»ö« Welcome to Perl 6! | perl6.org/ | evalbot usage: 'p6: say 3;' or /msg camelia p6: ... | irclog: irc.perl6.org or colabti.org/irclogger/irclogger_log/perl6 | UTF-8 is our friend! 🦋
Set by Zoffix on 25 July 2018.
irced m: for <'bleh' blah bloop' 'bleep meep> { say $_ } 00:01
camelia 'bleh'
blah
bloop'
'bleep
meep
irced m: for <'bleh' blah bloop' 'bleep meep> { say $_ ~~ / .* / } 00:02
camelia ｢'bleh'｣
｢blah｣
｢bloop'｣
｢'bleep｣
｢meep｣
00:02 rindolf left
irced m: for <'bleh' blah bloop' 'bleep meep> { say $_ ~~ / <[']>? (.*) <[']>? / } 00:02
camelia ｢'bleh'｣
0 => ｢bleh'｣
｢blah｣
0 => ｢blah｣
｢bloop'｣
0 => ｢bloop'｣
｢'bleep｣
0 => ｢bleep｣
｢meep｣
0 => ｢meep｣
irced eyes cross. 00:03
m: for <'bleh' blah bloop' 'bleep meep> { $_ ~~ / <[']>? (.*) <[']>? /; say ~$0; } 00:04
camelia bleh'
blah
bloop'
bleep
meep
00:04 pecastro left
irced m: for <'bleh' blah bloop' 'bleep meep> { $_ ~~ / <[']>? (.*) <[']>? /; say ~$1; } 00:05
camelia Use of Nil in string context





in block at <tmp> line 1
Use of Nil in string context
in block at <tmp> line 1
Use of Nil in string context
in block at <tmp> line 1
Use of Nil in string context
in block at <tmp> l…
irced m: for <'bleh' blah bloop' 'bleep meep> { $_ ~~ / <[']>? ([^']*) <[']>? /; say ~$0; }
camelia 5===SORRY!5=== Error while compiling <tmp>
Unable to parse expression in metachar:sym<( )>; couldn't find final ')' (corresponding starter was at line 1)
at <tmp>:1
------> 3leep meep> { $_ ~~ / <[']>? ([^']*) <[']7⏏5>? /; say ~$0; }…
irced m: for <'bleh' blah bloop' 'bleep meep> { $_ ~~ / <[']>? (<-[']>*) <[']>? /; say $0; } 00:06
camelia ｢bleh｣
｢blah｣
｢bloop｣
｢bleep｣
｢meep｣		 00:07
irced that's somewhat perly
maybe I can go sed on it
but i'll spare you the spam, any suggestions on making that more perly? 00:08
00:08 kotrcka left 00:09 nepugia left
timotimo m: for <'bleh' blah bloop' 'bleep meep> { say m/"'"? <( .*? )> "'"? } 00:09
camelia 5===SORRY!5=== Error while compiling <tmp>
Couldn't find terminator / (corresponding / was at line 1)
at <tmp>:1
------> 3bleep meep> { say m/"'"? <( .*? )> "'"? 7⏏5}
expecting any of:
/
timotimo m: for <'bleh' blah bloop' 'bleep meep> { say m/"'"? <( .*? )> "'"?/.Str }
camelia



timotimo m: for <'bleh' blah bloop' 'bleep meep> { say m/"'"? <( .*? )> "'"?/ }
camelia ｢｣
｢｣
｢｣
｢｣
｢｣
irced begins to applaud but stops mid air.
timotimo yeah that didn't work 00:10
m: for <'bleh' blah bloop' 'bleep meep> { say $_ ~~ /"'"? <( .*? )> "'"?/ }
camelia ｢｣
｢｣
｢｣
｢｣
｢｣
irced checks his watch.
timotimo m: for <'bleh' blah bloop' 'bleep meep> { say $_ ~~ /"'"? <( .* )> "'"?/ }
camelia ｢bleh'｣
｢blah｣
｢bloop'｣
｢bleep｣
｢meep｣
00:10 candr joined
irced looks up from his watch face for an ephemeral moment. 00:10
timotimo the trim methods and subs are only for whitespace 00:11
m: for <'bleh' blah bloop' 'bleep meep> { say $_ ~~ /^"'"? <( .*? )> "'"?$/ } 00:12
camelia ｢bleh｣
｢blah｣
｢bloop｣
｢bleep｣
｢meep｣
timotimo that's the important bit
irced hmm
AlexDaniel tobs: nice
timotimo anchoring it, else it'll just get the shortest match it can 00:13
irced golf claps.
candr does a nested module have access to it's parent's modules private subs and variables? or does it even make sense to nest modules in perl6?
timotimo "private" is something only methods have 00:14
subs meanwhile are usually lexically scoped
irced timotimo: upon inspection, that is indeed quite perly. thanks!
i'll give the floor to candr though there are some wrinkles I think worth ironing 00:15
timotimo if you put the one module lexically into the other, then it will have access, too 00:16
candr maybe a better question is this.. i want to break up some source code into multiple files and wars thinking of making a few new .pm6 files with a single module in each. Those modules need to share a few boolean variables though so i war thinking of making a module Foo with what needs to be shared and a then a Foo::A and Foo::B 00:18
timotimo not sure what the bast way for that is, maybe it's better to have your users handle that? dunno 00:25
00:27 Manifest0 left
thundergnat m: say "'bleh' blah bloop' 'bleep meep 'can't' '''''up-to-date' jack-o'-lantern 1990's''" ~~ m:g| \w \w* [\S+ \w*]* \w |; # without other comment 00:27
camelia (｢bleh｣ ｢blah｣ ｢bloop｣ ｢bleep｣ ｢meep｣ ｢can't｣ ｢up-to-date｣ ｢jack-o'-lantern｣ ｢1990's｣)
thundergnat actually, that probably should have been jack-'o-lantern. ah well, same, same. 00:29
irced m: say "'bleh' blah bloop' 'bleep meep 'can't' '''''up-to-date' jack-o'-latnern 1990's''" ~~ m:g| ^^"'"? <(.*?)> "'"?$$|
camelia (｢bleh' blah bloop' 'bleep meep 'can't' '''''up-to-date' jack-o'-latnern 1990's'｣)
irced m: say "'bleh' blah bloop' 'bleep meep 'can't' '''''up-to-date' jack-o'-lantern 1990's''" ~~ m:g| ^^"'"? <(.*?)> "'"?$$| 00:30
camelia (｢bleh' blah bloop' 'bleep meep 'can't' '''''up-to-date' jack-o'-lantern 1990's'｣)
irced ack
err, that's right in a way tho 00:31
00:31 Manifest0 joined
irced m: say "'bleh' blah bloop' 'bleep meep 'can't' '''''up-to-date' jack-o'-lantern 1990's''" ~~ m:| ^^"'"? <(.*?)> "'"?$$| 00:31
camelia 5===SORRY!5=== Error while compiling <tmp>
Colons may not be used to delimit quoting constructs
at <tmp>:1
------> 3to-date' jack-o'-lantern 1990's''" ~~ m:7⏏5| ^^"'"? <(.*?)> "'"?$$|
expecting any of:
colon pair (restr…
irced m: say "'bleh' blah bloop' 'bleep meep 'can't' '''''up-to-date' jack-o'-lantern 1990's''" ~~ / ^^"'"? <(.*?)> "'"?$$ /
camelia ｢bleh' blah bloop' 'bleep meep 'can't' '''''up-to-date' jack-o'-lantern 1990's'｣
irced thundergnat: good work, still dissecting it thanks 00:32
m: say "'bleh' blah bloop' 'bleep meep 'can't' '''''up-to-date' jack-o'-lantern 1990's''".split(" ") ~~ / ^^"'"? <(.*?)> "'"?$$ / 00:33
camelia ｢bleh｣
