|
🦋 Welcome to the IRC channel of the core developers of the Raku Programming Language (raku.org #rakulang). This channel is logged for the purpose of history keeping about its development | evalbot usage: 'm: say 3;' or /msg camelia m: ... | Logs available at irclogs.raku.org/raku-dev/live.html | For MoarVM see #moarvm Set by lizmat on 8 June 2022. |
|||
|
00:00
reportable6 left
00:02
reportable6 joined
01:02
greppable6 left,
linkable6 left,
committable6 left,
nativecallable6 left,
evalable6 left,
bloatable6 left,
reportable6 left,
statisfiable6 left,
sourceable6 left,
squashable6 left,
benchable6 left,
quotable6 left,
coverable6 left,
notable6 left,
tellable6 left,
bisectable6 left,
releasable6 left,
unicodable6 left,
shareable6 left,
unicodable6 joined,
reportable6 joined,
notable6 joined,
sourceable6 joined
|
|||
| ugexe | Anyone have any suggestions to speed up this pure raku sha1 implementation? its pretty slow - gist.github.com/ugexe/06469f5d3e06...ceed814883 | 01:02 | |
|
01:03
tellable6 joined,
nativecallable6 joined,
releasable6 joined,
linkable6 joined,
quotable6 joined,
coverable6 joined
01:04
benchable6 joined,
greppable6 joined,
statisfiable6 joined,
shareable6 joined,
bloatable6 joined,
squashable6 joined,
bisectable6 joined,
committable6 joined
01:05
evalable6 joined
|
|||
| ugexe | infix:<+&> | 02:26 | |
| SETTING::src/core.c/Int.pm6:465 | |||
| there is a bunch of allocations like that, but why does it appear to be using the Int multi and not the int multi? | |||
|
02:38
MasterDuke joined
|
|||
| MasterDuke | you have to type pretty much every operand. i.e., `my int $a = 3; $a + 5` will use the Int multi. `my int $a = 3; $a + (my int $ = 5)` should use the int multi | 02:41 | |
| i'm not sure if we convert multiply/divide/mod by powers of 2 to shifts. might try doing that manually to see if there's a difference | 02:42 | ||
| ugexe | m: BEGIN my int $ff = 0xFF; | 03:00 | |
| camelia | ===SORRY!=== Error while compiling <tmp> An exception X::Comp::AdHoc occurred while evaluating a BEGIN: Lexical with name '$ff' has wrong type. real type 8 wanted type -1 at <tmp>:1 Exception details: ===SORRY!=== Error while com… |
||
| ugexe | m: my int $ff = 0xFF; | ||
| camelia | ( no output ) | ||
| ugexe | replacing the various constant literal ints like 64 with $i64 where `my int $i64 = 64;` doesn't seem to have helped | 03:18 | |
| gist.github.com/ugexe/dcd034259f1c...3525164bd7 | 03:19 | ||
| MasterDuke | well, a profile shows a decent amount of time spent in line 53 | ||
| ugexe | infix:<+&> | ||
| SETTING::src/core.c/Int.pm6:459 | |||
| 1164435 | |||
| like where are all those coming from | |||
| MasterDuke | maybe `@words[$t] = (@words[$t - 3] +^ @words[$t - 8] +^ @words[$t - 14] +^ @words[$t - 16]);` would by faster if all those array accesses were pulled out into explicit natives | 03:20 | |
| ugexe | in the second gist i have `@words[$t] = ((@words[$t] +< $i1) +& $ffffffff) +| (@words[$t] +> $i31);`, and since @words is `int @words` shouldn't that all be native accesses? | 03:21 | |
| oh, re: "where are all those coming from" i was assuming because it was Int.pm6 that it was using the Int candidates, but the int candidates are in there too | 03:22 | ||
| if i write that same algorithm in perl it does the 500000 character string in less than a second, but it takes the raku code 3.6 seconds | 03:25 | ||
| MasterDuke | but the literals might be be native | 03:35 | |
| might *not* | |||
| ugexe | i replaced all the literals like 32 with $i32 (which was `my int $i32 = 32`). are you suggesting that $i32 might not be a native? | 03:38 | |
|
03:38
evalable6 left,
linkable6 left
|
|||
| MasterDuke | oh, i kind of missed that `$ffffffff` was already a native | 03:39 | |
| for example | 03:40 | ||
|
03:40
linkable6 joined
|
|||
