»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.
00:31 Khisanth left 00:32 Khisanth joined 00:33 cdg left, cdg joined 00:38 cdg left 00:41 Khisanth left 00:42 Khisanth joined 00:59 Zero_Dogg left 01:00 Zero_Dogg joined, Zero_Dogg left, Zero_Dogg joined, BenGoldberg joined 01:01 rgrinberg joined 01:12 Actualeyes joined 01:15 kurahaupo left 01:16 kurahaupo joined 01:19 stevieb left 01:20 kalkin- joined 01:21 adhoc joined, adhoc left 01:22 molaf left 01:23 kalkin-_ left 01:24 bjz left 01:25 sammers joined
sammers good morning from Japan 01:26
diakopter o/
sammers is there a way to get the current line / row index when looping over slurped file lines? 01:27
01:31 kurahaupo left 01:32 kurahaupo joined, kurahaupo left 01:34 ssotka left 01:35 molaf joined 01:41 yqt left 01:42 sufrostico left 01:46 ilbot3 left 01:47 kid51 left, ilbot3 joined 01:48 tomboy64 joined 02:01 Ven left 02:05 Lazarus joined, nebuchadnezzar joined 02:06 Lazarus is now known as Guest2253
Guest2253 help 02:07
help anyone 02:08
02:08 Guest2253 left 02:12 tbrowder_ joined
tbrowder_ someone needs help? what can we do for you? 02:14
geekosaur they left already 02:17
tbrowder_ I think most of the heavy hitters have left for the evening
nite all 02:18
02:18 tbrowder_ left 02:19 Score_Under joined 02:33 sQuEE joined, cdg joined 02:35 noganex joined 02:38 noganex_ left
dj_goku p6: for ['a','b','c'].kv -> $i, $v { say "index $i has a value of $v" } 02:46
camelia rakudo-moar 1254bb: OUTPUT«index 0 has a value of a␤index 1 has a value of b␤index 2 has a value of c␤»
02:46 cdg left
dj_goku sammers: ^^ 02:46
02:47 cessna left 02:48 assgrass joined 02:49 assgrass left, assgrass joined 02:50 assgrass left 02:51 grassazz joined, grassass left, Actualeyes left 02:54 huggable left, huggable joined 02:56 skids joined 02:58 woodruffw joined 03:03 NEveDko_ left 03:09 raiph joined 03:11 raiph left, NEveDko joined 03:12 raiph joined 03:13 raiph left, raiph joined, sortiz joined
sortiz \o #perl6 03:14
yoleaux 5 Jun 2016 12:51Z <FROGGS> sortiz: I you wanna tackle something in XML::LibXML, pick either a still failing test file or grab a new test file from t/to-be-ported and move it to t
dj_goku sammers: probably this is what you want: for $file.lines.List.kv -> $index, $line { say $index, $line }
sortiz .tell FROGGS Sure, test-driven is the better approach. 03:16
yoleaux sortiz: I'll pass your message to FROGGS.
03:17 raiph left 03:18 raiph joined 03:27 raiph left, raiph joined, grassazz left 03:28 grassaSS joined 03:31 raiph left, raiph joined 03:32 raiph left, raiph joined 03:36 ssotka joined 03:38 mr-foobar left 03:39 mr-foobar joined 03:43 raiph left 03:44 raiph joined 03:50 skids left 03:57 Actualeyes joined 04:06 sortiz left 04:08 jack_rabbit joined 04:11 mr-foobar left, molaf left 04:17 khw left 04:20 Cabanossi left 04:22 Cabanossi joined 04:33 grassaSS left 04:51 kaare_ joined 05:06 araujo_ left 05:08 Zero_Dogg left 05:09 Zero_Dogg joined 05:20 sno left 05:31 rgrinberg left, MadcapJake left 05:37 wamba joined 05:38 jrusso joined, jrusso is now known as MadcapJake 05:39 domidumont joined 05:43 domidumont left 05:44 domidumont joined 05:57 domidumont left 06:04 rurban joined 06:05 BenGoldberg left 06:08 domidumont joined 06:13 rurban_ joined 06:16 vermiforme joined
vermiforme hey hey 06:16
newbie here
nine vermiforme: welcome! 06:17
yoleaux 7 Jun 2016 11:33Z <llfourn> nine: I've tested the bugfix branch, results: RT #128156. Sorry I didn't respond sooner. I only just saw your request for tests. RT doesn't seem to be forwarding me follow-ups for this thread for some reason =(.
synopsebot6 Link: rt.perl.org/rt3//Public/Bug/Displa...?id=128156
vermiforme wondering if there was interest in SPARQL and RDF here
thanks nine 06:18
06:19 sno joined
nine .tell jdv79 do you happen to use Perl 5.24? 06:21
yoleaux nine: I'll pass your message to jdv79.
vermiforme I'm sick of procedual database/web stuff. I want , for lack of a better term, "Lazy Functional reasoning"
nine vermiforme: sounds intriguing 06:22
vermiforme Yes nine
I work/hate .NET and PHP, and there are good ideas going on there 06:23
LINQ is a way forward, but I want more 06:24
nine That's the spirit :) 06:25
vermiforme The tricky part is the balance between parsing results, and composing the query 06:26
The thing I find with dealing with query result, is the freking loop 06:30
'night nine, will explore some other related channels. might come back 06:32
\q
06:32 firstdayonthejob joined 06:33 vermiforme left 06:34 darutoko joined 06:41 firstdayonthejob left
Xliff vermiforme has some nice ideas, but it will be tricky building a <yourlanghere> -> SQL parser. 06:52
06:52 lordbanana joined 06:53 domidumont left 06:54 huggable left, huggable joined
Xliff Tony-O's Slang::SQL is about the closest out there: github.com/tony-o/perl6-slang-sql 06:55
sammers, you can do this: for $file_handle.lines.kv - > $k, $v { ... }. Where $k is your index and $v is the line text. 07:01
07:07 _mg_ joined 07:20 abraxxa joined
masak good morning, #perl6 07:22
07:22 azawawi joined
azawawi Good morning 07:22
is there a way in Perl 6 to access a private attribute in a subclass other than making read/write to the outside world? 07:23
moritz azawawi: you can have a private accessor method, and then in the parent class explicitly trust the subclass 07:24
07:25 rindolf joined
azawawi moritz: do you have an example snippet please? 07:26
07:29 abraxxa left, breinbaas joined, zakharyas joined 07:30 abraxxa joined, g4 joined, bjz joined, wamba left
moritz azawawi: perlpunks.de/paste/show/5757c9d4.6bd2.391 07:31
azawawi moritz: but that means a parent needs to know about its children :) 07:32
moritz azawawi: sure; otherwise *any* class could declare it to be a children, making the privacy of the private attribute worthless 07:33
s/children/child/
azawawi wishes if Perl 6 had protected
moritz azawawi: if you allow any odd subclass to access the private attribute, it's part of the public API
and you could just make it public 07:34
azawawi i will give u the problem im facing right now
moritz from an encapsulation standpoint, protected isn't better than public. It just gives you a false sense of "this is not *really* public"
azawawi github.com/azawawi/gtk-simple/blob...Button.pm6
jast well, 'protected' is a hint to developers, mainly
azawawi now RadioButton should be a child of CheckButton per GTK documentation 07:35
jast except in languages where 'protected' limits access to the same namespace and extending a namespace from an third-party module is impossible
azawawi making $!gtk_widget an accessor = changing all the GTK::Simple:: classes 07:36
github.com/azawawi/gtk-simple/blob...Button.pm6
github.com/azawawi/gtk-simple/blob...Button.pm6
07:37 ssotka left
moritz azawawi: nah, it's enough if you write a public accessor in CheckButton, and then use that in it's subclasses 07:37
you don't have to touch every GTK::Simple class for that
masak I'm with moritz on this one; protected is a sham and a dilution of what little purity OOP has
azawawi ideally RadioButton is a CheckButton which is a ToggleButton which does a Widget
moritz azawawi: or you could give it a public accessor, but continue to use the private attribute where you can 07:38
azawawi cool
i will try that
07:46 azawawi left
DrForr Egads. I've been nperez'd. 07:49
07:50 bjz left
moritz DrForr: what does that mean? 07:50
07:51 Ven joined
DrForr I submitted 4 talks so ::EU could choose, I just got all 4 accepted. 07:51
Fixing this momentarily, or at least explaining.
moritz DrForr: well, if that's too much, you could always only accept some of them, and talk to the organizers :-) 07:52
the earlier, the easier for the organizers
DrForr Well, luckily I sit 3 desks away from the organizer in question.
moritz hated last minute schedule changes at GPW2016 07:53
07:56 Ven left 08:02 zakharyas left 08:03 TreyHarr1 joined, TreyHarris left
DrForr Hrm. my $T̶e̶s̶t̶-foo = 1; # I see skewer-case in our future. 08:04
08:06 bjz joined
Xliff .tell FROGGS If you need more help with XML::LibXML test cases, let me know! 08:19
yoleaux Xliff: I'll pass your message to FROGGS.
DrForr Heyo - apparently I've forgotten the invocation in here-docs to suppress interpolating {} blocks. 08:20
:c maybe?... 08:21
Xliff DrForr, you mean q:to[something] still does interpolation?
DrForr Didn't remember if it was the ;to bits or not, looking now. 08:22
I've got p6 code that I need to write inside p6 code and escaping braces is annoying.
Xliff *nods* 08:23
Xliff smells the need for eval()
08:23 dakkar joined
DrForr Well, what I really need are p6 macros but I'll take what I can get. 08:23
Xliff I wouldn't mind eval() so much but... MONKEY-SEE-NO-EVAL 08:24
08:24 RabidGravy joined
Xliff That lil ugly bit has forced me to rethink several scripts I've developed over the past few months. 08:25
\o RabidGravy 08:26
RabidGravy harr
Xliff How goes?
RabidGravy fabulous 08:27
Xliff OK, here's my problem with OAuth2. It depends on a GUI. Well.... you could do without, but it would be a pain. 08:28
08:29 rindolf left
Xliff I've spent the past week trying to work around that pesky requirement. Only way I can think of doing it is grabbing the HTML, parsing the form fields, querying the user (if necessary) and then packaging the request back to the server. 08:29
jast the only thing it depends on is user interaction
08:29 vytas joined, dakkar left
jast yeah, usually that's your only option unless the provider offers a separate interface 08:29
Xliff jast: Yeah, but if you already have the credentials, it doesn't even need that.
