crates-io 0.36.1 breaks semver

[lu-zero]

Problem

At least rust_version got added to NewCrate.

Steps

Open 0.36.0 and 0.36.1 and compare.

Possible Solution(s)

yank 0.36.1 and release it as 0.37.0

Notes

No response

Version

No response

[pietroalbini]

Yanked cargo 0.72.0 and crates-io 0.36.1, until the Cargo team decides how to handle this.

[obi1kenobi]

👋 I noticed this via the rust-lang Zulip, I'm the maintainer of cargo-semver-checks.

Any interest in adopting cargo-semver-checks for the lib crates in this repo? If so, I'd be happy to help!

It's designed to be runnable as cargo semver-checks && cargo publish and one of its existing lints would have flagged and prevented this accidental semver break:

     Parsing crates-io v0.36.1 (current)
     Parsing crates-io v0.36.0 (baseline)
    Checking crates-io v0.36.0 -> v0.36.1 (minor change)
   Completed [   0.015s] 39 checks; 38 passed, 1 failed, 6 unnecessary

--- failure constructible_struct_adds_field: externally-constructible struct adds field ---

Description:
A pub struct constructible with a struct literal has a new pub field. Existing struct literals must be updated to include the new field.
        ref: https://doc.rust-lang.org/reference/expressions/struct-expr.html
       impl: https://github.com/obi1kenobi/cargo-semver-check/tree/v0.22.1/src/lints/constructible_struct_adds_field.ron

Failed in:
  field NewCrate.rust_version in /.../cargo/crates/crates-io/lib.rs:60
       Final [   0.015s] semver requires new major version: 1 major and 0 minor checks failed

You can replicate this output locally right now by running cargo semver-checks --baseline-version 0.36.0 from the directory containing the crates-io manifest.

[epage]

@obi1kenobi we are slowly working towards that. Our release process is a bit scattered and we need to carefully improve our release process to not have unintended affects

e.g. one simple step we've taken is to prevent accidentally forgetting to prep a crate for release. See #12352 as an example where CI caught an "issue" (strictly speaking, that change doesn't need to be released but oh well)

[obi1kenobi]

Makes sense, thanks for the context! Let me know if I can do anything to help. If it'd be useful, I'd be happy to try plugging in the cargo-semver-checks GitHub Action into the CI pipeline you pointed me to in the linked PR.