»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.
00:00 adu left 00:04 adu joined 00:11 huggable left 00:12 huggable joined 00:13 buharin left 00:17 mantovani left 00:19 mantovani joined 00:26 Celelibi joined 00:28 sufrostico left 00:30 redhands98100 left, redhands98101 joined 00:31 redhands98100 joined
awwaiid Zoffix: definately. I am soon to submit a PR on the Nil/"" behavior 00:32
yoleaux 23 Jun 2016 22:25Z <Zoffix> awwaiid: if you have any time, would you be able to take a look into rt.perl.org/Ticket/Display.html?id=128470 ? It was introduced by the loopy REPL commit, so maybe it'll be easier for you to fix it than the rest of us. Thanks! github.com/rakudo/rakudo/commit/6b...125250a18b
Zoffix awwaiid++
awwaiid Zoffix: reproduced locally, and forcing fallback to nqp-repl shows correct behavior (not exiting but still showing error). Hard part done! 00:34
00:34 redhands98101 left 00:38 rindolf joined 00:45 cdg left, cdg joined 00:50 cdg left, pmurias_ left 01:00 buharin joined 01:14 BenGoldberg joined
Xliff Zoffix: It was a merge from upstream. I've seen CVS make merges more complex than this. 01:17
It's fixed now, however I had to make the same fixes like three times before it took. 01:18
Xliff goes back to grumbling quietly.
tbrowder ref doc and space after sub/method names: please check PR: <github.com/perl6/doc/pull/630> 01:19
awwaiid Zoffix: ahh... it's because the .map is lazy and not really executed until you try to print the result, at which time it raises the exception, which is outside of the catch (which is around the eval, not the print) 01:21
Zoffix \o/
Xliff awwaiid++ # Code Sherlock 01:24
01:24 redhands98100 left 01:29 cognominal joined 01:33 abruanese left 01:40 kid51 left 01:46 ilbot3 left, mantovani left 01:47 ilbot3 joined
Zoffix m: .Int does role { method Str { (self + 2).Str } } for ^1000; say "2 + 1 is {2 + 1}"; say "2 * 0 is {2 * 0}"; 01:54
camelia rakudo-moar 9b579d: OUTPUT«2 + 1 is 15␤2 * 0 is 16␤»
Zoffix m: {{{{{{{{{{ "foo" does role { method say { "not foo".say } }; }}}}}}}}}}; {{{{{{{{{{ "foo".say }}}}}}}}}} 02:00
camelia rakudo-moar 9b579d: OUTPUT«not foo␤»
02:02 BenGoldberg left 02:04 araujo_ left
adu hi all 02:05
02:05 BenGoldberg joined, araujo_ joined
Zoffix m: {{{{{ &say does role { method CALL-ME ($s) { "You wanted to print `$s`, eh? Well... YOU CAN'T! Muahaha".print } } }}}}}; {{{{{ say "Foo" }}}} } 02:05
camelia rakudo-moar 9b579d: OUTPUT«You wanted to print `Foo`, eh? Well... YOU CAN'T! Muahaha»
02:08 ssotka left, BenGoldberg left, BenGoldberg joined
rindolf adu: hi, how are you? 02:12
adu good
rindolf adu: nice
adu just checking in
any new developments in NativeCall or Inline::C? 02:13
02:13 lizmat joined 02:17 _mg_ joined, noganex joined, _mg_ left 02:20 noganex_ left, _mg_ joined 02:21 aborazmeh joined, aborazmeh left, aborazmeh joined, _mg_ left 02:30 mantovani joined 02:35 mantovani left
dalek osystem: c63a3f7 | (Brian Duggan)++ | META.list:
Add Slang::Mosdef to ecosystem

See github.com/bduggan/mosdef		 02:37
osystem: 4b15234 | (Zoffix Znet)++ | META.list:
Merge pull request #220 from bduggan/master

Add Slang::Mosdef to ecosystem
Zoffix Wow, looks like UK will actually leave EU 02:39
timotimo OK, it'll be fun to see the world burn :\
Zoffix The 'Leave' has been leading by 1 point since the start and it's already 50% votes counted.
timotimo it's apparently really easy to rally people behind a few simple lies 02:42
Zoffix The current US election is the prime example of that :)
adu timotimo! 02:43
timotimo ohai
i'm up later than i should be :\ 02:44
adu timotimo: me too
Zoffix Me three :")
timotimo at least i'm working the piano keys a little bit
adu I don't think the UK/EU thing is going to affect much
I think the precious metal mines holding back inventory is going to make the world burn 02:45
Zoffix It's going to be a precedent for others to leave. France is already saying they'll have their own referendum.
timotimo fortunately, the referendum means jack shit
and 70% of deciders are stay, 10% undecided
adu timotimo: what exactly is a decider? 02:47
I think, in the U.S. we call them "officials"
timotimo members of parliament i think? 02:48
"The answer is that technically MPs could block an EU exit - but it would be seen as political suicide to go against the will of the people as expressed in a referendum. The referendum result is not legally binding" 02:50
adu Zoffix: US election is an example of what?
Zoffix adu, [22:42:21] <timotimo> it's apparently really easy to rally people behind a few simple lies
adu as in Hillary and Drumph?
Zoffix Yeah 02:51
Sniper Fire Shillary :)
adu I don't think their lies are simple
and besides, it's really difficult to decide between these two, because one of them is always menstrating, and the other one is a woman 02:52
02:54 ssotka joined
Zoffix The fact that 300,000,000 people have to decide between these two is already a good sign of how messed up Murican government is. 02:54
adu I could vote third party, but it wouldn't do any good, since my state has been blue since the beginning of time 02:55
Zoffix That's exactly why Drumpf will become president. No one cares :) 02:56
02:56 freezerburnV left
timotimo sleep time, gnite! 02:56
Zoffix night
adu oh I care
but I'm also not delusional 02:57
timotimo US election procedure is one of the silliest clusterfucks in the history of ever
oh well.
ShimmerFairy timotimo: the gerrymandering, First-Past-The-Post voting system, and electoral college are all _positives_, I assure you :V .
adu I just wish there were choices other than killing infants and killing poor people 03:00
03:01 freezerburnV joined
tony-o poor people could eat the infants 03:01
ShimmerFairy adu: fortunately for me, the choice isn't hard. After all, voting for Hitler is not something I'm keen on doing ever. :)
03:02 aborazmeh left
adu tony-o: you're wierd 03:02
tony-o adu: the comment is based in satire, one that originated during the famines in ireland 03:03
adu tony-o: wow, I didn't know...
Zoffix ShimmerFairy, what's funny is I can't tell which Hitler you're talking about. The Drumpf who wants a wall or the Shill who can't get enough of her regime changes :)
MasterDuke it's only a modest proposal
tony-o MasterDuke++
ShimmerFairy Zoffix: nah, Hillary at worst is your standard "I don't like 'er" politician. Drumpf, however, is pretty much Hitler's re-election campaign. I'd be doing my German ancestry a disservice if I was deluded enough to vote Adolf. 03:04
adu I just had a great idea for an election
instead of voting for people, we would all contribute search queries, like (money_stolen < 10^3), (is_pedophile == False), (num_mexican_insults == 0) 03:06
03:07 addison left
adu then if the set of results is empty, have another election to remove your friends search queries 03:08
ShimmerFairy adu: heh :) . Or any of the other million things that would make elections work.
adu and repeat
or just be populist and automatically elect the US citizen with the most facebook likes 03:10
Zoffix :)
03:13 fhorck left
ShimmerFairy "I [Like] The President", a new sitcom Thursdays this Fall on ABC. 03:14
03:15 lizmat left 03:16 cdg joined 03:24 harmil left
Xliff m: sub f { }; f( '//a:foo') 03:29
camelia rakudo-moar 9b579d: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Calling f(Str) will never work with declared signature ()␤at <tmp>:1␤------> 3sub f { }; 7⏏5f( '//a:foo')␤»
Xliff m: sub f($a) { }; f('//a:foo')
camelia ( no output )
Xliff m: my @a = (); sub f($a) { }; f('//a:foo')
camelia ( no output )
03:30 boyo joined, boyo is now known as Guest84838
tony-o imgur.com/gallery/zUba3 03:38
03:51 rgrinberg left
Guest84838 Getting "Expected MAST::Frame, but didn't get one" from t/client.t while installing JSON::RPC using panda? Anyone know what would cause that and how to fix? 03:52
(I'm trying to install Rakudo Star for the first time in a while after YAPC) 03:53
Of note: this is a newish Linux Mint installation so there may be some packages missing. e.g., I had to install libssl-dev just to get this far 03:54
04:01 khw left 04:11 huggable left 04:12 huggable joined 04:21 mantovani joined 04:26 inokenty left 04:28 freezerburnV left 04:29 mantovani left 04:39 molaf joined 04:41 inokenty joined 04:45 cdg left 04:48 cyphase left 04:49 mantovani joined 04:53 cyphase joined 04:55 sortiz left 04:58 Guest84838 left, Cabanossi left 05:00 finanalyst joined, Cabanossi joined 05:01 grondilu joined 05:03 BenGoldberg left 05:04 cyphase left 05:09 cyphase joined 05:10 molaf left 05:21 adu left 05:29 lizmat joined 05:35 kerframil left 05:40 kerframil joined 05:47 cpage_ joined 05:49 _mg_ joined 05:56 ufobat joined 05:58 rindolf left 06:05 rindolf joined 06:06 finanalyst left 06:07 wamba joined 06:09 dupek joined
dupek masak, hey 06:12
06:15 setty1 joined, kerframil left 06:16 finanalyst joined 06:24 skids left 06:28 skids joined 06:29 skids left 06:33 MilkmanDan left, firstdayonthejob joined 06:34 jjido joined 06:36 jjido left, CIAvash joined 06:37 mantovani left 06:39 firstdayonthejob left 06:41 jjido joined 06:42 sno left, abraxxa joined 06:44 MilkmanDan joined 06:46 domidumont joined 06:50 domidumont left 06:51 domidumont joined 06:52 jjido left 06:54 jjido joined, ssotka left 06:56 darutoko joined
dupek hey someone 06:59
is here?
moritz it really depends on the "here" 07:00
that's always a bit tricky in distributed systems
07:02 setty1 left
dupek :) 07:02
I remember that someone told be 07:03
:parameter
if not defined is treated as True
is it True?
moritz dupek: can you think of a way to find out by asking rakudo? 07:04
dupek moritz, believe me I am trying 07:05
sub f(:bin) { say $bin }; f(:bin); 07:06
07:06 Zoffix left
moritz m: sub f(:bin) { say $bin }; f(:bin); 07:06
camelia rakudo-moar 9b579d: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Missing block␤at <tmp>:1␤------> 3sub f(:7⏏5bin) { say $bin }; f(:bin);␤»
moritz m: sub f(:$bin) { say $bin }; f(:bin);
camelia rakudo-moar 9b579d: OUTPUT«True␤»
dupek good :D 07:07
moritz, thnks bro
moritz you're welcome dude 07:08
dupek :)
07:08 dupek left 07:10 cyphase left, cyphase joined 07:12 girafe2 left 07:29 Zoffix joined 07:32 cyphase left, sno joined, cyphase joined 07:43 finanalyst left
ufobat morning perl6 :D 07:45
eyck mornin' 07:46
Xliff \o 07:47
Why is rakudo saying I can't set an attribute that isn't read-only?
"Cannot assign to an immutable value" 07:48
moritz Xliff: please show the code 07:49
Xliff Lemme gist it.
07:50 jjido left 07:51 lizmat left
Xliff gist.github.com/Xliff/96e48eaad469...e-value-md 07:53
07:54 lizmat joined
nine Very highly recommended: vimeo.com/71278954 07:55
moritz Xliff: are you binding to $!node anywhere? 07:56
Xliff Hrm....
Maybe. Lemme grep.
....possibly..... but the attribute itself doesn't appear on the RHS anywhere in the code. 07:57
07:58 finanalyst joined 07:59 finanalyst left
Xliff I just pulled all of the binds out. They were unnecessary anyway. 08:03
Changed $.node to is rw and set it normally and now I am getting: "Cannot modify an immutable xmlNode" 08:11
08:11 bjz joined
Xliff Oh well. I'm tired. This can wait until tomorrow. 08:11
08:11 huggable left 08:12 huggable joined
Xliff_zzzz g'night! o7 08:14
08:16 eugen left 08:17 jjido joined 08:21 bjz left, donaldh joined
dalek href="https://perl6.org:">perl6.org: 57c3528 | nige123++ | source/documentation/index.html:
Add SixFix link
 08:22
08:26 araujo_ left, salv0 left 08:27 bjz joined 08:38 bjz left, RabidGravy left 08:39 salva joined 08:46 bjz joined 08:52 araujo joined 08:55 jjido left 09:02 jjido joined 09:05 bjz_ joined 09:06 donaldh left 09:07 bjz left 09:12 iH2O joined
iH2O great defeat for the New World Order!! 09:13
09:16 AlexDaniel joined, donaldh joined
nine will never understand people 09:21
iH2O the vote was visceral, nine 09:24
the pro-brexit voted in very large number
nine Does "visceral" in this context mean "on an emotional basis"? 09:25
iH2O on a nationalistic/patriotic basis
nine ah ok
iH2O though im not sure lol
may be just anti-European too, or more probably a mix of both 09:26
gregf_ unpredictable ... like the weather out here :| 09:27
the first day of summer.... and theres a downpour 09:28
masak I'm sad to see the UK leave like this
iH2O u in the UK or in continental Europe?
masak from what I understand, we have to cut the anchoring that holds the islands stationary next to the continent, and let them drift free on the high seas 09:29
iH2O: neither. I'm in Scandinavia :)
we're also mildly sessecionist, but less so
iH2O Sweden should divorce from the New World Order too
secede from the NWO 09:30
gregf_ iH2O: sorry, whats the New World Order?
isolating refugees?
