhash on parrot | |||
FROGGS | pmurias: so the tests are wrong? or rakudo? | ||
pmurias | rakudo has an #ifdef parrot | 20:59 | |
rakudo-parrot depends on a specific version of nqp kept in a file? | 21:00 | ||
FROGGS | yes, tools/build/NQP_REVISION | ||
pmurias | and it hasn't just been updated right | 21:01 | |
? | |||
FROGGS | yes | 21:02 | |
that happens regularly | 21:03 | ||
21:03
dalek left,
dalek joined,
ChanServ sets mode: +v dalek
|
|||
pmurias | ahh, you just updated that and the env tests now break :/ | 21:04 | |
FROGGS | pmurias: the NQP_REVISION is nqp's revision you nead *at least* | ||
21:05
lue left
|
|||
FROGGS | is should always be possible to use a better version than then | 21:05 | |
than that* | |||
sure this isnt always possible | |||
pmurias | --gen-nqp doesn't use the one from NQP_REVISION? | 21:06 | |
FROGGS | it does use that version | ||
pmurias | we could introduce nqp::getenvpmc() to have the old semantics | 21:07 | |
FROGGS | cant we just fix rakudo? so it does not now-invalid operations? | 21:09 | |
use* | |||
jnthn | Fix Rakudo and we can have the same code-path for Parrot and JVM on this stuff. | ||
21:09
Khisanth joined
|
|||
FROGGS | yeah, I mean, it is just a hash after all | 21:10 | |
pmurias | I need to make nqp::shell on parrot recpect the env argument | ||
and the problem with that is the parrot portability os independence layer doesn't support calling a command with a specified set of env variables | 21:11 | ||
FROGGS | ahh, so you rely on modifying the env and passing this through? | 21:12 | |
pmurias | rakudo-jvm passes the env as a 3 argument to nqp::shell(...) | 21:13 | |
rakudo-parrot just modifies the env and calls the spawnw op | 21:14 | ||
FROGGS | hmmm, not ideal | ||
pmurias | jnthn: if I made nqp::shell work on rakudo-parrot@linux, would you help debug nqp::shell on rakudo-parrot@windows | 21:15 | |
? | |||
jnthn: as I don't have a windows machine set up for development | 21:16 | ||
I can write code for that I *suspect* will work ;) | 21:17 | ||
dalek | kudo-star-daily: 46d4bc7 | coke++ | log/version.log: today (automated commit) |
21:18 | |
21:18
dalek left
|
|||
FROGGS | pmurias: I have a windows box to test on | 21:19 | |
21:19
dalek joined,
ChanServ sets mode: +v dalek
|
|||
jnthn | pmurias: I can if nobody else can...I'm tied up with teaching these next 2 days, so FROGGS++ is probably a better bet. :) | 21:21 | |
FROGGS | yes, I'll have time for that :o) | ||
jnthn | \o/ | 21:22 | |
[Coke] | which airport should I be picking? | 21:23 | |
jnthn | 'night, #perl6 | 21:24 | |
pmurias | 'night | ||
colomon | \o | ||
donaldh is backlogging. | 21:27 | ||
21:28
araujo left
|
|||
donaldh | FROGGS: pmurias: It's trivial to change rakudo to have the same behaviour between parrot and jvm, i.e initialise @*ENV with nqp::getenvhash() and nothing more. | 21:28 | |
FROGGS | donaldh: but this won't get passed to a subprocess created by nqp::shell, which is a pir::spawn_IP or so | 21:29 | |
21:29
xenoterracide joined
|
|||
dalek | rl6-roast-data: 012193b | coke++ | / (5 files): today (automated commit) |
21:30 | |
21:30
araujo joined,
Nuru_Ziya joined
|
|||