08:30 TEttinger left
jast well, the idea with OAuth is having the user confirm that they're okay with sharing their data or something like that 08:30
DrForr chomp(q:to[END]) ~ "($a,$b)" # works, but *HOO* boy does that look ugly.
Xliff jast: Certainly could do without the need for a browser and a webserver. But that's just me. 08:31
jast otherwise websites could secretly steal your data (e.g. your e-mail address, in many cases)
Xliff Heh. I understand the purpose, but the implementation is overly complex.
jast I'm not exactly a fan of OAuth 2, either
Xliff =)
DrForr Xliff: This is why I had someone else do the implementation.
jast and don't get me started on the many different implementations in practice 08:32
Xliff DrForr: LOL! And what did you use, if you don't mind me asking?
DrForr And if you think that's bad try CORS for banking sites.
jast the worst offender is twitter where an app needs to register and use a secret key (for a public API...)
Xliff DrForr: Oh no.... I stay clear of the financial sector if I can help it!!!
jast and they block your key if they find it online, so you can't have an open source app with twitter support unless you cheat
Xliff jast: Eve-Online is just as bad. 08:33
DrForr What I said above. chomp(q:to[END]) ~ "($a,$b)" ~ '}'; # with the rest of the statement below.
08:33 raiph left 08:34 raiph joined
jast I just know that when I tried to write a tiny CLI twitter client, that's where I threw my hands up in defeat 08:34
timotimo DrForr: did you figure out it's :!c to not get curlies interpolated? 08:36
Xliff timotimo to the rescue.
I was just about to suggest using Q:to[]
timotimo with Q you don't get any interpolation whatsoever 08:37
08:37 Ven joined
timotimo but it looks like DrForr still wants to have $a and $b interpolated 08:37
Xliff Oh. Yikes!
I didn't catch that requriement. 08:38
timotimo have to read between the lines :)
Xliff Yes, but it is almost 5am here and the lines are starting to run together. ;) =P
timotimo ah 08:39
08:40 _mg_ left
DrForr timotimo: Yeah, it's kind of a "want it both ways" situation. What I'd really like is $x,(,$a,,$b) like CLISP macros :) 08:40
timotimo well, i don't know clisp macros at all
i just know its quote language is ridiculous :)
Xliff What is this sudden craving for sunflower seeds and chocolate?!?!?! 08:43
masak Xliff: maybe you're pregnant...? 08:44
Xliff knew someone was going to make that joke...
No. I may have a beer belly, but I am not pregnant....
I did have face-hugger nightmares recently.... hmmmm....
masak or could just be that sunflower seeds and chocolate are tasty 08:45
Xliff Let's go with that.
And the auto-doc....
08:45 dakkar joined
Xliff goes to find some chocolate. 08:45
08:46 rindolf joined
psch m: my ($a, $b) = 1, 2; say q[have some curlies: { }, but also have \qq[$a and $b]] 08:47
camelia rakudo-moar 1254bb: OUTPUT«have some curlies: { }, but also have 1 and 2␤»
psch DrForr: ^^^ that's how i'd do that. as in, drop into qq for what you want interpolated inside q
masak whoa.
didn't know we had that. 08:48
psch yeah, that happens to me quite a lot too :) 08:49
timotimo whoa. i also didn't realize you could \qq 08:50
Xliff \o/ 08:51
psch++
Woodi so, who implemented that ? aliens ?? 08:52
Xliff qq inside of q inside of..... Wait...almost got carried away!
Woodi is sometimes shocked that some ppls actually know *a lot* of Perl6 :) 08:53
08:55 rurban1 joined 08:56 Ven left
DrForr If it lets me do the entire block in one heredoc I'm all for it, other routes are uglier. 08:56
08:57 rurban left 08:58 araujo joined
DrForr Aah, much cleaner, thanks. 09:00
Feels like the old ${\{}} or whatever it was in p5, but at least the qq[] makes it more visible. 09:01
09:04 pat_js joined 09:07 Ven joined
psch DrForr: well, it works the other way around as well, fwiw 09:07
m: my ($a, $b) = 1, 2; say qq[have some curlies: \q[{ }], but also have $a and $b]
camelia rakudo-moar 1254bb: OUTPUT«have some curlies: { }, but also have 1 and 2␤»
psch DrForr: i guess the choice depends which is the more common desired behavior :) 09:08
+on
DrForr Pfft, it lets me lay out the code as I want without getting my eyes continually backslashed. 09:09
09:12 ocbtec joined 09:18 Actualeyes left 09:23 konobi joined
konobi saw something kinda cool that might be a useful tool for use 09:23
www.joyent.com/blog/550-regression...yent-manta
09:36 autarch joined 09:41 rindolf left 09:47 rindolf joined 09:49 vytas left
Ven .tell brrt interesting LuaJIT talk by one of the current DartVM engineers: mrale.ph/talks/vmss16 09:53
yoleaux Ven: I'll pass your message to brrt.
rurban1 brrt: note that I already argued with the dartvm devs about their bad usage of the int vs ptr tag usage. lua and most lisps did it better. ints need to be tagged, ptrs need to have 0b00. dartvm need to mask all ptrs, and int arith is basically the same with or without tags. 09:55
nan tagging also helps a lot, and relying on fast SSE arith 09:56
the optional continuation/meta ptr in the lua callframe is also much better for dynamic langs. 09:58
09:58 zakharyas joined
konobi Ven: terralang is also pretty interesting too 09:58
09:59 wamba joined
rurban1 and potion, tvmjit, wren, … and most lisps. 10:00
and lua 5.2 broke all this for proper int32 types 10:01
which is the same problem as in perl5. cannot use tagged ints 10:02
10:02 kid51 joined 10:04 rurban_ left
konobi rurban1: would an alternate mapping work? 10:04
10:05 rurban joined, tbrowder_ joined 10:06 rurban1 left 10:10 tbrowder_ left, labster left 10:13 bjz_ joined, bjz left
masak is there a way to `use` all of some module, except for one sub which I want to define myself? 10:14
if I just `use` it in the normal way, I get a redeclaration error
10:14 pat_js left
masak let's say for concreteness that I want all of `use Test;`, except for the `is` routine 10:14
10:20 g4 left 10:22 mr-foobar joined 10:24 aries_liuxueyang left 10:26 tbrowder left
RabidGravy m: use Test <!&is>; # something like this would be nice 10:27
camelia rakudo-moar d86dc4: OUTPUT«5===SORRY!5=== Error while compiling /tmp/BAyjjGrmei␤Error while importing from 'Test':␤no EXPORT sub, but you provided positional argument in the 'use' statement␤at /tmp/BAyjjGrmei:1␤------> 3use Test <!&is>7⏏5; # something like this woul…»
RabidGravy m: use Test :except<&is>; # something like this would be nice 10:28
camelia rakudo-moar d86dc4: OUTPUT«5===SORRY!5=== Error while compiling /tmp/SDNsKFh4N8␤Error while importing from 'Test': no such tag 'except'␤at /tmp/SDNsKFh4N8:1␤------> 3use Test :except<&is>7⏏5; # something like this would be nice␤»
masak I'm very far from requiring new syntax or anything of the sort
but I do have a real use case right now where it'd be nice to be able to selectively override some stuff from an imported module 10:29
RabidGravy oh no, it's all doable, but I suspect it's import only the things you want
masak I don't care how it ends up looking in the end. as long as it works
currently getting `no EXPORT sub, but you provided positional argument in the 'use' statement`
where can I read more about this EXPORT sub? S11 doesn't seem to have the full scoop 10:30
masak checks the spectests
konobi maybe an exclusion set? 10:32
10:33 cpage_ left
RabidGravy right, off out now 10:34
toodles
masak \o
by reading the Rakudo source code I found out what the EXPORT sub is and what it's expected to return 10:37
but I've found neither synopsis nor spectest to explain this
could be just me being bad at searching, of course
10:39 noninc joined
arnsholt The docs being incomplete wouldn't be *terribly* surprising either, though =) 10:42
masak oh, there's one `sub EXPORT` in the spectests
it's a bugfix
but it does show the general principle, which I already guessed correctly
arnsholt Heh 10:44
masak yay, now it works! :D
I can selectively import things :>
arnsholt That's one of the most useful things I've learned from the Perl 6 project, really. Effective source-code spelunking
masak yes, but that should never be the accepted status quo or the steady state
every time that happens, something else should happen too 10:45
10:45 wamba left, wamba joined
masak the Ember project doesn't have a "documentation" team, they have a "learning" team 10:45
because they figure it's never just about the document, it's about a comprehensive cross-cutting effort to bring people up the learning curve more effectively
konobi rurban: would an intermediary representation work for lua JIT and ints? 10:46
10:46 rindolf left 10:48 vytas joined 10:49 vytas left
arnsholt masak: Ooh, yeah. Calling it learning instead of documentation is a much better name 10:50
rurban no, luajit has no intermediate compile stage, no SSE. The ints it can unbox it unboxes, the rest are refs. 10:51
In lua 5.2 all ints are now refs, hence luajit refused to update to the 5.2 language.
10:51 aries_liuxueyang joined
rurban NAN tagging helps, as a double can hold all ints. But what about the ptr - int distuingishment for the ffi then 10:53
konobi rurban: no, but terralang.org does
10:54 huggable left, huggable joined 10:55 bjz joined, bjz_ left 11:00 gnull joined 11:04 wamba left 11:05 vytas joined 11:06 vytas left 11:07 vytas joined 11:13 pat_js joined 11:15 tomboy64 left 11:17 cpage_ joined
konobi rurban: with reflection too, iirc 11:18
11:19 buharin left 11:28 lordbanana left 11:32 noninc left
Xliff masak: Re selective export - What worked out in the end? 11:34
sjn also likes the "learning" perspective masak mentioned 11:35
that's a really good way to think about docs
Xliff sjn: What? Code diving?
sjn Xliff: re: masak's comment at 12:45 today 11:36
masak Xliff: defining that `sub EXPORT`, and having it put together a custom hash based on the parameter list.
