»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.
labster and another great job, lizmat++ 00:04
00:06 mcmillhj joined 00:07 cpage__ joined 00:10 mcmillhj left 00:19 cyphase left
BenGoldberg m: my method is-even (Int:D: --> Bool:D) { self %% 2 }; 42.^add_method("is-even", &is-even); say .is-even for 3, 12; 00:20
camelia False
True
BenGoldberg finds it funny that no .compose is needed for this.
00:24 sjn_ joined 00:25 greenspaceman joined
greenspaceman Hi! 00:25
skids o/
greenspaceman I'm trying to understand the behaviour of the .clone method, and it's been elluding me. Here's a minimal example of what looks like weird behaviour to me: pastebin.com/8enBKtdX 00:26
timotimo clone in perl6 is shallow
Xliff Um.
Here's an odd thing.
greenspaceman That explains a bit. But why the non-change in the first one? 00:27
Xliff m: my $idx = 0; say $idx"; $idx++
camelia 5===SORRY!5=== Error while compiling <tmp>
Two terms in a row
at <tmp>:1
------> 3my $idx = 0; say $idx7⏏5"; $idx++
expecting any of:
infix
infix stopper
postfix
statement end
statem…
Xliff m: my $idx = 0; say $idx; $idx++
camelia 0
00:27 jast left
Xliff Why does my code say "Use of Nil in numeric context" when I do that in a loop. 00:28
timotimo what's your loop look like, exactly?
Xliff unless %market{$k}<sell>[$idx]<vol_remain> > $minVol {
say "I: $idx";
$idx++;
next;
}
00:28 sjn_ is now known as sjn
timotimo hm, you gave $idx a value before that loop, yeah? 00:28
skids greenspaceman: The uninitialized attribute gets cloned unautovivified I think. 00:29
Xliff my ($minVol, $idx) = ($count / 10, 0);
So yes. That's done before the loop.
skids So you clone an autovivifier (a WHENCE)
00:29 ChoHag left
timotimo Xliff: too tired to think, sorry. will now fall into bed 00:29
Xliff LOL 00:30
MasterDuke_ Xliff: where's the loop?
Xliff The only other thing that might be odd is:
my $count := $i[1];
MasterDuke_: Scroll up
greenspaceman Interesting. So, if I want deep-cloning I should implement it as a method for my classes, right?
00:30 mcmillhj joined
timotimo yeah, that's probably necessary 00:30
Xliff But I do work on $count.
And I like self-documenting code. 00:31
skids tries to rmember if there is a deep clonish thing allready.
MasterDuke_ Xliff: but unless isn't a loop...? 00:33
00:34 Cabanossi left 00:35 mcmillhj left
Xliff Oh. Sorry. That's more code than I want to flood so lemme fpaste. 00:35
skids greenspaceman: yeah I'm not seeing anything built in. Well, deepmap for lists, but no attribute traverser.
Xliff And changing the "$idx++" to "$idx = $idx + 1" doesn't fix the issue.
00:36 Cabanossi joined
timotimo might be able to >>.clone or something 00:36
since hyperops try to preserve structure
greenspaceman skids: Thanks! I'll implement it myself, then. I don't really need a full deep-clone so I should have done it anyway. 00:38
Xliff gist.github.com/Xliff/4bdd5ba85334...0802640af8
00:39 jast joined
MasterDuke_ if you say $idx right before the increment? 00:40
Xliff That is the original version, but yes, if I "say $idx" before the increment, I get a value. 00:41
So.... WTF
timotimo maybe the error's line number is wrong
Xliff Probably so.
00:45 sufrostico left 00:46 khw left, mcmillhj joined 00:50 mcmillhj left
Xliff Problem is... why am I getting the wrong line number? 00:51
MasterDuke_ does --ll-exception say anything useful? 00:52
Xliff I'll check.
After this I will probably rakudobrew to see if it's just a commit oddness.
00:54 greenspaceman left
Xliff Renamed the variable and still getting same behavior. 00:56
This makes no sense.
01:00 Actualeyes joined
Xliff It was the line above. 01:03
Well that was LTA
01:05 Cabanossi left 01:06 Cabanossi joined
Xliff *sigh* -- also data issues. Fixed now. Thanks. 01:14
01:17 Dunearhp__ joined, Dunearhp left 01:27 Dunearhp_ joined 01:29 Dunearhp__ left 01:31 Dunearhp___ joined 01:32 Dunearhp_ left
Xliff If you sort something and assign it to a variable does that give you an array or a sequence? 01:32
I am assuming a sequence.
That has a hidden gotcha.
Because this line: %market{$k}<sell> = %m<quicklook><sell_orders><order>.sort: $sorter
I am assuming will return a sorted list, and it looks like that is not the case. 01:33
m: my %a; %a{list} = (1 2 3 4); %a{sorted_list} = %a.sort; dd %a
camelia 5===SORRY!5=== Error while compiling <tmp>
Two terms in a row
at <tmp>:1
------> 3my %a; %a{list} = (17⏏5 2 3 4); %a{sorted_list} = %a.sort; dd %
expecting any of:
infix
infix stopper
statement end
Xliff m: my %a; %a{list} = (1 2 3 4); %a{sorted_list} = %a.sort; dd %a; 01:34
camelia 5===SORRY!5=== Error while compiling <tmp>
Two terms in a row
at <tmp>:1
------> 3my %a; %a{list} = (17⏏5 2 3 4); %a{sorted_list} = %a.sort; dd %
expecting any of:
infix
infix stopper
statement end
Xliff m: my %a; %a{list} = q(1 2 3 4); %a{sorted_list} = %a.sort; dd %a;
camelia 5===SORRY!5=== Error while compiling <tmp>
Unable to parse expression in argument list; couldn't find final ')'
at <tmp>:1
------> 3my %a; %a{list} = q(17⏏5 2 3 4); %a{sorted_list} = %a.sort; dd %
expecting any of:
infix…
Xliff m: my %a; %a{list} = (1, 2, 3, 4); %a{sorted_list} = %a.sort; dd %a;
camelia 5===SORRY!5=== Error while compiling <tmp>
Undeclared routine:
sorted_list used at line 1
Xliff m: my %a; %a<list> = (1, 2, 3, 4); %a<sorted_list> = %a.sort; dd %a;
camelia Hash %a = {:list($(1, 2, 3, 4)), :sorted_list((:list($(1, 2, 3, 4)),).Seq)}
Xliff m: my %a; %a<list> = (1, 2, 3, 4); %a<sorted_list> = %a.sort.list; dd %a;
camelia Hash %a = {:list($(1, 2, 3, 4)), :sorted_list($(:list($(1, 2, 3, 4)),))}
Xliff YEP. 01:35
So to extend that....
