timotimo OK 00:00
travis-ci Rakudo build passed. Zoffix Znet 'Fix .= initialization of attributes for types w/ `::` in name 01:12
travis-ci.org/rakudo/rakudo/builds/194639582 github.com/rakudo/rakudo/compare/6...0a07747a39
b2gills I implemented Haskell's ｢takeWhile｣, and I was wondering where to put it ( I used the name ｢accept-while｣ because it seems more Perlish ) 03:02
m: gist.github.com/b2gills/b877658f38...dc0990f581
camelia rakudo-moar 700a07: OUTPUT«ok 1 - ␤ok 2 - ␤ok 3 - ␤ok 4 - ␤ok 5 - ␤ok 6 - ␤ok 7 - ␤ok 8 - ␤ok 9 - ␤ok 10 - ␤»
TimToady b2gills: maybe you could call it 'scan', to go with 'skip', and really confuse the Haskellers :) 03:59
'scan' being one of the worse-named Haskell features... 04:00
b2gills The main thing is I didn't want to go with take-while, as someone may think that it goes with gather
TimToady I think I first saw skip/scan in Vax assembly language :) 04:01
b2gills I have a copy of The Second Book of Machine Language, and a 8085 users manual 04:03
TimToady and it seems a bit anti-social to make the positive scanner so much longer than the negative...
also, accept has some unfortunate socket overloadings 04:05
b2gills I tried it with adding prefix:<!>, and just using `-> |args { not condition |args }`
I'm not sure which would be the better of the three as far as speed/memory are concerned 04:06
TimToady though I guess our skip just takes a number of elements to discard 04:09
b2gills perhaps obtain or acquire 04:10
TimToady well, head is the positive counterpart of skip, I guess
m: say (1..100).head(10) 04:11
camelia rakudo-moar 700a07: OUTPUT«(1 2 3 4 5 6 7 8 9 10)␤»
TimToady m: say (1..100).skip(10)
camelia rakudo-moar 700a07: OUTPUT«(11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 …»
TimToady is gonna have to learn this language one of these days... 04:12
b2gills I like discard-while for dropWhile
TimToady what are you doing about the necessity of 1-element lookahead? 04:13
and is that gonna impact laziness/caching? 04:14
b2gills I forgot to re-add .lazy 04:15
TimToady that's what .lazy-if is for :)
b2gills I meant `method is-lazy() { $!source.is-lazy }` 04:17
antonyms of reject: retain admit embrace uphold keep save choose 04:20
TimToady well, allow is often the opposite of reject 04:30
'course we could go with head-while and skip-while, just for consistency, but that's a little odd 04:31
b2gills Technically since head and skip currently only know about numberish values, we could just add a Callable candidate ... or not 04:45
TimToady maybe even with a :while named arg 04:48
MasterDuke if anyone is interested, here's a perf report of building rakudo with target=parse. gist.github.com/MasterDuke17/aff44...6d1d0ba537 05:02
samcv would be nice is we could rewrite some of the nqp tests to use the new TAP stuff i heard pmurias added 05:09
b2gills .oO( wish I had taken a break 3 hours ago, would have liked to see todays episode of Lucifer ) 05:12
samcv looks like a lot of them do, gonna ignore the other ones that i'm getting hangs on 05:16
|Tux| This is Rakudo version 2017.01-34-g700a07747 built on MoarVM version 2017.01-10-g76872dbe 08:29
csv-ip5xs 2.883
test 12.521
test-t 4.993
csv-parser 13.719
lizmat Files=1166, Tests=56206, 191 wallclock secs (11.42 usr 4.52 sys + 1156.13 cusr 112.10 csys = 1284.17 CPU) 08:33
morning, #perl6-dev 08:55
lizmat wonders where timotimo added the redo support for whenever 08:59
Geth kudo/nom: f6531d366f | (Elizabeth Mattijsen)++ | src/core/Any-iterable-methods.pm
Allow Any.batch(:elems)

To make it more in line with the reactive Supply.batch(:elems) counterpart.		 09:09
