»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.
TimToady samcv: be sure your script pulls the latest; I just pr'd Grammar::Debugger to add some of those fields, for instance 00:01
samcv cool
yeah ofc
TimToady always figures it's better to say a thing too many times than too few :)
samcv yep 00:02
00:03 cpage_ joined
TreyHarris It looks like most well-behaved classes ignore any named argument they don't recognize, but some do barf. When you only define BUILD, though, that's the behavior you get, so that's what should be considered default behavior, yes? (Wondering how to write the test "I implement this flag" before I get to the behavior that can exercise it, but I guess I can't and should just accept that.) 00:05
samcv ok cool got my automated script to scrape all these that RabidGravy said failed the META license test 00:07
so many variations
also idk what 'perl' license means. perl 5 ? perl 6? heh
ugexe "yes" 00:08
samcv ;)
here's the list gist.github.com/samcv/9177c43f2a78...54c63d28e5
ugh so many saying just "Artistic" and many of those i've seen have 2.0, but. no way to automate that without dling any license file in the project and checking that. 00:09
ugexe no urls then i assume?
samcv url's?
00:09 mcmillhj joined
samcv to the meta's? well. 00:09
ugexe as the license some people use a url to the license text
samcv my script has them. that's just the printout
anarchy! 00:10
TimToady well, anything that /Artistic.*2/ can probably change to 'Artistic-2.0', which I believe is the standard espoused by the website
samcv yeah
exactly
ugexe i can't filter urls effectively, so i'm all for some ENUM of values 00:11
samcv hmm?
TimToady perl probably means it was originally "same as Perl itself", which at the time would only have been Artistic
samcv yeah. though. many say Artistic. but mean Artistic 2.0. but i'm sure some mean Artistic 1.
TimToady well, Artistic | GPL actually
it was a quantum superposition
samcv :) 00:12
Artistic-1.0-Perl OR GPL-1.0
TimToady but since Perl 6 is still "perl", we could try to tweak them to Artistic 2.0, with a love note saying they should do something else if this is unacceptable
samcv using SPDX spec notation
hahaha
00:13 Actualeyes joined
TreyHarris I'm trying to port a package with many Perl 5 classes with named boolean switches in the constructors. Since they were standard blessed hashes, their failure to define an accessor method for those flags means doesn't register any inten; I can define the switches as public attributes or as hidden attributes only settable at constructor time. What makes sense to me is to make them all attributes with "is ro" if 00:15
their behavior can't be changed usefully after construction, 'is rw' if they can. That's the right way to do it, yes?
s/means.*intent/doesn't indicate any intent/
00:17 japh__ left, japh__ joined 00:18 iyra left
samcv ugexe, updated the gist to have url's and url's to meta files gist.github.com/samcv/9177c43f2a78...54c63d28e5 00:18
00:19 AlexDaniel left
TimToady "Artistic-1.0-Perl OR GPL-1.0" is not quite right, since the README itself allows you to use any subsequent version of the GPL 00:20
samcv hm
TimToady but then, I think GPL-1.0 also contains that provision
so maybe it doesn't matter
samcv isn't that in the GPL-1.0?
then it'd be Artistic-1.0-Perl OR GPL-1.0 WITH foobarclausename
or something like that
TreyHarris Also: is there any basic rule of thumb about when to replace underscores in P5 names with hyphens? 00:21
samcv uhm. do it if you want to make it more perl6y?
TimToady well, Perl always added a lot of foobar clauses to the GPL :)
samcv you don't have to, not really any rules
TimToady, dev.perl.org/licenses/ it seems to just say GPL 1 or Artistic 00:22
TreyHarris samcv: I'm just trying to do a "straight but Perl 6-ish" port. I guess Test::More -> Test is the canonical case of that, so the answer is "yes" on underscores to hyphens and "yes" on attributes
samcv i usually used dashes for perl 6 stuff and sometimes name variables with underscores when they are going to be used in non perl 6 programming languages. but that's just me 00:23
TimToady we generally reserve underscores for internal names in p6
samcv do we though?
hm
TimToady well, "reserve" is perhaps too strong
we encourage hyphens for public-facing APIs, let's put it that way 00:24
samcv yeah
00:24 japh____ joined
TimToady the presence of any underscores can be taken as a mild warning of <mumble> 00:25
TreyHarris TimToady: undrstd. 00:26
00:28 japh__ left 00:30 gdonald left, gdonald joined 00:45 Cabanossi left 00:48 Cabanossi joined 00:49 kurahaupo left, kurahaupo joined
samcv nice. i automated it! 01:01
first automated PR down github.com/zostay/perl6-IO-Glob/pull/6
now i will uh. just let it go crazy!
and it checks to make sure json is correct before doing it 01:02
01:03 shdb left 01:04 shdb joined, mcmillhj left
samcv so should be fairly safe 01:04
01:05 mcmillhj joined
TreyHarris erm... isn't there a shorthand to toggle a `has Bool $.blinking is rw` attribute that doesn't require saying the attribute name twice? 01:08
samcv toggle it programatically without typing it into the program?
not sure the question
mst $_ = !$_ for $obj.blinking; or something? 01:09
TreyHarris like $p.blinking = not $p.blinking, but without saying "$p.blinking" twice.
01:09 mcmillhj left
TreyHarris I thought there was a concise and clear way I saw once that was better than just writing it out. But if not, no worries. 01:09
01:12 sufrostico left, labster left
TreyHarris I've had to learn too many languages in the past 16 months--I feel like the semester I took Spanish, Japanese and Russian. I keep "remembering" features from the wrong language. 01:14
01:16 Cabanossi left 01:18 Cabanossi joined 01:20 sufrostico joined
samcv so many PR gist.github.com/samcv/b7809f404778...f08efeb6bd 01:21
that was fun though. once i tried to relax and that i had already built in json validity safechecks and other things that it wouldn't do something embarassing to 30+ repos 01:22
BenGoldberg m: my $x = False; [not=] $x; say $x; 01:28
camelia 5===SORRY!5=== Error while compiling <tmp>
Prefix not requires an argument, but no valid term found
at <tmp>:1
------> 3my $x = False; [not7⏏5=] $x; say $x;
expecting any of:
prefix
BenGoldberg m: my $x = False; $x not= ; say $x; 01:29
camelia 5===SORRY!5=== Error while compiling <tmp>
Two terms in a row
at <tmp>:1
------> 3my $x = False; $x7⏏5 not= ; say $x;
expecting any of:
infix
infix stopper
statement end
statement modifier
…
BenGoldberg m: my $x = False; $x not= (); say $x;
camelia 5===SORRY!5=== Error while compiling <tmp>
Two terms in a row
at <tmp>:1
------> 3my $x = False; $x7⏏5 not= (); say $x;
expecting any of:
infix
infix stopper
statement end
statement modifier…
skids samcv: merged X::Protocol, Control::Bail, Proc::Screen. Also, the urls stop showing up halfway down the gist. 01:30
BenGoldberg m: my $x = False; $x .= not; say $x;
camelia True
BenGoldberg m: class Blinkable { has $.blinking }; my Blinkable $obj .= new; $obj.blinking.=not; dd $obj; 01:31
camelia Cannot modify an immutable Any
in block <unit> at <tmp> line 1
BenGoldberg m: class Blinkable { has $.blinking is rw }; my Blinkable $obj .= new; $obj.blinking.=not; dd $obj;
camelia Blinkable $obj = Blinkable.new(blinking => Bool::True)
BenGoldberg TreyHarris ^
01:40 kurahaupo left 01:41 cpage_ left 01:42 labster joined 01:45 ilbot3 left 01:47 Cabanossi left 01:48 ilbot3 joined, ChanServ sets mode: +v ilbot3
samcv skids, the ones without url's don't have PR yet 01:48
01:48 Cabanossi joined
samcv because they didn't have artistic and 2 in the license field 01:48
the other ones will have to be done more carefully
02:08 noganex_ joined 02:10 noganex left 02:15 Cabanossi left 02:17 Cabanossi joined
TimToady m: my $x = False; $x ?^= True; say $x; $x ?^= True; say $x; 02:19
camelia True
False
TimToady you can use exclusive or as well
02:21 japh__ joined 02:24 japh____ left
BenGoldberg m: use Benchmark; 02:26
camelia ===SORRY!===
Could not find Benchmark at line 1 in:
/home/camelia/.perl6
/home/camelia/rakudo-m-inst-2/share/perl6/site
/home/camelia/rakudo-m-inst-2/share/perl6/vendor
/home/camelia/rakudo-m-inst-2/share/perl6
CompUnit::R…
02:27 jrac1001 joined
MasterDuke_ star: use Benchmark 02:30
camelia ===SORRY!===
Could not find Benchmark at line 1 in:
/home/camelia/.perl6
/home/camelia/star-2016.10/share/perl6/site
/home/camelia/star-2016.10/share/perl6/vendor
/home/camelia/star-2016.10/share/perl6
CompUnit::Repository::Ab…
MasterDuke_ BenGoldberg: trying to do some benchmarking?
02:32 jrac1001 left
TimToady m: my $x = False; $x .= not for ^1000000; say now - INIT now 02:34
camelia 2.5071128
TimToady m: my $x = False; $x ?^= True for ^1000000; say now - INIT now
camelia 1.27069904
TimToady m: my int $x = 0; $x +^= 1 for ^1000000; say now - INIT now 02:36
camelia 0.60105261
02:40 ChoHag joined 02:47 japh____ joined 02:49 mcmillhj joined 02:51 japh__ left 02:54 Actualeyes left, mcmillhj left 02:56 mr-foobar left 03:00 japh__ joined 03:02 titsuki_ joined 03:03 japh____ left, BenGoldberg left, titsuki left 03:04 BenGoldberg joined 03:07 Actualeyes joined 03:17 sufrostico left, japh____ joined 03:20 japh__ left 03:27 zapwai left, Sgeo joined, Sgeo__ left
samcv phase 2 attack of the automated PR has begun 03:40
now if it has artistic or perl in the license it checks to make sure there's only one license file, and that, that file is at max one line copyright line different from my copy
03:43 kevin joined, kevin is now known as Guest8657, perlawhirl left 03:44 Cabanossi left
Guest8657 Hi all, is there any module in perl6 can parse excel file 03:44
03:44 labster left 03:46 Cabanossi joined, perlawhirl joined, perlawhirl__ joined, perlawhirl_ joined, perlawhirl___ joined
samcv good question Guest8657 03:47
03:47 Guest8657 left
samcv TimToady, looks especially nice if we sort it by length heh gist.github.com/samcv/b7809f404778...f08efeb6bd 03:47
i guess that's the lazy way to sort it by whether or not there's a url there 03:48
still 50 more repos i need to cover though 03:55
u-ou hello 03:57
samcv hi u-ou :) 03:59
u-ou (:
04:00 wamba joined 04:02 u-ou is now known as no-n 04:03 japh__ joined 04:06 japh____ left
no-n tfw your program is tidy and resonable but then you need to add more features 04:10
samcv :-D 04:12
no-n :)
04:14 skids left 04:15 Cabanossi left 04:17 Cabanossi joined
samcv how many PR until github shuts you down?. lol. 04:17
maybe they have an algorithm? 04:18
since idk what prevents people creating Pull requests or issues with spam in it. hm. or mabye it's not really an issue
no-n hehe 04:19
perlawhirl_ fwiw, i fixed up mine < Net::Netmask Pod::To::Markdown::Fenced wig > 04:21
04:24 Possum left 04:25 perlawhirl_ left
samcv oh cool 04:28
thank you perlawhirl :) 04:30
04:30 no-n is now known as u-ou 04:31 Possum joined 04:34 labster joined 04:46 perlawhi1l joined 04:50 BenGoldberg left 05:00 Cabanossi left 05:01 Cabanossi joined 05:02 japh__ left, japh__ joined
Geth doc: 4bc72d5114 | (Samantha McVey)++ | doc/Language/modules.pod6
Add updated information on putting license in META6.json

  design.perl6.org/S22.html#license		 05:02
doc: eb608b1a1f | (Samantha McVey)++ | .travis.yml
Add fast_finish to .travis.yml

