»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.
timotimo m: sub dies() { die "dead meat" }; dies orelse say "oh no! $_" 00:00
camelia rakudo-moar 74d0e3: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Undeclared routine:␤ orelse used at line 1␤␤»
timotimo m: sub dies() { die "dead meat" }; dies() orelse say "oh no! $_"
camelia rakudo-moar 74d0e3: OUTPUT«dead meat␤ in sub dies at <tmp> line 1␤ in block <unit> at <tmp> line 1␤␤»
geekosaur or smartmatch against Failure, or several other ways
timotimo m: sub dies() { die "dead meat" }; (try dies()) orelse say "oh no! $_"
camelia rakudo-moar 74d0e3: OUTPUT«Use of Nil in string context␤ in block at <tmp> line 1␤oh no! ␤»
timotimo :\
m: sub dies() { die "dead meat" }; (try dies()) orelse say "oh no! $!"
camelia rakudo-moar 74d0e3: OUTPUT«oh no! dead meat␤»
timotimo there we go, that could be right
samcv cool 00:01
timotimo i don't work with this style of writing error handling often 00:02
m: sub dies() { die "dead meat" }; (try dies()) or say "oh no! $!"
camelia rakudo-moar 74d0e3: OUTPUT«oh no! dead meat␤»
timotimo ok, with or it also works
samcv hmm. not sure why False or die $! doesn't work. is that a bug? 00:03
is that what it's supposed to do
samcv or seemed to work for Juerd using an older rakudo 00:04
timotimo there is no $! for False or die $! 00:05
samcv well. true. but 00:05
timotimo i mean, what is it supposed to be set to?
samcv for open or whatever
timotimo and what's supposed to set it?
well, if open dies (instead of failing), it'll go through exception stuff like stack unwind and such
samcv well disregard what i said about False, but with open 'foo' or die $!, it worked for Juerd but not on my rakudo
timotimo so it interacts with try/catch, not with and/or
samcv but orelse works
timotimo juerd's rakudo is really old o_O 00:06
also ... using the REPL is a bad idea
Juerd If it's a bad idea, why is it the default thing that opens?
samcv well open 'foo' orelse die "$!"; doesn't work on my rakudo eeither :\ 00:07
timotimo, i haven't been using REPL, just runing the file normally
timotimo Juerd: because it's not as bad as i make it out to be 00:10
when it works, it works
however, it has many cases in which it causes very strange behavior that you won't be able to reproduce when you run code from a file
samcv ^
timotimo it's bedtime for me now 00:11
adios! :)
samcv bye :)
samcv well sub dies() { die "dead meat" }; (try dies()) or say "oh no! $!" 00:14
that works too
samcv i kinda think it should work. because open 'foo'; say $!; works fine 00:15
but i'm not an expert
[Coke] I disagree that using the REPL is a bad idea. 00:22
dalek c: 12d4ef5 | coke++ | doc/Language/nativecall.pod6:
#828 - remove reference to zavolaj
 00:26
dalek c: 03ed944 | (Tom Browder)++ | doc/Language/tables.pod6:
remove restriction 5

users **may** be warned, though		 00:41
c: 9bb8ed9 | (Tom Browder)++ | doc/Language/tables.pod6:
remove a no-longer-valid bad example
 00:45
c: bd8f673 | (Tom Browder)++ | doc/Language/tables.pod6:
correct results due to lifting of restriction 5
 00:52
iBakeCake tbrowder++ thanks. 00:53
tbrowder iBakeCake: you're welcome, I bow to more perceptive minds 00:55
iBakeCake I think the future will be the real test :)
tbrowder I hope the results are good for mankind. 00:56
Herby__ o/ 02:47
b7j0c p6: say 3; 03:04
camelia rakudo-moar 74d0e3: OUTPUT«3␤»
b7j0c p6: my uint8 $i = 255; 03:05
camelia ( no output )
b7j0c p6: my uint8 $i = 254;
camelia ( no output )
b7j0c p6: my uint8 $i = 255;
camelia ( no output )
b7j0c p6: my uint8 $i = 255; say $i; $i++; say $i;