m: my %a; %a<list> = (1, 2, 3, 4); %a<sorted_list> = %a.sort.list; dd %a; my @a = %a<sorted_list; @a.elems.say
camelia 5===SORRY!5=== Error while compiling <tmp>
Unable to parse quote-words subscript; couldn't find right angle quote
at <tmp>:1
------> 3%a; my @a = %a<sorted_list; @a.elems.say7⏏5<EOL>
expecting any of:
postfix
Xliff m: my %a; %a<list> = (1, 2, 3, 4); %a<sorted_list> = %a.sort.list; dd %a; my @a = %a<sorted_list>; @a.elems.say
camelia Hash %a = {:list($(1, 2, 3, 4)), :sorted_list($(:list($(1, 2, 3, 4)),))}
1
Xliff Is this the intended behavior? 01:36
TimToady yes, values in arrays and lists are itemized, so sometimes you have to .list them when you take them out 01:37
s/lists/hashes/
01:37 mr-foobar joined
TimToady m: my %a; %a<list> = (1, 2, 3, 4); %a<sorted_list> = %a.sort.list; dd %a; my @a = %a<sorted_list>[]; @a.elems.say 01:38
camelia Hash %a = {:list($(1, 2, 3, 4)), :sorted_list($(:list($(1, 2, 3, 4)),))}
1
TimToady hmm, looks like an extra layer of $ there though
01:40 cyphase joined
TimToady m: my %a; %a<list> = (1, 2, 3, 4); %a<sorted_list> = %a<list>.sort.cache; dd %a; my @a = %a<sorted_list>[]; @a.elems.say 01:42
camelia Hash %a = {:list($(1, 2, 3, 4)), :sorted_list($(1, 2, 3, 4))}
4
01:42 gdonald left, gdonald joined
TimToady m: my %a; %a<list> = (1, 2, 3, 4); %a<sorted_list> = %a<list>.sort.list; dd %a; my @a = %a<sorted_list>[]; @a.elems.say 01:42
camelia Hash %a = {:list($(1, 2, 3, 4)), :sorted_list($(1, 2, 3, 4))}
4
TimToady m: my %a; %a<list> = (1, 2, 3, 4); %a<sorted_list> = %a<list>.sort.list; dd %a; my @a = %a<sorted_list>; @a.elems.say
camelia Hash %a = {:list($(1, 2, 3, 4)), :sorted_list($(1, 2, 3, 4))}
1
TimToady yes, that all seems about right 01:43
u-ou does .sort on a scalarised list return another scalarised list? 01:44
01:45 wamba joined 01:46 ilbot3 left 01:47 ilbot3 joined, ChanServ sets mode: +v ilbot3
Xliff TimToady: But shouldn't assigning a sequence to an array give you a list off sorted values rather than a single valued list (where that value is the sequence). 01:50
m: my %a; %a<list> = (1, 2, 3, 4); %a<sorted_list> = %a.sort.list; dd %a; my @a = %a<sorted_list>; dd @a; 01:51
camelia Hash %a = {:list($(1, 2, 3, 4)), :sorted_list($(:list($(1, 2, 3, 4)),))}
Array @a = [(:list($(1, 2, 3, 4)),),]
Xliff m: my %a; %a<list> = (1, 2, 3, 4); %a<sorted_list> = %a.sort; dd %a; my @a = %a<sorted_list>; dd @a;
camelia Hash %a = {:list($(1, 2, 3, 4)), :sorted_list((:list($(1, 2, 3, 4)),).Seq)}
Array @a = [Seq.new-consumed(),]
Xliff Seq.new-consumed() ?!?
geekosaur because the dd %a ate it 01:52
u-ou I think assigning a sequence to a @ variable evaluates the whole sequence
Xliff m: my %a; %a<list> = (1, 2, 3, 4); %a<sorted_list> = %a.sort; my @a = %a<sorted_list>; dd @a;
camelia Array @a = [(:list($(1, 2, 3, 4)),).Seq,]
Xliff ^^ That's array containing a single Seq, not (1, 2, 3, 4).
m: my %a; %a<list> = (1, 2, 3, 4); %a<sorted_list> = %a<list>.sort; my @a = %a<sorted_list>; dd @a; 01:53
camelia Array @a = [(1, 2, 3, 4).Seq,]
Xliff Minor error
Still a seq.
m: my %a; %a<list> = (1, 2, 3, 4); %a<sorted_list> = %a<list>.sort; my @a = %a<sorted_list>; dd @a; @a.elems.say
camelia Array @a = [(1, 2, 3, 4).Seq,]
1
MasterDuke_ m: my %a; %a<list> = (1, 2, 3, 4); %a<sorted_list> = %a<list>.list.sort; my @a = %a<sorted_list>; dd @a; 01:54
camelia Array @a = [(1, 2, 3, 4).Seq,]
MasterDuke_ m: my %a; %a<list> = (1, 2, 3, 4); %a<sorted_list> = %a<list>.sort; my @a = %a<sorted_list>.list; dd @a;
camelia Array @a = [1, 2, 3, 4]
MasterDuke_ m: my %a; %a<list> = (1, 2, 3, 4); %a<sorted_list> = %a<list>.sort; my @a = %a<sorted_list>[]; dd @a;
camelia Array @a = [1, 2, 3, 4]
Xliff "This Seq has already been iterated, and its values consumed 01:55
(you might solve this by adding .cache on usages of the Seq, or
by assigning the Seq into an array)"
^^ So am I misunderstanding this message?
MasterDuke_ m: my %a; %a<list> = (1, 2, 3, 4); %a<sorted_list> := %a<list>.sort; my @a = %a<sorted_list>; dd @a;
camelia Array @a = [1, 2, 3, 4]
Xliff Ahhh...
MasterDuke_ the hash value is containerized 01:56
Xliff Still.... weird.
OK. That will probably solve it.
MasterDuke_ m: my %a; %a<list> = (5, 1, 2, 3, 4); %a<sorted_list> = %a<list>.sort; my @a = %a<sorted_list>[]; dd @a;
camelia Array @a = [1, 2, 3, 4, 5]
MasterDuke_ m: my %a; %a<list> = (5, 1, 2, 3, 4); %a<sorted_list> = %a<list>.sort; my @a = %a<sorted_list>; dd @a;
camelia Array @a = [(1, 2, 3, 4, 5).Seq,]
Xliff Using a bind in my code breaks it. 01:58
MasterDuke_ what about `%a<sorted_list>[]`? 01:59
Xliff m: my %a; %a<list> = (5, 1, 2, 3, 4); %a<sorted_list> = (%a<list>.sort)[]; dd %a
camelia Hash %a = {:list($(5, 1, 2, 3, 4)), :sorted_list((1, 2, 3, 4, 5).Seq)}
Xliff m: my %a; %a<list> = (5, 1, 2, 3, 4); %a<sorted_list> = %a<list>.sort.list; dd %a 02:00
camelia Hash %a = {:list($(5, 1, 2, 3, 4)), :sorted_list($(1, 2, 3, 4, 5))}
Xliff OK. Now here's the problem with .list.
m: my %a; %a<list> = ({ a=> 5}, { a=> 1}, { a=> 2 }); %a<sorted_list> = %a<list>.sort({ $^a<a> <=> $^b<a>}).list; dd %a 02:02
camelia Hash %a = {:list($({:a(5)}, {:a(1)}, {:a(2)})), :sorted_list($({:a(1)}, {:a(2)}, {:a(5)}))}
Xliff m: my %a; %a<list> = ({ a=> 5, b => 'a'}, { a=> 1, b => 'b' }, { a=> 2, b => 'c' }); %a<sorted_list> = %a<list>.sort({ $^a<a> <=> $^b<a>}).list; dd %a
camelia Hash %a = {:list($({:a(5), :b("a")}, {:a(1), :b("b")}, {:a(2), :b("c")})), :sorted_list($({:a(1), :b("b")}, {:a(2), :b("c")}, {:a(5), :b("a")}))}
Xliff Hmm... 02:03
camelia isn't doing it... dammit
02:03 kurahaupo left
Xliff m: my %a; %a<list> = ({ a=> 5, b => 'a'}, { a=> 1, d => 'b' }, { a=> 2, e => 'c' }); %a<sorted_list> = %a<list>.sort({ $^a<a> <=> $^b<a>}).list; dd %a 02:03
camelia Hash %a = {:list($({:a(5), :b("a")}, {:a(1), :d("b")}, {:a(2), :e("c")})), :sorted_list($({:a(1), :d("b")}, {:a(2), :e("c")}, {:a(5), :b("a")}))}
u-ou is it bad practice to call a Role's .new method, you're supposed to make a class that "does" that role first? 02:08
samcv heh nice. my rakudo-appimage gh-pages branch is now 465MB so far
wonder how big a repo has to get for github to get mad at you?
if you only keep like 50MB of files at a time in each revision 02:09
geekosaur u-ou, you can do it and an empoty class will be composed with it first
u-ou automagically like?
