»ö« Welcome to Perl 6! | perl6.org/ | evalbot usage: 'p6: say 3;' or /msg camelia p6: ... | irclog: irc.perl6.org or colabti.org/irclogger/irclogger_log/perl6 | UTF-8 is our friend! 🦋
Set by Zoffix on 25 July 2018.
MasterDuke timotimo: attempting to --profile Juerd's test file segfaults, is that a known problem? 00:34
vrurg For a type .^name gives its full name, .^shortname – bare type name. But how do I get the containing module name directly, not by cutting off shortname from name? 00:40
MasterDuke Juerd: i can confirm, your test file takes about the same amount of time as before jnthn++'s commits today 00:55
AlexDaniel: ^^^
but the isolated test case (which may not be all that representative after all) is faster 00:56
benchable6: 2018.11,HEAD my $r = q|\d+|; "asdf123fdsa321" ~~ /<$r>/ for ^50_000; 00:57
benchable6 MasterDuke, starting to benchmark the 2 given commits
MasterDuke, ¦2018.11: «1.7040» ¦HEAD: «1.2530»
MasterDuke, benchmarked the given commits and found a performance difference > 10%, now trying to bisect
MasterDuke, ¦2018.11: «1.7040» ¦52c2af5: «1.6786» ¦b781f8c: «1.6252» ¦af96fbb: «1.6870» ¦5cc0e02: «1.6349» ¦fecfb22: «1.6294» ¦e747b19: «1.3147» ¦adb8560: «1.3888» ¦HEAD: «1.2530» 00:58
Juerd MasterDuke: I've found things making my benchmarks harder; I've switched to KDE Plasma and for reasons I haven't figured out in the past hour or so, performance is different under KDE than it was under XFCE, and less reliably similar :( 01:33
Everything is noise and I can't find why. No processos show up in top, so it can't be much... 01:34
s/processos/processes/
leont IME benchmarks can be disturbed by a lot of things 01:41
Once had bizarre results that were apparently caused by having a youtube window in the background 01:42
Juerd I found that my script ran faster when I had firefox running. 01:45
But not under KDE
Er, not under XFCE
It was faster with firefox running under KDE
So this is sufficiently fucked up that I cannot trust any benchmark anymore.
(Note: firefox running means just that: firefox, with no extensions loaded, on about:blank) 01:46
Really, the regex.t from Net::MQTT runs in 14 seconds without firefox in KDE, and about 9 with firefox running. Or 7 to 8 seconds in XFCE regardless of whether firefox is running. 01:47
I give up
timotimo MasterDuke: i wonder if it's memory exhaustion? 02:24
buggable New CPAN upload: PDF-Class-0.3.2.tar.gz by WARRINGD modules.perl6.org/dist/PDF::Class:cpan:WARRINGD 02:34
mowcat hi recently picked up perl after a long break and realised perl6 is a thing! :) nice to meet you... 02:38
Kaiepi welcome o/ 02:50
timotimo cool, hi! 02:58
vrurg actually, Perl6 is not 'a thing', it's 'The Thing'... ;) 03:11
timotimo haven't you heard perl6 is the word? 03:12
vrurg does it contradic to each other? 03:13
*contradict
bl4ckm4n_ where is larry
timotimo it's hard to not be a thing when you are also the thing
bl4ckm4n_ it's easy to be a tooll 03:14
vrurg it's also hard not to be a god if you're The God. But some may not agree with me... 03:16
;)
vrurg BTW, seriously. I defined own metamodel HOW module to have a specific kind of classes. How can I register those classes with another module when modules with them a being require'd? Obviously, compose works only once. 03:19
In other words, I need a place where my code would be executed each time when a module is being required 03:20
mowcat _vrurg you're not wrong... lol 03:59
MasterDuke timotimo: the --profile segfault? 04:14
MasterDuke i don't think so, i've still got a couple gb free when it dies 04:19
timotimo: now i succeeded creating a profile, but the test file died after 74 tests with `===SORRY!=== P6opaque: no such attribute '$!restart' on type NQPMu in a Perl6::Grammar when trying to get a value` 04:20