iH2O the NWO favors refugees coming to the west
gregf_ dont find anything wrong with that. do you know what its like to lose a loved one? in war? 09:32
masak iH2O: I do fear you're being at least partly serious
iH2O: may I remind you that this is #perl6, not #politics6
gregf_ out here in the west we care more about our possessions and pets.. less about human beings :|
iH2O need a nap z..... 09:34
gregf_ masak: *never prefers politics*
iH2O re-z.....
masak let's stick to ushering in New World Orders using Perl 6 instead ;) 09:36
gregf_ Perl6++ 09:38
hahainternet hola sad brits
iH2O hi ha 09:39
diakopter and happy brits?
hahainternet there are no happy brits
just sad brits and stupid brits
iH2O 52% are happy i presume
diakopter happy welsh?
iH2O logicallly
hahainternet the welsh vote was the most bizarre
iH2O was it rigged?
hahainternet in very many ways yes, but by the media 09:40
diakopter is any election not rigged?
hahainternet it's a very sad day for us :(
iH2O poor hahainternet
hahainternet poor UK 09:41
that name won't last
iH2O lol
hahainternet soon it will be England
iH2O hmm
England will secede from the UK?
interesting
hahainternet no, scotland have already called for a new referendum
diakopter lol
hahainternet which they will win
sinn fein have called for a vote on reunification
which could go either way
huf and then it's only wales left 09:42
hahainternet wales isn't its own nation
iH2O could Russia have seceded from the Soviet Union at the time of the cold war?
huf it should be!
if we're lucky, the danelaw will also secede eventually :D
hahainternet huf: haha welsh nationalists do exist!
but no this is real bad
really really unbelievably bad
pain and misery to tens of thousands bad 09:43
iH2O danelaw??? that still exists
huf the people eat what the politicians cook
iH2O ???
hahainternet no the people eat what the media cooks
huf iH2O: the north is still the north
hahainternet this referendum was bought by lies
printed on buses, advertised on TV
iH2O who controls the media in the UK haha?
hahainternet they were permitted to lie to the public on an industrial scale
iH2O: the largest owner is likely murdoch
iH2O its the pro-remain side
huf iH2O: once norn iron and scotland go, the biggest divides will be wales and north vs south, no? :) 09:44
hahainternet iH2O: i'm talking about lies, not politics
iH2O the same
huf or london vs everyone else :D
iH2O london isnt the UK huf, i hear that all the time 09:45
huf yeah, so there's more left to fall apart here
this could be as fun as yugo
iH2O wat
no
huf everything ends at some point... 09:46
the uk had a good run :)
incidentally, when are they going to ban these strange foreign italian letters? :)
iH2O be positive, huffy
:) 09:47
ShimmerFairy positive: popcorn sales go up :P
nine ~/win 15
iH2O perl 7 is coming soon for Windows 15 09:48
nine All the more funny because both Perl 6 and Windows 10 are considered the last major versions respectively 09:50
09:51 TEttinger left
iH2O thats because they want to avoid having Windows 11, Windows 12, then *Windows 13* 09:51
they dont want "Windows 13*
after what happened to Apollo 13 09:52
huf they can just jump to years again and nobody will notice :)
windows 17 next year, oslt
iH2O if they jumped that big everybody would notice huf 09:53
nine Windows hasn't had any kind of consistent (marketed) versioning scheme since NT 4 09:54
Woodi I hear some Japanees fears number 4...
araujo in some parts of Asia yeah 09:55
they do
iH2O if you have to fear a number, 13 is better because u bump into it less often than 4
araujo because the number 4 pronunciation is quite similar to the pronunciation of dead
iH2O so perl6 has a future in Japan 09:56
(perl4 never had a chance)
araujo hehe
huf araujo: yeah, but that's so *so* weird :D
i mean, not weirder than the fear of 13, but ...
to get hung up on a coincidence in pronunciation...
am i supposed to fear fish now?
Woodi anyway, advice for politicians: don't make total cesspool from your own place; you will hate it yourself and realy liked if someone other take over over you :> 09:57
09:57 andreoss joined
araujo huf, sure, superstitions ... 09:58
09:58 andreoss left
araujo and they change from culture to culture :D (I think I heard once the 13 is actually lucky numbers for some cultures?) 09:59
huf gotta be :)
10:03 kaare__ joined
nine If I were to fear a number (and why would I? I love them!), it'd be something like 3^^^3 because I cannot imagine running into that ever. 10:03
huf i dont know, i think 3 is pretty indecent and sexually licentious 10:05
10:07 trnh joined, rindolf left
iH2O the 3 body problem has already been solved, huffy 10:11
by astronomers and astrologists
AlexDaniel just use binary…
and you're not going to stumble upon this bullshit :/
iH2O q-bits instead of bits? 10:12
q-bits are the future, right?
masak believe it's spelled "qubit"
iH2O perl6 needs builtin support for qbits now that google claims to have a working quantum computer 10:13
hahainternet didn't you see damian's keynote? 10:14
iH2O ?
hahainternet Quantum::Superposition module
10:15 pmurias joined, _mg_ left 10:17 CIAvash left 10:24 rindolf joined
iH2O im rather new to perl6 and i dont know how much vim has support, so i invented my own support 10:25
that is, i defined the keys z and t this way:
when i type 5z, perl6 is called on the next 5 lines and to paste the output into my editing session i type t 10:26
if i type only z, then perl6 is called on the whole file being edited
thats it 10:27
thats very straightforward, i have it implemented on both linux and windows
i can post the code if anybody is interested
Zoffix wakes up 10:29
AlexDaniel Zoffix: good morning! How is your IRC thing going?
:)
Zoffix UK left EU! Heh. Called it!
hahainternet it's not a thing to be excited about Zoffix
it's hell for us now
iH2O the nightmare is after you wake up? 10:30
hahainternet this isn't waking up
this is a stupidity induced coma
one of the world's most notable nations is likely to be split up 10:31
and the lives of tens of millions will be seriously affected
it's an incredibly sad day for the citizens of the UK
iH2O (of course above i used "5z" as an example. you can use any number instead of 5)
10:31 jjido left
iH2O are you a citizen of the UK 10:32
hahainternet i am
iH2O poor haha, u doomed
abraxxa this has a great explanation of how UTF-8 is stored/encoded/works (International text (in networking)): arstechnica.com/apple/2016/06/ipv6-...os-sierra/
10:33 jjido joined
hahainternet i've also been looking for work, presuming the referendum would be remain, so now i'm going to be in an interesting position soon 10:34
it's extremely depressing, a few old people have ruined our lives
iH2O you wanted to work in France, now its illegal
hahainternet not for another 2 years
but it's possible i will become a foreigner to europe
even without, the UK is swinging radically to the right
iH2O the UK will recapture its centuries old strong identity 10:35
tadzik following the trends of the rest of europe, sadly
iH2O isnt that good haha?
hahainternet iH2O: the UK will not survive to have an identity
scotland will leave if they at all can, which they probably can
there's call for a reunification vote in ireland 10:36
tadzik oh, survive it will
tons of countries are not in the EU, and poorer than UK
hahainternet tadzik: the UK without scotland or northern ireland is known as 'England'
or 'Britain'
iH2O Welsh?
hahainternet the identity we'll be reclaiming is our roman identity, with border controls on the wall
iH2O u forget them
hahainternet iH2O: they are not an independent nation 10:37
iH2O shame!
hahainternet even though they'd like to be
stmuk I don't wish to inject too much politics into #perl6 but the UK is likely to withdraw from only the poltical aspects of the EU. It's likely to end up in EFTA/EAA
Zoffix AlexDaniel, I haven't been working on it. Gonna wait for after Damian concurrency talks
hahainternet stmuk: it's not at all clear that's the case
stmuk we shall see
hahainternet stmuk: the leading voices negotiating from this side of the channel are quite radical
i do apologise for bringing politics in though 10:38
i'm just incredibly annoyed and frustrated
stmuk I'm very familiar with those voices
hahainternet i wouldn't mind if there was a good factual basis for this
profan i woke up to swedish nationalists taking advantage of the british exit vote already
hahainternet but that's the real rub
profan someone send help
hahainternet profan: yep that's happening across europe
i think everyone under 40 or 50 in the UK would like to apologise for our terrible terrible mistakes :( 10:39
iH2O Britain rules the 7 seas!
stmuk the UK is likely to end up like Norway or Switzerland
tbrowder AlexDaniel: ref comment on my PR, that "Notice the space..." comment I think is OBE (I made it when I first thought the example needed it--I'll check and correct if the comment is now incorrect 10:40
stmuk hahainternet: you don't speak for everyone
profan the divide is the scary part
hahainternet stmuk: yes i'm aware
and we may end up like switzerland, but it's going to be luck if we do
iH2O Switzerland is a tax heaven
Norway is a oil heaven 10:41
hahainternet iH2O: we have a lot of corrupt banking groups, that's valuable right? :)
anyway i'll shut up and we can get back to perl :)
iH2O banksters grrr
tadzik hahainte1net: well, we're talking about *with* this time though :) 10:42
they all left, the entire UK
timeless m: say "abc".comb.map({ "b" ?? "d" !! "e"});
camelia rakudo-moar 9b579d: OUTPUT«(d d d)␤»
hahainternet tadzik: that's not clear at all yet
timeless where is that form of `comb` documented on doc.perl6.org?
tadzik hahainte1net: how so? 10:43
timeless AlexDaniel: ?
hahainternet tadzik: scotland will make every effort they can to remain in the EU, NI may reunify (that is very unlikely though)
Zoffix timeless, docs.perl6.org/routine/comb
timeless, "If no matcher is supplied, a list of characters in the string (e.g. $matcher = rx/./) is returned."
tadzik hahainte1net: ah, that may be interesting to see, yes
hahainternet tadzik: i was born in england, but i would move to scotland in a heartbeat to remain in the EU, my long term life goal was to buy loads of land in france and retire there 10:44
timeless Zoffix: oh, so i'm not really using comb
so much as map?
timeless sighs
AlexDaniel timeless: yeah, it's just comb + map :)
timeless ok...
AlexDaniel m: ‘abc’.comb.perl.say
camelia rakudo-moar 9b579d: OUTPUT«("a", "b", "c").Seq␤»
AlexDaniel m: ("a", "b", "c").Seq.map({ "b" ?? "d" !! "e"}).say; 10:45
camelia rakudo-moar 9b579d: OUTPUT«(d d d)␤»
timeless m: say "abc".comb.map({ "b" ?? "d" !! "e"}).join;
camelia rakudo-moar 9b579d: OUTPUT«ddd␤»
Zoffix timeless, you're not really using anything, since you're just discarding the comb results and stuffing
hahainternet "b" there isn't testing anything but the truth of "b" is it?
Zoffix 'd' for each item
hahainternet, right
timeless Zoffix: i simplified what i was actually doing for the question 10:46
AlexDaniel m: ("a", "b", "c").Seq.map({ $_ ~~ "b" ?? "d" !! "e"}).say
camelia rakudo-moar 9b579d: OUTPUT«(e d e)␤»
Zoffix m: say 'd' x 'abc'.comb
camelia rakudo-moar 9b579d: OUTPUT«ddd␤»
timeless unfortunately i had a 3 line expression and tried to replace it w/ a 1 line expression 10:47
but i lost the value of one of the lines (namely `join`)
10:47 iH2O left
AlexDaniel comb is actually well-documented :) 10:47
Zoffix And well-blogged: perl6.party/post/Perl6-Comb-It 10:48
dalek c: 42ad323 | (Tom Browder)++ | doc/ (26 files):
remove space between sub/method names and opening paren
 10:49
c: b46490b | (Tom Browder)++ | doc/Language/functions.pod:
remove incorrect statement
c: f5f6c50 | (Aleks-Daniel Jakimenko-Aleksejev)++ | doc/ (26 files):
Merge pull request #630 from tbrowder/sub-space

remove space between sub/method names and opening paren
timeless tbrowder: github.com/perl6/doc/commit/42ad32388f 10:50
your commit included something else... 10:51
AlexDaniel timeless: it was fixed in the next commit
timeless (the first change in the diff)
AlexDaniel oh
you mean that comma?
timeless yeah
AlexDaniel any problems with that?
timeless bad summary :)
AlexDaniel timeless: yeah, that happens
timeless: also people tend to remove trailing whitespace on other lines sometimes 10:52
tbrowder the comma was added to use the accepted form of e.g., but I did not state that
timeless (working on dirty working directories when trying to make mass changes is messy)
tbrowder: i definitely agree the comma should be added 10:53
AlexDaniel in an ideal world we probably should not do that, but having a separate commit for one comma is also not so awesome :)
tbrowder I will strive to do better in the future... 10:54
Zoffix tbrowder, your commit was fine. timeless is just being a pedant :)
tbrowder++ # fixing all the docs
AlexDaniel being a pedant is also fine :)
as long as it is all about friendliness and love ;) 10:55
(and I hope that's the case)
timeless definitely trying to be friendly 10:56
tbrowder I love pedants, being kind of OCD myself
but i'm also too careless much of the time, but sorry afterwards... 10:57
10:58 jjido left
AlexDaniel tbrowder: by the way, I think that you can commit --amend to your own fork 10:58
am I right?
hahainternet has it been decided that there will be no space permitted between sub name and the signature? or is this just for consistency?