| ugexe | yeah forcing all those literals to be native didn't seem to improve the runtime (it was actually slightly worse) | 03:40 | |
|
03:41
evalable6 joined
|
|||
| MasterDuke | idle curiosity, but why do you have some `BEGIN my int @a = <...>` and some `my int @b = BEGIN <...>` | 03:41 | |
|
06:00
reportable6 left
06:02
reportable6 joined
|
|||
| Geth | rakudo/main: 4b2c40728c | ab5tract++ | src/Raku/ast/package.rakumod RakuAST: Fix role inheritence not working |
06:59 | |
| rakudo/main: 0572d44140 | niner++ (committed using GitHub Web editor) | src/Raku/ast/package.rakumod Merge pull request #5363 from ab5tract/add-role-inheritence RakuAST: Fix role inheritence not working |
|||
|
07:02
codesections left
07:40
sena_kun joined
09:24
squashable6 left,
squashable6 joined
09:35
Xliff joined
|
|||
| Xliff | Raku-GLib suite timing statistics for Raku 2023.08.51.gec.2076759.b | 09:35 | |
| Using the following processor: AMD Ryzen 9 5950X 16-Core Processor | |||
| Total number of projects: 34 (506509 loc) | |||
| Total non-parallel compile times: 10790.200 (317.359 avg) | |||
| tellable6 | 2023-08-27T08:30:44Z #raku <lizmat> Xliff: please provide sample and --ll-exception trace :-) | ||
| Xliff | Total parallel compile times: 1948.092 (57.297 avg) 5.539x speedup | ||
| .tell lizmat gist.github.com/Xliff/f9658ed834dc...c36f869ac3 # As requested. | 10:26 | ||
| tellable6 | Xliff, I'll pass your message to lizmat | ||
| nemokosch | this "no space on the device" came up a couple of days ago | 10:28 | |
| lizmat | it doesn't ring a bell with me... | 10:29 | |
| nemokosch | you said it then as well, Xliff | ||
| "Turns out the problem was NOT drive space, but open apps." | |||
| Xliff | Yeah. In this case, pulsar (spiritual successor to the Atom Editor) | 10:32 | |
| But why is pulsar creating a false "no space left on device" error in Raku? | |||
| nemokosch | I don't know if Atom was also this huge but the size of Pulsaar is just beyond ridiculous | 10:33 | |
| Xliff | Yes. Atom was that huge. | ||
| The previous machine had something similar, except it was "too many open files" | |||
| nemokosch | What makes you use it? It's pretty basic as well | ||
| Xliff | It's not if you have the right plugins (and I've written a few for my own personal use!) | 10:34 | |
| nemokosch | I remember I thought "VSC started off from this, it can't be that bad", and yes, it was that bad, ugly and stupid and all | ||
| got rid of it very soon | |||
| Xliff | What? VSC? | ||
| nemokosch | Atom | ||
| VSC is absolutely rocket science compared to it | |||
| Xliff | Yes, welll.... color me spoiled. It's treated me well over the last decade. | ||
| I'm not sold on VSC (color me crazy). | 10:35 | ||
| Pulsar works, but has teething issues. | |||
| I would switch if I had a decent alternative that I haven't used (and abandoned) before. And that's a LOT. | |||
| nemokosch | I took part in migrating the code of the package manager back to JS | ||
| Xliff | Syntax highlighting is decent. | ||
| nemokosch: Ah! Nice! | 10:36 | ||
| nemokosch | it wasn't a huge work but it took a couple of months to be integrated | ||
| but it happened at least | |||
| it really makes a difference how your work is received, especially when it was at least agreed on | 10:37 | ||
| Xliff | True enough. Now if only I can get my own work out there. | 10:39 | |
| Only time will tell. | |||
| Stuff keeps getting in the way. Now it's medical (both mine and mom's) | |||
| nemokosch | take care... | 10:40 | |
| Xliff | Insurance dropped me after only 41 days. I was ready to pay 2 weeks late. | ||
| Now I have no insurance and can't get any until Open Enrollment. This whole thing strikes me as a big con. Why does medical insurance need an Open Enrollment period? | |||
| Especially when they are ready to kick people off of their roles, early! | |||
