🦋 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 01:40 squashable6 left, greppable6 left, notable6 left, reportable6 left, unicodable6 left, benchable6 left, committable6 left, quotable6 left, bisectable6 left, bloatable6 left, sourceable6 left, releasable6 left, shareable6 left, bloatable6 joined, notable6 joined, reportable6 joined 01:41 squashable6 joined, unicodable6 joined, quotable6 joined, committable6 joined, releasable6 joined 01:42 sourceable6 joined, greppable6 joined, bisectable6 joined, benchable6 joined 01:43 shareable6 joined 04:45 linkable6 left, evalable6 left, linkable6 joined 04:48 evalable6 joined 05:36 Xliff joined 06:00 reportable6 left 06:03 reportable6 joined 06:17 Xliff left
gfldex m: class Programmer { has $.fun if $*alive }; 07:00
camelia ( no output )
gfldex filed as GH#5268 07:04
That's the most fun I ever had with a compiler bug. :-> 07:05
07:05 linkable6 left 07:08 linkable6 joined 07:44 sena_kun joined 07:51 sena_kun left 08:23 codesections joined
Voldenet m: class gfldex { has $.fun with $*compiler-bug }; 08:36
camelia ( no output )
gfldex :) 08:37
08:42 codesections1 joined 08:44 codesections left, codesections1 is now known as codesections 09:58 codesections1 joined 10:00 codesections left, codesections1 is now known as codesections
Geth nqp/main: 8a925a02f7 | (Jonathan Worthington)++ | src/QRegex/Cursor.nqp
Reuse bstack and cstack when restarting a cursor

While cursors and their match extents are immutable and we must make a new one when backtracking, it seems there's no situation where we will backtrack twice into the exact same cursor. Assuming this is correct (I can't think of a case, the tests don't bring up a case), then it is safe to pass ownership of them from one cursor to another. ... (7 more lines)
10:23 Xliff joined 10:47 squashable6 left 10:49 squashable6 joined
Geth rakudo/main: c88291e2a2 | (Jonathan Worthington)++ | src/Raku/ast/rakuast-prologue.nqp
Publish caches for RakuAST nodes

So that we can do type checks much faster. This means that things like deparse now candidate sort and dispatch far faster.
10:58 Xliff left
Geth rakudo/main: 2c6c0095d5 | (Elizabeth Mattijsen)++ | tools/templates/NQP_REVISION
Bump NQP to get jnthn's regex fix
rakudo/main: d320deedf0 | (Stefan Seifert)++ | 3 files
RakuAST: Avoid DEPARSE during compilation for better performance

DEPARSE can be surprisingly expensive, so when the compiler needs the actual source string, just use the origin instead.
11:15 Xliff joined 11:17 Xliff left 11:41 codesections left 12:00 reportable6 left, reportable6 joined 12:13 codesections joined 12:30 sena_kun joined
Geth rakudo/main: 4a46ee1781 | (Stefan Seifert)++ | 2 files
RakuAST: fixup regex stringification on synthesised ASTs

We don't always have a source code to access. Fall back to DEPARSE in that case.
nqp/main: 7f434701fe | (Jonathan Worthington)++ | src/QRegex/Cursor.nqp
Don't try to reuse the cstack

This is used to form the Match object, and so reusing and clearing it can cause issues.
13:28 discord-raku-bot left, discord-raku-bot joined
lizmat .tell lizmat foo 13:35
tellable6 lizmat, I'll pass that message to your doctor
13:36 finanalyst joined
finanalyst .tell guifa codesections suggested I contact you about internationalisations. 13:39
tellable6 finanalyst, I'll pass your message to guifa
13:39 discord-raku-bot left, discord-raku-bot joined 13:46 codesections left 13:56 discord-raku-bot left, discord-raku-bot joined 14:04 codesections joined
Geth rakudo: Leont++ created pull request #5269:
Add connect-path and listen-path to IO::Socket::Async
rakudo/main: 6eafacd0bc | (Elizabeth Mattijsen)++ | src/core.c/core_epilogue.pm6
RakuAST: add :grammar/:actions to .AST

So that you can specify which actual grammar to use for creating a RakuAST tree. Defaults to the standard Raku grammar/actions.
rakudo/main: 5da5c60527 | (Elizabeth Mattijsen)++ | src/core.c/RakuAST/Fixups.pm6
RakuAST: make =begin rakudo new alternative for =begin pod

Instead of the less descriptive =begin doc. Suggested by
rakudo/main: 760f4640d4 | (Elizabeth Mattijsen)++ | tools/templates/NQP_REVISION
Bump NQP to fix a recent regex regression, jnthn++
rakudo/main: 566f878e0a | (Elizabeth Mattijsen)++ | src/core.c/RakuAST/Fixups.pm6
RakuAST: make sure L<>/X<> deparse ok if there is no meta
rakudo/main: b76e2bc04f | (Elizabeth Mattijsen)++ | src/core.c/core_epilogue.pm6
RakuAST: make sure defaults to .AST are accepted

The default Raku::Grammar is not a HLL object, so needs a Mu. Same for Raku::Actions
rakudo: Leont++ created pull request #5270:
Make IO::Handle.Supply read character-by-character
16:37 evalable6 left, linkable6 left 16:38 linkable6 joined 16:40 finanalyst left, evalable6 joined 16:43 codesections left 16:46 finanalyst joined 16:52 finanalyst left 17:01 Xliff joined 18:00 reportable6 left 18:01 reportable6 joined 19:07 codesections joined
Geth roast/cando-typed-arrays: 7bbc449be1 | (Leon Timmermans)++ | S05-capture/call.t
Add test for #4730
roast: Leont++ created pull request #835:
Add test for #4730
20:58 evalable6 left 20:59 evalable6 joined 21:02 codesections left 21:33 Xliff left 21:50 sena_kun left