Should enable build status before ignored build statuses have finished		 05:07
05:09 japh____ joined 05:10 Possum left 05:11 japh__ left 05:17 perlawhirl___ left 05:18 azawawi joined
azawawi good-morning 05:18
05:18 perlawhirl__ is now known as perlawhirl_, perlawhirl_ is now known as 07EAAQYAC 05:19 07EAAQYAC left, perlawhirl left, perlawhi1l is now known as perlawhirl 05:20 japh____ left, japh____ joined 05:22 Possum joined 05:29 Cabanossi left 05:31 Cabanossi joined 05:34 Possum left 05:36 japh__ joined 05:39 japh____ left 05:40 Possum joined
masak good morning, #perl6 05:43
m: my $program = q[say "lol, so unsafe!"]; EVAL $program
camelia 5===SORRY!5=== Error while compiling <tmp>
EVAL is a very dangerous function!!! (use the MONKEY-SEE-NO-EVAL pragma to override this error,
but only if you're VERY sure your data contains no injection attacks)
at <tmp>:1
------> 3 q[say "lol,…
samcv hii
masak m: use MONKEY-SEE-NO-EVAL; my $program = q[say "lol, so unsafe!"]; EVAL $program # bah 05:44
camelia lol, so unsafe!
masak m: my $program = q[say "lol, so unsafe!"]; my &eval = &EVAL; eval $program
camelia lol, so unsafe!
samcv freaks out and runs around screaming
masak ah, so *that's* how you do it
two birds with one stone, really
05:44 xtreak joined
samcv hahahha 05:44
masak (a) I get the non-silly spelling back
(b) I trivially circumvent the hysterical error
this almost deserves a blog post, although I fear it might be a slightly grumpy one 05:45
samcv hahaha 05:46
m: ("'foo'" ~ '.EVAL').EVAL.say
camelia 5===SORRY!5=== Error while compiling /home/camelia/EVAL_1
Undeclared routine:
foo used at line 1
samcv m: Q<'foo'.EVAL>.EVAL.say 05:47
camelia 5===SORRY!5=== Error while compiling /home/camelia/EVAL_1
Undeclared routine:
foo used at line 1
masak fwiw, in the book "JavaScripdt: The Definitive Guide", it says that (in JS) `eval()` oughta have been a unary (prefix) operator, not a function. this is why.
samcv m: 'foo'.EVAL.say
camelia 5===SORRY!5=== Error while compiling /home/camelia/EVAL_0
Undeclared routine:
foo used at line 1
masak (because you basically cannot statically analyze first-class function values)
TimToady: hm, why was it at one point believed that only `:=` would work for assigning functions in Perl 6, not `=`? 05:48
(this is back to the Pugs or pre-Pugs days. ISTR that Pugs proved in practice that `=` worked. but I'm unclear on why it was not obvious at the time that it'd work.) 05:49
05:50 ufobat joined, RabidGravy joined
moritz I guess the thinking was that & variables wouldn't have a level of container 05:51
but I mostly remember that I didn't understand the reasoning at that time 05:53
samcv m: my $n = 5; my $var = "foo"; for ^$n { $var = "‘‘" ~ $var ~ "’’" ~ ".EVAL" }; $var.say; ($var ~ '.EVAL' x $n -1 ).EVAL.say
camelia ‘‘‘‘‘‘‘‘‘‘foo’’.EVAL’’.EVAL’’.EVAL’’.EVAL’’.EVAL
foo
samcv nice.
hahaha
m: my $n = 5; my $var = "foo"; for ^$n { $var = "‘‘" ~ $var ~ "’’" ~ ".EVAL" }; $v2 = ($var ~ '.EVAL' x $n -1 ); $v2.say; $v2.EVAL.saysay 05:54
camelia 5===SORRY!5=== Error while compiling <tmp>
Variable '$v2' is not declared
at <tmp>:1
------> 3 $var = "‘‘" ~ $var ~ "’’" ~ ".EVAL" }; 7⏏5$v2 = ($var ~ '.EVAL' x $n -1 ); $v2.say
samcv m: my $n = 5; my $var = "foo"; for ^$n { $var = "‘‘" ~ $var ~ "’’" ~ ".EVAL" }; my $v2 = ($var ~ '.EVAL' x $n -1 ); $v2.say; $v2.EVAL.saysay
camelia ‘‘‘‘‘‘‘‘‘‘foo’’.EVAL’’.EVAL’’.EVAL’’.EVAL’’.EVAL.EVAL.EVAL.EVAL.EVAL
No such method 'saysay' for invocant of type 'Str'
in block <unit> at <tmp> line 1
samcv m: my $n = 5; my $var = "foo"; for ^$n { $var = "‘‘" ~ $var ~ "’’" ~ ".EVAL" }; my $v2 = ($var ~ '.EVAL' x $n -1 ); $v2.say; $v2.EVAL.say
camelia ‘‘‘‘‘‘‘‘‘‘foo’’.EVAL’’.EVAL’’.EVAL’’.EVAL’’.EVAL.EVAL.EVAL.EVAL.EVAL
foo
samcv so many evals
m: my $n = 5; my $var = "foo"; for ^$n { $var = "‘“" ~ $var ~ "”’" ~ ".EVAL" }; my $v2 = ($var ~ '.EVAL' x $n -1 ); $v2.say; $v2.EVAL.say 05:55
camelia ‘“‘“‘“‘“‘“foo”’.EVAL”’.EVAL”’.EVAL”’.EVAL”’.EVAL.EVAL.EVAL.EVAL.EVAL
foo
samcv slightly easier to read like this. but. really not XD
06:04 mr-foobar joined 06:10 domidumont joined
RabidGravy samcv, so when I asked you *not* to send me PRs about this license thing ... 06:12
samcv i'm sorry RabidGravy, i automated it, and it applied it to all of them. sorry about that
06:17 domidumont left
moritz RabidGravy: you can like, close them :-) 06:18
RabidGravy I have
samcv they did get closed ;)
RabidGravy I am also going to remove all the modules from the ecosystem at 9PM BST unless I see some evidence that this has been discussed among a group wider than half a dozen people on here 06:19
TimToady sorry, we don't play the nuclear blackmail game here 06:20
moritz we have 304 nicks in here, so maybe ~250 people who had the opportunity to speak up 06:21
RabidGravy I know, it's not a threat, I just don't think this is the way to proceed
moritz then propose alternative processes to deal with such things
ideally some that aren't a huge overhead
06:21 japh____ joined 06:22 japh__ left 06:23 xtreak left, xtreak joined
samcv RabidGravy, i'm sorry for the PR. and i looked at them and changes from "perl" to "Artistic-2.0" or from "Artistic" to "Artistic-2.0" offer more information about the modules. i mean there could be a way that we could have package managers resolve names that may not be exact matches, but "perl" and "Artistic" don't show full information 06:24
TimToady Perl 6 has succeed to the extent it has because it has been a convergent (on average) process over the years.
samcv also open to suggestions about how to preced of course. and i've blacklisted you on my script 06:25
TimToady we haven't forced rapid convergence, (obviously, in retrospect), but we do want to keep a gentle pressure in that direction
if this was not gentle enough, then we need to figure out a better way, as moritz points out 06:26
samcv agreed
TimToady perhaps "blacklisted" has some undesirable negative connotations :) 06:27
samcv was not meant that way. but it will skip any repos of user jonathanstowe on github now
06:27 xtreak left
moritz "not-nag-listed" :-) 06:27
samcv added an exception?
06:28 nadim joined
TimToady actually, given how opinionated we all are, I'm surprised we've managed the degree of convergence we have, even if it did take a decade and a half (so far) :) 06:29
samcv :) 06:30
docs.perl6.org/language/modules#in...META6.json i've updated the license section of here to provide more guidance. mostly says the same as the synopsis. need to clean up formatting a bit but 06:33
or template now has a license field so hopefully that will help people out as well making new modules
06:33 darutoko joined 06:35 domidumont joined 06:37 wamba left 06:38 imcsk8 joined 06:40 domidumont left 06:41 domidumont joined 06:44 azawawi left 07:01 japh____ left
Geth Inline-Perl5: 45cd7ec289 | (Samantha McVey)++ | META6.json
Use SPDX identifier in license field of META6.json

For more details see design.perl6.org/S22.html#license		 07:01
Inline-Perl5: 8cb8c3254f | niner++ (committed using GitHub Web editor) | META6.json
Merge pull request #95 from samcv/license