donaldh | FROGGS: yeah, I wasn't sure of pmurias' goal with porting nqp::shell/3 to parrot. If parrot doesn't have the underlying op for passing env to spawn. | 21:30 | |
21:30
araujo left,
araujo joined
21:31
Nuru_Ziya left
|
|||
pmurias | donaldh: I plan to extend the nqp::shell op to pass the env to spawn tomorrow | 21:33 | |
donaldh | pmurias: extending parrot ? | ||
pmurias | nqp ops are written in C | 21:36 | |
21:36
virtualsue joined
21:38
pmurias left
21:42
donaldh left
|
|||
[Coke] hopes he has one of the slots! | 21:43 | ||
21:52
hummeleBop left
|
|||
[Coke] | .to lizmat Which airport to fly to, FRA or HHN?, and also, any rooms left at the hotel? | 21:54 | |
yoleaux | [Coke]: I'll pass your message to lizmat. | ||
21:55
colomon left
22:01
colomon joined
|
|||
timotimo | berekuk: i like what you did with the "grouped" news feed | 22:02 | |
22:02
xenoterracide left,
xenoterracide joined,
Rotwang left
|
|||
berekuk | timotimo: thanks! you can also "like" it here: questhub.io/realm/meta/quest/51cabd...f17a00001a :) | 22:02 | |
22:07
BenGoldberg joined
|
|||
Mouq rakudo PR #200 sum-bitted | 22:07 | ||
timotimo | can you throw a typed NYI exception instead? | 22:09 | |
Mouq | I was just following suit from an example above it. Lemme figure out how... | 22:11 | |
22:12
lue joined
|
|||
Mouq is becoming an expert at not yet implementing things | 22:12 | ||
timotimo | something like die X::NYI.new(what => '~> operator') | ||
22:13
virtualsue left
|
|||
Mouq | I see | 22:13 | |
timotimo | r: die X::NYI.new(what => "this thing"); | 22:14 | |
camelia | rakudo 38a0f9: OUTPUT«use of uninitialized value of type Any in string context in block at /tmp/YorxUsSYXr:1use of uninitialized value of type Any in string context not yet implemented. Sorry.  in block at /tmp/YorxUsSYXr:1» | ||
timotimo | "what" is apparently wrong. | ||
r: X::NYI.new(what => "this thing").^methods.perl.say | |||
camelia | rakudo 38a0f9: OUTPUT«(method message(X::NYI : Mu *%_) { ... }, method <anon>(Mu $self: Mu *%_) { ... }, method backtrace(Exception : Mu *%_) { ... }, method+{<anon>} throw(Exception : Mu *%_) { ... }, method+{<anon>} rethrow(Exception : Mu *%_) { ... }, method resumable(Exception : Mu … | ||
timotimo | r: X::NYI.new(what => "this thing").^methods.say | 22:15 | |
camelia | rakudo 38a0f9: OUTPUT«message <anon> backtrace throw rethrow resumable resume fail is-compile-time Str gist» | ||
Mouq | r: X::NYI.new(feature => "thing").throw | ||
camelia | rakudo 38a0f9: OUTPUT«thing not yet implemented. Sorry.  in block at /tmp/DJ25l2vT30:1» | ||
timotimo | ah, yes | ||
masak | r: X::NYI.new(feature => "closed timelike paths through wormholes").throw | 22:16 | |
camelia | rakudo 38a0f9: OUTPUT«closed timelike paths through wormholes not yet implemented. Sorry.  in block at /tmp/0KJEbyeg7s:1» | ||
dalek | rl6-roast-data: 7f0daa3 | coke++ | / (5 files): rename csv file to *.csv |
22:18 | |
22:19
PacoAir left
|
|||
timotimo watches the weird perl5/6 shakespeare/txtspk talk | 22:21 | ||
masak | 'night, #perl6 | 22:23 | |
timotimo | 'night masak :) | ||
Mouq | g'night | 22:25 | |
timotimo | strange how he uses . instead of -> in a later slide, but -> in perl6 earlier | 22:29 | |
Mouq | r: macro orcam ($a) { quasi { "{{{$a}}}" } }; say orcam 'perl' | ||
