Sender | Message | Time |
---|---|---|
29 Nov 2024 | ||
m1-s changed their profile picture. | 02:28:36 | |
m1-s changed their profile picture. | 02:29:54 | |
30 Nov 2024 | ||
@julius:mtx.liftm.de left the room. | 12:14:45 | |
1 Dec 2024 | ||
shawn8901 joined the room. | 00:09:33 | |
mmkaram joined the room. | 21:32:39 | |
2 Dec 2024 | ||
Madalin joined the room. | 11:33:59 | |
entheogenesis | Hello. I am attempting to build a Rust application which uses a crate named
I noticed that this file is excluded in the Cargo.toml file: https://github.com/liushuyu/xdelta3-rs/blob/7bca8bc72548bd88a92a308dcd01225c3e842bc7/Cargo.toml#L10-L12 Does anyone know what I can to do fix this? | 20:30:59 |
tnias | Do you have | 20:35:29 |
entheogenesis | tnias: Yes. It doesn't make a difference. | 20:48:31 |
tnias | entheogenesis After adding fetchSubmodules the hash needs to be updated. If you don't update the hash, it will continue to use the old src (the one, that matches the old hash). | 20:50:11 |
tnias | Not sure, if that is something you forgot, but it's something slightly unexpected I ran into ^^ | 20:51:14 |
entheogenesis | I invalidated the hash, but it remained the same in the end:
Specifically, I invalidated it by changing the first character from | 20:56:25 |
tnias | That's strange. I got differen hashes.
But more importantly: the py-compile is a symlink to some automake thing. oof | 21:02:25 |
tnias | * That's strange. I got differen hashes.
But more importantly: the py-compile is a symlink to some automake thing. oof
| 21:02:40 |
entheogenesis | Oh, good find! I am not trying to build xdelta3 directly, as it is a dependency of an application. Here is what I am attempting to do. | 21:19:39 |
entheogenesis | I suppose what I want to know is, if the
Why is nixpkgs trying to copy an excluded file? | 21:37:49 |
tnias | I am afaid I can't help with that. | 21:47:51 |
entheogenesis | I appreciate you looking. Is there a better Matrix room in which to pose this question? | 21:56:36 |
tnias | That sounds like an issue specific to how buildRustPackage is implemented. I believe the room is correct. I am just not very experienced with the rust ecosystem in nix, yet. :) | 22:02:44 |
3 Dec 2024 | ||
getchoo changed their profile picture. | 06:06:31 | |
4 Dec 2024 | ||
@orzklv:matrix.org | I’m trying to build a tauri application on 24.11 and running into this issue:
| 08:26:39 |
@orzklv:matrix.org | is there any chance that this is because of pkgs.lib.makeLibraryPath openssl_3? | 08:28:03 |
@orzklv:matrix.org | yup, it is. thanks for ya’ll attention. | 08:32:08 |
entheogenesis | K900: Do you happen to know what I need to do fix the aforementioned build issue? You said this:
| 21:03:14 |
entheogenesis | * K900: Do you happen to know what I need to do fix the aforementioned build issue? You said this:
But I do not know how to do that | 21:03:20 |
entheogenesis | * K900: Do you happen to know what I need to do fix the aforementioned build issue? You said this:
But I do not know how to do that. | 21:03:29 |
5 Dec 2024 | ||
djacu | What's the current best practice for using buildRustPackage with a private registry that requires authentication? I am working with a private Artifactory instance and have been able to get builds working using the cargoLock.lockFile by vendoring import-cargo-lock.nix and modifying the registry URL. This works though because of a secret in the netrc file that the sandbox has access to. It does not work when using cargoHash or cargoVendorDir because cargo gets involved and is looking for registry credentials either via the config.toml in CARGO_HOME or the environment variable that stores the token for a registry. | 01:29:23 |
Danny Gerhardt joined the room. | 09:58:19 | |
Morj | I've been lazily trying to solve this for two years now. For now the solution is to vendor all private deps as git submodules, and patch the project to use those instead. This separates checkout of private repos and build with everything publicly available, but honestly it's a silly solution | 11:47:14 |
Morj | * I've been lazily trying to solve this for two years now. For now my solution is to vendor all private deps as git submodules, and patch the project to use those instead. This separates checkout of private repos and build with everything publicly available, but honestly it's a silly solution | 11:47:21 |