Use SPDX identifier in license field of META6.json
07:02 japh__ joined, samcv left, samcv joined 07:03 raschipi joined 07:07 cpage_ joined 07:13 parv joined 07:15 astj joined
TEttinger I'm surprised nobody's made an Autistic License that is just like the Artistic License but has a clause that specifies it must be taken literally 07:22
it's redundant with it being a legal document, but it never hurts to be clear 07:23
07:26 wamba joined 07:27 rindolf joined 07:29 TheAthlete joined 07:32 qeesung joined, qeesung left 07:33 astj left, TheAthlete left
raschipi TEttinger: aiming to increase redudancy is an infinite loop. 07:42
TEttinger you're in #perl6, there are how many redundant implementations of it? 07:53
raschipi Yes, but if increasing redudancy was our *aim*, everyone would have to start multiple nem implementations... 07:58
new implementations...
tyil nibbles on samcv 08:03
08:04 xtreak joined
raschipi I'm joking, of course, but I think you meant it's better to be explicit instead of implicit, not that just increasing redudancy is a good thing in itself. 08:04
08:06 xtreak left, xtreak joined 08:07 abraxxa joined
TEttinger no I think I meant it to tie back to the characteristic of many people with autism who have difficulty understanding metaphors or non-formal documents and prefer things specified pedantically for clarity 08:07
which is not a disadvantage for programmers
or lawyers!
samcv aren't licenses already pretty precise? 08:08
raschipi I think we just said the same thing in too different ways, there's redudancy for you.
TEttinger hehe
samcv :O
u-ou too?
samcv also
raschipi two*
TEttinger samcv: I think there are lots of shoddy licenses out there. I encountered one 08:09
samcv WTF license?
www.wtfpl.net/txt/copying/
raschipi Arguing for the everything explicit in #perl6, that's asking for a flamewar. 08:10
samcv i'm still not sure what it actually grants as the license… though i know what the spirit of it is. i still don't know
eater why not force everyone to use Artistic-2.0? "If you run this program under perl6 it is licensed under Artistic-2.0"
u-ou what license do I need to use if I release some Perl 6 code?
TEttinger like this one is just terrible oryxdesignlab.com/license
samcv u-ou, Artistic 2.0 is the most common one
that's what rakudo moarvm and nqp are licensed as, and the majority of our licensed modules
TEttinger (that license appears to prevent any usage of the things you buy from oryx)
08:11 xtreak left, japh__ left
parv u-ou, depends mostly on your own motives & anybody else's involved 08:11
samcv i would argue the WTF license doesn't grant anything new. and people already "do what the fuck they want to"
TEttinger AGPL, watch the world burn
samcv but i've seen argued that GPL and AGPL at least in europe still allow linking to programs of any license without forceing them to be GPL 08:12
though in other juristictions it could be different
TEttinger AGPL is a pretty fiendish one though
especially with stuff like Opa, which had an AGPL standard lib for a long time 08:13
samcv yes. but at least in europe. if it's linked then you only need to release the source of the library itself that was modified
TEttinger so any code that used strings, or numbers, etc. became AGPL. and it was for web dev
so your crypto, needs to be OSS
08:13 pmurias joined
raschipi working as intended. 08:13
08:13 zakharyas joined
TEttinger no surprise, Opa is dead 08:14
08:14 japh__ joined
masak moritz: "no level of container" makes a modicum of sense, I guess 08:14
or rather, I don't understand why it was thought it wouldn't/shouldn't have one, but... if that was the case, then binding'd be the only option 08:15
samcv TEttinger, see joinup.ec.europa.eu/community/eupl...sing-ghost
As a conclusion, it looks that in most cases, linking two programs or linking an existing software with your own work does not – at least in Europe – produce a derivative or extends the coverage of the linked software licence to your own work
^ the gist of it
TEttinger I'm not in Europe
samcv me either 08:16
in other places in is not quite as certain
TEttinger also, this is just hilarious opalang.org/
there's no download link, just some build instructions with a significant omission for windows 08:17
samcv but it hinges on what is a derivitive work TEttinger
in any jurisdiction
TEttinger mm 08:18
samcv which is the language the GPL and AGPL use
TEttinger also any software that needs, as part of its build process, for you to run `sudo ln -s /sbin/md5 /usr/local/bin/md5sum`
maaaaay have a problem
samcv uhm
i don't see why
how is that a derivitive work... 08:19
TEttinger oh that was the Opa thing
samcv oh
TEttinger Opa changed from AGPL stdlib to just AGPL compilr
samcv i should make a "don't do anything" license. as a parady of the do what the fuck you want to license
which will say you can't do anything. even look at the software
or think about it
or maybe "don't do anything you want to" 08:20
raschipi can't copy it even to your mind.
samcv which is equally meaningless to the wtf license
haha
because what does "don't do anything you want to" even mean...
what if you don't want to do something, are yo uthen allowed to do it?
parv samcv: is that not for every closed source software & tractors? 08:21
samcv and conversely with the WTF license
haha
TEttinger I should make a license that requires anyone who seeks to enforce a patent against the licensor represent themselves in court with a blood alcohol level of 1%
samcv LOL
idk if that's legal
TEttinger if they don't drive
samcv just make it at least 3/4 of the legal limit
TEttinger prohibition was repealed samcv
samcv for a court room
08:21 japh__ left
samcv but. yo ucan't be drunk in court... 08:21
if you fail a breath test they'll throw you own in contempt 08:22
08:22 japh__ joined
TEttinger public drunkenness yeah 08:22
jast after a blood alcohol level of 1% I'm fairly sure you'll never drive again
samcv court is in public
TEttinger hmm, what would be a better requirement
samcv no not true jast
TEttinger oh!
samcv but it is not healthy
TEttinger must present entire case standing on head
samcv dunno then no way for case to go forward 08:23
TEttinger if you fail, you must concede
samcv don't think judge would allow
LOL
TEttinger, or how to get parts of your license thrown out in court easily
RE these things. lol
TEttinger yes.
but standing on your head must be legal in court, they need to reenact things like assaults where someone's wrestling 08:24
raschipi The other day a judge threw someone in contempt in Brazil because they were using flip-flops. The Superior Court awarded them reparation from the judge that did it.
samcv haha
raschipi A judge won't enforce things they don't want to.
08:24 Cabanossi left 08:25 Cabanossi joined
jast samcv: AFAIK the lethal dose is around 0.3-0.4% for "unpracticed" drinkers. there have been occasional measurement results of up to 0.7% I believe, but 1% seems a bit much 08:25
08:26 araujo left
TEttinger i suppose in that case you would weigh your options and decline to sue 08:26
jast historical find of 0.77%: www.tagesspiegel.de/weltspiegel/alk...71216.html :)
08:27 raschipi left
samcv After a short sobering, the drunkard had already left the hospital. 08:27
seems fine to me 08:28
but i would argue jast 1% would likely the person be unconscious without a very high tolerance
and could be dangerous. so would be hard to argue in court
jast yeah, and note the "I've never seen a value this high in my 28 years of doing this job" and "you might survive this once but it'll cause a lot of damage" 08:29
samcv so to meet the requirements you'd have to spend months gaining a tolerance to fight in court
jast and it'd still be almost 30% more than has ever (?) been encountered 08:30
samcv .ask Zoffix is there anyway we could have `dir :d` or `dir :f` to get all directories or all files etc? 08:31
yoleaux samcv: I'll pass your message to Zoffix.
08:32 domidumont left, robertle joined, domidumont joined
samcv i can do dir.grep({.d}) but dir :d would be much more convenient 08:33
moritz dir.grep("d" => True) might work 08:34
.grep(*.d) 08:35
samcv but it's a very commonly wanted thing i would think
to want all folders or all files
moritz but there's more to it 08:36
what about symlinks?
should it follow them? treat them as dirs if it points to a dir?
samcv there is more to it. but.... :d would be like .grep(*.d)
or dir :d, :f would do dirs and files. would perform like the .d or .f things do 08:37
it just works exactly the same way things already do
El_Che hi 08:39
08:40 domidumont left
moritz ho 08:40
nadim hej
El_Che 'Trump has to go'
oh, we're not doing that :)
08:41 domidumont joined, gdonald left
u-ou does $s does Role run all of Role's initialisers on $s? 08:41
08:41 gdonald joined
u-ou so that you get all of the role's variables reset? 08:41
I hope that makes sense
if you do it multiple times, does it keep resetting variables through the initialisers? 08:42
samcv pyeah it does u-ou
that's the thing about roles which make them cool 08:43
u-ou yeah, they are cool
samcv it carries over many more things than `is` does
u-ou so you don't have to do self.stats = Stats.new, you can just do self does Stats
m: role Stuff { has $.i is rw = 0 }; class Thing does Stuff { method inc { $.i++ }; method get { $.i }; method reset { self does Stuff } }; my $t = Thing.new; $t.inc for ^10; $t.get.say; $t.reset; $t.get.say 08:46
camelia 10
0
u-ou so, every time 'does' is used the initialisers run? 08:48
i.e. = 0
08:56 domidumont left, domidumont joined 08:59 espadrine joined
u-ou weird. if i change method get to $!i i get 10 again 08:59
m: role Stuff { has $.i is rw = 0 }; class Thing does Stuff { method inc { $.i++ }; method get { $!i }; method reset { self does Stuff } }; my $t = Thing.new; $t.inc for ^10; $t.get.say; $t.reset; $t.get.say
camelia 10
10
09:02 wamba left
u-ou but why? :I 09:03
09:06 gregf_ joined 09:07 espadrine left
timotimo oh you really don't want to "does" your object multiple times 09:08
m: role Omg { }; my $i = 10; for ^10 { say $i.perl; $i does Omg }
camelia 10
10
10
10
10
10
10
10
10
10
timotimo bad example
m: role Omg { }; class Test { }; my $i = Test.new; for ^10 { say $i.perl; $i does Omg } 09:09
camelia Test.new
Test+{Omg}.new
Test+{Omg}+{Omg}.new
Test+{Omg}+{Omg}+{Omg}.new
Test+{Omg}+{Omg}+{Omg}+{Omg}.new
Test+{Omg}+{Omg}+{Omg}+{Omg}+{Omg}.new
Test+{Omg}+{Omg}+{Omg}+{Omg}+{Omg}+{Omg}.new
Test+{Omg}+{Omg}+{Omg}+{Omg}+{Omg}+{Omg}+{Omg}.new…
timotimo you're creating a new class every time
u-ou ahhhh
timotimo that'll also cause deoptimization because we can't be sure about types involved in existing code any more
u-ou so to reset everything you've mixed in you just have to reset everything you've mixed in 09:10
timotimo you can keep the original object around and use "but" instead of "does" 09:11
09:12 wamba joined
u-ou m: role R {}; class C {}; my $i = C.new; for ^10 { say $i.perl; $i = $i but R } 09:13
camelia C.new
C+{R}.new
C+{R}+{R}.new
C+{R}+{R}+{R}.new
C+{R}+{R}+{R}+{R}.new
C+{R}+{R}+{R}+{R}+{R}.new
C+{R}+{R}+{R}+{R}+{R}+{R}.new
C+{R}+{R}+{R}+{R}+{R}+{R}+{R}.new
C+{R}+{R}+{R}+{R}+{R}+{R}+{R}+{R}.new
C+{R}+{R}+{R}+{R}+{R}+{R}+{R}+{R}+{R}.…
u-ou oh, "the original object" 09:14
09:16 Ven joined, Ven is now known as Guest37478
timotimo oh, but that won't keep changes around of course 09:17
u-ou I'm probably going to just write a "reset" method for my role :P
timotimo i wonder how to properly delegate to a method of one name in all roles, i don't think we had something for that yet
09:24 sena_kun joined 09:25 xtreak joined
eater self but Stuff? 09:26
u-ou what's the diff. between does and but? 09:28
I think it's just that but clones the object and does does not
09:32 wamba left 09:34 bjz joined 09:38 TEttinger left 09:48 Guest37478 left, salv0 joined 09:50 AlexDaniel joined 09:51 Ven joined, Ven is now known as Guest2485
tbrowder .tell ugexe zef fails to install Text::Abbrev 09:52
yoleaux tbrowder: I'll pass your message to ugexe.
09:52 wamba joined 10:00 Cabanossi left, Cabanossi joined 10:03 xtreak left
Geth ecosystem: the-eater++ created pull request #323:
Add travis tests for newly added packages		 10:04
10:22 aindilis left 10:24 labster left 10:28 xinming left 10:29 xinming joined 10:30 xtreak joined 10:34 bjz left 10:37 bjz joined 10:38 ilogger2 joined, ChanServ sets mode: +v ilogger2 10:46 Cabanossi joined
samcv opened an eco issue github.com/perl6/ecosystem/issues/324 so we can track the modules which have nonstandard license keys, now that i'm done with the automable stage. full summary and detail in the summary at the top 10:47
i must go to sleep now. night everybody o/
10:55 salv0 joined 11:02 Ven joined 11:03 Ven is now known as Guest98008 11:13 Guest98008 left
w4and0er96 assume `my @array = < 3 1 4 5>; ` i wanna know how `@array[*-1] == 5`, what is `*` means and how can i use it 11:21
11:21 mr-foobar joined
moritz w4and0er96: *-1 creates a callback, just like sub ($x) { $x - 1 } 11:21
w4and0er96: and if you pass such a callback to the [] indexing operator, it calls it with the number of elements as an argument 11:22
so it becomes @array[@array.elems - 1], basically
w4and0er96 is it a special syntactic sugar? or where else can i use such trick 11:26
moritz m: say map *-1, 1..5; 11:27
camelia (0 1 2 3 4)
moritz w4and0er96: wherever you want to construct a simple function
m: say map *.uc, <a b c>
camelia (A B C)
w4and0er96 thx, maybe i got some about it 11:28
Woodi hi #perl6 11:39
question: if I buy software licence (with code) then what I can do with it ? looks that just buying is MUCH less trouble then the best OOS licence... 11:41
Oracle and MS just sells per cpu or even core. what about "pay $2^m" where m is number of managers you have in company ? ;) 11:42
samcv open source software is free though. and depending on the license you can modify and include in software that you charge money for without needing to release the source
Woodi just looks OOS licences are main problem with OOS :) maybe we do that to yourself ? 11:44
eater .f scr
eh
samcv Woodi, non open source don't allow you the source PERIOD 11:45
and the licenses don't let you sell it for money
in certain ways 11:46
so they are more restrictive period. regarding MIT licenses or Artistic licenes
Artistic 2.0 is pretty nice
Woodi so selling source code to distributor to resell is impossible business ? :)
why painters don't have so much troubles ? 11:48
Geth doc: 466cc01ac1 | (Samantha McVey)++ | META.info
Use SPDX identifier in license field of META.info

Use the standardized identifier for the license field. For more details see design.perl6.org/S22.html#license
Woodi I guess purpose of Artistic license is to have similiar case...
samcv you can sell proprietary software with open source parts in it easily 11:49
pmurias samcv: there are non open source things that allow you access to the source 11:50
yoleaux 18 Apr 2017 00:58Z <MasterDuke_> pmurias: i think NQP_VERBOSE_EXCEPTIONS=1 is what gives better jvm errors
samcv just keep the proprietary stuff not open source if you want to remain ability to pursue people distributing it or something idk. depends what profit model is
pmurias, that's not possible
do you mean free software?
sena_kun is it just me or zef install File::Ignore doesn't work on latest rakudo? 11:51
samcv open source... definition means you can read it
uh sena_kun passed for me uh
3 days ago?
oh wait. nope
it fails! 11:52
sena_kun samcv, well, I'll open an issue. Thanks!
samcv forgot the bug here hack.p6c.org/~samcv/install_log_live.txt but search for "Testing [FAIL]: File::Ignore"
if you have the power add it to here github.com/perl6/ecosystem/issues/318
if you're in p6 group. i need to sleep
sena_kun samcv, sure.
pmurias samcv: is "you can change the source code but must give as 5% of your profits for anything you use it" Open Source?
samcv or just reply if not
sena_kun samcv, sleep well.
samcv yes pmurias
open means you can read it 11:53
free means that you can freely copy and alter it
also. that's not a license that's a contract
what you described
very different legally. one is copyright law other is contract law 11:54
also night o/
i should have been to bed hours ago
also this has agood answer softwareengineering.stackexchange....t-you-cant 11:55
but gpl 3 is not that bad, but companies don't like it because of uh. they don't understand fully when it is and isn't covered. so. but that makes sense
anyway
it talks about contract and copyright differences (as long as i linked the right thing). night 11:56
sena_kun the list is updated 11:58
samcv also for extra points, search for `===> Testing [FAIL]` on that log txt hack.p6c.org/~samcv/install_log_live.txt and add them too ;) 12:01
sena_kun is there anyone who can give me a hint about new braid(?) policy? I mean the "define_slang" thing. It was implemented about two months ago(as I remember), but I was too slow to update my slang that time.
samcv Zzz
sena_kun samcv, roger!
jnthn I somewhat fear the File::Ignore thing is going to turn out to be a Rakudo regression 12:02
Because I've installed it not all that long ago
And even if I go back in the module's history some way, stuff still fails 12:03
So it's not a regression in the module due to recent changes 12:04
sena_kun regressions are unavoidable, I guess. though, for two months of difference rakudo feels slightly faster for me now, so everyone++
jnthn m: say 'resultA.txt' ~~ /'result-' <[AB]-[/]> '].txt'<?before "/" | $> / 12:07
camelia Nil
jnthn commitable6: 2017.02 say 'resultA.txt' ~~ /'result-' <[AB]-[/]> '].txt'<?before "/" | $> /
committable6 jnthn, ¦2017.02: «Nil»
jnthn ah, d'oh
12:07 Ven joined, nowan joined
jnthn commitable6: 2017.02 say 'result-A.txt' ~~ /'result-' <[AB]-[/]> '].txt'<?before "/" | $> / 12:08
committable6 jnthn, ¦2017.02: «Nil»
12:08 Ven is now known as Guest49446
jnthn oh hmm 12:08
yeah, that's wrongly generated, jsut spotted the ]
It's taking a little time to golf, but I've got an idea what might be going on 12:15
12:15 mr-foobar left 12:16 wamba joined 12:19 mr-foobar joined
sena_kun >Cannot import symbol DOM from Template::Anti, because it already exists in this lexical scope 12:25
jnthn This is super weird
sena_kun it seems that there is a lot of broken modules. :|
jnthn The mere presence of an action method with an empty body causes a later match to be wrong
12:26 grondilu joined
jnthn m: gist.github.com/jnthn/f26b1ed53e6e...256523af0e 12:26
camelia ｢[AB].txt｣
matcher => ｢AB｣
matcher => ｢].txt｣
jnthn commitable6: HEAD gist.github.com/jnthn/f26b1ed53e6e...256523af0e 12:27
committable6 jnthn, It looks like a URL, but mime type is ‘text/html; charset=utf-8’ while I was expecting something with ‘text/plain’ or ‘perl’ in it. I can only understand raw links, sorry.
jnthn Slacker!
committable6: HEAD gist.githubusercontent.com/jnthn/f...ismatch.p6
committable6 jnthn, Successfully fetched the code from the provided URL.
jnthn, ¦HEAD(736be4d): «｢[AB].txt｣␤ matcher => ｢AB｣␤ matcher => ｢].txt｣»
jnthn committable6: 2017.02 gist.githubusercontent.com/jnthn/f...ismatch.p6
committable6 jnthn, Successfully fetched the code from the provided URL.
jnthn, ¦2017.02: «｢[AB].txt｣␤ matcher => ｢AB｣␤ matcher => ｢.txt｣»
jnthn Note the ] that has sneaked in to the final matcher capture
bisectable6: gist.githubusercontent.com/jnthn/f...ismatch.p6 12:28
bisectable6 jnthn, Successfully fetched the code from the provided URL.
jnthn, Bisecting by output (old=2015.12 new=736be4d) because on both starting points the exit code is 0
jnthn, bisect log: gist.github.com/119eb26d8ca245bf59...bc014c9fc2
jnthn, There are 20 candidates for the first “new” revision. See the log for more details
12:28 Cabanossi left
jnthn Aww 12:29
committable6: 2017.03 gist.githubusercontent.com/jnthn/f...ismatch.p6
committable6 jnthn, Successfully fetched the code from the provided URL.
jnthn, ¦2017.03: «｢[AB].txt｣␤ matcher => ｢AB｣␤ matcher => ｢.txt｣»
jnthn committable6: 2017.04 gist.githubusercontent.com/jnthn/f...ismatch.p6
committable6 jnthn, Successfully fetched the code from the provided URL.
jnthn, ¦2017.04: «｢[AB].txt｣␤ matcher => ｢AB｣␤ matcher => ｢].txt｣»
12:31 Cabanossi joined
jnthn committable6: e0e0800 gist.githubusercontent.com/jnthn/f...ismatch.p6 12:31
committable6 jnthn, Successfully fetched the code from the provided URL.
jnthn, ¦e0e0800: «｢[AB].txt｣␤ matcher => ｢AB｣␤ matcher => ｢.txt｣»
12:31 domidumont joined
jnthn committable6: fbf19d88dac7 gist.githubusercontent.com/jnthn/f...ismatch.p6 12:31
committable6 jnthn, Successfully fetched the code from the provided URL.
jnthn, ¦fbf19d8: «Cannot test this commit (Commit exists, but a perl6 executable could not be built for it)»
jnthn hmmm
That was the uncurse merge
Which I'm suspecting but can't easily prove
committable6: 63e76d7 gist.githubusercontent.com/jnthn/f...ismatch.p6 12:32
committable6 jnthn, Successfully fetched the code from the provided URL.
jnthn, ¦63e76d7: «Cannot test this commit (Commit exists, but a perl6 executable could not be built for it)»
jnthn files a rakudobug with what he's managed to figre out
*figure
12:33 bjz_ joined
sena_kun jnthn++ 12:33
12:33 xtreak joined
jnthn sena_kun++ # reporting it :) 12:33
12:34 Guest49446 left 12:35 Ven_ joined
sena_kun jnthn, I just merely tried to install Ddt to fix my Slang module, and File::Ignore(the dependency of Ddt) was broken and... Well, the dependency chain goes here. Yet I still don't know how to fix my module. 12:35
12:35 japh____ joined
jnthn Filed it; hoping TimToady++ might have an idea what's up :) 12:42
12:42 domidumont left 12:43 domidumont joined 12:45 Actualeyes joined
Geth doc: 6172df409c | (Zoffix Znet)++ (committed using GitHub Web editor) | doc/Type/IO.pod6
Mention interaction between indir...

