fix(publish): Move .crate out of final artifact location
#15915
+12
−26
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
|
r? @weihanglo rustbot has assigned @weihanglo. Use |
rustbot
added
Command-package
S-waiting-on-review
epage
commented
Sep 3, 2025
Comment on lines
+163
to
+164
| let dir = ws.build_dir().join("package").join("tmp-crate"); | ||
| let dst = dir.open_rw_exclusive_create(&filename, gctx, "package scratch space")?; |
There was a problem hiding this comment.
As this will cause a behavior change, Ive marked this as a draft while we wait for #15910 to propagate to users so they can prepare.
weihanglo
reviewed
Sep 26, 2025
There was a problem hiding this comment.
For myself, if we want an intermediate period, I feel like more than one release is probably needed.
This comment has been minimized.
This comment has been minimized.
When `target_dir == build_dir`, ensure `cargo publish` doesn't put intermediate artifacts in the final artifact location of `cargo package`. If anyone was relying on this behavior of `cargo publish`, it will break them. We could avoid this and instead consider the location change to be part of the opt-in of using `build-dir` (until we make it opt-out). Note that we expect to be able to change the layouf of content written to `build-dir` even if users aren't opting in. On the other hand, this will help identify people relying on intermediate artifacts. While there aren't any performance benefits to this, it consolidates all of the uplifting logic and avoids dealing with overlapping `target_dir` and `build_dir`. We could optimize this further by doing a `rename` and only doing a copy if that fails.
epage
force-pushed
the
publish-move-crate
branch
from
d742813 to
8cc523d
Compare
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
When
target_dir == build_dir, ensurecargo publishdoesn't put intermediate artifacts in the final artifact location ofcargo package.What does this PR try to resolve?
In #15910, users could identify that
.cratefiles fromcargo publishare not final artifacts by setting a custombuild-dir. This extends that to all users, ie whenbuild-dir = target-dir(the default currently), making it clear that these files are internal.This also cleans things up by consolidating all of the uplifting logic and avoids dealing with overlapping
target_dirandbuild_dir.How to test and review this PR?
Notes
We could optimize this further by doing a
renameand only doing a copy if that fails, effectively arename_or_copyas opposed to ourhardlink_or_copywe normally use for uplifting. The difference is that we don't do change tracking for.cratefiles but fully re-generate, so we don't benefit from keeping the.cratearound in the original location.