»ö« Welcome to Perl 6! | perl6.org/ | evalbot usage: 'p6: say 3;' or rakudo:, or /msg camelia p6: ... | irclog: irc.perl6.org or colabti.org/irclogger/irclogger_logs/perl6 | UTF-8 is our friend!
Set by moritz on 22 December 2015.
00:06 japhb joined 00:16 BenGoldberg joined 00:22 AlexDaniel left 00:43 labster left 00:44 labster joined 01:06 z3ndrag0n left 01:08 kalkin- joined 01:09 z3ndrag0n joined, Actualeyes joined 01:12 kalkin-_ left 01:14 AndyBotwin left 01:26 Roamer` left, wisti left 01:27 ssotka left 01:28 cdg_ joined 01:29 tbrowder left 01:31 cdg left 01:41 sortiz joined
sortiz \o #perl6 01:41
Zoffix \o
01:45 ilbot3 left
Zoffix Damn... I either have throat cancer or I've never noticed a botched tonsil removal job for 23 years :/ 01:47
01:47 nchambers left
Zoffix If I'll stop showing up, you know why ^_^ 01:47
01:47 ilbot3 joined 01:49 cdg_ left
gfldex m: Int.^method_table<abs>:delete; Int.^compose; say (-5).abs 01:50
camelia rakudo-moar 55c359: OUTPUT«Memory allocation failed; could not allocate 79872 bytes␤»
gfldex Zoffix: ^^^
01:50 grondilu left
Zoffix gfldex, doesn't seem to be working? :) 01:52
gfldex please don't pull the rug out from under objects 01:53
Xliff m: use NativeCall; class A is repr('CStruct') { has uint8 $.a; has uint8 $.b; has uint32 $.c; has uint32 $.d; has uint32 $.e; }; say nativesizeof(A) 01:55
camelia rakudo-moar 55c359: OUTPUT«16␤»
Xliff ^^ Expecting 14. 01:56
m: say 1 + 1 + 4 * 3
camelia rakudo-moar 55c359: OUTPUT«14␤»
geekosaur nope 01:57
sortiz Xliff, All C structs use some padding.
Zoffix Doesn't it need to pad?
geekosaur C compilers force alignment
Xliff Ooooo
Well that sucks.
geekosaur a and b will be contiguous but c will be aligned
to the next 4-byte boundary, hence 2 slack bytes 01:58
Xliff That explains some things.
I'm trying to read from a data file. I don't want alignment. That throws stuff off.
geekosaur (on x86/x86_64 at least; actual alignment depends on platform ABI)
gfldex m: use NativeCall; class A is repr('CStruct') { has uint32 $.a; has uint32 $.b; has uint32 $.c; has uint8 $.d; has uint8 $.e; }; say nativesizeof(A)
camelia rakudo-moar 55c359: OUTPUT«16␤»
geekosaur unpack should be in the ecosystem
Xliff So I can't avoid pack/unpack in this case? 01:59
sortiz Nope
Xliff kk
So is there any way to compute the size of a class/record WITHOUT the alignment issues?
Or do I need *gasp* naked values! 02:00
02:02 noganex_ joined 02:03 kurahaupo left 02:04 rgrinberg joined 02:05 noganex left 02:06 maybekoo2 left
Zoffix Can someone with access to OSX try out the hanging examples in this ticket? I can't reproduce it on neither Windows nor Linux. Wondering if we can close that ticket: rt.perl.org/Ticket/Display.html?id=126390 02:09
02:28 vendethiel joined 02:35 vendethiel left, pat_js left 02:36 addison left
dj_goku Xliff: so I am getting a segfault with the changes to inet_ntop, turning on debugging in moar so it'll be a few before I know more. 02:41
sortiz dj_goku, what is your problem with inet_ntop? 02:43
Xliff dj_goku, If you missed nine's correction, here it is: sub inet_ntop(int32, Pointer, Str, uint32) returns Str is native { * } 02:44
And make the corresponding change to the other sub.
02:46 espadrine left
dj_goku Xliff: still same seq fault 02:51
sortiz m: use NativeCall; sub inet_ntop(int32,Buf,utf8,uint32 --> Str) is native {*}; my $a = utf8.allocate(16); say inet_ntop(2,Buf.new(200,33,54,1), $a, 16);
camelia rakudo-moar 55c359: OUTPUT«200.33.54.1␤»
02:52 AndyBotwin joined
dj_goku sortiz: lol 02:52
sortiz: I have spent a few hours last night trying to figure this all out. Just for fun.
lol
balh
02:53 ssotka joined
sortiz dj_goku, the important detail is that the 'dst' argument should be a pre-allocated buffer. 02:56
02:56 wamba left
sortiz m: use NativeCall; sub inet_ntop(int32,Buf,utf8,uint32 --> Str) is native {*}; my $a = utf8.allocate(16); inet_ntop(2,Buf.new(127,0,0,1), $a, $a.elems); $a.decode.say; # Other way 02:58
camelia rakudo-moar 55c359: OUTPUT«127.0.0.1␀␀␀␀␀␀␀␤»
dj_goku m: use NativeCall; sub inet_pton(int32, Str, Buf is rw) returns int32 is natvie {*}; my $a = Buf; inet_pton(2, '127.0.0.1', $a); say $a.WHAT; 02:59
camelia rakudo-moar 55c359: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Can't use unknown trait 'is natvie' in a sub+{callable[int32]} declaration.␤at <tmp>:1␤ expecting any of:␤ rw raw hidden-from-backtrace hidden-from-USAGE␤ pure default DEPRECATED inl…»
dj_goku m: use NativeCall; sub inet_pton(int32, Str, Buf is rw) returns int32 is natvie {*}; my $a = Buf; inet_pton(2, '127.0.0.1', $a); say $a.WHAT;
camelia rakudo-moar 55c359: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Can't use unknown trait 'is natvie' in a sub+{callable[int32]} declaration.␤at <tmp>:1␤ expecting any of:␤ rw raw hidden-from-backtrace hidden-from-USAGE␤ pure default DEPRECATED inl…»
dj_goku m: use NativeCall; sub inet_pton(int32, Str, Buf is rw) returns int32 is native {*}; my $a = Buf; inet_pton(2, '127.0.0.1', $a); say $a.WHAT;
camelia rakudo-moar 55c359: OUTPUT«(signal SEGV)»
dj_goku m: use NativeCall; sub inet_pton(int32, Str, Buf is rw) returns int32 is native {*}; my $a = Buf.new; inet_pton(2, '127.0.0.1', $a); say $a.WHAT;
camelia rakudo-moar 55c359: OUTPUT«(signal SEGV)»
03:00 huggable left, huggable joined
dj_goku m: use NativeCall; sub inet_pton(int32, Str, Buf is rw) returns int32 is native {*}; my $a = Buf.new; inet_pton(2, '127.0.0.1', $a); 03:00
camelia rakudo-moar 55c359: OUTPUT«(signal SEGV)»
03:00 noganex joined
sortiz m: use NativeCall; sub inet_pton(int32, Str, Buf is rw) returns int32 is native {*}; my $a = Buf.allocate(4); inet_pton(2, '127.0.0.1', $a); say $a 03:01
camelia rakudo-moar 55c359: OUTPUT«Buf:0x<7f 00 00 01>␤»
03:04 noganex_ left
dj_goku ahh 03:05
sortiz m: use NativeCall; sub inet_pton(int32, Str, Buf --> int32) is native {*}; my $a = Buf.allocate(4); inet_pton(2, '127.0.0.1', $a); say $a; # In fact no 'is rw' required. 03:06
camelia rakudo-moar 55c359: OUTPUT«Buf:0x<7f 00 00 01>␤»
Xliff sortiz++ # Knowledge!
03:10 kid51 left
dj_goku m: use NativeCall; sub inet_pton(int32, Str, Buf is rw) returns int32 is native { * }; sub inet_ntop(int32, Buf, utf8, uint32 --> Str) is native { * }; my $s_addr = Buf.allocate(4); inet_pton(2, '127.0.0.1', $s_addr); my $ip = utf8.allocate(16); say inet_ntop(2, $s_addr, $ip, 16); # to put it all together. 03:11
camelia rakudo-moar 55c359: OUTPUT«127.0.0.1␤»
dj_goku sortiz: wow, so thanks. now I have to go figure out why that all worked and why. :D 03:12
sortiz The important detail with those calls is that all buffers need to be properly allocated. And you can consider .allocate of Blob/Buf as Perl6's malloc. 03:14
dj_goku ok. I haven't messed with C in ages and have forgotten all that knowledge. 03:15
03:22 Actualeyes left
MasterDuke .tell timotimo another minor update to add the line number in the source file as a column, github.com/MoarVM/MoarVM/pull/382 03:32
yoleaux MasterDuke: I'll pass your message to timotimo.