  ...$*CWD and lazily evaluted results returned from indir		 12:46
12:49 Ven_ left 12:50 sufrostico joined 12:51 parrot_bird joined 12:52 darutoko- joined, wamba left, parrot_bird left 12:54 mcmillhj joined, Ven_ joined 12:58 Cabanossi left 12:59 kurahaupo joined 13:00 xtreak left 13:01 Cabanossi joined 13:26 astj joined
sena_kun .tell samcv I have raided install_log_live.txt a bit, but, hmm... "cat install_log_live.txt | grep 'Testing \[FAIL\]' | sort | uniq | nl" gives me 132. Of course, some of them are failing because of missing native library, but it is approximately 70-100 modules to re-check and add to issue description. I wonder if it can be automated somehow. 13:30
yoleaux sena_kun: I'll pass your message to samcv.
grondilu mans nl 13:31
13:33 Ven_ left
sena_kun grondilu, it should be, hmm, wc -l, but I'm more used to nl somehow. 13:33
13:34 Hor|zon joined 13:35 Ven joined 13:36 Ven is now known as Guest54785 13:42 astj left, skids joined 13:43 Cabanossi left 13:45 Guest54785 left 13:46 Cabanossi joined 13:48 astj joined 13:52 astj left
SmokeMachine anyone ever tried to use native call with OpenCL on mac? 13:53
13:54 Ven_ joined
SmokeMachine im trying to run the PerlCL: github.com/kmwallio/PerlCL/blob/master/play.p6 13:55
13:56 cowens joined 14:00 mcmillhj left 14:05 mcmillhj joined 14:06 darutoko joined 14:08 darutoko- left
ugexe tbrowder: I'm surprised Text::Abbrev ever worked - `unit module Abbrev;` 14:10
yoleaux 09:52Z <tbrowder> ugexe: zef fails to install Text::Abbrev
14:10 darutoko left 14:12 Ven_ left
tbrowder ugexe: I didn't check it closely, thsnks. 14:12
w4and0er96 is there some usual use of `$_` and `@_` 14:13
new to perl, I find it too odd to understand
b2gills If you put an object in ｢$_｣ you can call methods on it without mentioning it ｢.method｣ is short for ｢$_.method｣ 14:15
cowens $_ is default variable it is targeted by some functions and syntax by default 14:16
I think Perl 6 calls it the “subject”
@_ is where your arguments to a function are stored if you don’t specify a signature
14:19 Ven joined, Ven is now known as Guest78706
b2gills There is also ｢$/｣ which is mostly used for regex results. ｢$<key>｣ is short for ｢$/<key>｣ and ｢$0｣ is short for ｢$/[0]｣ 14:19
grondilu m: say sub { $_ }("hello")
camelia Too many positionals passed; expected 0 arguments but got 1
in sub at <tmp> line 1
in block <unit> at <tmp> line 1
grondilu m: say { $_ }("hello") 14:20
camelia hello
grondilu works for blocks but not subroutines, though
14:21 wamba joined
cowens m: say sub { @_ }("hello") 14:21
camelia [hello]
b2gills m: {say @_; say $_}('hello')#If you have both @_ and $_ in a block, then @_ takes precedence
camelia [hello]
(Any)
b2gills w4and0er96: In case you didn't know we have docs online docs.perl6.org/syntax/$_ please let us know if you have trouble understanding anything in them so that we can improve them 14:26
w4and0er96 ok i'll try to understand it, thank you all 14:28
b2gills Those of us who understand the language can only easily see logical errors in the docs, not what is missing, so we need new people to let us know.
14:30 dct joined
samcv sena_kun, sounds good to me 14:32
yoleaux 13:30Z <sena_kun> samcv: I have raided install_log_live.txt a bit, but, hmm... "cat install_log_live.txt | grep 'Testing \[FAIL\]' | sort | uniq | nl" gives me 132. Of course, some of them are failing because of missing native library, but it is approximately 70-100 modules to re-check and add to issue description. I wonder if it can be automated somehow.
samcv feel free to restructure to allow easier updatinng 14:33
samcv &
sena_kun samcv, well, I have some free time anyway, so I will think about it. The github issue's form looks a bit strange for me, but anyway. I will restructure and update the issue then later. 14:34
samcv yeah it is weird. and i totally agree it could be a lot better 14:35
samcv heads back to bed 14:37
sena_kun samcv, sleep well! # hint: just `poweroff` that damn internet-connected thing already! 14:38
14:42 Guest78706 left
ugexe anyone know github:softmoth / Tim Smith irc handle? 15:07
TreyHarris :m class Point { has $.x; has $.y; }; my $p = Point.new(:3x, :5y); "x: {.x} y: {.y}" given $p; 15:08
agh
m: class Point { has $.x; has $.y; }; my $p = Point.new(:3x, :5y); "x: {.x} y: {.y}" given $p;
camelia WARNINGS for <tmp>:
Useless use of "x: {.x} y: {.y}" in expression "x: {.x} y: {.y}" in sink context (line 1)
tadzik I think softmoth used to hang around here, no?
.seen softmoth
yoleaux I saw softmoth 20 Oct 2015 06:17Z in #perl6: <softmoth> Explicitly specifying modules (what looks like Crust::Runner's current approach) puts the burden on the end user, which is also brittle and tedious, but it works and puts the solution at the same level as the error will show up.
tadzik yeah, years ago :) 15:09
ugexe heh
TreyHarris m: class Point { has $.x; has $.y; }; my $p = Point.new(:3x, :5y); my $out = "x: {.x} y: {.y}" given $p;
camelia ( no output )
TreyHarris Is that the most concise way to interpolate several stringy attributes of the same object? 15:10
m: class Point { has $.x; has $.y; }; my $p = Point.new(:3x, :5y); my $out = "x: {.x} y: {.y}" given $p; say $out;
camelia x: 3 y: 5
15:11 Zoffix joined
Zoffix m: class Point { has $.x; has $.y; }; my $p = Point.new(:3x, :5y); say $p.Capture; 15:12
camelia \(:x(3), :y(5))
TreyHarris And, .sprintf doesn't support named parameters--am I missing the way to do this if I need additional format control?
15:12 dct left
TreyHarris Zoffix: sorry, I wasn't literally talking about like that. Assume 15:12
avuserow TreyHarris, try using `with` instead of `given`
TreyHarris m: class Point { has $.x; has $.y; }; my $p = Point.new(:3x, :5y); my $out = "My x-val is {.x}, and as for y: {.y}" given $p; say $out;
camelia My x-val is 3, and as for y: 5
Zoffix TreyHarris: looks pretty consise to me...
avuserow m: class Point { has $.x; has $.y; }; my $p = Point.new(:3x, :5y); my $out = "My x-val is {.x}, and as for y: {.y}" with $p; say $out; 15:13
camelia My x-val is 3, and as for y: 5
Zoffix `with` won't work with all objects
m: class Point { has $.x; has $.y; }; my $p = Point.new(:3x, :5y); my $out = sprintf "My x-val is %s, and as for y: %s", |$p.Capture<x y>; say $out; 15:14
camelia My x-val is 3, and as for y: 5
TreyHarris Zoffix: agreed. But what if I want to do formatting, is there a pythonish named-parameter variant of sprintf I'm not seeing lurking somewhere?
So I can embed the names in the format rather than in the args?
Zoffix m: class Point { has $.x; has $.y; }; my $p = Point.new(:3x, :5y); my $out = {"My x-val is $:x, and as for y: $:y"}(|$p.Capture); say $out 15:15
camelia My x-val is 3, and as for y: 5 15:16
TreyHarris Zoffix: yep, I've got lots of ways to do straight interpolation. But if I want column and alignment control, sprintf with a capture is the best I'm going to do? 15:17
perlpilot TreyHarris: There's always the Form module. 15:18
Zoffix TreyHarris: .Capture won't always give all the attrs. I'd just use sprintf '%s.... %s', .x, .y, .z with $foo 15:19
buggable: eco Form
buggable Zoffix, Found 37 results: Math::FourierTransform, Text::Diff, BioPerl6, BioInfo, DateTime::Format. See modules.perl6.org/#q=Form
Zoffix buggable: dude, learn to give exact matches, bruh 15:20
TreyHarris Zoffix: yeah. I have a table of formats I need to select on with varying number of columns depending on the underlying implementation
Zoffix No docs.
TreyHarris I can convert the table into multis returning strings across all those implementations, but I'd kinda like to keep them together 15:21
15:23 devmikey joined, iyra joined
TreyHarris perlpilot: Form? I don't see it in ecosystem... I remember it from the apocalypse and exegesis.... 15:24
m: use Form;
camelia ===SORRY!===
Could not find Form at line 1 in:
/home/camelia/.perl6
/home/camelia/rakudo-m-inst-1/share/perl6/site
/home/camelia/rakudo-m-inst-1/share/perl6/vendor
/home/camelia/rakudo-m-inst-1/share/perl6
CompUnit::Reposi…
Zoffix TreyHarris: it *is* in the ecosystem 15:25
15:25 japh__ joined
Zoffix TreyHarris: modules.perl6.org/repo/Form 15:25
TreyHarris Zoffix: oh crap, I was searching on '\bForm', forgot that despite my -i when you use a capital with ack it turns -i off
Zoffix: sorry, that was like one of the first things I looked for before asking anything here, sorry :-/ 15:26
perlpilot TreyHarris: but, as Zoffix said ... no docs. You'd have to read the docs from Perl 5's Perl6::Form
TreyHarris perlpilot: That's fine, I've used it before.
15:28 buggable joined, ChanServ sets mode: +v buggable
eater m: multi sub prefix:<7>(Any --> Any) { say "yo" }; 7 4 15:28
camelia yo
eater :D
15:28 japh____ left
Zoffix eater: I'll raise ya 15:29
m: sub infix:<⁣> { $^a + $^b }; say 2⁣2;
camelia 4
Zoffix :)
eater :')
Zoffix btw what do you think about that MR I sent, too harsh or okay? 15:30
TreyHarris What's the right way to write a method whose multiple underlying OS-dependent native implementations will return either an iterable thingy or a list all at once? Flatten the iterator ones, turn the array ones into iterators, neither and just return what you get assuming people will use it in a way that works either way?
Zoffix buggable: eco Form
buggable Zoffix, Form 'A Perl 6 implementation of Perl 6-style string formatting.': github.com/perl6/form 37 other matching results: modules.perl6.org/#q=Form 15:31
Zoffix yey
TreyHarris Zoffix: nice, thanks
15:31 buggable left, buggable joined, ChanServ sets mode: +v buggable
Zoffix m: say List ~~ Iterable 15:32
camelia True
TreyHarris Zoffix: right. Is this where "Slip" comes in handy?
Zoffix TreyHarris: you mean a Seq by iterable thingy? I'd convert it all to one return value, so it'd be predictable.
15:33 bjz_ left
Zoffix Slip is a List that flattes itself into outer iterables 15:33
m: say Slip ~~ List
camelia True
Zoffix eater: didn't have time to review any of it; other than the $line < 6 stuff I mentioned this morning.
eater Zoffix: that's fine :) It's currently pretty horrible (faking a Test environment and stuff) but yeah take your time 15:34
Zoffix TreyHarris: just stick a .List or .Seq as a coercer
15:35 azawawi joined
TreyHarris Zoffix: some OSes implement a readdir-like one-at-a-time interface, others implement an array-return interface 15:35
azawawi Hi
TreyHarris I could make the readdir ones lazy 15:36
Zoffix eater: well, I don't think I'll be reviewing it. Just merge it. It's just for travis stuff, right? I don't see much need to deliberate it in PR form. Merge it. Iron out the bugs. Done
eater: I sent you an invite to perl6 GitHub org, so you should be able to merge your PR yourself
eater Zoffix: thanks :D
Zoffix TreyHarris: yeah, and return those as Seq. And for those that are all-at-a-time, I'd coerce them to Seqs just to keep up the consistency with the return value. 15:37
Zoffix goes to do things
\o
15:37 Zoffix left
TreyHarris Zoffix++ 15:37
Er, now anyone lurking knows that I agree with Zoffix anyway :-) 15:38
15:39 itaipu joined
eater :D 15:40
15:42 japh__ left
iyra can someone please tell me the most efficient way of using a regex to parse a line starting with : but not a line starting with \:? 15:43
ChoHag /^:/ ? 15:45
sena_kun m: ":haha".starts-with(':')
camelia ( no output )
sena_kun m: say ":haha".starts-with(':')
camelia True
sena_kun m: say "\:haha".starts-with(':')
camelia True
sena_kun m: say "\\:haha".starts-with(':')
camelia False
sena_kun If this check is everything you want, of course. 15:46
iyra thanks a lot, that helps!
I just realised after I posted it that it's obvious :)
sena_kun iyra, yw
and simple string methods are much, much faster than any regex-related work, afaik 15:47
TreyHarris sena_kun: in most cases. Not all. When ratcheting can be usefully applied to reject large swaths of work, the regex will be faster than the string operation unless you write extra stuff that does the equivalent of ratcheting. 15:53