AlexDaniel tbrowder: this way we wouldn't need to have “oops” commits in the main repo 10:59
hahainternet: it's probably better to have no space in docs
hahainternet: so yeah, just for consistency
m: sub foo ($x, $y) { say $x, $y }; foo(42, 69); 11:00
camelia rakudo-moar 9b579d: OUTPUT«4269␤»
hahainternet i tend to like a space in the definition so the name is a bit more clearly distinguished, but i just wanted to know :)
AlexDaniel m: sub foo ($x, $y) { say $x, $y }; foo (42, 69);
camelia rakudo-moar 9b579d: OUTPUT«Too few positionals passed; expected 2 arguments but got 1␤ in sub foo at <tmp> line 1␤ in block <unit> at <tmp> line 1␤␤»
hahainternet no i understand that distinction, just wondered if the issue had been decided since a couple days ago
AlexDaniel tbrowder: so that's something you can probably try next time :) 11:02
11:03 jjido joined 11:10 iH2O joined, trnh left 11:16 jjido left 11:19 a3r0 joined, _mg_ joined
timeless AlexDaniel: ok, so fun edge from the link checker 11:25
11:25 RabidGravy joined
timeless view-source:doc.perl6.org/routine/elems.html 11:25
<p>From <a href="/type/Uni#elems">Uni</a></p> 11:26
-- link checker says that Uni does not have #elems
but, presumably that was generated by something...
moritz timeless: likely #method_elems would be correct 11:28
11:28 kid511 joined
timeless moritz: from the output, it seems like this is systemic 11:28
is it likely the pod to html thing itself? 11:29
timeless goes searching for elems
oh
11:29 rindolf left
timeless ok 11:29
moritz timeless: it's likely because we don't always know if something is a method or a sub 11:30
or what the anchor is
timeless ./htmlify.p6:835: my $p = pod-with-title(
"Documentation for method $type.$name",
pod-block('From ', pod-link($type, "/type/{$type}#$name")),
@$pod,
)
thankfully i have enough logging to be able to see where things are from :) 11:31
oh
timeless ponders
moritz: so... 11:32
we could probably have the generated .html for Uni include an extra <a id="elems">
moritz timeless: that would be nice 11:33
though we have to take care that we don't duplicate the id
timeless it's a lot of extra content though
moritz because include methods from superclasses and from roles
which can have the same name 11:34
gfldex there are no linkable anchors for X<> too
timeless moritz: so,...
we could get a count pretty easily 11:35
to figure out if something is unique
gfldex maybe some JS magic that highlights search/link terms would help to deal with dupes
timeless in fact
the code already thinks about duplicates
my $subkind = @subkinds.elems == 1 ?? @subkinds.list[0] !! $kind;
moritz oh nice 11:36
timeless is that the only thing it needs to not mess up?
can something have both a method and a const or ... for the same name?
gfldex in github.com/gfldex/pod-to-bigpage i ended up using numerical link targets to deal with duplicated index entries 11:37
timeless ponders 11:39
11:41 rindolf joined 11:46 cyphase left 11:49 iH2O left, cyphase joined 11:50 cpage_ left
timeless ok 11:51
we have duplicate ids... in spades
a quick sample www.irccloud.com/pastebin/XlpCN8nc/ 11:54
method_ACCEPTS is quite popular
html/routine/ACCEPTS.html
method_ACCEPTS: 11,
especially there :o 11:55
html/routine/new.html
class_Nil: 2, method_new: 26,
<h2 id="method_new"><a class="u" href="#___top" title="go to top of document"><a href="/routine/new#class_Backtrace">method new</a></a></h2> 11:58
ok, so um... where does `go to top of document` come from?
11:58 rgrinberg joined
timeless oh 11:58
timeless sighs 11:59
timeless needs to fix make clean-html
dogbert17 moritz: hi, are you still around? 12:00
12:01 jjido joined
timeless gfldex: any idea where that go to top thing comes from? 12:02
unit class Pod::To::HTML;
multi sub node2html(Pod::Heading $node) {
grr 12:03
so that's a rakudo-bug for Pod::To::HTML ?
12:05 donaldh left
moritz dogbert17: yes-ish 12:05
BrokenRobot Do any toxic deadly chemicals smell like manure? I just rode the bus through several blocks and it reeked. Still stinks at $workplace. I wonder if I should evacuate :|
gfldex timeless: i can't find it
dogbert17 moritz: if you're bored :) maybe you could take a look at gist.github.com/dogbert17/56d39157...46f9261355
timeless gfldex: i found it 12:06
the two lines i quoted after i asked you are from the relevant file
(yay for grepping in rakudobrew)
moritz dogbert17: looks good, at a glance
timeless so..., how the heck does one fix that? 12:07
gfldex does it come from a module?
12:07 jjido left
timeless i don't really really want to hack Pod::To::HTML from outside it 12:07
BrokenRobot AlexDaniel: but that's not really what tbrowder++'s commit was about though, was it? I glanced at it and just saw a lot of spaces in the *definitions* removed. As in sub foo ($a, $b) {} changed to sub foo($a, $b) {}
BrokenRobot prefers a space there themselves.
timeless i'm pretty sure that *is* the module
BrokenRobot Yes, it is.
dogbert17 moritz: I'll fix 'invert' later, it lacks info on how list values are handled 12:08
gfldex you would have to make the current bahaviour a default and allow a parameter, then use that from perl6-doc
BrokenRobot modules.perl6.org/repo/Pod::To::HTML
It's community-manged, so I think you have access to hack on it, timeless
12:08 brrt joined
BrokenRobot There's an open Issue for duplicate <a>s 12:09
timeless hrm, since i can lock the issue, i guess that means i can hack it
eh? 12:10
github.com/perl6/Pod-To-HTML/issues
BrokenRobot github.com/perl6/doc/issues/561#is...-223566753
timeless only shows something for poorly formed <ul>s
tbrowder re spaces between sub names and (): I'm at TPC::NA::2016 and had the good fortune to ask Larry about the space and sub names and () and he said that they might revisit that issue later
12:11 huggable left 12:12 huggable joined 12:13 cooper_ joined, cooper_ left, cooper_ joined, lizmat left
tbrowder also asked Damian, Larry, and Patrick about best practices and Damian said he hasn't written enough p6 code yet but he is thinking about it, Larry said it's probably premature at the moment, and Patrick thinks we ought to start BP idea lists and votes on a github wiki 12:14
timeless view-source:doc.perl6.org/routine/new.html
github.com/perl6/Pod-To-HTML/issues/14
dalek c: 70fc547 | (Jan-Olof Hendig)++ | doc/Type/Map.pod:
Added docs for Map.antipairs. moritz++
 12:18
moritz timeless: github.com/moritz/perl6-wtop exists, though not very active at the moment 12:19
timeless moritz: was that for me? 12:20
buharin masak: is here? 12:21
12:21 woolfy left
dogbert17 .seen masak 12:21
yoleaux I saw masak 10:12Z in #perl6: <masak> believe it's spelled "qubit"
timotimo i think it was meant for tbrowder
moritz erm no, sorry, for tbrowder 12:22
timeless which isn't to say i didn't start this a few days ago, because i might have /sorta/ :)
moritz tbrowder: github.com/moritz/perl6-wtop an early attempt at gathering good practises
timeless although i think someone else specifically complained about it causing problems
and thus was a more significant impetus
gfldex / BrokenRobot : so... 12:24
BrokenRobot ?
timeless i can't follow this
BrokenRobot The double link issue?
timeless yeah
i mean, i understand why it's happening
12:24 stevieb9 joined
BrokenRobot Pod::To::Html generates links. Whatever else is generating links in htmlfy.p6 needs to stop doing it. 12:24
timeless no
12:24 cpage_ joined
BrokenRobot That's my interpretation of the problem, but if I knew the exact cause, I'd fixed it already 12:25
timeless here only Pod::To::HTML is at fault
this isn't that kind of double-link
this is double-id
BrokenRobot Ah
12:25 aries_liuxueyang joined
timeless or in routine/new.html 26x id= 12:25
because doubling is boring :)
BrokenRobot No idea.
gfldex the only lasting solution would be to generate uniq ids for ids, what will clash with perma-links
timeless the problem is... i can't figure out where `class X` comes from
because it doesn't seem to come from this module 12:26
if it knew that, it could generate {class}_{method}
hrm 12:27
gfldex maybe you could subclass from the module?
timeless maybe i need to look at the def of `Pod`
nah, at the end of the day, someone (might as well be me) needs to fix the module
so, where is `Pod` as a class/module? 12:28
./moar-nom/src/core/Pod.pm ?
(that's unhelpful) 12:30
gfldex most of the work is done in the grammar / action class 12:32
timeless so Pod::To::Text looks at a .HOW
but Pod itself doesn't have one 12:33
psch m: say Pod.HOW 12:34
camelia rakudo-moar 9b579d: OUTPUT«Perl6::Metamodel::PackageHOW.new␤»
psch m: say Pod::.keys
camelia rakudo-moar 9b579d: OUTPUT«(FormattingCode Item Heading Config Raw Block)␤»
timeless waits for a decoding 12:35
psch timeless: ...it's just a package
timeless HOW?
timeless feels like this could be whos-on-first 12:36
psch timeless: no, Pod
timeless: HOW is "higher order workings", as in "HOW does it know what it does" (or something
timeless psch: right, it's a package, i read it
psch )
timeless Pod::To::Text reads from a .HOW 12:37
i was hoping to see .HOW in Pod to explain what it knew
but it wasn't there
psch timeless: the different HOWs are in src/Perl/Metamodel/
12:37 araujo_ joined
psch timeless: and nqp/src/how has the bootstrap HOWs 12:38
kid511 cite-sciences.ubicast.tv/lives/journees-perl/ live stream of talk on Perl 6 from French Perl workshop
psch well, nqp HOWs more precisely
i think the KnowHOW is always VM specific, and bootstraps everything else
m: say Any.HOW 12:39
camelia rakudo-moar 9b579d: OUTPUT«Perl6::Metamodel::ClassHOW.new␤»
psch m: say Any.HOW.HOW
camelia rakudo-moar 9b579d: OUTPUT«Method 'gist' not found for invocant of class 'NQPClassHOW'␤ in block <unit> at <tmp> line 1␤␤»
psch m: say Any.HOW.HOW.HOW
camelia rakudo-moar 9b579d: OUTPUT«Method 'gist' not found for invocant of class 'KnowHOW'␤ in block <unit> at <tmp> line 1␤␤»
12:39 araujo_ left
psch so the KnowHOW is the bare bones metaclass of everything, from which we build the NQPClassHOW, which we use to build the Perl 6 level ClassHOW 12:39
12:39 araujo_ joined
timeless ok... 12:40
psch has a strong feeling that didn't actually help...
timeless right! 12:41
psch timeless: what's the actual thing you're trying to do or find out?
12:41 araujo left
timeless whether or not Pod::To::Html should have enough information to know the class it's talking about 12:41
12:41 araujo_ left
psch if it has the type object it knows everything there is to know 12:41
timeless so that it can avoid generating the same ambiguous id
psch or, well, can find out
12:42 araujo_ joined
timeless multi sub node2html(Pod::Heading $node) { 12:42
id => escape_id(node2rawtext($node.contents)),
[Coke] tryperl6.org/ - my google alert notification found this
timeless $node is presumably the class you're talking about?
psch timeless: $node is clearly a Pod::Heading :) 12:43
timeless m: say Pod
camelia rakudo-moar 9b579d: OUTPUT«(Pod)␤»
[Coke] m: say 1™
camelia rakudo-moar 9b579d: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Bogus postfix␤at <tmp>:1␤------> 3say 17⏏5™␤ expecting any of:␤ infix␤ infix stopper␤ postfix␤ statement end␤ statement modifier␤ statem…»
timeless m: say Pod.Heading
camelia rakudo-moar 9b579d: OUTPUT«Method 'Heading' not found for invocant of class 'Pod'␤ in block <unit> at <tmp> line 1␤␤»
timeless m: say Pod::Heading
camelia rakudo-moar 9b579d: OUTPUT«(Heading)␤»
[Coke] (the try site allows that, ah well)
timeless m: say Pod::Heading.HOW
camelia rakudo-moar 9b579d: OUTPUT«Perl6::Metamodel::ClassHOW.new␤»
timeless sighs 12:44
psch: i think i'm going to go back to being totally lost
12:45 wamba left
psch timeless: i still don't know what you're trying to do, honestly 12:45
timeless: i mean, i think it's about caching Pod::Headings..?
timeless psch: view source of doc.perl6.org/routine/new.html
look for `id="method_new"` 12:46
when you've hit say the 25th instance, let me know
[Coke] If the person who runs tryperl6.org is here, a link to report issues on the site would be appreciated.
12:46 wamba joined
timeless note that in XML, and realistically, HTML, you're only supposed to have 1 id= 12:47
(you can do whatever you like in html, but browsers will start thinking you're crazy pretty quickly if you do...)
psch timeless: and $node.contents doesn't give you anything to differentiate the ids?
timeless the generated content indicates it's just the `new` 12:48
somewhere presumably is more useful stuff like the container
but gosh, navigating this is painful
psch did check in the generating code what $node.contents contains? 12:49
timeless ?
i can't figure out where any of that is
Pod.pm is close enough to empty
Pod::To::Html isn't where things come from
psch 12:42 < timeless> multi sub node2html(Pod::Heading $node) {
12:42 < timeless> id => escape_id(node2rawtext($node.contents)),
timeless that's from Pod::To::Html 12:50
12:50 kerframil joined
psch so am i understanding that righ 12:50
timeless it's possible assemble-list-items is relevant
psch t
that Pod::To::Html isn't where the html comes from..?
timeless github.com/perl6/Pod-To-HTML/blob/...To/HTML.pm 12:51
is that file if you want to try to read it
instead of relying on my sleep-deprived....
it's possible that assemble-list-items is doing stuff... 12:52
psch hm, yeah, that does look like the Heading.contents is really just "method new"
curious where the link to another class comes from then
timeless that's doc/htmlify.p6 12:53
pod-block(
"$podname inherits from class ",
pod-link($class.name, "/type/{href_escape ~$class}"),
", which provides the following methods:",
),
%routines-by-type{$class}.list,
github.com/perl6/doc/blob/master/htmlify.p6 12:54
psch ah, alright 12:56
i guess the easiest way to solve the dup id issue is by adding a state hash that tracks all the existing ones and has an Int as value that increments which each use 12:57
starting at 0 probably :)
timeless so...
i'd much rather it actually retain useful information
because if it manages to know x is a class X
12:58 Sgeo_ left
timeless then it really should be able to generate id="X_method_x" 12:58
but i can't follow this to figure out if it knows X
psch doesn't look like it
at least not in Pod::To::HTML
awwaiid Is there an easy way to get perl6-code whitespace detection via something like / <.ws> / ? I want it for the REPL so that if you enter space or a comment it doesn't really eval it
timeless but...
psch htmlify seems to know, but then you're back to s/// on the html directly
timeless htmlify is calling p2h/pod2html with an entire pod, right? 12:59
psch now idea, this is literally the first time i ever looked at either :)
BrokenRobot m: 'foo' does role { method uc { say "We're in! Someone's trying to uppercase {self}"; callsame; } }; say 'foo'.uc; 13:00
camelia rakudo-moar 9b579d: OUTPUT«We're in! Someone's trying to uppercase foo␤FOO␤»
BrokenRobot How come this works? I thought roles were composed into classes and the above callsame wouldn't work, since the method would be replaces?