kudo/nom: 56947275c5 | (Elizabeth Mattijsen)++ | src/core/Supply.pm
Add Supply.rotor($batch) candidate

To prevent confusion with Any.rotor($batch) candidate.		 09:14
lizmat afk for a few hours& 09:34
timotimo lizmat: it lives in/near the add-whenever-block thing
nine -win 10 09:54
brokenchicken RT's weird. RT#130629 doesn't have lizmat's response that I see on mailing list but does have smsl's response, which isn't on mailing list :/ 10:30
synopsebot6 Link: rt.perl.org/rt3//Public/Bug/Displa...?id=130629
timotimo ugh :(
brokenchicken m: dd (Empty xx 12) 10:45
camelia rakudo-moar 569472: OUTPUT«(Empty, Empty, Empty, Empty, Empty, Empty, Empty, Empty, Empty, Empty, Empty, Empty)␤»
brokenchicken m: dd (slip() xx 12)
camelia rakudo-moar 569472: OUTPUT«()␤»
brokenchicken m: dd slip() === Empty
camelia rakudo-moar 569472: OUTPUT«Bool::True␤»
lizmat m: dd slip() =:= Empty 11:41
camelia rakudo-moar 569472: OUTPUT«Bool::True␤»
lizmat wow 11:42
hmmm
Empty is defined as a constant, maybe that has something to do with that
brokenchicken Yeah, one version goes through the thunked version and the other doesn't 11:44
Not sure if it's really a problem though. If I literally typed (Empty xx 12) I'd want a bunch of empties and not nothing
m: my $e = Empty; dd ($e xx 12) 11:45
camelia rakudo-moar 569472: OUTPUT«()␤»
brokenchicken m: constant z = Empty; dd (z xx 12) 11:46
camelia rakudo-moar 569472: OUTPUT«(Empty, Empty, Empty, Empty, Empty, Empty, Empty, Empty, Empty, Empty, Empty, Empty)␤»
brokenchicken hmmm
lizmat timotimo: are you sure you pushed it? your last commit is github.com/rakudo/rakudo/commit/f9...fc8427bfce 11:48
which only implements next ?
timotimo it's in a branch 11:49
i was wondering why you said it was already in
github.com/rakudo/rakudo/pull/998
lizmat aahh :-) 11:53
timotimo: I said it because it was in the P6W :-)
timotimo oh, well, the p6w is lying, clearly! ;) 11:54
lizmat not anymore :-) 11:55
timotimo that's good 11:56
lizmat timotimo: so what happens if a redo happens, and it's in $redoable, and then a next is done? 12:04
timotimo well, hopefully that works. i didn't write a test for that combination
lizmat I'm trying to see if I can apply the same trick to Any-iterable and speed up all map/fors 12:05
timotimo m: react { whenever Supply.from-list(^10) { say "doing $_"; redo unless $++ %% 2; next unless $++ %% 3 } }
camelia rakudo-moar 569472: OUTPUT«doing 0␤doing 1␤===SORRY!===␤redo without loop construct␤»
timotimo oh, it's not merged, of course
right, that'd be cool if that could work
lizmat $ 6 'react { whenever Supply.from-list(^10) { print "$_ "; last } }' 12:28
0 1 2 3 4 5 6 7 8 9
timotimo: is that what we expect? I would think it would only say "0" ? 12:29
m: react { whenever Supply.from-list(^10) { print "$_ "; last } }
camelia rakudo-moar 569472: OUTPUT«0 ===SORRY!===␤last without loop construct␤»
lizmat ok, so it would be an improvement
timotimo: added some comments 12:33
jnthn Looks like it's not untapping the supply or something 12:36
lizmat jnthn: so you agree it should just say "0" with the last ? 12:37
jnthn yes
timotimo good thing i pr'd and marked "needs review" :) 13:12
brokenchicken likes to write tests first for stuff like that 13:26
Geth kudo/nom: a2fbe20bdc | (Elizabeth Mattijsen)++ | 3 files
Reorganize Set/SetHash/Setty a bit

The base functionality is the .iterator method. By making sure Set and SetHash have their own iterator method, we can more .pairs to Setty. .antipairs could be moved to Setty anyway, as there is no difference between Set and SetHash for that. Also make sure that SetHash.values|pairs|kv have their own iterator.