camelia rakudo-moar 74d0e3: OUTPUT«255␤256␤»
b7j0c p6: my uint8 $i = 255; say $i; $i++; $i++; say $i;
camelia rakudo-moar 74d0e3: OUTPUT«255␤257␤»
b7j0c p6: my uint8 $i = 255; say $i; say ++$i; 03:06
camelia rakudo-moar 74d0e3: OUTPUT«255␤0␤»
dalek c: 31cf6cb | (Francis Grizzly Smit)++ | doc/Type/Pair.pod6:
fixed Pair method fmt it was missing the format sting in the signature
 03:18
grondilu m: say Inf == Inf 07:20
camelia rakudo-moar 74d0e3: OUTPUT«True␤» 07:21
grondilu thought that was supposed to return False
grondilu may have confused it with NaN though 07:28
m: say NaN == NaN
camelia rakudo-moar 74d0e3: OUTPUT«False␤»
daxim_ andrewalker_, thp wants to see your qt5 bindings 07:30
masak grondilu: yes, there's only one Inf. it's supposed to be equal to itself. 07:31
masak well, there's -Inf too, I guess. but it's pretty far from Inf, and so there's no risk of confusion. 07:32
moritz unless you're making a sphere out of the complex plane :-) 07:35
FROGGS .oO( will it still fly? ) 07:39
[ptc] FROGGS, timotimo: about the coverity scan stuff for MoarVM: could you guys create a generic link to MoarVM on Coverity please? I think if one is admin of the MoarVM group on GitHub it should work. 07:51
I don't think it's a good idea to use my account directly for the Coverity scan stuff. 07:52
Also, with the generic account (or an account directly linked to the MoarVM/MoarVM repo), then it's possible to get pushes to be automatically checked, which isn't possible with my account
FROGGS ohh niec 07:54
nice*
andrzejku hello :) 08:20
moritz hi ho 08:33
El_Che who are you calling a ho'? 08:34
moritz a hollistic person? 08:38
DrForr Morning... just barely. 08:54
andrewalker daxim_: Ohh... Cool :D I'll push them somewhere and ping him. 10:37
smls m: my $a:b = 42; say $a:b; 10:56
camelia rakudo-moar 74d0e3: OUTPUT«42␤»
smls ^^ is thta supposed to be allowed? 10:57
lizmat smls: I think so, as I remember patches for that being added
smls m: my $a:</i> = 42; say $a:</i>; 10:59
camelia rakudo-moar 74d0e3: OUTPUT«42␤»
smls ^^ This too?
Certainly a WAT in string interpolation.
timotimo those are allowed, yes 11:03
m: my $foo:<hey> = 10; say $foo:<hey>
camelia rakudo-moar 74d0e3: OUTPUT«10␤»
timotimo compare: 11:04
m: say &infix:<+>
camelia rakudo-moar 74d0e3: OUTPUT«sub infix:<+> (Mu $?, Mu $?) { #`(Sub+{<anon|70919312>}+{Precedence}|39580704) ... }␤»
smls But 'infix' is a known category 11:05
what's the point of allowing this format with arbitraty prefixes? 11:06
jnthn In action methods for protoregex candidates, the prefixes defined by the user 11:08
*are defined
token whatever-i-want:<foo> { <sym> } etc 11:09
For the tokens themselves too
timotimo m: my $foobar:<with spaces allowed, too?>; 11:11
camelia ( no output )
timotimo m: my $foobar:<with spaces allowed, too?> say $foobar;
camelia rakudo-moar 74d0e3: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Two terms in a row␤at <tmp>:1␤------> 3my $foobar:<with spaces allowed, too?>7⏏5 say $foobar;␤ expecting any of:␤ infix␤ infix stopper␤ statement end␤ state…»
timotimo m: my $foobar:<with spaces allowed, too?>; say $foobar;
camelia rakudo-moar 74d0e3: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Variable '$foobar' is not declared␤at <tmp>:1␤------> 3foobar:<with spaces allowed, too?>; say 7⏏5$foobar;␤»
timotimo right, requires it every time
m: my $foobar:<with spaces allowed, too?>; say $foobar:<with spaces allowed, too?>;
camelia rakudo-moar 74d0e3: OUTPUT«(Any)␤»
smls jnthn: That's still all & stuff though. Why would you need to name a $ variable like that?