Xliff Yeah. Marked 6:45 here cos.... EDT
sjn ah :)
masak Xliff: I can package it as a commit and link it here.
Xliff masak: Is there a link to the example you found?
sjn assumes everyone lives in central Europe :D
Xliff kk 11:37
sjn: And I just assume everyone knows GMT and just do the conversions in my head.
Even though I live East Coast US, I have learned that GMT is.... definitive.
masak Xliff: github.com/masak/007/commit/5b4eaa...e807cb4073 11:39
huf_ Xliff: GMT has been superseded by UTC because there's always another standard :) 11:40
Xliff masak++: Didn't know about the hash assignment-and-fetch. Nice to know. 11:45
11:45 xinming_ joined
Xliff huf_, until UTC becomes global... I'm ignoring it. 11:45
huf_ it's not global? 11:46
Xliff Not enough for me. Honestly I don't know how wide it's spread now but I am still seeing GMT instead of UTC...so
11:46 kid51 left
huf_ oh come on 11:46
i'm still seeing IV but the roman empire has fallen a while ago 11:47
Xliff And from Google: "There is no time difference between Greenwich Mean Time and Coordinated Universal Time"
So I think I'm safe! ;)
huf_ yes, which is why my original line had a :) on the end
Xliff Ah. Missed that.
11:48 xinming left
Xliff And just because something is "dead" doesn't mean that there aren't lingering organs still in use! =) 11:48
huf_ with strange aeons even death may die 11:49
Xliff I mean... there's still a Visual BASIC -- Haven't computers evolved enough to ignore that? It's like the human tail bone! It's there... but not.
11:53 kyrose joined
rurban I just had to fix schwern's time64 tests because he based those tests on PTC (Pacific) and they failed in CET, in certain hours in the evening, where europe is already in the new day. With UTC (greenwhich time) the tests are now much more stable to such off-by-ones. 11:59
does per6 use time64? many langs do. we fixed a lot there recently. see the pull requests. 12:00
but nothing serious, just performance and better tests
kyrose m: say "Hello everyone!\nHow it's going!"; 12:07
camelia rakudo-moar 9942c4: OUTPUT«Hello everyone!␤How it's going!␤»
masak Xliff: turns out there was already a hash with what I wanted, so I managed to get rid of some code: github.com/masak/007/commit/b98fb6...a0fa274026 12:08
Xliff: I can think of a way to get rid of the `%exports` variable and the for loop, too. but I haven't tried yet.
12:12 kyrose left 12:14 araujo_ joined
Woodi I just make naive code diving hour ago... I give up in 2 minutes or something like that :) rakudo/src/Perl6/Actions.nqp/[wanted|WANTED] - I even have no set of ideas what it can do :) 12:15
12:16 araujo_ left 12:17 araujo_ joined
masak Woodi: that's when you `git blame` and read the commit message. 12:17
Woodi also discovered that semicate problem is already discovered... however main solution is: drop C, use something with "option type" :) started to thinking C is simplest posible language ever... even asm can do more ;) 12:18
masak: o, excuse to learn git blame :)
thanx :)
12:18 araujo_ left 12:19 araujo left, araujo_ joined
masak in this case, it'll lead you to f95c14402, with the topic line "Great Sink Refactor" 12:19
so "wanted nodes" are connected to the whole "useless use" thing
12:20 rurban left, rurban joined
Woodi *semipredicate there... 12:21
masak excuse me? 12:22
oh, that backlog. got it. 12:24
12:25 rurban left
Xliff "drop C" -- *gasp* -- how could you?! =) 12:25
masak there are many excellent reasons not to write things in C nowadays :) 12:26
also, you're not considering language usage a zero-sum game, are you? because it mostly isn't
Woodi yea, Perl was announced in 80's ? :)
12:28 azawawi joined
Xliff masak: Yes, C can be a pain however...there was sarcasm. 12:28
Woodi masak: no, just simplicistic view
azawawi pasteboard.co/1wSaC4JE.png # GTK::Simple::RadioButton in action :)
Xliff And nowadays you use the language that's best for the task. Language interoperability is high!
Woodi and git blame requires a big screen...
masak or a small typeface
azawawi: neat! 12:29
12:29 iH2O joined
Xliff azawawi++ 12:30
Woodi any idea what new MONKEY-* can be later ?
Xliff How much of GTK is covered in GTK::Simple?
perlpilot Woodi: MONKEY-MONKEY ;) 12:31
Woodi perlpilot: but I asked about *code* :) 12:32
azawawi: any way to make examples to not be so damn gray ? I see IDE in screenshot is't such ugly... 12:34
Woodi have brain burned out by gtk industrial gray...
azawawi :)
github.com/perl6/gtk-simple/pull/49 12:41
azawawi going home & :)
12:42 azawawi left, pmurias joined
pmurias masak: has anything really taken the high performance thing that's easy to use from other language niche? 12:43
s/taken/taken over/
masak pmurias: not really, no. that's probably also a good use for C++.
pmurias: note that I'm not saying there are no uses left for C. that'd be a ridiculous statement at this point. 12:44
pmurias: I'm just saying there are reasons to choose something else in many cases. :)
12:44 _mg_ joined
perlpilot notes that fortran was once "the high performance thing that's easy to use from other language" 12:45
(and often the other language was C IME :)
12:45 rurban joined
stmuk_ I think go and rust have similar performance to C for many things 12:49
12:52 dj_goku left 12:54 rurban1 joined, Sgeo_ left 12:58 rurban left
arnsholt perlpilot: Yeah, I think Fortran is still the high-performance language of choice for those who are super into performance (climate sims and such, for example) 12:59
IIRC it's something to do with Fortran disallowing pointer aliasing entirely, which makes it easier to optimize
13:00 araujo__ joined, rgrinberg joined 13:01 skids joined
jnthn iirc, also recursion is fairly constrained 13:02
So you don't need call frames.
13:03 iH2O left
pmurias masak: C++ has it's own set of problems 13:03
13:03 araujo_ left
perlpilot prior to fortran 90, fortran didn't have recursion, so ... F77 anf F67 were really constrained in that regard :) 13:04
rurban1 can moarvm already do tail-call elimination?
masak pmurias: you keep making counterpoints to things I didn't really say ;) 13:05
pmurias: yes, C++ has its own set of problems, distinct from C's
rurban1 @stmuk_: go not, rust yes. pony even better than C++ with openmp, and safer and easier than rust
avalenn_ rurban1: is your pony www.ponylang.org/ ? 13:07
timotimo rurban1: moarvm cannot do that yet, no 13:08
13:16 wamba joined 13:24 stevieb joined 13:26 Khisanth left, RabidGravy left 13:27 cdg joined 13:32 sftp left, araujo_ joined, araujo__ left 13:34 sftp joined
rurban1 avalenn_: yes 13:35
13:35 azawawi joined
azawawi reporting for Perl 6 fun duty :) 13:35
github.com/perl6/gtk-simple/pull/49 # Any feedback before merging pull? 13:36
rurban1 timotimo: I'm also still working on proper tce in cperl. (tail call elimination). at least cperl has now proper tailcalls, reusing the stack and pads (locals and args).
13:36 rindolf joined
timotimo that's good 13:36
just yesterday i heard someone say "well, everybody complains about the GIL, but i think it really only costs like 10% performance, because when you're using all cores, it'll be faster, obviously!" 13:38
13:38 RabidGravy joined
azawawi RabidGravy: boom :) 13:39
13:39 Khisanth joined 13:40 rurban joined, rurban1 left
dalek k-simple: 648855f | azawawi++ | / (3 files):
More places sidebar properties
 13:48
k-simple: 3b2c302 | azawawi++ | lib/GTK/Simple/Raw.pm6:
Tidy code a bit and fix copy & paste

Add LinkButton widget
13:48 dalek left 13:49 pmurias_ joined, dalek joined, ChanServ sets mode: +v dalek 13:51 brrt joined 13:52 pmurias left
ilmari how does "unit class Foo" differ from "class Foo"? I can't find anything on docs.perl6.org 13:57
psch m: class Foo;
camelia rakudo-moar 2bd421: OUTPUT«5===SORRY!5=== Error while compiling /tmp/kyhpVNrXb4␤Semicolon form of 'class' without 'unit' is illegal. You probably want to use 'unit class'␤at /tmp/kyhpVNrXb4:1␤------> 3class Foo;7⏏5<EOL>␤»
psch that's the only difference i know of
ilmari m: unit class Foo { } 13:58
camelia rakudo-moar 2bd421: OUTPUT«5===SORRY!5=== Error while compiling /tmp/ykRHMpQ5rR␤Cannot use 'unit' with block form of class declaration␤at /tmp/ykRHMpQ5rR:1␤------> 3unit class Foo7⏏5 { }␤ expecting any of:␤ generic role␤»
timotimo what about "sub EXPORT must be in unit scope"?
i think that means if you have a unit class Foo, it'll go inside the class (by necessity) and if you have a class Foo { ... } it'll go outside
ilmari but where is "unit" documented?
