🦋 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 00:03 reportable6 joined 00:34 NemokoschKiwi left 00:58 AlexDaniel left 02:57 samebchase2 joined 02:58 samebchase left, samebchase2 is now known as samebchase
MasterDuke yeah, --target=optimize is different for int vs uint in this case. uint is actually calling `&infix:<..>`, but int is getting turned into a while loop 03:01
huh, a minor change in Optimizer.nqp and a profile shows the same runtime, but `time` shows ~1.1s for uint and ~0.2s for int 03:16
and --target=optimize looks the same
ha. so the --target=optimize is now the same, but wallclock is slower?! because before my change, time reports ~0.6s for uint (and the same ~0.2s for int) 03:20
and before my change a profile reports uint being *much* slower 03:21
so now it looks like we're also in some spesh/jit differences
Geth nqp/main: 953b0ee3a6 | (Daniel Green)++ | CREDITS
Add myself to the credits
06:00 reportable6 left 06:01 reportable6 joined 06:49 squashable6 left 06:52 squashable6 joined 09:10 sena_kun joined 09:50 sena_kun left 10:14 sena_kun joined 10:18 sena_kun left 10:55 sena_kun joined
Geth rakudo/lizmat-Rat-Str-digits: 5a1abdc618 | (Elizabeth Mattijsen)++ | src/core.c/Rational.pm6
Introduce Rational.Str(:digits)

Allows one to specify number of digits for stringification of Rat/FatRats. Inspired by discussion by tadzik++ at
rakudo: lizmat++ created pull request #5223:
Introduce Rational.Str(:digits)
11:53 codesections left
Geth rakudo/main: 376eb9180e | (Elizabeth Mattijsen)++ | src/core.c/RakuAST/Raku.pm6
RakuAST: don't bother with () if there are no arguments

In the .raku handling of arguments to methods.
12:00 reportable6 left 12:01 reportable6 joined 12:11 thundergnat joined 12:14 thundergnat left 12:43 jdv left 12:44 jdv joined 12:56 epony left, linkable6 left, samcv left, nativecallable6 left, tellable6 left, benchable6 left, sourceable6 left, bloatable6 left, unicodable6 left, statisfiable6 left, bartolin_ left, japhb left, jjatria left, sivoais left, nebuchadnezzar left 13:01 epony joined, linkable6 joined, samcv joined, nativecallable6 joined, tellable6 joined, benchable6 joined, sourceable6 joined, bloatable6 joined, unicodable6 joined, statisfiable6 joined, bartolin_ joined, japhb joined, jjatria joined, sivoais joined, nebuchadnezzar joined 13:04 epony left 13:05 epony joined 13:20 epony left 13:30 nebuchad` joined 13:31 nebuchadnezzar left
lizmat looking at taking a stab at moving Code.assuming to RakuAST 13:35
and it makes me wonder whether we need a way to take a Parameter object, and turn it into a RakuAST::Parameter object 13:36
or otherwise create a RakuAST::Parameter object that would just take the Parameter object as the product of PRODUCE-META-OBJECT 13:37
13:53 epony joined
lizmat also: maybe "use v6.e.PREVIEW" should imply "use experimental :rakuast", as the PREVIEW in itself is not set in stone yet either ? 13:55
sjn wonders what point of "PREVIEW" is. If the list of "PREVIEW" features today is different from "PREVIEW" features next week, then this name doesn't help much in distinguishing between them, does it? 14:22
vrurg PREVIEW means you're ready to experiment. 14:28
lizmat: I'd be ++ for implying.
Nemokosch yes... PREVIEW is not a version, it's a disclaimer 😛
vrurg Well said. 14:29
It is indeed.
Nemokosch it actually goes well with the things that "need to be shouted"
[Coke] lizmat++ adding rakust to 6.e seems fine to me. 14:30
14:43 Altai-man joined 14:44 sena_kun left 15:52 epony left 15:53 epony joined 18:00 reportable6 left 18:02 reportable6 joined
Geth rakudo/lizmat-RakuAST-EVAL: 349a7fbcaa | (Elizabeth Mattijsen)++ | 2 files
Allow EVAL of RakuAST::Nodes without MONKEY

