Can't run `cargo test --all` for `error-chain` anymore. The tests don't
compile because of `#[deny(invalid_doc_attributes)]`. Here's the error
message:
```
error: this attribute can only be applied at the crate level
--> tests/tests.rs:508:7
|
508 | #[doc(test)]
| ^^^^
|
= note: read <https://doc.rust-lang.org/nightly/rustdoc/the-doc-attribute.html#at-the-crate-level> for more information
= note: `#[deny(invalid_doc_attributes)]` on by default
help: to apply to the crate, use an inner attribute
|
508 | #![doc(test)]
| +
```
rustfmt currently has a runtime dependency on the sysroot. So when we
build a standalone rustfmt binary we need to set `LD_LIBRARY_PATH` so
each rustfmt binary knows where to find it's dependencies.
When running our Diff-Check job to test PRs for breaking changes it's
often the case that both the master rustfmt binary and the feature
branch binary have the same runtime dependencies so we only need to
set `LD_LIBRARY_PATH` once.
However, when running the diff-check job against a subtree sync PR that
assumption doesn't hold. The subtree sync PR bumps the required
toolchain used to build rustfmt and therefore the binary that gets built
for the subtree sync PR has a different runtime dependency than the
master rustfmt binary.
Now we set `LD_LIBRARY_PATH` twice to account for this potential
difference.
These changes mostly improve logging out the cargo version and version
of the two rustfmt binaries that are compiled. Some other minor logging
changes were made as well to add some whitespace to improve visual
clarity when looking at the logs in the GitHub Actions console.
Previously we were only building rustfmt with default features in CI. We
recently received a report that rustfmt was unable to compile with the
`generic-simd` feature, which is not enabled by default. To prevent a
similar situation in the future we'll start build nightly rustfmt with
all features enabled.
There was an issue with the script when passing optional rustfmt configs
without specifying a commit hash. Because these optional values are
passed via positional arguments the configs ($4) would be used in place
of the commit hash ($3). Now that we set a default value for the
optional commit hash we avoid this problem.
The `set -e` option is used to immediately exit if any command exits
with a non zero exit status. This will help us catch errors in the
script, for example, needing the `LD_LIBRARY_PATH` to be set.
There were some upstream changes made a while back that requires this to
be set when building rustfmt from source like we do in the
`check_diff.sh` script.
See issue 5675 for more details.
This new action is intended to help us maintainers determine when feature
branches cause breaking formatting changes by running rustfmt (master)
and the feature branch on various rust repositories.
Over time I expect the list of checked projects to increase.
With this action in place we can more easily test that a new feature or
bug fix doesn't introduce breaking changes. Although this action needs to
be manually triggered right now, we might consider adding it to our CI
runs in the future.
There are some tests in the rustfmt test suite that are ignored by
default. I believe these tests are ignored because they have caused
issues with the the `rust-lang/rust` test suite.
However, we recently experienced an issue (5395) that would have been
avoided had these tests been running.
With the introduction of the new `#[rustfmt_only_ci_test]` attribute
macro we can run these tests when the `RUSTFMT_CI` environment variable
is set, which will presumably only be set during rustfmts CI runs.
When the environment variable is not set the `#[rustfmt_only_ci_test]`
will be replaced with an `#[ignore]`.
* config_proc_macro: fix failing doctests
* ci: include config_proc_macro crate in ci
* [review] working native windows ci
* [fix] add --locked file for ci
* [fix] quoting of cmd variables
There was recently an issue where `cargo install` was installing a newer
version of a dependency than the one listed in our Cargo.toml. The newer
version added deprecation warnings that caused our continuous integration
tests to break.
As mentioned in the `cargo help install` docs, passing the `--locked`
flag should force cargo to use the `Cargo.lock` file included with
the repository.
The `cargo test --all` command failed and exited the main process with a
SIGINT. Trapping the signal or trying to get the code of a subshell
didn't work.
Close#2724
This commit adds integration tests against some crates in the
nursery.
Each integration test is added as a separate build-bot, where
the rust-lang-nursery/${CRATE} is first downloaded and its tests run.
Afterwards, `cargo fmt --all` is applied to the crate, and the tests
are re-run. If the tests fail after formatting, the integration test
fails.
The crates that currently fail are added as allowed-to-fail, but the
intent is that either these crates or rustfmt should be fixed such
that the tests pass.