timotimo sounds vaguely multithreading related 04:21
MasterDuke that might make sense, right? because of the evalling that's happening 04:22
it was a big profile too, 6.5mb sql file
timotimo does it actually eval from multiple threads? 04:24
MasterDuke don't think so, at least not explicitly 04:25
timotimo OK
MasterDuke huh, just got the SORRY! when running under valgrind, but it didn't have anything to say about it 04:29
anyway, off to sleep... 04:32
no-n waiting for the mailman to deliver my copy of Learning Perl 6 05:51
very excite
jmerelo Pod6 are now being rendered in Github!!!! github.com/perl6/Pod-To-HTML/blob/...EADME.pod6 06:22
yoleaux 13 Dec 2018 17:47Z <masak> jmerelo: yup. :)
Geth Pod-To-HTML: a0ebb8c91a | (JJ Merelo)++ | t/test.pod6
Adds from @kivikakk to test #55. Thanks for the issue
 06:26
Geth Pod-To-HTML: 50502ca1ab | (JJ Merelo)++ | t/011-external.t
Adds test

Refs #55, for the time being it does not reproduce the error. Will need more tests.		 06:42
Pod-To-HTML: 4209c9a83c | (JJ Merelo)++ | 2 files
Adds another test for #55
 06:55
Geth Pod-To-HTML: e4cebe3ceb | (JJ Merelo)++ | 2 files
Move pragmas to the top
 07:13
Pod-To-HTML: 74b964473b | (JJ Merelo)++ | t/011-external.t
Removes debugging info
buggable New CPAN upload: pod-load-0.0.1.tgz by JMERELO cpan.metacpan.org/authors/id/J/JM/...-0.0.1.tgz 12:34
New CPAN upload: Pod-Load-0.0.1.tgz by JMERELO cpan.metacpan.org/authors/id/J/JM/...-0.0.1.tgz
lucasb m: my @a.push(1,2,3); say @a # It never occurred to me before that this was valid syntax 12:40
camelia [1 2 3]
buggable New CPAN upload: POD-Load-0.0.1.tgz by JMERELO cpan.metacpan.org/authors/id/J/JM/...-0.0.1.tgz 12:43
jnthn m: [email@hidden.address] say @a # and this too :) 13:32
camelia [1 2 3]
jnthn Now, does anyone know *why* these two parse fine? :-) 13:34
sena_kun m: [email@hidden.address] say @a; 13:35
camelia [1 2 3]
sena_kun somebody has been golfing a bit too much for this universe to handle. D: 13:36
jnthn I think I've realized it before, but I ran into it just now because I lost the space when editing, then spotted it, and was like "grmbl, why's Comma not say that's a syntax error", then checked with Rakudo to see and...well, it ain't one. :-) 13:41
It's because the parser wants <.ws> there, and that is satisfied with <!ww> (that is, not between two word chars), and @ isn't a word char, so so far as <.ws> is concerned, it's happy between r and @ :) 13:42
And the reason for the .push working is that a declaration is just a term, and we can parse a postfix after any term. 13:43
lucasb This is interesting. Is it related with the fact the syntax 'my Foo $obj .= new' works? 13:45
I spotted the snippet in the post of day 12: Building a flexible grammar 13:46
for some reason I was expecting the need of parens there: (my @a).method(...)
tobs` yeah, declarations being "parsed as terms" means that the parens are there implicitly 13:48
jnthn lucasb: No, that's different; that is parsed as an initializer. The initial "assignment"/"binding" is always parsed specially, since it actually isn't a straight assignment in many cases. 13:51
Like `has Blah $.foo .= new` has to actually do it once per object creation 13:52
alnk Is the slack group still active? If so, how do I get an invite? 14:44
timotimo last message is three people joining #general on march 29th 2017 14:47
tobs` m: role A[$type] { method of { $type } }; role B[$n] { method n { $n } }; role C[$n] does A[B[$n]] {}; C[5].of.n.say 15:27
camelia (Mu)
tobs` it seems like the $n in A[B[$n]] is not in the same scope as the parameter $n to role C? Shouldn't it be?