geekosaur as if you'd said (iirc): (class :: does TheRole {}).new 02:10
samcv rakudo is 41031KB, nqp is 110874KB
u-ou I see
samcv stats you can get in json here api.github.com/repos/perl6/nqp
geekosaur and yes, automagically
roles can't be instantiated, so it has to compose a class with it
u-ou cool
02:11 pytuger left
MasterDuke_ it's called punning 02:12
02:12 shady0wl joined, pytuger joined, pytuger left, pytuger joined 02:16 noganex joined
Xliff I am getting this emitted every time I run perl6, now. 02:18
HERE:
- sym: :
- O:
02:19 noganex_ left
Xliff m: my %a; %a<list> = ({ a=> 5, b => 'a'}, { a=> 1, d => 'b' }, { a=> 2, e => 'c' }); %a<list>.sort; dd %a; 02:19
camelia Hash %a = {:list($({:a(5), :b("a")}, {:a(1), :d("b")}, {:a(2), :e("c")}))}
Xliff m: my %a; %a<list> = ({ a=> 5, b => 'a'}, { a=> 1, d => 'b' }, { a=> 2, e => 'c' }); %a<list>.sort: { $^a<a> <=> $^b<a>}; dd %a; 02:22
camelia Hash %a = {:list($({:a(5), :b("a")}, {:a(1), :d("b")}, {:a(2), :e("c")}))}
Xliff m: my %a; %a<list> = ({ a=> 5, b => 'a'}, { a=> 1, d => 'b' }, { a=> 2, e => 'c' }); sort { $^a<a> <=> $^b<a>} %a<list>; dd %a; 02:23
camelia 5===SORRY!5=== Error while compiling <tmp>
Missing comma after block argument to sort
at <tmp>:1
------> 3' }); sort { $^a<a> <=> $^b<a>} %a<list>7⏏5; dd %a;
Xliff m: my %a; %a<list> = ({ a=> 5, b => 'a'}, { a=> 1, d => 'b' }, { a=> 2, e => 'c' }); sort { $^a<a> <=> $^b<a>}, %a<list>; dd %a;
camelia Hash %a = {:list($({:a(5), :b("a")}, {:a(1), :d("b")}, {:a(2), :e("c")}))}
02:26 dugword joined, dugword left, dugword joined 02:30 ilmari left 02:31 ilmari joined
MasterDuke_ that's from a TimToady commit on the 10th 02:33
m: say "hi": 02:34
camelia HERE:
- sym: :
- O:

hi
02:35 cpage__ left
Xliff m: say 1 02:36
camelia 1
Xliff m: say 1: 02:37
camelia HERE:
- sym: :
- O:

1
Xliff What triggers it?
Or is it always there and camelia is suppressing it without the trailing ':'
AlexDaniel e: say 1 02:38
evalable6 1
AlexDaniel e: say 1:
evalable6 HERE:
- sym: :
- O:

1
02:38 cpage_ joined
MasterDuke_ github.com/rakudo/rakudo/blob/nom/....nqp#L4504 02:39
github.com/rakudo/rakudo/commit/cd...a1c4018b7b 02:40
Xliff So it's in my code somewhere?
I am having a hard time reading that.
MasterDuke_ i think it's a debugging thing that got accidentally left in
Xliff OK
MasterDuke_ .tell TimToady if you backlog here irclog.perlgeek.de/perl6/2017-04-18#i_14443061, was this intended? 02:42
yoleaux MasterDuke_: I'll pass your message to TimToady.
02:46 mcmillhj joined 02:50 wamba left 02:51 mcmillhj left 02:55 Sgeo_ joined 02:56 Sgeo__ joined 02:59 Sgeo left 03:00 Sgeo_ left
AlexDaniel MasterDuke_: definitely not :D 03:01
03:02 mcmillhj joined 03:05 Cabanossi left 03:06 Cabanossi joined 03:08 mcmillhj left
MasterDuke_ seems unlikely for sure 03:14
AlexDaniel would be nice if somebody fixed it right away 03:19
so that we don't wake up with a hundred of complaints :)
03:24 dugword left 03:47 Cabanossi left 03:50 Cabanossi joined 03:57 zapwai left 04:09 xtreak joined 04:14 xtreak left 04:21 mcmillhj joined 04:24 skids left 04:26 mcmillhj left
Xliff *sigh* 04:33
If I have sub MAIN($a, $b, *%c), will the option --d be put in %c as %c<d> ?
Or will the main choke on "--d" 04:34
I have a script that will depend on a WHOLE passel of optional arguments and I don't want to clutter the signature.
Assuming I have that option.
Just tried it and MAIN throws back "--c=..." in USAGE. 04:35
And yes, it chokes on --d.
04:37 mcmillhj joined 04:42 mcmillhj left 04:47 CIAvash joined 04:53 mcmillhj joined 04:58 mcmillhj left 05:03 BenGoldberg left 05:09 mcmillhj joined 05:17 Cabanossi left, mcmillhj left 05:20 wamba joined, Cabanossi joined 05:22 incredible-- joined 05:25 SwoleGymBro- left, lowbro joined, lowbro left, lowbro joined, dugword joined 05:27 labster left 05:29 dugword left 05:37 AlexDaniel left, parv joined
Xliff m: sub a (Int $a Where (1..10)) { say $a }; a(5); 05:37
camelia 5===SORRY!5=== Error while compiling <tmp>
Malformed parameter
at <tmp>:1
------> 3sub a (Int $a7⏏5 Where (1..10)) { say $a }; a(5);
expecting any of:
constraint
Xliff m: sub a (Int $a Where 1..10) { say $a }; a(5);
camelia 5===SORRY!5=== Error while compiling <tmp>
Malformed parameter
at <tmp>:1
------> 3sub a (Int $a7⏏5 Where 1..10) { say $a }; a(5);
expecting any of:
constraint
Xliff m: sub a (Int $a Where * %% 2) { say $a }; a(5);
camelia 5===SORRY!5=== Error while compiling <tmp>
Malformed parameter
at <tmp>:1
------> 3sub a (Int $a7⏏5 Where * %% 2) { say $a }; a(5);
expecting any of:
constraint
Xliff m: sub a (Int $a Where * > 0) { say $a; }; a(5).say 05:38
camelia 5===SORRY!5=== Error while compiling <tmp>
Malformed parameter
at <tmp>:1
------> 3sub a (Int $a7⏏5 Where * > 0) { say $a; }; a(5).say
expecting any of:
constraint
Xliff m: subset Range of Int Where { .any(1..10) }; sub a (Range $a) { say $a; }; a(5).say 05:40
camelia 5===SORRY!5=== Error while compiling <tmp>
Two terms in a row
at <tmp>:1
------> 3subset Range of Int7⏏5 Where { .any(1..10) }; sub a (Range $a)
expecting any of:
infix
infix stopper
postfix
s…
Xliff m: subset Range of Int Where { .any(1..10) }; sub a (Range $a) { say $a; }; a(5).say;
camelia 5===SORRY!5=== Error while compiling <tmp>
Two terms in a row
at <tmp>:1
------> 3subset Range of Int7⏏5 Where { .any(1..10) }; sub a (Range $a)
expecting any of:
infix
infix stopper
postfix
s…
Xliff m: subset Range of Int Where { .any(1..10); }; sub a (Range $a) { say $a; }; a(5).say;
camelia 5===SORRY!5=== Error while compiling <tmp>
Two terms in a row
at <tmp>:1
------> 3subset Range of Int7⏏5 Where { .any(1..10); }; sub a (Range $a
expecting any of:
infix
infix stopper
postfix
s…
CIAvash Xliff: Where → where 05:43
Xliff *groan*
CIAvash: Can you use slurpy named args in MAIN?
Right now it doesn't seem to be working for me. 05:44
(Thanks for the help, btw!)
05:50 RabidGravy joined 05:51 geekosaur left 05:53 geekosaur joined 05:56 jraglin left
CIAvash Xliff: It seems *%foo works 05:57
Xliff Everytime I try that, USAGE says "--foo=..."
But I don't want "--foo", I want "--a_name_not_specified_in_the_signature" 05:58
And I will write custom USAGE
That's how slurpy's work in subs.
CIAvash It's only written in the usage, all options will be stored in %foo 05:59
Xliff When I try that, the USAGE gets called.
06:00 xtreak joined
Xliff Oh SNAP! 06:00
I was doing --noname blah, not --noname=blah
CIAvash :)
Xliff I wish there was a way of doing the former in P6 without too much work, buut.
'kay well 06:01
I will save it for tomorrow.