Final result: SetHash.iterator|pairs|antipairs|kv|values 20x faster		 13:36
kudo/whenever_last_redo: 48 commits pushed by lizmat++
review: github.com/rakudo/rakudo/compare/c...334b8b2176
lizmat oops, that last bit wasn't intended :-)
but guess doesn't hurt ?
timotimo wow, those are all the commits to nom that happened since i started that branch?! 13:43
that's crazy i didn't realize i started it so far back
lizmat apparently
timotimo feels like it was just the day before yesterday
brokenchicken well, almost, it's two days before yesterday :) 13:44
timotimo maybe i accidentally started the branch from a very old commit 13:45
brokenchicken No, there have been that many commits in the past 4 days. 13:46
NeuralAnomaly: status 13:56
NeuralAnomaly brokenchicken, [✘] Next release will be in 3 weeks and 4 days. Since last release, there are 10 new still-open tickets (10 unreviewed and 0 blockers) and 41 unreviewed commits. See perl6.fail/release/stats for details
brokenchicken NeuralAnomaly: status 13:57
NeuralAnomaly brokenchicken, [✘] Next release will be in 3 weeks and 4 days. Since last release, there are 10 new still-open tickets (10 unreviewed and 0 blockers) and 33 unreviewed commits. See perl6.fail/release/stats for details
brokenchicken And 33 of them were since release
timotimo wow. 14:00
lizmat yeah, I had a bit of backlog that I didn't want to commit before the release :-) 14:02
timotimo mhm 14:03
mst wins 14:38
brokenchicken: hrm, benchable6 and friends are on a different host to the others? 14:41
brokenchicken yeah
That's AlexDaniel's server
Geth kudo/nom: e995107538 | (Elizabeth Mattijsen)++ | src/core/Baggy.pm
Make Baggy.kv about 15% faster

This was actually more about getting the .kv in line with all of the other implementations of .kv		 14:43
mst AlexDaniel: any chance you can get a consistent ident for your bots so I can +v them with a single mask? 14:45
AlexDaniel mst: well, ideally they should be identified as “whateverable”, but only 5 simultaneous logins are allowed on freenode 14:47
if that is what you're asking
so what should I do? 14:48
mst eh?
AlexDaniel: what-bug, what-com, etc. (for buggable6, commitable6) would work nicely 14:49
just like zoffix's bots all match *!*zofbot-*@perl6.party etc. 14:50
AlexDaniel mst: like this? 15:04
b2gills m: say <include exclude> X~ '-' X~ <while until> # TimToady how about include/exclude 15:07
camelia rakudo-moar e99510: OUTPUT«(include-while include-until exclude-while exclude-until)␤»
AlexDaniel I thought that you can do it by nick and just use “whateverable” for that, but okay
mst blam 15:08
# access #perl6-dev add *!~*-able@2001:41d0:2:5eb5:: +V 15:09
brokenchicken those are pretty long names 15:10
Fitting Perl 6 code into 80-col width is already somewhat a game due to scopes being added quite fast 15:11
mst AlexDaniel: I can, but then I'd have to add access entries for every bot manually, and since you don't bother telling me when you add a bot, what we had was 'half a set of manual entries and some missing ones'
now I've cleaned everything up and consolidated down to three entries for each channel, two for zoffix's bots and one for yours
and now neither of you need to (ahem) bug me to get configuration for a new one
*and* we've freed up a bunch of access list slots, which are not infinite 15:12
I mean, sure, I *can* "do it by nick", but I don't like copy-paste administration any more than I like copy-paste coding :P
AlexDaniel it doesn't work when you're identified for one nick but use another? 15:13
mst what?