camelia | rakudo 38a0f9: OUTPUT«AST<-1471433078>» | ||
timotimo | probably prepared the slides like i usually do, as in hastily and only a few hours before the talk is supposed to be given | ||
Mouq | r: macro orcam ($a) { quasi { "{ {{{$a}}}.uc }TA" } }; say orcam 'spar' | 22:30 | |
camelia | rakudo 38a0f9: OUTPUT«===SORRY!=== Error while compiling /tmp/GEstGqpjgAError while compiling block : Error while compiling block orcam (source text: "{ quasi { \"{ {{{$a}}}.uc }TA\" } }"): Error while compiling block (source text: "{ \"{ {{{$a}}}.uc }TA\" }"): Error… | ||
Mouq | r: macro orcam ($a) { quasi { "{ <<<$a>>>.uc }TA" } }; say orcam 'spar' | 22:31 | |
camelia | rakudo 38a0f9: OUTPUT«No such method 'uc' for invocant of type 'Mu' in block at /tmp/0w1oRQPvaW:1 in any at /tmp/0w1oRQPvaW:1 in block at /tmp/0w1oRQPvaW:1» | ||
timotimo | i'm having a little bit of trouble understanding most things said, due to acoustics not being terribly good | 22:33 | |
there was really not much content there :( | 22:43 | ||
oh yeah! masaks talk is up! | |||
www.youtube.com/watch?v=3xE32d32z7c | 22:44 | ||
pretty rough acoustics on the mic in that video, too :( | |||
22:48
berekuk left
|
|||
timotimo | we can't see masak, but there's also a whole bunch of wasted video space around the slide :( | 22:49 | |
there masak is! | |||
ah, some friendly person is adjusting the camera | |||
as if they read my messages | |||
diakopter | timotimo: they did; they did! | 22:50 | |
dalek | rl6-roast-data: 5c4485f | coke++ | cull: normalize CSV output |
22:52 | |
rl6-roast-data: 932e676 | coke++ | perl6_pass_rates.csv: eliminate extranous space |
|||
22:56
berekuk joined
|
|||
timotimo | whenever masak is standing near the camera, the slides become pretty unreadable and the image gets much brighter | 23:00 | |
:( | |||
23:01
sftp left
23:02
berekuk left
23:03
cognominal__ joined
23:04
berekuk joined
23:05
berekuk left
|
|||
[Coke] | bah. no #comments in github csv | 23:05 | |
23:06
cognominal left
|
|||
timotimo | aaw :( | 23:07 | |
so we need an extra field? | 23:08 | ||
oh, you mean adding comments in the middle? | |||
[Coke] | github.com/coke/perl6-roast-data/b..._rates.csv | 23:09 | |
need to pull the info in the comments into the csv lines. | 23:10 | ||
23:10
FROGGS left
|
|||
timotimo | i'm now convinced that auto-focus should be disabled for talks like that in every case | 23:14 | |
ooooh i made a brief cameo! :) | 23:17 | ||
23:26
jaldhar left
23:29
jaldhar joined
23:31
denis_boyun left
23:40
_jaldhar joined
23:41
jaldhar left
|
|||
timotimo | one thing that annoys me about the reactive programming stuff if you do it with lazy lists and gather blocks in perl6 is that you have to do manual work to resume such a thingie | 23:42 | |
and if it does things like round-trip to web servers or something you have to know about that and simulate those or something :| | |||
Mouq | I'm still trying to figure out what colonpair is | 23:50 | |
23:51
abnorman joined
23:52
dmol left
|
|||
[Coke] | r: say :10<23> | 23:56 | |
camelia | rakudo 38a0f9: OUTPUT«23» | ||
[Coke] | r: say :23<99> | ||
camelia | rakudo 38a0f9: OUTPUT«216» | ||
Mouq | Ohhhhhhh. Thank you | 23:58 |