🦋 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: ... | log inspection situation still under development | For MoarVM see #moarvm
Set by lizmat on 22 May 2021.
00:02 reportable6 left, reportable6 joined 00:54 frost joined 01:44 Xliff left 02:56 zostay left 02:57 zostay joined 03:18 squashable6 left 04:41 kawaii_ left, rypervenche left, kawaii_ joined, rypervenche joined, rypervenche left, rypervenche joined 04:45 ilogger2 left, tonyo_ left, ilogger2 joined 04:50 tonyo joined 05:30 squashable6 joined 06:02 reportable6 left 06:05 reportable6 joined 06:16 nebuchadnezzar joined 07:44 linkable6 left, evalable6 left 07:46 linkable6 joined
nine t/02-rakudo/15-gh_1202.t is a flapper and cannot work as it is written. That's because nothing guarantees that 5 concurrent `echo $_` shell commands will put their output nicely in 5 rows. Output is bound to get mixed like 12\n\n instead of 1\n2\n 08:19
08:45 frost left 08:47 evalable6 joined 09:09 patrickb joined 09:55 patrickb left 10:11 frost joined
Altai-man can anyone go to dev.azure.com/rakudo/rakudo/_build...dArtifacts and download the archive under `build-result`? 10:20
raydiak I click on it and nothing happens. no javascript error, just no response of any sort 10:23
same in both ff and chromium on linux 10:28
lizmat Altai-man: no response on Safari either 10:39
10:51 patrickb joined 10:56 rakuUser joined 11:04 nebuchadnezzar left, nebuchad` joined 11:09 nebuchad` is now known as nebuchadnezzar 11:23 Geth joined 11:24 sena_kun joined
Geth rakudo/new-disp: 10e37b8706 | (Jonathan Worthington)++ | src/vm/moar/dispatchers.nqp
Allow nextsame et al in detached method calls

That is, when we look up a method and then separately invoke it. We supported this historically, so keep it working. This also fixes the case where we have an NQP-based type with a Raku method installed in it that then does nextsame, which shows up in various modules that built upon the MOP, such as Method::Also (which is fixed by this change).
11:47 patrickb left 12:02 reportable6 left 12:28 patrickbkr joined
patrickbkr Altai-man: re the download button: When clicking on the three dots on the far right a menu with a working download option opens for me. 12:31
sena_kun patrickbkr, is it possible for you to do uploading? 12:33
patrickbkr Yes. 12:36
Shall I!
sena_kun patrickbkr, if it won't bother you too much...
patrickbkr ok. Will do in the evening.
sena_kun thanks
patrickbkr I'm back from vacation as of today btw. :-D 12:37
sena_kun hope you were able to rest well 12:38
patrickbkr Thanks. I did. 12:41
12:48 frost left
Geth ¦ problem-solving: lizmat assigned to codesections Issue Moving distributions between distribution backends "duplicates" them github.com/Raku/problem-solving/issues/295 12:49
¦ problem-solving: lizmat unassigned from codesections Issue Moving distributions between distribution backends "duplicates" them github.com/Raku/problem-solving/issues/295 12:50
13:03 reportable6 joined
lizmat notable6: weekly 13:20
notable6 lizmat, 2 notes: 2021-08-27T01:48:08Z <melezhik>: news.ycombinator.com/item?id=28321812 ; 2021-08-28T14:26:08Z <elcaro>: www.youtube.com/watch?v=UVUjnzpQKUo
lizmat notable6: weekly reset 13:29
notable6 lizmat, Moved existing notes to “weekly_2021-08-30T13:29:47Z”
lizmat And yet another Rakudo Weekly News hits the Net: rakudoweekly.blog/2021/08/30/2021-...perseding/ 13:30
ugexe decuplicating distributions is entirely a per-ecosysem thing 13:44
basically there is a UI problem with raku.land and we are looking to solve it with additional specced meta data, which smells like "when all you have is a hammer" 13:46
there is a lowest common demoninator that is always ignored with those types of solutions: someones local cache of files or a companys dark-pan, 13:47
14:14 b2gills left, b2gills joined
Geth rakudo/new-disp: 6664c423de | (Jonathan Worthington)++ | lib/NativeCall.rakumod
Compile code object unwrapping for native calls
nqp/new-disp: d5830b4b57 | Altai-man++ (committed using GitHub Web editor) | tools/templates/MOAR_REVISION
Bump new-disp MoarVM
rakudo/new-disp: 09105c197b | Altai-man++ (committed using GitHub Web editor) | tools/templates/NQP_REVISION
Bump new-disp nqp
15:14 evalable6 left, linkable6 left 15:15 linkable6 joined 15:16 evalable6 joined
timo Does not reach a complete line within 1 megabyte of data 15:39
Then the server may send a may send a JSON line with an object containing err and then close the connection.
^- jnthn, your Log::Timeline module readme has a little duplicated text in it
16:26 patrickbkr left
Geth rakudo/new-disp: 0567c24317 | (Jonathan Worthington)++ | src/vm/moar/dispatchers.nqp
Fix protos that control multiple candidate lists

For example, when the proto is cloned because multi candidates are defined in a nested scope.
rakudo/new-disp: 6d62a2cb40 | (Jonathan Worthington)++ | src/vm/moar/dispatchers.nqp
Make sure to run `proto` for complex signatures

For example, where there is an unpack or constraint on a parameter in the `proto`, we need to run it in order to enforce that.
17:57 sena_kun left 18:02 reportable6 left
Geth nqp/new-disp: 7440bc8717 | (Stefan Seifert)++ | src/vm/moar/QAST/QASTOperationsMAST.nqp
Fix miscompilation of sized integers as loop condition

We were allocating a register for a coercion result but never actually generated code for doing the actual coercion. This resulted in us reading an uninitialized register which could have caused infinite loops in the worst case.
Fix by doing exactly the same as for if/unless ops.
19:05 reportable6 joined 19:44 patrickb joined
patrickb Altai-man: The binary files have been uploaded. I have also created and uploaded M1 Mac release build. 19:46
Altai-man patrickb, thanks! 19:51
19:54 patrickb left 19:57 discord-raku-bot left, discord-raku-bot joined 22:33 dogbert11 left 23:01 dogbert11 joined
Geth rakudo/new-disp: ed93e37dd4 | (Jonathan Worthington)++ | src/vm/moar/dispatchers.nqp
Don't run every proto during CORE.setting build

We can't properly do the analysis as to whether or not we need to run the proto since we're not yet properly bootstrapped. However, assume that we don't need to run any of them. This saves rather a lot of time, because to run the proto involves compiling it. It also at present triggers the compilation of all multi candidates and would compile them in duplicate.
rakudo/new-disp: f7b641f36a | (Jonathan Worthington)++ | src/Perl6/World.nqp
Don't compile all candidates with a proto

This seems to have been a workaround for a problem that no longer exists. It was also buggy and re-compiled the same candidates many times over; since it's apparently not needed any longer, however, it can be debugged simply by deleting it.