m: role B[$n] { method n { $n } }; role C[$n] does B[$n] {}; C[5].n.say # shorter 15:38
camelia (Mu)
lucasb In 'role C[$n] does B[$n] {}', the $n in B[$n] is supposed to be the same value that was passed to C[5] ? 15:39
tobs` lucasb: yes 15:41
tobs` in that case, I want to print "5" 15:41
lucasb m: role B[::T] { method n { T } }; role C[::U] does B[U] {}; C[5].n.say 15:42
camelia (Int)
lucasb I was going to write C[Int] :) 15:42
but looks like types *are* captured, while scalar values are... not? Would you agree? 15:43
jnthn I think it only lets type captures through at present.
tobs` seems like it. Maybe it's a compile time issue?
jnthn I implemented that lot, but it was long enough ago I can't remember how possible it'd be to have the var case work; I guess if we can find a way to make them be treated a bit like compile-time constants it could maybe be made to work. 15:44
Or something along those lines...effectively, they need to be made subject to generic instantiation, iirc.
tobs` m: role B[$n] { method n { $n } }; role C {}; my $r = rand; (C.new does B[$r]).n.say 15:45
camelia 0.059105454651730116
tobs` this business happens all at runtime, right? 15:46
jnthn That's a runtime mixin, so a completely different case.
tobs` Ok, maybe I can rely on that in the meantime. Should I create a ticket about the capturing?
tobs` well, potential "meantime" :) 15:47
jnthn tobs`: Yes; at worst the result will be "can't make that work because X" 15:48
But it's worth a try; it'd be nice if it did work.
tobs` R#2546 16:09
synopsebot R#2546 [open]: github.com/rakudo/rakudo/issues/2546 RFC: Make role parameters reusable in `does` clauses
SmokeMachine m: role R[$a] { method bla { $a } }; class C does R[42] {}; C.bla.say 16:10
camelia 42
jmerelo Can someone help Vadim Belman in StackOverflow? stackoverflow.com/questions/537732...odel-class 16:22
buggable New CPAN upload: Pod-Load-0.0.1.tar.gz by JMERELO cpan.metacpan.org/authors/id/J/JM/...0.1.tar.gz 16:34
lucasb m: class C is Mu { method FALLBACK(|c) { dd c } }; C.foo; C.Numeric 16:35
camelia \("foo")
Use of uninitialized value of type C in numeric context
in block <unit> at <tmp> line 1
lucasb ^^ I want to intercep all method calls, even those in Mu
lucasb any way to "delete" those methods or inherit from a completely empty object? 16:36
*intercept
timotimo you should be able to override ^find_method 16:40
and probably also ^publish_method_cache
check out Grammar::Debugger, i think it has a few pieces that can help you out, lucasb
lizmat lucasb: modules.perl6.org/dist/InterceptAllMethods 16:43
lucasb timotimo, lizmat: thanks! I'll check out those 16:56
timotimo liz' dist is all you need i think 16:57
Geth Pod-To-HTML: cdc6266fab | (JJ Merelo)++ | 4 files
Pod::To::HTML eliminated from class names in external files.

Refs #55. Still the same problem with strings, but that's not a blocker right now. Please @kivikakk download new version to check this works correctly, ping me if it does not.		 17:00
buggable New CPAN upload: Pod-Load-0.0.2.tar.gz by JMERELO cpan.metacpan.org/authors/id/J/JM/...0.2.tar.gz 17:04
lucasb ^^ has anyone ever downloaded a CPAN distribution tarball announced by buggable? Is there a more useful URL to show? 17:08
jmerelo lucasb: it looks like it's legit...
lucasb: you mean like instructions for zef or something like that?