AlexDaniel ok-ok nevermind, I don't want you to scream at me again…
mst oh, you have an *account* for them that they all log into? 15:14
AlexDaniel yes
mst yeah, had you told me that, I could've done it that way instead 15:15
b2gills brokenchicken: to be fair it is *only* 1½ times longer than Haskell's takeWhile and dropWhile
AlexDaniel no not really, what you did is right
as I've said, it's limited to 5 simultaneous logins only
and I'm already at 6
mst ah, *that's* what you were talking about
we were discussing nicks and idents and hostnames at the time
there's all sorts of limits for various shit, so I didn't catch what you meant 15:16
the trouble with me having spent a long time running IRC stuff is that I use the language precisely, and then when somebody who's not as experienced uses terms that appear to them to be about right, I sometimes completely fail to map it across
(that is -not- meant to be a criticism, except perhaps of me)
brokenchicken b2gills: what are these methods for again? 15:18
Geth kudo/nom: ad908065e3 | (Elizabeth Mattijsen)++ | src/core/Rakudo/Iterator.pm
Introducing R:It.Mappy-kv-from-pairs

Abstract handling of .kv logic on hashes in which the actual key/value
  are stored in a Pair of the value (such as object hashes and baggies)
kudo/nom: 4ef37cf9a5 | (Elizabeth Mattijsen)++ | 2 files
Use new R:It.Mappy-kv-from-pairs where possible
 15:19
mst AlexDaniel: anyway, thanks for poking things, and shout if there's anything else like this that isn't doing the right thing
AlexDaniel mst: what's the idea behind voicing the bots? 15:20
mst it's fairly traditional, and you'll now find that if you do '/names' (or /n on most clients) the bots are listed at the top
makes it easier to find 'em
AlexDaniel mst: for example, there's dalek, synopsebot6 and ilbot3
and even camelia 15:21
they don't seem to be voiced
mst nobody asked yet
synopsebot6 is on the same host as the zofbots, so that could be sorted just by changing ident, same as dalek
and camelia I can special case 15:22
seems camelia was setup for perl6 but not here
b2gills brokenchicken: ｢(^20).include-while(* !== 5).List eqv (^5).List｣ and ｢(^20).exclude-while(* !== 5).List eqv (5..^20).List｣ 15:23
AlexDaniel shadowpaste is also a bot on #perl6 15:24
mst brokenchicken: is synopsebot6 yours? whose is dalek?
brokenchicken nope 15:26
mst mm
brokenchicken I think they're "community managed"
yeah, both dalek and synopse are on hack 15:30
TIL synopsebot is a p6 bot :o 15:33
oh, but it doesn't use IRC::Client
mst ah, I see you already did dalek, excellent 15:34
brokenchicken and done
[Coke] Do we even want dalek here anymore now that we have geth? 15:37
AlexDaniel for now, yes
some repos are still using it 15:38
.tell dalek Please switch this bot to Geth
yoleaux2 AlexDaniel: I'll pass your message to dalek.
travis-ci Rakudo build errored. Elizabeth Mattijsen 'Make Baggy.kv about 15% faster 17:07
travis-ci.org/rakudo/rakudo/builds/194838757 github.com/rakudo/rakudo/compare/a...9510753809
buggable [travis build above] ✓ All failures are due to timeout (1), missing build log (0), or GitHub connectivity (0).
