| 17 May 2024 |
 puck | we can have compat pkg-config entries if we really care | 00:00:05 |
 jade_ | we should not have compat pkg-config | 00:00:14 |
| jade_ | because we don't have api compat | 00:00:18 |
| jade_ | we would just like to not explode peoples actual application code that badly for no reason i guess | 00:00:31 |
| puck | In reply to @jade_:matrix.org
maybe the c api should be in nix/ still oh, the C api is in the root of -isystem anyways; but i think we can discuss that; i might do a 1:1 with horrors about this some point soon | 00:01:01 |
| puck | (i'd like to get the C api in soonish because it'd mean we can run zilch on it) | 00:01:13 |
| puck | (mind you, https://puck.moe/git/zilch/tree/core/nix/default.nix?id=ae0f39a7b7eeef67442f96dd461648369fc47e0f#n3) | 00:01:34 |
| jade_ | anyway i guess im inclined to explode C++ API really hard on purpose. but we can stage the move include/nix/ to include/lix/ to a second change? idk what you prefer | 00:01:47 |
| puck | * (mind you, i miight've already done this~) | 00:01:51 |
 Qyriad | no one external is using the C API right now I'm not sure if it matters or not | 00:03:52 |
| puck | In reply to @puck:puck.moe
i think we'll likely break the C api but i have plans to try and convince robert (we can probably have a convincing point if we have a nice API and proper rust/python bindings) | 00:04:52 |
 delroth | anyway, https://gerrit.lix.systems/c/lix/+/1151 enough rabbit hole for tonight | 00:06:53 |
| delroth | also first bug I fix in Lix which is actually a Nix bug and not a regression in Lix /hides | 00:07:27 |
 julia | oh yeah for Gerrit if a change is all approved who presses the submit button? is it still one of the reviewers or is it supposed to be me | 00:13:19 |
| puck | the author, preferably | 00:13:47 |
| jade_ | author, yes | 00:14:04 |
| jade_ | it prevents the classic github blunder of not being able to decide as the submitter if you want to action a review | 00:14:34 |
| jade_ | unless you have merge rights. which is goofy! | 00:14:42 |
| jade_ | (at my last employer people had merge rights and the practice was to self-merge with an approval for exactly this reason) | 00:15:05 |
| jade_ | * (at my last employer people had github merge rights and the practice was to self-merge with an approval for exactly this reason) | 00:15:11 |
| julia | ah | 00:15:13 |
| Qyriad | This also means the author has the final say in whether their code is merged | 00:15:50 |
| julia | Gerrit is so nice it makes so much more sense than GitHub/hitlab merge request workflow I can't believe I "got taught it" (read; died having group members learn it) at uni | 00:16:11 |
| Qyriad | I know right | 00:16:34 |
| delroth | every time I wonder if the gerrit workflow is actually better or if I've just been conditioned by 9+ years of Critique | 00:17:30 |
| delroth | and then I think for 15s and realize that no, github actually really just sucks | 00:17:38 |
| puck | yeah | 00:17:42 |
| puck | like, gerrit reviews are so good, they nailed all of it by just .. replicating the email workflow but in a web ui | 00:17:56 |
| delroth | disagree: email workflow doesn't have the "resolved" bit | 00:18:23 |
| delroth | which makes it greatly inferior | 00:18:28 |