06:17 gdonald left, gdonald joined 06:23 cpage_ left
RabidGravy boom! 06:26
samcv hi RabidGravy
i'm gonna have to open a lot of github tickets for these failing modules. having travis try installing every single eco module 06:27
and i think it will time out before installing all of them. but. i could be wrong. but it seems likely
06:30 darutoko joined 06:31 imcsk8 left
RabidGravy I'm surprised it doesn't time out after about twenty to be honest 06:32
06:33 bioexpress joined 06:37 pytuger left, shady0wl left 06:38 shady0wl joined, pytuger joined, pytuger left, pytuger joined 06:39 shady0wl left 06:41 ChoHag joined
bioexpress Hello, I have the module "MyModule": use v6; unit class MyModule; use MyModule::Kurses; "MyModule::Kurses" looks like this: use v6; unit module MyModule::Kurses; use LibraryCheck; if library-exists('ncurses', v5) {} When I try to install "MyModule" I get this error: ===> Install [FAIL] for MyModule: ===SORRY!=== 06:45
Cannot invoke this object (REPR: Null; VMNull) What could be the reason this fail? I get the same error when I use NCruses (instead of MyModule::Kurses) 06:46
RabidGravy dunno, do you have a reasonably recent rakudo? 06:52
bioexpress rakudo star 2017.01 06:54
RabidGravy Hmm that should be fine I guess, can you gist your actual code? 06:55
06:56 incredible joined
TimToady how are you installing? I'm currently having a bit of trouble with zef... 06:57
bioexpress Can you tell me what "gist the code" means?
geekosaur post it at gist.github.com
which is essentially temporary git repos
for multi-file pastes
bioexpress How could I do that?
zef --force install /home/mm/lib/PERL6/MyModule-p6 06:58
Should I create a git repo?
TimToady my zef doesn't seem to want to talk to github at the moment 06:59
RabidGravy :(
bioexpress, just go to gist.github.com/ and paste your code 07:00
then tell us the resulting URI
bioexpress ok 07:01
RabidGravy TimToady, well 'perl6 -MHTTP::UserAgent -e 'say HTTP::UserAgent.new.get("github.com/")'' seems okay from here so it may be local :) 07:03
07:03 Cabanossi left 07:04 cpage_ joined 07:05 Cabanossi joined
TimToady heh, I get: Could not find HTTP::UserAgent at line 1 07:05
07:06 perlawhirl__ left
gfldex TimToady: you can use zef to install that :-> 07:06
07:06 perlawhirl left
bioexpress gist.github.com/anonymous/bfb46bc3...01b233bc29 07:06
07:07 EAAQC48 joined, NAAEU0P joined, perlawhirl joined, perlawhirl_ joined 07:13 z-b is now known as zb
RabidGravy bioexpress, I can confirm what you are seeing 07:13
bioexpress Thx, I'm not alone. 07:14
TimToady reinstalling zef seems to have fixed my download problem; I probably nuked something it thought it could rely on at some point 07:15
07:17 perlawhirl_ left, perlawhirl left 07:19 zakharyas joined 07:20 perlawhirl joined
RabidGravy bioexpress, it looks like something funny with library-exists() being used where it is, I'm not quite sure why though 07:22
bioexpress Is this a bug? 07:27
07:34 robertle left 07:35 committable6 left, committable6 joined, ChanServ sets mode: +v committable6 07:37 wamba left, committable6 left 07:38 benchable6 left, bisectable6 left, evalable6 left, unicodable6 left, statisfiable6 left, bloatable6 left 07:40 xtreak left 07:41 xtreak joined, labster joined, evalable6 joined, committable6 joined, bloatable6 joined, unicodable6 joined, ChanServ sets mode: +v evalable6, ChanServ sets mode: +v committable6, ChanServ sets mode: +v bloatable6, ChanServ sets mode: +v unicodable6, benchable6 joined, ChanServ sets mode: +v benchable6, bisectable6 joined, ChanServ sets mode: +v bisectable6, statisfiable6 joined, ChanServ sets mode: +v statisfiable6 07:45 xtreak left 07:46 committable6 left, AlexDaniel joined 07:48 Cabanossi left, xtreak joined 07:50 ChoHag left, Cabanossi joined, committable6 joined, ChanServ sets mode: +v committable6 07:54 domidumont joined 07:55 xtreak left 07:56 zoosha left, domidumont left, xtreak joined, domidumont joined 07:57 yqt joined 08:00 jonas1 joined 08:01 bioexpress left
RabidGravy bioexpress, it's probably a bug, but I'm not sure who's bug it is, inasmuch as the library-exists definitely works in all other circumstances and if you use like that in a module that is directly used it is fine 08:01
08:02 abraxxa joined
RabidGravy I think it's because it is being called while the MyModule is being compiled 08:03
08:04 mcmillhj joined 08:08 mcmillhj left 08:10 mcmillhj joined 08:15 AlexDaniel left, mcmillhj left 08:26 mcmillhj joined 08:29 e1 joined 08:30 mcmillhj left
RabidGravy bioexpress, right, you can work around this by adding "no precompilation;" at the top of the MyModule::Kurses - it's not ideal but should get you moving along 08:32
08:35 cyphase left 08:38 bioexpress joined
bioexpress RabidGravy, thanks for the help! 08:39
lizmat clickbaits p6weekly.wordpress.com/2017/04/17/...-the-road/ 08:40
08:40 lizmat left 08:41 cyphase joined, mcmillhj joined 08:42 bioexpress left 08:45 xtreak left
RabidGravy I suspect that this is an alternative manifestation of rt.perl.org/Ticket/Display.html?id=127034 08:45
08:46 mcmillhj left 08:47 bjz joined 08:48 Cabanossi left 08:50 Cabanossi joined 08:55 robertle joined 09:04 xtreak joined 09:06 abraxxa left 09:07 abraxxa joined 09:09 xtreak left 09:13 xtreak joined 09:16 xtreak left 09:23 lowbro left 09:24 lowbro joined, xtreak joined, lowbro left, lowbro joined 09:30 xtreak left 09:31 gdonald left 09:32 gdonald joined
eater is the definition of Build.pm documentated somewhere? 09:32
09:33 Cabanossi left 09:35 Cabanossi joined
samcv build.pm? 09:37
there's no build.pm in rakudo or nqp repos
oh is that in panda or zef maybe? 09:39
eater samcv: some zef / panda packages have them ye 09:40
samcv which? 09:42
Panda::Builder?
i'm getting 'Build' cannot inherit from 'Panda::Builder' because it is unknown. personally
09:43 labster left
eater samcv: f.x. Noiseline uses it 09:44
github.com/search?utf8=%E2%9C%93&a...;type=Code
samcv can seorch this way github.com/search?utf8=%E2%9C%93&a...;type=Code 09:48
09:48 domidumont left
samcv not all perl 6 files have use v6 09:48
also we have github.com/search?utf8=%E2%9C%93&a...;type=Code
09:49 incredible left
samcv or all of them: github.com/search?utf8=%E2%9C%93&a...;type=Code 09:49
eater, so the exact eror you are getting is? is it the same as mine? 09:52
eater no, I just was looking if there was actually a spec/rfc
samcv oh. so you don't get any error
ok
well. you can install stuff without zef i mean. i think. or maybe it's special 09:53
eater samcv: this is for: github.com/ugexe/zef/issues/152#is...-294594127
samcv idk perl6 -Ilib bin/zef install .
i guess it uses itself to install itself
which is weird but cool i guess
eater samcv: there needs to be a /global/ way to define build parameters 09:54
and currently there isn't
samcv ah
it's called CFLAGS
:P
eater fite me
samcv ok a non ENV variable way to set CFLAGS? 09:55
eater yea
samcv also i very much think is within the scope of zef
it builds packages
like saying it's out of scope for make
unless it's using some other package to build?
eater so a binary package manager can inject those build parameters
dalek b-rakudo: e98193a | (Zoffix Znet)++ | download-page.php:
Fix sorting of point releases on rakudo.org/downloads/

PHP's glob() gives them to us AFTER non-point releases, so when a point release is the latest release, we end up with non-point release in the "latest" box.
Fix by shoving them into a separate array first and prepend to file list when we get the first non-point release (or we're done looking through files).
