lizmat m: class A { method ^while($) { "foo" } }; dd A.^while # a .while method excluding the endpoint ? 00:00
camelia "foo"
lizmat yes, I know it's doing something else than what you expect :-)
jnthn o.O 00:01
lol? :)
AlexDaniel: Don't know why, but I can reproduce it 00:05
AlexDaniel everything non-ascii does not work, right?
jnthn Seems. Though it works fine with -e
Somebody clueful about this things patched env vars and command line args 00:06
*these things
(That somebody wasn't me)
lizmat m: use MONKEY; augment class Seq { method ^while(Seq:D $self) { "foo" } }; say (^10).map( * + 2 ).^while # jnthn: worth pursuing or not? 00:07
camelia foo
jnthn What, you mean actually using that? :P 00:08
I think the ... operator already provides a nice way to solve the problem, and I'm not so keep on .while and .until as method names 00:09
*so keen
lizmat well, from a language point of view, it would make sense?
ok, I disagree about the ... operator being a nice way to do it 00:10
but I will pursue this no further
jnthn What do you find disagreeable about it, ooc? 00:11
lizmat it's not something that comes naturally
.while and .^while are (to me at least)
AlexDaniel for myself: RT#130363 RT#131925
synopsebot RT#130363 [open]: rt.perl.org/Ticket/Display.html?id=130363 [@LARRY] there's no match for Any aka /./ codegens differently from m/./
RT#131925 [open]: rt.perl.org/Ticket/Display.html?id=131925 [BUG] [REGEX] Smartmatch against `m//` operator mishandles junctions
lizmat m: say ((^10).map( * + 2 ) ... * == 5).map: * * 2 # jnthn: also need another set of parens when chaining 00:17
camelia (4 6 8 10) 00:18
lizmat goes to bed
jnthn An argument for a method form would be that we could implemnet it on Supply also 00:21
And also potentially hyper and race but the semantics there would be very interesting :)
I'm more uncomfortable with the names while and until, though I can't put my finger on why. TimToady maybe has some thoughts on that. 00:22
And abusing .^ is not going to fly :)
wander dogbert17: yes you can fix it temporarily in that way, but our purpose is showing the link properly. The main problem is our `htmlify.p6`(and somewhere module Pod::To::HTML) be too work-only. 01:27
`htmlify.p6` can be modified directly without hesitation, while the module's design is so ugly that we even need change its interface, painful. 01:34
IMO Pod::To::HTML should only produce the <body>, and provide a dynamic wrapper variable if user wants customization. 01:38
oops, i removed my try to build p2h. i will bring this topic up again with codes :P 01:43
b2gills m: gist.github.com/b2gills/b877658f38...dc0990f581 # lizmat I was talking about .accept-while and .accept-until back in Jan 02:09
camelia 5===SORRY!5=== Error while compiling <tmp>
Confused
at <tmp>:1
------> 3https:7⏏5//gist.github.com/b2gills/b877658f38dae5
expecting any of:
colon pair
Zoffix How do you even notice that... 02:57
AlexDaniel I don't know, but it's really infuriating :) 02:59
Zoffix: there's another page with big “Concise” title on it, and there's a “Try this out!” block. This particular green text color does not work very well on that red-ish background 03:01
I don't know why it is though…
Zoffix I'll try to remember these changes the next time I'm working on it 03:02
You can also open an Issue on the repo 03:03
(don't wanna do anything today; spent the whole day running errands and got an hour to relax before sleep)_ 03:04
AlexDaniel GODDAMN LOGGER
alright I'll just inline it…
done 03:07
Zoffix Thanks.
AlexDaniel Zoffix: fwiw I like it a lot
Zoffix Thanks. 03:08
AlexDaniel another fwiw: stock images do create a strange feeling, but I don't think I care about it enough. It's ok 03:09
Zoffix ¯\_(ツ)_/¯ maybe years of staring at them twisted my mind, but I find sexy people smiling, looking at content, or pointing to it, make the piece more personable and friendlier 03:11
I have one woman at work who hates stock models and always asks me to change them to stock of flowers... Maybe you two should meet :P 03:13
AlexDaniel ( three? :) ) 03:14
Zoffix three? three what? 03:15
AlexDaniel well, I thought stmuk didn't like the stock models too?
Zoffix Ah :) I should start a list :) 03:16
AlexDaniel yeah! Let's make a party!
japhb Zoffix: Note that you say "sexy people smiling". This indicates that they are window dressing, there for their attractiveness, not because they are essential. 03:18
(Essential would be for example if they are demonstrating how something is done, but for programming that's not as likely to be valuable.) 03:19
Zoffix japhb: sure. That's what stock photos are for.
japhb Some people find that use of people in marketing material off-putting, or even offensive. Others like it. 03:20
I'm saying that someone who always asks you to change out the stock (human) models for something else may have an actual narrative in their head about why they don't like that -- and some portion of the audience is likely to share that narrative. 03:21
Zoffix Some people like beef. Others like pork.
japhb I guess it depends on whether your audience is vegetarian, kosher, or omnivore then. And whether you are wanting to appeal one or both of the former groups. 03:23
*to appeal to
Zoffix I'm not removing stock pictures just because you think using pictures of people as decoration is offensive. 03:24
japhb I didn't ask you to. I'm offering you more viewpoints. 03:25
(You'll often find that I don't argue a point directly, but merely explain why there is a reasonable argument to be made.)
Zoffix Well, make your own fucking brochure then. 03:26
japhb *I'm not trying to pick a fight*
I'm not trying to pick on your work either.
It just seemed like you didn't understand where others were coming from, and I was trying to explain it 03:27
FWIW, I'm not about to pick on someone for putting out honest effort to do good. 03:28
Some things work perfectly, some don't, but I certainly don't want you to feel like I'm saying your work isn't appreciated.
japhb realizes he neglected to ask if Zoffix was even interested in his help before starting 03:31
Sorry about that, Zoffix.
MasterDuke this is nice, seems like firefox 57 loads rakudo profiles noticeably faster 03:32
Geth rakudo: 5a29a0ce41 | (Elizabeth Mattijsen)++ | src/core/Any-iterable-methods.pm
Make .head and .tail only take WhateverCode