irced m: say "'bleh' blah bloop' 'bleep meep 'can't' '''''up-to-date' jack-o'-lantern 1990's''".split(" ").lines ~~ / ^^"'"? <(.*?)> "'"?$$ /
camelia ｢bleh' blah bloop' 'bleep meep 'can't' '''''up-to-date' jack-o'-lantern 1990's'｣
irced m: say "'bleh' blah bloop' 'bleep meep 'can't' '''''up-to-date' jack-o'-lantern 1990's''".split(" ").lines ~~ / ^^"'"? <(.*)> "'"?$$ / 00:35
camelia ｢bleh' blah bloop' 'bleep meep 'can't' '''''up-to-date' jack-o'-lantern 1990's''｣
irced m: say "'bleh' blah bloop' 'bleep meep 'can't' '''''up-to-date' jack-o'-lantern 1990's''".split(" ").lines ~~ / ^^"'"? <(-[ ' ])> "'"?$$ /
camelia 5===SORRY!5===
Unrecognized regex metacharacter - (must be quoted to match literally)
at <tmp>:1
------> 03990's''".split(" ").lines ~~ / ^^"'"? <(7⏏5-[ ' ])> "'"?$$ /
Unable to parse regex; couldn't find final '/'
at <tmp>:1
---…
irced m: say "'bleh' blah bloop' 'bleep meep 'can't' '''''up-to-date' jack-o'-lantern 1990's''".split(" ").lines ~~ / ^^"'"? (<-[ ' ]>) "'"?$$ /
camelia Nil
irced m: say "'bleh' blah bloop' 'bleep meep 'can't' '''''up-to-date' jack-o'-lantern 1990's''".split(" ").lines ~~ / ^^"'"? (<-[ ' ]>*) "'"?$$ / 00:36
camelia Nil
irced m: say "'bleh' blah bloop' 'bleep meep 'can't' '''''up-to-date' jack-o'-lantern 1990's''".split(" ").lines ~~ / ^^"'"? (<-[ ' ]*>) "'"?$$ /
camelia 5===SORRY!5=== Error while compiling <tmp>
Unable to parse expression in metachar:sym<assert>; couldn't find final '>' (corresponding starter was at line 1)
at <tmp>:1
------> 3'".split(" ").lines ~~ / ^^"'"? (<-[ ' ]7⏏5*>) "'"?$$ /
irced anywayz
thanks all
00:42 aborazmeh left, reach_satori_ joined 00:48 Manifest0 left 00:53 Manifest0 joined 01:00 Manifest0 left
AlexDaniel m: my $s = (1…∞); say $s.iterator.pull-one; say $s.iterator.pull-one 01:01
camelia 1
The iterator of this Seq is already in use/consumed by another Seq
(you might solve this by adding .cache on usages of the Seq, or
by assigning the Seq into an array)
in block <unit> at <tmp> line 1
AlexDaniel eh, what am I missing? I want to pull values from $s one by one when I need
so I can store the iterator, right, but… I'm not sure that I like that
m: my $s = (1…∞); say $s[0]; say $s.skip(1) 01:03
camelia 1
(...)
AlexDaniel m: my $s = (1…∞); say $s[0]; say $s.skip(1); say $s[0]; say $s.skip(1)
camelia 1
(...)
1
(...)
AlexDaniel whoops, not what I wanted :)
thundergnat m: my $s = cache (1…∞); say $s.iterator.pull-one; say $s.iterator.pull-one 01:04
camelia 1
1
AlexDaniel thundergnat: no, I don't want caching
thundergnat Right but pulling one didn't change the sequence.
returned 1 each time.
01:04 empee0 left 01:05 Manifest0 joined
AlexDaniel thundergnat: it kinda did, except that it gave you a new iterator for the cached Seq 01:05
thundergnat Ah. D'oh. Right. 01:06
AlexDaniel m: my $s = (1…∞); $it = $s.iterator; $it.pull-one; say Seq.new($it) 01:07
camelia 5===SORRY!5=== Error while compiling <tmp>
Variable '$it' is not declared
at <tmp>:1
------> 3my $s = (1…∞); 7⏏5$it = $s.iterator; $it.pull-one; say Seq
AlexDaniel m: my $s = (1…∞); my $it = $s.iterator; $it.pull-one; say Seq.new($it)
camelia (...)
AlexDaniel m: my $s = (1…∞); my $it = $s.iterator; say $it.pull-one; say Seq.new($it)[^10]
camelia 1
(2 3 4 5 6 7 8 9 10 11)
AlexDaniel ok so I need to rewrap the iterator?
I don't mind that, I guess…
01:11 khisanth_ left 01:12 Manifest0 left 01:17 noisegul_ joined 01:19 Manifest0 joined 01:21 noisegul left 01:24 khisanth_ joined 01:30 Manifest0 left 01:31 kaare_ joined
tobs AlexDaniel: that seems to be required, yeah. I once did this github.com/taboege/p6-Seq-Partial (without putting it into the ecosystem it seems :/) 01:33
irced anywayz 01:34
01:34 molaf left
irced i ended up using the UCD code point quote as in my $regex-pattern = rx / ^<:!quote>? (.*?) <:!quote>? /; 01:36
01:37 Manifest0 joined
irced which covers the other quote character as well 01:38
amazingly, but i can't see where it's explicitly documented at www.unicode.org 01:43
01:45 Manifest0 left
irced any1 know where the exhaustive list of such names Unicode properties can be found? 01:47
01:47 molaf joined 01:50 Manifest0 joined
vrurg irced: did you try wikipedia's 'unicode character property' topic? There is a link to a PDF at the references section. Could be what you're looking for. 01:50
irced vrurg: cool let me see
vrurg To justify myself: I googled it! ;) 01:51
I mean, to excuse myself... 01:52
irced vrurg: I don't see it in there
vrurg www.unicode.org/versions/Unicode12.0.0/ch04.pdf – this one? 01:53
irced vrurg: i tried duck duck go on some working names
vrurg en.wikipedia.org/wiki/Unicode_char...r_property
irced vrug: yep yep
vrurg The wiki page might have even more info. 01:54
vrurg still prefers google over duckduck due to better relevance
irced oh, I am allergic to google and microsoft. 01:55
found something that might work with perl6 01:58
found something that might work with perl6 02:00
02:01 jaldhar left
irced nope, could only find the code points for specific unicode, I am interested in these names that cover complex matches i.e. several code points lumped into one 02:02
i see many code points that can match via \c[LONG NAME FOR UNICODE]
which is cool, so thanks for the help 02:03
AlexDaniel m: my $i = 3; my $it = ((1…∞).rotor(1 => 1).flat.rotor(2 => 1).flat).skip(2).iterator; say (1, 3, { $i++; my $val = $it.pull-one; $it = Seq.new($it).rotor($val - $i => 1, ($val - 1 => 1) xx ∞).flat.iterator; $val} … ∞)[^100] 02:05
camelia (1 3 7 9 13 15 21 25 31 33 37 43 49 51 63 67 69 73 75 79 87 93 99 105 111 115 127 129 133 135 141 151 159 163 169 171 189 193 195 201 205 211 219 223 231 235 237 241 259 261 267 273 283 285 289 297 303 307 319 321 327 331 339 349 357 361 367 385 391 3…
irced vrurg: i also see some ranges so that's cool.
