|
🦋 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. |
|||
|
00:43
guifa joined
04:26
guifa left
11:36
sjm_ joined
11:38
sjm__ left
13:23
sjm_ left
15:09
finanalyst joined
15:47
donaldh left
|
|||
| ab5tract | nine: we seem to be going through a moment of distros changing their packaging standards en masse | 16:17 | |
| What worked 8 years ago clearly wasn’t up to Debian standards or else timo++ wouldn’t have had to go through a hell or two to keep us available | 16:18 | ||
|
16:19
finanalyst left
|
|||
| ab5tract | timo: I think DB::SQLite already provides such a pool | 16:19 | |
| nine | ab5tract: but what changed? | 16:25 | |
| ab5tract | This requirement for reproducible builds seems to have come from nowhere, to be honest | 16:28 | |
| But now you can’t stay in without it | 16:29 | ||
| nine | No, it really hasn't. It's somewhere between 5 and 10 years old. Can't say more precisely without checking | ||
| Where do you think I got the term "reproducible builds" from? | |||
| lizmat | fwiw, I think the fact that we do not have reproducible builds yet (apparently), is still a development issue as well (for me at least) | 16:30 | |
| ab5tract | I never noticed you using it, so… | ||
| nine | Well the test file is called t/02-rakudo/reproducible-builds.t :) | ||
| ab5tract | All I know is that I never saw a ticket about reproducible builds until last year | ||
| So then what has changed is that it went from soft to hard requirement | 16:31 | ||
| That’s all that has changed, afaik | |||
| nine | If it's a hard requirement, then there's no immediately applicable solution. | ||
| Because there are bugs that make our build non-reproducible. | 16:32 | ||
| ab5tract | But maybe there is something collateral fallout for distros related to that crazy supply chain hack attempt from last year | 16:33 | |
| nine | * One may get around that for packaged module files by just not precompiling them at all, but the same bugs affect rakudo itself | ||
| ab5tract | The xz one | ||
| Hmmm, that’s interesting. I know timo worked very hard on keeping us (rakudo) in the repos. I had assumed he had fully addressed the issue. But since you are saying that this isn’t actually possible, I realize I don’t have a clear picture on what work needed to be done to satisfy the distro folks | 16:36 | ||
| But for sure things have changed in a bunch of distros in what felt like overnight | 16:37 | ||
| nine: I understand your position about the yyjson patch. I want to gauge the veracity of your opposition before I add the same for the JVM | 16:55 | ||
| nine | I'm not gonna rage quit over that :D | 17:12 | |
| ab5tract | Ok :) | 17:28 | |
| I do agree in principle though. It would obviously be preferable if we could speed up the whole runtime instead of the is sort of special casing via libraries | 17:29 | ||
|
17:53
donaldh joined
20:03
librasteve_ joined
22:41
guifa joined
23:23
librasteve_ left
|
|||