09:55 xtreak joined
samcv LibraryMake right? 09:55
09:55 bjz left
eater that's currently one that's being weird 09:56
but I have the odd feeling the project is dead
09:57 bjz joined
samcv wait but zef has no dependencies 09:58
hm
eater samcv: what do u mean? 09:59
09:59 incredible_ joined
samcv ok what is pulling in that package? 10:00
the noiseline does i guess. and then that calls library make
eater yea
samcv well i'd make a PR
10:01 xtreak left, gregf_ joined
eater samcv: to where? LibraryMake? 10:01
samcv github.com/retupmoca/P6-LibraryMak...e.pm6#L129 # %vars<CCFLAGS> = %*ENV<CCFLAGS> // $*VM.config<cflags>;
yes
10:02 xtreak joined, Cabanossi left
samcv and if no response i mean can always fork it. multiple releases by the same name can exist. just make sure you put an :auth part in the module 10:02
eater cool
samcv iand get that used by the other one
but yeah
that's what i'd do
eater will look into it tonight
samcv thank you :)
eater after that I'll still look into actually defining Build.pm tho 10:03
because this is weird
El_Che hello's 10:05
10:05 Cabanossi joined
samcv now has TEN travis builds running travis-ci.org/samcv/rakudo-appimage 10:07
trying to install all of the modules in the ecosystem
hopefully this is doable cause i only get 50 mins per build
so sorted them by name, then split them into 10 pieces
doesn't take dependencies into account so there's gonna be some overlap probably. but. simplest way to do it 10:08
10:13 e1 left 10:18 parv left 10:20 bolangi left 10:26 TEttinger left 10:32 wamba joined 10:33 bolangi joined
Geth doc: 9b184cb458 | (Zoffix Znet)++ (committed using GitHub Web editor) | doc/Language/glossary.pod6
Toss empty `handles` entry from glossary
10:45
10:48 Cabanossi left 10:50 Cabanossi joined 10:51 wamba left 11:16 AlexDaniel joined 11:20 shlomif joined, rindolf left 11:21 movl left, Zoffix joined
Zoffix El_Che: you may wanna hold on with your packages until 2017.04.2 is out 11:21
(should be within next 10 hours) 11:22
11:24 xtreak left 11:26 Xliff_ joined 11:27 xtreak joined 11:29 Xliff left 11:30 pytuger left 11:36 xtreak left 11:42 pytuger joined 11:48 Cabanossi left 11:49 Stamm joined 11:50 Cabanossi joined, Khisanth left 12:04 Khisanth joined 12:19 shlomif2 joined 12:20 Zoffix left 12:23 shlomif left 12:35 AlexDaniel left 12:37 shlomif2 is now known as rindolf
El_Che tell 12:39
!tell Zoffix Thx, I'll do that. I was just starting to create the pkgs, but indeed beter to wait 12:40
nope
MasterDuke_ El_Che: .tell 12:42
El_Che .tell El_Che you suck
yoleaux El_Che: Talking to yourself is the first sign of madness.
El_Che :)
.tell Zoffix Thx, I'll do that. I was just starting to create the pkgs, but indeed beter to wait
yoleaux El_Che: I'll pass your message to Zoffix.
12:46 mcmillhj joined 12:52 movl joined 12:57 domidumont joined
Geth ecosystem: 74a4b9dd5f | (Carl Mäsak)++ (committed using GitHub Web editor) | META.list
Take tardis off the ecosystem list

Project being dormant/unmaintained for a long time now without achieving escape velocity.
12:59
13:02 Cabanossi left 13:05 Cabanossi joined 13:16 raschipi joined 13:18 sufrostico joined 13:28 ChoHag joined 13:29 jonas1 left 13:33 sammers left, kybr left 13:43 Stamm left
eater is panda btw abandoned? or does it just look like that? 13:44
13:45 tokomer left
moritz what's the difference? 13:45
eater my observation skills 13:47
13:48 Cabanossi left 13:50 Cabanossi joined
tadzik it's effectively abandoned 13:55
Geth Inline-Perl5: 81c8a1f4eb | (Stefan Seifert)++ | lib/Inline/Perl5.pm6
Fix accessing the helper shared lib when installed via Staging repo

Rakudo will now return Distribution::Resource objects from %?RESOURCES lookups. NativeCall knows how to deal with those, so we don't have to stringify manually anymore. This allows for deferred creation of the path and thus fixes issues when the path differs between compile time and run time (as it does when installed via Staging repo).
13:59
14:00 xtreak joined 14:01 lowbro left 14:05 sammers joined, wamba joined 14:09 xtreak left 14:12 xtreak joined, xtreak left 14:15 zoosha joined 14:22 azawawi joined 14:24 azawawi left 14:35 cdg joined
alphah Hi 14:37
m: my %h; %h<a> = { k1 => "v1", k2 => "v2" }; %h<b> = { k2 => "v3", k4 => "v4" }; %h<a> ,= %h<b>
camelia ( no output )
alphah m: my %h; %h<a> = { k1 => "v1", k2 => "v2" }; %h<b> = { k2 => "v3", k4 => "v4" }; %h<a> ,= %h<b>; say %h
camelia {a => (\List_75262736 = (List_75262736 {k2 => v3, k4 => v4})), b => {k2 => v3, k4 => v4}}
alphah why the output has \List_, and how can I make it show the key'values normally 14:38
moritz the \List_ stuff is because you have a circular reference in your data structure
and you can show it normally by deleting circular references first 14:39
alphah mortiz: can you point out where is that circular refernce? 14:40
moritz or not necessarily circular, but referencing the same hash twice
m: my %h; %h<a> = { k1 => "v1", k2 => "v2" }; %h<b> = { k2 = "v3", k4 => "v4" }; %h<a> ,= %h<b>; say %h<a> 14:42
camelia 5===SORRY!5=== Error while compiling <tmp>
Preceding context expects a term, but found infix = instead
at <tmp>:1
------> 3k1 => "v1", k2 => "v2" }; %h<b> = { k2 =7⏏5 "v3", k4 => "v4" }; %h<a> ,= %h<b>; say
moritz m: my %h; %h<a> = { k1 => "v1", k2 => "v2" }; %h<b> = { k2 => "v3", k4 => "v4" }; %h<a> ,= %h<b>; say %h<a>
camelia (\List_77711440 = (List_77711440 {k2 => v3, k4 => v4}))
moritz m: my %h; %h<a> = { k1 => "v1", k2 => "v2" }; %h<b> = { k2 => "v3", k4 => "v4" }; %h<a>.push(%h<b>.list); say %h<a> 14:43
camelia {k1 => v1, k2 => [v2 v3], k4 => v4}
moritz m: my %h; %h<a> = { k1 => "v1", k2 => "v2" }; %h<b> = { k2 => "v3", k4 => "v4" }; %h<a>.push(%h<b>.list); say %h
camelia {a => {k1 => v1, k2 => [v2 v3], k4 => v4}, b => {k2 => v3, k4 => v4}}
14:44 bjz left
raschipi m: my %h; %h<a> = { k1 => "v1", k2 => "v2" }; %h<b> = { k2 => "v3", k4 => "v4" }; %h<a> ,= %h<b>; put %h<a> 14:44
camelia MoarVM panic: Memory allocation failed; could not allocate 64 bytes
14:45 ifim joined
moritz m: my %a = a => 1; my %b = b => 2; %a ,= %b; say %a.perl 14:48
camelia {:a(1), :b(2)}
moritz m: my $a = {a => 1}; my $b = {b => 2}; $a ,= $b; say $a.perl
camelia (my \List_61638160 = $(List_61638160, ${:b(2)}))
moritz that's a simple way to reproduce 14:49
basically the ,= doesn't flatten the hash, but creates a two-element list, the first of which is itself
alphah hmm 14:50
makes sense now
14:50 ChoHag left
moritz more or less 14:51
I'm not sure it's desired behavior
m: my $a = {a => 1}; my $b = {b => 2}; $a = $a, $b; say $a.perl
camelia WARNINGS for <tmp>:
Useless use of $b in sink context (line 1)
${:a(1)}
moritz m: my $a = {a => 1}; my $b = {b => 2}; $a = ($a, $b); say $a.perl
camelia (my \List_73800528 = $(List_73800528, ${:b(2)}))
moritz but I guess it is; the evaluation order thingy is pretty well-tested 14:52
raschipi Couldn't .gist at least return something more descriptive there? 14:55
moritz easier said than done 14:56
(especially if you also want it to be fast)
14:56 kybr joined 15:03 Cabanossi left 15:05 Cabanossi joined 15:06 CIAvash left 15:12 perlpilot joined
timotimo alphah: can you tell us where you found the suggestion to use ,= to merge hashes? 15:18
15:20 incredible-- left
alphah timotimo: it was here on the channel, I'll see if I can find it in log 15:22
timotimo irclog.perlgeek.de/perl6/2017-04-12#i_14418748 15:26