llfourn probably isn't tbh
timotimo doc.perl6.org/language/modules
oh 13:59
not documented
but used
ilmari yeah
14:00 zakharyas left
timotimo so seems like the docs want to be written about this 14:00
14:00 zakharyas joined
llfourn there a number of package topics NYD 14:02
14:03 araujo_ left
perlpilot still isn't sure about the necessity of "unit" 14:03
14:04 _mg_ left
perlpilot at least I don't remember why we went from "class Foo;" to "unit class Foo;" 14:04
14:04 zakharyas left
azawawi because 'unit class' is cooler? :) 14:05
psch perlpilot: the commit (14e680733) alludes to one-pass parsing
perlpilot If it was to remove the ambiguity of what "class Foo;" means, that's good, but I don't quite see how that jibes with the NYI scope declarator
geekosaur forward declarations
perlpilot m: unit $a;
camelia rakudo-moar 2bd421: OUTPUT«5===SORRY!5=== Error while compiling /tmp/r4fdmXaX7P␤unit scoped variables not yet implemented. Sorry. ␤at /tmp/r4fdmXaX7P:1␤------> 3unit $a7⏏5;␤ expecting any of:␤ constraint␤»
ilmari thought it was something like scala's "sealed trait", i.e. can't be extended outside the compilation unit it's declared in
psch notably the "A semicolon [...] is too late" bit 14:06
timotimo we should add "no return;" so that people can ask "what's the point of 'no return;'?" 14:09
14:11 tharkun left
tadzik hah! 14:11
I convinced $employer to have a thing ported to perl 6 while I do the refactoring of it :D 14:12
perlpilot tadzik++
tadzik praises nine++ for Inline::Perl5
14:13 tharkun joined
perlpilot indeed, Inline::Perl5 (and the other work nine has done) are the best shot at "bootstrapping" Perl 6 into a Perl 5 world. 14:13
14:13 kalkin- left, kalkin-_ joined, ptolemarch joined 14:14 Ven left 14:16 araujo joined, araujo left 14:26 brrt left 14:35 donaldh joined 14:37 rurban left 14:38 khw joined 14:40 dj_goku joined
gregf_ m: sub upcase($i,$c) { $i %2 == 0 ?? $c.uc !! $c; }; say "greetings perl6".comb.kv.map(&upcase).join 14:41
camelia rakudo-moar 2bd421: OUTPUT«GrEeTiNgS PeRl6␤»
14:44 rgrinberg left, rgrinberg joined
Xliff Eesh! Reading the GTK docs is hard. 14:45
timotimo yes, it is :(
14:45 pat_js left 14:46 palad joined 14:49 pyrimidine joined
Xliff deprecated 14:49
oops
timotimo, also incomplete. But then again GTK is huge.
14:54 noninc joined, huggable left, huggable joined, brrt joined 15:02 rgrinberg left
azawawi Xliff: github.com/perl6/gtk-simple/tree/m...r/examples :) 15:02
15:02 rgrinberg joined 15:04 rgrinberg left
azawawi tadzik: ping 15:04
15:04 rgrinberg joined
Xliff azawawi: What's that? Current progress? 15:04
azawawi Xliff: Perl 6 examples that you can use 15:05
Xliff Ah...
Already forked. ;)
azawawi noooo lol
:)
timotimo NO LOL | NO PROJECT EULER 15:06
azawawi .tell tadzik Please take a look at github.com/tadzik/panda/issues/316
yoleaux azawawi: I'll pass your message to tadzik.
azawawi timotimo: :)
~ 1 hr 45 min till breaking fast... low on sugar :) 15:07
Xliff azawawi: Only forked coz....pull requests. :P
15:07 wamba left
Xliff Assuming my ambitions aren't more ambitions than me. 15:08
azawawi i hate that 3055 error travis-ci.org/perl6/gtk-simple/job...7602#L1197 15:10
with passion
15:13 ssotka joined
azawawi starts working on developer.gnome.org/gtk3/stable/Gt...utton.html 15:16
15:17 rgrinberg left 15:19 aries_liuxueyang left 15:20 mr-foobar left 15:21 aries_liuxueyang joined, xinming_ left, xinming joined, araujo joined 15:24 Actualeyes joined 15:25 rurban joined 15:28 rurban_ joined 15:32 rgrinberg joined 15:40 tharkun left, tharkun joined
azawawi pasteboard.co/1x5h09il.png # GTK::Simple::LinkButton :) 15:48
15:52 raiph left, raiph joined
Juerd gist.github.com/Juerd/b6531810922d...50be4915ca # b2gills suggests using .^name instead of .WHAT, but I've only ever seen .WHAT in examples as far as I can remember. Any opinions about which is best for example codes gear toward beginning Perl 6 programmers? 15:55
s/gear/geared/
b2gills The main thing is .WHAT on an undefined object is a no-op, whereas .^name returns a Str 15:57
Juerd b2gills: What do you mean? "say Str.WHAT" does say "(Str)" 15:59
b2gills Str.WHAT is exactly the same as Str
Juerd Why is that a problem?
b2gills m: say Str; say Str.WHAT
camelia rakudo-moar 2bd421: OUTPUT«(Str)␤(Str)␤»
Juerd It's like "A".uc, I think 16:00
retupmoca .WHAT returns an object and .^name returns a string, IIRC
m: say Int.WHAT.WHAT; say Str.^name.WHAT; 16:01
camelia rakudo-moar 2bd421: OUTPUT«(Int)␤(Str)␤»
b2gills I just find it annoying
Juerd b2gills: I'm more sensitive to technical arguments than emotional ones :) 16:02
retupmoca m: say Int.WHAT.WHAT; say Int.^name.WHAT; # both Int makes it more clear
camelia rakudo-moar 2bd421: OUTPUT«(Int)␤(Str)␤»
16:02 palad_ joined
b2gills and in example code, it makes new Perl 6 programmers think it does something other than what it actually does 16:02
Juerd What do you reckon they will think it does?
b2gills Return the name
16:02 palad left
Juerd But if you use .^name in example code, they might be more inclined to do $foo.^name eq 'Str' than $foo.WHAT ~~ Str 16:03
I don't have a smart match like that in there yet, but I intend to 16:04
b2gills and they may think that ｢'the type is ' ~ 42.WHAT｣ might work ( it doesn't )
They should do $foo ~~ Str
Juerd No, but then they run into a very helpful error message...
b2gills: Yes, they should. Hm.
b2gills or $foo.WHAT =:= Str to make sure it is of the same type 16:05
Juerd I will not use =:= in any example code if I can avoid it at all.
b2gills nor should you for most purposes
literal wasn't there some shorthand for <[0..9 a..z A..Z]> ? 16:06
err. I mean a..f A..F
16:06 palad_ left
azawawi food & :) 16:07
16:08 azawawi left, raiph left
ilmari don't perl6 regexes have named posix character classes, e.g. [:xdigit:]? 16:08
16:09 raiph joined, rurban left
b2gills m: say 'aF' ~~ /<xdigit>+/ 16:10
camelia rakudo-moar 2bd421: OUTPUT«｢aF｣␤ xdigit => ｢a｣␤ xdigit => ｢F｣␤»
literal ah ok
ilmari m: say 'af' ~~ /<:Hex_Digit>/ 16:11
camelia rakudo-moar 2bd421: OUTPUT«｢a｣␤»
ilmari m: say 'af' ~~ /<:Hex_Digit>+/
camelia rakudo-moar 2bd421: OUTPUT«｢af｣␤»
ilmari m: say 'axf' ~~ /<:Hex_Digit>+/
camelia rakudo-moar 2bd421: OUTPUT«｢a｣␤»
ilmari (that's the unicode property name)
16:12 raiph left 16:13 raiph joined, pullphinger joined
b2gills m: ((0.. 0x10FFFF).grep: *.uniprop: 'Hex_Digit').map: *.chr.print 16:15
camelia rakudo-moar 2bd421: OUTPUT«0123456789ABCDEFabcdef０１２３４５６７８９ＡＢＣＤＥＦａｂｃｄｅｆ»
k-simple: da08826 | azawawi++ | / (6 files):
Merge pull request #50 from azawawi/master

Add LinkButton widget
16:16 sno left
tony-o where did $*OS move to? 16:16
b2gills m: ((0.. 0x10FFFF).grep: {.chr ~~ /<xdigit>/}).map: *.chr.print
camelia rakudo-moar 2bd421: OUTPUT«0123456789ABCDEFabcdef» 16:17
tony-o ah, brain fart
16:18 _mg_ joined 16:19 azawawi joined
azawawi m: say $*DISTRO.perl 16:19
camelia rakudo-moar 2bd421: OUTPUT«Distro.new(release => "13.2", is-win => Bool::False, path-sep => ":", name => "opensuse", auth => "opensuse.org/", version => v13.2.Harlequin, signature => Blob, desc => "2016-06-08T18:19:46.210396+02:00")␤»
azawawi tony-o: $*OS -> $*DISTRO # if i remember correctly 16:20
tony-o yes sir
16:21 azawawi left
sjn should using $*OS give a warning? 16:21
tony-o it's been that way for a long time :) i'm working on doing automated test reports when you install modules with `zef` - does panda have a plugin type system i can write something like that for?
16:30 raiph left, raiph joined 16:35 raiph left 16:36 raiph joined
gregf_ dumb question: is it possible to have abstract methods in a role that force the consumer to implement them? 16:43
nine gregf_: not dumb at all and yes it's possible
gregf_: just declare them like this: method foo() { ... }
Xliff builds rakudo, again. 16:44
sjn really likes the … method 16:45
m: …;
camelia rakudo-moar 2bd421: OUTPUT«Stub code executed␤ in block <unit> at /tmp/zcXsP_Bi8x line 1␤␤»
jdv79 nine: just give me a hint and i'll investigate (github.com/niner/Inline-Perl5/issues/63)
yoleaux 06:21Z <nine> jdv79: do you happen to use Perl 5.24?
gregf_ nine: oh, so like, role PrefixBuilder { method buildPrefix(){ die "TBI"; }; } ; class DatePrefixBuilderImpl does PrefixBuilder { method buildPrefix(){} }; class StringPrefixBuilder does PrefixBuilder { method buildPrefix(){} }?