Geth ast: d00fc0f99e | (Zoffix Znet)++ | README.md
Mention fudge --keep-exit-code option
 19:40
ast: 18ca2c1273 | (Zoffix Znet)++ | README.md
Fix typo
 19:43
p: MasterDuke17++ created pull request #343:
Support Nd's in sprintf format strings		 19:54
brokenchicken MasterDuke: did TimToady say we didn't want this stuff on fastpath? I wasn't really paying attention but I recall something said on this 19:56
Also, every time I see nqp::radix() I'm reminded of Zefram and the issue of cut off numbers 19:58
m: printf "%033", 42
camelia rakudo-moar 4ef37c: OUTPUT«Directive 033 is not valid in sprintf format sequence %033␤␤»
brokenchicken m: printf "%033d\", 42
camelia rakudo-moar 4ef37c: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Unable to parse expression in double quotes; couldn't find final '"' ␤at <tmp>:1␤------> 3printf "%033d\", 427⏏5<EOL>␤ expecting any of:␤ postfix␤»
brokenchicken ugh
m: printf "%033d", 42
camelia rakudo-moar 4ef37c: OUTPUT«000000000000000000000000000000042»
brokenchicken m: printf "%03\x[308]3d", 42
camelia rakudo-moar 4ef37c: OUTPUT«Directive d not applicable for type Int␤␤»
brokenchicken that's a crappy error but I'd guess the new version will silently succeed, using effectively "%03d" 20:02
samcv i would like to see a benchmark of new vs old speed tbh 20:16
wouldn't making ascii numbers fast in radix be the best way to fix any potential performance issues? 20:17
Geth p: 518a38ac87 | MasterDuke17++ | src/HLL/sprintf.nqp
Support Nd's in sprintf format strings
 20:23
p: c22e9c65cb | lizmat++ | src/HLL/sprintf.nqp
Merge pull request #343 from MasterDuke17/support_Nds_in_sprintf

Support Nd's in sprintf format strings
samcv u: trillion 20:25
unicodable6 samcv, U+16B61 PAHAWH HMONG NUMBER TRILLIONS [No] (𖭡)
samcv m: say 𖭡
camelia rakudo-moar 4ef37c: OUTPUT«1000000000000␤»
samcv m: 0x1CD1.uniprop('sc').say 20:31
camelia rakudo-moar 4ef37c: OUTPUT«Inherited␤»
moritz m: say 0x16B61.unival
camelia rakudo-moar 4ef37c: OUTPUT«1000000000000␤»
samcv m: 0x0363.uniprop('sc').say 20:32
camelia rakudo-moar 4ef37c: OUTPUT«Inherited␤»
Geth kudo/nom: 736ef1ab45 | (Elizabeth Mattijsen)++ | 4 files
Create specific skip-one's for "difficult" iterators

Specifically the consumers of Rakudo::Iterator::Mappy that cannot have a "simple" skip-one.		 20:41
kudo/nom: 54f647d127 | (Elizabeth Mattijsen)++ | src/core/Rakudo/Iterator.pm
Add a generic skip-one to R:It:Mappy

This makes all uses of .skip on iterators based on this role about 15% faster.
raiph m: role A { method m {} }; role B does A {}; class C does A does B {} # curious message; at least LTA 20:48
camelia rakudo-moar 4ef37c: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Method 'm' must be resolved by class C because it exists in multiple roles (B, A)␤at <tmp>:1␤»
samcv i've been thinking about implementing the script extensions property
and how to store the data in MVM. maybe have a bitmap type thing where 1 = Deva, 2=Dupl, 4=Greek etc, and then just add up the total 20:49
and then we can fast path checking if that property includes the requested name
which is the most common case of using it 20:50
it's the only property I know of which holds multiple values
raiph ^^^ my `m: role ...` line above is a golf of bdfoy's recent SO Q 20:52
hi samcv, i'm curious what "script extensions" are and who uses them? 20:54
(looking right now) 20:55
samcv for example
0951 ; Beng Deva Gran Gujr Guru Knda Latn Mlym Orya Shrd Taml Telu # Mn DEVANAGARI STRESS SIGN UDATTA
this one character is used with all these scripts
so if you wanted to do /<:Deva>/ to match it, it would currently not work
m: 0x0951.uniprop('sc').say
camelia rakudo-moar 54f647: OUTPUT«Inherited␤»
samcv imo we should check script and script extension when doing <:scriptname> 20:56
but script values can only have one value
thouhg you could still do <:Script<Deva>> and force it to only check script property if for some reason you want that 20:57
raiph i now feel up to speed :) 20:58
samcv cool 20:59
m: 0x32C0.uniprop('sc').say
camelia rakudo-moar 54f647: OUTPUT«Common␤»
samcv so that is a Hani symbol that means month january
if you did <:Hani> it wouldn't match, but the user probably would be suprised by why it did not match
raiph does the script extensions list include a dupe of the original script? 21:00
samcv no
that's why it's called script extensions :P
an addition to
raiph so what i'd perlishly expect is 21:01
1. <:Deva> matched either script or any entry in extensions
samcv what comes after 1? 21:02
raiph 2. <:Script<Deva>> matched Deva in script
samcv scx is the short name of script extensions btw 21:03
yeah that's what I feel too raiph
raiph 3. <:SCX<Deva Beng>> matched extension containing Deva and Beng
or something like that
samcv and? 21:04
raiph i think that's it
samcv not or?