*replaced
jnthn BrokenRobot: That's for compile-time composition. 13:01
BrokenRobot Ah. Alright. Thanks.
jnthn BrokenRobot: The `does` infix operator does a runtime mixin, which is done by making an anonymous subclass and mixing the role into that and then reblessing the object into that subclass
BrokenRobot Ah. Now it makes sense. jnthn++ 13:02
awwaiid It's like I want Perl6::Grammar::ws
timeless psch: i think i need to find the thing that contributes `indexList indexList2`
ok, that's pod2html 13:03
because do-toc knows about classes and things 13:05
13:07 mcmillhj joined 13:09 gnull` joined
timeless eh 13:09
doc.perl6.org/routine/new.html
doc.perl6.org/routine/new 13:10
new.html links to new.html + new
new just links to new
13:11 mcmillhj left, mcmillhj joined
timeless (there's only one underlying file: new.html, the system remaps new as new.html) 13:12
13:16 eugen joined
timeless ok, so, i think i /sorta/ understand this 13:17
my $new-head = Pod::Heading.new(
in htmlify
13:17 donaldh joined
BrokenRobot m: &put does role { method CALL-ME (*@args) { print "We're in! Someone's trying to say @args[]\n"; nextsame; } }; put 'foo', 'bar'; 13:23
camelia rakudo-moar 9b579d: OUTPUT«We're in! Someone's trying to say foo bar␤»
BrokenRobot How come in this case nextsame doesn't work?
jnthn BrokenRobot: Because there's no CALL-ME to delegate to?
Routines are "natively" invokable, rather than going via CALL-ME 13:24
BrokenRobot Ohhh. Thanks, jnthn++
jnthn Though we could stick a method CALL-ME(|c) { self(|c) } or so up in Code and that'd make it work.
Uh...or would it... :)
It may have to look a tad different to that :) 13:25
13:25 cdg joined
gfldex m: my &orig-put = &put; put does role { method CALL-ME (*@args) { print "We're in! Someone's trying to say @args[]\n"; nextsame; } }; put 'foo', 'bar'; orig-put 'test'; 13:25
camelia rakudo-moar 9b579d: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Undeclared routine:␤ does used at line 1␤␤»
BrokenRobot Is there a way to call the original put above or do I have to emulate it's behaviour?
gfldex you could use .wrap 13:26
BrokenRobot looks that up
jnthn suggets wrap also 13:27
timotimo you mean &put does role ...
not "put does role"
but "does" works on the object, not the container, so &orig-put will also see the change 13:28
literal Zoffix: I see you're doing "».say" here: blogs.perl.org/users/zoffix_znet/20...ation.html
13:28 _mg_ left
literal that's arguably wrong :) 13:28
BrokenRobot m: &put.wrap(sub (*@args) {say "We're here!"; nextsame}); put 'foo';
camelia rakudo-moar 9b579d: OUTPUT«We're here!␤»
timotimo correct, >>.say will, in the future or the past, give you different order
literal not just different order, but mix up the strings, right? 13:29
as all are being written at the same time
13:29 zakharyas joined
BrokenRobot m: &put.wrap(sub (*@args) {say "We're here!"; callsame}); put 'foo'; 13:30
camelia rakudo-moar 9b579d: OUTPUT«We're here!␤»
BrokenRobot Doesn't seem to be working/>?
psch m: say &put.signature 13:31
camelia rakudo-moar 9b579d: OUTPUT«(| is raw)␤»
psch m: &put.wrap(sub (| is raw) {say "We're here!"; callsame}); put 'foo';
camelia rakudo-moar 9b579d: OUTPUT«We're here!␤foo␤»
13:31 Sgeo_ joined
psch BrokenRobot: {call,next}{same,with} only dispatch amongst other candidates that also match the signature/args 13:31
13:32 aries_liuxueyang left
psch is a bit fuzzy on the details 13:32
BrokenRobot psch: thanks
The .wrap doesn't work for my case. Rakudo segfaults :|
psch that's a ticket! 13:33
gfldex BrokenRobot: see doc.perl6.org/language/functions#Routines
dalek c: 79b1cce | (Jan-Olof Hendig)++ | doc/Type/Map.pod:
Added missing info and example to the docs for Map.invert
 13:34
BrokenRobot This is royally annoying. I now have two blog posts that I can't publish due to bugs in rakudo 13:36
13:37 Khisanth joined
dalek c: e1a5b23 | (Josh Soref)++ | Makefile:
split clean-html by lines and sort
 13:37
c: 1437ec6 | (Josh Soref)++ | Makefile:
add programs dir to clean-html
psch BrokenRobot: start a draft on jvminterop ;)
y'know, cause good things come in threes..? :S 13:38
13:38 user9 left
jnthn If you're wrapping a sub, you must also do "use soft" 13:39
13:39 user9 joined
BrokenRobot segfaults with use soft too; 13:39
13:39 araujo_ left
BrokenRobot I'm wrapping in C.pm6, B.pm6 uses C.pm6, and A.pm6 uses B.pm6 and I try to call put in A.pm6 and it segfaults 13:40
jnthn I can't see the segfault in the backlog?
BrokenRobot Yeah, it only happens when spread in modules
jnthn Ah :/
psch m: module A { my $old = &say.wrap(sub (| is raw) { print "well, "; callsame }) }; import A; say "foo" 13:41
camelia rakudo-moar 9b579d: OUTPUT«well, foo␤»
psch import's not enough vOv
13:41 hanekomu joined
BrokenRobot precompilation bug! 13:42
:)
It actually doesn't happen even if you just use 1 module. It happens when one module uses the other. 13:43
RT: rt.perl.org/Ticket/Display.html?id=128476 13:44
Slightly wrong description 13:45
BrokenRobot hates that RT doesn't have an edit button
timeless brokenrobot: how's this? www.irccloud.com/pastebin/jLF6LDe6/ 13:46
things look a bit funny: 13:47
# class Mu
. Mu method new
# class Backtrace
. Backtrace method new
but, i think the id's are unique
[Coke] AlexDaniel: based on the chat yesterday, how's this: gist.github.com/coke/3a34284c5f591...bcf50e9e5a 13:48
timeless doc.perl6.org/type/Pod::Block has a list of useful subclasses 13:49
should those have docs? 13:50
timeless wants Pod::Block::Heading
AlexDaniel [Coke]: LHF label is probably needed too
13:50 aries_liuxueyang joined
[Coke] AlexDaniel: I tend not to find those helpful personally, but happy to include if it gets more eyes on tickets. 13:50
AlexDaniel [Coke]: I think that we should try it. 13:51
13:51 acrussell joined
timeless AlexDaniel: is it Pod::Block::Heading or Pod::Heading? 13:51
gfldex Pod::Heading
[Coke] added.
gfldex see github.com/gfldex/pod-to-bigpage/b...e.pm6#L326
[Coke] er, it got lost. one sec.
timeless gfldex: ok, so the Pod::Block doc is wrong 13:52
AlexDaniel what is “needs line editing”? What was the idea behind this label?
[Coke] done.
AlexDaniel: no clue.
gfldex timeless: the Pod classes are quite flexible. It allows the pod author to use formatting code that no renderer understands :)
AlexDaniel ha
[Coke] AlexDaniel: it's only on one ticket opened by Zoffix 3 days ago 13:53
AlexDaniel oh well
dalek c: 2fbaeec | (Josh Soref)++ | doc/Type/Pod/Block.pod:
fix Pod::Block ref to Pod::Heading
 13:54
timeless gfldex: flexible is fine, but incorrectly documented isn't helpful
AlexDaniel [Coke]: let's see. So type graph issues are ｢site｣, right? 13:55
timeless the fact that the pod things aren't doc'd much at all doesn't help matters
AlexDaniel [Coke]: 404-s are ｢site｣, broken links are ｢docs｣
[Coke] AlexDaniel: seems reasonable to me.
AlexDaniel or maybe some 404-s are docs :)
gfldex i did a lot of dd-ing to make sense of Pod::* 13:56
[Coke] broken links might turn out to be ｢build｣, but I think the default would be to assume site, yes.
timeless gfldex / AlexDaniel : doc.perl6.org/type/Pod::Block has a L<Pod::Item>
doc.perl6.org/type/Pod::Item exists
BrokenRobot AlexDaniel: my interpretation of "line editing" was just minor edits
timeless so ... why isn't it a link?
AlexDaniel BrokenRobot: ok, meaning LHF? :)
[Coke] I mean, the goal is to close all the ticket so we don't have to care. :)
BrokenRobot Yeah, it can be categorized as that
AlexDaniel [Coke]: let's say we have an issue that has no label, what would that mean? 13:57
[Coke]: that somebody has to add a label? 13:58
[Coke] I'd say
13:59 sufrostico joined
AlexDaniel [Coke]: OK, I think that your list totally makes sense 13:59
14:02 zakharyas left, pmurias left, pierrot joined
timeless github.com/pantsbuild/pants/pull/3048 14:03
timeless sighs
i don't suppose anyone here knows what a pantsbuild is, or someone who does? 14:04
BrokenRobot You tried to get into someone's pants but access was denied? :P
timotimo … 14:05
what's with technology and innuendo jokes :)
timeless i tried to offer them patches for their pants which had way too many holes
it was just plain embar-ass-ing
but, well, they can keep their ass-es 14:06
AlexDaniel [Coke]: is there any easy way to refactor existing labels? 14:08
there are only 300, sure I can handle it manually, but… :)
in fact, there's probably no need to touch closed issues 14:09
BrokenRobot "Before posting your first review, you need to both subscribe to the pants-reviews Google Group and create an RBCommons account."
shees... your *first* review? :) That's a hellalot of hoops to jump for typo fixes
14:09 woolfy joined 14:10 girafe joined
BrokenRobot AlexDaniel: I think you can rename the labels 14:10
14:11 andreoss joined
AlexDaniel BrokenRobot: well, I probably can't :) 14:11
BrokenRobot AlexDaniel: "Name has already been taken" never mind :/ 14:12
AlexDaniel BrokenRobot: where's that?
BrokenRobot AlexDaniel: I tried it on my own project. Creating a label and trying to change it to an already-existing label in an attempt to merge 14:13
For docs it'd be this page: github.com/perl6/doc/labels
AlexDaniel BrokenRobot: ah, so we cannot merge but we can rename 14:14
that's good
14:14 lizmat joined
timeless BrokenRobot: i think your comments support my "WTF" 14:15
thanks, i appreciate that
14:15 pmurias joined, brrt left
[Coke] AlexDaniel: right, only do open issues is fine. 14:16
timeless gfldex: github.com/perl6/doc/pull/632 in case you want to express an opinion 14:17
AlexDaniel [Coke]: what now? Let's wait for more comments?
timeless ... since you seem to understand this more than most
14:18 khw joined
[Coke] AlexDaniel: eh. maybe give it a few hours for someone to go "nooooooo"... or just do it and let them fix it if they don't like it. 14:18
"Remember, no matter how kind you are, German children will always be Kinder." 14:19
BrokenRobot :) 14:20
AlexDaniel hear that, people? If you *love* existing labels in perl6/docs repo, please go “noooo”.
gfldex timeless: .subst(:g, /\s+/, '_') could also create a dupe if there are two names that only differe in a ' ' and '_'. Doubling all _ before the ' ' subst should solve that. I may be nitpicking tho.
timeless gfldex: for now, i'm mostly interested in the change i myself made 14:21
your concerns are probably reasonable
gfldex timeless: however, doctors have a saying: "if you cure, you are right."
timeless killing/scaring the patient generally isn't welcomed 14:22
although in the UK, perhaps it's ok?