| It's not like there's a waiting list! | 10:41 | ||
| lizmat | profit ? | ||
| Xliff | I guess. | ||
| But I doubt it. Big Pharma is going to make its money whether I'm on its roles or not! | |||
|
11:04
NemokoschKiwi joined
|
|||
| lizmat | Xliff: looking at your gist, not getting much further as a .DEPARSE fails as well in a weird way | 11:22 | |
| and *that* appears to be related to the ColonPair::Value | 11:23 | ||
| ugexe | MasterDuke: the ones with BEGIN before ‘my’ are basically constant, while the other BEGIN goes to an array that is modified (but the initial default values are always the same) | 11:30 | |
| lizmat | m: BEGIN my @a = 1,2,3; @a.push(42); dd @a | ||
| camelia | Array @a = [1, 2, 3, 42] | ||
| lizmat | m: my constant @a = 1,2,3; @a.push(42); dd @a | 11:31 | |
| camelia | Cannot call 'push' on an immutable 'List' in block <unit> at <tmp> line 1 |
||
| lizmat | MasterDuke ^^ the difference | ||
| ugexe | In my code the difference wouldn’t be that error though. The values of the array would just be wrong | ||
| It’s more like using BEGIN to initial the array but then doing @a[1] = 42 | 11:33 | ||
| the second time the function is called @a won’t have the expected values | 11:34 | ||
| lizmat | Xliff: the RakuAST::ColonPair::Value.new(key => "\$upc") is missing a value | 11:44 | |
| Xliff: did you mean RakuAST::ColonPair::True.new(key => '$upc') ?? | 11:45 | ||
| ab5tract | m: my $content = 'role A { method m() { self.x} }; class B does A { has $.x = 42; method m() { self.A::m() } }; dd B.new.m'; shell Q:scalar<RAKUDO_RAKUAST=1 $*EXECUTABLE -e '$content'> | 11:50 | |
| camelia | No such method 'A::m' for invocant of type 'B' in method m at -e line 1 in block <unit> at -e line 1 The spawned command 'RAKUDO_RAKUAST=1 /home/camelia/rakudo-m-inst-1/bin/perl6-m -e 'role A { method m() { self.x} }; class B does A { has… |
||
| ab5tract | nine: any thoughts on how one might go about fixing this? ^^ | ||
| there's so many little corners just to getting 't/spec/S14-roles/basic.t' to compile and run suncessfully, this is one of them | 11:51 | ||
| lizmat | Xliff: if you fix the ColonPair::Value issue, the code EVALs ok | 11:52 | |
| Geth | rakudo/main: 48838af431 | (Elizabeth Mattijsen)++ | src/Raku/ast/pair.rakumod RakuAST: ::ColonPair::Value *must* have a value specified |
11:53 | |
|
12:00
reportable6 left
12:01
reportable6 joined
13:09
NemokoschKiwi left
|
|||
| ab5tract | any idea how I can convert a RakuAST::Call::Method object ot a RakuAST::Method object? | 13:27 | |
| I can't seem to figure out how to create an expression out of a RakuAST::Postfixish, despite it living in expressions.rakumod | 13:35 | ||
|
13:49
jgaz joined
|
|||
| ab5tract | got it :D | 14:33 | |
| m: role A { method m() { self.x} }; class B does A { has $.x = 42; method m() { self.A::m() } }; dd B.new.m | 14:41 | ||
| camelia | 42 | ||
| ab5tract | I've never seen this `self.A::m()` syntax before. | 14:42 | |
| unfortunately the exception is thrown late enough that I don't see any RakuAST module references in the stacktrace, so I'm really not sure where to begin. | 14:45 | ||
| but... if we fix that, then I get a single (failing) test file to run without a bad exit code. I'm aiming high! | 14:47 | ||
| ugexe | m: my @HEXBYTES[16] = flat 0..9, "A".."F"; say @HEXBYTES | 14:58 | |
| camelia | [0 1 2 3 4 5 6 7 8 9 ...] | ||
| ugexe | m: BEGIN my @HEXBYTES[16] = flat 0..9, "A".."F"; say @HEXBYTES | ||
| camelia | (Array) | ||
| ugexe | m: BEGIN my @HEXBYTES = flat 0..9, "A".."F"; say @HEXBYTES | ||
| camelia | [0 1 2 3 4 5 6 7 8 9 A B C D E F] | ||
| [0 1 2 3 4 5 6 7 8 9 A B C D E F] | 14:59 | ||
| ugexe | BEGIN + shaped array seems to not work right | ||
| ab5tract | m: my $content = 'my @HEXBYTES[16] = flat 0..9, "A".."F"; say @HEXBYTES'; shell Q:scalar<RAKUDO_RAKUAST=1 $*EXECUTABLE -e '$content'> | 15:32 | |