jdv79 yes. yes i do. 16:46
sjn gregf_: you can actually use the three dots (...) or it's utf8-version …
16:46 donaldh left
sjn they even give a nice message 16:46
"Stub code executed" 16:47
gregf_ m: (...).^name
camelia rakudo-moar 2bd421: OUTPUT«Stub code executed␤ in block <unit> at /tmp/cgYxW8tfpu line 1␤␤»
16:47 brrt left, dakkar left
gregf_ hmm, i was of the impression that ... was for ranges/lists 16:48
sjn it is
gregf_ ah - ok, that was what nine meant ;)
sjn but if you write it where a term is expected, you get the "not yet implemented" function
gregf_ rather than a die use ... #brill ;) 16:49
sjn yeah, it throws an exception too
gregf_ nine++ sjn++
16:50 Ven joined
sjn wonders how to introspect that method 16:50
16:52 pmurias joined 16:56 flaviusb joined
TimToady a ... in a stub will also suppress any redefinition warnings 16:59
(on subs or classes) 17:00
17:03 stevieb left 17:04 rgrinberg left 17:09 molaf joined
jdv79 nine: i believe i encounter a similar error with 5.22.2 17:09
17:10 pmurias left, pmurias_ left
geekosaur m: &term:<...>.^name.say 17:10
camelia rakudo-moar 2bd421: OUTPUT«5===SORRY!5=== Error while compiling /tmp/78YI5W5d66␤Undeclared routine:␤ term:<...> used at line 1. Did you mean 'term:<now>'?␤␤»
geekosaur hm
sjn that was weird 17:13
nine jdv79: I asked because there's a github issue with similar error messages from a 5.24 user 17:14
jdv79: I guess, I should just perlbrew 5.24 and test here 17:15
b2gills Perl 5 copied part of the ... feature ｢perl -MO=Deparse -e '...'｣ => ｢die 'Unimplemented';｣
nine jdv79: it's just that on my desktop I don't even have a working rakudo currently as I'm still working on unbreaking perl6-j
17:16 firstdayonthejob joined
[Coke] if $*OS moved before christmas... nope 17:16
b2gills m: $^O 17:18
camelia rakudo-moar 2bd421: OUTPUT«5===SORRY!5=== Error while compiling /tmp/QBLJBPlGw6␤Unsupported use of $^O variable; in Perl 6 please use $?DISTRO.name or $*DISTRO.name␤at /tmp/QBLJBPlGw6:1␤------> 3$^O7⏏5<EOL>␤»
17:20 sno joined 17:21 keix left
sjn huh, I get a completely different warning 17:21
m: say $*PERL.perl 17:22
camelia rakudo-moar 2bd421: OUTPUT«Perl.new(compiler => Compiler.new(id => "73CD636A1ABD7B788919C77C797DC2E21F4C48B0.1465392618.02908", release => "", codename => "", name => "rakudo", auth => "The Perl Foundation", version => v2016.05.68.g.2.bd.4211, signature => Blob, desc => Str), name =…»
sjn rebuilds moar 17:23
17:23 _mg_ left, keix joined
ugexe 5.24 worked ok for Inline:Perl5 on my VM, and 5.22 had the same error on jdv79s VM when I had access 17:23
17:24 firstdayonthejob left 17:25 firstdayonthejob joined 17:31 Urchin joined 17:32 FROGGS joined 17:34 zakharyas joined
FROGGS o/ 17:35
yoleaux 03:16Z <sortiz> FROGGS: Sure, test-driven is the better approach.
08:19Z <Xliff> FROGGS: If you need more help with XML::LibXML test cases, let me know!
lizmat FROGGS o/
FROGGS Xliff: sure, if you want to hack on it... let me know your github-ID and I'll welcome you :o) 17:36
hi lizmat
lizmat FROGGS: long time no see :-)
FROGGS aye :o) 17:39
17:39 Actualeyes left
FROGGS will attempt to join every evening again 17:39
lizmat ++FROGGS :-)
17:41 rgrinberg joined
masak FROGGS: yes, please do! 17:41
17:42 rgrinberg left, rgrinberg joined
moritz \o 17:45
masak if I had a nickel for every "missing or wrong dependency" 17:46
17:46 rgrinberg left
masak yes, compiler, I know you're ===SORRY!===... 17:46
17:46 rgrinberg joined 17:47 TreyHarris joined 17:48 llfourn left
FROGGS masak: but at least I filed that issue just the other day :o) 17:48
17:49 rgrinberg left, rgrinberg joined
masak FROGGS++ 17:50
17:50 TreyHarr1 left 17:51 llfourn joined 17:53 rgrinberg left
masak chalk this one up as a first: fix bug, write regression test for bug, see it pass, interactive rebase to put the commits in the reverse order, `edit` after the test, re-run to see red, rebase --continue, rebase -i again, squash the two commits together 17:53
:)
17:53 rgrinberg joined
masak man, I love Git 17:53
I do not love that precomp bug. fifth time I get it now in the past 15 minutes
apparently it bites you if your modules `use` each other :/ 17:54
17:54 buharin joined
buharin hi 17:54
:)
mst, hello
perlpilot buharin: greetings.
mst buharin: yes? 17:55
[Coke] masak: you have circular deps?
buharin mst, don't ban me please
masak [Coke]: heavens, no
mst buharin: you had a 24h temporary quiet
masak ...that's not what I meant
mst buharin: I told you it was only 24h
buharin: I lifted it after 24h
buharin ok
mst buharin: what are you even talking about?
buharin thnks
masak I meant "if you have modules which use other modules which you have"
17:55 rgrinberg left
perlpilot buharin: Is that why you greet mst specifically each time you join here? 17:55
buharin yes
mst well, stop it 17:56
17:56 rgrinberg joined, zakharyas left
[Coke] If you don't wish to be tempbanned, don't act in a way which requires it 17:56
… crud, I have a meeting in 3m. 17:57
17:58 timeless joined
mst buharin: look. if I tempban you, once the tempban is up, it's done. 17:59
buharin: just don't do the same thing again, and we're fine
buharin ok, I am sorry
I am good
mst buharin: making a big deal of it is not necessary and bloody annoying
right. so let's move on.
timeless hi, do people take feedback on doc.perl6.org/language/5to6-nutshell here? 18:00
18:00 donaldh joined
perlpilot timeless: the best feedback comes in the form of a pull request :) 18:00
timeless the syntax highlighting on this line is *odd*: return scalar(<H>), scalar(<H>);
perlpilot: that's great. sometimes i'll do them, sometimes i won't
today i'm offering bug reporting as opposed to fixes, since i don't know the language 18:01
perlpilot timeless: yeah, the highlighting there is weird.
timeless also here: my $href = { A => 98, Q => 99 }; # Both Perl 5&6 [*See Note*]
i don't understand the coloring in doc.perl6.org/language/5to6-nutshe..._and_!~~_.
also, sometimes `See S0x/....` is a link, as in `See S05/Substitution.` 18:02
timotimo the highlighter is derping around, i'd guess
timeless but often, it isn't, as in `See S32/Containers`
(i suppose the highlighter could want a trailing period, but i haven't finished reading, so i haven't built a mental model) 18:03
In Perl 6 regexen, | does LTM, which decides which alternation wins an ambiguous match based off of a set of rules, rather than about which was written first. 18:06
the `about` doesn't make sense there
18:06 muraiki joined
timeless > no warnings is currently NYI, but putting things in a quietly {} block will silence. 18:06
timeless has no clue what NYI means
lizmat Not Yet Implemented
S99:NYI
synopsebot6 Link: design.perl6.org/S99.html#NYI
timeless would it hurt for that to be linkified? 18:07
18:08 rgrinberg left
Xliff FROGGS: Here's my github profile -- github.com/Xliff 18:11
timeless wonders why everything in this page has a `link to top of documetn`
surely web browsers let users do that...
Xliff So I think my github-id is the same as my IRC nick.
timeless: Some IRC clients do it for you.... 18:12
Xliff hides.
timeless Xliff: eh?
Xliff: load doc.perl6.org/language/5to6-nutshell#-s
Xliff You said something about linkifying.
timeless hover over `-s`
why would you want clicking on `-s` to take you to doc.perl6.org/language/5to6-nutshell#___top
Xliff Ooh! I see now. 18:13
timeless doc.perl6.org/language/5to6-nutshell#Operators
Xliff But where else should it take you?
timeless scroll up one line
"See S32/Containers"
that should be a link, but no irc client will make it one...
nemo timeless: well actually my MATE terminal included -s in the link ☺ 18:14
timeless Xliff: having a larger anchor for me to click, or, having the anchor available consistently to the left of the topic would be nice -- for sharing
Xliff Would need to add that format to the bots.
nemo timeless: it didn't do so well with the link that ended in a . though
Xliff S32:Containers 18:15
synopsebot6 Link: design.perl6.org/S32.html#Containers
Xliff ^^ or that
nemo doc.perl6.org/language/5to6-nutshe..._and_!~~_. ← this link
Xliff Unfortunately, that link doesn't exist
timeless Xliff: ok, nice bot
doc.perl6.org/language/5to6-nutshell#-s
is it really the convention to mix file content w/ shell commands w/ shell output? 18:16
geekosaur timeless, that's a common (but not universal) convention: in the table of contents a link takes you to the section, and the corresponding link in the section takes you back to the table of contents
some wikis do this automatically as well (trac, at least)
timeless geekosaur: i personally hate trac, but i'm not going to get into that
Xliff .tell FROGGS Here's my github profile -- github.com/Xliff 18:17
yoleaux Xliff: I'll pass your message to FROGGS.
timeless if you can accept the pieces that can be fixed and overlook places where i'm whining, i'd appreciate it :)
> Removed. See S19.
missing link
S19
synopsebot6: S19
...
Xliff S19:top 18:18
synopsebot6 Link: design.perl6.org/S19.html#top
18:18 sherwool joined
Xliff Need the : 18:18
timeless S19:
S19:_
synopsebot6 Link: design.perl6.org/S19.html#_
timeless um, ok...
anyway, that doesn't tell me what it is, i'd have to click the link
18:18 tomboy64 joined
Xliff Someone would have to update synopsebot6 with the titles. 18:19
geekosaur spitting possibly large chunks of text into irc is not generally a good plan...
FROGGS Xliff: done
yoleaux 18:17Z <Xliff> FROGGS: Here's my github profile -- github.com/Xliff
Xliff FROGGS, Cool!
timeless geekosaur: titles shouldn't be long =b 18:20
FROGGS Xliff: I hope you get the idea of how the code is structured
geekosaur one would hope. but give someone something to abuse and someone will abuse it :(
Xliff geekosaur, just a title for the associated S article and the link.
Limited abuse potential. 18:21
FROGGS: OK. Got a link for the repo?
FROGGS Xliff: github.com/FROGGS/p6-XML-LibXML 18:22
timeless > (Please note that not having the module statement isn't consistent with Synopsis 11, however it works.) 18:24
that should have a link
18:26 tbrowder joined
Xliff FROGGS: Cool! I'll try to work on some of these tonight. 18:26
18:27 Urchin left 18:28 telex left
timeless doc.perl6.org/language/5to6-nutshell#-s 18:29
Switch parsing is now done by the parameter list of the MAIN subroutine.
doc.perl6.org/language/5to6-nutshe...topt::Long
Switch parsing is now done by the parameter list of the MAIN subroutine.
would it make sense to have `-s` link to the other thing?
