🦋 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. |
|||
00:59
buildable6 left
01:00
buildable6 joined
02:00
buildable6 left
02:01
buildable6 joined
03:01
buildable6 left
03:02
buildable6 joined
03:48
Geth left
03:49
RakuIRCLogger left
03:50
lizmat left,
[Tux] left
03:58
lizmat joined
04:02
buildable6 left,
[Tux] joined
04:03
buildable6 joined
04:07
buildable6 left,
buildable6 joined
04:12
buildable6 left,
buildable6 joined
05:03
buildable6 left
05:04
buildable6 joined
06:04
buildable6 left
06:07
buildable6 joined
07:07
buildable6 left
07:10
buildable6 joined
07:15
buildable6 left,
buildable6 joined
08:10
buildable6 left
08:12
buildable6 joined
08:50
sena_kun joined
09:12
buildable6 left
09:15
buildable6 joined
09:26
Geth joined
10:15
buildable6 left
10:17
buildable6 joined
10:22
buildable6 left,
buildable6 joined
10:39
finanalyst left
11:17
buildable6 left
11:21
buildable6 joined
|
|||
[Coke] | nemokosch - i would love to replace the full expanded bootstrap with templating, so I can say "put a box here with this title" and not have to cut and paste the html everywhere. I will avoid doing this as I know you're working on updating the templating. But once we have new templating, i'd love to use more of it. | 12:07 | |
(on raku.org0 | |||
Looks like there was an effort to go to Bootstrap 4 that didn't complete. 5 is out now, opened a ticket for that as well. (another change that can probably wait until after your PR is complete) | 12:08 | ||
12:21
buildable6 left,
buildable6 joined
|
|||
nemokosch | yes, perhaps this could be done better. The same includes/ system is probably enough for that | 12:23 | |
[Coke] | making small content updates, I see each page is using slightly different styling, having it templated would just mean we don't have to think about it. :) | 12:24 | |
next big content things: clean up resources page, add RSC to community page. | 12:25 | ||
lizmat | bisectable6: $_="0X";say m/\d/,m/X/ | 13:15 | |
bisectable6 | lizmat, Will bisect the whole range automagically because no endpoints were provided, hang tight | ||
lizmat, ¦6c (73 commits): «「X」「X」» | 13:16 | ||
lizmat, Nothing to bisect! | |||
Geth | rakudo/main: ec57a59fbb | (Elizabeth Mattijsen)++ | src/Raku/ast/code.rakumod RakuAST: make sure $/ is deconted This fixes #5454 for RakuAST. Leaving it as an exercise for someone else to fix this in the legacy grammar (although I don't think it's really necessary, given that this bug existed since at least Christmas |
||
13:21
buildable6 left
13:24
buildable6 joined
|
|||
ab5tract_ | nine_: 'Cannot compile lookup of attributes yet' <— what would compiling lookups of attributes look like? | 13:30 | |
13:36
vrurg_ is now known as vrurg
14:24
buildable6 left
14:26
buildable6 joined
15:26
buildable6 left
15:27
buildable6 joined
|
|||
ab5tract_ | nine_: also looking for feedback on what could replace the container initializer for attributes with where blocks. Subsets cannot be instantiated, so the normal approach of creating an ApplyPostfix that calls new on a resolved constant declaration of the type doesn’t work. | 16:22 | |
Or maybe that part is actually irrelevant (ie, that codepath shouldn’t be run) and there’s some other tiny bit flip missing in the meta-object setup for attributes | 16:25 | ||
16:27
buildable6 left
|
|||
ab5tract_ | though it looks like everything that could possibly be set for a new attribute is set.. | 16:28 | |
16:30
buildable6 joined
17:30
buildable6 left
17:33
buildable6 joined
17:37
buildable6 left,
buildable6 joined
|
|||
[Coke] | anyone want the cast warnings that windows is complaining about? | 18:20 | |
lizmat | would those be MoarVM build warning? | 18:21 | |
s | |||
[Coke] | Right, the c level ones. | 18:22 | |
I am not capturing them right now but could re-run and do so. I think we do this every year or so and they don't get fixed. | 18:23 | ||
lizmat | well, the MoarVM build isn't clear of warnings on Linux / MacOS either | 18:24 | |
[Coke] | ah. OK, I assume it'll be easier to fix linux first, so I won't worry about windows for now (assuming that'll also cover some of the same ones). lizmat++ | 18:27 | |
ugexe | the warnings are also visible in rakudos ci from what i recall | 18:29 | |
so you can easily link people to them if you can find them in there | 18:30 | ||
18:33
buildable6 left
18:35
buildable6 joined
|
|||
[Coke] | ugexe: You made some progress on long windows paths but it still needed something, yes? | 18:57 | |
I'm assuming that's why JSON::Fast is dying on install right now | 18:58 | ||
ugexe | I don't really have anything useful to show, but I did prove locally that if we pass the paths to moarvm with the \\?\ prefix that long paths work | 19:06 | |
i got lost trying to implement it properly though | |||
to properly implement it requires IO::Spec::Win32's various methods (like the one for splitting and concating) and the regex to be updated to understand that prefix so that when we pass moarvm "\\?\C:\foo\bar".IO.absolute, it doesn't misparse the string (it doesn't know about \\?\) and reconstruct the absolute path incorrectly | 19:08 | ||
the other issue being when using \\?\ we have to normalize paths from / to \ which we don't do in every single place (we do it some places, but not all) | 19:10 | ||
when is the next ground of tpf grants? i could probably make time to do in that scenario | 19:27 | ||
[Coke] | "we're always accepting applications" | 19:33 | |
19:33
elcaro left,
elcaro joined
|
|||
[Coke] | no one here is on the committee anymore, I think. | 19:33 | |
... and it doesn't look like they are posting the "we'll review the next round..." blog posts anymore. | 19:35 | ||
19:35
buildable6 left
|
|||
[Coke] | might be worth tagging codesections and seeing if he can follow up with whoever the GC chair is these days (I think Crome stepped down) | 19:35 | |
19:36
buildable6 joined
|
|||
[Coke] hurls news.perlfoundation.org/post/may-2...plications by Saif Ahmed, looks like it's just open ended now? | 19:37 | ||
I will heartily endorse your proposal to fix this issue. | |||
20:36
buildable6 left
20:39
buildable6 joined
21:27
finanalyst joined
21:39
buildable6 left
21:43
buildable6 joined
22:43
buildable6 left
22:45
buildable6 joined
22:49
buildable6 left,
buildable6 joined
22:53
buildable6 left,
buildable6 joined
23:45
buildable6 left,
sena_kun left
23:48
buildable6 joined
|