| camelia | [0 1 2 3 4 5 6 7 8 9 ...] | ||
| ab5tract | ugexe: fixed in RakuAST?? :O :D | ||
| ugexe | heh | 15:38 | |
| m: sub foo(str $a) {}; my $arg = "xxx"; say $arg; foo($arg) | |||
| camelia | xxx | ||
| ugexe | m: proto sub foo(|) {*}; multi sub foo(str $a) {}; my $arg = "xxx"; say $arg; foo($arg) | ||
| camelia | xxx Cannot resolve caller foo(Str:D); none of these signatures matches: (str $a) in block <unit> at <tmp> line 1 |
||
| ugexe | why does it being changed to a multi have that effect? | 15:39 | |
| ab5tract | m:my $content = 'proto sub foo(|) {*}; multi sub foo(str $a) {}; my $arg = "xxx"; say $arg; foo($arg)'; shell Q:scalar<RAKUDO_RAKUAST=1 $*EXECUTABLE -e '$content'> | 15:48 | |
| evalable6 | (exit code 1) xxx Cannot resolve caller foo(Str… |
15:49 | |
| ab5tract, Full output: gist.github.com/c2fb62050ddc8033d8...59c9110feb | |||
| ab5tract | twice in one day would be too much ;) | ||
|
16:11
linkable6 left,
evalable6 left,
evalable6 joined
16:12
linkable6 joined
|
|||
| Xliff | lizmat: Ah! Thanks so much! | 17:02 | |
| ab5tract | m: my $content ='class A:auth<Foo> {}; dd A.^auth'; shell Q:scalar<RAKUDO_RAKUAST=1 $*EXECUTABLE -e '$content'> | 17:03 | |
| camelia | ===SORRY!=== Error while compiling Undeclared routine: A used at line -1. Did you mean ''? The spawned command 'RAKUDO_RAKUAST=1 /home/camelia/rakudo-m-inst-1/bin/perl6-m -e 'class A:auth<Foo> {}; dd A.^auth'' exited unsuccessfully… |
||
| ab5tract | m: class A:auth<Foo> {}; dd A.^auth | ||
| camelia | "Foo" | ||
| ab5tract | I guess Foo:auth<Bar> is NYI in RakuAST... getting so closed to getting this test file to pass | 17:04 | |
| Xliff | m: RakuAST::ColonPair::True.new("upc").gist.sau | 17:16 | |
| camelia | ===SORRY!=== Error while compiling <tmp> Use of RakuAST is experimental; please 'use experimental :rakuast' at <tmp>:1 ------> RakuAST::ColonPair::⏏True.new("upc").gist.sau expecting any of: statement list … |
||
| Xliff | m: use experimental :rakuast; RakuAST::ColonPair::True.new("upc").gist.sau | 17:17 | |
| camelia | No such method 'sau' for string 'RakuAST::ColonPair::True.new("upc")'. Did you mean 'say'? in block <unit> at <tmp> line 1 |
||
| Xliff | m: use experimental :rakuast; RakuAST::ColonPair::True.new("upc").gist.say | ||
| camelia | RakuAST::ColonPair::True.new("upc") | ||
| Xliff | m: use experimental :rakuast; RakuAST::ColonPair::True.new("upc").DEPARSE.say | ||
| camelia | :upc | ||
| lizmat | Xliff: use v6.* is shorter :-) | 17:38 | |
| Geth | rakudo/main: ea40e4c626 | (Elizabeth Mattijsen)++ | src/core.c/RakuAST/LegacyPodify.pm6 RakuAST: introduce =numitem/=numhead As alternatives to =item # / =head # , aka an alternative to specifying # as an alternative to :numbered. As per discussion in github.com/finanalyst/Rakudoc-BETA/issues/47 |
17:58 | |
|
18:00
reportable6 left
18:01
reportable6 joined
18:59
Xliff left
20:32
NemokoschKiwi joined
20:53
samcv_ joined
21:01
djinni`_ joined
21:02
discord-raku-bot left,
djinni` left,
samcv left,
JRaspass left
21:07
JRaspass joined
21:41
sena_kun left
22:53
NemokoschKiwi left
23:53
quotable6 left,
statisfiable6 left,
coverable6 left,
sourceable6 left,
linkable6 left,
greppable6 left,
committable6 left,
tellable6 left,
releasable6 left,
reportable6 left,
nativecallable6 left,
benchable6 left,
bisectable6 left,
unicodable6 left,
notable6 left,
squashable6 left,
bloatable6 left,
evalable6 left,
shareable6 left
23:54
coverable6 joined,
sourceable6 joined,
quotable6 joined,
statisfiable6 joined
23:55
releasable6 joined,
unicodable6 joined,
shareable6 joined,
squashable6 joined,
evalable6 joined,
benchable6 joined,
bisectable6 joined,
notable6 joined
23:56
linkable6 joined,
committable6 joined,
reportable6 joined,
bloatable6 joined,
greppable6 joined,
nativecallable6 joined,
tellable6 joined
|
|||