03:45 skids left 03:49 dj_goku left, molaf joined 03:50 canopus left, canopus joined 03:57 dj_goku joined 04:17 rgrinberg left, cbk left 04:28 khw left 04:34 Senji joined 04:46 Actualeyes joined 04:48 Cabanossi left 04:50 jack_rabbit joined 04:51 Cabanossi joined 04:56 Sgeo left 04:57 Sgeo joined 04:59 Upasaka left 05:00 Sgeo left, Sgeo joined 05:11 Sgeo left, zacts joined, Sgeo joined 05:14 BenGoldberg left 05:18 Sgeo left 05:20 CIAvash joined 05:24 vendethiel joined 05:36 Zoffix left 05:54 M-matthew left 05:56 kerframil left 05:57 zacts left, Sgeo joined 06:00 vendethiel left, z3ndrag0n left 06:01 M-matthew joined, z3ndrag0n joined 06:12 sortiz left 06:15 jack_rabbit left 06:17 ssotka left 06:22 nemo left 06:23 nemo joined, nemo is now known as Guest94483 06:31 firstdayonthejob joined 06:33 jack_rabbit joined
Xliff Does anyone know the best way to take data from unpack and properly convert it to a float? 06:38
m: use NativeCall; say nativecast(num32, Buf.new(<00 01 00 01>)); 06:39
camelia rakudo-moar 55c359: OUTPUT«2.35106044812595e-38␤»
Xliff m: use NativeCall; say nativecast(num32, Buf.new(<01 01 00 0>));
camelia rakudo-moar 55c359: OUTPUT«3.60133705331478e-43␤»
Xliff m: use NativeCall; say nativecast(num32, Buf.new(<00 01 00 00>)); 06:40
camelia rakudo-moar 55c359: OUTPUT«3.58732406867153e-43␤»
Xliff m: use NativeCall; say nativecast(num32, Buf.new(<00 00 01 00>));
camelia rakudo-moar 55c359: OUTPUT«9.18354961579912e-41␤»
06:42 firstdayonthejob left
Xliff m: use NativeCall; say nativecast(num32, Buf.new(<80 3f 00 00>)); 06:43
camelia rakudo-moar 55c359: OUTPUT«Type check failed in initializing element #1 to Buf; expected uint8 but got Str ("3f")␤ in any at gen/moar/m-Metamodel.nqp line 1736␤ in block <unit> at <tmp> line 1␤␤Actually thrown at:␤ in any at gen/moar/m-Metamodel.nqp line 3055␤ in a…»
Xliff m: use NativeCall; say nativecast(num32, Buf.new([80, 3f, 00, 00]>)); 06:44
camelia rakudo-moar 55c359: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Confused␤at <tmp>:1␤------> 3ll; say nativecast(num32, Buf.new([80, 37⏏5f, 00, 00]>));␤ expecting any of:␤ whitespace␤»
Xliff m: use NativeCall; say nativecast(num32, Buf.new(80, 3f, 00, 00));
camelia rakudo-moar 55c359: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Unable to parse expression in argument list; couldn't find final ')' ␤at <tmp>:1␤------> 3all; say nativecast(num32, Buf.new(80, 37⏏5f, 00, 00));␤ expecting any of:␤ whitespace␤»
06:47 domidumont joined
raydiak m: use NativeCall; say nativecast num32, Buf.new: 0x80, 0x3f, 0x00, 0x00 06:49
camelia rakudo-moar 55c359: OUTPUT«2.27795078360642e-41␤»
Xliff D'oh 06:51
m: use NativeCall; say nativecast(num32, Buf.new(0x3f, 0x80, 00, 00));
camelia rakudo-moar 55c359: OUTPUT«Potential difficulties:␤ Leading 0 does not indicate octal in Perl 6.␤ Please use 0o0 if you mean that.␤ at <tmp>:1␤ ------> 3nativecast(num32, Buf.new(0x3f, 0x80, 007⏏5, 00));␤ Leading 0 does not indicate octal in Perl 6.␤…»
06:51 domidumont left
Xliff m: use NativeCall; say nativecast(num32, Buf.new(0x3f, 0x80, 0x00, 0x00)); 06:52
camelia rakudo-moar 55c359: OUTPUT«4.60060298822481e-41␤»
Xliff m: use NativeCall; say nativecast(num32, Buf.new(0x80, 0x3f, 0x00, 0x00));
camelia rakudo-moar 55c359: OUTPUT«2.27795078360642e-41␤»
06:52 domidumont joined
Xliff m: use NativeCall; say nativecast(num32, Buf.new(0x00, 0x00, 0x80, 0x3f)); 06:52
camelia rakudo-moar 55c359: OUTPUT«1␤»
Xliff radiak++ 06:53
raydiak :)
07:00 huggable left 07:06 cpage_ left 07:09 Amnez777 left 07:11 Amnez777 joined 07:13 darutoko joined 07:22 zakharyas joined
timotimo MasterDuke: is it correct to remove the ; before the function? because in the line before that there's still a ! for some reason ... 07:28
yoleaux 03:32Z <MasterDuke> timotimo: another minor update to add the line number in the source file as a column, github.com/MoarVM/MoarVM/pull/382
07:28 aries_liuxueyang left 07:30 aries_liuxueyang joined 07:33 RabidGravy joined, sno left, abraxxa joined 07:39 bjz joined 07:42 AlexDaniel joined 07:43 bjz left 07:56 AlexDaniel left 08:02 jonas1 joined 08:06 maybekoo2 joined
buharin hello :) 08:08
08:13 dakkar joined
DrForr Mornin'. 08:14
hahainternet hola 08:18
08:19 Actualeyes left 08:22 Roamer` joined
RabidGravy BOOM! 08:30
08:32 jack_rabbit left
Xliff Here comes the Boom! 08:33
Ready or not, here comes the boys from the South
Boom!
Here comes the Boom!
How you like me now?
(love that song...)
nighty, #perl6! 08:34
08:34 Xliff is now known as Xliff_zzzz 08:43 sno joined 08:53 jgrabber joined
jgrabber Hello, how do I create a subset of an Type that only accepts Object instances, so neither Type Objects nor .undefined Objects? 08:58
"subset Defined of Real where *.defined" does not seem to work
RabidGravy In what way doesn't iit seem to work? 09:00
m: subset Defined of Real where *.defined; my Defined $a = Int
camelia rakudo-moar 2ff5ad: OUTPUT«Type check failed in assignment to $a; expected Defined but got Int (Int)␤ in block <unit> at <tmp> line 1␤␤»
RabidGravy is what I would expect
psch m: subset RealReal of Real:D where * > 1.0; # is what i'd do vOv
camelia ( no output )
psch i mean, like, if you want a subset of Real that only limits to .defined, just use Real:D in the first place 09:01
but if you want to constrain that further, subset of *:D
...not literally *, though :)
m: subset RealReal of Real:D where * > 1.0; sub f(RealReal $) { }; f Real 09:02
camelia rakudo-moar 2ff5ad: OUTPUT«Constraint type check failed for parameter '<anon>'␤ in sub f at <tmp> line 1␤ in block <unit> at <tmp> line 1␤␤»
RabidGravy Oooh when did type smileys work there?
psch not sure, i dimly recall a change about the resulting type about 3 or so weeks ago..?
that is, the type that results from smilying a type 09:03
RabidGravy the advantage of the subset however is that it doesn't require an initialiser
gregf_ m: my Real $val = -0.1;
camelia ( no output )
jgrabber I see, Real:D is the better option. I did not think of it 09:04
m: subset Defined of Real where *.defined; my Defined $a = Real 09:05
camelia ( no output )
jgrabber why does this work though? 09:06
timotimo Real is a role, maybe roles are defined? 09:07
m: say Real.defined 09:08
camelia rakudo-moar 2ff5ad: OUTPUT«False␤»
timotimo no, they're not :\
09:08 wamba joined
psch m: subset DInt of Int where *.defined; my DInt $a = Int 09:08
camelia rakudo-moar 2ff5ad: OUTPUT«Type check failed in assignment to $a; expected DInt but got Int (Int)␤ in block <unit> at <tmp> line 1␤␤»
psch it is something about roles apparently
m: subset DInt of Int where *.defined; my DInt $a = 1
camelia ( no output )
RabidGravy confirmed 09:10
m: subset DD of Stringy where *.defined; my DD $a = Stringy
camelia ( no output )
RabidGravy BUG? FEATURE? PLACE YOUR BETS NOW! 09:11
timotimo i'd say bug.