18:30 telex joined 18:31 tbrowder left, donaldh left
timeless reaches the end of the document 18:31
nemo timeless: and are now a perl6 expert 18:32
timeless goes back to um.. something else
(C apparently) 18:33
nemo timeless: ah. just in time, since you're doing perl5 → perl6 time to go C → Rust ☺
timeless no thanks
this isn't my project 18:34
(it's tmux)
nemo ah
18:34 donaldh joined 18:35 labster joined
FROGGS Xliff++ 18:36
damn 18:40
I believe I removed my login from the sudoers group -.-
18:44 FROGGS left, tbrowder joined
tbrowder hi, perl people 18:44
anyone here going to YAPC::NA::2016 besides lizmat, DrForr, timtoady, damian, and util (and moi)? 18:46
18:46 noninc left, FROGGS joined
Xliff Where in NA will the YAPC be? 18:47
timeless perlpilot: anyway, were my notes helpful?
tbrowder Orlando, Florida, home of Disney World and such 18:48
18:48 wamba joined 18:49 Ven left
tbrowder I hope lizmat can spare some time to talk about debugging P6 and working with the grammar 18:50
lizmat debugging yes, working with grammars is definitely not my forte
tbrowder debugging is good! 18:51
perlpilot tbrowder: I'll be at YAPC::NA
tbrowder good! 18:52
it helps to put a face to irc, and maybe chat over a beverage of choice 18:53
perlpilot timeless: maybe stick them in an issue on github? I don't have time right now to do anything about them. 18:54
18:54 huggable left, huggable joined
timeless regrets not having claimed `timeless` on github :/ 18:54
18:54 hankache joined 18:55 FROGGS left
timeless perlpilot: is the pod really the original source? it feels like it might not be... 18:56
perlpilot tbrowder: you can see a 10 year old picture of me on my github profile if you want to put a face to a nick. I've lost some hair since then, but it's all under the hat in that picture anyway. I basically look the same though :)
hankache good evening #perl6 18:57
lizmat hankache o/
hankache: so what would I need to create a Dutch version of perl6intro ?
*do
hankache hiya lizmat
you need to know the syntax (asciidoc). 18:58
18:58 win32 joined
win32 Hi here =D 18:58
perlpilot timeless: aye, it is. (at least last time I looked, the html was being built from the pod. someone could have changed it when I wasn't paying attentiong :-)
lizmat hankache: but that would be pretty clear from the source, no? 18:59
perlpilot s/ong/on/
18:59 donaldh left
hankache lizmat yes 18:59
hello win32
perlpilot win32: greetings
hankache lizmat github.com/hankache/perl6intro
lizmat hankache: I was more thinking along: clone the english version, and then what ?
win32 hankache> , perlpilot thnx =D
perlpilot oops, I'm late for a meeting
win32 what upz .
timeless perlpilot: ok, and presumably the steps for making the html are easy? (and don't require perl6)?
if so, i might send a pr for /something/
literal enum die_faces ("⚀".."⚅"); say die_faces.enums 19:00
m: enum die_faces ("⚀".."⚅"); say die_faces.enums
win32 who have for spend a short time for me =D
camelia rakudo-moar 30b33e: OUTPUT«WARNINGS for /tmp/E2AKgn89kz:␤Useless use of ".." in expression "\"⚀\"..\"⚅\"" in sink context (line 1)␤Map.new(("⚀" => 0,"⚁" => 1,"⚂" => 2,"⚃" => 3,"⚄" => 4,"⚅" => 5))␤»
lizmat hankache: ah, RTFM :-)
hankache: will do!
literal how can I do this without the warning?
hankache lizmat create a new file nl.perl6intro.adoc
lizmat ack, gotcha
hankache: should I do a PR or do you want to give me a commit bit? 19:01
hankache lizmat: i'll give you commit
TimToady ooh, a useless use of useless use bug :)
hankache hiya TimToady 19:02
win32 i wanna to get a code read a file Answer questions , questions its format > NY |New york end msg to # channel.
who can help me! thanks u all
lizmat hankache: cool
19:03 labster left
TimToady m: constant tmp = "⚀".."⚅"; enum die_faces (tmp.list); say die_faces.enums # one workaround 19:04
camelia rakudo-moar 30b33e: OUTPUT«Map.new(("⚀" => 0,"⚁" => 1,"⚂" => 2,"⚃" => 3,"⚄" => 4,"⚅" => 5))␤»
literal difference being that now they're numbered 0 to 5 :) 19:05
timeless ok, stupid question...
literal oh wait, so was my solution
hankache lizmat: I added you as collaborator.
timeless are the t/*.t files in perl6/doc supposd to pass?
lizmat hankache++
timeless s/supposd/supposed/ :(
19:06 tbrowder left 19:07 FROGGS joined 19:08 TEttinger joined
timeless also, anyone here familiar w/ the trig stuff? 19:08
timeless is reading doc/Type/Cool.pod and is rather puzzled
timeless hasn't really played w/ trig in nearly 2 decades ...
timeless decides it's just wrong and moves on 19:09
masak heh, if you read Cool.pod and your reaction was "trig", then maybe the focus of that document is a bit skewed... :) 19:10
TimToady literal: testing a fix for the bogus warning 19:11
masak timeless: I'm familiar with the trig stuff. heck, you could probably throw me a cosecant and I'd know what to do with it ;)
literal TimToady: nice
timeless masak: here's the text from the pod. www.irccloud.com/pastebin/9sDnbaqM/
you mentioned `cosecant` which is something i recognize
am i right to believe that the pod just doesn't know how to spell cosecant? 19:12
19:12 yqt joined
masak timeless: yes, you are 19:13
timeless awesome! thanks
19:13 FROGGS left
timeless will arrange for a pr eventually 19:13
sorry, i was worried my trig knowledge was rotting ...
19:15 FROGGS joined
timeless masak: funnier one 19:15
en.wikipedia.org/wiki/Trigonometri...#Etymology
is `cotangens ` there intentional?
lizmat sounds like a Dutchism 19:16
timeless goes to read OED
interesting, apparently it was once spelled that way in hal.inria.fr/inria-00543938/document
19:17 darutoko left
timeless i think this is perhaps old-french 19:17
masak it'd be "cotangens" in Swedish too
timeless Descriptio Canonis.
masak oddly enough, that didn't even cross my mind as I read it in the paste and crossed it off as being Wrong in English :) 19:18
timeless shrugs
I didn't specialize in romance languages
for all i know this is actually Latin
timeless isn't used to seeing æ 19:19
anyway, i'm going to suggest enforcing the English spelling, since the document is in English and it references the English wikipedia which almost* exclusively uses the English spelling...
19:19 rurban_ left 19:21 Ven joined
[Coke] timeless: yes, pleas. 19:21
*e
timeless btw, do you prefer en-us or en-gb? 19:22
for a large project, i'm tempted not to enforce since the last thing i need are complaints over minor things
[Coke] in general, we lean towards -us
timeless but i can enforce
[Coke] but I don't think we'd mind either.
mst *I* prefer en_GB. However, en_US is probably the correct answer.
having spent a year or so using QBASIC during the day and BBC BASIC at night
I am already quite sufficiently traumatised by varying spellings of COLOUR 19:23
timeless heh
win32 i wanna to get a code read a file Answer questions , questions its format > NY |New york end msg to # channel. ^_^ 19:24
hankache What do you mean en_us? There's only one English and that is her Majesty's ;)
timeless hankache: having read en-in amongst others, i must disagree
and i blame her Majesty's relatives for that 19:25
=b
b2gills Just imagine that the en in en_us stands for something other than the word english
masak yes, en_US is the correct answer. it's always the correct answer, unless your shop is so overwhelmingly British-speaking that it'd be silly to do otherwise *and* you don't expect contributions/patches from the open-source world.
hankache :p
huf_ there are more americans. they're easier to hit. i mean, the target is easier to hit. 19:26
i mean... oh dear
mst huf_: except when you give americans weapons they usually manage to shoot the nearest british soldier instead
b2gills .oO( many of them are overweight )
hankache huf_ haha
huf_ mst: cos you eejits keep going near them 19:27
mst how to avoid being hit by the US army: (1) stay away from the british forces (2) hide in a urinal
timeless fwiw, the doc's spelling error ratio is actually quite tolerable
timeless is in the h's, whereas one project hit 200
huf_ mst: or really, just 1) adopt rincewind tactics
hankache is it bad if you mix both en_us and en_gb in your writing? English being my 3rd language 19:30
timeless hankache: yes
for various reasons
1. it means you don't know how to use a spell checker
(i.e. it just makes you look bad)
2. it causes cognitive dissonance for anyone reading it 19:31
mst c. it brutalises people and makes them turn a funny color
timeless basically, your word processor and browser both have spell checkers, so there's very little excuse not to let one or the other help you out
timeless is using irccloud, so the browser is the spell checker here
huf_ i dont think i'd notice it :)
mst I turn all spell checkers off
and my word processor is ex-vi, which doesn't really have one 19:32
timeless mst: for this exercise, my primary spell checker is /usr/share/dict/words 19:33
hankache timeless this is not a spell checker issue. For example I like to write color instead of colour for a reason that i ignore. It feels more natural. But for other words it may be the opposite
19:33 labster joined
timeless the secondary is my brain, the the tertiary is google-search-suggest 19:33
mst I agree with proof reading
I agree with spelling tests 19:34
I do not agree with the 'very little excuse not to use a spell checker while writing' part
timeless ah, fair enough
as long as what you publish has been checked, i don't care if you check-as-you-go or check-before-you-post
the point is that you have the ability to get-it-checked for minimal cost
mst but generally I proofread and then publish 19:35
19:35 donaldh joined
huf_ hankache: yeah, i have mixed spellings too :) 19:35
mst since I'm as worried about 'wrong valid word' problems as speling erors
huf_ a base of american with weird spots of uk
19:35 ufobat left
hankache huf_ exactly 19:35
timeless doc/Language/quoting.pod:=head2 X<Heredocs: :to|qoute,heredocs :to> 19:37
please say that `qoute` should just be `quote` and that i'm not missing some perl6 knowledge...