AlexDaniel .tell lizmat maybe this is a good benchmark: colabti.org/irclogger/irclogger_lo...07-06#l154 02:06
yoleaux AlexDaniel: I'll pass your message to lizmat.
irced in fact, looks like i was mistaken about the specific named property :quote
AlexDaniel .tell lizmat it's really slow but I don't think it should be
yoleaux AlexDaniel: I'll pass your message to lizmat.
candr if i have an `our class A` in a `module Foo`, how can I bring A into the current scope? if I `use Foo` i need to access A with Foo::A everywhere. I don't want to `class A is export` in Foo because that forces it into the current scope which doesn't always seem appropriate. I would think there is something `use Foo::A` or `from Foo use A` .... can't
find it in the docs though...
irced nevertheless, such named properties are available in perl6 without \c[CODE POINT NAME] via <:concise property name>
vrurg candr: use import classes. like is export(:FOO) 02:08
irced: I didn't work that much with unicode classes. Usually I'm ok with what's documented for Regex'es. 02:09
irced vrurg: you helped me get a more proper solution anyway as follows 02:11
candr So if i want to be able to arbitrarily bring any thing from Foo into the current scope I have to assign a separate import class to every "our" in the module ahead of time? Seems... redundant...
irced my regex quote { <[\c[QUOTATION MARK]\c[APOSTROPHE]]> }; my $regex-pat = rx / ^<quote>? (.*?) <quote>?$ /; ~~ $regex-pattern etc. i yield for other discussions now. 02:12
vrurg candr: Look at docs.perl6.org/language/modules#is_export – perhaps you'll have some better ideas. Besides, you can always define your own EXPORT sub for the purpose. 02:13
candr: moreover, EXPORT would be the best solution for you, perhaps.
candr i am just not liking how export leaves to trace as to where something came from... seems like that could get confusing quick... 02:15
02:15 MasterDuke left
AlexDaniel .tell lizmat I looked at the profile but I don't see anything obvious. I think it's an interesting case, unless I'm missing something obvious, it should be just a bunch of iterators glued together in interesting ways. The logic basically skips values at certain indexes (that's it!). Even though it's not fun to have 100 iterators on top of each other, it should still be really really fast. 02:16
yoleaux AlexDaniel: I'll pass your message to lizmat.
vrurg candr: Usually it doesn't.
candr: If one needs to know where something came from then there're ways to introspect concrete objects and type objects as well. 02:17
candr just seems like you should be able to know where something came from by just reading the source code... either from the use statement or by a fully qualified name. 02:21
"Note: there currently is no way for the user to import a single object if the module author hasn't made provision for that, and it is not an easy task at the moment (see RT #127305). One way the author can provide such access is to give each export trait its own unique tag. (And the tag can be the object name!)."
synopsebot RT#127305 [open]: rt.perl.org/Ticket/Display.html?id=127305 EXPORT ignored when unit module/package is used
candr ^ found that in the docs
seems it has been thought of, but is a tricky issue... 02:22
vrurg candr: That ticket is not valid. EXPORT must be defined outside of any module/package/class declaration. 02:23
Note that EXPORT can be multi. With that one has numerous possibilities of exporting symbols in any combinations, including aliasing and re-exporting from other modules. That's pretty much enough, I think. 02:24
candr: You can have a look at Red sources. It's using EXPORT and does't pretty much in a self-explanatory way. 02:25
s/does't/doest it/ 02:26
candr I'll take a look... it does seem to have a similar file structure as what I am working on now.. 02:28
02:30 irced left
candr thanks vrurg 02:30
vrurg candr: welcome :) 02:31
02:41 Cabanossi joined 02:42 Manifest0 left, Cabanoss- left 02:47 Manifest0 joined 02:54 jaldhar joined, Manifest0 left 03:00 candr left, Manifest0 joined 03:02 kaare_ left 03:06 jaldhar left, jaldhar joined
AlexDaniel .tell lizmat here's a more readable version: github.com/ajs/perl6-Math-Sequence...69f6c12b8d 03:10
yoleaux AlexDaniel: I'll pass your message to lizmat.
03:23 Manifest0 left 03:29 Manifest0 joined 03:36 Manifest0 left 03:37 candr joined
candr Ok I must be missing something simple.. I have a Foo.pm6 with a module Foo and an `our sub a()` in the module... from main.p6 I `use Foo;` and call the sub with `Foo::a;` , but get a runtime error: Could not find symbol '&a' 03:40
03:41 Manifest0 joined 03:45 Manifest0 left 03:52 Manifest0 joined 03:56 Manifest0 left
candr interesting... seems to be a conflict in use... i have some other modules that use the one i need and when i comment those out the sub is found... seems like i might need the equivelent of #ifndef #define.... guards... so modules don't get used more than once... 03:58
AlexDaniel candr: it should be working just fine 03:59
candr: can you golf it? 04:00
what's the minimal set of files to reproduce the issue?
candr yeah... let me write something up...
04:02 dominix_ joined, Manifest0 joined
AlexDaniel I need to get some sleep, will check it later 04:04
04:05 dominix__ left 04:06 Manifest0 left
candr I got a minimal test if you have just a sec... it is interesting... seems module order also matters... 04:11
04:11 Manifest0 joined
candr pastebin.com/yvcgzjza 04:16
main.p6 (pastebin.com/yvcgzjza) Foo.pm6 (pastebin.com/gFs5whC6) Foo_Bar.pm6 (pastebin.com/XVa9UJF0) 04:17
oh shoot... I think I just figured it out... I think I need `use lib` in Foo_Bar also .... 04:19
nope... that gives another error ... 'use lib' may not be pre-compiled 04:20
ah got it.... the `use Foo;` in Foo_Bar.pm6 needs to be outside of the module. 04:24
04:54 marmor joined 05:05 Manifest0 left 05:09 Manifest0 joined 05:17 candr left 05:22 Manifest0 left 05:25 marmor left 05:26 Manifest0 joined 05:31 Manifest0 left 05:37 Manifest0 joined, aborazmeh joined, aborazmeh left, aborazmeh joined 05:45 domidumont joined 05:49 Manifest0 left 05:51 molaf left, ryn1x joined
ryn1x is this a bug? gist.github.com/ryn1x/360e63f89075...879ffc507a 05:54
05:55 ryn1x left, ryn1x joined, Manifest0 joined
ryn1x using the traditional if statement works fine, but the postfix style errors with a strange List error... 05:56
06:04 Manifest0 left 06:07 lizmat left
Kaiepi m: module M { our sub foo { say 'foo' } }; M::foo() if 1 06:07
camelia Cannot find method 'Nil' on object of type List
in block <unit> at <tmp> line 1
Kaiepi oh
m: module M { our sub foo { say 'foo' } }; &M::foo() if 1
camelia foo
Kaiepi that's how you use it ryn1x
the problem is `our sub foo` is actually named &foo 06:08
m: module M { OUR::<foo> := anon sub foo { say 'foo' } }; M::foo if 1 06:09
camelia Could not find symbol '&foo'
in block <unit> at <tmp> line 1
06:09 Manifest0 joined
ryn1x ah ok... but the & is only required with the if... 06:09
06:10 marmor joined
Kaiepi m: module M { our sub foo { say 'foo' } }; M::foo 06:10
camelia foo
ryn1x something about the postfix if needs it's "real" name...