timotimo why would we forbid $ variables but not method names and sub names? 11:12
smls because things like string interpolation
where strict parsing allows different syntaxes to step on each other's toes as little as possible 11:13
"<i>$foo:</i>" # my use-case 11:14
timotimo html tags in an interpolated string is already a well-known trap; i'd personally recommend anyone who uses string interpolation for html like that to put braces around their variables from the get-go
since $foo</i> is already a trap
smls true
kinda devalues sigils though, if you have to interpolate with {} all the time anyway. 11:15
dalek c: 251e333 | (Tom Browder)++ | doc/Language/tables.pod6:
remove text no longer valid

remove reference to an exception for uneven columns per row ( no longer true with removal of former restriction five)		 11:22
smls Maybe the strictness should be applied on the string interpolation side. I.e. instead of allowing interpolation with
< $ @ % & > <ident> <postcircumfix>?
only allow:
'$' <simple-ident>
'@' <simple-ident> <bracket-postcircumfix>
'%' <simple-ident> <brace-postcircumfix>
'&' <simple-ident> <paren-postcircumfix>
timotimo that should definitely be doable with a slang 11:23
iBakeCake Mixing markup with code is the real trap :} 12:08
dalek c: 4fd5eb7 | coke++ | CONTRIBUTING.md:
match number
 13:11
c: 8899ab3 | coke++ | CONTRIBUTING.md:
add a note about make test/make xtest
c: fa27d10 | coke++ | xt/aspell.t:
spellcheck all pod6 files and md files
c: cab6b77 | coke++ | / (4 files):
pass spellcheck on all MD files

  - fix typos
  - learn words/code
  - avoid some made up words
[Coke] doc - predefined-search-items should probably be moved into 'templates' 13:17
FROGGS o/ 13:30
iBakeCake \o\
[Coke] iBakeCake: latest message on perl6-users about log(23,0) seems to be something in your current wheelhouse 13:31
iBakeCake [Coke]: what is it? 13:32
Isn't log base 0 undefined
moritz log to base 0?
[Coke] www.nntp.perl.org/group/perl.perl6....g3027.html
moritz m: say log(23, 0) 13:33
camelia rakudo-moar 71a01e: OUTPUT«-0␤»
moritz that's the complaint
iBakeCake m: say log( my num $ = 23, my num $ = 0) 13:34
camelia rakudo-moar 71a01e: OUTPUT«This type cannot unbox to a native number: P6opaque, Int␤ in block <unit> at <tmp> line 1␤␤»
iBakeCake m: say log( my num $ = 23e0, my num $ = 0e0)
camelia rakudo-moar 71a01e: OUTPUT«-0␤»
iBakeCake Well, I recently checked the log() stuff for nums...