19:38 grondilu joined 19:39 cdg left
[Coke] almost certainly quote 19:39
timeless: github.com/rakudo/rakudo/search?ut...mp;q=qoute 19:40
timeless thx
mst hankache: if that's for IRC or for your blog, I, personally, give no fucks
[Coke] (so, not some weird hidden thing)
mst thx OTOH is a sign of illiteracy in all languages in my head
timeless [Coke]: shortcircuits -- would you write that as one word or two? 19:41
(and would you accept the alternate)
mst (pet hate)
hankache: if it's for an OSS project, whatever spelling they're already using is preferred
of course, I sometimes write method names in en_US and docs in en_UK
timeless mst: tmux had stuff like that 19:42
it's a minor pita when trying to write fixes
but, i understand it 19:43
and don't object to it in general
19:43 buharin left
timeless (it just makes the task i'm doing now slower) 19:43
mst timeless: if you find a project of mine like that, a patch to USify all the spelling with a test to make sure I don't backslide wouldn't upset me, but I doubt I'd be bothered doing so myself 19:44
timeless :-)
doc/Language/glossary.pod:=head2 Thnunk 19:45
um.... i know about thunks, ... but thnunks?
timeless decides it's just a typo
mst would think so, yes 19:46
masak it's like thunks, but with nunchucks
japhb Those would make quite a thunk when they hit you, I would think ... 19:47
timeless doc/Language/testing.pod:# tests requiring functional wombling 19:48
is wombling a technical term? (it wasn't in the glossary, and i don't know it)
TimToady fix pushed for literal++
timeless (note: i don't expect to change it, it's just odd)
only 23 changes (not counting checking for en-gb) 19:50
not bad
hankache m: { our $var = 'Text'; say $var.WHERE; }; { our $var = 123; say $var.WHERE; }; say ::('$var').WHERE; 19:51
camelia rakudo-moar 965ca6: OUTPUT«139708347882696␤139708348067648␤139708348067648␤»
hankache should this be allowed? ^^
stmuk_ wombling probably means tidying 19:54
19:56 brrt joined
masak hankache: yes. that's correct. 19:57
m: { our $var = "OH HAI" }; { our $var; say $var }
camelia rakudo-moar 965ca6: OUTPUT«OH HAI␤»
timeless masak: hrm, `rubbish` is en-gb 19:58
masak the value gets shared through the underlying package scope, but it's two distinct variables.
timeless stmuk_: i'm trying to read doc/Language/testing.pod and Womble+Wombling are used a lot
but i can't figure out what it is
because of the rubbish, i'm somewhat confident that womble is also en-gb 19:59
class GreatUncleBulgaria is Womble {}
19:59 rindolf left
hankache masak thanks 20:00
mst o/ Underground, overground, wombling free / The Wombles of Wimbledon Common are we o/
timeless mst: see, that's *definitely* not in my corpus
class Tobermory is Womble does Invent {} 20:01
mst timeless: see enwp.org/Wombles
timeless is Tobermory also en-gb?
mst that's a name of a character
en.wikipedia.org/wiki/The_Wombles#Known_Wombles 20:02
timeless mst: thx
20:02 buharin joined
mst timeless: AAAAAAAAAAAAAAAAAAAA 20:02
timeless didn't know about enwp.org either, neat
b2gills hankache: The GC system uses a copying GC for the nursery so short lived objects tend to move around a fair bit even in the same scope
timeless > Although Wombles supposedly live in every country in the world, Beresford's stories are concerned with the lives of the inhabitants of the burrow on Wimbledon Common in London, England. 20:03
mst can live with 'ty', or the en_UK 'ta', but please don't use 'thx' at me, that particular not-even-an-abbreviation is like fingernails down a blackboard to me :P
20:03 win32 left
timeless mst: sorry 20:03
i rarely abbreviate, that's one of my few exceptions, but i can avoid it :)
mst it's ok, I know it's weird, but AAAAAAAAAAAAAAAAAAA
timeless alias as: `:-)`
DrForr . o ( 1138? )
timeless 😃
... yeah, they definitely didn't seem to travel the world 20:04
mst timeless: the think that annoys me is it isn't even a proper abbreviation due to the x
timeless mst: interesting
mst but, yeah, I'm not claiming I'm entirely rational about this
timeless oddly, i'm not a big fan of it either
i wonder when my brain internalized it
mst 'ty' doesn't bug me, 'tks' doesn't bug me, 'ta' doesn't bug me
timeless `interned`
mst but 'thx' makes me imagine a 12 year old typing on a nokia candybar 20:05
timeless `ty` and `tks` bug me
`ta` i could live w/
mst I'm not *fond* of 'ty' or 'tks' but I don't actively hate them like 'thx'
hankache b2gills thanks
timeless mst: did Sesame Street make it across the pond?
mst yes
there's a reason Nicholas Clark announced RC releases when he was pumpking with 'Vun. Ha! Ha! Ha!' etc. 20:06
timeless would it be a reasonable replacement for the Wombles in the .pod?
timeless doesn't know `nicholas clark` or `pumpking`
timeless also hasn't followed perl6
mst pumpking is the release manager for perl5
and general leader of p5p
timeless ah 20:07
20:07 cdg joined
timeless i think i mostly used p5.2..5.6 occasionally 5.8 (?) 20:07
b2gills hankache: If you want more information www.infoq.com/presentations/termin...-collector / jnthn.net/papers/2013-bs-secret-life-of-gc.pdf
mst I started off on 5.005_03
timeless yeah 5.005_* feels very familiar 20:08
mst also, I feel like the wombles have a certain whimsicality that's very in keeping with the perl6 ethos
timeless well
20:08 cdg left
timeless if you want them to stick around, then i think they need to be added to the glossary 20:08
b2gills I didn't even hear about Perl until 5.6 ( I was on DOS and Windows )
20:09 cdg joined
mst timeless: I would suggest a link to the wikipedia page? 20:09
timeless mst: ok, but straight into the pod that's using them?
mst I, personally, don't really see a problem with that, but would not argue if other people felt that was insufficient
literal is there a method or idiom to do this in a single expression? my %counts; %counts{$_}++ for @things; 20:10
b2gills @things.Bag
masak %counts{@things}>>++;
what b2gills said :)
lizmat b2gills++
literal oh right, forgot about bags 20:11
mst but I do think "preserving whimsicality" is something we should value
masak I often forget about bags
lizmat bag @things
masak .oO( I need you to... take care of @things, see? capishe? )
lizmat literal: also note that the keys in a bag are not stringifications, but the actual objects
literal yeah
DrForr bag @your-face; # Not that anyone here remotely qualifies, save myself ) 20:12
s/.$/:)/
timeless frowns 20:13
ok, so um... apparently i need perl6 in order to make the docs 20:14
mst rakudobrew to the rescue?
20:16 colomon left
Xliff ♥ rakudobrew 20:16
timeless Xliff: design.perl6.org/S19.html#Removed_...c_Features 20:17
there's no way for me to get the anchor for that section on that page *grrr* 20:18
20:19 girafe joined
Xliff S19:Removed Syntactic Features 20:21
synopsebot6 Link: design.perl6.org/S19.html#Removed_S...c_Features
20:21 donaldh left
Xliff pokes timeless -- "Werked fer meee!" 20:22
<mst> I am already quite sufficiently traumatised by varying spellings of COLOUR
timeless Xliff: load the link, is there an anchor you can hover over to get the link for that section?
timeless doesn't see one
Xliff hands mst a pint of his favourite beer.
timeless: The link synopsebot6 just posted doesn't work for you? 20:23
timeless wrong problem
say i'm on that page and want to get a link to a subsection w/o going to the toc
Xliff ooooooo 20:25
Xliff finally understands.
timeless :-)
Xliff You'd be SOL. But why would you not use the TOC?
timeless because i'm usually scrolling through a document, finding a problem and wanting to point people to it 20:26
having to jump to the top, try to figure out what section it is...
=> very painful
Xliff Yeah. The layout of those pages would need to be slightly changed to fix that problem.
timeless (not necessarily a problem, i could be trying to point to a solution)
20:27 hankache left
timeless anyway, i'm willing to try to fix the 5to6 page, but i really want someone else to fix the design pages to at least be no more annoying than the 5to6 page 20:27
Xliff I don't know of a good solution. Leave the section headers as they are but have a "→" link placed unobtrusively nearby that is a link to the current section? 20:29
20:29 cdg left
Xliff But then... how do you make that intuitive? 20:29
20:30 AlexDaniel joined
timeless Xliff: in the main thing if you hover over a section it gets an :after() thing 20:33
b2gills timeless: if you right click => inspect on Chrome you can find out the name of a link
timeless b2gills: i'm effectively a retired browser develoepr
s/epr/per/ 20:34
i could do lots of things. i shouldn't have to.
and these two sites should behave similarly, that one makes it easy to get anchors and the other doesn't is disappointing
and please don't ask me to do the work, i've just sent 3prs which is well over my average quota
mst oh, yeah, that's been a thing with POD rendering forever and a day 20:37
mst does ctrl-click, ctrl-tab, right click ToC entry and copy, ctrl-W without thinking now 20:38
but the fact that doesn't annoy me is blatantly stockholm syndrome
timeless mst: at this point, what bothers me is consistency, there's clearly a better engine being used for doc.perl6.org than design.perl6.org
20:39 ocbtec left
timeless appreciates that mst is aware he's suffering from stockholm syndrome :) 20:39
mst I suspect that design.perl6.org is using one of the common POD -> HTML renderers on CPAN for perl5
I'm also now wondering why the fsck POD renderers don't do a sensible thing with the headers and make them links-to-self 20:40
timeless yay!
thanks for feeling my pain 20:41
20:42 molaf left 20:43 kaare_ left
mst huh 20:43
metacpan.org/pod/Safe::Isa#SYNOPSIS
timeless: so, if you mouse over the word SYNOPSIS you can see a little link thing
I wonder if one could port that to design.
stmuk_ the design pages aren't currently maintained in terms of technical content so I don't think tidying up the web design is worth it
timeless mst: doc.perl6.org/language/5to6-nutshell#Whitespace 20:44
if you hover over a header there, you get a :after() section mark
§
mst timeless: aha
timeless personally, the metacpan.org :before() is better
mst timeless: well, at least we've found you half a solution
timeless because it's where I expect it to be!
mst in that now you can be less annoyed and not have to go up to the ToC ;)
timeless anyway, i don't care if you copy the doc version or the metacpan version 20:45
mst oh, I see, that's doc. and not design. sorry
gah
timeless but please don't leave design as is :)
stmuk_: so... the problem is that doc. links to design. 20:46
if you make design. not reachable from the web (doc. + google) then i wouldn't care
but, ...