Kaiepi hm
06:13 domidumont left
Kaiepi m: module M { OUR::<foo> := anon sub { say 'foo' } }; M::foo 06:14
camelia Could not find symbol '&foo'
in block <unit> at <tmp> line 1
Kaiepi m: module M { OUR::<foo> := anon sub { say 'foo' } }; M::foo()
camelia Could not find symbol '&foo'
in block <unit> at <tmp> line 1
06:14 domidumont joined
Kaiepi m: module M { BEGIN OUR::<foo> := anon sub { say 'foo' } }; M::foo() 06:14
camelia foo
Kaiepi m: module M { BEGIN OUR::<foo> := anon sub { say 'foo' } }; M::foo
camelia WARNINGS for <tmp>:
Useless use of constant value M::foo in sink context (line 1)
Kaiepi m: module M { BEGIN OUR::<foo> := anon sub { say 'foo' } }; M::foo() if 1 06:15
camelia foo
Kaiepi there's another way to go about it
06:16 Manifest0 left, domidumont left
ryn1x adding the & works fine... I just wonder if the inconstancy is a bug... 06:16
06:20 Manifest0 joined
Kaiepi possibly 06:22
it's hard to say until someone more familiar with that part of perl 6's internals logs in
06:25 marmor left 06:35 kybr left 06:37 Manifest0 left 06:38 kybr joined
ryn1x github.com/rakudo/rakudo/issues/3034 opened an issue just in case 06:38
06:39 lizmat joined 06:42 Manifest0 joined 06:47 Manifest0 left
ryn1x huh... not only do you need to add the & when calling the sub with the if statement modifier, but you also have to add the parens.... without the parens there is no error, but the sub does not do anything.... 06:48
06:48 rindolf joined 06:51 Manifest0 joined 06:58 Manifest0 left 07:01 lizmat left 07:03 Manifest0 joined 07:08 Manifest0 left 07:10 antoniogamiz joined 07:12 MasterDuke joined 07:14 Manifest0 joined 07:18 domidumont joined 07:21 Manifest0 left 07:25 marmor joined, Manifest0 joined 07:26 vrurg left 07:34 marmor left, domidumont left 07:35 domidumont joined 07:44 Manifest0 left 07:47 Celelibi left, domidumont left 07:49 Manifest0 joined 08:02 molaf joined 08:09 aborazmeh left 08:18 telex left 08:19 Manifest0 left, telex joined 08:24 Manifest0 joined 08:31 Manifest0 left 08:38 Manifest0 joined 08:42 lancew joined 08:43 molaf left 08:49 Manifest0 left 08:54 Manifest0 joined 08:55 antoniogamiz left 08:57 ryn1x left 09:00 antoniogamiz joined, antoniogamiz left 09:06 Manifest0 left 09:12 Manifest0 joined 09:14 rfold joined 09:16 Manifest0 left 09:17 kylese joined 09:20 mniip left 09:21 Manifest0 joined 09:29 kylese left, kylese joined 09:31 Manifest0 left 09:35 Manifest0 joined 09:40 Manifest0 left 09:41 kylese left 09:44 Kaiepi left 09:45 Manifest0 joined, Kaiepi joined 09:46 Sgeo__ joined 09:49 Sgeo_ left 09:50 sena_kun joined 09:51 Manifest0 left 10:00 Manifest0 joined 10:01 Kaiepi left 10:05 Kaiepi joined 10:07 nepugia joined 10:21 Kaiepi left, Kaiepi joined 10:24 Black_Ribbon left 10:29 Manifest0 left 10:37 Manifest0 joined 10:38 Kaiepi left 10:39 Kaiepi joined 10:45 Manifest0 left 10:49 aborazmeh joined, aborazmeh left, aborazmeh joined, Manifest0 joined 10:54 Manifest0 left 11:01 Manifest0 joined
Kaiepi > wcswidth "\c[FACE WITH TEARS OF JOY]\0", 1 11:10
4294967295
still really buggy but wide string support for nativecall is coming along! 11:11
11:11 Manifest0 left
AlexDaniel squashable6: status 11:15
11:15 lancew left
squashable6 AlexDaniel, 🍕🍕 SQUASHathon is in progress! The end of the event in 1 day and ≈2 hours. See github.com/rakudo/rakudo/wiki/Mont...Squash-Day 11:15
Kaiepi how can i pass NULL to a nativecall sub that takes a string?
11:17 lancew joined
timotimo just pass Str 11:19
11:20 Manifest0 joined
Kaiepi thanks 11:21
11:23 pecastro joined 11:30 Manifest0 left 11:35 Manifest0 joined 11:39 Manifest0 left
AlexDaniel aaand I just merged a bunch of PRs 11:43
11:46 Manifest0 joined
lancew :) 11:51
lucs My perl6 is installed by root. 11:53
I can install a module as non-root with 'zef --install-to=/some/dir install Some::Module'.
How do I make the module visible to my programs; something like PERL6LIB=...?
Is this the correct approach? 11:54
12:01 Manifest0 left 12:06 Manifest0 joined
ugexe ryn1x: my &foo — how else would one test this for truth *without* invoking it? 12:09
hence needing to invoke it to test the result 12:10
otherwise yku are testing the symbol for truth not the result 12:11
12:17 noisegul_ left, Manifest0 left 12:18 kaare_ joined, lancew left 12:19 marmor joined, aborazmeh left 12:22 Manifest0 joined 12:26 goon_ left, Manifest0 left 12:33 Manifest0 joined 12:40 kaare_ left, marmor left 12:41 denis666 joined 12:43 denis666 left
lucs ...? 12:43
El_Che .tell lizmat: sungo.wtf/2019/07/05/leaving-irc.perl.org.html 12:45
yoleaux El_Che: What kind of a name is "lizmat:"?!
El_Che .tell lizmat sungo.wtf/2019/07/05/leaving-irc.perl.org.html
yoleaux El_Che: I'll pass your message to lizmat.
timotimo lucs: i think you address it with -I inst#/some/dir 12:47
lucs timotimo: I'm kinda looking for a solution where I don't have to specify the directory on each invocation. 12:49
Also, what if I install more than one module locally? 12:50
AlexDaniel wheeeew I think things are alright now 12:55
there's a lot of fun going on in github.com/ajs/perl6-Math-Sequences :)
please join :)
sena_kun AlexDaniel++ 12:56
AlexDaniel we might not have a webhook but that's not going to stop us :)
13:00 Manifest0 left 13:01 jeromelanteri joined
lucs attempts a cheap trick... 13:04
So I guess it's impossible to install modules locally and to have them visible by default to my (root owned) perl6 executable, eh.
El_Che export PERL6LIB ?
(didn't read the backlog)
thundergnat AlexDaniel: It may be worth putting up a file somewhere editable like this: gist.github.com/thundergnat/474ec2...f3f4bbad7d so people can "claim" a sequence and not step all over each other. 13:05
And update it every time you push.
13:06 Manifest0 joined
lucs El_Che: Yeah, something like that, but not sure how. 13:06
El_Che lucs: yeah, the infamous sha1 13:07
woolfy Claudio Ramirez: Thanks, that is one heavy blog by sungo... wow... Anyway lizmat is on the road driving home from another country after 4 days of intense work and she might not read / react soon. 13:09
13:09 marmor joined
El_Che woolfy: hi, have fun. Take some time to decompress! 13:09
woolfy Claudio Ramirez: I am at home, not with lizmat (does not happen often that we are 4 days apart from eachother, we both had different things to do), and I will make sure lizmat will take time to decompress. 13:12
El_Che woolfy: good she's coming back then!
13:13 marmor left
timotimo lucs: yeah, PERL6LIB will also accept that value 13:16
woolfy Claudio Ramirez: indeed. 13:18
13:19 Manifest0 left
AlexDaniel thundergnat: maybe, or maybe we should just merge things faster 13:19
thundergnat AlexDaniel: Or that. :-) 13:20
lucs timotimo: Maybe I should have tried :/ Thanks, I'll let you know how it goes.
AlexDaniel thundergnat: I wonder, do you actually need `lazy` ? 13:21
timotimo lucs: if ".say for $*REPO.repo-chain" outputs an inst# in your home, you can also probably use --to with that 13:22
MasterDuke AlexDaniel: is there a way to tell what sequences are missing/wanted?
AlexDaniel MasterDuke: grep -B 1 '&NOSEQ ...' ./lib/Math/Sequences/Integer.pm6 13:23
thundergnat AlexDaniel: Some do, some don't. At least one of the sequences I added hung without it so I started adding it defensively.
13:23 Manifest0 joined
AlexDaniel thundergnat: there are just 69 sequences that are not implemented yet. Should I file a ticket for each of them? 13:23
thundergnat Or perhaps superstitiously may be a better term...
