| 29 Jul 2025 |
 emily | I notice that the derivation in Nixpkgs already did this devdoc stuff and I just wasted a non-trivial amount of time :) | 20:14:37 |
| emily | is there no systematic sync between those two? | 20:14:44 |
| emily | I'll leave what I have up since it's an improvement over HEAD, but it seems like it would be good to minimize the drift between those two to avoid redundant work. | 20:16:09 |
 Marie | hi, I do a applyPatches thing to apply PR's to my nixpkgs and I've been getting really annoyed that lix copies nixpkgs into the store on every invocation, even though it is already in the store.
From googling around this seems to be a known issue. If I call my patched thing "source" the problem goes away.
Does anyone know if there is a good reason to only allow "source" here: https://git.lix.systems/lix-project/lix/src/commit/93acdd40f6c3eab1a25ea03d7a160433ed551677/lix/libfetchers/path.cc#L138 | 21:57:17 |
| Marie | Seems to work fine when I remove the restriction, but I don't know much about the codebase, so no idea if this could break something | 22:11:36 |
 jade_ | i think there is an issue about this somewhere | 23:56:34 |
| jade_ | i am not sure lol | 23:56:40 |
| jade_ | i think you should just change your derivation to be named "source" tbh | 23:56:58 |
| jade_ | iirc if you put an input-addressed path into the path:... flake fetcher it throws an assert in lix internals. flakes, man. | 23:57:27 |
| 30 Jul 2025 |
| jade_ | * i think there is an issue about this somewhere because i filed it in cppnix 5 years ago | 00:16:00 |
| jade_ | * i think there is an issue about this somewhere because i filed it in cppnix nearly 5 years ago | 00:16:06 |
| jade_ | * i think there is an issue about this somewhere because i filed it in cppnix nearly 5 years ago and we probably imported it | 00:16:12 |
 llakala | In reply to @jade_:matrix.org
was i working on a new one? i don't remember having done so you mentioned wanting to do it in the "You don't have to use Nix to manage your dotfiles" blogpost | 10:53:43 |
| llakala | 
Download Screenshot_2025-07-30-06-54-19-29_3aea4af51f236e4932235fdada7d1643.jpg | 10:54:25 |
 WeetHet | auto prefix = useChroot ? chrootRootDir : tmpDir; is incorrect on macOS | 15:50:17 |
| WeetHet | This is correct however:
#if __linux__
/* If we are using no sandboxing, we still need to use the physical prefix. */
useChroot ? logicalTargetPath : prefix + logicalTargetPath;
#else
prefix + logicalTargetPath;
#endif
| 15:50:51 |
| WeetHet | I think this patch works for me:
diff --git a/lix/libstore/build/local-derivation-goal.cc b/lix/libstore/build/local-derivation-goal.cc
index e0e14b09f..33f46dc93 100644
--- a/lix/libstore/build/local-derivation-goal.cc
+++ b/lix/libstore/build/local-derivation-goal.cc
@@ -918,7 +918,11 @@ void LocalDerivationGoal::setupConfiguredCertificateAuthority()
{
if (settings.caFile != "") {
if (pathAccessible(settings.caFile)) {
+#if __linux__
auto prefix = useChroot ? chrootRootDir : tmpDir;
+#else
+ auto prefix = tmpDir;
+#endif
debug(
"rendering visible configured CA '%s' in the builder (prefix directory: '%s')",
settings.caFile,
@@ -1249,8 +1253,6 @@ void LocalDerivationGoal::runChild()
"non-functional."
);
}
-
- setupConfiguredCertificateAuthority();
}
for (auto & i : ss) pathsInChroot.emplace(i, i);
@@ -1406,11 +1408,8 @@ void LocalDerivationGoal::runChild()
}
#endif
- if (!useChroot) {
- /* When chroot is not used, FODs still requires a CA to be available as well. */
- if (!derivationType->isSandboxed()) {
- setupConfiguredCertificateAuthority();
- }
+ if (!derivationType->isSandboxed()) {
+ setupConfiguredCertificateAuthority();
}
if (chdir(tmpDirInSandbox.c_str()) == -1)
| 15:51:36 |
| WeetHet | chrootRootDir is "" on macOS even with sandbox enabled | 15:52:19 |
 raitobezarius | can you submit this? | 15:53:40 |
| WeetHet | My laptop is dead and will be for the next couple hours but after that sure | 16:05:46 |
| WeetHet | You can also beat me to it and submit it yourself | 16:05:57 |
| WeetHet | I don't care about attribution | 16:06:20 |
| emily | what am I holding wrong that just setup fails with lix/lix-doc/meson.build:1:7: ERROR: Subproject exists but has no Cargo.toml file. | 16:52:06 |
| emily | (yes I have been doing non-incremental builds all this time) | 16:52:44 |
| jade_ | oh it's related yes. one of the things i want to know is "how long did this eval spend copying to the store and why/where" | 17:46:55 |
| jade_ | i have no idea. is this some gitignored nonsense state? | 17:47:28 |
| emily | I don't believe so. but it is Jujutsu :) | 17:56:27 |
| emily | lix/lix-doc/Cargo.toml does in fact exist. | 17:56:37 |
| emily | but it's complaining about rnix | 17:56:55 |
| emily | sorry I should have included more output | 17:57:00 |
