Use `i32` rather than `usize` as "default integer" in library template
`usize` was renamed from `uint` in order to convey that it's not the "default integer type". Guide new users to use `i32` instead of `usize`.
`usize` was renamed from `uint` in order to convey that it's not the
"default integer type". Guide new users to use integers with specific
bit width instead of `usize`.
fetch specific commits even if the github fast path fails
### What does this PR try to resolve?
This PR fixes#13555, which describes a regression from 1.64.0 to 1.65.0 where the inability to fetch commit information from api.github.com (the "GitHub fast path") silently changes Cargo's behavior.
Cargo can fetch a specific Git commit from a remote without having to fetch all refs. Prior to #10807, this functionality required a repository hosted on github.com and providing the full commit hash (usually available from the Cargo.lock); after that change, any revision (including abbreviated revisions) that could be resolved by GitHub's API could be fetched directly. However, this logic requires the "GitHub fast path", which was not intended to be robust, to successfully return the resolved commit hash; if a client is currently rate-limited by api.github.com (very common in CI and shared cloud / corporate environments) this fails and Cargo falls back to fetching all refs.
Usually this is not noticeable. However, GitHub allows fetching commits that are related to the repository but not actually part of any of its refs, including commits pushed to a fork. This results in the same command working fine in some environments where api.github.com is accessible, and not working in other environments that are rate-limited, which is very confusing and difficult to debug.
This change adds another branch to cover the regression case: if we are going through the GitHub fast path with a full commit hash, return early indicating that we need to fetch it. (Previously: ~~when the GitHub fast path was unsuccessful, the user is not using the unstable shallow clone options, and we have a full commit hash and expect to be able to fetch it directly because we know it's a github.com repository.~~)
### How should we test and review this PR?
I have been testing this PR by temporarily adding a `0.0.0.0 api.github.com` entry to my `/etc/hosts`, which causes the GitHub fast path to always fail, then running:
```
target/release/cargo install --git https://github.com/haha-business/unstable-test-repo.git --rev c9040898c9183ddbb9402dcbf749ed06d6ea90ad
```
This refers to [a particular commit on a fork of the repo](c9040898c9) which won't be found by the fallback path or current Cargo.
**Note** that you will need to delete `~/.cargo/git/checkouts/unstable-test-repo-*` and `~/.cargo/git/db/unstable-test-repo-*` after a successful run with this change in order to reproduce the broken behavior of the current release.
I am having trouble getting the test suite to run at all on my system so I haven't experimented with writing a specific test for this case, but I probably should.
### Additional information
This uses the same logic as the unstable shallow clone support to detect if the revision is a full commit hash. This is not compatible with SHA-256 commit hashes; `git2::Oid` specifically expects a 40-character hexadecimal string. Given that the change introducing this bug was meant to future-proof SHA-256 support (despite only doing so for GitHub repositories), it might be good to make the logic more explicit within Cargo and allow either 40- or 64-character hex strings.
I wanted to keep this change focused on the regression fix, but in testing, pretty much every Git repository I could think of (including non-forges, like git.kernel.org and some repositories I host on my own infrastructure with cgit) supports fetching directly from a commit, so it would be ideal to eventually relax the GitHub requirement for this functionality. However, it would need some sort of fallback logic because I suspect the HTTP [dumb protocol](https://git-scm.com/book/en/v2/Git-Internals-Transfer-Protocols) doesn't support commit references, and I haven't researched when this functionality was added to the smart protocol.
Load `libsecret` by its `SONAME`, `libsecret-1.so.0`
The library name `libsecret-1.so` is only found in `-dev` or `-devel` packages on many distros, and it's usually only a symlink. For example, this is what I have on Fedora:
/usr/lib64/libsecret-1.so -> libsecret-1.so.0
/usr/lib64/libsecret-1.so.0 -> libsecret-1.so.0.0.0
/usr/lib64/libsecret-1.so.0.0.0
$ rpm -qf /usr/lib64/libsecret*
libsecret-devel-0.21.4-2.fc40.x86_64
libsecret-0.21.4-2.fc40.x86_64
libsecret-0.21.4-2.fc40.x86_64
The middle `libsecret-1.so.0` filename matches the actual `SONAME` in the library, which you can see with `readelf -d`, and this is what programs built directly against this library will use. It's a better choice for dynamic loaders too, so devel packages aren't needed.
docs(ref): Simplify check-cfg build.rs docs
### What does this PR try to resolve?
After looking over the rust documentation updates, decided to look over Cargo's build.rs documentation.