but "starts-with" is definitely not such a case unless the string you pass to start-with is very long (and maybe not even then, lemme see how it's implemented...) 15:54
sena_kun TreyHarris, well, I guess you are right. I am not an expert in such questions, just a useless young folk who sometimes can do some easy-tiredsome work. :) 15:55
s/tiredsome/tiresome/ 15:56
buggable, eco Tardis 15:58
buggable sena_kun, Nothing found
sena_kun eeeeeh.
16:00 devmikey left 16:03 cowens left 16:08 azawawi left 16:09 Hor|zon left
sena_kun how that great and cool and useful channel with Zoffix's bots is called again? I need to assault buggable a bit. 16:13
ah, stop, nevermind.
buggable, eco 007 16:15
buggable sena_kun, 007 'Small experimental language with a license to macro': github.com/masak/007
sena_kun well, it doesn't work in private for me, so the channel still will be greatly appreciated.
MasterDuke_ sena_kun: i think #zofbot 16:19
sena_kun MasterDuke_, thanks a lot
16:20 domidumont left 16:22 imcsk8 joined 16:24 astj joined 16:27 john51_ joined 16:30 Cabanossi left, robertle joined 16:31 Cabanossi joined
iyra how can I sort a hash by key, using a custom sort function (lambda)? 16:44
sena_kun .tell samcv I have much or less updated github.com/perl6/ecosystem/issues/318 - it was done automatically, so troubles may occur, but now we have some pattern to investigate further. And I believe that it is not, well, "our" issue, to be honest. There are some modules that has already working PRs with fixes, but authors just don't merge them. Some libraries are easily abandoned. We need some policy from our side about it, imho, 16:48
yoleaux sena_kun: I'll pass your message to samcv.
sena_kun though I cannot come up with any just now.
.tell samcv also, there are some false positives in your link, please, check it and consider removing of some entries(I've made appropriate comments there). 16:49
yoleaux sena_kun: I'll pass your message to samcv.
sena_kun s/link/links/
16:51 mcmillhj left
sena_kun let's call it a day. thanks for everyone's help today! \o 16:52
16:53 sena_kun joined, Actualeyes left
sena_kun .tell samcv p.s. and I doubt that travis log's are needed. There are false positives("job exceed...", you know what I mean) and false negatives(the package is "green", but with a new rakudo it still fails). 16:54
yoleaux sena_kun: I'll pass your message to samcv.
16:54 sena_kun left 16:58 Cabanossi left
TreyHarris m: my %h = gather {for (1..20) { take $_ => 'f' }}; %h.sort({$^a.key <=> $^b.key}); 17:01
camelia ( no output )
TreyHarris m: my %h = gather {for (1..20) { take $_ => 'f' }}; %h.sort({$^a.key <=> $^b.key}); say %h.perl;
camelia {"1" => "f", "10" => "f", "11" => "f", "12" => "f", "13" => "f", "14" => "f", "15" => "f", "16" => "f", "17" => "f", "18" => "f", "19" => "f", "2" => "f", "20" => "f", "3" => "f", "4" => "f", "5" => "f", "6" => "f", "7" => "f", "8" => "f", "9" => "f"}
17:01 Cabanossi joined
TreyHarris m: my %h = gather {for (1..20) { take $_ => 'f' }}; say %h.sort({$^a.key <=> $^b.key}); 17:01
camelia (1 => f 2 => f 3 => f 4 => f 5 => f 6 => f 7 => f 8 => f 9 => f 10 => f 11 => f 12 => f 13 => f 14 => f 15 => f 16 => f 17 => f 18 => f 19 => f 20 => f)
17:01 astj left
TreyHarris iyra: ^^ ignore the first 17:01
iyra thanks a lot TreyHarris :) 17:03
TreyHarris iyra: there may be a simpler P6ical way, but that's the Perl 5-ish way. If you wanted to sort by values, not keys, you shouldn't use that method though. See docs.perl6.org/type/Hash for examples of that. 17:04
iyra: and, unlike in Perl 5, if you're using a Hash merely for its keys, you should be using Bag, Set, BagHash or BagSet instead. 17:05
MasterDuke_ m: my %h = gather {for (1..20) { take $_ => 'f' }}; say %h.sort(*.key) 17:07
camelia (1 => f 10 => f 11 => f 12 => f 13 => f 14 => f 15 => f 16 => f 17 => f 18 => f 19 => f 2 => f 20 => f 3 => f 4 => f 5 => f 6 => f 7 => f 8 => f 9 => f)
MasterDuke_ m: my %h = gather {for (1..20) { take $_ => 'f' }}; say %h.sort(+*.key)
camelia (1 => f 2 => f 3 => f 4 => f 5 => f 6 => f 7 => f 8 => f 9 => f 10 => f 11 => f 12 => f 13 => f 14 => f 15 => f 16 => f 17 => f 18 => f 19 => f 20 => f) 17:08
TreyHarris MasterDuke_: I do not understand. Why does hyperwhatever cause a non-cmp sort? Or is something else happening? 17:10
MasterDuke_ it defaults to string comparison 17:11
m: my %h = gather {for (1..20) { take $_ => 'f' }}; say %h.sort({$^a.key cmp $^b.key});
camelia (1 => f 10 => f 11 => f 12 => f 13 => f 14 => f 15 => f 16 => f 17 => f 18 => f 19 => f 2 => f 20 => f 3 => f 4 => f 5 => f 6 => f 7 => f 8 => f 9 => f)
jnthn The assignment to %h stringifies the keys 17:14
So while sort may use cmp, which will happily sort integers, the keys are not integers any more by that point.
m: my %h{Int} = gather {for (1..20) { take $_ => 'f' }}; say %h.sort(*.key) 17:15
camelia (1 => f 2 => f 3 => f 4 => f 5 => f 6 => f 7 => f 8 => f 9 => f 10 => f 11 => f 12 => f 13 => f 14 => f 15 => f 16 => f 17 => f 18 => f 19 => f 20 => f)
MasterDuke_ better explanation 17:17
TreyHarris I understand that bit. But why does a string-keyed hash sort numerically when **.key is used?
17:17 nadim joined
TreyHarris Oh! That's +*.key.... my eyesight.... gah 17:18
MasterDuke_ +*, not **
17:18 mcmillhj joined
jnthn :) 17:18
TreyHarris And I was embarrassingly about to try using bisectable6 to figure out when that started working b/c it doesn't work on my installation.
TreyHarris sighs 17:19
17:29 Cabanossi left 17:31 Cabanossi joined 17:32 wamba left 17:39 astj joined 17:43 iyra left 17:48 roguelazer joined 17:49 labster joined 17:53 ufobat_ joined 18:07 dct joined 18:12 astj left 18:22 shdb joined 18:23 iyra joined 18:24 dct left 18:37 domidumont joined
timotimo a little much to backlog today, so i'll skip some parts 18:42
El_Che #perl6. The good parts 18:47
19:01 itaipu left 19:02 gdonald joined
moritz just offers #perl6. The comprehensive guide. 19:03
timotimo what options do i have on an ios phone to offer a website and have it pretend to be an app? 19:14
19:15 nadim left
timotimo i've found the "add to home screen" button in safari, which is a start 19:15
El_Che that's it 19:16
timotimo m( 19:17
El_Che there are certainly frameworks to turn a webpage into a "real" app, but you'll need to pay to get it into the closed garden 19:18
timotimo my mom wants a super simple thing, and i'm def not going to shell out what 60 dollars or so for a developer's license and find a macbook to work with
haha, i'll develop it for Nintendo DS and give her my old device :D
El_Che well, it depends if you love your mom or not :) 19:19
timotimo don't question my love for my mommy 19:20
El_Che true 19:21
timotimo i'm going to do javascript for her sake
El_Che she can play mario as well
so everyone wins!
timotimo arguably a fate worse than death
XML is broken because it's using IO in a signature 19:31
oh 19:32
it wasn't XML
19:32 dct joined
timotimo er, or it was XML and the function in question just wasn't tested in XML itself 19:33
what the heck 19:34
iyra can someone please help me understand what $x.Int returns if $x can't be converted to an integer, and how to check for it? 19:38
timotimo m: say "hello".Int.perl 19:39
camelia Failure.new(exception => X::Str::Numeric.new(source => "hello", pos => 0, reason => "base-10 number must begin with valid digits or '.'"), backtrace => Backtrace.new)
timotimo m: say ("hello" // -1) 19:40
camelia hello
timotimo oops
m: say ("hello".Int // -1)
camelia -1
timotimo m: say ("hello".Int ~~ Nil)
camelia True
timotimo any of those ways will be fine, iyra
iyra thanks :) 19:41
TimToady PSA: the File::Ignore module is working again (and maybe any other modules that were failing due to use of the <( )> construct)
.tell sena_kun File::Ignore is working again
yoleaux TimToady: I'll pass your message to sena_kun.
19:45 Cabanossi left 19:46 Cabanossi joined
iyra timotimo, what is a good way of checking if multiple variables can be converted to Int or not? I have 7 or 8 variables to check this way 19:46
[Coke] should they be separate variables or in an array/ 19:47
iyra i have them as separate variables 19:48
eater theres not zef init :(
AlexDaniel m: say ("58", "25c", "60).any.Int ~~ Nil
camelia 5===SORRY!5=== Error while compiling <tmp>
Unable to parse expression in double quotes; couldn't find final '"'
at <tmp>:1
------> 3say ("58", "25c", "60).any.Int ~~ Nil7⏏5<EOL>
expecting any of:
double quotes
t…
AlexDaniel m: say ("58", "25c", "60").any.Int ~~ Nil
camelia True
AlexDaniel m: say ("58", "25", "60").any.Int ~~ Nil
camelia False
[Coke] m: my @a = 1.0,3/2,4/2,1.5,π; say @a.map{$_.narrow}.perl;
iyra thanks!
camelia Cannot resolve caller map(Array: ); none of these signatures match:
($: Hash \h, *%_)
(\SELF: &block;; :$label, :$item, *%_)
(HyperIterable:D $: &block;; :$label, *%_)
in block <unit> at <tmp> line 1
AlexDaniel iyra: maybe junctions again 19:49
iyra junctions are one thing I think are most confusing about perl 6 for me, anyway
[Coke] m: my @a = 1.0,3/2,4/2,1.5,π; say @a.map({$_.narrow}).perl;
camelia (1, 1.5, 2, 1.5, 3.14159265358979e0).Seq
[Coke] there's a difference between .narrow and ~~ Int
so, depending on which question you're trying to answer... 19:50
m: my @a = 1.0,3/2,4/2,1.5,π; say @a.grep({$_.narrow ~~ Int}).perl;
camelia (1.0, 2.0).Seq
iyra testing converting to Int is what I care about, but I don't really care about the result
AlexDaniel junctions then \o/ 19:51
[Coke] then you want that last one - is the .narrow'd result ~~ Int.
iyra thanks
[Coke] You probably don't want junctions.
(unless you only care about one of the variables)
(or all of them) 19:52
19:53 domidumont left, dct left 19:54 gdonald left, gdonald joined 19:58 itaipu joined
TreyHarris m: my @a = 1.0,3/2,4/2,1.5,π,i,i*i; say @a.map({$_.narrow.WHAT}) 20:05
camelia ((Int) (Rat) (Int) (Rat) (Num) (Complex) (Int))
TreyHarris m: my @a = 1.0,3/2,4/2,1.5,π,i,i*i; say @a.map({($_.WHAT, $_.narrow.WHAT)}) 20:08
camelia (((Rat) (Int)) ((Rat) (Rat)) ((Rat) (Int)) ((Rat) (Rat)) ((Num) (Num)) ((Complex) (Complex)) ((Complex) (Int)))
20:08 skids left 20:13 astj joined
iyra AlexDaniel, it's strange, I can't get something like this to work: if (($year, $month, $day, $hour, $min, $sec).any.Int ~~ Nil) { ... } 20:14
I get error: "No such method 'Int' for invocant of type 'Any' 20:15
" instead
20:15 mcmillhj left
AlexDaniel iyra: one of your variables is Any 20:15
geekosaur betting they were expecting undefined to be Nil instead of Any 20:16
20:17 astj left 20:21 devmikey joined
iyra yeah, thanks, some of my variables were undefined.. 20:22
20:23 itaipu left
timotimo phew, i was somehow working with an outdated version of exemel even though i was kind of certain i had gotten the newest one 20:25
crisis averted
iyra why can't I index hashes using strings? For example, if i have: my %langs = { 'en' => 'English', 'ja' => 'Japanese' }; then why doesn't %langs<"en"> return the value? Instead it return (Any) 20:29
ilbelkyr you want %langs<en>, or %langs{"en"}
iyra I have to use %langs<en> but I don't know how to do that if I'm using a string variable to index (such as %langs<$my_language_code>)
ah, okay, thanks 20:30
I thought they were equivalent for some reason
TreyHarris m: my $foo = "hiya"; my %h; %h<$foo> = 3; say %h.perl 20:31
camelia {"\$foo" => 3}
TreyHarris iyra: <...> indexing is for literals
{} indexing for anything else
geekosaur <> is like single quotes, << >> is like double quotes, { } is any expression 20:32
iyra interesting, thanks for the explanation
geekosaur or « » if you hate texas :p
timotimo you're unfortunately* not allowed to use any unicode quoting pair here :P 20:33
20:33 TEttinger joined 20:35 espadrine joined
TreyHarris m: my $foo = "hiya"; my %h; %h<$foo> = 3; %h<<$foo>> = 9; %h{$foo ~ "there"} = 2; say %h.perl 20:36
camelia {"\$foo" => 3, :hiya(9), :hiyathere(2)}
TreyHarris m: my $foo = "hiya"; my %h<<$foo ~ "there">> = 7; say %h.perl 20:38
camelia 5===SORRY!5=== Error while compiling <tmp>
