🦋 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. |
|||
02:10
kjp left
02:11
kjp joined
02:12
kjp left,
kjp joined
|
|||
nine | flapper? | 06:46 | |
ab5tract: t/spec/S29-os/system.t fails very often when the system is under load and there's a recent change on it from you | |||
Geth | roast: a83b2d452f | (Stefan Seifert)++ | spectest.data Remove deleted spectest file from the list of files to run The spectest was deleted in commit ae6135510104bf036bdca913471f1242bcf13587 |
06:49 | |
nine | ab5tract: oh, the missing MVMROOT is still not fixed | 06:50 | |
ab5tract | nine: is that surprising? I asked whether you were expecting a patch from me or would take care of it yourself and I recall something about needing to wait on it | 06:57 | |
nine | ab5tract: I also (later) said "if in doubt, an MVMROOT would be preferable" | 07:04 | |
pushed the fix | |||
ab5tract | That didn’t tag me, so for all I knew it was a continuation of a different conversation. Didn’t mean to leave it hanging. | 07:07 | |
nine | Well, bug is gone, so all's well that ends well. Let's just hope that system.t is stable now. | 07:09 | |
Looking through the IRC log, this test has historically been troublesome | |||
I shouldn't have worried about performance in this particular case anyway. Spawning is a super heavy operation. You ain't gonna do that a million times a second anyway. | 07:16 | ||
Geth | rakudo/main: 1586e96ae3 | (Stefan Seifert)++ | 3 files RakuAST: add dotty:sym<.=> Fixes $_ = -1; .=abs; |
07:37 | |
07:46
djinni` left
07:52
djinni` joined
07:57
evalable6__ joined
08:02
evalable6 left
08:22
lizmat left,
gfldex left,
zostay left,
coleman left,
JRaspass left
08:34
lizmat joined,
gfldex joined,
zostay joined,
coleman joined,
JRaspass joined
14:14
MasterDuke left
|
|||
nine | So a few hours into debugging a mis-parse that swallows the orelse part of $x orelse .=new andthen .=new: 43 I finally found the culprit | 15:10 | |
lizmat | andthen? | 15:11 | |
nine | EXPR-reduce tries to access the operator symbol via nqp::ifnull( nqp::atkey( nqp::atkey($op,'OPER'), 'sym'), '') | 15:12 | |
Alas, infixes do not have the common <sym> key anymore. | |||
token infix:sym<after> { <sym> » } became token infix:sym<after> { <.infix-after> » } | 15:13 | ||
lizmat | aha! why did it become that ? | 15:14 | |
not localization related, is it ? | |||
nine | # These tokens replace bare strings so that they can be localized by mixing in tokens with translated localizations. | 15:15 | |
lizmat | ah, I read the regex wrong: yes, indeed localization :-( | 15:17 | |
so: { $<sym>=<.infix-after> » } instead ? | 15:23 | ||
nine | <sym=.infix-after> works | 15:28 | |
lizmat | *phew*, so that's overseeable | 15:36 | |
Geth | rakudo/main: 8 commits pushed by (Stefan Seifert)++
|
15:41 | |
nine | 1093 | 15:42 | |
lizmat | 1092 for me on MacOS | 15:55 | |
17:42
sena_kun joined
|
|||
Geth | rakudo/main: af7acf0c45 | (Stefan Seifert)++ | src/Raku/ast/expressions.rakumod RakuAST: fix logic error in creating curried expression's signature Whether we ought to descend into a WhateverCode when renumbering arguments depends on the operator of the outer expression, not the one used by the WhateverCode. |
18:40 | |
rakudo/main: 081147eeae | (Stefan Seifert)++ | src/Raku/ast/expressions.rakumod RakuAST: also thunk code arguments for thunk type t Operators like xx can also be used to process code objects, e.g. {$_} xx 2 In this case the supplied block has to be thunked. Otherwise the operator will just call the block itself assuming that it's a thunk. |
|||
nine | 1100! | ||
ab5tract | Nice! \o/ | 18:57 | |
That curried expression fix looks like it might have been a pain to surface | |||
nine | Actually it wasn't that bad. The currying code is quite concise nowadays so there couldn't have been many places where we'd go wrong :) Was harder to find out what (* o (* * 2)(* + 1)(3) actually should do | 19:07 | |
lizmat | 1099 for me | 19:23 | |
Geth | rakudo/main: 0b4048f267 | (Stefan Seifert)++ | src/Raku/ast/expressions.rakumod RakuAST: support using bracketed infixes as list infixes |
20:19 | |
nine | With that it should be 1100 even for you :) And 1101 for people on more reasonable OSes :D | 20:20 | |
lizmat | nine: indeed: still I wonder which one I'm missing | 20:30 | |
these are my 1100: gist.github.com/lizmat/590a2d6dcc6...7261804582 could you tell me which one I'm missing ? | |||
I have a severe case of FOMO :-) | |||
nine | Huh....it says it's t/spec/S12-attributes/augment-and-initialization.rakudo.moar | 20:34 | |
lizmat | when run manually, it passes | 20:40 | |
Geth | rakudo/main: 1a616bfb13 | (Stefan Seifert)++ | src/Raku/Grammar.nqp RakuAST: bring back explanations for Syntax::Confused errors |
20:53 | |
20:55
sena_kun left
21:21
MasterDuke joined
|