lucasb hmm, site doesn't know about modules.perl6.org/dist/Pod::Load:cpan:JMERELO 17:11
jmerelo: I mean the line the IRC bot just announced
I don't think a tarball link is very useful 17:13
jmerelo lucasb: two different things, I guess. Buggable prints some information it gets from, really, I don't know where. That can be changed. But it's not buggable who uploads the information to modules.perl6.org 17:15
jmerelo lucasb: to tell you the truth, I'm not sure where that's taken from. 17:16
jmerelo lucasb: it might take a while... 17:19
lucasb ah, maybe it's here: github.com/zoffixznet/perl6-buggab...m6#L39-L44 17:21
jmerelo yep, that's the line it announces new uploads. But I have no idea where it reads it or why it's not yet on modules.perl6.org 17:24
lucasb: took a while, but it got there: modules.perl6.org/dist/Pod::Load:cpan:JMERELO 17:25
lucasb cool, thanks for publishing modules, jmerelo++ :) 17:27
I see you made Pod::To::HTML depend on that 17:28
jmerelo lucasb: right. It fixes a bug that was kind of silly, but bug anyway github.com/perl6/Pod-To-HTML/issues/55
lucasb: part of a bug, anyway.
lucasb: when that gets fixed, we might have pod6 displayed on GitHub :-) 17:29
lucasb just a nitpick, to avoid the hard-coded temporary path: 17:33
m: dd $*TMPDIR.add('perl6-pod-load')
camelia IO::Path.new("/tmp/perl6-pod-load", :SPEC(IO::Spec::Unix), :CWD("/home/camelia"))
lucasb and maybe assign it to $*POD-LOAD-TMPDIR and use that? :) 17:34
jmerelo lucasb: hum
Good idea
lucasb jmerelo++, I see you made to commit, but IO::Path.add only creates "logical" paths, you still has to mkdir it IIUC :) 17:40
jmerelo lucasb: it actually created the directory, I think 17:43
lucasb: let me check
lucasb maybe it's not need if CompUnit::PrecompilationStore::File creates the dir... 17:46
*needed
[Coke] jmerelo: commits messages are not a great place to put messages about process/workflow to other developers. 17:47
jmerelo [Coke]: um. Sorry. Which one do you mean? 17:50
lucasb: ah, OK, right...
[Coke] github.com/perl6/Pod-To-HTML/commit/cdc6266fab 17:51
jmerelo [Coke]: sorry, but I don't follow you. You mean I shouldn't have pinged the OP of the issue? Because he's going to be getting the message anyway? 17:52
AlexDaniel I agree with [Coke], but I also understand what you wanted to achieve 17:57
perhaps write a comment next time?
I mean here github.com/perl6/Pod-To-HTML/commi...t_comments 17:58
jmerelo AlexDaniel: I would appreciate if you told me why, really. I'm not arguing against, it's only that I don't understand what's the problem. Can you please elaborate? 18:00
AlexDaniel [Coke] I will try not to do it again... as long as I understand what's wrong.
AlexDaniel jmerelo: it just feels inappropriate, like saying in the commit message that the weather is nice today 18:01
jmerelo AlexDaniel: you are talking about the part on strings having the same problem. 18:02
AlexDaniel: it was the motivation for not closing the issue. The OP had been addressed, actually.
AlexDaniel jmerelo: no, about “Please @kivikakk download new version to check this works correctly, ping me if it does not.”
jmerelo AlexDaniel, [Coke] OK. 18:03
Now what _I_ don't appreciate is that I am putting today a good amount of work on getting pod6 to be rendered in GitHub and that's the only comment I get about it. 18:04
But OK, so be it.
lucasb hehe :) 18:05
I'm sure we *all* appreciate what you are doing! I sure do! Pod 6 in... *GitHub*. Thank YOU for that.
lizmat jmerelo++ 18:06
lucasb folks are just commenting it looks strange personal messages in commits. Commits messages are more permanent than issues. :) 18:07
AlexDaniel jmerelo: welcome to open source :) we do appreciate your work, of course, people just suck at expressing that
jmerelo: recently I especially liked the stuff you do for the advent calender, like pinging authors to make sure posts happen 18:08
jmerelo lucasb: OK, I understand perfectly everyone's concerns. This person is not really familiar with Perl 6, I understand. I just thought it would be better for him to instruct on what to do.