Shaped variable declarations not yet implemented. Sorry.
at <tmp>:1
------> 3y $foo = "hiya"; my %h<<$foo ~ "there">>7⏏5 = 7; say %h.perl
TreyHarris oops
m: my $foo = "hiya"; my %h; %h<<$foo ~ "there">> = 7; say %h.perl
camelia {:hiya(7), :there(Any), "~" => Any}
20:38 mcmillhj joined
iyra if $year is a string, is it valid to make $year into an Int (assuming it works) by doing $year = $year.Int? or will $year still by of type Str? 20:39
TreyHarris the point is, the three aren't interchangeable except for <<$var>> and {$var}...anything more complicated than that is no longer equivalent
geekosaur m: my $y = '5'; $y .= Int; dd $y 20:40
camelia Int $y = 5
iyra thank you :D
geekosaur although if you declared it as my Str $year then it'd throw an error 20:42
iyra TreyHarris, an example in the docs is: DateTime.new(:year<10000>, :month<2>).days-in-month; but I seem to be getting errors when I do something like: DateTime.new(:year<$year>, :month<$month>).days-in-month; even though I know $year and $month are integers 20:45
samcv .
yoleaux 16:48Z <sena_kun> samcv: I have much or less updated github.com/perl6/ecosystem/issues/318 - it was done automatically, so troubles may occur, but now we have some pattern to investigate further. And I believe that it is not, well, "our" issue, to be honest. There are some modules that has already working PRs with fixes, but authors just don't merge them. Some libraries are easily abandoned. We need some policy from our side about it,
iyra I also tried with {} instead of <> with no luck
yoleaux imho,
16:49Z <sena_kun> samcv: also, there are some false positives in your link, please, check it and consider removing of some entries(I've made appropriate comments there).
samcv .
yoleaux 16:54Z <sena_kun> samcv: p.s. and I doubt that travis log's are needed. There are false positives("job exceed...", you know what I mean) and false negatives(the package is "green", but with a new rakudo it still fails).
samcv .tell sena_kun thanks! 20:46
yoleaux samcv: I'll pass your message to sena_kun.
TimToady TreyHarris: even those aren't interchangeable 20:48
m: my %h = :1a, :2b, :3c, 'a b c' => 42; my $var = "a b c"; say %h«$var»; say %h{$var}
camelia (1 2 3)
42
TreyHarris iyra: that's because :year<2017> isn't a hash index, it's a pair literal. you write DateTime.new(year => $yearvar) 20:49
TimToady m: my %h = :1a, :2b, :3c, 'a b c' => 42; my $var = "a b c"; say %h«'$var'»; say %h{$var}
camelia (Any)
42
TreyHarris TimToady: yes...but I was worried that would confuse iyra even more.
TimToady m: my %h = :1a, :2b, :3c, 'a b c' => 42; my $var = "a b c"; say %h«"$var"»; say %h{$var}
camelia 42
42
TreyHarris m: DateTime.new(year => "2001") 20:52
camelia ( no output )
TreyHarris m: say DateTime.new(year => "2001")
camelia 2001-01-01T00:00:00Z
TreyHarris iyra: and for the record, DateTime is cool so this works
iyra do other functions work in the same way with named arguments? or is it just datetime's new method? 20:53
TreyHarris didn't the bot at one point print an evaluation of the last expression in sink context?
iyra: see the documentation for Cool. 20:54
iyra thanks, I'll take a look
TreyHarris In Perl 5 you generally didn't need to worry about what was a number and what wasn't--except when you did. Cool is how Perl 6 does the same (in a way requiring less special-casing) 20:55
timotimo and in perl6 looking at a variable can't make it flip what it thinks it is 20:56
TreyHarris iyra: See NumStr if you absolutely must control this (but you almost certainly don't need to) 20:57
[Coke] zoffix, do you still plan to make 'xtest' pass? 20:58
TreyHarris m: my $x = "1999"; say ++$x; say $foo.WHAT 21:02
camelia 5===SORRY!5=== Error while compiling <tmp>
Variable '$foo' is not declared
at <tmp>:1
------> 3my $x = "1999"; say ++$x; say 7⏏5$foo.WHAT
TreyHarris ack
m: my $x = "1999"; say ++$x; say $x.WHAT
camelia 2000
(Str)
TreyHarris That's slightly surprising to me 21:03
Ahh, all is explained in docs.perl6.org/routine/succ 21:05
If I _wanted_ a strictly lexicographic succ to "1999", i.e., "199:", could I get it easily without stripping the string apart? 21:07
21:07 iyra left, iyra joined 21:09 mcmillhj left
timotimo don't know of a way to make that happen with any single method we have 21:09
TreyHarris timotimo: it doesn't seem useful. i was just curious. Before Cool, like way back in pre-Synopsis days, initially the idea would have been that multi dispatch would have chosen to succ numerically because the string looks numeric, right, so you would have been able to force it by specifying which succ you wanted, but with succ as a method on Str that checks if it looks numeric, you can't? 21:12
timotimo true 21:14
21:19 kent\n joined 21:20 emeric joined 21:22 nadim joined 21:23 itaipu joined
robertle what are people doing to determine test coverage? I found MVM_COVERAGE_LOG, but how di I convert the output to something more understandable? is there anything else? 21:23
timotimo there's moarvm's tools/parse_coverage or whatever it's called 21:25
21:25 Zoffix joined
timotimo it's ... kinda rough still 21:25
Zoffix robertle: there's a slightly dated blog post describing the procedure: perl6.party/post/Perl-6-Core-Hacki...Moar-Cover
robertle: this is the updated version of the script mentioned in the blog post that I used to generate coverage for perl6.WTF gist.github.com/zoffixznet/09a7eb7...a2e0d3dfa3 21:26
robertle: and that's just for core stuff. More work needs to make it show coverage for any script
robertle: I think to make it show coverage, it could work by just tweaking the coverage reporter script to look for lines from file? Like right now it looks for setting files: github.com/MoarVM/MoarVM/blob/mast...rt.p6#L128 21:27
21:28 itaipu left
timotimo well, it could just ignore #?file lines, right? 21:29
then you'd still have to run it once for each file of you rcode
though what you could totally do is concatenate your code files and put the #?file lines in there manually, and also concatenate the moar --dump output for each file there 21:30
that's what i mean by rough :D
hm, did i write my more precise coverage stuff before or after the release ... 21:31
Zoffix timotimo: after 21:32
timotimo um, i didn't even merge it to master yet! 21:34
that's a bad miss %)
Zoffix m: .[*-1]++, dd $_.».chr.join with '1999'.ords.Array 21:35
camelia "199:"
Zoffix oh oh
Zoffix rakes the toolbox for .substr-rw
robertle tools/parse_coverage_report.p6 looks promising, but I seem unable to run it correctly: gist.github.com/anonymous/5f33670d...e5fcfcb2d0 21:36
what's the "annotations" argument meant to do?
timotimo it uses what moar --dump outputs to get a clear picture of what lines even have code in 'em
so the percentage can actually reach 100%
Zoffix m: my $x = '1999'; $x.substr-rw($x.chars-1, 1) = $x.comb.tail.ord.&[+](1).chr; dd $x
camelia Str $x = "199:"
Zoffix hehe neat
timotimo can't you (*-1,1) for substr-rw? 21:37
Zoffix Oh, probably
timotimo actually, isn't that also equivalent to (*-1)?
Zoffix m: my $x = '1999'; $x.substr-rw(*-1) = $x.comb.tail.ord.&[+](1).chr; dd $x
camelia Str $x = "199:"
Zoffix timotimo++
timotimo because if you only have one arg it's "to the end of string"
yay
oh but wait there's more
Zoffix m: my $x = '1999'; $x.substr-rw(*-1) .= ord.=&[+](1).=chr; dd $x 21:38
camelia Str $x = "199\x[6]7"
Zoffix heh
timotimo yes!
wait, no :D
m: my $a = "hello"; my $x = $a.substr-rw(2); $x .= uc; $x .= flip; say $a 21:39
camelia hello
timotimo m: my $a = "hello"; my $x := $a.substr-rw(2); $x .= uc; $x .= flip; say $a
camelia heOLL
timotimo m: my $a = "hello"; my $x := $a.substr-rw(2); $x .= uc.= flip; say $a
camelia heOLL
timotimo why doesn't it work properly in your case?
Zoffix m: my $x = '1999'; $_ = chr .ord + 1 with $x.substr-rw: *-1; dd $x
camelia Str $x = "199:"
timotimo that's the cleanest so far, nice! 21:40
but wait
Zoffix timotimo: it assigns the .ord to string and then continues with the 6 or whatever
timotimo let's try .ord.succ
Zoffix m: my $x = '1999'; $_ = .ord.succ.chr with $x.substr-rw: *-1; dd $x
camelia Str $x = "199:"
robertle so how do I create this annotations file for some perl6 code? this seems to work at a slightly lower level
TreyHarris geekosaur: I may have been netsplit because I can't find it in scrollback, but lucs tells me you answer my "$x .= not" question. Thanks
timotimo you'll have to perl6 --target=mbc --output=something.moar mycode.p6
then you can moar --dump something.moar 21:41
robertle which is confusing, the output produced by MVM_COVERAGE_LOG seems to reference lines in my code! why do I need the annotations?
timotimo they are kind of optional
robertle ok, will try
timotimo the coverage parser tool could run without
it won't give proper percentages in that case, though
Zoffix robertle: I don't think you *need* them. You'll just get unreachable lines marked as uncovered
timotimo because it'll think every empty line or comment is not being covered by your tests
but it seems like if you don't specify the annotation file it'll complain 21:42
robertle: an empty file will work just fine btw
Zoffix It will just `note` a warning message without the annotations: github.com/MoarVM/MoarVM/blob/mast...rt.p6#L140
21:43 Zoffix left
timotimo robertle: could you change a line for me in the coverage tool? 21:44
robertle timotimo: totally 21:45
timotimo try putting Any:D in front of $ann where ...
in line 125
robertle better, now I can run it without the --annotations arg! thank! 21:46
timotimo okay, i'll commit that to moarvm
robertle but no matter whether I do that or pass an empty file, it finishes now and says "Coverage report read: 76099 lines covered."
nadim Hi all, is there a way to dump a grammar? .perl gives it's name, s .gist say it is a grammar, and stringification ruturns Any.
robertle but the resulting coverage report is empty, somehow it must fail to match the lines in the coverage input to the source provided?
another minor thig: if the "coverage" output directory does not exist, it is not created but the tool exits with a semi-helpful message 21:47
the message is actually not that bad all things considered, but...
timotimo btw, you will want to pull master from moarvm now, it has a fix that makes the line coverage more precise 21:48
(it used to skip some lines)
hah, it would only have accepted annotations from SETTING: anyway
robertle hah! I noticed that! I though it skips lines that can be inferred from the previous ones, where tehre is no possible branches...
timotimo okay can you put a line like this in the beginning of the source file you're trying to cover: 21:49
#line 1 SETTING:src/blah
robertle only for parse_coverage_report.p6 or do I need that when creating the coverage list itself? 21:52
timotimo oh, every line from the annotations file and from the coverage report will be offset by one
MasterDuke_ nadim: what do you mean "dump a grammar"? a grammar is a class. or do you mean dump the results of parsing something with a grammar? 21:53
robertle timotimo: still no data in the final coverage report 21:54
nadim I wrote Data::Dump::Tree, it dumps Matches but it is the definition of the grammar itself I'd like to dump
MasterDuke_ what does D::D::T do for "regular" classes? 21:56
nadim it dumps its guts
jnthn We don't keep the source code around to compile time, so there's no way to dump the source 21:57
Though I think for regexes we do keep it around as a special case
Which is a bit of a dubious exception
timotimo robertle: one more try:
nadim anyway to gets to its components programmatically?
timotimo can you put the same kind of line in the end of your file? with a different mock-filename, please
jnthn Like what rules it has?
Those are methods, so .^methods 21:58
MasterDuke_ jnthn: just `regex`s, or `rule`s and `token`s also? for keeping the source that is
jnthn MasterDuke_: I'm not sure
timotimo m: regex a { hello there }; say &a.perl;
camelia 5===SORRY!5===
Undeclared routine:
a used at line 1