Instead of Callable. Being able to take a Callable is not documented at the moment, and restricting it to WhateverCode does not break any tests. Should allow for more uses of Callable with .head/.tail.		 08:03
tbrowder .tell timotimo mind talking table pod cell formatting sometime? 12:09
yoleaux tbrowder: I'll pass your message to timotimo.
tbrowder .tell timotimo see gist gist.github.com/tbrowder/d8a48ae70...f0c0c4192e 12:29
yoleaux tbrowder: I'll pass your message to timotimo.
timotimo wasn't there some commits that threw out twines in favor of just requiring the user to typecheck individual elements? 12:30
yoleaux 12:09Z <tbrowder> timotimo: mind talking table pod cell formatting sometime?
12:29Z <tbrowder> timotimo: see gist gist.github.com/tbrowder/d8a48ae70...f0c0c4192e
timotimo because a whole lot of empty strings kept ballooning the memory usage or something?
tbrowder i don't know anything about that, but if that's the case, it needs to be in the docs IMHO. doesn't seem to be an elegant way to handle it (i.e., LTA). 12:31
timotimo if you're already traversing a pod tree with multi subs and such, you're already typechecking stuff implicitly 12:32
tbrowder another way to handle it is change to structure of the table class to have an addition data member which will be empty unless there is a twine to be used instead of the string 12:33
for each cell maybe... 12:35
timotimo i'm not sure what that will solve
tbrowder as far as the user type-checking, there are notes inside Pod::To::* griping about cell formatting 12:36
anyway, any past problem with memory and empty strings i'll bet is not as much a problem now with a more mature code base. i'll be glad to work on this if it's not shot down. 12:43
BTW, this would address Rakudo issue #1244 12:50
also maybe RT #126966 and others: #126654, etc. 12:58
synopsebot RT#126966 [open]: rt.perl.org/Ticket/Display.html?id=126966 [POD] empty strings in .contents with single formatting code 12:59
liztormato jnthn: I was thinking of using .head( { condition } ) instead of .while 14:03
As long as the block returns true, it will
pass on values. And end on first False 14:04
Vice versa, .tail would skip while False, and start passing on 14:05
on first True
And possibly .skip could 14:06
be considered the opposite of .grep 14:07
stmuk ZofBot: I was joking about the stock models 14:23
ZofBot stmuk, I called after him that it was only Fred and me, but at that moment a gun went off with a tremendous bang, so he couldn't have heard me
stmuk damm tab complete 14:24
tbrowder timotimo: lizmat would like someone more familiar with Pod.nqp to accept my PR #1240, would you mind considering it? 14:43
(or deny, of course)
*accept (or deny) 14:44
to pass spectests you also have to use my roast PR #353 14:45
Geth nqp: 98f6e54c02 | pmurias++ | 2 files
[js] Implement some uniprop matching
 14:46
nqp: 626e60c81f | pmurias++ | src/vm/js/RegexCompiler.nqp
[js] Remove captures from cstack when backtracking a capture
pmurias bartolin: found a new bug bug where the fix can be copied over to the JVM: github.com/rakudo/rakudo/issues/1251
Geth nqp: af3b775a7c | pmurias++ | src/vm/js/RegexCompiler.nqp
[js] Implement zerowidth anchor type
 15:46
AlexDaniel releasable6: status 16:01
releasable6 AlexDaniel, Next release in 2 days and ≈2 hours. No blockers. 109 out of 205 commits logged
AlexDaniel, Details: gist.github.com/997ac6ce10947260fe...df1fb4c589
Geth rakudo: 028b7d22ca | (Moritz Lenz)++ | tools/commit-stats.pl
Remove outdated tool (issue #1252)
 16:02
synopsebot RAKUDO#1252 [open]: github.com/rakudo/rakudo/issues/1252 [meta] Rakudo repo cleanup (mainly tools/ directory)
AlexDaniel \o/ 16:03
Geth rakudo: 5baede5915 | (Moritz Lenz)++ | tools/progress-graph.pl
Remove outdated tool (issue #1252)
rakudo: 825a8b0d09 | (Moritz Lenz)++ | tools/available-pm-names.pl6
No need for available-pm-names.pl6

