5c87c14f9a
This is to bring us into conformance with the [Rust crate ownership policy](https://forge.rust-lang.org/policies/crate-ownership.html). Items of note - `cargo-credential-1password` is declared as Experimental as it is intended for the community but I was unsure if we wanted to commit to full support for it. In my mind, the ideal thing to do would be to expatriate this to 1password. - `home` is declared as Internal despite its wide use within the ecosystem. - `cargo-credential` is declared as Intentional as its an API intended for the wider ecosystem and I didn't see a reason to declare it experimental. - `cargo-platform`, `cargo-util-schemas`, and `crates-io` are declared as Intentional as they are both used internally and intended for others to use for logic that integrates with cargo/registries. I wondered about these being Experimental or Internal instead.
1011 B
resolver-tests
The aim
This crate aims to test the resolution of Cargo's resolver. It implements a SAT solver to compare with resolution of Cargo's resolver.
This ensures that Cargo's dependency resolution is proven valid by lowering to SAT problem.
This crate is maintained by the Cargo team, primarily for use by Cargo and not intended for external use (except as a transitive dependency). This crate may make major changes to its APIs or be deprecated without warning.
About the test
The Cargo's resolver is very sensitive to what order it tries to evaluate constraints. This makes it incredibly difficult
to be sure that a handful of tests actually covers all the important permutations of decision-making. The tests not only needs
to hit all the corner cases, it needs to try all of the orders of evaluation. So we use fuzz testing to cover more permutations.