(they did just ruin their currency)
14:22 donaldh left
gfldex timeless: let me quote myself from a discussion i had earlier today. "the vote is not law nor does it have law character. Both chambers of parliament may have a vote on the matter. Further the queen may refuse Royal Assent. As nice a story it would be to get rid of the brits, it may never be more then that." 14:23
timeless isn't in EU 14:24
timeless does have EUR investments/retirement bits
amusingly, i'm not sure how those would be impacted
since their mostly EUR holdings against non EU items
so i guess they'd go up in EUR value and cashing them out would be a wash
(this assumes that the EUR collapses in addition to the GBP) 14:25
gfldex i'm only comfortible with predicting things in the past. :) 14:26
14:27 perlpilot_ left
grondilu off topic: rocket launch in 30s youtu.be/nFxTr00Mtdw 14:29
timotimo oh 14:30
i haven't heard of this one
grondilu ULA though, not SpaceX
timotimo these days, it's like an upgoer goes up every couple of weeks
US forces on the moon!
oh, no, not actually
AlexDaniel eh, no pretty explosion 14:32
gfldex there is something horrible wrong with how we commute. If we would be satelites, it would take us never more then 10 minutes to get to work. 14:33
AlexDaniel and slightly more than that to get to the satellite ;) 14:34
dalek c/tab-test: 19f2efe | coke++ | / (16 files):
tabs -> spaces
 14:41
c/tab-test: ec16a20 | coke++ | t/tabs.t:
give t/tabs.t a plan
14:41 perlpilot joined 14:42 go|dfish left
[Coke] That should make that branch pass all tests. 14:42
literal how do I create a baghash that allows non-string keys? I thought "$foo = baghash :{}" would do it but apparently there's no such routine :)
timotimo no that's not how you do it 14:43
:{ } will give you a Hash object with non-string keys
but when you coerce that to a BagHash, you aren't telling the BagHash itself to also have non-string keys
literal I see
timotimo you may want something like BagHash[Any]
Woodi thinks Perl6 should be advertised as very good for protoyping new projects...
timotimo though i wasn't aware BagHash would stringify everything
14:43 hanekomu left
timotimo m: my BagHash $foo; $foo{1} = 100; say $foo.keys>>.perl 14:44
camelia rakudo-moar 9b579d: OUTPUT«Type check failed in assignment to $foo; expected BagHash but got Hash (${})␤ in block <unit> at <tmp> line 1␤␤»
timotimo m: my BagHash $foo .= new; $foo{1} = 100; say $foo.keys>>.perl
camelia rakudo-moar 9b579d: OUTPUT«(1)␤»
14:44 go|dfish joined
timotimo m: my BagHash $foo .= new; $foo{hey => 1234} = 100; say $foo.keys>>.perl 14:44
camelia rakudo-moar 9b579d: OUTPUT«(:hey(1234))␤»
literal timotimo: oh, actually, BagHash seems to work for non-strings just fine, I was just doing something else wrong
timotimo doesn't seem stringifying
14:45 wamba left
AlexDaniel [Coke]: pft… yeah. Was about to commit the same thing 14:45
[Coke] AlexDaniel: whoops, sorry
AlexDaniel :)
[Coke] I have a trailing whitespace test I can throw in as well, but figured I'd see if the tab one got merged. :)
literal as for adding a List to a BagHash, the only way to do it is $baghash{$(@list)} = 1, right? 14:46
because it likes to flatten the list otherwise
[Coke] AlexDaniel: you want?
I can put it in the same branch in about 5m if you like.
14:47 brrt joined, abruanese joined
BrokenRobot literal: the parens are not necessary 14:47
psch m: my BagHash $b .= new; $b{[1,2,3]} = 1; # this looks strange :o
camelia rakudo-moar 9b579d: OUTPUT«Use of Nil in numeric context in block <unit> at <tmp> line 1␤Use of Nil in numeric context in block <unit> at <tmp> line 1␤Use of Nil in numeric context in block <unit> at <tmp> line 1␤Use of Nil in numeric context in block <unit> at <tmp> line …»
BrokenRobot psch: that's a slice 14:48
timotimo right
it expects you to assign three things
psch BrokenRobot: but how many Nils do i get
AlexDaniel [Coke]: well, I haven't decided if that's a good idea
[Coke]: do we actually want travis to scream about trailing whitespace?
psch m: my BagHash $b .= new; $b{[1,2,3]} = 1,2,3; # this looks strange :o
camelia ( no output )
Xliff_zzzz Anyone have any ideas? -- gist.github.com/Xliff/96e48eaad469...e-value-md
psch minus comment
AlexDaniel are we hoping that existence of such tests will make people leave less trailing whitespace?
huggable: dunno
huggable AlexDaniel, ¯\_(ツ)_/¯
literal BrokenRobot: ah, indeed 14:49
BrokenRobot m: my BagHash $b .= new; $b{[1,2,3]} = 1,2; 14:50
camelia rakudo-moar 9b579d: OUTPUT«Use of Nil in numeric context in block <unit> at <tmp> line 1␤Use of Nil in numeric context in block <unit> at <tmp> line 1␤»
psch oh, that's just the STORE dance isn't it
BrokenRobot prolly
AlexDaniel [Coke]: put it in, let's see how it goes :P
14:50 skids joined
[Coke] AlexDaniel: new branch so you can decide later. one moment. 14:50
AlexDaniel ah or that, sure 14:51
timeless Xliff: i hit that sort of thing regularly :/
Xliff timeless: How do you fix it? 14:52
dalek c/tws-test: aeea5b2 | coke++ | t/trailing_whitespace.t:
add (failing) trailing whitespace test
timeless i tend to create extra variables...
Xliff Near as I can tell, the object isn't immutable. So setting an attribute shouldn't be a problem.
timeless but that won't help you
Xliff Not in this situation, no. 14:53
dalek c/tws-test: 1259c40 | coke++ | doc/ (20 files):
remove trailing whitespace, test passes
 14:54
14:57 travis-ci joined
travis-ci Doc build passed. Will "Coke" Coleda 'give t/tabs.t a plan' 14:57
travis-ci.org/perl6/doc/builds/140042463 github.com/perl6/doc/compare/4432b...16a20fca2c
14:57 travis-ci left
timeless hrm 15:00
ok, so, my fix for id='s seems to work
but, not well enough 15:01
15:01 travis-ci joined
travis-ci Doc build failed. Will "Coke" Coleda 'add (failing) trailing whitespace test' 15:01
travis-ci.org/perl6/doc/builds/140045076 github.com/perl6/doc/commit/aeea5b2d5450
15:01 travis-ci left
BrokenRobot Xliff: received parameters are immutable by default. You want is copy if you don't want to affect caller or is rw if you do: perl6.party/post/Perl-6-There-Are-T...t-1#iscopy 15:02
Xliff: oh, nevermind. I misread what you were assigning
timeless when :(Str $ where {m/^trait\s+(\S+\s\S+)$/}) { 15:05
# Infix Foo
shouldn't that be # trait Infix Foo ?
15:05 brrt left
dalek c: 3d70d4a | coke++ | t/tabs.t:
Add (failing) tab test
 15:11
c: 4432bf4 | coke++ | t/tabs.t:
Skip more files
c: 19f2efe | coke++ | / (16 files):
tabs -> spaces
c: ec16a20 | coke++ | t/tabs.t:
give t/tabs.t a plan
c: b1493ab | (Aleks-Daniel Jakimenko-Aleksejev)++ | / (32 files):
Merge branch 'master' into tab-test
c: aa0e22f | (Aleks-Daniel Jakimenko-Aleksejev)++ | / (17 files):
Merge branch 'tab-test'
AlexDaniel omg… 15:12
dalek c: 297f94d | (Josh Soref)++ | htmlify.p6:
fix find-definitions trait case comment
 15:13
AlexDaniel missed one conflict
Xliff AlexDaniel: Been there. Done that. Yesterday. 15:14
Except I missed lots!
Fortunately... my own fork.
dalek c: 32cdea9 | (Aleks-Daniel Jakimenko-Aleksejev)++ | doc/Language/variables.pod:
Removed missed conflict
AlexDaniel git should've complained about it, not sure how I missed that 15:15
Xliff Well, the only thing I can do now is try to reproduce this bug in a smaller context. 15:16
Which might not bring up the error.
15:17 webstrand joined
[Coke] AlexDaniel: we have a lot of merged branches we could probably kill: "git checkout master && git branch -la --merged" 15:17
AlexDaniel [Coke]: yep 15:18
Xliff If anyone is curious looking at the entire context then you can check it out here: github.com/Xliff/p6-XML-LibXML.git and checkout branch 08findnodes. 15:19
15:20 travis-ci joined
travis-ci Doc build failed. Aleks-Daniel Jakimenko-Aleksejev 'Merge branch 'tab-test'' 15:20
travis-ci.org/perl6/doc/builds/140049602 github.com/perl6/doc/compare/2fbae...0e22f65193
15:20 travis-ci left, travis-ci joined
travis-ci Doc build failed. Josh Soref 'fix find-definitions trait case comment' 15:20
travis-ci.org/perl6/doc/builds/140049965 github.com/perl6/doc/compare/aa0e2...7f94d2a446
15:20 travis-ci left
timeless AlexDaniel: please say i can ignore that 15:21
oh, tabs,
timeless ignores
15:21 ssotka joined
Xliff timeless: *poof* wish granted! 15:22
timeless hrm, i think graphviz.org has fallen off the internet 15:23
timeless doesn't like pod
timotimo graphviz.org goes offline every few days :( 15:24
timeless timotimo: how do they manage that? 15:25
timotimo i haven't a clue
perhaps they don't have money? :(
BrokenRobot sudo apt-get install graphviz 15:27
timeless do pod objects have a dom-tree like thing? (it doesn't look like it)
timeless is tempted to do the running-list approach :/ 15:28
moritz timeless: yes 15:29
timeless moritz: what's it called?
moritz timeless: all the pod nodes derive from Pod::Block, which has a @.contents attribute 15:30
and that basically forms the document tree
timeless err
can a node find its parent?
or only its children?
moritz only its children 15:31
15:31 skids left
timeless not very dom-ish 15:31
(no cycles, less gc pain, but.. not dom-ish)
moritz DOM just means Document Object Model
basically an OO-approach to modeling a document
dogbert17 something is fishy with the doc builds. My last three commits have built ok according to travis but I can't see my changes on doc.perl6.org 15:33
gfldex dogbert17: not just yours. I got changes that didn't make it since yesterday. 15:35
timeless m: my @a=[1,2,3,4,5]; @a=@a[0..1]; @a.append([7]); say @a
camelia rakudo-moar 9b579d: OUTPUT«[1 2 7]␤»
[Coke] possible the doc build got hung. 15:36
AlexDaniel dogbert17: indeed
[Coke] tries to remember where that lives...
dogbert17 could travis be affected by github.com/perl6/doc/issues/545
AlexDaniel [Coke]: interestingly, last commit hash is reported in footer
timeless is there a better way to do what i did above? 15:37
AlexDaniel but other changes are not there
BrokenRobot Didn't travis announce build failures for the last few commits?
AlexDaniel BrokenRobot: last commit is fine 15:38
psch m: my @a = ^7; @a.splice(2, 7, 7); say @a
camelia rakudo-moar 9b579d: OUTPUT«[0 1 7]␤»
dogbert17 problem has been there since yesterday at the very least
AlexDaniel BrokenRobot: see here travis-ci.org/perl6/doc/builds
15:39 kerframil left
gfldex doc.perl6.org/build-log/build-2016...5+0000.log 15:39
build log looks ok. My guess would be util/sync 15:40
[Coke] AlexDaniel: yah, seems like it's working. for someone who says a commit is missing... which one?
AlexDaniel [Coke]: doc.perl6.org/language/variables.h...Declarator github.com/perl6/doc/commit/32cdea...bfd19d69df 15:41
[Coke]: still has “f ()”
dogbert17 [Coke]: github.com/perl6/doc/commit/39a5d14482 15:42
AlexDaniel [Coke]: doc.perl6.org/routine/invert github.com/perl6/doc/commit/79b1cc...3b751e4227 15:43
[Coke] AlexDaniel: just one example was fine. Verified.
AlexDaniel :)
[Coke] the doc build directory on hack is at 32cdea9 15:44
BrokenRobot That's latest
AlexDaniel and that's what I see in the footer
[Coke] the generated HTML in the build dir is wrong, however. 15:45
<span class="k">sub</span> <span class="n">f</span> () {
so we could probably duplicate this failure to update locally.
dogbert17 [Coke] is there a precomp directory there by any chance?
BrokenRobot dogbert17: yup
dogbert17 could we try wiping it? 15:46
BrokenRobot doesn't have perms to wipe it
dogbert17 :(
[Coke] we can try, but that isn't going to be a long term fix.
dogbert17 agreed 15:47
[Coke] there is also a "precompiled" directory
dogbert17 precompiled it is, my bad
[Coke] that's a more likely culprit, our manual precomp attempt
(rather than perl6's builtin precomp) 15:48
BrokenRobot :S
[Coke] /me: mv precompiled/ precompiled-broken 15:49
dogbert17 another thing with the doc builds is this comment in the travis log ... 15:51
240/240: doc/Type/X/TypeCheck/Splice.pod => type/X::TypeCheck::Splice
/bin/sh: 1: pygmentize: not found
15:52 perlpilot left
AlexDaniel isn't it ok? 15:52
15:52 perlpilot joined
[Coke] dogbert17: yes, that's fine. 15:52
the code first tries the executable, then tries inline::python verson
dogbert17 [Coke]: ok, had me worried for a sec there :)
[Coke] github.com/perl6/doc/blob/master/h...fy.p6#L859 15:53
that logic could probably be trimmed to avoid trying to call out to pygmentize all the time.
dogbert17 with some luck, a lot of changes should show up after the next commit ... 15:54
15:58 webstrand left, webstrand_ joined 15:59 domidumont left 16:00 pmurias_ joined 16:01 webstrand_ is now known as webstrand
timeless thx psch 16:03
16:03 pmurias left 16:11 huggable left 16:17 boyo_ joined 16:18 domidumont joined 16:19 huggable joined 16:21 salva left
BrokenRobot The wrap segfault goes away if I add `no precompilation` to the module with the wrap. 16:27
[insert joke blaiming nine for everything] :) 16:28
pyrimidine nine was responsible for the brexit, Zika virus, and Chernobyl (last one depending on nine's age) 16:30
16:31 dupek joined
BrokenRobot :) 16:32
gfldex pyrimidine: please note that nine is a time traveller
perlpilot gfldex: well, why didn't he bring back the perfect Perl 6 from the future?
pyrimidine gfldex: there may be times that nine wished he was a time traveller :) 16:33
gfldex it would spoil the fun of making the perfect language
pyrimidine nine++ # you get waaay too much guff for working on very difficult things 16:34
lizmat indeed, nine is a ten!