we do not name our releases after PM groups. Issue #1252		 16:14
synopsebot RAKUDO#1252 [open]: github.com/rakudo/rakudo/issues/1252 [meta] Rakudo repo cleanup (mainly tools/ directory)
|Tux| did I post a timing today? 16:40
japhb |Tux|: Not that I've seen. 16:45
|Tux| Rakudo version 2017.10-208-g825a8b0d0 - MoarVM version 2017.10-80-ga0dc2747c
csv-ip5xs1.075 - 1.106
csv-ip5xs-2012.606 - 12.852
csv-parser12.063 - 12.099
csv-test-xs-200.435 - 0.440
test11.801 - 12.126
test-t3.175 - 3.238
test-t --race1.235 - 1.331
test-t-2058.480 - 59.791
test-t-20 --race19.480 - 20.186
 16:52
japhb buggable: speed :4
buggable japhb, ▃ ▂ ▃ ▅ █ ▃↑▃ ▃ ▅ █ ↑ dates: 2017-10-26–2017-11-16
japhb, █ ▂ █ ▁█ ▆█ █▇ ▆███ █▅ ▇ █ █▁▄█▅▂ range: 19.341s–23.185s
japhb, ▆█▄█ █▆▆▅▇██ ██▄████████▇ ██▆█▇█ ██████▇▃▂ speed: 1% slower (widths: 9/3)
japhb, ▁▂████▅▇▇███████▆▇████████████▄██████▃█████████▂
japhb More noise than anything else, it looks like. 16:53
ilmari buggable: speed :4 test-t 16:58
buggable ilmari, ▅ ↑ ↑█ ↑ ▂ dates: 2017-10-26–2017-11-16
ilmari, █ ▄ ▂▄ ▂█ ▄▂▃▁██ █ █ ▇ ▂ range: 3.067s–3.701s
ilmari, ▂ ▂ █ █▄▅▃▅██▅███▇██████▅██▃▃▆▆▅█▆▂██▅▇▇█▂▄ ▅ speed: 4% slower (widths: 9/3)
ilmari, ▃▃▁▃▇█▅█▃█▇███████████████████████████████████████
sjn thinks zoffix's buggable bot would be a little less noisy if it's output wasn't as wide as it is now.
ilmari sjn: here's a penny, get yourself a wider terminal ;-P
|Tux| or a smaller font 16:59
ilmari fits four 120x38 terminals with a comfortable font size on this 23" monitor
sjn ideal output width would be max 78-25-$nick.chars :)
|Tux| thinks ideal is in the eyes of the beholder 17:00
ilmari you have 25 characters of cruft on each line? 17:01
sjn that sounds painful
ugexe just turn the graph 90 degrees :P
ilmari m: "16:58 <+buggable> ".length
camelia No such method 'length' for invocant of type 'Str'. Did you mean any of these?
chars
codes

in block <unit> at <tmp> line 1
ilmari m: "16:58 <+buggable> ".chars
camelia ( no output )
ilmari m: "16:58 <+buggable> ".chars.say
camelia 18
ilmari okay, if you have second-precision timestamps that does add up 17:02
|Tux| ----+----1----+----2----+----3----+----4----+----5----+----6----+----7----+----8----+----9----+----0----+----1----+----2----+----3----+----4----+----5----+----6----+----7----+----8----+----9
I see that my text area is ± 136 characters wide
sjn nah, I actually have 18 chars, like you :)
|Tux| m: "16-11 17:14| synopsebot|".chars.say 17:04
camelia 24
|Tux| I win
sjn see there, I knew 25 was a reasonable expectation :) 17:05
|Tux| => *$HOME 17:07
japhb
.oO( $person ==> &commute ==> $person.home )
 17:47
[Coke] uses vi & 80x25, as required by the Michigan Terminal System 17:49
Geth nqp: c918b9d8e0 | pmurias++ | 2 files
[js] Implement some negated uniprops
 19:32
nqp: 8292e8b721 | pmurias++ | t/nqp/031-grammar.t
Test a uniprop in a regex

Powered by Cro and the Raku® Programming Language.

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

Thank you!