AlexDaniel thundergnat: that'd serve as a todo list and as a discussion platform… also with a mechanism to assign yourself :) 13:24
thundergnat AlexDaniel: Seems like it might be helpful but I don't want to make more work for you (and JJ), you guys do a lot as it is. 13:25
AlexDaniel that's just 3 pagefuls of tickets
thundergnat burnout is a thing
AlexDaniel well, I have a script! Let me try that
thundergnat: says a person who implemented a jaw-dropping amount of sequences :) 13:27
sena_kun agrees with thundergnat and wants to express gratitude for such amounts of work
thundergnat AlexDaniel: Yeah, but I'm not in public and don't have people expectantly watching me. 13:29
thundergnat watches expectantly :-)
13:30 nepugia left
lucs timotimo: Makes sense. A problem I have: for personal project handling reasons, even as $USER, I change the value of $HOME a lot. 13:31
MasterDuke AlexDaniel: i don't know. wasn't there some picture posted of you programming on the bus? that seems pretty public to me
AlexDaniel MasterDuke: yes :D
lucs Hmm... Maybe I can simply symlink those $HOME/.perl6 to /home/$USER/.perl6, that should work.
AlexDaniel thundergnat: what should I write in the ticket body? 13:32
qq:to/TEMPLATE/
Sequence [$sequence](oeis.org/$sequence) is not implemented yet.
TEMPLATE
anything else?
13:32 rindolf left
thundergnat AlexDaniel: Um. Not really... Though it might be nice if you can capture the comment line above each unimplemented sequence. Has the common name for the sequence to ease wikipedia searching. I suppose users can get that from OEIS though. 13:35
tbrowder AlexDaniel: i just submitted pr for math-seq 13:36
AlexDaniel thundergnat: yeah, the name of the sequence is kinda important, I agree 13:38
thundergnat: let's see what info OEIS module can provide… 13:39
13:39 rindolf joined 13:44 Manifest0 left 13:49 Manifest0 joined
AlexDaniel thundergnat: ok, more feedback please? gist.github.com/AlexDaniel/2c62bc0...ues-p6-L11 13:51
thundergnat: like, I have access to all of the fields: modules.perl6.org/dist/OEIS:cpan:T...y.pm6#L232 13:52
so I can put anything in the ticket description
13:52 mowcat joined
thundergnat AlexDaniel: That looks pretty good to me. Really, the critical parts to me are the ID, the NAME and a link directly to the oeis page. Everything else is going to be a little manual anyway as there is o much variation between sequences 13:56
* so 13:57
tobs AlexDaniel: I wish that module was more helpful already. OEIS supports searching for all sequences tagged 'core' for example (wrt github.com/ajs/perl6-Math-Sequence...08919554), but I couldn't design a cute interface to the advanced search syntaxes in time.
AlexDaniel: other than that, this script would be really helpful. I had the same idea but suppressed it yesterday :/ 14:00
AlexDaniel tobs: testing it right now :)
14:00 jeromelanteri left
AlexDaniel tobs: it's good that we first implemented a bunch of sequences before populating the ticket list, less spammy this way 14:01
tobs: is there a short name of some sort?
tobs: right now the format of the issue title is like A000005 / d(n) (also called tau(n) or sigma_0(n)), the number of divisors of n. 14:02
tobs this doesn't exist afaik. Struck me as something ajs came up with
AlexDaniel tobs: no way? They have no short identifiers whatsoever??? 14:03
tobs let me double check 14:04
AlexDaniel: no, there is only the ID and the full name (oeis.org/eishelp1.html) 14:05
AlexDaniel yes, I guess you're right… ew
alright 14:06
MasterDuke: can't find that pic btw, somewhere on Zoffix's twitter I think. But it's really old, couldn't scroll that far
tobs I'll be back later, have to arrange and catch a train 14:07
AlexDaniel thundergnat: anyway, ready for the first automated ticket? 14:08
thundergnat AlexDaniel: There is a hash %oeis-core in ./lib/Math/Sequences/Integer.pm6 that has ajs's short names
AlexDaniel thundergnat: yeah but uh… do I want to import that and do a lookup?
thundergnat AlexDaniel: Woot! lay it on us!
AlexDaniel wait, let's try with short naes
long titles are really annoying 14:09
thundergnat watches expectantly. SCNR
AlexDaniel thundergnat: but is there a short-name <=> id mapping? 14:10
thundergnat The hash is short_name => seq_id pairs. 14:11
in Integer.pm6, if that wasn't clear.
AlexDaniel thundergnat: the value is the variable, so maybe… 14:12
thundergnat Could invert it and get id => short_name pretty easily
AlexDaniel thundergnat: there's no id, just the @ variable 14:13
I can format it with macros in emacs but I'd prefer something that can work live…
m: say @foobar; say @foobar.VAR
camelia 5===SORRY!5=== Error while compiling <tmp>
Variable '@foobar' is not declared
at <tmp>:1
------> 3say 7⏏5@foobar; say @foobar.VAR
AlexDaniel m: my @foobar; say @foobar.VAR
camelia []
AlexDaniel what's the trick to get the variable name? 14:14
Kaiepi m: my @foobr; say @foobar.^name
camelia 5===SORRY!5=== Error while compiling <tmp>
Variable '@foobar' is not declared. Did you mean '@foobr'?
at <tmp>:1
------> 3my @foobr; say 7⏏5@foobar.^name
sena_kun .name
Kaiepi m: my @foobar; say @foobar.name
camelia @foobar
sena_kun m: my @foobar; say @foobar.VAR.name;
camelia @foobar
Kaiepi m: my @foobar; say @foobar.^name
camelia Array
Kaiepi oh right 14:15
AlexDaniel weird I'm getting %oeis-core 14:16
it's like: our %oeis-core is export = ( "groups" => @A000001, ) 14:17
and I want A000001
m: my @A000001; my @A000002; my %oeis-core = ("groups" => @A000001, "kolakoski" => @A000002,); for %oeis-core { say .value.VAR.name } 14:18
camelia %oeis-core
%oeis-core
MasterDuke m: my @A000001; my @A000002; my %oeis-core = ("groups" => @A000001, "kolakoski" => @A000002,); for %oeis-core { say .value.name }
camelia @A000002
@A000001
AlexDaniel pfffffffffft 14:19
14:19 HarmtH_ left
AlexDaniel MasterDuke: thank you :) 14:19
MasterDuke heh
14:19 HarmtH_ joined
thundergnat Ah. too slow. gist.github.com/thundergnat/e2021b...b3c7ccfdd4 14:21
14:21 HarmtH_ is now known as HarmtH
AlexDaniel yeah I think it's just this or something: my %short-names = %oeis-core.map({.value.name.substr(1) => .key}); 14:22
thundergnat alex
14:22 goon_ joined
AlexDaniel yes? :) 14:22
thundergnat AlexDaniel: ^^^ see above gist
Sorry, fatfingers 14:23
14:27 vrurg joined
AlexDaniel thundergnat: github.com/ajs/perl6-Math-Sequences/issues/52 14:27
what do you think?
the “name” has a dot at the end :S
14:28 mniip joined
AlexDaniel thundergnat: github.com/ajs/perl6-Math-Sequences/issues/53 ? 14:28
thundergnat AlexDaniel: Looks good to me. 14:29
MasterDuke wow. the initial compile of Math::Sequences::Integer is slow!! 14:30
AlexDaniel thundergnat: I can also put these keywords as labels. Should I?
thundergnat AlexDaniel: May be useful. Can't see how it would hurt. 14:31
MasterDuke was it this slow for everybody else? `perl6 -I lib/ -e 'use Math::Sequences::Integer'` has been running for about 3 min now...
thundergnat Even after the squashathon, any tickets for sequences left undone will be useful for bookkeeping. 14:32
MasterDuke: Hmm. Compiled in about 25 seconds for me earlier. Sounds like there may be a sequence that needs a lazy. 14:33
Let me try pulling again and compiling
14:33 lancew joined
AlexDaniel MasterDuke: checkout an earlier version and measure again? Then bisect? 14:36
thundergnat MasterDuke: I got a bunch of timeouts in individual sequences on latest master so something is up
MasterDuke took just a couple seconds at 082c440702f1c9db41d75614380ceb0d78143b52 14:39
thundergnat Hmm. I'm getting a timeout in A000793 that wasn't happening earlier. (only the one actualy)
AlexDaniel thundergnat: is there a description for OEIS keywords somewhere? 14:42
thundergnat: yeah, that test is ticketed here: github.com/ajs/perl6-Math-Sequences/issues/48
but nothing should be using that sequence during the compilation so I don't think it's the problem 14:43
MasterDuke huh. now HEAD just takes a couple seconds
AlexDaniel MasterDuke: heh… very precompiled now? :P
MasterDuke: and if you purge your .precomp folder?