iBakeCake looks
Ah, ok, now I remember
m: say log(23)/−∞ 13:35
camelia rakudo-moar 71a01e: OUTPUT«-0␤»
lizmat commute to NR.pm meeting& 13:37
iBakeCake ln(0) by 2008 IEEE 754 is −∞; and log₀(23) = ln(23)/ln(0) = ln(23)/−∞ = -0
iBakeCake So it's not so much an artefact of the log() but rather how we divide by infinity 13:37
[Coke] btw, my pod issue with E<> before is that it doesn't support html entities (and that's what the pod docs had used) - however, it does support unicode names and codepoints... so I'm not going to open a ticket on that, I'm fine with E<171> and E<LEFT DOUBLE ANGLE BRACKET> 13:40
iBakeCake IEEE section 6.1 tells about division by −∞ would give the limit, with the sign of ∞, so -0 is right. and section 7.3 says that ln(0) is −∞ 13:41
If anyone wanted to respond to that person... I don't have a perl6-users subscriptions
dalek c: 015c9b9 | (Zoffix Znet)++ | doc/Type/Mix.pod6:
Indicate Mix -> Bag(Hash) coersion loses negative weights
 14:10
c: 459310f | (Zoffix Znet)++ | doc/Type/MixHash.pod6:
Indicate MixHash -> Bag(Hash) coersion loses negative weights
 14:11
iBakeCake m: my BagHash $b; $b{42}++; dd $b 14:12
camelia rakudo-moar 79bb86: OUTPUT«BagHash $b = (42=>1).BagHash␤»
iBakeCake skids: ^ fixed
skids iBakeCake++ 14:12
tailgate Can I write to std out from a DOC INIT block? Like DOC INIT { say 'hello'; } tp check if it's running? 14:35
jnthn Sure 14:36
$ perl6-m --doc -e 'DOC INIT say 42'
42
tailgate Ah, it only runs if it's in doc mode 14:39
crucialrhyme i have a super easy question. what is the idiomatic way to iterate over the characters in a Str? 14:46
jnthn Provided you really mean characters (as in graphemes), for $str.comb { } 14:48
(If you mean the Unicode codepoints, then for $str.ords { }) 14:49
crucialrhyme it's good to know both
i hope i get some really gnarly unicode text just so i can revel in not having to worry about it 14:50
dalek c: 6e97839 | (José Albert Cruz Almaguer)++ | doc/Language/subscripts.pod6:
Sigil typo in MixHash example
 15:19
c: ad392db | (Zoffix Znet)++ | doc/Language/subscripts.pod6:
Merge pull request #974 from polyglot-plt/master

Sigil typo in MixHash example
El_Che yeah 15:30
for people wanting to speak at fosdem: our dev room will have 80 places (problably full most of the day) and it's in the K block (same one as the booth)
dalek c: e3c7aa3 | (Armand Halbert)++ | / (3 files):
Added external link SVG image.
 15:52
c: 9cb2744 | (Zoffix Znet)++ | / (3 files):
Merge pull request #877 from ahalbert/SVG

Added external link SVG image.
titsuki Hi #perl6 16:02
I'm facing the circular module loading issue while creating a MeCab (famous Japanese morphological analyzer) bindings.
Does anyone know how to declare stub CStruct classes ?
gist: gist.github.com/titsuki/664c94a991...7d08538171
timotimo should work just like you did it 16:08
just drop the "is repr" for the second time
titsuki timotimo: Thx! I'll try it ! 16:09
timotimo good luck!