I don't usually do that for the Perl6 crew 18:09
AlexDaniel: OK, OK, no problems. Thanks for pointing that out anyway. AlexDaniel is right that it would be better to put that in a comment, it's more transparent and public
Never mind me, really :-) 18:10
jmerelo feels sorry for lashing out.
AlexDaniel it's fine, it's an actual problem
not sure how to improve the situation, people should just learn to say “thank you” more often, I guess? 18:11
many times I appreciate someone's work but I say nothing 18:12
lucasb I confess I rarely say it too... it's like its implicit! These are all gifts, of course I'm grateful for all of this :) 18:13
jmerelo And thanks, AlexDaniel, lucasb, lizmat, [Coke] :-) 18:14
Now a problem
m: say "This ,is, a =large string , and !ugly".words.map( .substr(1) ).grep( /:alpha/ )
camelia 5===SORRY!5=== Error while compiling <tmp>
Unrecognized regex modifier :alpha
at <tmp>:1
------> 3 !ugly".words.map( .substr(1) ).grep( /:7⏏5alpha/ )
expecting any of:
argument list
term
jmerelo m: say "This ,is, a =large string , and !ugly".words.map( .substr(1) ).grep( /:alpha:/ ) 18:15
camelia 5===SORRY!5=== Error while compiling <tmp>
Unrecognized regex modifier :alpha
at <tmp>:1
------> 3 !ugly".words.map( .substr(1) ).grep( /:7⏏5alpha:/ )
expecting any of:
argument list
term
jmerelo m: say "This ,is, a =large string , and !ugly".words.map( .substr(1) ).grep( /\w/ )
camelia No such method 'substr' for invocant of type 'Any'
in block <unit> at <tmp> line 1
jmerelo m: say "This ,is, a =large string , and !ugly".words
camelia (This ,is, a =large string , and !ugly)
jmerelo m: say "This ,is, a =large string , and !ugly".words.map( *.substr(1) )
camelia (his is, large tring nd ugly)
jmerelo m: say "This ,is, a =large string , and !ugly".words.map( *.substr(1,1) )
camelia (h i l t n u)
jmerelo m: say "This ,is, a =large string , and !ugly".words.map( *.substr(1,1) ).join("") 18:16
camelia hiltnu
jmerelo I'm trying to get an unique filename out of a string
jmerelo Could use SHA, but I don't want any external dependencies. Doing that above would make sense? 18:16
Unless there are many words, that would work, right? 18:17
lizmat is too tired to think straight, so goes afk&
jmerelo There's the nice Temp::Path by zoffix, but first it uses SHA1, and second it's another external dependency 18:18
jmerelo m: say "This ,is, a =large string , and !ugly".words.map( *.substr(1,1) )[^3].join("") 18:19
camelia hi
vrurg jmerelo: thanks for taking care of my question! 18:38
jmerelo vrurg: still no answer, though...
vrurg I have a simple solution of imposing a requirement of placing a registration call in the body of the class. Though I don't like it... 18:40
jmerelo vrurg: did you look up EnumHOW? It's kind of curious, but it's got also some kind of "registration" if I remember correctly... 18:42
vrurg jmerelo: not yet. Thanks for the idea! 18:44
Xliff m: say 3 == True 18:45
camelia False
Geth Pod-To-HTML: 8b32091110 | (JJ Merelo)++ | 2 files
Mostly cosmetic changes
 18:46
tobs m: say 3 +^ True # :P 18:47
camelia 2
jmerelo Can someone please check if zef install Pod::Load works and what version it returns? 19:07
jmerelo I don't get why it's not working here: travis-ci.org/perl6/Pod-To-HTML/jobs/468129339 19:07
jmerelo Also, check out this on how multis work stackoverflow.com/questions/537856...cting-them 19:12
lucasb jmerelo: can't install here: No candidates found matching identity: Pod::Load 19:14
jmerelo Um 19:16
But it's right here: modules.perl6.org/dist/Pod::Load:cpan:JMERELO 19:17
jmerelo Maybe it takes a while to generate new cpan lists? 19:18
tobs jmerelo: re Constraining multis and its use for selecting them: that's because the where clause constrains an Any type (`$file where .IO.e`) and the Str type of the other candidate already wins other Any. If you change the second signature to `Str $file where .IO.e` it'll work.