MasterDuke i rm -rf'ed lib/.precomp
heh. seems the culprit is my change 14:44
AlexDaniel found it: oeis.org/wiki/Keywords
thundergnat AlexDaniel: Maybe the OEIS glossary? oeis.org/wiki/Index_to_OEIS:_Section_Cor <-- Core sequences entry
Arrgh. to slow again. (and wrong too.)
MasterDuke -our @A000364 is export = 1, &NOSEQ ... *; 14:45
+our @A000364 is export = lazy @A000111[0, 2 ... *];
14:48 lancew left 14:53 Manifest0 left
AlexDaniel thundergnat: it works :) github.com/ajs/perl6-Math-Sequences/issues/55 14:58
thundergnat AlexDaniel: Cool! May not be strictly necessary but cool nonetheless. 15:00
MasterDuke oh wow. `@A000111` is slow! `say @A000111[0]` is instant, but `say @A000111[1]` hasn't finished after a 1.5min
AlexDaniel github.com/ajs/perl6-Math-Sequences/issues 15:02
MasterDuke: ↑ which ones are not implemented yet :)
thundergnat MasterDuke: That sound suspicious. I would it expect it to get slow quickly, but not that quickly.
MasterDuke is it the same for you? 15:03
15:03 Manifest0 joined
AlexDaniel just imagine ajs coming back from vacation or whatever, and seeing what we've done to their repo :D 15:04
thundergnat: look, this one doesn't have a short name: github.com/ajs/perl6-Math-Sequences/issues/77 15:05
maybe add one? I'll take a small break
15:10 Manifest0 left
thundergnat AlexDaniel: Hm. Looks like there is collision with A005588 which is also named binary-rooted-trees. 15:11
MasterDuke an implementation of @A000111 done with multi Entringer subs takes 3s for the first 10 numbers
AlexDaniel thundergnat: very cool that we noticed it :) 15:13
thundergnat AlexDaniel: Seems like A005588 should be named "free-binary-rooted-trees" and A002572 "binary-rooted-trees". I'll make a pull request. 15:15
15:17 Manifest0 joined
AlexDaniel tobs: it's a relatively short script: github.com/ajs/perl6-Math-Sequence...-issues.p6 15:17
tobs: I ♥ Perl 6 :)
squashable6: status 15:18
squashable6 AlexDaniel, 🍕🍕 SQUASHathon is in progress! The end of the event in ≈22 hours. See github.com/rakudo/rakudo/wiki/Mont...Squash-Day
MasterDuke m: my @a = lazy (^Inf); say @a[^5]; my @b = lazy @a[0,2...*]; say @b[^5] 15:19
AlexDaniel Now with an issue tracker! github.com/ajs/perl6-Math-Sequences/issues
camelia (timeout)(0 1 2 3 4)
MasterDuke huh, i thought you could make a lazy list from a lazy list
AlexDaniel MasterDuke: btw you might also like this one: github.com/ajs/perl6-Math-Sequence...69f6c12b8d 15:20
MasterDuke: some explanation: colabti.org/irclogger/irclogger_lo...07-06#l174 15:21
MasterDuke: and then on top of that it causes this: github.com/ajs/perl6-Math-Sequences/issues/47
rfold p6: sub f { my @xs = 1, 2, 3; my $x = 1; [<=] @xs>>.sqrt, $x; }; say f.perl
camelia Bool::False
rfold Amn't I supposed to slip @xs>>.sqrt before using it with [<=]? Like so: [<=] |@xs>>.sqrt, $x 15:22
timotimo if you want the $x to be part of that comparison, you'll need to flatten the @xs>>.sqrt list into it somehow, yeah 15:23
rfold Then why doesn't it crash?
MasterDuke AlexDaniel: can you assign github.com/ajs/perl6-Math-Sequences/issues/100 to me?
rfold p6: say([1, 2, 3] <= 3); 15:24
camelia True
rfold Apparently <= with a list and an integer compares the length of the list. 15:25
MasterDuke thundergnat, AlexDaniel: have you tried `say @A000111[1]`? 15:26
rfold Which happens because it calls .Real on the argument:
p6: say [1, 2, 3].Real;
camelia 3
rfold timotimo: thanks I get it! 15:27
thundergnat MasterDuke: Yes. it IS ridiculously slow, but haven't investigated it yet.
MasterDuke k. i'll submit a PR with an alternate implementation. still slow, but not as much
AlexDaniel MasterDuke: yeah but I wonder why tests are skipping that sequence… 15:28
down to 0 PRs again! :) 15:30
thundergnat AlexDaniel: github.com/ajs/perl6-Math-Sequence...r.pm6#L158 15:31
AlexDaniel ah
so is it broken or just slow?
MasterDuke ha!
AlexDaniel well, I guess both 15:32
15:33 Cabanossi left 15:35 Cabanossi joined 15:37 natrys joined
thundergnat AlexDaniel: github.com/ajs/perl6-Math-Sequences/issues/74 is already done. (That was one of my pull requests while you were still writing the generator script.) 15:44
15:46 Manifest0 left
thundergnat AlexDaniel: Also, assign #66 to me when you have a moment. github.com/ajs/perl6-Math-Sequences/issues/66 Thanks! 15:49
15:51 Manifest0 joined
MasterDuke afk, going to take my daughter to a festival in the nearby park 15:55
AlexDaniel thundergnat: just leave a comment saying you're working on it 16:00
thundergnat: again, a weird limitation of github… can't assign anyone who doesn't have write permissions to the repo
thundergnat AlexDaniel: Ah. Ok. Weird, but not a real problem. 16:01
16:02 dominix__ joined 16:05 dominix_ left 16:12 Manifest0 left 16:17 Manifest0 joined 16:36 Manifest0 left 16:41 Manifest0 joined 16:47 Manifest0 left 16:53 Manifest0 joined 17:03 Manifest0 left 17:08 Manifest0 joined 17:12 vike left 17:14 Joe65 joined 17:21 Manifest0 left
Joe65 Anyone have a good example of p6 script that implements the usual command-line input behavior (stdin by default, one or more files on cmdline, "-", etc)? Do I have to explicitly parse arguments or is there a simpler way? I can't seem to find a complete example for perl6. 17:22
17:23 vike joined 17:26 Manifest0 joined
rfold Joe65: here's a full program that reads stdin by default, otherwise reads files given as arguments, again treating - as stdin: $*ARGFILES.slurp.uc.say; 17:28
17:32 mowcat left
Joe65 Thanks! Let me play `with that a bit. I had trouble using $*ARGFILES when I tried it previously. 17:34
tobs AlexDaniel: yay, I did something useful \o/ 17:36
17:45 mniip left 17:50 mniip joined 17:51 Manifest0 left 17:56 Manifest0 joined
Joe65 Here's an example (I forgot to mention named cmdline args): 18:12
Oops. Best way to paste a few lines of code (I forget)? 18:13
sena_kun any paste service? github gists? 18:14
Joe65 nm thx 18:20
18:20 Joe65 left 18:22 Manifest0 left 18:26 Manifest0 joined 18:34 Joe65 joined
Joe65 ok, sample code here: <script src="gist.github.com/nd3i/80e5da536fa73...script> 18:34
sena_kun Joe65, can you post a link to gist page here? 18:36
Joe65 I'm trying to get named args and standard cmdline input filename(s). Running the script with a named argument, --fmt='%5d' seems to interfere with getting a following file name into $*ARGFILES 18:37
Oops. Sry. I pasted the embed link. Share link: gist.github.com/nd3i/80e5da536fa73...1d32ce7dc7 18:38
tbrowder AlexDaniel: i added another minor pr to math-seq 18:40
sena_kun Joe65, the thing is, MAIN gives you a new signature that overrides the default one 18:41
Joe65 ok 18:44
sena_kun I am not very familliar with MAIN, but playing with it now... 18:46
to get something you might want...
