|
🦋 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. |
|||
|
08:50
finanalyst joined
11:42
finanalyst left
|
|||
| lizmat | could someone confirm for me that JSON::Fast currently does not install on Windows ? | 14:34 | |
| github.com/lizmat/App-Rak/actions/...8054089893 | |||
| [Coke] | how recent a raku do I need to test with? :) | 14:44 | |
| I already have it installed on windows, doing a forced install | 14:45 | ||
| just did an install on git bash on windows with raku v2023.10+ | 14:50 | ||
| looks like I'll need to test with something more recent. | 14:51 | ||
| lizmat | well, I guess it's more important to test it with the most recent version of zef | 15:15 | |
| [Coke] | updating rakudo, will update zef, then try to reinstall | 15:17 | |
| (Been a while, forgot how I was doing the builds on windows here) | |||
| seeing lots of new(?) compier warnings in winbase.h that I don't recall seeing before | 15:19 | ||
| (which is something on the OS, not rakudo) | |||
|
15:31
vrurg joined
15:34
vrurg_ left
15:36
MasterDuke left
|
|||
| lizmat | looks like Github CI is using zef 0.22.0 rather than 0.22.1 | 15:37 | |
| [Coke] | I get the same error with rakudo-latest, zef latest, and trying to install JSON::Fast 0.19 | 16:00 | |
| I'm on a win 10 box. | 16:01 | ||
|
16:07
sena_kun joined
|
|||
| ugexe | what is the actual failure? you'll have to use --verbose or --debug | 16:09 | |
| nevermind, see it in lizmats link | |||
| ye old path length issue | |||
| it looks to be using the `C:\Users\RUNNER~1\AppData\Local\Temp\.zef` temp directory and not /tmp, so maybe that isn't set correctly? | 16:10 | ||
| [Coke] | agreed - this avoids the issue until the next path increase: | 16:11 | |
| mkdir C:\tmp; SET TEMP=C:\tmp; zef install JSON::Fast | |||
| I had a similar tmp dir here | 16:12 | ||
| lizmat | I wonder whether the path length issue has to do with the fact that the module lives on CPAN ? | 16:13 | |
| [Coke] | Anything we do to avoid it is just a stopgap, need to fix the actual root cause or it'll keep showing up | 16:15 | |
| (in fact, use C:\t to avoid it for two more characters) | |||
| I know we had some WIP on this, I ended up avoiding needing it as a user by moving my scripts that were dying to it to run on mac instead. | 16:17 | ||
| lizmat | notable6: weekly | 16:19 | |
| notable6 | lizmat, 1 note: 2024-07-27T18:23:16Z <antononcube>: youtu.be/LmLk9UB4_Ak | ||
| lizmat | notable6: weekly reset | ||
| notable6 | lizmat, Moved existing notes to “weekly_2024-07-29T16:19:55Z” | ||
| ugexe | lizmat: it has to due with every part of those paths being relatively long | 16:28 | |
| lizmat | but I have other modules with long names that work fine on Windows CI | 16:29 | |
| ugexe | what matters is the total path length, and the path length is also influenced by the compunit repository that backs it, i.e. what the path prefix is | 16:30 | |
| lizmat | and yet another Rakudo Weekly News hits the Net: rakudoweekly.blog/2024/07/29/2024-...drianally/ | ||
| ugexe | for instance how we use sha1s in many spots when something shorter might technically work | 16:31 | |
| of course the real fix here is to make long paths work on windows | |||
| github.com/MoarVM/MoarVM/pull/1776 for instance | 16:32 | ||
| lizmat | we definitely have a too low bus factor there :-( | 16:35 | |
| Geth | rakudo/main: 6e6ce83a5c | (Elizabeth Mattijsen)++ | tools/contributors.raku Include contributions to doc website as well This would have included Richard Hainsworth as a contributor to the 2024.07 release |
17:06 | |
| lizmat | jdv: ^^ fyi | ||
| jdv | cool | 17:27 | |
| i didn't know there was that distinction | 17:28 | ||
| [Coke] | lizmat++ finanlyst++ | 17:48 | |
|
19:19
finanalyst joined
19:55
dawids joined
|
|||
| Geth | rakudo/rakuast-phase-cleanup: c987ce13d1 | (Stefan Seifert)++ | src/Raku/ast/variable-declaration.rakumod RakuAST: fix generated accessor methods looking at wrong signature Current scope for the method and it's signature's BEGIN handlers was the surrounding scope (usually the package) of the declaration, not the method itself. |
19:58 | |
|
19:58
vrurg_ joined
|
|||
| nine | Turns out, I was completely off in my earlier attempt to fix role R { my $.e; } in rakuast-phase-cleanup. Sometimes a two month break and a fresh start can help a lot :) | 19:59 | |
|
20:01
dawids left
20:03
vrurg left
|
|||
| [Coke] | ^_^ | 20:27 | |
|
21:19
sena_kun left
22:16
finanalyst left
|
|||