Other potential difficulties:
Space is not significant here; please use quotes or :s (:sigspace) modifier (or, to suppress this warning, omit the space, or otherwise change the sp…
timotimo m: regex a { hello there }; say a.perl;
camelia 5===SORRY!5===
Undeclared routine:
a used at line 1

Other potential difficulties:
Space is not significant here; please use quotes or :s (:sigspace) modifier (or, to suppress this warning, omit the space, or otherwise change the sp…
robertle timotimo: sure! are you positive that it's a single colon after SETTING? I noticed other (core) entries in my coverage that have SETTING::...
nadim but ethods will not give the structure I guess
jnthn No
that said 21:59
timotimo oh you're right, it has to be two :
nadim OK, I put that idea back in the freezer
jnthn You can do .file and .line on a method I think
And then if the source file exists track it down
timotimo m: my regex blah { "hello there" }; say &blah.perl;
camelia regex blah { "hello there" }
timotimo ^- there it is
m: say / foo .* bar /.perl 22:00
camelia / foo .* bar /
timotimo there as well
MasterDuke_ m: my rule blah { "hello there" }; say &blah.perl;
camelia rule blah { "hello there" }
MasterDuke_ m: my token blah { "hello there" }; say &blah.perl; 22:01
camelia token blah { "hello there" }
TreyHarris and TimToady's '?= True' solution--but I think I'll take the 50% execution time hit for the clearer '.= not' solution. Though given that "$obj.truth .= not" is 12 times the execution that "$obj.truth = not $obj.truth" (!?!), maybe I'll just use the wordy one after all.
er, '?^= True' that should have been 22:02
22:02 dct joined
samcv does anyone know what caused gentoo to fail instlaling rakudo? 22:02
and if there is a fix?
timotimo no clue, do you have some output to look at? 22:03
TimToady m: say '1999' ... '~~~~' # TreyHarris: but note this
camelia (1999 199: 199; 199< 199= 199> 199? 199@ 199A 199B 199C 199D 199E 199F 199G 199H 199I 199J 199K 199L 199M 199N 199O 199P 199Q 199R 199S 199T 199U 199V 199W 199X 199Y 199Z 199[ 199\ 199] 199^ 199_ 199` 199a 199b 199c 199d 199e 199f 199g 199h 199i 199j 199…
nadim timotimo: doesn't work with a Grammar, it says "Any"
samcv ./libmoar.so: undefined reference to `mp_set_long'
timotimo can you show a bit of example code, nadim?
samcv ebuild gitweb.gentoo.org/repo/gentoo.git/....03.ebuild
tyil, is having issues
TimToady m: say '1999' ... '::::'
camelia (1999 199: 19:9 19:: 1:99 1:9: 1::9 1::: 2999 299: 29:9 29:: 2:99 2:9: 2::9 2::: 3999 399: 39:9 39:: 3:99 3:9: 3::9 3::: 4999 499: 49:9 49:: 4:99 4:9: 4::9 4::: 5999 599: 59:9 59:: 5:99 5:9: 5::9 5::: 6999 699: 69:9 69:: 6:99 6:9: 6::9 6::: 7999 799: 79:…
samcv and i remember someone else came in a week ago
Geth doc: 61e749f14c | (Zoffix Znet)++ | 5 files
Fix white-space and nbsp xtests
timotimo m: grammar Test { rule A { hey }; rule B { yo } }; for Test.^methods -> $m { say $m.perl }
camelia rule A { hey }
rule B { yo }
method subparse (Grammar $: \target, :$rule, :$args, Mu :$actions, *%_) { #`(Method|33440992) ... }
method parsefile (Grammar $: Cool $filename, :$enc, *%_) { #`(Method|33441144) ... }
method parse (Grammar $: \target…
tyil samcv: it happens on .02 and .01 as well 22:04
timotimo nadim: when you do it like this, it'll work
jnthn samcv: At a wild guess, is it using a packaged libtommath?
tyil same error, all on the moarvm package
samcv that's what i thought jnthn
jnthn What happens if we make it use the bundled one?
samcv is there any other ebuild though?
i don't see any --use-* things in that ebuild
TreyHarris TimToady: because a range doesn't iterate over "succ"?
timotimo hm, does --backends=moar, work? 22:05
TimToady not when the strings are the same length
TreyHarris ohh
TimToady m: say '000' ... '???'
camelia (000 001 002 003 004 005 006 007 008 009 00: 00; 00< 00= 00> 00? 010 011 012 013 014 015 016 017 018 019 01: 01; 01< 01= 01> 01? 020 021 022 023 024 025 026 027 028 029 02: 02; 02< 02= 02> 02? 030 031 032 033 034 035 036 037 038 039 03: 03; 03< 03= 03> 0…
jnthn samcv: I have no idea, I just have a hazy memory about that function not being in some older version of libtommath
samcv yep
that is 100% accurate
TimToady so there's hex in a weird notation
m: say '000' ... '777' 22:06
camelia (000 001 002 003 004 005 006 007 010 011 012 013 014 015 016 017 020 021 022 023 024 025 026 027 030 031 032 033 034 035 036 037 040 041 042 043 044 045 046 047 050 051 052 053 054 055 056 057 060 061 062 063 064 065 066 067 070 071 072 073 074 075 076 0…
TimToady but it's more for octal and such
timotimo Date: Wed Dec 10 18:59:31 2014 +0100
add mp_get_long_long() and mp_set_long_long()
oh, no, wait
Date: Sun Nov 13 19:17:46 2011 +0100 22:07
Add set_long and get_long functions
that's really old
what the heck is gentoo shipping? :)
jnthn samcv: gitweb.gentoo.org/repo/gentoo.git/...ebuild#n45
It's doing it in the MoarVM ebuild
timotimo yeah, bad idea. that'll also get you an expmod that can infinitely recurse (or iterate?) 22:08
samcv version 1.0?
idk packages.gentoo.org/packages/dev-l...libtommath
looking.
TreyHarris Why do the .= not and ?^= True solutions take so much longer than the fully-written-out one? Because simple assignment doesn't have lots of dispatching possibilities?
timotimo +SRC_URI="github.com/libtom/libtommath/relea...V}.tar.xz"
so that must be a 1.0 tarball then? 22:09
but v1.0 is from 2016 .. ?!
-- Moritz Lenz contributed a fix to mp_mod()
tyil 1.0 seems the latest available in ports
timotimo and provided mp_get_long() and mp_set_long()
i know that guy!
tyil but 0.4 is marked as the latest stable
robertle timotimo: got it! it's a bit fickle, but that does work!
timotimo oh, look
tyil 0.42 *
timotimo yeah, 0.42 is from 2010 22:10
tyil Im compiling 1.0 rn
nadim m: grammar XXX { token TOP { abc } } ; say &XXX.perl ;
tyil see if that helps
camelia 5===SORRY!5=== Error while compiling <tmp>
Illegally post-declared type:
XXX used at line 1
22:10 dct left
timotimo robertle: yeah, we'll rework the tool so it can be used for usercode 22:10
the reason why it requires SETTING::src/ is because i didn't want the reports to show what lines in the test suite were covered
samcv who is in an admin on the perl6 group? we need to enable travis for travis-ci.org/perl6/ecosystem
timotimo it is done 22:11
samcv \o/
tyil moar builds with libtommath 1.0
lets see if nqp and rakudo itself also work now
timotimo tyil: cool, we should tell the maintainer of the ebuild to require that version
nadim m: grammar MY::GR { token TOP { abc } } ; say &MY::GR.perl ;
camelia 5===SORRY!5=== Error while compiling <tmp>
Cannot use pseudo package MY in package name
at <tmp>:1
------> 3grammar MY::GR7⏏5 { token TOP { abc } } ; say &MY::GR.per
expecting any of:
generic role
timotimo nadim: no, you don't get methods like that 22:12
tyil timotimo: I'll send out a mail to him
timotimo also, GR doesn't have a & sigil
m: grammar My::Gr { token TOP { abc } }; say My::Gr.perl
camelia My::Gr
timotimo m: grammar My::Gr { token TOP { abc } }; say My::Gr.^methods.grep(*.name eq "TOP").perl
camelia (token TOP { abc },).Seq
timotimo TYVM, TYIL 22:13
TreyHarris TimToady: my recollection of how you originally envisioned numeric/str polymorphism was correct, right? Multi dispatch via value dispatch that would make a looks-like-number string closer to the numeric multi than the string multi?
nadim right, it's a class with methods
timotimo but the error you were getting was because you're not allowed to use MY like that
nadim which means we can't dump anything that makes sense
timotimo has to be My or my or MEH
nadim timotimo: I didn't even know MY was special :) 22:14
timotimo it's a pseudopackage that has your lexical variables in it
22:14 dct joined
nadim timotimo++ 22:14
timotimo m: my $test; sub inner { my $innertest; say MY::.keys }; inner()
camelia ($_ $/ $innertest $*DISPATCHER $! $¢)
timotimo MY only has the one that are defined in the current lexical scope 22:15
tyil timotimo: sent, lets see how quickly it updates :> 22:16
timotimo \o/
samcv tyil, link? 22:17
tyil samcv: its an email
samcv also would be good if it didn't --use-libffi as well. since that backend is less well tested
ok. fw me
tyil its already sent
:'D
samcv that's why it's called a forward
:)
tyil are you implying I keep records of sent mail? 22:18
samcv yes
tyil you are correct
nadim jnthn: I send a pull request for Grammar::Tracer some times ago, there's another pull request there btw, I think I'll just create another module. I was thinking about having it under Grammar::Tracer::something namespace, is that ok for you?
TreyHarris TimToady: I ask just because I wondered if the ultimate (/current) solution has any functional disadvantages you miss, despite its advantages
tyil samcv: [email@hidden.address] right?
jnthn nadim: I think you wanted to take the output format in a direction I disagreed with, so that's probably best.
samcv oh btw everyone i'm running another full module build hack.p6c.org/~samcv/install_log_live2.txt right now
no 22:19
[email@hidden.address]
tyil >ham
nice
nadim jnthn: it's just a different output format, I prefer mine because it is much more compact, it's just a question of taste, functionally it is the same
tyil samcv: sent 22:20
jnthn nadim: Maybe we could make it an option?
timotimo nadim: is there an example where i can see that output format in action?
nadim sure, 1mn to post a screenshot
jnthn use Grammar::Tracer (format => 'foo') or something
22:22 bjz joined
tyil samcv: if I make any module, I'll be sure to use "perl" as the license name 22:23
samcv XD
22:23 wamba joined
TreyHarris samcv: I just created a new package with mi6--it creates an Artistic License 2.0 by default but doesn't give any license info in the metadata. 22:24
samcv thanks for the heads up TreyHarris 22:25
Geth doc: 06cb6c9a93 | (Zoffix Znet)++ | doc/Type/IO/Path.pod6
Fix for space-after-comma xtest
doc: 144f0be23e | (Zoffix Znet)++ | 2 files
Fix typo in xtest filename
TreyHarris samcv: your massive PR was scanning for LICENSE files when there was no metadata though, wasn't it? 22:26
samcv no. not *no* metadata. just wrong or nonstandard naming
nadim timotimo: imgur.com/3SnOKm9 that's the grammar that was in one of the advent calender entries
samcv will have to do ones for no metadata as well though
TreyHarris samcv: ahh. in this case then you'd not have fixed it then. it creates a LICENSE file with Artistic 2 filled out 22:27
samcv but for now i'm correcting ones from like "perl" >> "Artistic-2.0" or "The Artistic License 2.0" >> "Artistic-2.0" or
timotimo nadim: not a bad look
samcv a url to the license and replace it with the license identifier
nadim I am actually thinking about outputing it via Data::Dump::Tree to get some filtering and highlighting options but I am afraid the buffering will make it slow for large ASTs
timotimo: if the string get capped it lookes even better 22:28
timotimo right, with backtracking and such it might very well give a few thousand lines of output easily
samcv TreyHarris, next step would be applying the script i have to ones with *no* license field, and checking the license file in the project etc
TreyHarris nods.
samcv the PR i sent out, took the field in the metadata, and corrected it. and if the metadata was not precise "artistic" or "perl" would check the license file and ensure there was only one license/licence/copying file and also check the text matched
Geth doc: 9efb25fda1 | (Zoffix Znet)++ | doc/Language/io-guide.pod6
Reword
 22:29
nadim timotimo: I am trying to optimize Date::Dump::Tree but my last effort made little to no difference, althought in my example I reduced the number of total call with 3-4%
eater how does NativeCall work with an Enum as argument?
TreyHarris I think the other thing I've seen--certainly on CPAN--is no license file but a LICENSE section of POD (usually only used for "Module::Name is made available under the same licensing terms as Perl itself.")
samcv eater, you need to make an enum. you should try gpxtrixie
gptrixie* 22:30
nadim timotimo: it's worth a shot, I'll create a few classes for the tracer, I could also add some statistics about what matched and what didn't and how many times in the dump 22:31
jnthn: if you want it as an option, you have the pull request already, although it may need to be rebased 22:32
22:33 cpage_ joined
timotimo nadim: i wonder if it'd be interesting to have a view at the target string showing little flags where different regexes were trying to parse 22:34
robertle timotimo: one last quetion: how do I get parse_coverage_report to look at multiple source? or merge the results?
TreyHarris Oh, dear... a user (not a bot I don't think, the wording varies) is responding to every single mesage sent by anyone on Twitter asking @United for customer service pointing them at my story with a different disability-rights hashtag every time. This is troubling since I didn't authorize it. (If it's any of you, don't tell me, it's probably best I don't know, but please stop it, it isn't helping.) 22:36
!!!
...and that was the wrong channel, I'm SO sorry
timotimo yeah, it's hard when people on twitter just go ahead and do whatever while referencing you 22:37
there's a person talking about perl6 and NSA and xkeyscore and i'm like "okay, but ... what?"
TreyHarris timotimo: :-P 22:38
TreyHarris hides under a big Perl-shaped rock.
timotimo camelia lets you hide under her wings
(two meters wingspan, after all!)
sorry, it's three meters 22:39
nadim timotimo: it certainly would be nice, Damian had a regex debugger that I like, may be something like that would be nice, do you know which one I mean?
timotimo en.wikipedia.org/wiki/Homona_coffearia - um, why didn't i know before that a camellia exists that is actually a moth?
nadim: no clue. for reference, i do zero perl5 at all
gfldex is &slurp lazy?
timotimo nope 22:40
we don't have lazy strings yet
22:40 iyra left 22:41 r3m joined
gfldex m: my @fs = <file1.ext file2.ext file3.ext>; dd [//] try (@fs».IO».slurp); 22:41
camelia Failure.new(exception => X::AdHoc.new(payload => "Failed to open file /home/camelia/file3.ext: no such file or directory"), backtrace => Backtrace.new)
TreyHarris would bare slurp be lazy even if we did? I'd think that would be an option or a different method
gfldex does [//] shurt circuit? 22:42
timotimo even if it does, it can't properly thunk the list you're writing, so any code you write after the [//] will be evaluated to create the list [//] will operate on (is my assumption)
22:43 iyra joined
iyra I hope I'm not being too much of a bother, can anyone help me understand why this is returning False? "TITLE: a" ~~ m/^':'(\w+)' '(.+)$/ 22:44
timotimo because there's no ':' at the beginning of a line
iyra ah crap, you're right, thanks..
timotimo :)
nadim timotimo: for reference search.cpan.org/~dconway/Regexp-Deb...ebugger.pm
22:45 robertle left
timotimo heatmapped visualization log! 22:45
nadim: is there a screenshot or screencast you can recommend? 22:46
22:46 cpage_ left 22:47 Garland joined
gfldex m: my @fs = <file1.ext file2.ext file3.ext>; my $s = slurp([//] try (@fs».IO».open)); 22:47
camelia Failed to open file /home/camelia/file3.ext: no such file or directory
in block <unit> at <tmp> line 1