Geth kudo/nom: 5db0b2964b | (Elizabeth Mattijsen)++ | src/core/Baggy.pm
Make Baggy.kxxv about 30% faster

By rewriting using nqp:: ops
samcv well I guess you can do <:scx<Deva> + :scx<Beng>> i think that's the syntax
or something, haven't done it in a bit, would have to check my module 21:05
idk why the hell you'd want to match multiple script extension values though... really a corner case
so I don't think we should focus too much on that. can do <:scx<Deva>>|<:scx<Beng>> or something 21:06
raiph yeah the `and` rather than `or` was not based on RL needs
samcv i mean 21:07
<:Script<Beng Devi>> i would think that would match both of them
since only scx has multiple values
raiph it was just an arbitrary pick so let's pretend I didn't suggest that bit :)
samcv kk
idk maybe wa can already do this, i've never tried to put multiple things there
m: say 'a' /<:Script<Arabic Latin>>l 21:08
camelia rakudo-moar 54f647: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Two terms in a row␤at <tmp>:1␤------> 3say 'a' /<:Script<Arabic Latin>>7⏏5l␤ expecting any of:␤ infix␤ infix stopper␤ postfix␤ statement end␤ state…»
samcv m: say 'a' /<:Script<Arabic Latin>>
camelia rakudo-moar 54f647: OUTPUT«Value of type List uselessly passed to val()␤ in block <unit> at <tmp> line 1␤Cannot convert string to number: base-10 number must begin with valid digits or '.' in '3⏏5a' (indicated by ⏏)␤ in block <unit> at <tmp> line 1␤␤Actually throw…»
samcv m: say 'a' /<:Script<Arabic>>
camelia rakudo-moar 54f647: OUTPUT«Cannot convert string to number: base-10 number must begin with valid digits or '.' in '3⏏5a' (indicated by ⏏)␤ in block <unit> at <tmp> line 1␤␤Actually thrown at:␤ in block <unit> at <tmp> line 1␤␤»
samcv m: say 'a' /<:Script<Arabic>>/
camelia rakudo-moar 54f647: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Missing required term after infix␤at <tmp>:1␤------> 3say 'a' /<:Script<Arabic>>/7⏏5<EOL>␤ expecting any of:␤ prefix␤ term␤»
samcv m: say 'a' ~~ /<:Script<Arabic>>/
camelia rakudo-moar 54f647: OUTPUT«Nil␤»
samcv hits self on head
m: say 'a' ~~ /<:Script<Arabic Latin>>/
camelia rakudo-moar 54f647: OUTPUT«Nil␤»
raiph heh
samcv m: say 'a' ~~ /<:Script<Latin Arabic>>/ 21:09
camelia rakudo-moar 54f647: OUTPUT«Nil␤»
samcv m: say 'a' ~~ /<:sc<Latin>>/
camelia rakudo-moar 54f647: OUTPUT«Nil␤»
samcv uhmm
raiph when it works, i agree it's best with OR logic as the assumption if it's a list
samcv m: 'a'.uniprop('sc').say
camelia rakudo-moar 54f647: OUTPUT«Latin␤»
samcv m: say 'a' ~~ /<:Latin>/ 21:10
camelia rakudo-moar 54f647: OUTPUT«｢a｣␤»
samcv m: say 'a' ~~ /<:Script<Latin>>/
camelia rakudo-moar 54f647: OUTPUT«｢a｣␤»
samcv oh yeah. there's a bug where sc != Script
m: say 'a' ~~ /<:Script<Latin>>/
camelia rakudo-moar 54f647: OUTPUT«｢a｣␤»
samcv m: say 'a' ~~ /<:Script<Arabic Latin>>/
camelia rakudo-moar 54f647: OUTPUT«Nil␤»
samcv m: say 'a' ~~ /<:Script<Latin Arabic>>/
camelia rakudo-moar 54f647: OUTPUT«Nil␤»
samcv idk what it tries to even do if anything 21:11
may just not know wtf to do with two values
m: say 'a' ~~ /<:Script<Latin Latin>>/
camelia rakudo-moar 5db0b2: OUTPUT«Nil␤»
samcv yeah
raiph m: say 'a' ~~ /<:Script<'Latin Arabic'>>/
camelia rakudo-moar 5db0b2: OUTPUT«Nil␤»
raiph m: say 'a' ~~ /<:Script<<'Latin Arabic'>>>/ 21:12
camelia rakudo-moar 5db0b2: OUTPUT«Nil␤»
raiph m: say 'a' ~~ /<:Script('Latin Arabic')>/
camelia rakudo-moar 5db0b2: OUTPUT«Nil␤»