jgrabber From a noob's perspectives, this seems pretty counter-intuitive since Real.defined is False
timotimo yes 09:12
m: my Real:D $foo = Num
camelia rakudo-moar 2ff5ad: OUTPUT«Type check failed in assignment to $foo; expected Real:D but got Num (Num)␤ in block <unit> at <tmp> line 1␤␤»
timotimo m: my Real:D $foo = 1
camelia ( no output )
timotimo you can use that for the time being, it seems like
it's also going to be 1000x more performant
jgrabber I see, thanks! 09:13
gregf_ sorry for interrupting. was wondering if there was documentation these: :D :D:, :U etc 09:14
timotimo those are the type smileys
gregf_ heh
so are there ones like :p :P :o :O ? 09:16
timotimo not yet
but there's :_
(not implemented, iirc) 09:17
psch m: my Int:_ $x;
camelia ( no output )
RabidGravy where :O auto-initialises with some unexpected value and :P with just the one you didn't want
psch ponders slanging that
jgrabber Ok, ':D' is in the FAQ and few other places, could've checked that one. I did not see any documentation that tells you not to subset via *defined for roles - although ':D' seems way more suitable 09:19
gregf_ timotimo++ :) 09:22
m: :_.^name
camelia ( no output )
gregf_ m: say [:_, :D, :D:, :U].map(*.^name) 09:23
camelia rakudo-moar 2ff5ad: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Confused␤at <tmp>:1␤------> 3say [:_, :D, :D:7⏏5, :U].map(*.^name)␤ expecting any of:␤ colon pair␤»
hahainternet all i have to say to ^^ is :O 09:24
;)
timotimo what is :D: supposed to do there?
psch m: say [:_, :D, :D, :U].map(*.^name)
camelia rakudo-moar 2ff5ad: OUTPUT«(Pair Pair Pair Pair)␤»
gregf_ dunno, i think i've seen :D: as well in some docs 09:26
psch the second : is the invocant marker
timotimo except that's only syntax inside type definitions 09:27
and such
psch m: my method foo(Int:D: ) { "foo" }; say foo 2
camelia rakudo-moar 2ff5ad: OUTPUT«foo␤»
psch right
it's not valid anywhere else
gregf_ ah - ok
RabidGravy++ 09:29
psch++
09:30 labster left
hahainternet oh so that reminds me 09:34
Inline::Go would be fun
i was reading about Inline::Perl5 again
the problem is, you can't embed perl6 in another binary yet can you, i guess i need to figure out how perl6 in perl5 gets communicated with 09:35
anyway i already implemented a semi equivalent in TCL so could be a fun project 09:36
timotimo um, we load perl as a shared library
so it is "in another binary"
we just have a tiny helper .c file, i believe
hahainternet timotimo: no i mean the opposite way around, when using perl6 from perl5
timotimo ah
jnthn hahainternet: I think Inline::Perl6 does that
So may be worth looking at for hints
hahainternet jnthn: precisely
but Go has runtime reflection, and a sufficient enough type system to permit this 09:37
timotimo yeah, inline::perl6 starts a perl6 with inline::perl5 and wraps it all backwards
hahainternet and i have working code that instantiates custom types at runtime and dispatches to them with reflection
aah ok
i'll just leave that out for now then, until perl6 can be linked in as a .so or what have you it's just too much effort imho
09:37 wamba left
hahainternet morning all btw 09:37
09:43 wamba joined 09:47 tbrowder joined
jnthn hahainternet: I doubt you'll ever link perl6 in as a .so, unless we come up with a way to fat-pack bytecode into a .so too, and then I suspect distros will find a way to hate that. :P 09:49
hahainternet jnthn: being unable to link to perl6 like you can to perl5 would be a little bit of a smell, but you know FAR more about the internals of compiler implementation than i ever will
so please don't take that as criticism :)
timotimo we'll want to be able to fatpack bytecode into an executable anyway 09:50
jnthn hahainternet: Well, we certainly need an embedding story, but it becomes a little more interesting to figure out given the Perl 6 compiler isn't written in C (or something else that compiles into a .so) but rather Perl 6 (which compiles to bytecode). As opposed to Perl 5 which is written in C, so it's relatively easy to see how to provide a .so to link against. 09:51
hahainternet jnthn: i would love to say something interesting and erudite here, but i'm afraid i have nothing 09:52
i doubt i can help, but if i can, i'd love to
timotimo when will we AOT compile bytecode? :P
jnthn Maybe it'd end up being a library that links in MoarVM and provides a nice C-level API for loading/invoking the Perl 6 compiler.
hahainternet doing the perl6 -> Go portions should be somewhat straightforward at least 09:53
Go's reflection makes that part not too bad
anyway i'm not ognna be starting that anytime soon, just hit me in the head last night as i tried to sleep
timotimo ouch
09:56 jgrabber left 10:01 CIAvash left 10:10 wamba left 10:12 rindolf joined 10:13 wamba joined 10:17 TEttinger left 10:22 wamba left
nine Sooo...which hotel in Cluj will feature the most Perl 6 devs? 10:23
10:23 wamba joined
DrForr That's an excellent question. 10:23
According to the wiki, Capitolina City Chic Hotel is currently in the lead for the total number of days followed by (and now I'm going to have that damn song stuck in my head) Hotel Transilvania. 10:25
nine Sample size is just very, very small though :) 10:26
hahainternet small sample sizes are fine 10:28
just give your confidence interval :)
10:30 Zoffix joined 10:35 RabidGravy left
DrForr I would guess either Capitolina or Hotel Capitol, if I were looking at the map. Plenty of Pensiunes in the area, but those aren't going to appeal to a YAPC crowd, I think. 10:35
10:36 z3ndrag0n left
DrForr (pensiunea being the rough equivalent of a B&B) 10:36
10:37 iH2O joined
DrForr Nice thing is it's all in walking distance for me :) 10:37
10:38 z3ndrag0n joined
iH2O u live in Manhattan i gather 10:38
DrForr No, but I do use that metric :) 10:39
10:57 colomon joined 11:00 huggable joined 11:02 pmurias joined 11:12 sftf left 11:17 maybekoo2 left 11:18 rindolf left 11:22 bjz joined 11:32 rindolf joined 11:48 bjz left, sergot joined
buharin my friend? 11:52
11:53 kaare_ joined 11:55 grondilu joined
DrForr I'm sure they're around. 11:56
11:56 iH2O left 11:58 skids joined 12:00 skids left 12:02 ggoebel116 joined 12:04 pmurias left 12:07 kid51 joined
BrokenRobot m: try { EVAL "say 09"; CATCH { default { .^name.say }}} 12:11
camelia rakudo-moar d1ebac: OUTPUT«Potential difficulties:␤ Leading 0 is not allowed. For octals, use '0o' prefix, but note that 9 is not a valid octal number␤ at /home/camelia/EVAL_0:1␤ ------> 3say 097⏏5<EOL>␤9␤»
BrokenRobot Is this not catchable or am I doing it wrong? 12:12
12:12 bjz joined
gfldex it is a compile time exception but inside an eval. I vote for bug. 12:13
jnthn Note it's only a warning
BrokenRobot Oh
m: my $x = 09; say ++$x
camelia rakudo-moar d1ebac: OUTPUT«Potential difficulties:␤ Leading 0 is not allowed. For octals, use '0o' prefix, but note that 9 is not a valid octal number␤ at <tmp>:1␤ ------> 3my $x = 097⏏5; say ++$x␤10␤»
jnthn Not sure if it's actually thrown as a control exception though. 12:14
Perhaps it should be if not
m: try { EVAL "say 09"; CONTROL { default { .^name.say }}}
camelia rakudo-moar d1ebac: OUTPUT«Potential difficulties:␤ Leading 0 is not allowed. For octals, use '0o' prefix, but note that 9 is not a valid octal number␤ at /home/camelia/EVAL_0:1␤ ------> 3say 097⏏5<EOL>␤9␤»
gfldex m: try { my $a = '0' ~ (0..9).pick; EVAL "say $a"; CATCH { default { .^name.say }}}
camelia rakudo-moar d1ebac: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤EVAL is a very dangerous function!!! (use MONKEY-SEE-NO-EVAL to override,␤but only if you're VERY sure your data contains no injection attacks)␤at <tmp>:1␤------> 3my $a = '0' ~ (0..9).pick; EVAL "say…»
jnthn Yeah, seems not
BrokenRobot It's a X::Worry::P5::LeadingZero is X::Worry::P5
Um... it there a way to capture warnings? :) 12:15
jnthn Those thrown by warn, yes...with a CONTROL block and match on CX::Warn.