[Coke] TimToady++ emailing 16:28
briandfoy Does anyone have ideas about how to fix a subset type failure from throwing X::AdHoc? I asked about it on Stackoverflow stackoverflow.com/q/40097868/2766176 but it's also on RT from last year: rt.perl.org/Ticket/Display.html?id=126763 16:35
[Coke] briandfoy: hio. Not sure if you got an alert, but combined your recent RT about prefix= with an older one. 16:36
iBakeCake m: subset Z where 2; sub (Z)(42) 16:38
camelia rakudo-moar b1f77c: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Missing block␤at <tmp>:1␤------> 3subset Z where 2; sub (Z)7⏏5(42)␤ expecting any of:␤ new name to be defined␤»
iBakeCake m: subset Z where 2; sub (Z){}(42)
camelia rakudo-moar b1f77c: OUTPUT«Constraint type check failed for parameter '<anon>'␤ in sub at <tmp> line 1␤ in block <unit> at <tmp> line 1␤␤»
iBakeCake m: sub (Str){}(42) 16:39
camelia rakudo-moar b1f77c: OUTPUT«Type check failed in binding to <anon>; expected Str but got Int (42)␤ in sub at <tmp> line 1␤ in block <unit> at <tmp> line 1␤␤»
[Coke] briandfoy: I would definitely agree adhoc isn't right. (in general, adhoc is "we hadn't gotten around to putting a typed exception here.") 16:40
iBakeCake briandfoy: well, it's thrown here, FWIW: github.com/rakudo/rakudo/blob/nom/...P.nqp#L502 16:41
Actually that's where the message gets created 16:42
briandfoy There's an X::TypeCheck::Parameter in Exception, or is that for something else (like an error specifying a signature)
I saw the message in bootstrap, but it looks like all the other origins of exceptions. Higher up it looks like those get translated into something else.
ObCodeIsNotDocumentation :) 16:43
This matters for the parts of Learning Perl 6 that will talk about subsets and signatures. People really want these cool type features! 16:44
[Coke] might be a good idea to keep a list of RTs that fixing would help with the book. 16:45
iBakeCake suspects that'd be a large list :)
briandfoy I will do that. Or, maybe I can have RT permissions to add a tag?
[Coke] a fake subject tag like [BOOK] ? 16:46
briandfoy I'm happy to be a person who goes through RT and acknowledges, closes, or whatever
[Coke] sure, give me 5s.
briandfoy I wasn't thinking a subject line tag. Doesn't RT have other features for that? 16:47
perigrin it has straight out tags if I recall.
briandfoy maybe something in "depended on by" 16:48
[Coke] added.
perigrin been a while since I was the RT monkey.
iBakeCake m: subset Z where { $_ ~~ 2 or X::TypeCheck.new(:operation<assigning-stuff>, :got($_), :expected(Z)).throw }; sub (Z){}(42)
camelia rakudo-moar b1f77c: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Illegally post-declared type:␤ Z used at line 1␤␤»
[Coke] it doesn't have freeform tags.
iBakeCake m: subset Z where { $_ ~~ 2 or X::TypeCheck.new(:operation<assigning-stuff>, :got($_), :expected("Z")).throw }; sub (Z){}(42)
camelia rakudo-moar b1f77c: OUTPUT«Type check failed in assigning-stuff; expected Str but got Int (42)␤ in block <unit> at <tmp> line 1␤␤»
[Coke] (our queue, anyway)
mst [Coke]: well, yeah, what you'd do, normally, is create a ticket to use as a tag
then use the ticket relationships
[Coke] you can have a ticket that has child tickets, that works.
iBakeCake briandfoy: ^ awkward workaround. There's also Subset::Helper that does custome messages
mst very common approach
[Coke] right, what mst said. 16:49
my only issue with that is that your ticket count is never accurate, but since we're over 1400, I guess that doesn't matter. :)
mst - any progress on getting your build fixes pushed upstream? we have tickets at each level complaining about spaces in paths, which i thought your presentation said you solved. 16:50
iBakeCake briandfoy: when do you think the book will be ready, BTW? 16:51
mst [Coke]: no, my presentation said I broke it worse 16:53
briandfoy The book will be ready by Christmas.
[Coke] mst: oh, whoops. Ok. then I will start digging back into at least making sure that spaces don't kill anything. Thanks.
briandfoy But O'Reilly wants it before OSCON.
[Coke] briandfoy: we're not falling for THAT old trick.
iBakeCake briandfoy: which Christmas? :D 16:54
mst [Coke]: prompting your hilarious response of "surprise! you're wrong! it was already broken" or so
briandfoy For what it's worth, I'm not going to stop to fix bugs and whatnot. If there's a wart, I'll just note that in the text and move on.
iBakeCake Sounds reasonable.