raiph m: say 'a' ~~ /<:Script('Latin')>/
camelia rakudo-moar 5db0b2: OUTPUT«｢a｣␤»
Geth kudo/nom: 2922ec49ff | (Elizabeth Mattijsen)++ | src/core/Baggy.pm
Make Baggy.antipairs about 25% faster

By rewriting it in nqp:: ops, and using Pair.antipair		 21:17
kudo/nom: e5ca5c278c | (Elizabeth Mattijsen)++ | 3 files
Give Baggy an iterator, move .pairs to QuantHash

Because now we can.		 21:41
samcv curses the heavens 21:47
m: 129342.uniprop('Prepend').say
camelia rakudo-moar 2922ec: OUTPUT«E_Base␤»
samcv ugh. my constant issue lol. that the property _value_'s and the property _names_ comingle 21:48
jnthn samcv: I'm guessing in your re-impl of ucd2c we shall *not* be repeating this? :) 22:14
samcv yep
I still haven't heard back on my grant proposal though :( 22:15
brokenchicken :/ weird
samcv gonna send him another email 22:16
any opinion on the other thing i mentioned jnthn , regarding script extensions 22:17
storing them like 1 = Greek 2 = Latin. so Greek+Latin would be 3 22:18
at least for storing in MVM
lizmat samcv: I was wondering about your grant application, haven't seen anything about it :-( 22:25
samcv yeah i sent another email. last i heard back was a week ago
jnthn samcv: It sounds reasonable at first blush. 22:26
But I'm more than a little tired, so... :)
brokenchicken samcv: ask [Coke] to poke around about it. He's a TPF member IIRC
[Coke] I'm guessing samcv missed the submission deadline for this run 22:33
everything goes through makoto, and he hasn't mentioned it.
jnthn Hm, can we do it from the Perl 6 core dev fund? It meets those reqs, IMO. 22:34
samcv so i may have to wait an extra month or how long is the next time period?
[Coke] every two months is the general fund.
samcv i tried to submit it as an ian hague grant. and he said:
I learned that Ian Hague fund has no allocatable money. I am discussing the
best option with the Board.
idk if what i did was wrong
[Coke] I don't see why it couldn't be addressed from one of the other funds - but then it isn't a GC grant.
the core dev fund is different. 22:35
jnthn samcv: It was sensible in so far as that *was* the right thing to do. But recently that fund was indeed exhausted.
samcv ah ok
brokenchicken :(
jnthn (But you couldn't have reasonably known that from what's on the TPF site.)
samcv yea
jnthn But the Perl 6 core dev fund was the longer term solution for funding Perl 6 core dev work
[Coke] jnthn: well, based on prposed work - not sure if all the grants allocated are still *really* in progress.
samcv i sent the app to karen which is what TPF site said. and then just didn't hear back
[Coke] karen doesn't work there anymore.
samcv oh 22:36
the site needs to be. updated
jnthn There's More Than One Way To Be Outdated :P
[Coke] are you talking with Makoto now?
samcv: it has been
samcv yeah i am [Coke]
it has not. see www.perlfoundation.org/ian_hague_pe...ent_grants
[Coke] ok. Keep doing that, he's the best person to talk to
samcv: see www.perlfoundation.org/contact_us
samcv grant page: Please submit this form when completed to Karen Pauley, President of TPF, karen-at-perlfoundation.org 22:37
[Coke] I'll let M. know about the email on the hague page.
samcv thanks
jnthn [Coke]++
[Coke] samcv: also, feel free to ask people here when you have questions, as there are least 3 of us who are involved somehow. (like you're doing now.)
jnthn has no idea of the current status of the Perl 6 core dev fund, tbh.