16:34 lizmat left 16:35 dvinciguerra joined 16:39 jast joined 16:41 jrusso joined, lizmat joined, jrusso is now known as MadcapJake
awwaiid "...it is vitally important not just to 16:41
have a complete metasystem, but to have fences that help guard the crossing
of metaboundaries." -- Alan Kay, lists.squeakfoundation.org/pipermai...17019.html 16:42
perigrin metafences. 16:43
awwaiid indeed
AlexDaniel
.oO( metafeces… )
 16:46
16:48 kurahaupo joined
lizmat
.oO( tower of poo )
 16:48
16:48 webstrand left 16:49 dj_goku_ left
nine pyrimidine: I barely remember Chernobyl, so who knows what exactly I did back then ;) 16:50
[Coke] nine: it's all the radiation damage.
nine That would explain quite a lot
Though I won't tell what exactly... would spoil the suspense ;)
dupek masak, hello 16:51
:))
16:51 cdg left
AlexDaniel [Coke]: it was much easier than I thought. 16:53
[Coke]: OK, so I have a little problem with “needs planning” github.com/perl6/doc/labels/needs%20planning 16:54
[Coke]: it feels like we need ｢big ass issue｣ label
[Coke]: note that it can be removed without because these issues already have a ｢build｣ or ｢docs｣ label 16:55
[Coke]: that being said, it feels like a different category… what do you think?
[Coke]: by the way, I've also relabeled some of the closed issues. There were too few of them 16:56
16:57 avenj joined
AlexDaniel [Coke]: ah, and one more problem: some issues are about search problems that are associated with the build. Should we label these as ｢search｣ or as ｢build｣? Or both? 16:57
16:58 Amnez777 joined 17:01 avenj left, avenj joined
timeless sighs 17:02
i'm sure i've seen the magic for making an array of x elements
python []x3 or something...
psch m: my @a[3]; @a[3] = "foo"; 17:03
camelia rakudo-moar 9b579d: OUTPUT«Index 3 for dimension 1 out of range (must be 0..2)␤ in block <unit> at <tmp> line 1␤␤»
timeless ok
17:04 dj_goku joined, dj_goku left, dj_goku joined
timeless psch: ok, this isn't working well 17:05
i have a list of n items
i want it to have j items where the last item has some other value
in js, i'd set the length to j and then stomp on the last value
splice doesn't do what i want if the array is too smal
l
m: my @a=1..7; my @b[9]; @b[8]=100; say @a || @b 17:06
camelia rakudo-moar 9b579d: OUTPUT«[1 2 3 4 5 6 7]␤»
timeless there's probably an or operator that would work, but i haven't found it
psch ...what?
what do you want to have as result there?
AlexDaniel 1..8 ? 17:07
ah
1..7, 100
grondilu just concatenate
use comma and slips
timeless m: my @a=1..7; my @b[9]; @b[8]=100; say @a[0..4] || @b
camelia rakudo-moar 9b579d: OUTPUT«(1 2 3 4 5)␤»
timeless slips?
grondilu oh hang on 17:08
AlexDaniel m: my @a=1..7; my @b[9]; say (|@a, 100)
camelia rakudo-moar 9b579d: OUTPUT«(1 2 3 4 5 6 7 100)␤»
AlexDaniel m: my @a=1..7; say (|@a, 100)
camelia rakudo-moar 9b579d: OUTPUT«(1 2 3 4 5 6 7 100)␤»
grondilu m: my @a = 1..7; my @b[9]; @b[8] = 100; say @a Z// @b
camelia rakudo-moar 9b579d: OUTPUT«(1 2 3 4 5 6 7)␤»
grondilu m: my @a = 1..7; my @b[9]; @b[8] = 100; say (|@a, |Mu xx *) Z// @b 17:09
camelia rakudo-moar 9b579d: OUTPUT«(1 2 3 4 5 6 7 (...))␤»
grondilu hum
jdv79 Zoffix: i only saw a ~4s parse time improvement from my last build to today. last was june 2 or so. 17:10
grondilu m: my @a = 1..7; my @b[9]; @b[8] = 100; say (|@a, |(Mu xx *)) Z// @b
camelia rakudo-moar 9b579d: OUTPUT«Type check failed in binding x; expected Any but got Mu (Mu)␤ in block <unit> at <tmp> line 1␤␤»
jdv79 still cool that its trending down
grondilu m: my @a = 1..7; my @b[9]; @b[8] = 100; say (|@a, |(Any xx *)) Z// @b
camelia rakudo-moar 9b579d: OUTPUT«(1 2 3 4 5 6 7 (Any) 100)␤»
jdv79 68s to 64s
17:11 sno left, spider-mario joined
grondilu m: my @a = 1..7; my @b[9]; @b[8] = 100; say @a <<//<< @b 17:11
camelia rakudo-moar 9b579d: OUTPUT«[1 2 3 4 5 6 7 1 2]␤»
grondilu m: my @a = 1..7; my @b[9]; @b[8] = 100; say @a >>//<< @b 17:12
camelia rakudo-moar 9b579d: OUTPUT«Lists on either side of non-dwimmy hyperop of infix:<//> are not of the same length␤left: 7 elements, right: 9 elements␤ in block <unit> at <tmp> line 1␤␤»
grondilu gives up
timeless that last error needs help
oh
grondilu no it's normal
timeless it's actually correct
psch m: my @a = ^7; my @b[9]; @b[8] = 100; @a <<[//=]<< @b; say @b # this needs help :P
camelia rakudo-moar 9b579d: OUTPUT«Non ast passed to WANTED: NQPMu␤Non-QAST node visited NQPMu␤Weird node visited: NQPMu␤Weird node in analyze: NQPMu␤===SORRY!===␤Unknown QAST node type NQPMu␤»
grondilu that was just me making a desperate attempt 17:13
psch similarly here
grondilu lol, "weird mode", never seen that before.
psch *n*ode :)
still not complete sure i understand correctly, but it does sound like "leave all defined elems alone, appends if more exists on the rhs"
grondilu oh yeah. (jeez I need to change glasses) 17:14
AlexDaniel what are you trying to do? I still don't understand
psch AlexDaniel: last line is my best guess vOv
timeless i'm trying to maintain a TOC
clipping elements deeper than my current point
and not worrying if i'm missing things higher
grondilu AlexDaniel: he kind of want to do Z// but greedily (if that makes sense)
timeless not a tree toc, a flat "who's my parent" view
psch so given @a = 0, 1, 2; and @b = Any, Any, Any, Any, 4; the result should be @a = 0, 1, 2, Any, 4
that's how i understood it 17:15
grondilu like from 1, 2, 3 and Any, Any, Any, pi how do you get 1, 2, 3, pi
timeless psch: right
[Coke] AlexDaniel: both, I guess. 17:16
and needs planning can be the opposite of LHF, sure.
grondilu do we need a greedy zip metaop? 17:17
AlexDaniel what means greedy?
do you mean roundrobin?
grondilu oh, maybe
17:17 firstdayonthejob joined, kid511 left
timotimo undersg: yourule ntand td prog dont 1 of threaderamminumber hreaderammind progg 17:17
AlexDaniel it's not an operator though
grondilu I don't know that word actually
AlexDaniel m: my @a = 1..4; my @b = 5..6; roundrobin @a, @b 17:18
camelia ( no output )
AlexDaniel m: my @a = 1..4; my @b = 5..6; say roundrobin @a, @b
camelia rakudo-moar 9b579d: OUTPUT«((1 5) (2 6) (3) (4))␤»
grondilu m: my @a = 1..7; my @b[9]; @b[8] = 100; say flat roundrobin @a, @b 17:19
camelia rakudo-moar 9b579d: OUTPUT«(1 (Any) 2 (Any) 3 (Any) 4 (Any) 5 (Any) 6 (Any) 7 (Any) (Any) 100)␤»
AlexDaniel stupid roundrobin has no :with parameter 17:20
timotimo robertoconcerto.blogspot.de/2013/0...t-bug.html - a fun read
grondilu by greedy I meant that it would act like the zip operator except that it would fill the shortest parameter with Any or something
AlexDaniel oh right #127992
rt.perl.org/Ticket/Display.html?id=127992
grondilu <1 2> GreedyZ~ <a b c> would give 1a 2b c 17:21
AlexDaniel m: say roundrobin <1 2>, <a b c>
camelia rakudo-moar 9b579d: OUTPUT«((1 a) (2 b) (c))␤»
AlexDaniel except that roundrobin has no :with so you can't really pass ~
grondilu yeah it's not a metaop 17:22
AlexDaniel it does not have to be a metaop to have :with
m: say zip <1 2>, <a b c>, :with([~])
camelia rakudo-moar 9b579d: OUTPUT«No such method 'CALL-ME' for invocant of type 'Str'␤ in block <unit> at <tmp> line 1␤␤»
grondilu ok *
AlexDaniel m: say zip <1 2>, <a b c>, :with(&[~])
camelia rakudo-moar 9b579d: OUTPUT«(1a 2b)␤»
17:22 addison joined
grondilu oh that's cool 17:22
AlexDaniel [Coke]: any better name than “needs planning”? It looks a bit weird 17:28
psch m: my @a = ^4; my @b = Any, Any, Any, Any, Any, 5; @a = |@a, |@b[+@a..*]; say @a 17:29
camelia rakudo-moar 9b579d: OUTPUT«[0 1 2 3 (Any) 5]␤»
psch been thinking waaay to complicated there 17:30
17:31 brrt joined
grondilu m: my @a = 1..7; my @b[9]; @b[8] = 100; say @a ... * Z// @b 17:31
camelia rakudo-moar 9b579d: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Only identical operators may be list associative; since '...' and 'Z//' differ, they are non-associative and you need to clarify with parentheses␤at <tmp>:1␤------> 037; my @b[9]; @b[8] = 100; say @a ..…»
grondilu m: my @a = 1..7; my @b[9]; @b[8] = 100; say (@a ... *) Z// @b
camelia rakudo-moar 9b579d: OUTPUT«(1 2 3 4 5 6 7 8 9)␤»
grondilu well that's not what I expected at all 17:32
17:32 fhorck joined
grondilu m: my @a = 1..7; my @b[9]; @b[8] = 100; say (@a ... *)[^10] 17:32
camelia rakudo-moar 9b579d: OUTPUT«(1 2 3 4 5 6 7 8 9 10)␤»
grondilu oh
psch m: my @a = 100, 200, 300; say @a ... 10
camelia rakudo-moar 9b579d: OUTPUT«()␤»
grondilu oh course
*of
m: my @a = 1..7; my @b[9]; @b[8] = 100; say (@a, Any ... *)[^10]
camelia rakudo-moar 9b579d: OUTPUT«Use of uninitialized value $b of type Any in string context␤Any of .^name, .perl, .gist, or .say can stringify undefined things, if needed. in block <unit> at <tmp> line 1␤===SORRY!===␤Method 'pred' not found for invocant of class 'Any'␤»
17:32 grassass joined
grondilu how do I make an infinite list from a finite one? should be easy 17:33
psch xx * vOv
psch goes back to poking the jvm issue 17:34
grondilu m: my @a = 1..7; my @b[9]; @b[8] = 100; say (flat @a, (Any xx *))[^10]
camelia rakudo-moar 9b579d: OUTPUT«(1 2 3 4 5 6 7 (Any) (Any) (Any))␤»
grondilu m: my @a = 1..7; my @b[9]; @b[8] = 100; say (flat @a, (Any xx *)) Z// @b
camelia rakudo-moar 9b579d: OUTPUT«(1 2 3 4 5 6 7 (Any) 100)␤»
AlexDaniel [Coke]: to me it sounds like it creates a wall for contributors
[Coke]: “oh, it needs planning… I'm not going to hack on it!” 17:35
BrokenRobot
.oO( Planning for what? )
AlexDaniel BrokenRobot: github.com/perl6/doc/issues?q=labe...lanning%22 17:38
17:39 gtodd joined
AlexDaniel it's not ｢refactoring｣, it's not necessarily ｢plans｣… 17:39
gtodd must ... have ... damian's ... vim config ...
even though it will not make me smart muscular and funny ...
BrokenRobot AlexDaniel: well, what has to be done for "needs planning" tickets. Who is doing the planning? IMO, "feedback needed" or something like that is more apropo
m: 'foo' does role { method uc { say "We're in! Someone's trying to uppercase {self}"; nextsame; } }; say 'foo'.uc; 17:40
camelia rakudo-moar 9b579d: OUTPUT«We're in! Someone's trying to uppercase foo␤FOO␤»
17:40 brrt left
AlexDaniel BrokenRobot: it would be nice to have a label for “big” issues 17:40
BrokenRobot Is the reason the above doesn't cross module boundaries is because each file has it's own namespace so a "foo" in a different file is a different object? 17:41
17:41 brrt joined
BrokenRobot AlexDaniel: the label is useless if no one bothers to act and think on the big issues. "needs planning" tells me there's something blocking progress on the issue and someone somewhere is meant to do something to unblock it. It's entirely unclear who that someone should be. 17:43
AlexDaniel BrokenRobot: yep, that's why I dislike it
｢think big｣ ? :) 17:44
andreoss Configure.pl requires perl 5.10, but modules it uses were not in CORE until 5.014
BrokenRobot I stand by my proposal :)
andreoss in rakudo
AlexDaniel BrokenRobot: ｢feedback needed｣? but this kinda applies to small issues as well 17:45
psch how about "someone has to make impactful design decisions so this can be solved" 17:46
'cause that's what i think "needs planning" means
gtodd I can haz the vim! youtu.be/9u6O0dLuqhI
psch but it's a bit clearer and all :P
BrokenRobot psch: that's bad, 'cause it places the decision on a non-existant entity :) 17:47
andreoss: what would be a good way to resolve it?
psch fair enough
AlexDaniel bump it to perl 5.14?
BrokenRobot :S
andreoss AlexDaniel: yep, consider my PR
BrokenRobot Just over a module?
-1 from me 17:48
AlexDaniel v5.10 is ancient anyway
just like v5.14 is
gtodd maybe plain "feedback needed" and "design feedback needed" are different enough
BrokenRobot That's the only thing I have on one of the hosts I manage
[Coke] AlexDaniel: if the goal isn't to create a wall there, what is it? 17:49
I'm fine with "BIG" to contrast with "LHF"
AlexDaniel [Coke]: to make it easier to track big issues (as these tend to have lots of useful discussions) 17:50
grondilu gtodd: Damian's talk about Vim as a Perl IDE is nice but how well does it work in Perl 6?