- Removed link to the blog post as that is not evergreen documentation
- Collapsed down the example to put more emphasis on the more complete example
### How should we test and review this PR?
### Additional information
Make `git::use_the_cli` test truly locale independent
### What does this PR try to resolve?
The current `git::use_the_cli` test part the output of Git but that output is locale dependant, making it fail on my non-english system.
Specifically this part of the test is local-dependent:
```
From [..]
* [new ref] [..] -> origin/HEAD[..]
```
Adding the `LC_ALL=C` env to the cargo invocation solve the issue, making the test locale independent.
### How should we test and review this PR?
Trying putting a different `LC_ALL` locale and the test will fail without this PR and will not fail with my PR.
Silence warnings running embedded unittests.
The `util::toml::embedded::test_expand` unittests were printing warnings directly to stderr, circumventing the libtest override. They were displaying ``warning: `package.edition` is unspecified, defaulting to `2021` ``. That warning doesn't look particularly important for those tests, so this changes it to swallow the output.
We may want to have some kind of utility function for generating a GlobalContext for testing purposes that does this, since this is a repeated pattern.
Fix warning output in build_with_symlink_to_path_dependency_with_build_script_in_git
The test `build_with_symlink_to_path_dependency_with_build_script_in_git` was emitting a large warning block (in my case, about init.defaultBranch) because it was running `git` without filtering its output. It's not clear to me why this test was shelling out to `git` instead of using the built-in test support functions. From what I can tell, this should be exactly equivalent, and silences the warning output.
Temporarily fix standard_lib tests on linux.
This fixes the standard_lib tests which are broken in the latest nightly. The latest nightly now requires rust-lld to be in the sysroot for x86_64-unknown-linux-gnu. This broke these tests which were trying to verify that the standard library is not required. This temporarily removes this validation, but we should have some way of enforcing it (https://github.com/rust-lang/wg-cargo-std-aware/issues/31).
cc https://github.com/rust-lang/rust/issues/125246
The library name `libsecret-1.so` is only found in `-dev` or `-devel`
packages on many distros, and it's usually only a symlink. For example,
this is what I have on Fedora:
/usr/lib64/libsecret-1.so -> libsecret-1.so.0
/usr/lib64/libsecret-1.so.0 -> libsecret-1.so.0.0.0
/usr/lib64/libsecret-1.so.0.0.0
$ rpm -qf /usr/lib64/libsecret*
libsecret-devel-0.21.4-2.fc40.x86_64
libsecret-0.21.4-2.fc40.x86_64
libsecret-0.21.4-2.fc40.x86_64
The middle `libsecret-1.so.0` filename matches the actual `SONAME` in
the library, which you can see with `readelf -d`, and this is what
programs built directly against this library will use. It's a better
choice for dynamic loaders too, so devel packages aren't needed.
Add special `check-cfg` lint config for the `unexpected_cfgs` lint
### What does this PR try to resolve?
This PR adds a special `check-cfg` lint config for the `unexpected_cfgs` lint, as it was decided by T-cargo (in today's meeting).
The goal of this lint config is to provide a simple and cost-less alternative to the build-script `cargo::rustc-check-cfg` instruction.
```toml
[lints.rust]
unexpected_cfgs = { level = "warn", check-cfg = ['cfg(foo, values("bar"))'] }
```
### How should we test and review this PR?
I recommand reviewing commit by commit; and looking at all the new tests added in `check_cfg.rs`, I tried making them as exhaustive as I could, many of them are very similar to their non-config counterpart.
### Additional information
I didn't add *(actually removed from the 1st version of this PR)* the possibility to omit the `level` field if `check-cfg` is specified, https://github.com/rust-lang/cargo/pull/13913#discussion_r1600609229.