it also shows as a way to merge hashes by searching google 15:27
Geth doc: 03e435eb5b | (Brad Gilbert)++ (committed using GitHub Web editor) | doc/Language/quoting.pod6
Add missing Q adverbs to table
15:28
perlpilot TimToady: If you're feeling generous, #perl wonders why only arrays and scalars interpolate in Perl 5 (and 4). There's speculation that it's because of the printf() formatters (and I guessed that also there was no "natural" way to interpolate keys and values that made sense) 15:31
15:33 Cabanossi left
perlpilot TimToady: er, to be slightly clearer ... why don't hashes interpolate in double quoted strings? 15:33
15:35 Cabanossi joined
Geth doc: ae4e48799c | (Zoffix Znet)++ (committed using GitHub Web editor) | doc/Language/unicode_texas.pod6
Add note that 「」 cannot be swapped to Q// in regexes
15:35
15:38 eroux joined 15:43 cdg left 15:44 Resol joined, cdg joined
TimToady perlpilot: mostly just because it interfered with printf, but also because the use case is weak, and we hadn't thought of requiring some kind of subscript as we did in p6 15:45
there's never only one reason for anything in Perl :) 15:46
timotimo alphah: can't find it easily, what were your keywords? 15:47
15:50 sufrostico left
TimToady timotimo: didn't you see the link? 15:51
oh, you were asking about the line after, nevermind 15:52
TimToady heads back to the coffeemaker
timotimo yup :) 15:53
alphah: there's a difference between what you linked to and what you tried
your hashes are items inside another hash, i expect 15:54
and putting two items together with , makes things not flatten
alphah searching in google shows this at first page stackoverflow.com/questions/350018/...es-in-perl and I guess %hash1 = (%hash1, %hash2) is same as ,=
timotimo m: my %h; %h<a> = { k1 => "v1", k2 => "v2" }; %h<b> = { k2 => "v3", k4 => "v4" }; my %result = %h<a>, %h<b>; say %result.perl
camelia {"k1\tv1\nk2\tv2" => ${:k2("v3"), :k4("v4")}}
timotimo haha
m: my %h; %h<a> = { k1 => "v1", k2 => "v2" }; %h<b> = { k2 => "v3", k4 => "v4" }; my %result = @(%h<a>), @(%h<b>); say %result.perl 15:55
camelia {"k1\tv1 k2\tv2" => (:k4("v4"), :k2("v3")).Seq}
timotimo m: my %h; %h<a> = { k1 => "v1", k2 => "v2" }; %h<b> = { k2 => "v3", k4 => "v4" }; say %h.perl; my %result = @(%h<a>), @(%h<b>); say %result.perl
camelia {:a(${:k1("v1"), :k2("v2")}), :b(${:k2("v3"), :k4("v4")})}
{"k1\tv1 k2\tv2" => (:k4("v4"), :k2("v3")).Seq}
timotimo m: my %h; %h<a> = { k1 => "v1", k2 => "v2" }; %h<b> = { k2 => "v3", k4 => "v4" }; my %result = flat %h<a>.list, %h<b>.list; say %result.perl
camelia {:k1("v1"), :k2("v3"), :k4("v4")}
alphah Actually what I'm trying to do is: I have two grammars, 1 grammar pasrse a config file and cretae a hash of configs, second grammar parses command line args, and I need to merge these two hashes into one hash
timotimo oh, what did you want keys existing in both hashes to do?
15:56 abraxxa left
alphah in case duplicates keys, the second hash overwrites the first hash 15:57
timotimo m: my %h; %h<a> = { k1 => "v1", k2 => "v2" }; %h<b> = { k2 => "v3", k4 => "v4" }; %h<a> <<=<< %h<b>; say %h.perl
camelia {:a(${:k1("v1"), :k2("v3"), :k4("v4")}), :b(${:k2("v3"), :k4("v4")})}
timotimo m: my %h; %h<a> = { k1 => "v1", k2 => "v2" }; %h<b> = { k2 => "v3", k4 => "v4" }; %h<a> <<=<< %h<b>; say %h<a>.perl 15:58
camelia ${:k1("v1"), :k2("v3"), :k4("v4")}
timotimo more like that
that looks right, doesn't it?
15:58 sufrostico joined
alphah looks right, will test on my code... 15:59
timotimo it probably descends into substructures 16:00
alphah I remeber, I tried that earlier,, it works in some cases, and in other cases it throws error regarding left side or right side values,, one is bigger than the other
timotimo so if any configured things are lists, better see how that influences things 16:01
alphah i'll try to reproduce the error
it was something like this (can not reproduce it now thugh): 16:07
Lists on either side of non-dwimmy hyperop of infix:<=> are not of the same length while recursing
left: 2 elements, right: 1 elements
16:07 sufrostico left
timotimo did you write >>=<<? 16:08
16:08 setty1 joined
alphah i wrote <<=<< 16:08
timotimo interesting, this might be a bug in the implementation of hyper, then
16:08 mcmillhj left 16:10 sufrostico joined
alphah just to be clear, not in the above code, I got this error earlier in my code, when change the config file contents (which get parsed by a grammar) 16:10
timotimo mhm, if you get that error again, you could output the .perl (or even .&to-json from JSON::Fast) both hashes beore attempting the merge
then clean up any strings with sensitive data and put it up somewhere for us to see
alphah will do 16:11
16:15 sufrostico left 16:16 sufrostico joined 16:18 Xliff_ left 16:19 incredible_ left 16:22 sufrostico left 16:23 committable6 left 16:24 sufrostico joined 16:34 Cabanossi left 16:35 Cabanossi joined 16:37 robertle left 16:46 mcmillhj joined 16:49 lostinfog joined 16:53 ifim left 16:59 setty1 left, eater left 17:01 eater joined 17:05 robertle joined, mr-foobar left 17:17 zakharyas left 17:22 sufrostico left 17:23 pat_js joined 17:28 Resol left 17:29 sufrostico joined 17:32 domidumont left 17:33 grondilu_ joined 17:36 grondilu left, domidumont joined 17:40 domidumont left, domidumont joined 17:41 domidumont left, domidumont joined 17:46 TEttinger joined 17:50 ovechkin joined 17:51 incredible joined
Geth Inline-Perl5: d03317170d | (Stefan Seifert)++ | META6.json
Version 0.25
17:57
17:58 ovechkin left 18:01 grondilu_ is now known as grondilu, grondilu left, grondilu joined
grondilu I was trying to golf FFT on RC. I got a neat short code that was just slightly inaccurate. Turns out I had to tweak &cis to make it better: paste.debian.net/928234/ 18:03
relevant part is:
sub my-cis($a) { $a < 0 ?? -cis($a + pi) !! cis($a) }
m: say cos(pi/4) + cos(pi + pi/4)
camelia -1.11022302462516e-16
18:03 ovechkin joined
grondilu m: say cos(pi/4) - cos((pi + pi/4) % pi) 18:04
camelia 0
grondilu seems a bit LTA imgo 18:05
*imho
basically I'm wondering if trigonometric functions could not use symmetric properties explicitely 18:06
though it seems no programming language seems to do it. 18:09
timotimo that'd be a little crazy :)
grondilu "crazy" sounds a bit harsh 18:10
slightly overkill, maybe
geekosaur you're ore in the realm of stuff like mathematica here 18:11
*more
18:11 cdg_ joined
grondilu I dunno 18:11
not really
geekosaur or numerical methods libraries
which aren't the default because they tend to be slower 18:12
18:13 cdg left
grondilu It bugs me that I have a shorter program for FFT but I can't use it because cis(x) is not exactly -cis(x + pi) :/ 18:13
18:18 cdg_ left
grondilu m: proto my-cos($) {*}; multi my-cos($x where * < 0) { my-cos(-$x) }; multi my-cos($x where $x > tau) { my-cos($x % tau) }; multi my-cos($x) { cos($x) }; say my-cos(pi/4) - my-cos(pi/4 + pi) 18:19
camelia 1.4142135623731
grondilu failed, nvm 18:20
could be done with polymorphism though: paste.debian.net/928235/ 18:43
grondilu tries it in Real.pm and recompiles rakudo
18:46 darutoko left
sjn \o 18:48
raschipi heya
sjn is getting a fail during the configuration step when rebuilding moar (I'm upgrading from an older version using rakudobrew) 18:49
has there been any important changes there lately? 18:50
geekosaur does it mention libtommath?