Though we'd much prefer if you'd stop and fix :D 16:55
mst "As of [release], there's currently [bug], requiring [workaround]" seems completely sane, especially if [bug] contains an RT number so it's easy for readers to look up if they still need to care 16:56
[Coke] mst: ah, yes. My memory, as always, not entirely helpful.
mst but having attempted to write a book and been fairly terrible at it, I'm pretty sure "whatever brian decides" is more likely to be the right answer than my thoughts 16:57
[Coke] someone with a recent star install care to test this: rt.perl.org/Ticket/Display.html?id=122839
briandfoy Heh, I was trying to get the recent star installed, but filed a bug on that too. That's still on my to do list for today. 17:00
mst [Coke]: oh, yeah, also, bear in mind star's build system is a fork of the rakudo one with ... about half of the patches since merged down, probably 17:01
I suspect the 'correct' solution will turn out to be 'fix rakudo, nqp, moarvm, then burn the star build system down and redo it'
timotimo what mst says. 17:03
iBakeCake \o/
iBakeCake briandfoy: FWIW, we also have #perl6-dev 17:06
[Coke] mst: the plan was to fix them from the ground up; moarvm, then nqp, then rakudo 17:29
(and then maybe R*)
mst right 17:30
mst the recent dramastorm around DBIC has rather left me having trouble focusing on technical stuff 17:31
I am feeling increasingly guilty about not having got back to this stuff for you
but sadly guilt doesn't actually make it easier to focus while in the middle of complicated political bollocks
El_Che mst: I have the impression the issue has been resolved, no? 17:37
iBakeCake Has it? Damn! So goes my popcorn sales business idea... 17:38
mst El_Che: mostly. I fully expect there to be one last loud blip before we're done
El_Che iBakeCake: well, there could be some plot changes. mst having a long lost twin and the like 17:39
iBakeCake :) 17:40
El_Che mst: probably. But it won't probably change the direction of the discussion. Hopefully, things will be handled without too much personal damage to the people involved. Although, not likely 17:44
mst El_Che: my rational mind concurs entirely. my adrenal system, less so. 17:45
nine Is there a writeup of the DBIC thing somehwere? 17:46
nine would love to see lwn.net coverage of Perl issues 17:48
mst what would be the point? I'd rather we got the governance issues resolved and move on, he'd likely attack any attempt at writing it up as partisan, and attracting more drama to the situation would benefit nobody
I mean, I avoided explicitly naming the conflict in question specifically because of that
nine mst: yeah, I guess you're right. Since I've opted out of getting involved in any way so far and rather focused on coding, I should really swallow my curiosity and stick to that decision. 17:50
El_Che mst: the "upside", although not for you, is that the personal damage seem to be limited to 2 participants 17:54
mst that was basically the idea
El_Che so it's nog a full on flameware, luckily for the project
andrzejku_ hey 17:58
El_Che nine: Most of the discussion was pretty much non sensical imho. I don't think LWN readers would care about that 17:59
iBakeCake andrzejku_: hey 18:05
[Coke] doc folks: why do we have predefined-search-terms? Why is it not inlined in template/search_template.js ? 18:24
FROGGS o/ 18:32
iBakeCake \o
[Coke] also doc: perl6 htmlify.p6 --parallel=3 # eventually dies with fork: Resource temporarily unavailable
moritz uhm, I don't understand the --parallel implementation 18:34
I would have expected it to create a ThreadPoolScheduler with this as the max thread count 18:35
[Coke] I was merely hoping it would make things go faster. :)
AlexDaniel [Coke]: why not? 18:54
[Coke] because it's an extra file, causes extra work to happen during the build, and the file is in the top level when it shouldn't be. Why not split search.js out into 30 different files? 18:55
dalek c/remove-predefined: 42c2cd7 | coke++ | / (3 files):
inline predefined-search-items