But for the "potential difficulties" spat out by the compiler, I'm not so sure.
gfldex m: try { quietly EVAL "say $a"; CATCH { default { .^name.say }}}
camelia rakudo-moar d1ebac: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Variable '$a' is not declared␤at <tmp>:1␤------> 3try { quietly EVAL "say 7⏏5$a"; CATCH { default { .^name.say }}}␤»
gfldex m: try { EVAL "quietly say $a"; CATCH { default { .^name.say }}} 12:16
camelia rakudo-moar d1ebac: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Variable '$a' is not declared␤at <tmp>:1␤------> 3try { EVAL "quietly say 7⏏5$a"; CATCH { default { .^name.say }}}␤»
BrokenRobot k, I'll use is_run from roast for it.
hmhmhmh this kinda sucks. I didn't realize it was just a warning so this warning is LTA: 12:21
m: say 09
camelia rakudo-moar d1ebac: OUTPUT«Potential difficulties:␤ Leading 0 is not allowed. For octals, use '0o' prefix, but note that 9 is not a valid octal number␤ at <tmp>:1␤ ------> 3say 097⏏5<EOL>␤9␤»
BrokenRobot Since leading 0 is allowed
m: quietly say 09
camelia rakudo-moar d1ebac: OUTPUT«Potential difficulties:␤ Leading 0 is not allowed. For octals, use '0o' prefix, but note that 9 is not a valid octal number␤ at <tmp>:1␤ ------> 3quietly say 097⏏5<EOL>␤9␤»
BrokenRobot m: say quietly 09
camelia rakudo-moar d1ebac: OUTPUT«Potential difficulties:␤ Leading 0 is not allowed. For octals, use '0o' prefix, but note that 9 is not a valid octal number␤ at <tmp>:1␤ ------> 3say quietly 097⏏5<EOL>␤9␤»
BrokenRobot shrugs 12:22
gfldex m: say &quietly.WHAT 12:23
camelia rakudo-moar d1ebac: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Undeclared routine:␤ quietly used at line 1␤␤»
gfldex no wonder it doesn't work!
BrokenRobot It does work 12:24
m: say Nil + Nil
camelia rakudo-moar d1ebac: OUTPUT«Use of Nil in numeric context in block <unit> at <tmp> line 1␤Use of Nil in numeric context in block <unit> at <tmp> line 1␤0␤»
BrokenRobot m: quietly say Nil + Nil
camelia rakudo-moar d1ebac: OUTPUT«0␤»
gfldex you may have missed a joke there 12:25
BrokenRobot doesn't see a joke
12:26 flaviusb left 12:30 ggoebel116 left 12:36 obfusk joined 12:39 Guest94483 is now known as nemo 12:43 rgrinberg joined 12:44 harmil left 12:49 RabidGravy joined 12:50 AlexDaniel joined
tbrowder gfldex: I will file with RT the pod issues you mentioned yesterday if you have no objections. You can add or subtract after you see them. 12:53
gfldex tbrowder: thanks a lot
tbrowder you're welcome!
12:54 AlexDaniel left, AlexDaniel joined
[Coke] tbrowder: do you find you get better help on questions on email vs. irc? 12:56
dalek c: 1267395 | (Wenzel P. P. Peppmeyer)++ | html/js/main.js:
#__debug__ will display the viewport size in the bottom left corner
c: d0f5aaf | (Wenzel P. P. Peppmeyer)++ | doc/Type/Capture.pod6:
Merge branch 'master' of github.com/perl6/doc
tbrowder [Coke]: I do in some cases, especially non-time-sensitive or transient ones. I have asked some things here (of specific folks in some cases) that don't get answered and I'm reluctant to ping again. Whereas, with e-mail, I have a better record of queries and I feel they will usually be answered at some point. 13:00
dalek c: e742369 | (Wenzel P. P. Peppmeyer)++ | CONTRIBUTING.md:
tell contributors about new #__debug__ feature
 13:01
tbrowder [Coke]: BTW, I have successfully gotten RT to recognize 'tbrowder' (thanks to help from RT support), but the issue you assigned to me doesn't show up on my top ten list. 13:02
[Coke]: Can you fix that or do I need to make another RT support request? 13:03
[Coke] tbrowder: as I said before, I can't help you with that. 13:04
dalek c: a8ebdc2 | (Wenzel P. P. Peppmeyer)++ | html/js/main.js:
external links may throw an exception, catch those
 13:05
[Coke] when you go to the ticket, do you see it assigned to you there?
13:05 travis-ci joined
travis-ci Doc build failed. Wenzel P. P. Peppmeyer 'Merge branch 'master' of github.com/perl6/doc' 13:05
travis-ci.org/perl6/doc/builds/143038280 github.com/perl6/doc/compare/3e694...f5aaf5ed80
13:05 travis-ci left, ggoebel116 joined, ggoebel116 left
[Coke] gah, empty catch block! 13:05
dalek c: 675d073 | (Zoffix Znet)++ | html/js/main.js:
Remove trailing whitespace
 13:07
13:08 bjz_ joined 13:09 bjz left
tbrowder [Coke]: Sorry, I missed the first response (that's another reason I like e-mail for those kinds of things). I asked RT support. 13:09
13:09 travis-ci joined
travis-ci Doc build failed. Wenzel P. P. Peppmeyer 'tell contributors about new #__debug__ feature' 13:09
travis-ci.org/perl6/doc/builds/143039824 github.com/perl6/doc/compare/d0f5a...42369390b4
13:09 travis-ci left
tbrowder [Coke]: I do see it assigned to me on the ticket. 13:10
13:11 iH2O joined
dalek c: 907e647 | (Wenzel P. P. Peppmeyer)++ | html/js/main.js:
fix trailing WS
 13:13
c: ad9e1be | (Wenzel P. P. Peppmeyer)++ | /:
Merge branch 'master' of github.com/perl6/doc
c: c891da2 | (Wenzel P. P. Peppmeyer)++ | html/js/main.js:
Revert "fix trailing WS"

This reverts commit 907e6473d4f915ef94c4f48290d37dc211c0f732.
13:13 T-Bone joined
c: f12e45a | (Wenzel P. P. Peppmeyer)++ | html/js/main.js:
explain empty catch block
13:13 travis-ci joined
travis-ci Doc build failed. Wenzel P. P. Peppmeyer 'external links may throw an exception, catch those' 13:13
travis-ci.org/perl6/doc/builds/143040928 github.com/perl6/doc/compare/e7423...ebdc2ec4ee
13:13 travis-ci left 13:16 eone joined
AlexDaniel Zoffix: not sure if there's any need to interrupt someone's workflow, people can solve such things themselves (as the git log shows) 13:19
gfldex AlexDaniel: thanks for complaining for me :) 13:21
i should outsource all my complaining. Would save tons of time. :D
AlexDaniel gfldex: by the way, this revert was unnecessary, as now the problem is there again 13:22
13:24 T-Bone left, eone left
gfldex t/trailing_whitespace.t works locally for me and I can't push changes. travis might be confused 13:27
13:27 travis-ci joined
travis-ci Doc build passed. Zoffix Znet 'Remove trailing whitespace' 13:27
travis-ci.org/perl6/doc/builds/143041596 github.com/perl6/doc/compare/a8ebd...5d073dfd44
13:27 travis-ci left
gfldex goes to study git hooks 13:28
13:28 skids joined
BrokenRobot [Coke]: on this ticket you commented that it can be closed with tests because the current behaviour matches the one requested, but it doesn't. The ticket said REPL must print "^D" when it's pressed before exiting, but we don't do that. Should I write the tests for current behaviour or do we need to add the ^D? rt.perl.org/Ticket/Display.html?id...xn-1399431 13:30
And I don't get a newline either on bleed rakudo on Debian (via ssh) 13:31
[Coke] tbrowder: are you sure the expansion icon next to the list is open? 13:32
BrokenRobot: I think it's answering the spirit of the request. also, see Faye's update on jul 18, 2015. 13:33
masak's around somewhere, he can say if he's happy.
tbrowder: looks like the icon has to be pointing up for the box to be open. 13:34
BrokenRobot k, I'll leave this one alone for now.
grondilu what's the P5 equivalent of the rw trait for sub parameters? 13:37
[Coke] it's not "is rw" ? 13:38
BrokenRobot grondilu: $_[0] = 42
[Coke] er, I read that backwards
BrokenRobot As in, @_ is alias
AlexDaniel gfldex: I'm not sure what git magic you have done locally, but main.js line 93 is still there
gfldex: you can't push because there's nothing to push or because git complains about something? 13:39
13:40 RabidGravy left
grondilu that works indeed 13:40
dalek c: d4097fe | (Wenzel P. P. Peppmeyer)++ | html/css/style.css:
CSS changes depeding on github.com/perl6/Pod-To-HTML/commi...c87b589b26

this will fix
   - fixes duplicate anchors from X<> around/inside a =head
   - fixes missing links on toc entries for =head4 and higher
The layout has changed and may trigger a popular vote.		 13:43
c: a8199d5 | (Wenzel P. P. Peppmeyer)++ | html/js/main.js:
fix trailing WS
gfldex AlexDaniel: there was confusion on my side, should be fixed with the push I just did 13:44
AlexDaniel gfldex: by the way, what editor do you use?
gfldex AlexDaniel: about magic: "The pre-push hook runs during git push, after the remote refs have been updated but before any objects have been transferred. It receives the name and location of the remote as parameters, and a list of to-be-updated refs through stdin. You can use it to validate a set of ref updates before a push occurs (a non-zero exit code will abort the push)."