timotimo if you want, you can use *@*ARGS as the last argument to your MAIN 18:47
sena_kun timotimo, hmm, I am doing `$fmt = '%d ', *@*ARGS`, but on `perl6 perl6.pl6 test.lua` @*ARGS is empty 18:48
timotimo that's not where to put the * :) 18:49
sena_kun is confused
ok, made it working... 18:50
timotimo++
gist.github.com/Altai-man/44aee256...a16539c837 <- Joe65 18:51
18:53 robertle joined 18:59 lizmat joined
timotimo i didn't come up with that trick :) 18:59
i think it might have been bdf?
sena_kun Ballroom Dancers' Federation? 19:01
Joe65 ok tnx! Any advantage using $*ARGFILES there in place of $*ARGS? It seems that either works. 19:02
sena_kun Joe65, unrelated to MAIN(which changes things a bit), @*ARGS are just all arguments passed to script, while $*ARGFILES is an object of docs.perl6.org/type/IO::ArgFiles class, constructed from @*ARGS, so they are different beasts. 19:04
but considering you are overloading this processing with a MAIN subroutine, any name can be there, actually, even e.g. @files.
so I'd use *@files in signature and just @files in body. 19:05
I'd even say that my gist is bad because it uses @*ARGS, it is really unrelated here, just any suitable name will be less confuseable for the reader. 19:06
19:13 Black_Ribbon joined 19:15 Manifest0 left
Joe65 ok, i feel like there should be something more concise @files.IO.lines but that seems to give the whole file for each $line. 19:17
AlexDaniel thundergnat: if you say in your commit message “Resolves #123” then #123 will be closed automatically when your PR is merged
Joe65 Thanks for the help. something to start with at least. TYVM
AlexDaniel squashable6: status 19:18
squashable6 AlexDaniel, 🍕🍕 SQUASHathon is in progress! The end of the event in ≈18 hours. See github.com/rakudo/rakudo/wiki/Mont...Squash-Day
AlexDaniel going strong! Still 65 tickets to resolve! github.com/ajs/perl6-Math-Sequences/issues
thundergnat AlexDaniel: Ah. Well that's handy to know. Sorry for not doing that. My git foo is not so great. 19:19
AlexDaniel thundergnat: it's not git foo, it's github foo :)
thundergnat Indeed. :-/
19:20 Joe65 left, Manifest0 joined 19:25 anon joined
AlexDaniel thundergnat: it'd be really cool to get that ticket count to 0 (so that all core sequences have a perl6 implementation) 19:25
anon hello
anyone here?
AlexDaniel anon: hi!
sena_kun anon, hi 19:26
19:26 anon is now known as Guest96979
AlexDaniel thundergnat: by the way now I'm wondering, are any of the sequences already implemented on rosettacode? 19:26
Guest96979 i love scripting and love writing bash scripts the most
i am considering perl
could any1 please highlight whether perl is better for scripting tasks for a system? 19:27
than bash
sena_kun Guest96979, perl or perl 6? they are pretty different languages.
Guest96979 and i am also confused about perl5 or perl6
sorry
is perl6 just improvement on perl5 or complete change in syntax? 19:28
El_Che Bash is a great language for procedural, step-by-step talks
thundergnat AlexDaniel: Yep, several of them.
sena_kun no problem, there is a bit of confusion around this topic. perl(5) is one language, perl 6 is a different one. it is a complete change of a lot of things.
lizmat weekly: perlweeklychallenge.org/blog/meet-...mpion-014/
yoleaux 02:06Z <AlexDaniel> lizmat: maybe this is a good benchmark: colabti.org/irclogger/irclogger_lo...07-06#l154
02:06Z <AlexDaniel> lizmat: it's really slow but I don't think it should be
notable6 lizmat, Noted!
yoleaux 02:16Z <AlexDaniel> lizmat: I looked at the profile but I don't see anything obvious. I think it's an interesting case, unless I'm missing something obvious, it should be just a bunch of iterators glued together in interesting ways. The logic basically skips values at certain indexes (that's it!). Even though it's not fun to have 100 iterators on top of each other, it should still be really really fast.
03:10Z <AlexDaniel> lizmat: here's a more readable version: github.com/ajs/perl6-Math-Sequence...69f6c12b8d
12:23Z <AlexDaniel> lizmat: another weird bug: github.com/ajs/perl6-Math-Sequences/issues/47
El_Che Perl and other programming languages are better suited when you need to transform data or when the logic is more complicated
yoleaux 12:23Z <AlexDaniel> lizmat: (with the same code)
Guest96979 we can use parallel command for running tasks in concurrently..is perl capable of that?
yoleaux 12:45Z <El_Che> lizmat: sungo.wtf/2019/07/05/leaving-irc.perl.org.html
sena_kun Guest96979, it sure is. 19:29
El_Che Guest96979: Perl 6, yes, very nicely. Perl 5 can use forks
AlexDaniel well, piping in bash is much easier
Guest96979 El_Che, is it like awk then?
El_Che bigger, with a bigger standard library 19:30
Guest96979 ok
El_Che more datatypes
Guest96979 sounds interesting..
which one would u recommend to a newbie ? perl5 or perl6
El_Che They are both different languages, so it depends on you usage. As a bash replacement I think perl6 is more fun 19:31
(but you're in a perl 6 channel)
AlexDaniel also bash is not a bad choice for a lot of small tasks
Guest96979 how does piping look like in perl6 though? 19:32
thundergnat AlexDaniel: E.G. A000602 hydrocarbons is on there, but it called Paraffins on Rosettacode rosettacode.org/wiki/Paraffins#Perl_6
Guest96979 is it as simple as chaining |
"|"
AlexDaniel a bit more verbose 19:33
Guest96979 would it require constructing objects and passing them as parameter?'
AlexDaniel Guest96979: see some examples here: docs.perl6.org/type/Proc
sena_kun .oO ( need to write a library with | operator, as Haskell has )
thundergnat AlexDaniel: But it isn't implemented as a lazy list, and I'm not sure how easy it may be to convert it. (You have to specify how many terms you want up front)
El_Che Guest96979: you tend to do a lot of less piping in programming languages in my experience 19:34
you transform the data along the way
and more often than not you don't need to run shell commands
AlexDaniel thundergnat: you can make almost any code lazy with gather/take
Guest96979 i guess then recommended way is to get required output through bash and transform using perl/perl6 and again filter using pipes in bash? 19:35
thundergnat afk for a bit
AlexDaniel thundergnat: actually, it already uses gather take, so why is it not lazy?
oooh because you have to provide N?
Guest96979: why not do the processing in perl 6 ? 19:36
thundergnat: fwiw github.com/ajs/perl6-Math-Sequences/issues/87 is the only remaining sequence that is covered on wikipedia 19:37
Guest96979 ok.if it can get the job done then that's great
but in that doc page, is it really required to construct your command as "run process, arg1, arg2, arg3,..."? 19:38
why not a simple string of entire command?