 18:56
AlexDaniel I kinda liked the way these terms were separate from the actual js stuff, but whatever. 18:57
[Coke] puts out github.com/perl6/doc/pull/980 instead of just merging it. 18:58
gfldex it was a kludge before the inlining and it still is a kludge 18:59
it would have made the link rewrite stuff really hard to read so I just dropped those 3 items in there
[Coke] if we decide to keep it, we should move it into template/thing.js instead of a top level file. 19:00
AlexDaniel yea, sure
travis-ci Doc build passed. Will "Coke" Coleda 'inline predefined-search-items' 19:17
travis-ci.org/perl6/doc/builds/169322248 github.com/perl6/doc/commit/42c2cd782ad3
moritz fwiw there's a python library that aspires to provide MAIN-like functionality: begins.readthedocs.io/en/latest/guide.html 20:17
El_Che moritz: I am a fan of docopt. Started in the python world, although I use it in Go 20:22
El_Che moritz: docopt.org/ 20:23
moritz El_Che: the idea of generating code (even if it's just a signature) from docs somehow scares me 20:25
El_Che moritz: it works very well, though 20:26
Go has the same time limited getopt builtin. Very un-UNIX like.
and docopt fixes that and you don't have to keep code and doc in synx 20:27
generating doc from code would be nicer, though 20:28
moritz PSA: I'll reboot the *.p6c.org machines soonish, because of CVE-2016-5195 20:50
seatek I really love how perl's generating usage messages from MAIN signatures now
that combined with being able to attach #|{ } comments to methods has made things so much easier for documenting as I got
go 20:51
moritz just hopes all the servers come up again after the reboot
timotimo uh oh 20:52
seatek chaos 20:53
perlpilot too many secrets 20:57
moritz reboot status: www and irc didn't come up automatically after the host reboot 20:59
I've changed the KVM config to do that next time
and now I have to reboot my IRC machine (and the IRC logger) as well :( 21:00
bbl&
MasterDuke moritz, [Coke]: i tried to make the --parallel option to htmlify.p6 relatively straightforward, what do you find confusing? 21:19
[Coke] MasterDuke: well, if I run it with --parallel, it fork bombs. 21:23
[Coke] er, --parallel=3 21:23
MasterDuke yeah, i get lots of crashes any time i try to use start() 21:24
though i found that running --parallel=10 a couple times until it didn't crash was still faster than running serially
actually, i think that fork error is new, i used to just get segfaults 21:25
[Coke] Laters. 21:30
timotimo twitter.com/ckolderup/status/789002880246177792 - let's not F things up this badly please :) :) 22:13
regreg how do i print the version of the perl interpreter from within perl6? 22:48
timotimo m: say $*PERL.version; say $*VM.version 22:50
camelia rakudo-moar b1f77c: OUTPUT«v6.c␤v2016.10.15.g.715.e.39.a␤»
dalek rl6-most-wanted: e7a4871 | (David Warring)++ | most-wanted/modules.md:
Want modules Pod::Checker and Test::Pod
 22:59
samcv is there a module i can use to download a page over http/https and then be able to process headers and only download more if the page is text? and then stop the request once i get the information i need out of the body of the html page? 23:02
not downloading the whole thing 23:03
gfldex samcv: you could subclass HTTP::Client github.com/supernovus/perl6-http-client 23:13
samcv how do i stop it once i have the data i need? 23:14
i need to request the page and read the header, and if it's not a text document not to download further. atm i'm using curl's command line program with -i, and then just terminating it if the header shows doc isn't text 23:15
-i will get the whole page but just put the headers first. so is easy to do it in one go, but it would be nicer to do it some other way 23:16
timotimo well, HTTP defines a HEAD (or HEADER?) command that'll tell the server not to transfer the actual document 23:33
geekosaur HEAD 23:34

Powered by Cro and the Raku® Programming Language.

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

Thank you!