Remove -Zasymmetric-token / cargo:paseto - #17486
Conversation
This feature has been stale for s significant time. Since then, we've decided to improve security through: - trusted publishing - deprecating storing tokens in plain text, preferring OS native credential stores Keeping it around has costs: - build time - test time - needs to be taken into account for any config system refactors See also https://rust-lang.zulipchat.com/#narrow/channel/246057-t-cargo/topic/Should.20we.20drop.20paseto.3F/with/624770308
|
r? @Muscraft rustbot has assigned @Muscraft. Use Why was this reviewer chosen?The reviewer was selected based on:
|
| struct Mutation<'a> { | ||
| mutation: &'a str, | ||
| name: Option<&'a str>, | ||
| vers: Option<&'a str>, | ||
| cksum: Option<&'a str>, | ||
| } | ||
| struct Mutation {} | ||
|
|
There was a problem hiding this comment.
I figured Some(Mutation) | None made intent in APIs more clear than true | false so I kept the struct around. I also considered an enum but felt this worked just as well and minimized the change
| // Please keep this list lexicographically ordered. | ||
| advanced_env: bool, | ||
| any_build_script_metadata: bool = ("Allow any build script to specify env vars via cargo::metadata=key=value"), | ||
| asymmetric_token: bool = ("Allows authenticating with asymmetric tokens"), |
There was a problem hiding this comment.
I'm not seeing precedence fr removing a -Z besides stabilization. I could instead keep it and error on use. I started the PR in the simpler state.
There was a problem hiding this comment.
I looked into what rustc does in this scenario and it appears they just go ahead and remove the -Z flag.
What does this PR try to resolve?
This feature has been stale for s significant time. Since then, we've decided to improve security through:
Keeping it around has costs:
See also https://rust-lang.zulipchat.com/#narrow/channel/246057-t-cargo/topic/Should.20we.20drop.20paseto.3F/with/624770308
How to test and review this PR?
LLM disclosure: the initial draft was written by an LLM which I then reviewed and changed. @weihanglo has agreed to review this.