AlexDaniel Guest96979: because then you're more likely to introduce shell injection bugs 19:42
Guest96979: but yes, you can do "shell 'some cmd | here | no prob'"
Guest96979 ok..thanks m8 19:44
will give it a try
any1 here uses scala here? 19:45
19:46 Guest96979 left 19:50 mowcat joined
TreyHarris Guest43313: I have, but it's been some years... 19:50
Geth ¦ problem-solving: AlexDaniel assigned to jnthn Issue Need a substitute for Perl 5 die with newline for raising end-user errors? github.com/perl6/problem-solving/issues/59 19:51
AlexDaniel TreyHarris: that's a wrong guest :) 19:52
TreyHarris oops 19:53
19:57 pecastro left 19:58 Manifest0 left 20:04 Manifest0 joined 20:06 mowcat left 20:11 Manifest0 left 20:17 |oLa| left, Manifest0 joined 20:22 Manifest0 left 20:26 Manifest0 joined 20:30 Manifest0 left 20:31 kaare_ joined 20:35 Manifest0 joined 20:42 lancew joined 20:45 lizmat_ joined 20:49 lizmat left 20:50 Manifest0 left 20:53 lizmat_ is now known as lizmat 20:56 Manifest0 joined 21:01 lancew left 21:05 Manifest0 left 21:11 Manifest0 joined 21:17 Manifest0 left 21:19 jaldhar left, jaldhar joined 21:23 Manifest0 joined 21:24 |oLa| joined 21:31 Manifest0 left 21:37 Manifest0 joined 21:43 Manifest0 left 21:46 kaare_ left, kaare_ joined 21:49 Manifest0 joined, rfold left 21:52 rindolf left, kaare_ left 22:03 Manifest0 left 22:10 Manifest0 joined 22:17 Manifest0 left 22:19 natrys left 22:22 Manifest0 joined 22:26 Manifest0 left 22:32 Manifest0 joined 22:36 Manifest0 left 22:42 Manifest0 joined, denix13 joined 22:46 denix13 left 22:47 _jrjsmrtn joined 22:48 __jrjsmrtn__ left, Manifest0 left 22:54 Manifest0 joined
AlexDaniel thundergnat: what's @_.clone ? 22:54
thundergnat: btw “Resolves #123” needs to be in the commit message, not in the PR title. I don't think it'll work :) 22:55
thundergnat AlexDaniel: Ah phooey. 22:58
AlexDaniel: @_ is the sequence up to that point. I'm cloning it so I can iterate it several times.
AlexDaniel thundergnat: that sounds weird… 22:59
thundergnat: isn't @ cached anyway?
thundergnat: what if you use @A000311 instead? Can you iterate it then?
thundergnat: or is this the problem: @seq.tail ? 23:00
you have $n, so you don't need .tail, just access it using the index
I just dislike the idea of cloning the whole sequence for every next element…
thundergnat AlexDaniel: you would think so but using @_ directly or the sequence id results in "sequence already iterated" errors.
AlexDaniel m: my @A000311 = 0, 1, { state $n = 1; my @seq = @_.clone; ++$n; -($n-1) * @seq.tail + sum (1 .. ($n - 1)).map: -> $k { ($n choose $k) * @seq[$k] * @seq[$n - $k] } } ... *; say @A000311[^10] 23:01
camelia 5===SORRY!5=== Error while compiling <tmp>
Two terms in a row
at <tmp>:1
------> 3l + sum (1 .. ($n - 1)).map: -> $k { ($n7⏏5 choose $k) * @seq[$k] * @seq[$n - $k] }
expecting any of:
infix
infix stopper
…
AlexDaniel ah sheesh no `choose`
thundergnat: isn't it because of .tail though?
thundergnat: accessing by the index should be just fine 23:02
thundergnat AlexDaniel: possibly... probably
AlexDaniel: Pushing a fix in few seconds 23:14
Kaiepi i'm hesitant to say i'm completely done implementing wide string support on moarvm since i'm wary of how nativecall handles a few things but i'm at the point where i can have a basic implementation working once i finish my work on the rakudo end!
23:15 Manifest0 left
AlexDaniel what do you guys think about github.com/ajs/perl6-Math-Sequences/pull/129 ? 23:16
thundergnat, MasterDuke
thundergnat Looking: 23:18
23:20 Manifest0 joined
thundergnat AlexDaniel: I agree. The only thing I would suggest is to add as many items as OEIS has to the sequence before &NOSEQ marker. (You may have already, I didn't dig that deep) 23:21
AlexDaniel thundergnat: yes, I did 23:22
these sequences tend to be very short though
that's what the PR does :)
OK I'll start merging stuff 23:23
thundergnat AlexDaniel: Ok, I pushed better code for A000311. No more cloning the entire sequence on each iteration. Thanks for the nudge. 23:24
AlexDaniel thundergnat: nice! 23:25
I merged everything
squashable6: status 23:26
squashable6 AlexDaniel, 🍕🍕 SQUASHathon is in progress! The end of the event in ≈14 hours. See github.com/rakudo/rakudo/wiki/Mont...Squash-Day
AlexDaniel thundergnat: the labels were very useful :) 23:27
thundergnat AlexDaniel: Turns out they were quite useful, I look smart in retrospect but I was just spitballing at the time ;-) 23:28
AlexDaniel :)
how many sequences are there in total? 23:29
thundergnat AlexDaniel: I am in moderate awe of how fast you knocked all the scripting together and really made the squashathon move along.
AlexDaniel 184, and only 50 left to implement 23:30
thundergnat: I stole some code from my past self
thundergnat: here, we did it before: github.com/perl6/ecosystem-unbitro...-issues.p6 23:31
thundergnat AlexDaniel: That past self guy is a damn useful fellow to have around!
AlexDaniel thundergnat: last time it was a bit more fun though, because unbitrot repo was in perl6 org and github used to autosubscribe all people in the organization
thundergnat: so we sent like more than 300 notifications, to each person out of 300 people subscribed, if I'm not mistaken 23:32
it was great :) 23:33
ahhh, there's no license in that repo?? 23:35
m: my %A001519; multi sub A001519(0 --> 1) { }; multi sub A001519(1 --> 1) { }; multi sub A001519($n) { %A001519{$n} //= 3 * A001519($n - 1) - A001519($n - 2) }; our @A001519 = lazy 𝕀.map: -> $n { A001519($n) }; say @A001519[^10] 23:39
camelia 5===SORRY!5=== Error while compiling <tmp>
Undeclared routine:
𝕀 used at line 1
thundergnat AlexDaniel: Hm. Indeed there isn't. That puts a little damper on my enthusiasm. Several of his other repos are MIT or Artistic licensed. Hopefully it's just an oversite. 23:40
AlexDaniel m: my %A001519; multi sub A001519(0 --> 1) { }; multi sub A001519(1 --> 1) { }; multi sub A001519($n) { %A001519{$n} //= 3 * A001519($n - 1) - A001519($n - 2) }; our @A001519 = lazy (1…∞).map: -> $n { A001519($n) }; say @A001519[^10]
camelia (1 2 5 13 34 89 233 610 1597 4181)
AlexDaniel thundergnat: yeah, I'm pretty sure it's a mistake
thundergnat: btw where are they, I saw some commits from them today
thundergnat AlexDaniel: shrug. Your guess is as good as mine. 23:41
AlexDaniel m: our @A001519 = 1, 1, -> $n2, $n1 { 3 * $n1 - $n1 }; say @A001519[^10] 23:42
camelia (1 1 -> $n2, $n1 { #`(Block|72919064) ... } (Any) (Any) (Any) (Any) (Any) (Any) (Any))
AlexDaniel m: our @A001519 = 1, 1, -> $n2, $n1 { 3 * $n1 - $n1 } … ∞; say @A001519[^10]
camelia (1 1 2 4 8 16 32 64 128 256)
AlexDaniel m: our @A001519 = 1, 1, -> $n2, $n1 { 3 * $n1 - $n2 } … ∞; say @A001519[^10]
camelia (1 1 2 5 13 34 89 233 610 1597)
23:48 sena_kun left 23:52 Manifest0 left 23:57 Manifest0 joined

Powered by Cro and the Raku® Programming Language.

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

Thank you!