(As in, how much is in the pot) 22:38
Guess I get to find out in a few weeks when I need to ask that. :)
brokenchicken I'm getting mixed info TBH. Some people say there's plenty of money. Others say there's no money.
samcv heh
maybe they're both right ;P
also aren't those kind of subjective words
[Coke] brokenchicken: there are different buckets of money. so they're probably all right.
(hague grant, core fund, general grants) 22:39
samcv was hoping to find out in a week or two so I could know whether or not I need to get or job, or could do that after working on the grant 22:40
brokenchicken [Coke]: is there a publicly available sum for "might be used for Perl 6 grants", regardless of which bucket the funds are from or is this stuff secret? 22:42
lizmat samcv: so which bucket were you going for ?
samcv i submitted an email for the ian hague grant according to instructions on www.perlfoundation.org/ian_hague_pe...ent_grants
so 1. the email it says to email on there, person doesn't work there anymore. 2. there's no money in that fund
brokenchicken I went with whatever the link was on P6Weekly; prolly a general grant. It asked for Perl 5 Pod submitted on a web page; no emails. 22:43
lizmat ah, ok, that's outside of the two-monthly grant applications, afaik
lizmat is tired and goes to bed 22:44
good night, #perl6-dev!
brokenchicken night
samcv night lizmat 22:45
timotimo gnite lizmat 22:46
jnthn 'night, lizmat o/ 22:47
[Coke] brokenchicken: as far as I know the buckets are controlled by the treasurer. I'm on the GC, and I have no idea how much is in that bucket. 22:52
so, no, it's not common knowledge. (I assume other board members and maybe the GC chair know) 22:53
timotimo GC is what again? Grand Committee?
Grand Coalition
Grant Committee, probably
[Coke] grants committee
samcv Garbage Collector
timotimo Garden Communists
samcv heh 22:54
[Coke] global
... and then I lost it
samcv it's faster to put functions before where they're actually used right? 22:55
as far as compiling speed, or does it not matter?
jnthn Barely matters
brokenchicken OK. Thanks.
jnthn (We poke an entry into a "check it later" hash if it isn't there, but it's really not particularly costly) 22:56
samcv TPF: i.ytimg.com/vi/odEIOnI64Hw/hqdefault.jpg
[Coke] samcv++ 23:02
samcv ah i found a bug. err. maybe this. yeah. i think i already knew this existed 23:03
m: my $var = 0x0600.chr ~ 'a'; say $var.chars; say $var.uninames
camelia rakudo-moar e5ca5c: OUTPUT«2␤(ARABIC NUMBER SIGN LATIN SMALL LETTER A)␤»
samcv m: my $var = 0x0600.chr ~ '2'; say $var.chars; say $var.uninames
camelia rakudo-moar e5ca5c: OUTPUT«1␤(ARABIC NUMBER SIGN DIGIT TWO)␤»
samcv it's cause numbers have NFG_QC = False, but numbers have True. and currently prepending characters only work for characters with NFG_QC=False after them 23:04
jnthn Don't prepends have NFG_QC=False? 23:07
That should be enough, no?
(Enough to end up doing the full set of checks, I mean)
jnthn goes to get some much-needed rest :) 23:08
'night
samcv nope jnthn 23:10
prepends have false, and if the one after them has True, then it thinks all is well and good
if it's false it checks and knows it shouldn't break
hm thinking i should try installing swift 23:36

Powered by Cro and the Raku® Programming Language.

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

Thank you!