AlexDaniel [Coke]: ｢BIG｣ or ｢big｣? :)
[Coke] don't care 17:51
I feel the same way about this tag as I do about LHF
gtodd grondilu: he uses vim with perl6 too afaict
grondilu yeah but he did not show anything in Perl 6 in this talk 17:52
I did not look at his tarball though
17:53 acrussell left
gtodd grondilu: he must be good at *.vim vimscripting ... 17:53
grondilu he sure is
I'm also intrigued by the fact that he does not use syntax highlighting. Any one here does not either?
BrokenRobot grondilu: mst doesn't
I tried it once... I think I could get used to it, but I don't really care one way or another. I think syn-highlkighte code looks prettier :) 17:54
grondilu I can understand that in the end it's a distraction but it's also useful to spot syntax errors.
AlexDaniel grondilu: yeah. The part about no syntax highlighting is a bit dumb :/ 17:55
AlexDaniel hides
gtodd grondilu: I think flashng informative stuff at the bottom of the buffer and automating the inserting of syntax bits makes it ieasier to avoid ...
grondilu though to be fair if I had a Vim script that could detect syntax errors and highlight them in real time I would maybe consider not using syntax highlighting. 17:56
[Coke] I don't particularly care for syntax highlight.
grondilu also sometimes a code can be correct but not at all what you wanted because you forgot some delimitor or something. For instance with here-doc. Syntax highlighting is very reassuring for that. 17:58
gtodd in vim the colorful syntax flashes so much that it exercises bugs in terminfo/ncurses (on local or remote machines and/or inside tmux/screen for extra fun) and scrambles my screen so Ctrl-L required so often I turn it off. Or set an event to autocommand :redraw!
Vim::X adds coolness for both vim and perl ... metacpan.org/pod/Vim::X ... would be neat if perl6 could be linked to vim somehow so writing extensions in perl6 would allow for tricked out vim bells/whistsl. Might be a way to gain mind share too ... "want to code in ruby with vim but not learn vimscript ? ... learn perl6 instead !" :) 18:03
18:03 boyo_ left
pmurias_ grondilu: I don't use syntax highlighting atm, I turn it one when I bored and want things to be more colorfull 18:05
18:05 acrussell joined 18:06 pmurias_ is now known as pmurias
gtodd OTOH vimeo.com/63116209 <-- great video about asynchronous dispatch with dispatch.vim plugin :) 18:06
there's more to vim than colour ;) 18:07
18:12 kalkin-_ joined
kalkin-_ hi 18:12
BrokenRobot \o
kalkin-_ I can't find any documentation how the module versioning should work
I tried S11, but i can't find there any specification of <ver:123123> keyword
BrokenRobot That'll be in... um S22? 18:13
Well, there are some mentions of it, but it's a historical document and may not match reality. 18:14
kalkin-_ BrokenRobot: I'm currently looking at Software::License. It has a class Software::License:ver<0.05>
BrokenRobot Yeah, that's how you specify the version for a module. 18:15
m: say 0.05 before 6.c
camelia rakudo-moar 9b579d: OUTPUT«Method 'c' not found for invocant of class 'Int'␤ in block <unit> at <tmp> line 1␤␤»
BrokenRobot m: say v0.05 before v6.c
camelia rakudo-moar 9b579d: OUTPUT«True␤»
kalkin-_ is 'ver' obsolete? and i should use meta data from S22, with full 'version<1.2.3>'?
BrokenRobot Oh
m: class Foo:ver<2.3> {} 18:16
camelia ( no output )
BrokenRobot m: class Foo:version<2.3> {}
camelia rakudo-moar 9b579d: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Cannot use adverb version on a type name (only 'ver' and 'auth' are understood)␤at <tmp>:1␤------> 3class Foo:version<2.3>7⏏5 {}␤ expecting any of:␤ generic role␤»
BrokenRobot kalkin-_: no, :ver<> seems to be the right one
kalkin-_ That's a very helpful error output.
BrokenRobot: thanks
18:22 kalkin-_ is now known as kalkin- 18:24 [particle] joined
kalkin- As far as I understand a module can only deliver one version, because the version is parsed from META6.json, right? I can't have a module which provides multiple version implementations, can I? 18:26
18:27 brrt left
gfldex grondilu: i switched off syntax highlighting for perl * because I found it unreliable. My brain is lazy and will fool me in believen a thing is a different thing or not there at all if it is not presented in the right colour. 18:28
no colours will solve that problem 18:29
18:29 captain-adequate joined, andreoss left
gfldex having wrongness in red has lots of appeal tho 18:29
18:30 cdg joined
pmurias grondilu: it would be great to have Perl 6 IDE support, it would be best if it was split into an editor agnostic part so it could be shared by the vim/emacs/atom/etc. peopel 18:31
18:34 wamba joined 18:35 [particle] left, Actualeyes left 18:36 [particle] joined 18:39 wtw left 18:42 brrt joined 18:48 itaipu joined 18:49 sufrostico left 18:50 pmurias left 18:52 pmurias joined, grondilu left 18:53 grondilu joined, grondilu left, grondilu joined 18:57 dvinciguerra_ joined
BrokenRobot What is (| is raw) is a signature? 18:59
19:00 dvinciguerra left
BrokenRobot m: say &say.signature 19:02
camelia rakudo-moar 9b579d: OUTPUT«(| is raw)␤»
BrokenRobot And I guess my real question is how to get args from it? 19:03
&say.wrap: -> (| is raw) { put "We're in! Someone's trying to print {...}"; nextsame; }; say 'foo';
lizmat | is that it doesn't touch the parameters basically, you see that in proto's 19:04
BrokenRobot Hum. Does that mean there's no way to wrap a &say and use the pass args, since the wrap must much the signature?
*passed 19:05
lizmat must the wrap match the signature ?
if you wrap the proto, you wouldn't need to /
?
which is what &say is, no?
m: say &say.dispatcher 19:06
camelia rakudo-moar 9b579d: OUTPUT«(Mu)␤»
lizmat hmmm...
m: proto foo {}; say &foo.dispatcher
camelia rakudo-moar 9b579d: OUTPUT«(Mu)␤»
BrokenRobot m: &say.wrap: -> (*@args) { put "We're in! Someone's trying to print @args[]"; nextsame; }; say 'foo';
camelia rakudo-moar 9b579d: OUTPUT«We're in! Someone's trying to print ␤Attempt to return outside of any Routine␤ in block <unit> at <tmp> line 1␤␤»
lizmat m: say &say.is_dispatcher
camelia rakudo-moar 9b579d: OUTPUT«True␤»
BrokenRobot Oh, I guess the sig needs to match for nextsame() 19:07
lizmat there you go
afk&
19:07 lizmat left
BrokenRobot looks up what is_dispatcher stuff is about 19:07
no docs :/ 19:08
19:09 mohae_ left, FROGGS joined, mohae joined
BrokenRobot Does anyone know what "there you go" meant? :) 19:09
19:09 lizmat joined
BrokenRobot feels like an idiot 19:09
lizmat sorry 19:10
gfldex BrokenRobot: github.com/rakudo/rakudo/search?ut...dispatcher
lizmat didn't make you want to feel like an idiot
gfldex no touchy touchy :)
AlexDaniel oh, that weird feeling when you see somebody talking about how great his emacs/vim config is, but whenever he needs to delete the last word he presses backspace 5 times… maybe that's not an issue to some people, but to me it feels like something went really wrong with that config… 19:11
geekosaur sometimes it's not config
I still have trouble using PC-style editing keys vs. emacs keys, so I end up arrowing to the end of a line instead of hitting End 19:12
BrokenRobot gfldex: you mean it's an internal thing? 19:14
lizmat BrokenRobot: no idea what's going on... and need to afk&
parabolize I backspace pretty often in vim. 19:15
BrokenRobot I was just trying to figure out how to wrap &say while still being able to call it from within the wrap and nextsame fails
parabolize BrokenRobot: its tested in S12-introspection/methods.t
BrokenRobot That doesn't really help me since I don't know what a dispatcher is :) 19:17
Or why it's relevant to the wrap thing :P
gfldex the dispatcher is the thing that picks the right multi candidate
or does some other magic with methods
parabolize is there a better way in vim? Esc c b seems a little unwieldy
geekosaur control-w is erase word backwards 19:18
perlpilot (and has been since ... forever) 19:19
parabolize oh, that is nice
BrokenRobot m: &say.wrap: -> (*@args) { put "We're in! Someone's trying to print @args[]"; say(@args); }; say 'foo';
camelia rakudo-moar 9b579d: OUTPUT«(timeout)We're in! Someone's trying to print ␤We're in! Someone's trying to print ␤We're in! Someone's trying to print ␤We're in! Someone's trying to print ␤We're in! Someone's trying to print ␤We're in! Someone's trying to print ␤We're in! Som…»
FROGGS m: &say.wrap: -> (*@args) { put "We're in! Someone's trying to print @args[]"; ::CORE::<say>(@args); }; say 'foo';
camelia rakudo-moar 9b579d: OUTPUT«We're in! Someone's trying to print ␤Cannot find method 'Nil' on object of type Array␤ in block <unit> at <tmp> line 1␤␤»
FROGGS m: &say.wrap: -> (*@args) { put "We're in! Someone's trying to print @args[]"; ::CORE::<&say>(@args); }; say 'foo'; 19:20
geekosaur it's an old BSD-ism, and quite a few younger folks simply don't know it's there
FROGGS m: &say.wrap: -> (*@args) { put "We're in! Someone's trying to print @args[]"; CORE::<say>(@args); }; say 'foo';
camelia rakudo-moar 9b579d: OUTPUT«(timeout)We're in! Someone's trying to print ␤We're in! Someone's trying to print ␤We're in! Someone's trying to print ␤We're in! Someone's trying to print ␤We're in! Someone's trying to print ␤We're in! Someone's trying to print ␤We're in! Som…»
rakudo-moar 9b579d: OUTPUT«We're in! Someone's trying to print ␤Cannot find method 'Nil' on object of type Array␤ in block <unit> at <tmp> line 1␤␤»
AlexDaniel hm, what is ctrl+backspace in vim?
does nothing here, I'm wondering if it actually does nothing? 19:21
geekosaur if you're running in a terminal, whether it does anything is up to the terminal 19:22
some of them don't send anything for ctrl-backspace
AlexDaniel ah
right
19:23 zakharyas joined
BrokenRobot Figured it out: 19:23
m: &say.wrap: sub (|c) { put "We're in! Someone's trying to print {c}"; nextsame; }; say 'foo';
camelia rakudo-moar 9b579d: OUTPUT«We're in! Someone's trying to print foo␤foo␤»
19:25 mantovani joined 19:26 darutoko left 19:32 zakharyas left 19:44 mantovani left
kalkin- Say i have a file lib/Foo/Bar.pm6 Contaning class Foo::Bar, then what do i import when i do use Foo::Bar? The class or the module? 19:47
timotimo "use" will use the name you give it to locate the file, then it imports everything in it 19:48
so you'll end up with a Foo::Bar you can then use 19:49
that's at least how i understand it :)
kalkin- i see
moritz note that Foo::Bar is resolved through GLOBAL:: though, not through imports 19:50
Xliff \o/
kalkin- camelia: class Foo::Bar:ver<0.4> {}; class Foo::Bar:ver<0.3> {};
Xliff I was able to reproduce my error in a simple use case
gist.github.com/Xliff/96e48eaad469...80a73a918a
kalkin- camelia: class Foo::Bar:ver<0.4> {}; class Foo::Bar:ver<0.3> {}; say "Yeah!" 19:51
m: class Foo::Bar:ver<0.4> {}; class Foo::Bar:ver<0.3> {};
camelia rakudo-moar 9b579d: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Redeclaration of symbol Foo::Bar␤at <tmp>:1␤------> 3Bar:ver<0.4> {}; class Foo::Bar:ver<0.3>7⏏5 {};␤ expecting any of:␤ generic role␤»
Xliff So if anyone can tell me rakudo can't manipulate the returned struct, please let me know! 19:52
kalkin- My assumption is that class Foo::Bar would be different if they have different versions
FROGGS Xliff: that rings a bell, let me think...
I bet a nqp::bindattr_o will do... 19:54
kalkin- So how do i defined two different versions of the same class, by reusing the :ver<1.2.3> smiley?
perlpilot kalkin-: normally you wouldn't define them in the same file. 19:55
FROGGS kalkin-: or at least not in the same scope
Xliff FROGGS: :-O 19:56
So THAT explains why you do that!
LOL! Mystery solved!
kalkin- so how would i split it up?
FROGGS :D
Xliff Well.... gotta test it first.
FROGGS Xliff: I've not enough brains right now to explain why that happens, or how to properly solve it
BrokenRobot New blog post ": perl6.party/post/Exploiting-Perl-6-...ency-Chain 19:57
New blog post "Exploiting Perl 6 Code From Down The Dependency Chain": perl6.party/post/Exploiting-Perl-6-...ency-Chain
.oO( fat fingers :( )
19:57 sno joined
kalkin- should i overwrite somehow EXPORT::DEFAULT ? 19:58
Xliff BrokenRobot: Oooh! You suffer from those, too? 20:00
I swear, my fingers need a diet!
BrokenRobot Xliff: the fingers is the last part of my body that is in dire need of a diet... but yeah :)
Xliff BrokenRobot: Oh... that sounds disturbingly familiar. 20:02
We should start a club =)
BrokenRobot :)
perlpilot BrokenRobot: I'm not sure if you deserve a ++ or a -- for that article ;)
BrokenRobot perlpilot: += 0 ? :) 20:03
perlpilot BrokenRobot++ for increased awareness.