Since there is no danger of injection with RakuAST::Node, there's no need for "use MONKEY-SEE-NO-EVAL" if you're EVALling RakuAST nodes.
Originally, this was a compile time error. Now it is a run-time error, because we cannot see the difference between a string and a RakuAST ... (7 more lines)
rakudo: lizmat++ created pull request #5224:
Allow EVAL of RakuAST::Nodes without MONKEY
18:10 Altai-man left
[Coke] I may reach out to core devs for some documentation requests as we try to slog our way through the backlog. 18:14
I will try to keep it to a reasonable number of asks. :)
lizmat [Coke]: sure, shoot :-) 18:15
[Coke] First, a general question: we have many asks for docs for things not in roast. I have been putting those at the bottom of the pile, but those might need someone to weigh in with "yes, we should add a test", or "no, this is an implementation detail" or "maybe document it anyway." I imagine most of these are one offs, but a small workflow on what to do there would be good to agree on. 18:17
I've been trying to tag these as NONSPECCED when I find them.
lizmat fwiw, a lot of things *do* need spectests *and* documentation 18:18
[Coke] ok. I wonder if we should have some sort of notation in roast that lets me point to a specific test or set of tests rather than using rak to find the text somewhere in a file. 18:19
lizmat it was basically the difficulty of specifying after which rakudo version some things became available
[Coke] like some notation of "test that follows is for Object.method" or "language:phaser:CATCH" or something.
(that's a lot of work) 18:20
lizmat: we have the same problem in the docs, there's a large discussion about that going on now.
lizmat and we have a solution for that now, right ?
[Coke] (revived from a 7 year old ticket.)
lizmat ah, no solution yet :-(
[Coke] I think we're actually having trouble agreeing on the problem.
lizmat where's the current discussion? 18:21
[Coke] github.com/Raku/doc/discussions/4206
btw, using discussions has been a big improvement for doc stuff, although we are having some growing pains figuring out the best way to use the interface. 18:22
it's very easy for something non-actionable to languish in a ticket for years, I feel like we're getting some momentum on the talking in discussions.
lizmat yeah, I like the momentum I'm seeing now :-) 18:23
tonyo coke++ 18:28
Geth rakudo/main: 97d15491ae | (Elizabeth Mattijsen)++ | src/core.c/RakuAST/Raku.pm6
RakuAST: add RakuAST::Node.EVAL method

To allow one to EVAL a RakuAST node without needing to do a
  "use SEE-NO-MONKEY-EVAL", similar to Cool.EVAL
18:35 linkable6 left
lizmat vrurg: re making RakuAST available in 6.e: I think it needs a $*WANT_RAKUAST := 1; somewhere, but I'm at a bit of a loss where 18:37
18:38 linkable6 joined
lizmat [Coke] perhaps Blin-report like issues could be made with things that should be documented ? 18:39
[Coke] can you show me a blin report? not sure what it looks like.
lizmat so people can tick off stuff done?
[Coke] we probably already have an xt/ that shows missing methods, at lest.
lizmat: see github.com/Raku/doc/issues?q=is%3A...Achecklist 18:40
there are tons of checklists that JJ (and now me) have added based on the rakudo releases.
lizmat cool... and checked means done, right ?
[Coke] We also have github.com/Raku/doc/blob/main/xt/c...s.rakutest which could probably generate similar checklists for inclusion in tickets. (right now it's just TAP, and lots of it is skipped) 18:41
yes. if an item is checked, the docs got merged. 18:42
once all the checks on a ticket are done, close the ticket.
lizmat then why is github.com/Raku/doc/issues/2673 still open ? 18:43
[Coke] github.com/Raku/doc/issues/2673#is...-683439870 18:44
lizmat ack
[Coke] yah, some of these are a mess. for some of them, I've closed out old ticket that was 99% done and opened a new one with the one thing left. 18:45
lizmat is there a resolution as to how to indicate something is in 6.e ? 18:46
Geth rakudo/main: fcdb78cef2 | (Elizabeth Mattijsen)++ | src/Perl6/World.nqp
use v6.e.PREVIEW or v6.* implies use experimental :rakuast

As this indicates people experimenting already, and when 6.e is released, RakuAST *will* be fully specced and standardly available.
lizmat vrurg: found it, I think ^^
19:02 djinni` left 19:08 linkable6 left 19:09 djinni` joined 19:10 linkable6 joined
Geth nqp/main: f14cc22047 | (Ben Davies)++ (committed using GitHub Web editor) | 3 files
[JVM] Implement the objprimunsigned and objprimbits ops (#781)

  * [JVM] Implement the objprimunsigned op
MoarVM has a couple ops related to objprimspec in that they expose metadata pertaining to anything an object boxes, this being one of them. Would be nice to have a portable way to determine the sign of an int! ... (8 more lines)
19:17 sena_kun joined
Geth nqp/main: 90ca9a6e82 | (Elizabeth Mattijsen)++ | tools/templates/MOAR_REVISION
Bump MoarVM to get Ben Davies' last fixes
rakudo/main: b721f0a04c | (Elizabeth Mattijsen)++ | tools/templates/NQP_REVISION
Bump NQP to get Ben Davies' last fixes
vrurg lizmat: that location is ok, but I would make it in the Grammar, right after <lang-version> at line 839 19:38
21:23 vrurg left 21:24 vrurg joined 21:45 Xliff joined
[Coke] aaaahahaha. realized the duplicate word checker in raku/docs was only doing pod6 & md. expanded it to all files. found a duplicate word *in the artistic license* 22:56
ah, it's ok, though. "offer to sell, sell" 22:57
lizmat hehe
23:00 kjp left 23:04 kjp joined 23:14 sena_kun left