sorry, I know you want to boost stackoverflow, but I don't have an account there. 19:19
El_Che hey jmerelo, you've been busy!
jmerelo: great work
jmerelo tobs: Good one. I hadn't thought about that
El_Che: still busy, but thanks.
tobs: wouldn't you consider opening one? 19:20
tobs (I guess there is less of an excuse to not have an account in the age of password managers...)
jmerelo tobs: but that's right, if I change the definition of the subset to Any it does not work :-) 19:21
tobs: answer it there and I'll give you lots of karma :-)
tobs and the subset you give constrains Str, so it call checks out
jmerelo El_Che: thanks!
tobs s/call/all/ 19:22
jmerelo tobs: totally correct. Thanks! 19:23
buggable New CPAN upload: Pod-Load-0.0.3.tar.gz by JMERELO modules.perl6.org/dist/Pod::Load:cpan:JMERELO 19:24
tobs jmerelo: maybe answer the question yourself in the meantime. These "might as well" / "might as well not" questions are always the hardest to decide :-) 19:24
jmerelo tobs: not really. That would be cheating, and it's your answer :-) I'll leave someone the chance to answer it herself :-)
jmerelo OK, the ecosystem is updated once a day: github.com/ugexe/Perl6-ecosystems. So it will take until tomorrow for the new modules to be available. 19:27
I'm finding out lots of stuff today about the Perl6 ecosystem...
Also, I wonder why that repo is not under the Perl 6 team... 19:28
jmerelo .seen masak 19:39
yoleaux I saw masak 13 Dec 2018 18:47Z in #perl6: <masak> b2gills: I guess my counterargument isn't that that's _wrong_, it's that it's nothing new -- a WhateverCode is just sugar for writing it out as a normal parameterized block
jmerelo .tell masak you ready to schedule your next advent calendar article during tomorrow, right? The 16th is yours! 19:40
yoleaux jmerelo: I'll pass your message to masak.
jmerelo .seen RabidGravy
yoleaux I saw RabidGravy 17 Dec 2017 01:40Z in #perl6: <RabidGravy> that's perl6advent.wordpress.com/2017/12/...ier-title/ if you want to get in early
jmerelo tobs: you've got everything you need for your Advent Calendar article? 19:46
Check out today's Advent calendar entry everyone, on processing ASN.1 perl6advent.wordpress.com/2018/12/...th-perl-6/ 19:47
tobs jmerelo: sadly no, but I'm working on it as we speak. I set up a schedule for myself and have that in mind. 19:48
jmerelo tobs: you have a Wordpress ID and it's added to the site? 19:49
tadzik that's a name I haven't heard in a looong time
tobs oh that, yes. You set me up some time ago :-)
jmerelo Ah, OK. :-)
Anything you need, just let us know :-)
tobs o7
Xliff Anyone around here with Cairo experience? 20:22
I need some help debugging some transliterated code.