vim
AlexDaniel: and failed tests produce a non-zero exit code :) 13:46
13:47 ptolemarch joined
AlexDaniel gfldex: is it possible to set up client-side hooks automatically for everyone who clones the repo? 13:50
gfldex AlexDaniel: i don't thinks so. .git is in .gitignore :)
AlexDaniel gfldex: last time I asked the same question about git filters I was told no, because that's pretty much remote code execution…
gfldex: and we can't have server-side hooks on github, so… 13:51
13:51 xinming_ left
gfldex however, one could write a t/is-there-a-git-hook.t 13:51
AlexDaniel this implies that contributors run tests :D
gfldex we do have travis to catch the lazy (like me) 13:52
13:52 xinming joined
AlexDaniel gfldex: anyway, surely we can have instructions on how to setup a hook in the README 13:52
gfldex AlexDaniel: nonono you got that wrong. README is the file folk read _after_ they caused problems. :) 13:54
13:55 dj_goku left, iH2O left
DrForr No, they post to stackoverflow and do something else. 13:55
AlexDaniel m: min(+'a', +'a').say 13:57
camelia rakudo-moar d1ebac: OUTPUT«Cannot convert string to number: base-10 number must begin with valid digits or '.' in '3⏏5a' (indicated by ⏏)␤ in block <unit> at <tmp> line 1␤␤Actually thrown at:␤ in block <unit> at <tmp> line 1␤␤»
AlexDaniel bisect: min(+'a', +'a').say
bisectable AlexDaniel: on both starting points the exit code is 1 and the output is identical as well
AlexDaniel bisect: good=2015.10 min(+'a', +'a').say 13:58
bisectable AlexDaniel: exit code is 1 on both starting points, bisecting by using the output
AlexDaniel: (2015-12-05) github.com/rakudo/rakudo/commit/3cb6ae9
13:58 RabidGravy joined, Actualeyes joined
AlexDaniel bisect: good=2015.10 try { min(+‘a’, +‘a’).say; CATCH { exit 1 } } 13:59
bisectable AlexDaniel: on both starting points the exit code is 1 and the output is identical as well
14:03 harmil joined
gfldex the ToC changes are now on the docs.perl6.org, if you got odd colours please shift-click the reload button 14:04
14:05 travis-ci joined
travis-ci Doc build passed. Wenzel P. P. Peppmeyer 'fix trailing WS' 14:05
travis-ci.org/perl6/doc/builds/143051072 github.com/perl6/doc/compare/f12e4...199d5cc7bb
14:05 travis-ci left
BrokenRobot gfldex: did you say you'll fix JS later or was the show/hide toc feature removed? ( github.com/perl6/doc/issues/217 ) 14:08
14:09 harmil left
BrokenRobot The table stuff is a definite -1 from me. Not only semantically (as discussed last night), but visually too. Can't see what roles/classes the thing inherits from and the bumble-bee row coloring is distracting. 14:12
dalek c: 02696a6 | (Wenzel P. P. Peppmeyer)++ | html/css/style.css:
hide heading numbers in ToC. They are not displayed in the main text
 14:13
c: f8d1d0b | (Wenzel P. P. Peppmeyer)++ | html/css/style.css:
indent heading > 1 in ToC
gfldex BrokenRobot: i will fix that now
BrokenRobot gfldex++
gfldex and i will display the heading numbering in #__debug__ so to make it easy to spot heading blurps 14:14
AlexDaniel gfldex: again, why does it have to be a table? 14:26
14:27 rgrinberg left
gfldex AlexDaniel: it does not have to be a table. If you don't like it, feel free to change it. Disabeling a column is quite easy though because ... it's a column. 14:27
BrokenRobot gfldex: but which column do we need to disable? 14:29
gfldex: e.g. this <ol> looks exactly like the <table> I see on docs site right now: jsfiddle.net/usgsstyf/
gfldex please stop 14:30
14:30 ggoebel116 joined
BrokenRobot gfldex: I was just trying to understand what problem you were attempting to solve. There's no point in saying "feel free to cahnge it" if I don't know why you changed it. 14:31
mst gfldex: people are asking questions because they don't understand yet, not to criticise you
BrokenRobot Exactly.
gfldex table#TOC td.toc-number { display: none; } 14:32
BrokenRobot The ol { list-style: none; } does the same thing. It can also do more, like change to roman numerals or use bullets instead of numbered things. 14:34
sammers m: my %dog = name => 'Elly', bark => -> $name { "$name says woof" }; say %dog<bark>(%dog<name>);
camelia rakudo-moar d1ebac: OUTPUT«Elly says woof␤»
BrokenRobot m: my %dog = name => 'Elly', bark => *.fmt: "%s says woof"; say %dog<bark>(%dog<name>); 14:36
camelia rakudo-moar d1ebac: OUTPUT«Elly says woof␤»
BrokenRobot \o/
sammers is there a way to reference a hash value from itself? like in that code object example.
vytas how can one access rakudo.org/rt/easy mentioned in rakudo.org/tickets/ ? 14:37
sammers like to reference name inside the code object without passing it to the sub
BrokenRobot sammers: is the value a sub? You can probably use &?ROUTINE
vytas agh it actually has button to proceed
AlexDaniel gfldex: I'm sorry, I'm just trying to understand your thought process - if it doesn't have to be a table, please take my question to mean “why did you decided to make it a table?” - I'm just trying to get my head around what's going on
14:38 sufrostico joined
gfldex BrokenRobot: the whole idea of proper heading numbering is to show the same numbers in the ToC then on he headings in the main body (we don't do that yet). To do so we have to do the rendering of numbers by hand. The benefit of auto numbering of <ol> is fairly slim in this case. 14:38
BrokenRobot We can easily add that to main body with the `counter` CSS property 14:39
gfldex further adding a 3rd column to a <ol> is possible, but i can think of better ways to spend my time
as I wrote 2 times already, if you don't like it, change it 14:40
BrokenRobot vytas: yeah, it's a bit bugged at the moment.
mst gfldex: people were trying to understand why you liked it, not try to change it - please stop being defensive, nobody was attacking you :( 14:41
14:42 masak joined
BrokenRobot gfldex: my changing it would be simply reverting your work. I don't know what you're attempting to do with your work. And when I asked, you're throwing a tantrum making me feel like an asshole. 14:42
masak hiya, #perl6
14:42 buharin left
mst BrokenRobot: dude, please chill, there's no need to escalate the misunderstanding 14:42
dalek c: 14271c9 | (Wenzel P. P. Peppmeyer)++ | html/ (2 files):
fix ToC toggle buttom
 14:43
gfldex BrokenRobot: please read the commit msg on github.com/perl6/Pod-To-HTML/commi...c87b589b26 14:44
14:45 user9 left
sammers m: my %dog = name => 'Elly', bark => { "%dog<name> says woof" }; say %dog<bark>(); 14:49
camelia rakudo-moar d1ebac: OUTPUT«Elly says woof␤»
BrokenRobot body { counter-reset: h1 } h1 { counter-reset: h2 } h2 { counter-reset: h3 } h1:before { counter-increment: h1; content: counter(h1) ". " } h2:before {counter-increment: h2; content: counter(h1) "." counter(h2) ". "} 14:50
But whatever... I'm not looking to make another enemy in #perl6, so I'll just move on
perlpilot "another enemy"?!? 14:51
gregf_ sammers: (my %dog = name => "Elly", bark => -> $name { "$name says woof" })<bark>.("Elly"); # because there can be only one 'name'?
sammers BrokenRobot, I guess we can just self-reference inside the hash code object.
BrokenRobot gfldex: and sorry for the "reverting your work" comment, I meant that I did not understand any of the changed, not that I was threatening to undo anything. 14:52
sammers gregf_ this seems to work: my %dog = name => 'Elly', bark => { "%dog<name> says woof" }; say %dog<bark>();
gfldex "my changing it would be simply reverting your work" he threatened to destroy my work _in_ public. I tried to move the discussion out of the public eye, what he ignored. If you ask me, he is correct with the assumtion that he makes himself enemies.
mst gfldex: nobody was threatening to destroy your work
gfldex: please stop, step back, and assume good faith 14:53
gregf_ cmon guys its not about indiviuals.. *points to Perl6*
gfldex what exactly does that phrase mean then?
mst gfldex: that was in response to your defensive "if you don't like it, change it", when he was trying to understand what you did
his point was that, not understanding, the only change he could make would be to revert it, and since obviously that would be silly, he wanted to try and understand what you did
gfldex: notice that's EXACTLY what he said - he didn't understand, and was trying to 14:54
I'm not sure how you jumped from "you're telling me to change it but I don't understand enough to change it except by reversion, so I'd like to understand" to that being a threat
gfldex i understand what it was in respose to, but there is a big difference between to change the html-elements in use and to "imply revert your work" 14:55
i am pissed and i have very good reason to do so
mst the point was completely clear to me, I'm not sure why you read a threat into it rather than helping him to understand :(
no, your getting pissy is what derailed this, when other people wanted to understand your work better :( 14:56
RabidGravy gah, ponies needed
gfldex mst: fine you try hard to understand him. Would you be so kind the spend the same energy in understanding me?
mst gfldex: I am doing so. that's why I asked him to calm down above as well.
nobody was attacking you. nobody was threatening your work.
two people cared about your work enough to try and understand it better
that's a compliment in open source, usually
perlpilot hugs gfldex and BrokenRobot 14:57
parabolize sammers: you've got state and functionality with that dog object. Wouldn't it be easier to use a class?
mst I don't know which projects you've been on when people would actually just delete somebody else's work without understanding it, but in my experience, this is not such a project
14:57 dogbert17 joined
sammers parabolize, just experimenting with code objects. 14:58
mst gfldex: if you step back, assume good faith, and re-read people's words with the assumption they just wanted to understand your work, I hope their words will be clearer to you
I think the issue here is that nobody else realised you'd take "the only change I could make right not would be to revert" as a threat rather than a statement of not understanding yet 14:59
sammers parabolize, what other containers can we use to pass around code objects?