Regarding the implementation, I tried making it is as straight forward as possible, nothing over-engineered or complex.
r? `@epage` (or `@weihanglo` maybe)
Those new tests tries to be as exhaustive as possible while being
reasonable in the number of them. To do so we try to check for
check/doc/test/build-script/features with a the `check-cfg` config.
Many of those tests are very similar to their non-config counterpart.
This permits things like this to be recognized and passed to
rustc/rustdoc.
```rust
[lints.rust]
unexpected_cfgs = { level = "warn", check-cfg = ["cfg(foo)"] }
```
test: set safe.directory for git repo in apache container
### What does this PR try to resolve?
Failure in container tests due to a new version of `git` package in Alpine Linux Package repository.
See also <https://github.com/rust-lang/cargo/pull/13913#issuecomment-2113712049>
### How should we test and review this PR?
Alpine Linux Package repository 3.19 just bumped git package to 2.43.4 from 2.43.0.
The docker image `httpd:2.4-alpine` we use in container tests failed due to the git version bump.
The `httpd` log shown that
```
240.10.0.1 - - [16/May/2024:03:52:36 +0000] "GET /repos/bar.git/info/refs?service=git-upload-pack HTTP/1.1" 500 -
[16/May/2024:03:52:36 +0000] 240.10.0.1 TLSv1.3 TLS_AES_256_GCM_SHA384 "GET /repos/bar.git/info/refs?service=git-upload-pack HTTP/1.1" -
fatal: detected dubious ownership in repository at '/repos/bar.git'
To add an exception for this directory, call:
git config --global --add safe.directory /repos/bar.git
```
git/git@f4aa8c8bb1 is likely the commit causing problems.
So I ended up adding `git config --system --add safe.directory '*'` to the Dockerfile of apache container.
Note that we need `--system` instead of `--global` because `httpd` are running under the other user www-data, not root.
refactor: misc refactors for `ops::resolve`
### What does this PR try to resolve?
This is a preparation for another `-Zpatch-files` experiment,
so that the future PR can move things around easier without too many conflicts.
### How should we test and review this PR?
Generally they shouldn't affect anything existing behavior.
a6230e348b786420e83cc8220e0d8f4c61084a12 might be a bit dubious,
though I believe preloading workspace members is kinda idempotent
and registering patches/lockfile never cares about it.
### Additional information
Preserve file permissions on unix during `write_atomic`
### What does this PR try to resolve?
Fixes#13896.
> When you run `cargo add`, it changes the file permissions of `Cargo.toml` to 600 (user read+write only). This is a little bit painful when you're building the code as a different user than the user writing the code, for example if you're running the code in a container. This applies to `cargo remove` as well. I tested this behaviour on Cargo 1.78.0 and nightly.
I'm not entirely sure how permissions are handled on Windows, but the tempfile lib [doesn't seem to support them](https://docs.rs/tempfile/3.10.1/tempfile/struct.Builder.html#windows-and-others), so I haven't changed the behaviour on Windows.
Only the user/group/other read/write/execute permission bits are copied.
This PR sets the permissions ~twice~ once:
~1. When creating the file. This has the umask applied, but means that we don't create a file that is more permissive than the original.~
2. After the file has been created. This doesn't apply the umask, resulting in the file having the same u/g/o r/w/x permissions as the original file.
Since this PR changes a util function, it has a wider scope than just changing the behaviour of `cargo add` and `cargo remove`. `write_atomic` is called from the following functions:
- [`migrate_manifests`](4de0094ac7/src/cargo/ops/fix.rs (L340))
- [`update_manifest_with_new_member`](4de0094ac7/src/cargo/ops/cargo_new.rs (L1008))
- [`LocalManifest::write`](4de0094ac7/src/cargo/util/toml_mut/manifest.rs (L299))
- [`gc_workspace`](4de0094ac7/src/bin/cargo/commands/remove.rs (L274))
### How should we test and review this PR?
Unit test was added (`cargo test -p cargo-util write_atomic_permissions`).