20:46 brrt left
mst timeless: if you remind me another day, I will try and see if it's a ten minute fix 20:49
if you don't remind me, I will try and remember to look another day anyway but no guarantees 20:50
timeless mst: ok
we'll see, i'm fairly flighty, but i should be able to return at least once
probably when someone accepts a pr or 2
dalek Heuristic branch merge: pushed 23 commits to doc by stmuk 20:58
20:59 zakharyas joined 21:00 buharin left 21:01 rurban joined 21:02 rurban_ joined
stmuk_ I've merged the spelling fixes .. less sure about the rest 21:02
maybe "wombling" should go in the glossary
glossary being the special case maintained bit of design. :) 21:03
21:03 pullphinger left
timeless goes to add NYI to the glossary 21:05
literal what should I do if I want rotor() but with overlapping combinations? so (a b c d) => (a b), (b c), (c d) 21:06
timeless err 21:08
what the heck?
someone told me NYI was in the design docs 21:09
but... i don't see it
retupmoca m: say ('a'..'d').rotor(2 => -1).perl;
camelia rakudo-moar 072a6c: OUTPUT«(("a", "b"), ("b", "c"), ("c", "d")).Seq␤»
AlexDaniel OMG, can we squish such commits next time?
literal retupmoca: oh, I should have read the rotor() docs better
timeless AlexDaniel: fwiw, the reason i don't squish myself is to let people exclude things, but i don't mind if they squish my commits when pulling them in 21:10
AlexDaniel timeless: that totally makes sense, I don't mind that at all
but why was it merged as is… come on, 22 commits!
timeless 22 is nothing :) 21:11
github.com/pwsafe/pwsafe/pull/106
try that :)
timeless tries to figure out the glossary 21:12
there are things in the glossary with no content...
doc.perl6.org/language/glossary#Cool
doc.perl6.org/language/glossary#Enum
AlexDaniel timeless: the idea is that somebody is going to write the content for these… 21:13
or something
timeless ...
[Coke] placeholders. "yes, we realize this is a confusing term" 21:14
timeless i do a couple of gimmes, but, if you're in that deep, i'd be open to contracting to fix the rest
AlexDaniel timeless: well, that's how it always been. Somebody adds an entry to S99, somebody else writes the definition
[Coke] I would argue that if you can search for the thing in docs.perl6.org, then you can remove the placeholder in S99
Hotkeys Is there enum documentation hidden somewhere 21:15
AlexDaniel yeah, nuke S99
Hotkeys Or will I have to look at the specs
AlexDaniel Hotkeys: yeah. It is hidden here: github.com/perl6/doc/issues/87
Hotkeys Ah I see 21:16
AlexDaniel ah, looks like gfldex++ fixed it
timeless AlexDaniel: do you want me to link to the type system entry? 21:17
i can do that i suppose
AlexDaniel timeless: I think that it is better than an empty entry :) 21:18
timeless although, tbh, i'd expect the glossary to have at least a one sentence entry beyond linking to the typesystem thing
AlexDaniel something is better than nothing
timeless ok, lemme see
timeless wonders how the docs link to other files... 21:19
AlexDaniel yay for using a spell checker on the whole thing! timeless++ 21:20
timeless 😃
AlexDaniel Tooolbar! :)
Hilites :D
timeless AlexDaniel: fwiw, the error count for per6 doc was really low, so your contributors are pretty good
(or you just are missing too much content) 21:21
pwsafe was pretty bad
AlexDaniel my contributions huh?
timeless "your contributors" = everyone contributing to perl6 doc 21:22
not "your contributions" :)
AlexDaniel :)
timeless is a very careful wordsmith
AlexDaniel “our” contributions then!
miminize :) 21:24
21:25 donaldh joined
timeless is it ok for me to copy stuff over from design? 21:25
design.perl6.org/S99.html#container 21:26
21:26 _4d47 joined
timeless of course, i can't copy over container 21:26
AlexDaniel timeless: github.com/perl6/doc/commit/1aacfb...fbdb01d36e 21:28
timeless: that's how it was created
timeless AlexDaniel: ok
hrm 21:29
is the pod thing smart enough to handle L<#twigil> v. =head1 Twigil ?
AlexDaniel timeless: if it is on the same page, then yeah. It has to be #Twigil though 21:32
timeless it's #twigil (lowercase) 21:33
AlexDaniel: should i just try to pull up the current pod for s99 and copy over the missing items?
it seems silly for me to manually try to recreate them
i mean, i can be selective about whether or not to pull them over
but, what you have now is ... unfortunate 21:34
lizmat timeless: copying *to* or *from* S99 ?
AlexDaniel timeless: right, and it doesn't work
timeless from
lizmat well, I did some work in that area a while ago
timeless AlexDaniel: ok... i can fix that one
lizmat wouldn't mind seeing S99 go completely
AlexDaniel timeless: so yeah, change it to #Twigil. Also see this one: github.com/perl6/doc/issues/561
timeless (and getting a tool to identify similar ones shouldn't be hard)
21:34 Ven left 21:36 Juerd joined
timeless lizmat: so, lemme find the source for s99 21:36
lizmat it lives in specs
timeless i think merging s99 over ito doc/glossary seems reasonable
lizmat yeah, that was the plan
timeless thx
lizmat timeless++
timeless (py)[timeless@gcc2-power8 perl6]$ hg clone git://github.com/perl6/specs 21:37
timeless waits
21:38 wamba left 21:39 spider-mario joined
AlexDaniel hg huh! 21:40
21:40 raiph left
timeless my toolchain/prefs are hg :) 21:40
the docs are better :-)
tadzik computer, messages 21:41
yoleaux 15:06Z <azawawi> tadzik: Please take a look at github.com/tadzik/panda/issues/316
tadzik hrm
timeless perl -ne 'next unless /=head|L<#.*>/; if (/=head/) {s/(=head\d+) (.*)/$2 $1/; print;next}; print "$1\n" for /L<(#[^>]*)>/' doc/Language/glossary.pod|sort|less 21:42
21:42 raiph joined
timeless AlexDaniel: that's what i'll use to find the mismatches/missings 21:42
AlexDaniel timeless: run it over all files
timeless ok, but for now, glossary, i want to take things baby-steps 21:43
AlexDaniel timeless: did I already mention that I love you? 21:45
lizmat timeless++
good night, #perl6!
timeless good night lizmat
oh brother
-=head Adverbial Pair 21:46
+=head adverbial pair
well, that explains why the links on doc/glossary were bad
someone got cute and capitalized things in the glossary at some point
lizmat runs for cover 21:47
timeless lizmat: if the build system didn't have a link checker, it isn't entirely your fault
tooling should have saved you
AlexDaniel timeless: well yeah, that's why #561 exists. But it does not check for links to the same page, as far as I can see 21:48
timeless: currently the problem is that we have a bunch of broken HTML, so the output of checklink is not as useful 21:49
timeless: but once that is solved it will be much easier to work on broken links :)
timeless nodes 21:51
timeless nods
21:51 mr-foobar joined
timeless ok, i think i'm going to do this differently 21:51
i suppose i should look for a pod parser, but i don't think i will
AlexDaniel timeless: also, if you want to prevent such problems you can add a test
timeless AlexDaniel: yeah, but i don't speak any particular .t languages
timeless hasn't spoken perl .t in >5 years 21:52
dj_goku timeless: so you have to learn then. :D
AlexDaniel dj_goku: shh, better bring the candies. That guy is going to fix our docs 21:53
:)
dj_goku I have been idling for ages sadly. sweet though. 21:54
timeless my pod splitter www.irccloud.com/pastebin/YkqnFHd9/0.pl 21:58
21:58 labster left, labster joined
timeless seems to work :) 21:58
(and no, that's not remotely perl6 friendly, i know, i already read the 5to6 page!) 21:59
22:00 zakharyas left, kid51 joined 22:05 setty1 left 22:09 cdg joined 22:10 girafe left
timeless oops 22:10
timeless needs to be a bit smarter about patching
22:10 skids left 22:14 firstdayonthejob left 22:21 Sgeo_ joined 22:30 raiph left, raiph joined 22:32 ptolemarch left 22:34 RabidGravy left
Xliff timeless++ # coz.....DOCS! 22:37
22:37 rurban1 joined
Xliff And I must crash. 22:37
22:38 rurban left 22:54 huggable left, huggable joined 22:56 donaldh left 23:12 spider-mario left 23:23 rurban_ left 23:30 aries_liuxueyang left 23:36 skids joined
timeless ok, gang 23:36
is it camelia or Camelia ?
see the glossary pod for camelia www.irccloud.com/pastebin/eXo01pjj...ossary.pod 23:37
23:42 rurban joined
timeless sighs 23:42
does the glossary get points for being circular? 23:43
23:45 rurban1 left, cdg left 23:46 cdg joined
AlexDaniel timeless: camelia is a bot 23:48
timeless: and Camelia is a butterfly
timeless AlexDaniel: yes, so, lowercase or uppercase in the glossary?
AlexDaniel timeless: raw.githubusercontent.com/perl6/mu...amelia.txt
both
with different descriptions
timeless ok 23:49
23:50 rurban1 joined 23:51 cdg left 23:54 rurban left
AlexDaniel timeless: it is probably not the most active time of a day here right now, so please don't worry if nobody answers your question :) 23:57
I should've been in the bed 3 hours ago too 23:58
timeless should have had lunch 8 hrs ago
fiddly and friends? 23:59

Powered by Cro and the Raku® Programming Language.

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

Thank you!