Xliff FROGGS: You mean this? -- github.com/perl6/nqp/blob/master/d...n#bindattr 20:04
My nqp-foo is non-existant.
parabolize wants a Hands-On Workshop: Weatherapp (Part 4)
20:05 kalkin- left
FROGGS Xliff: github.com/FROGGS/p6-XML-LibXML/bl...14N.pm#L25 20:05
Xliff: and yes, the doc is right
20:06 stevieb9 left, kalkin- joined
kalkin- did I miss something? 20:07
timotimo no, but we missed you 20:09
perlpilot we'll aim better next time ;)
kalkin- :) 20:10
Xliff FROGGS: LOL! That's exactly what I was trying to do. ^^ 20:11
20:12 _mg_ joined
timotimo if only i had known this yesterday ... 20:13
AFL_SKIP_CRASHES is a thing that i would really have needed
kalkin- Any examples of modules using different :ver<1.2.3> smileys? 20:14
20:15 sufrostico joined 20:18 ZoffixMobile joined, huggable left
ZoffixMobile I have a Gist...something in the ecosystem that has a :ver<> on it 20:19
Lately, I stopped using them since our installers seem to go just by META file anyway
And for a good reason. a dist should have one version not different files with different versions, IMO 20:20
20:22 TEttinger joined, acrussell left
FROGGS aye 20:24
timotimo so
i bet peopl in here know bash scripting
kalkin- Ok so it would be fine for me to release Foo::Bar which has in META6.json set version: 0.1, and then change API release the same way via META6.json the version 0.2. Later in some project i just define depends in META6.json to Foo::Bar:ver<0.2> and panda/zef will do the magic?
timotimo i'm for-looping over a thing, and i want to be able to abort the whole thing before it runs to completion
perlpilot timotimo: break
timotimo so, || break aftre the program?
i'm not 20:25
not sure it exits with nonzero when i ctrl-c it
ZoffixMobile kalkin-, yeah
kalkin- ZoffixMobile: yeah, like in theorey, yeah, or in practice? :)
will panda/zef do the magic?
timotimo OK, it's 1 when i ctrl-c
ZoffixMobile kalkin-, yeah in practice :) I think they may be ignoring :ver in classes altogether. 20:27
kalkin- ZoffixMobile: cool thanks
now let's just hope we never get in to a Left::Pad:ver<0.2> being removed from package repo :) 20:28
ZoffixMobile :)
mst don't worry, the people behind CPAN have a spine.
perlpilot Eventually, I would guess we'd obtain tools (like Dist::Zilla maybe?) so that we don't have the version information in two places
20:29 _mg_ left
kalkin- we need more p2p file storage 20:29
something like freenet or gnunet
ZoffixMobile Yeah, github is a temporary thing
FROGGS (for use as the module ecosystem)
ZoffixMobile Though it doesn't look like anyone has will/time to work on a better system :)
20:29 domidumont left
FROGGS ohh, I was working on using cpan... 20:30
kalkin- dreams some times of a panda/pip/gem like installer which is backed by something bittorrent like
20:36 acrussell joined, ZoffixMobile left 20:37 lichtkind joined 20:41 itaipu left 20:46 itaipu joined 20:48 acrussell left 20:50 _ramix_ joined, addison left
perlpilot anyone remember the other day (last week?) when I was talking about overhearing two of my coworkers mention that they wish they could use Perl 6? 20:56
AlexDaniel yea 20:57
20:57 itaipu left
perlpilot One of them now makes it point to say loudly in my direction "... if only it [Rakudo] were *fast* enough." 20:57
21:01 ptolemarch joined 21:04 _ramix_ left
mspo kalkin-: what does that gain? 21:04
21:05 inokenty left 21:09 FROGGS left 21:11 brrt left 21:12 sno left 21:13 sno joined 21:15 addison joined, mcmillhj left, wamba left 21:16 ZoffixMobile joined
ZoffixMobile perlpilot, tell them to get a faster computer :p 21:16
pmurias mst: what is the proper way to prevent a left-pad style situation?
ZoffixMobile might get to play around with a 30-core beast this weekend
kalkin- mspo: btw pmurias comment is one of benefits of censore free environment, which p2p networks try achieve 21:17
mst pmurias: herein the perl response to the last legal threat we got over a module name: p3rl.org/Nagios::Plugin 21:18
kalkin- mspo: also i keep to my heart the motto “Never trust the authorities” and a central repo is a “fashist” regime
take the fashist regime with a grain of salt :) 21:19
mspo who controls namespaces and stuff?
majority rules?
ZoffixMobile I still laugh at this "Honestly, I'd've put it in the public domain but the idea of somebody restributing it under the WTFPL amuses me too much."
kalkin- mspo: hash rules
ZoffixMobile mspo, which namespaces? :) 21:20
mspo, in Perl 6 multiple authors can have the same module name :)
21:21 inokenty joined
kalkin- And the authors name could you map to a gpg fingerprint. So you fetch the git commit tagged with version v1.2.3 and signed by the authors fingerprint 21:24
signed by the authors gpg key of course
21:25 dupek left
pmurias mst: wouldn't replacing left-pad with a package explaining the situation still lead to breakage? 21:27
21:27 abraxxa left
kalkin- pmurias: i think the only valid replacement would be any black box implementing exactly the same api 21:28
but even that would be questionable
mst pmurias: left-pad was only deleted because they nuked one of the author's other packages in response to a legal threat 21:29
pmurias: plus he wouldn't be able to delete from backpan, and you'd expect pause admins to re-upload a tombstone release that works while warning, I think (certainly I'd advocate for such a thing is somebody deleted a significant dependency) 21:30
21:31 kid51 joined
pmurias mst: the tombstone Nagios::Plugin authors seems to have moved to Monitoring::Plugin, I assume the left-pad author wouldn't agree to something like that 21:33
AlexDaniel [Coke]: there's one more category in issues. “duplicate of #155” 21:34
21:34 ZoffixMobile left
mspo I mean what if I wanted to take over a package 21:34
what's stopping me? 21:35
AlexDaniel [Coke]: I can see 9, but there are probably more
perlpilot mspo: why do you think you need to be stopped? ;-) 21:37
mst pmurias: they'd already moved, or I'd've played it a different way namespace wise 21:38
mspo perlpilot: because I'm a government actor who can easily create a majority stake in a "private" network
perlpilot: see tor
21:40 rgrinberg left
pmurias I trust modules from cpan much more than ones downloaded from tor ;) 21:45
21:46 wamba joined 21:51 ptolemarch left
Xliff m: dd ^4.cache 21:55
camelia rakudo-moar 9b579d: OUTPUT«Potential difficulties:␤ Precedence of ^ is looser than method call; please parenthesize␤ at <tmp>:1␤ ------> 3dd ^47⏏5.cache␤^1␤»
Xliff m: dd (^4).cache
camelia rakudo-moar 9b579d: OUTPUT«(0, 1, 2, 3)␤»
kalkin- but even that would be questionable
ups
wrgon channel
Xliff =)
I do that all the time. 21:56
jnthn m: dd ^4 .cache
camelia rakudo-moar 9b579d: OUTPUT«(0, 1, 2, 3)␤»
jnthn also :)
21:57 perlpilot left
Xliff What is the point of .cache? 21:57
jnthn Note much in that case :) 21:58
It's for use on Seq primarily
21:59 dvinciguerra_ left
jnthn m: my \a = ^4 .map(* + 2); .say for a; .say for a; 21:59
camelia rakudo-moar 9b579d: OUTPUT«2␤3␤4␤5␤This Seq has already been iterated, and its values consumed␤(you might solve this by adding .cache on usages of the Seq, or␤by assigning the Seq into an array)␤ in block <unit> at <tmp> line 1␤␤»
jnthn m: my \a = ^4 .map(* + 2); .say for a.cache; .say for a.cache;
camelia rakudo-moar 9b579d: OUTPUT«2␤3␤4␤5␤2␤3␤4␤5␤»
jnthn A Seq is a 1-shot iterable that discards values (important for building lazy pipelines); .cache lets you get a List that caches the values so you can obtain them multiple times 22:00
22:01 chrsteel joined
Xliff Ah! 22:01
chrsteel rakudo: react { whenever Supply.interval(1, 0) -> $v { say $v; done if $v >= 1; } } 22:02
camelia rakudo-moar 9b579d: OUTPUT«0␤1␤»
chrsteel rakudo: react { whenever Supply.interval(1, 0) -> $v { say $v; done if $v >= 0; } }
22:02 pmurias left
Xliff jnthn: But you wouldn't need .cache if you were assigning a Seq to an Array would you? 22:02
camelia rakudo-moar 9b579d: OUTPUT«(timeout)0␤»
Zoffix chrsteel, that was reported as a bug last night. I'm guessing you coming from the same workshop or class as the other person :)
chrsteel heh, just last night? :-) 22:03
I came across this bug just now.
Zoffix Yeah: irclog.perlgeek.de/perl6/2016-06-23#i_12723594 22:04
Though it appears to have existed for ages
chrsteel Ah! Funny... good to know I'm not the only one.
jnthn Xliff: Correct, assignment produces the values and stores them into the array.
Xliff: Though it'll do that production lazily (that is, on-demand as you access array elements) if the Seq is marked lazy 22:05
Xliff Gotcha. Thanks for the explanation!
I've only found this description via docs.perl6.org: docs.perl6.org/routine/cache#role_...ndFailover 22:06
It's not documented ion Seq or Positional
s/ion/in/
Although from grepping the code, it looks like it's defined in Any 22:07
22:08 chrsteel left
jnthn Please file a doc issue if it's not suitably documented or easy enough to find. :) 22:08
jnthn wanders off for some rest...'night
timotimo gnite jnthn :) 22:11
22:12 rindolf left, cpage_ left, cpage_ joined 22:13 cpage_ left 22:14 kanishka joined
Xliff \o/ 22:35
Fixed bug!
Zoffix \o/ 22:38
timotimo ooooh <3
Zoffix 🤘🤘🤘 www.youtube.com/watch?v=uVpU-OXrg1c 22:40
timotimo cat gave the back of my hand and my arm a proper cat wash %)
it was very scratchy 22:43
22:43 cpage_ joined
vendethiel- $ cat scratch 22:43
timotimo the day i started the fuzzing, / started being filled up a little bit over time 22:44
but ... why?
22:46 cpage_ left 22:47 kurahaupo left 22:48 kurahaupo joined 22:52 fhorck left 22:55 sufrostico left 23:00 firstdayonthejob left
Zoffix AlexDaniel, what does your "So?" comment mean? :) 23:00
(RE: github.com/perl6/doc/issues/167#is...-148887096 ) 23:01
AlexDaniel Zoffix: anything stopping us from renaming everything to .pod6?
Zoffix Don't think so ¯\_(ツ)_/¯ 23:02
AlexDaniel just trying to breathe some life into these old tickets :)
Zoffix AlexDaniel++
And cheers for finding a ticket where I say "I'm pretty new to the community" :P
AlexDaniel :D 23:03
23:03 sno left, sno joined
AlexDaniel by the way, we now have completely different set of labels in the doc repo 23:03
Zoffix sweet
AlexDaniel and I've even tagged some older issues
… practically all of them. Not that there was any need to do so, but it should give a better idea of what these labels are supposed to mean 23:04
23:04 cognominal left 23:05 fhorck joined
ShimmerFairy AlexDaniel++ # bug wrangling is always good :) 23:05
23:05 iH2O joined 23:07 iH2O left 23:08 [Sno] joined, sno left
timotimo i wonde if with "only method foo" you can get over the "already has a method named ..." when augmenting a core class 23:09
23:10 cpage_ joined
timotimo nope. 23:11
23:14 cpage_ left 23:17 jjido joined, girafe left
Zoffix m: class Foo { only method foo {} }; 23:18
camelia ( no output )
Zoffix :o
AlexDaniel what is only? 23:21
23:21 cpage_ joined
Zoffix ¯\_(ツ)_/¯ 23:23
AlexDaniel m: only sub what { say ‘what’ }
camelia ( no output )
timotimo i thought it was there to override "multi" from a parent class 23:26
23:29 cpage_ left
Juerd If I strace perl6, it turns out that it does an epoll_wait() after every print... Why's that? 23:31
timotimo because we're using libuv and its asynchronous stuff for printing ATM 23:32
kalkin- So if I want to specify an "interface" defining a new method with specfic arguments and some fields. The perl6 way would be to create a role, right? 23:33
Juerd timotimo: All prints, always?
kalkin- (btw the new method should be overwritten by the other implementations)
timotimo well, if you NativeCall into libc, then not 23:34
Juerd Heh, I might try that :)
23:36 cpage_ joined
Juerd Heh, write over nativecall is *slower* :) 23:37
$ perl6 -e'use NativeCall; sub write(int32, Str, int32) returns int32 is native { * }; my $x = now; write(1, "hoi", 3) for 1..1000000; note now - $x; $x = now; print "hoi" for 1..1000000; note now - $x;' >/dev/null
7.2371556
4.1859923
timotimo yeah, the overhead is still pretty bad
nativecall overhead, that is 23:38
Juerd $ perl -e'use Time::HiRes qw(time); my $x = time; print "hoi" for 1..1000000; warn time()-$x' >/dev/null 23:39
0.0617289543151855 at -e line 1.
0.17 with $| set to 1.
timotimo what does that do? 23:41
kurahaupo autoflush 23:42
a syscall for every print
23:43 ssotka left 23:46 ssotka joined 23:49 RabidGravy left
timotimo ah, ok 23:49
yeah, we're doing multiple syscalls for every print :) 23:50
dalek c: d5ba244 | (Jan-Olof Hendig)++ | doc/Type/Semaphore.pod:
Fixed typo and two missing C<>'s
 23:51
timotimo oh
dmesg is full of messages about the segfaults of my moar processes 23:52
the ones i'm fuzzing
it's not supposed to reach the system's logger ... that must be why / is filling up
i should probably not let it run for much longer? 23:54
23:55 cdg left
timotimo *sigh*, i wanted to see it get through the whole queue once at least :( 23:57
23:58 spider-mario left 23:59 jjido left

Powered by Cro and the Raku® Programming Language.

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

Thank you!