cargo test --doc --no-default-features currently fails with a compiler error:
failures:
---- src/lib.rs - (line 10) stdout ----
error[E0433]: failed to resolve: use of undeclared crate or module `rand`
--> src/lib.rs:13:5
|
6 | use rand::distributions::Distribution;
| ^^^^ use of undeclared crate or module `rand`
error[E0433]: failed to resolve: use of undeclared crate or module `rand`
--> src/lib.rs:14:13
|
7 | let mut r = rand::rngs::OsRng;
| ^^^^ use of undeclared crate or module `rand`
error[E0599]: no method named `sample` found for struct `Exp` in the current scope
--> src/lib.rs:16:16
|
9 | print!("{}", n.sample(&mut r));
| ^^^^^^ method not found in `Exp`
error: aborting due to 3 previous errors
Some errors have detailed explanations: E0433, E0599.
For more information about an error, try `rustc --explain E0433`.
Couldn't compile the test.
failures:
src/lib.rs - (line 10)
This was introduced in #275 and not noticed because CI doesn't run tests on all feature combinations. I think this only affects local doc builds, but it's still annoying that this was missed by CI. But testing all feature combinations with cargo-hack and --feature-powerset would likely increase the CI time dramatically. Not sure it's worth it since normal compiler errors will get caught by CI, it's just doctests that are not completely covered..
cargo test --doc --no-default-features
currently fails with a compiler error:This was introduced in #275 and not noticed because CI doesn't run tests on all feature combinations. I think this only affects local doc builds, but it's still annoying that this was missed by CI. But testing all feature combinations with
cargo-hack
and--feature-powerset
would likely increase the CI time dramatically. Not sure it's worth it since normal compiler errors will get caught by CI, it's just doctests that are not completely covered..