14:59 mohae_ joined
dogbert17 o/ #perl6 14:59
perlpilot sammers: What containers do you think won't hold code objects? :)
dogbert17: greetings! 15:00
15:00 huggable left
dogbert17 hi perlpilot 15:00
mst gfldex: does the misunderstanding make a little more sense, given that context?
dogbert17 it's time for todays stupid question
perlpilot dogbert17: 42 (the stupid answer)
parabolize gregf_: the smileys are documented in Signatures: docs.perl6.org/type/Signature#Cons...ned_Values and tested in roast/S{12-attributes,06-parameters,04-declarations}/smiley.t
dogbert17 in src/core/Capture.pm there are several methods, e.g. 'kv' and 'values' which I cant se if I do 'say Capture.^methods(:local)'. Why? 15:01
m: say Capture.^methods(:local)
camelia rakudo-moar d1ebac: OUTPUT«(new from-args BUILD hash list elems Capture FLATTENABLE_LIST FLATTENABLE_HASH WHICH Method+{<anon|59337488>}.new Method+{<anon|59337488>}.new Method+{<anon|59337488>}.new Str gist perl Bool Numeric Method+{<anon|59337488>}.new Method+{<anon|59337488>}.new…»
15:01 mohae left
gfldex BrokenRobot: my appology, my private message was eaten by my inability to type 15:02
sammers perlpilot, so I am pretty new to p6
BrokenRobot gfldex: I did receive your request to apologize and I apologized above
perlpilot m: Capture.^methods.map(*.name).say; # What do you mean, you can't see them? ;) 15:03
camelia rakudo-moar d1ebac: OUTPUT«(new from-args BUILD hash list elems Capture FLATTENABLE_LIST FLATTENABLE_HASH WHICH AT-KEY AT-POS EXISTS-KEY Str gist perl Bool Numeric keys kv values pairs antipairs)␤»
dogbert17 now I see them :) 15:04
perlpilot dogbert17: Oops, I forgot the :local, but it does the same. I think it's that methods don't always stringify to their names well (or at all)
dogbert17: especially multimethods 15:05
dogbert17 perlpilot: so chances are that the methods I'm looking for are one of the Method+{<anon|59337488>}.new?
perlpilot dogbert17: yep
dogbert17: caveat lector though, that's just my guess right now. 15:06
dogbert17 perlpilot: aha, so I should use your variant instead: Capture.^methods.map(*.name).say
perlpilot dogbert17: If you're interested in their names, you should ask them for their name, yes. :) 15:07
15:07 ssotka joined
dogbert17 perlpilot: will do, thx for explaining 15:07
15:11 rgrinberg joined
perlpilot dogbert17: btw, .^methods gives you back objects. You can call .^methods on those objects to see what else you can do besides "name" (though, again, some of them will be anonymous methods or something without a name) 15:12
BrokenRobot ^methodsception 15:14
Xliff_zzzz Whoop!
\0/
15:14 Xliff_zzzz is now known as Xliff 15:15 kurahaupo joined
Xliff I have Perl6 module to read Winamp Media Library files. 15:15
I need to write simple test cases and off to the ecosystem it will go.
15:15 harmil joined
Xliff Prolly this weekend. 15:15
And now.... MOAR SLEEP! 15:16
15:16 Xliff is now known as Xliff_zzz
gregf_ parabolize: cheers 15:16
15:18 khw joined
dalek c: c999f29 | (Jan-Olof Hendig)++ | doc/Type/Capture.pod6:
Added docs for Capture.Capture
 15:18
15:33 yaboo joined
yaboo is there microsoft sql server DB access for perl6 yet? 15:34
dogbert17 Ok, so I have concocted some docs for Capture.keys. Does it look ok to you? gist.github.com/dogbert17/3f7c3517...746ffe4603
[Coke] yaboo: it'd probably be here if so: github.com/perl6/DBIish - and I don't see it yet. 15:35
DBIsh readme has mojibake in it.
(near very bottom)
yaboo I saw DBIish... and didn't see SQL server as an option. I'm not familiar with mojibake. 15:36
perlpilot dogbert17: the all-caps CAPTURE looks like it might be meta-syntax, but it's not clear (i.e., someone might think that CAPTURE is a Perl 6 thing and try CAPTURE.keys literally)
dogbert17 perlpilot: I could remove that 15:37
perhaps it should be removed from the entire document...
[Coke] perlpilot: that's an issue with USAGE everywhere. 15:38
Not worth bugging this commit over.
USAGE syntax is meant to be intuitive, but isn't documented.
dalek Iish: 77ad8b2 | (Zoffix Znet)++ | README.pod:
Fix mojibake
 15:40
Iish: 010b74a | (Zoffix Znet)++ | README.pod:
Merge pull request #70 from zoffixznet/master

Fix mojibake
perlpilot Aye.
gfldex dogbert17: maybe use «$capture.keys» to indicate this is the variable part?
BrokenRobot Weird. The online editor had 'Commit' button greyed out when I tried to edit the original
dogbert17 gfldex: in the usage part or the example?
15:41 yaboo left
gfldex dogbert17: i would do it in the Usage part too, for anyone who made it that far it should be cleare what a scalar container is for 15:41
dogbert17: in examples i like to use variable names to indicate the type that is stored in them. The examples tend to be fairly abstract and the variable name isn't reused elsewhere anyways. 15:43
dogbert17 gfldex: could you reload the page, I've made some updates 15:44
[Coke] btw, I have been using text-based chat since 1985 or so, and still find it very easy to assume someone's intent. I try to follow mst's advice, and assume the best.
15:44 iH2O joined
[Coke] (assume incorrectly, I meant to say! stupid text-based chat.) 15:44
[Coke] heads out to lunch.
gfldex dogbert17: i would write it the same way
dogbert17 gfldex: then I'll commit it this way :) 15:45
gfldex dogbert17: also, if I need a speaking name I stick the type in the declaration. e.g. my Int $left; my Int $right;
dogbert17 gfldex: do you think I should write 'my Capture $capture ...'? 15:46
gfldex dogbert17: that would be redundant. Every time you have redundant info in a doc, the reader might spend time to figure out what that redundancy is for. If there is a good reason for that redundancy, it needs explaining words. So a lack of need of those words is a good indicator. 15:48
dogbert17 gfldex: point taken 15:49
dalek c: 83209fc | (Jan-Olof Hendig)++ | doc/Type/Capture.pod6:
Added docs for Capture.keys. gfldex++ perlpilot++
 15:51
15:52 canopus left, dj_goku joined 15:57 dj_goku left 15:58 canopus joined 15:59 Guest_84845 joined
Guest_84845 Allah is doing 15:59
sun is not doing allah is doing 16:00
moon is not doing Allah is doing
stars are not doing Allah is doing
tadzik I haven't see that spam in a while
BrokenRobot Who is he doing?
Guest_84845 planets are not doing Allah is doing
galaxies are not doing Allah is doing
BrokenRobot Oh, this is exciting. 16:01
tadzik Guest_84845: what about universes?