Actually thrown at:
in block <unit> at <tmp> line 1
timotimo searching google for rxrx screencast gives me a bunch of random semi-screencast-related videos
gfldex that's likely more sensible
22:47 cpage_ joined
gfldex m: my @fs = «$*HOME/.gitconfig $*HOME/.config/git/config»; my $s = slurp([//] try (@fs».IO».open)); 22:49
camelia Failed to open file /.config/git/config: no such file or directory
in block <unit> at <tmp> line 1

Actually thrown at:
in block <unit> at <tmp> line 1
gfldex that's what I actually want to do
m: say $*HOME 22:50
camelia "/home/camelia".IO
gfldex m: my @fs = «$*HOME/.gitconfig $*HOME/.config/git/config»; dd @fs; my $s = slurp([//] try (@fs».IO».open));
camelia Array @fs = ["/home/camelia", "/.gitconfig", "/home/camelia", "/.config/git/config"]
Failed to open file /.config/git/config: no such file or directory
in block <unit> at <tmp> line 1

Actually thrown at:
in block <unit> at <tmp> line 1
timotimo why is camelia allowing you to try to open stuff at all? 22:51
it should be prevented by the restricted setting
m: open("hello")
camelia open is disallowed in restricted setting
in sub restricted at src/RESTRICTED.setting line 1
in sub open at src/RESTRICTED.setting line 9
in block <unit> at <tmp> line 1
gfldex m: my @fs = $*HOME «~« </.gitconfig /.config/git/config>; my $s = slurp([//] try (@fs».IO».open));
camelia Cannot resolve caller slurp(IO::Handle); none of these signatures match:
(IO::ArgFiles:D $io = { ... }, :$bin, :$enc = "utf8", |c is raw)
(Cool:D $path, :$bin = Bool::False, :$enc = "utf8", |c is raw)
in block <unit> at <tmp> line 1
timotimo m: slurp("hello")
camelia Failed to open file /home/camelia/hello: no such file or directory
in block <unit> at <tmp> line 1

Actually thrown at:
in block <unit> at <tmp> line 1
timotimo yeah, no, it should definitely restrict that 22:52
huh, slurp was never in there in the first place? is this fine?
TreyHarris m: say glob('*') 22:53
camelia 5===SORRY!5=== Error while compiling <tmp>
Undeclared routine:
glob used at line 1
timotimo it's "dir" 22:54
TreyHarris say dir
m: say dir
camelia (".cpanm".IO ".local".IO ".npm".IO ".perl6".IO ".perlbrew".IO ".rcc".IO ".ssh".IO "Perlito".IO "evalbot".IO "log".IO "nqp-js".IO "p1".IO "p2".IO "perl5".IO "std".IO ".bash_history".IO ".bashrc".IO "mbox".IO ".lesshst".IO "evalbot.log".IO ".cpan".IO "dale…
TreyHarris m: slurp(".cpanm") 22:55
camelia '.cpanm' is a directory, cannot do '.open' on a directory
in block <unit> at <tmp> line 1

Actually thrown at:
in block <unit> at <tmp> line 1
22:55 cpage_ left
TreyHarris m: slurp(".bashrc") and say "yes" 22:56
camelia yes
TreyHarris m: slurp("nosuch") and say "yes"
camelia Failed to open file /home/camelia/nosuch: no such file or directory
in block <unit> at <tmp> line 1

Actually thrown at:
in block <unit> at <tmp> line 1
TreyHarris yeah, it's broken, Jim. try reversing the polarity.
geekosaur a tually I answered it in a different context 22:57
[21 20:40:00] <geekosaur> m: my $y = '5'; $y .= Int; dd $y an answer to [21 20:39:33] <iyra> if $year is a string, is it valid to make $year into an Int (assuming it works) by doing $year = $year.Int? or will $year still by of type Str? 22:58
TreyHarris geekosaur: answered what? why .= not is so much slower than "$obj.truth = not $obj.truth"? if so I don't see the through-line connecting the two 23:00
geekosaur oh. I didn't even touch that one, so I have no idea what lucs meant 23:01
just got back from the store and noticed the highlight
and that's the only thong I said that was in anyway relevant
*thing
gfldex there seams to be no candidate for &slurp(IO::Handle:D)
timotimo wonder if there should be; how are the other subs? do we have &lines(IO::Handle:D) for example? 23:02
23:03 ChoHag joined
gfldex lines defaults to $*ARGFILES 23:03
timotimo oh, does lines take any arguments ever? 23:04
23:04 Cabanossi left
gfldex and forwards to *.lines, so lines(IO::Handle:D) should work 23:04
23:04 BenGoldberg joined
timotimo argfiles is going to get a little overhaul anyway. do we have a slurp for argfiles? 23:05
gfldex in fact &slurp seams to be the only IO sub that is so strict
anyway, good * 23:06
23:06 Cabanossi joined
gfldex self.sleep() 23:06
23:07 dct left
timotimo gnite gfldex! 23:08
Geth doc: bb972365ea | (Zoffix Znet)++ | 15 files
Fix spelling xtest

Used `sort` on words files, so the diff shows some of the results of sorting along with new words added.		 23:11
Garland I've been getting an internal error that requests a report to rakudobug, but I can't make it happen consistently. I tried re-running it with -ll-exception like it said to, but I don't get an error at all. 23:20
<Internal error while creating backtrace: Cannot set attribute values when cloning a type object in method throw at SETTING::src/core/Exception.pm line 63 23:21
timotimo huh, interesting 23:22
Garland If I change this line at all, the error also disappears: 23:23
for @!subscribers.race( batch => 1 ) -> $sub {
AlexDaniel ah-ha, .race! 23:24
timotimo oh
yeah, race is kinda broken at the moment
as is .hyper
at least i think so?
Garland hyper doesn't have the issue
AlexDaniel I think jnthn was working on it
timotimo last time i tried (months ago) it tended to just give empty lists for no reason
Garland or at least, it didn't set off the long internal error 23:25
TreyHarris Anyone know of a good test suite to look at that mocks native calls?
AlexDaniel (e.g. 6guts.wordpress.com/2017/03/16/con...semantics/ )
Garland Even removing a space makes the error disappear eg: for @!subscribers.race( batch => 1) -> $sub {
where the space after the 1 is removed. 23:26
AlexDaniel .tell jnthn fwiw irclog.perlgeek.de/perl6/2017-04-21#i_14466196 23:27
yoleaux AlexDaniel: I'll pass your message to jnthn.
nadim timotimo: www.youtube.com/watch?v=zcSFIUiMgA...ature=plcp 23:29
timotimo: and I had a few ideas about how to make it slightly better rt.cpan.org/Public/Dist/Display.ht...p-Debugger 23:31
Garland Here's a link to the full error chain as well: www.dropbox.com/s/fexjtd1lswfu4y8/...perl6?dl=0 23:32
For whoever needs it
23:33 Garland left 23:40 skids joined
timotimo i think i'm going to bed real soon now 23:41
have a good one!
23:45 TimToady joined, wamba left 23:53 nadim left 23:55 bjz left
iyra what's the usual way to check if a hash already exists in a list of hashes? 23:59

Powered by Cro and the Raku® Programming Language.

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

Thank you!