[00:03] <Geth> ¦ rakudo/main: 22cf0082a9 | (Nick Logan)++ (committed using GitHub Web editor) | 4 files

[00:03] <Geth> ¦ rakudo/main: RakuAST: branch on a native-int condition directly (#6470)

[00:03] <Geth> ¦ rakudo/main: 

[00:03] <Geth> ¦ rakudo/main: Previously a condition that computed a native int still boxed a Bool

[00:03] <Geth> ¦ rakudo/main: for the branch to test. With compile-time dispatch inlining putting

[00:03] <Geth> ¦ rakudo/main: the native comparison in condition position, every iteration of a

[00:03] <Geth> ¦ rakudo/main: native-typed loop paid a boolification, a return-type check on the

[00:03] <Geth> ¦ rakudo/main: inlined operator, and the unboxing of the result. A bare native-int

[00:03] <Geth> ¦ rakudo/main: <…commit message has 16 more lines…>

[00:03] <Geth> ¦ rakudo/main: review: https://github.com/rakudo/rakudo/commit/22cf0082a9

[00:13] *** Geth left
[00:25] *** committable6 left
[00:25] *** japhb left
[00:25] *** committable6 joined
[00:26] *** japhb joined
[03:41] *** nine left
[03:41] *** nine joined
[11:00] <releasable6> Next release in ≈3 days and ≈7 hours. There are no known blockers. Please log your changes in the ChangeLog: https://github.com/rakudo/rakudo/wiki/ChangeLog-Draft

[12:50] <[Coke]> Reminder, anything for the legacy front end, please be really careful at this point in the release cycle.

[12:51] <[Coke]> If anyone is interested in learning how to do a release, let me know. (though at this point it's mostly following the guides)

[15:05] <[Coke]> m: sub a(Int --> True} { return True}; 

[15:05] <camelia> rakudo-moar 6c5175469: OUTPUT: «===SORRY!=== Error while compiling <tmp>␤Malformed return value␤at <tmp>:1␤------> sub a(Int --> True<HERE>} { return True};␤»

[15:05] <[Coke]> m: sub a(Int $a --> True} { return True}; 

[15:05] <camelia> rakudo-moar 6c5175469: OUTPUT: «===SORRY!=== Error while compiling <tmp>␤Malformed return value␤at <tmp>:1␤------> sub a(Int $a --> True<HERE>} { return True};␤»

[15:06] <[Coke]> slightly older version is giving a more explicit can't have a return value if you have it in the signature (and they complaining because that diagnostic used .perl)

[15:06] <[Coke]> *then

[16:21] *** Voldenet_ joined
[16:22] *** Voldenet left
[16:22] *** Voldenet_ is now known as Voldenet

[16:45] <ab5tract> ISTR the diagnostic being updated, but I’d have to check GitHub to be sure

[16:46] <ab5tract> Right now I have to finish prepping for my Star Wars RPG session … in fifteen minutes 😅

[16:48] <ab5tract> I expect the intention was just to ditch the .perl and save the user from some noise, so it may be worth double checking if the patch is also throwing a different(ly formed) exception

[16:52] <[Coke]> bisectable6: class A { method b(Int $a --> True} { return True} }

[16:52] <bisectable6> [Coke], Will bisect the whole range automagically because no endpoints were provided, hang tight

[16:52] <[Coke]> (was actually on a method, soo.

[16:53] <bisectable6> [Coke], Output on all releases: https://gist.github.com/e664ad799003c47947826b75ae26729c

[16:53] <bisectable6> [Coke], Bisecting by output (old=2024.10 new=2024.12) because on both starting points the exit code is 1

[16:53] <bisectable6> [Coke], bisect log: https://gist.github.com/7d909755499640d23333f9bec3b204b4

[16:53] <bisectable6> [Coke], (2024-11-22) https://github.com/rakudo/rakudo/commit/a917cf032144e2524fe798153f6e79b4dced35bc

[16:53] <bisectable6> [Coke], Bisecting by output (old=2017.09 new=2017.10) because on both starting points the exit code is 1

[16:54] <bisectable6> [Coke], bisect log: https://gist.github.com/4aba3e9bf366a751cee9ab9db445415b

[16:54] <bisectable6> [Coke], (2017-10-07) https://github.com/rakudo/rakudo/commit/1ed21c233cea9e9b87cef3a7b7c8e55ec98f4959

[16:54] <bisectable6> [Coke], Output on all releases and bisected commits: https://gist.github.com/ff786451f5d12118e175a69f79693df0

[16:54] <[Coke]> must have been some other code outside that golf. ah well

[16:55] <[Coke]> m: class A { method b(Int $a --> True} { } }

[16:55] <camelia> rakudo-moar 6c5175469: OUTPUT: «===SORRY!=== Error while compiling <tmp>␤Malformed return value␤at <tmp>:1␤------> class A { method b(Int $a --> True<HERE>} { } }␤»

[16:55] <[Coke]> m: class A { multi method b(Int $a --> True} { } }

[16:55] <camelia> rakudo-moar 6c5175469: OUTPUT: «===SORRY!=== Error while compiling <tmp>␤Malformed return value␤at <tmp>:1␤------> class A { multi method b(Int $a --> True<HERE>} { } }␤»

[16:56] <[Coke]> I'll see if I can duplicate it offline, but it's probably just user errors

[17:45] <ugexe> is the error we're seeing not legit enough given the signature isn't written correctly?

[17:45] <ugexe> m: class A { method b(Int $a --> True) { return True} }

[17:45] <camelia> rakudo-moar 6c5175469: OUTPUT: «===SORRY!=== Error while compiling <tmp>␤No return arguments allowed when return value Bool::True is already specified in the signature␤at <tmp>:1␤------>  method b(Int $a --> True) { return True<HERE>} }␤»

[17:46] <ugexe> m: class A { method b(Int $a --> True} { return True} }

[17:46] <camelia> rakudo-moar 6c5175469: OUTPUT: «===SORRY!=== Error while compiling <tmp>␤Malformed return value␤at <tmp>:1␤------> class A { method b(Int $a --> True<HERE>} { return True} }␤»

[19:02] *** notna joined
[19:38] *** notna left
[20:29] *** hurufu joined
[20:51] <[Coke]> releasable6: next

[20:51] <releasable6> [Coke], Next release in ≈2 days and ≈22 hours. There are no known blockers. 240 out of 240 commits logged

[23:45] *** hurufu left