BrokenRobot That's it? I was hoping we'd get to at least the multiverse!
tadzik BrokenRobot: yeah, that's anticlimatic :( 16:02
BrokenRobot nods
mst ooooh, an allahbot
Guest_84845 oceans are not doing Allah is doing
16:02 cdg joined
mst if nobody minds, I'm curious to see if the full lifecycle's the same as the other one I've seen 16:02
tadzik oh, we downgraded to earth stuff again 16:03
BrokenRobot :)
Guest_84845 mountains are not doing Allah is doing
trees are not doing Allah is doing
mst parents next, I think
Guest_84845 mom is not doing Allah is doing
BrokenRobot lol
Guest_84845 dad is not doing Allah is doing 16:04
mst now, capitalism time
Guest_84845 boss is not doing Allah is doing
tadzik that's cheating
BrokenRobot ...capitalism? :D
mst watch the next few :D
tadzik I'm no geologist, but afaik mountains aren't doing anything at all
Guest_84845 job is not doing Allah is doing
tadzik so allah just gets to fill his cv with no-work?
blasphemy
Guest_84845 dollar is not doing Allah is doing 16:05
degree is not doing Allah is doing
medicine is not doing Allah is doing
tadzik phah, that reminds me of Tim Minchin
Guest_84845 customers are not doing Allah is doing 16:06
you can not get a job without the permission of allah
mst and now, a brief segue into asking god nicely to be allowed to do things
iH2O troll bot alert!
mst iH2O: what do you mean 'alert', it's not like we didn't notice
Guest_84845 you can not get married without the permission of allah
gregf_ o_O
Guest_84845: now please automate this in Perl6 16:07
Guest_84845 nobody can get angry at you without the permission of allah
tadzik troll bot? So apparently we can't cross a bridge without the permission of allah either
Guest_84845 light is not doing Allah is doing
fan is not doing Allah is doing
16:07 ChanServ sets mode: +o jnthn
BrokenRobot Huh, why does he give permission to get angry? Damn, and all this time he seemed like a nice guy. Doing things for everyone else. 16:08
mst jnthn: hang on
16:08 Guest_84845 was kicked by jnthn (ALLAH IS DOING NOTHING, BOW TO THE ALMIGHTY JNTHN))
gregf_ he? bots are gender neutral :| 16:08
mst ...
jnthn mst: oh, oops
mst 17:02 < mst> if nobody minds, I'm curious to see if the full lifecycle's the same as the other one I've seen
tadzik how can you tell who they identify as
don't be insensitive
mst jnthn: YOU HAVE RUINED MY ANTHROPOLOGY STUDY
BrokenRobot :D 16:09
jnthn Aww :P
mst now somebody's going to have to listen to all of that again
jnthn heh
16:09 ChanServ sets mode: -o jnthn
mst goddamnit, I've been waiting weeks for that opportunity, and I asked the channel nicely, and nobody objected 16:09
gregf_ sure, run this *runs* loop { <sun moon stars planets galaxies oceans mountains trees mom dad boss job dollar degree medicine>.map: ->$who { say "$who is not doing God is doing"; sleep 1; } } 16:10
masak Guest_84845 is not doing, at least not in #perl6
mst but, no, you can't be bothered to actuallt read the scrollback before breaking out the hobnail boots
jnthn mst: I didn't actually read, I just saw the darn thing and was like "argh" :P
mst SULKS
tadzik masak: #rekt :P
dalek c: d190696 | (Jan-Olof Hendig)++ | doc/Type/Capture.pod6:
Added docs for Capture.values
masak pats mst, *pat pat*
mst jnthn: yeah. that's not how you +o :P
masak jnthn: I think this might be the first documented case of #perl6 over-kicking :P 16:11
jnthn Tough.
mst I should not be lecturing *other* people on showing restraint as an op, FFS
masak: I always expected that to be me :D
masak mst: guess you'll just have to write your own Allah bot :P :P :P
dalek c: f49b9fb | (Wenzel P. P. Peppmeyer)++ | / (2 files):
teach #__debug__ to show heading numbers
 16:12
mst I don't want my own, I want to find out if the lifecycle's standard
it was so far, but the last one I saw disconnected and reconnected part way through
and I was wondering if that was structural or just a bug
16:12 iH2O left
masak disconnected part way through is not doing, Allah is doing 16:12
mst LOL 16:13
masak you can not get disconnected without the permission of Allah 16:15
gfldex masak: do you indicate that telcos are not the work of the devil? 16:16
masak who is this Allah anyway? I mean I've heard the name before, but is he like a celebrity or something? 16:18
gfldex sorry can't help you, never met that fellow myself 16:19
16:19 freezerburnv joined
BrokenRobot He's the guy who flew to heaven on a flying horse. It was in all the newspapers. 16:20
16:22 FROGGS joined
masak thought that was Elton John 16:25
also, if your horse can fly, flying to heaven is not such a big achievement 16:26
16:33 sufrostico left 16:34 sufrostico joined
ugexe there is already an "Air Horse ONE" 16:35
timotimo m: say '^'.ord.base(16)
camelia rakudo-moar d1ebac: OUTPUT«5E␤»
tbrowder gfldex: see RT bugs #128569 and #128570 16:36
synopsebot6 Link: rt.perl.org/rt3//Public/Bug/Displa...?id=128569
Link: rt.perl.org/rt3//Public/Bug/Displa...?id=128570
tbrowder [Coke] I tried to take ownership of those bugs but am not allowed. 16:37
16:37 Possum left
gfldex .oO( I totally owned that bug! ) 16:38
16:41 cognominal joined 16:43 Guest_84845 joined, Guest_84845 left 16:50 neilhwatson joined 16:52 setty1 joined
tbrowder gfldex: do you have an account on RT? 16:53
gfldex: if you do, then take ownership if you wish, 16:54
16:55 sno left 16:57 abraxxa left 16:59 Possum joined 17:02 addison joined 17:03 dakkar left
dalek c: 747c2a9 | (Zoffix Znet)++ | assets/sass/style.scss:
Add warning about html/css/style.css being a generated file

Closes #662		 17:05
c: 71fa4e6 | (Zoffix Znet)++ | / (2 files):
Move style changed introduced into auto-generated style.css to style.scss

This moves work added in commits
  github.com/perl6/doc/commit/d4097f...e737e23f93
  github.com/perl6/doc/commit/02696a...27796e31cd
  github.com/perl6/doc/commit/f8d1d0...998d5638f0
  github.com/perl6/doc/commit/14271c...7d50a6fa3f
c: a8a84db | (Zoffix Znet)++ | / (2 files):
Improve auto-generation warning

Make it clear that it's html/css/style.css that is auto-generated, as the message has to appear in assets/sass/style.scss too and using "THIS FILE" is confusing.		 17:08
gfldex BrokenRobot: `make css` or `make assets` would be nice to have. 17:13
BrokenRobot nods
I'm completely unfamiliar with Makefiles tho
The dev app generates them automatically, if you have the extra modules installed. 17:14
17:18 addison left 17:20 freezerburnv left 17:23 firstdayonthejob joined
[Coke] jnthn++ #kicking that bot 17:24
tbrowder: what does it say in the upper left hand side for you? For me I get "Logged in as coke" 17:25
What's the reason not to just regen the files every build? 17:26
BrokenRobot [Coke]: the doc CSS? Because many contributors are contributing docs, not styles, and the extra prereqs for style generations can be avoided by including the generated styles. I know dogbert17 was having issues installing one of sass generating prereqs on debian 17:27
[Coke] ok. I tend to prefer not to have generated stuff checked in, but probably won't be able to help with a debian problem, so I'll try to let it go. :) 17:28
BrokenRobot Oh, it was Ubuntu not Debian: rt.cpan.org/Public/Bug/Display.html?id=115287 17:30
tbrowder [Coke]: It shows my email, not my user name. I've tried to change all the settings I can find, and I login with my username (not email), but nothing so far has worked.
17:32 user9 joined 17:36 neilhwatson left 17:37 buharin joined 17:43 rgrinberg left 17:44 maybekoo2 joined 17:51 zakharyas left 17:52 dj_goku joined, dj_goku left, dj_goku joined 17:58 dj_goku left 18:02 sufrostico left 18:04 sufrostico joined 18:13 sufrostico left 18:14 cdg left 18:18 rgrinberg joined 18:26 spider-mario joined 18:28 girafe joined
gfldex BrokenRobot: i do have problems on debian as well, META::Cpan debian package is to old 18:30
BrokenRobot I'm unsure which module requires that. 18:32
gfldex quite a few
BrokenRobot The alternative to CSS::Sass is having ruby-sass (at least in Debian) available
gfldex and a test in Mojolicious::Plugin::AssetPack hangs (investigating right now)
BrokenRobot And AssetPack uses it if CSS::Sass is not installed 18:33
18:34 labster joined
BrokenRobot Doesn't hang for me on Jessie. 18:36
And ::AssetPack is not needed; one can execute sass -w assets/sass/style.scss:html/css/style.css 18:37
And that'll watch the sass file for changes and process it.