sjn would prefer not to nuke stuff
gfldex moarvm does need a `make distclean` every now and then
18:52 azawawi joined
azawawi iH 18:52
geekosaur olleH 18:54
azawawi :)
gfldex ıɐɥo
18:55 AlexDaniel joined 18:56 agentzh_ left 18:57 committable6 joined, ChanServ sets mode: +v committable6, domidumont left, sufrostico left
azawawi is reading en.wikipedia.org/wiki/Abjad_numerals 18:58
18:59 agentzh joined
timotimo how long is the tommath thing going to follow me around ;_; 19:03
19:10 azawawi left
sjn geekosaur: yeah, there's a libtommath warning there 19:11
geekosaur foreverrrrrrrr 19:12
anyway, "someone" oopsed with libtommath and you need to remove it from the local git reference and let it be re-downloaded 19:13
sjn ook
19:13 dwarring joined
sjn does that 19:13
geekosaur (or possibly just nuke ~/.rakudobrew and start over, which I do every so often because it doesn't do useful things like track tags so when a new version is tagged it doesn;t notice and reports itself as an older version) 19:14
sjn hm. is there a sane way to record which modules have been installed and then reinstall those?
(after nuking moar, that is) 19:15
geekosaur that I don't know
sjn possible a question for #perl6-toolchain 19:17
possibly*
RabidGravy sjn, see App::ModuleSnap in the ecosystem, it is precisely for trhat 19:20
obviously you have to install and run it before nuking everything 19:21
sjn \o/ 19:22
well, that one didn't seem to work 19:28
RabidGravy in what way? 19:35
19:39 moray` joined
RabidGravy haha, just tested it, okay 19:39
one minute 19:40
sjn getting a "Type 'META6' is not declared" error
when running p6-module-snapshot
RabidGravy yes, just fixing
sjn sweet
eater d'aww //= is a thing but =// isn't :( 19:41
jnthn eater: R 19:42
19:43 AlexDaniel left
jnthn If you want to reverse the argument order, anyway :) 19:43
eater jnthn: what do you mean?
19:44 Dunearhp___ left
jnthn I'm assuming you wanted to test the RHS for definedness instead of the left? 19:44
If not, I don't know what you want
gfldex m: my $v; 55 R//= $v;
camelia ===SORRY!===
Cannot find method 'ann' on object of type NQPMu
jnthn m: my $v; 55 [R//]= $v; 19:45
camelia Cannot modify an immutable Int
in block <unit> at <tmp> line 1
jnthn m: my $v; 55 R[//]= $v;
camelia ===SORRY!===
Cannot find method 'ann' on object of type NQPMu
jnthn Hmm...darn
eater I wanted to do A =// B, if B is defined do assign to a
jnthn I thought that would work
eater d'aww :')
jnthn But yeah, that R thing should do it
RabidGravy sjn, should be fixed now 19:46
jnthn In theory :)
sjn :d
19:47 moray` left, Cabanossi left
eater jnthn: any documentation on that, my Moar is just telling me Im doing it wrong :') 19:49
19:50 Cabanossi joined 19:51 Zoffix joined
Zoffix .tell El_Che you're good for packages. 2017.04.2 is out 19:52
yoleaux Zoffix: I'll pass your message to El_Che.
19:52 Zoffix left
eater oh 19:52
because I'm doing it wrong
jnthn: Thanks it works :D 19:53
19:54 zakharyas joined 19:55 cdg joined
b2gills m: my $v; 55 R[R[//]]= $v; say $v; # That doesn't seem right 19:55
camelia 55
19:55 pat_js left
sjn wishes that perl6 could offer a warning when people are using unnecessary convoluted programming constructs... "WARNING: Possible future headache and/or homicide reason detected" 19:57
raschipi It does provide enough rope to hang oneself. 20:00
20:00 gdonald left
sjn is more worried about the people who come later and have to rummage through the rotting corpses 20:00
perlpilot sjn: perhaps you could write a module that implements that functionality 20:01
20:01 gdonald joined 20:03 wamba left
sjn perlpilot: maybe I will (just waiting for DrForr's linter^H^H^H^H^H^Hparser to become good enough for it :D ) 20:03
samcv my job limits are still exceeding the time limit 20:04
and i split it into 20 sections, to install all the eco modules 20:05
moritz run 'em on hack? 20:06
samcv yeah. i should
that's www.p6c.org?
moritz hack.p6c.org 20:07
samcv oh
i don't have an account there then. only at www.p6c.org
moritz you do
samcv oh ok
moritz at least there's a /home/samcv :-)
and a user of that name
samcv ok i'm connected :) 20:08
yay cpu's
though idk if zef can do parallel testing for bulids
gfldex samcv: zef itself is basically a set of modules. Shouldn't be to hard to make many things happen at the same time with it. 20:11
20:13 espadrine joined 20:14 kurahaupo joined 20:15 ChoHag joined 20:19 dominix joined
dominix hi there. 20:20
I am looking for damian conway perl6 module Quantum::Computation
gfldex m: say „ohai dominix!“;
camelia ohai dominix!
dominix aloha ! 20:21
moritz dominix: Damian doesn't seem to have published it on modules.perl6.org/ :( 20:22
which means I have no idea where to find it
[Coke] eco: Wuantum
eco: Quantum
moritz modules.perl6.org/#q=quantum
[Coke] thank you moritz-bot!
dominix: where did you hear about it? 20:23
timotimo but that isn't dconway's module
dominix he is talking about it in his famous talk "On The Shoulders Of Giants" YAPC-NA 2016
so this "may" exists 20:24
moritz yes, he's good at giving talks, but not very good at interacting the community
[Coke] don't see it on cpan, either, or github.
moritz like, following community wide conventions for what you do with your modules, and stuff :-)
timotimo he's hardly got time to hang around here and learn about how we do things 20:25
20:25 dct joined
timotimo and to be fair, how we do things changes every now and then 20:25
[Coke] eh. it's not surprising to have a module you whipped up for a talk that you didn't quite get to publishing. I can't give him crap for that, moritz.
moritz [Coke]: did I give him crap for it? 20:26
sjn Hm... is Task::Star still someting people use? It seems zef is having trouble installing it
gfldex sjn: known issue, will be resolved soon 20:27
sjn ah, ok
timotimo yeah we haven't done a star release in a while, so the modules aren't really up to snuff
sjn sees ufo is there (does anyone still use that?) 20:28
moritz sjn: not really
sjn thought so
timotimo i'm also not sure if the list of modules inside star and Task::Star have been kept in sync 20:29
dominix I have been an heavy user of Parse::Recdescent very usefuul to me 20:32
the talk www.youtube.com/watch?v=Nq2HkAYbG5o 20:33
[Coke] moritz: yes, that's how I read the "not very good" comment. 20:38
TEttinger huh, I thought quantum was related to the superposition stuff that would later be called junctions 20:39
timotimo TEttinger: nah, it's a bit different still 20:42
raschipi Boolean superposition is different from Quantum Superposition. 20:43
perlpilot dominix: oh, just extract the module from the slides ;-> 20:46
20:48 cdg left
dominix perlpilot: he didn't show the content ... 20:49
20:49 cdg joined
sjn RabidGravy: seems p6-module-snapshot ignores --directory 20:49
moritz [Coke]: ah, the fun not being a native language speaker. For me, "not very good" is far away from "terrible", for example. Getting the nuances right can be hard for me :/ 20:50
RabidGravy sjn, probably
20:51 cdg left, cdg joined
TEttinger moritz: double-plus ungood 20:52
20:52 cdg left
moritz TEttinger: nah, just ungood 20:53
sjn single-plus ungood
20:53 cdg joined
TEttinger half-plus ungood 20:54
sjn moritz: the term you might be looking for is "has room for improvement" :)
understatements <3 20:55
perlpilot [Coke]: I think you read it the right way the first time even if moritz didn't quite mean it that way. Damian is infamous for showing off modules that don't seem to exist in a public space. :)
20:56 Redrield joined
Redrield Are there any hacks or workarounds that I need to do in order to get the Padre plugin to install on Windows? 20:56
perlpilot dominix: yeah, but he *described* it, right? Surely you should be able to reverse engineer it ;)
moritz underestimatatemens <3 20:57
21:00 cdg_ joined
dominix perlpilot: i would love to be able to ... 21:01
Redrield This is my .cpanm/build.log when I try to install the Perl6 plugin hastebin.com/mecuruxije.php
How could I solve this?