buggable New CPAN upload: Tomtit-0.0.1.tar.gz by MELEZHIK cpan.metacpan.org/authors/id/M/ME/...0.1.tar.gz 20:34
buggable New CPAN upload: Tomtit-0.0.2.tar.gz by MELEZHIK cpan.metacpan.org/authors/id/M/ME/...0.2.tar.gz 20:54
New CPAN upload: Tomtit-0.0.3.tar.gz by MELEZHIK cpan.metacpan.org/authors/id/M/ME/...0.3.tar.gz 21:14
New CPAN upload: Tomtit-0.0.4.tar.gz by MELEZHIK cpan.metacpan.org/authors/id/M/ME/...0.4.tar.gz
pmurias is there a module to do qq:x in a safe manner? (without passing stuff to a shell) 21:19
pmurias it's a bit surprising that a failure as the last thing in a for is silently ignored :( 21:45
lucasb Do you have an example? 21:57
pmurias for @files -> $path_that_cant_be_create {spew($path_that_cant_be_created, "content")} 21:58
lucasb m: my $x = Failure.new; $x; say 'OK' 21:59
camelia WARNINGS for <tmp>:
OK
Useless use of $x in sink context (line 1)
lucasb you were suprised that failures don't throw in sink context? well, me too :) 22:00
is it 'spew' that returns the failure? spew(...).sink helps? 22:01
lucs 22:02
timotimo you can also use .self to explode failures
lucasb which makes of '.self' not an "identity" method, right :) 22:03
timotimo well ... i guess?
lucasb otherwise it would have returned the failure untouchted
*untouched 22:04
timotimo well, it would have returned the failure untouched
if it didn't throw an exception
lucasb yes. We are agreeing that '.self' activates the failure. I just wouldn't have made it that way :) 22:05
timotimo oh, no, .made is only for Match :) 22:06
lucasb haha you
lucasb the alternative could be: $obj.self(:throw-failures) :) 22:07
lucasb P6 should have abused more of adverbs to tune the behavior of things 22:08
timotimo i think necessity of adverbs is considered a design smell 22:16
quotable6: design smell
quotable6 timotimo, OK, working on it! This may take up to three minutes (4582161 messages to process)
timotimo, 4 messages (Inf⌁-Inf): gist.github.com/a52d547a49c4412d30...6d3edab687 22:17
lucasb but ok, good point :) 22:18
timotimo ha 22:25
quotable6: "design smell"
quotable6 timotimo, OK, working on it! This may take up to three minutes (4582161 messages to process)
timotimo, 138 messages (2006-03-20⌁2016-09-26): gist.github.com/eec57471562f6684d6...d04047d994 22:26
Geth ecosystem: ec20c4fe7a | Altai-man++ (committed using GitHub Web editor) | META.list
Add ASN::BER
 22:32
lucasb hmm, quotable6 could output the nickname for better context 22:33
timotimo yes, but then it'd have to have its database rebuilt :) 22:34
it's not a performance beast 22:35
m: say 4582161 / (60 - 28 + 19)
camelia 89846.294118
timotimo 90k regex matches per second.. could be worse i guess?
jn's improvements to regex performance yesterday could have an impact 22:36
Xliff timotimo: Heya. I've run into a problem with my latest example, that looks like it's Cairo based. Rather it looks like what I added to Cairo is messing things up. 22:38
timotimo: When you have time, could you take a look?
Now I think I understand why you left cairo_path_t as a CPointer. :/ 22:39
Problem is in this script: github.com/Xliff/p6-Pango/blob/mas...-twisted.t 22:40
Xliff Which was transliterated from: github.com/GNOME/pango/blob/master...otwisted.c 22:41
timotimo my brain is really not working very well this week, and that's a few screen pages of code :S 22:42
Xliff I know. Like I said... whenever you have some time.
I'm assuming it's my port of cairo_path_t that's the problem, however I can't see it, yet.
I do know one thing, It looks like the curves are missing the first point. 22:43
And that's just the curved part. I haven't even started the text part.
And to make things worse... the P5 version is lightning fast and my P6 version takes 70-80 times that. 22:47
timotimo: Oh well. Thanks In Advance. Hope you have a restful next week.
Oh, and for the curious. The p5 version is here: github.com/GNOME/perl-Cairo/blob/m...ed-text.pl 22:48

Powered by Cro and the Raku® Programming Language.

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

Thank you!