18:44 zakharyas joined, sortiz joined
sortiz \o #perl6 18:44
dalek c: 31b2b2e | (Zoffix Znet)++ | .gitignore:
Ignore extraneous files generated by `sass` command
 18:46
c: f6d3a24 | (Zoffix Znet)++ | CONTRIBUTING.md:
Document alternative SASS processing method

Using `sass` command		 18:47
AlexDaniel yaaaay! 18:49
BrokenRobot++
BrokenRobot: hmm, it looks like the output is a little bit different 18:50
BrokenRobot: gist.github.com/AlexDaniel/99fd823...955bca3b9a 18:52
18:58 Actualeyes left 19:00 huggable joined 19:08 darutoko left 19:14 buharin left 19:18 lostinfog joined 19:19 jonas1 left 19:22 sno joined 19:24 telex left 19:26 telex joined 19:32 domidumont left
BrokenRobot Any idea why this is giving me "Undeclared routine: is_run_repl"? gist.github.com/zoffixznet/672e8bb...875786ca02 19:33
timotimo BrokenRobot: that lib path is probably wrong? 19:34
isn't that already inside t/ somewhere? 19:35
19:35 neilhwatson joined
BrokenRobot timotimo: the bath is right when `make spectest`. In this case I'm including teh correct path with -I, and it can find the module, since the error is about the sub name and not missing modulke 19:36
*path :)
timotimo hum.
fair enough?
19:36 addison joined
BrokenRobot I'm just not following why the is export isn't taking 19:36
timotimo right 19:37
dunno :(
BrokenRobot ( full file github.com/perl6/roast/blob/master...st/Util.pm )
m: sub put { say "meow" }; ::CORE('&put')(42) 19:41
camelia rakudo-moar d1ebac: OUTPUT«Cannot invoke this object (REPR: Uninstantiable; CORE)␤ in block <unit> at <tmp> line 1␤␤»
BrokenRobot What was the way to do this? I want to call the core 'put' sub 19:42
m: sub put { say "meow" }; &CORE::put(42)
camelia rakudo-moar d1ebac: OUTPUT«42␤»
moritz not fast enough
timotimo yeah, you were calling CORE instead :) 19:45
BrokenRobot Oh, damn 19:46
timotimo++ the <t/spec/packages> was actually the issue :) along with a separate roast checkout I had it in my rakudo repo in t/spec with the OLD version of Test::Util and it was using that
timotimo yay 19:47
BrokenRobot Which reminds me of something I promised to fix before Christmas and didn't 19:48
m: use lib <foo bar>; say $*REPO; say $*REPO.next-repo
camelia rakudo-moar d1ebac: OUTPUT«file#/home/camelia/bar␤file#/home/camelia/foo␤»
BrokenRobot The <foo> should be first one looked in since it's the first one the programmer provided
Too late now I guess?
19:48 profan joined
BrokenRobot & # relocating 19:48
19:53 dj_goku joined, dj_goku left, dj_goku joined 19:59 dj_goku left 20:04 lue joined 20:06 ShimmerFairy left 20:13 kaare_ left 20:23 TEttinger joined 20:24 wamba left 20:25 lue is now known as ShimmerFairy 20:28 labster left 20:29 domidumont joined 20:30 rgrinberg left 20:31 dwarring joined 20:37 Technaton joined 20:41 neilhwatson left, labster joined
dalek c: 2d46e62 | (Jan-Olof Hendig)++ | doc/Type/Capture.pod6:
Added docs for Capture.kv
 20:42
Xliff_zzz m: use DateTime::Format 20:45
camelia rakudo-moar d1ebac: OUTPUT«===SORRY!===␤Could not find DateTime::Format at line 1 in:␤ /home/camelia/.perl6␤ /home/camelia/rakudo-m-inst-2/share/perl6/site␤ /home/camelia/rakudo-m-inst-2/share/perl6/vendor␤ /home/camelia/rakudo-m-inst-2/share/perl6␤ CompUn…»
moritz star-m: use DateTime::Format 20:46
camelia star-m 2016.04: OUTPUT«===SORRY!===␤Could not find DateTime::Format at line 1 in:␤ /home/camelia/.perl6␤ /home/camelia/star-2016.04/share/perl6/site␤ /home/camelia/star-2016.04/share/perl6/vendor␤ /home/camelia/star-2016.04/share/perl6␤ CompUnit::Repositor…»
moritz m: DateTime::Format
camelia rakudo-moar d1ebac: OUTPUT«Could not find symbol '&Format'␤ in block <unit> at <tmp> line 1␤␤Actually thrown at:␤ in block <unit> at <tmp> line 1␤␤»
Xliff_zzz *sob*
grondilu star-m: use DateTime 20:50
camelia star-m 2016.04: OUTPUT«===SORRY!===␤DateTime is a builtin type. You can use it without loading a module.␤»
grondilu lol 20:51
that sounded a bit condescending :)
lizmat grondilu: do you have a suggestion for better phrasing ?
grondilu Well, for a start, no error message should ever use the second person. 20:52
20:52 dj_goku joined, itaipu joined
grondilu DateTime is a builtin type, not an external module. 20:53
^quote this
20:53 itaipu left 20:54 fhorck joined, FROGGS left
grondilu though now that I think about it, there is the "did you mean xxxx?" after a suspected typo. And that is fine. 20:54
20:54 wamba joined 20:55 yqt joined 21:00 bjz_ left 21:03 ShimmerFairy left, ShimmerFairy joined 21:04 skids left, skids joined 21:05 rgrinberg joined
moritz I don't find it condescending, actually 21:06
21:06 zakharyas left
moritz but then I'm not an Englishman :-) 21:06
dalek c: 231c503 | (Wenzel P. P. Peppmeyer)++ | / (2 files):
lighten background greyness for <pre> to increase contrast
 21:07
c: c070180 | (Wenzel P. P. Peppmeyer)++ | / (3 files):
darken comments in <pre> for more contrast
c: b14a620 | (Wenzel P. P. Peppmeyer)++ | / (2 files):
Copy CSS around cause a casualty. The hide button is now where it
21:07 domidumont left 21:08 ShimmerFairy left 21:09 skids left, ShimmerFairy joined 21:10 as left 21:22 jjido joined 21:23 setty1 left, _dolmen_ joined 21:26 TimToady left 21:28 TimToady joined
lizmat github.com/rakudo/rakudo/commit/29a1107390 # grondilu++ 21:38
21:40 jjido left
timotimo that makes sense 21:40
21:41 wamba left 21:43 lostinfog left 22:01 rindolf left 22:07 RabidGravy left 22:11 rindolf joined
AlexDaniel hm, since when do we have two camelias on the home page 22:17
timotimo i only see one 22:19
dalek c: 274a88b | thundergnat++ | doc/Language/operators.pod6:
gcd should be divisor not denominator

3 justifications: 1) It coerces each argument to Int. Ints always have a "denominator" of 1 but may have larger common divisors. 2) The announcement from when it was added says it is greatest common divisor. github.com/rakudo/rakudo/blob/8cbb...011.05#L21 3) The actual code that implements the op says it is greatest common divisor. github.com/perl6/nqp/blob/27e99d4a..._gcd.c#L18		 22:25
22:32 jack_rabbit joined 22:41 spider-mario left 22:43 cognominal left 22:49 firstdayonthejob left
AlexDaniel timotimo: I mean here: docs.perl6.org/ 22:57
timotimo ah 22:58
that doesn't seem so bad?
AlexDaniel timotimo: if that was shadow, than it should be slightly below it… :D 22:59
22:59 _dolmen_ left
AlexDaniel it's not too bad, it's just that we didn't have it before 22:59
23:00 huggable left
gfldex it is, as soon as you hide the ToC. Then a fading Camelia starts to compete with text about the readers mind. 23:00
timotimo huh?
23:00 grondilu left, zacts joined
timotimo oh, indeed 23:01
gfldex depends a bit on the page ofc. Some pages don't have text span that far.
given that the ToC will wander to the left on screens with many pixels, shady camelia may have to go 23:02
AlexDaniel yeah, it is a little bit weird
gfldex timotimo: do you have the powers to update Pod::To::HTML on docs.perl6.org ?
timotimo oof. i may have that 23:03
23:04 john51 left, john51 joined
gfldex timotimo: please do so. I don't want to bug the same person twice every day. 23:04
timotimo let me first get a bite of something edible 23:05
gfldex please keep up your strength
timotimo my body is very insistent on me eating all the damn time 23:06
23:07 zacts left 23:12 avenj left 23:14 rindolf left 23:16 AlexDaniel left
gfldex i wonder if there should be links to presentation videos on perl6.org 23:17
23:17 zacts joined, xinming left 23:19 xinming joined
MasterDuke gfldex: perl6.org/documentation/, under "Miscellaneous sources" 23:20
23:21 avenj joined
MasterDuke but i think it would be good to call them out a bit more 23:21
gfldex it's just so much stuff on that site.
MasterDuke maybe their own section instead of just falling under misc sources?
gfldex i looked for them under Community 23:22
there is a Screencast section
23:31 john51_ joined 23:33 john51 left 23:36 BenGoldberg joined 23:44 AlexDaniel joined 23:47 maybekoo2 left, yqt left
timotimo sorry, gfldex, i'll have to delay this until after i had sleep 23:58
feel free to remind me again tomorrow 23:59
Zoffix timotimo, is sudo needed to do it?

Powered by Cro and the Raku® Programming Language.

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

Thank you!