perlpilot Redrield: that's probably more of a question for #perl than #perl6 21:03
Redrield Alright
21:04 cdg left 21:05 sufrostico joined, gdonald left, raschipi left 21:06 gdonald joined, agentzh left
grondilu learns about "numeric underflow" and is amused 21:07
m: say 10**-304 21:08
camelia 1e-304
grondilu o.O
21:08 cdg_ left
grondilu 10**-304 gives me a "Numeric underflow" on my machine 21:09
21:09 cdg joined
TEttinger m: say 10**-305 21:09
camelia 1e-305
TEttinger m: say 10**-1005 21:10
camelia Numeric underflow
in block <unit> at <tmp> line 1

Actually thrown at:
in block <unit> at <tmp> line 1
TEttinger there we go
21:10 cdg left, RabidGravy left
grondilu wonders if there is any use of such small numbers anywhere 21:11
perlpilot LHC? 21:12
dominix ok bye 6ers ...
21:12 dominix left
grondilu perlpilot: I doubt it, not even them. 21:12
perlpilot Just like the 100th or 1000th digit of pi then ... not all that useful :) 21:18
21:21 zakharyas left 21:26 dct left
grondilu btw in the core pi currently has 18 decimals. Is there any particular reason to stop there? 21:28
m: say 18*log(10)/log(2) 21:29
camelia 59.7947057079725
grondilu guesses it's the best accuracy with 64 bits 21:30
timotimo This heloþor.p6 is the example provided in this repo, which you should have cloned or downloaded in the usual way. Check out the use of þ. Cool, isn't it? You can use any other directory instead of /app
This heloþor.p6 is the example provided in this repo, which you should have cloned or downloaded in the usual way. Check out the use of þ. Cool, isn't it? You can use any other directory instead of /app 21:31
... whoops
samcv moritz, hack.p6c.org doesn't have fuse 21:43
also interestingly if you don't have fuse it says that you can use a --appimage-extract option and it'll extract the appimage for you or something. that sounds fancy
so i guess fuse isn't strictly required. just more convenient
21:45 rindolf left
samcv also i don't have the `tee` command X| 21:48
Redrield I can't install the Perl6 plugin for Padre 21:49
samcv i believe you
Padre. is. buggy
Redrield sighs
samcv i would recommend github.com/samcv/Atom-as-a-Perl6-IDE 21:50
Redrield I don't suppose there's a good VSCode Perl6 plugin that's gone under the radar?
samcv these packages with Atom
Redrield I really don't feel like getting atom
samcv please see the link.
:\
Redrield Alrigh
samcv it has the best perl 6 support 21:51
DrForr New post on theperlfisher.blogspot.ro on grammars and metadata.
Redrield Yeah, I just... 21:52
I really don't like Atom
Ah well
samcv it has gotten a lot better than it used to be
and i'm pretty comfortable with those plugins listed on that page 21:53
timotimo i do all my perl6 in vim
21:54 mcmillhj left
MasterDuke_ vim here too 21:54
samcv somebody should make a PR to atom-nav-panel-plus to add native support in for getting classes/routine/method names on the sid ebar 21:56
without a config file. hmm
similar to how this PR is github.com/0tho/atom-nav-panel-plu...ll/6/files
so we don't have to have a .nav-panel-plus file or whatever 21:57
22:04 eroux left 22:09 mcmillhj joined
samcv oh looks like i hosed my PATH var and that's why i could'nt use any programs. good going /me 22:10
22:14 mcmillhj left 22:20 skids joined
samcv i still need fuse though 22:21
22:24 lizmat joined
samcv nice. ok. well worked around and it now trying to install every module 22:24
22:27 perigrin left 22:29 lizmat_ joined, perigrin joined, muzza joined 22:31 lizmat left
Redrield I was just looking through the examples on the site 22:32
22:32 muzza left
Redrield Is there an easier way to express infinity other than ∞? 22:32
sjn easier than one character?
Redrield Wrestling with unicode to get it to output isn't a strong suit of mine 22:33
sjn Inf
Redrield So I resort to having to google for it
Alright
22:34 muzza joined
samcv Inf 22:37
m: say Inf === ∞
camelia True
samcv Redrield,
22:37 lostinfog left
samcv oh sorry that was said. ignore me 22:37
moritz, so i have www.p6c.org and how do i host a page on like www.p6c.org/samcv or whatever? or is that more work 22:43
22:43 BenGoldberg joined 22:45 yqt left 22:47 muzza left
timotimo samcv: i don't think we have a setup for people to put stuff into a place like that 22:48
samcv k
timotimo i can make a user for something like samcv.p6c.org or appimages.p6c.org or so
22:49 muzza joined
timotimo and you can put whatever you want into your ~/public_html on hack.p6c.org and that'll be on hack.p6c.org/~samcv/ 22:49
samcv i want somewhere to put module pass/failing conditions first and forement
since i have a gh-pages for appimages atm
22:49 muzza left
timotimo are we going to get historic appimages, too? 22:52
samcv this isn't about appimages just modules and if they pass or fail and their build logs 22:54
but. are we? sure
like for every release of rakudo?
or what?
though still really need a command line flag to chdir before trying to access variables profided on command line to perl 6. ugh 22:55
timotimo not every
just back a few releases
samcv then everything will be good. and appimages will be exactly the same as the normal 22:56
yeah sure
22:56 robertle left
timotimo have we tried giving moarvm a commandline switch like that? 22:56
samcv nope. that is one way maybe
22:56 ChoHag left
samcv that is fine by me though. and would be very nice 22:56
22:57 espadrine left
samcv but i mean. the path has to be correct in rakudo when it's doing precomp and other things. since they're relative. so idk 22:59
if we change it in moarvm before loading the bytecode? won't that be equivilant to chdiring into the users CWD it (which breaks it)
23:01 labster joined, dwarring left 23:05 bjz joined
samcv timotimo, we could try it and see what happens i guess :P 23:16
23:17 Cabanossi left
samcv timotimo, nice. i just fixed nqp::chdir 23:19
23:19 bjz left
samcv from chdir failed: Unknown system error 2 >> chdir failed: no such file or directory 23:19
was looking at the function and noticed some weirdness
23:20 mcmillhj joined, Cabanossi joined 23:25 mcmillhj left 23:30 itaipu joined
samcv timotimo, well it works fine in my own perl6 but we will see what happens when i try using it when everything is relative and see if it doesn't vomit 23:35
if it doesn't vomit i will be very very happy
23:36 mcmillhj joined 23:38 AlexDaniel joined 23:42 mcmillhj left 23:44 AlexDaniel left
samcv timotimo, yeah no dice. 23:46
it has to happen after perl6.moarvm is loaded
timotimo oh crap 23:47
23:47 kyan joined
samcv Unhandled exception: While looking for 'ModuleLoader.moarvm': no such file or directory 23:48
and that's if i put it right before perl6.morvm and give a full path to perl6.moarvm (which i would be able to do in the shell script fine)
so this is not going to solve it
at least i fixed something in MoarVM tho 23:49
timotimo mhyes
samcv nice. it's exactly the same number of lines. including the header file 23:50
plus i added a new function
heh
and it works
which is most important. not unknown system error bleh
23:53 mcmillhj joined
samcv i'm gonna take a tap 23:55
23:59 mcmillhj left