Tracking Issue for `panic_any`

[m-ou-se]

Feature gate: #![feature(panic_any)]

This is a tracking issue for std::panic::panic_any, a function to allow panicking with any value.

Context

Right now, it is possible to panic (that is, std::panic!) on any value: panic!(123). This will be disallowed in Rust 2021, when panic!(..) will always take a format string just like println!(..). At that point, this function will be the only way to panic without a (format) string. (See #80162 and rust-lang/rfcs#3007.)

Public API

mod std::panic {

    /// Panic the current thread with the given message as the panic payload.
    ///
    /// The message can be of any (`Any + Send`) type, not just strings.
    ///
    /// The message is wrapped in a `Box<'static + Any + Send>`, which can be
    /// accessed later using [`PanicInfo::payload`].
    ///
    /// See the [`panic!`] macro for more information about panicking.
    pub fn panic_any<M: Any + Send>(msg: M) -> !;

}

Steps / History

• Implementation: Add std::panic::panic_any. #74622
• Final commenting period (FCP)
• Stabilization PR: Stabilize std::panic::panic_any. #81194

Unresolved Questions

• None yet.

[m-ou-se]

@rfcbot merge

Once this is stable, we can add a lint about using this function instead of panic!(x) for non-strings, as part of RFC 3007 (see #80162).

[rfcbot]

Team member @m-ou-se has proposed to merge this. The next step is review by the rest of the tagged team members:

• @Amanieu
• @BurntSushi
• @KodrAus
• @dtolnay
• @m-ou-se
• @sfackler
• @withoutboats

No concerns currently listed.

Once a majority of reviewers approve (and at most 2 approvals are outstanding), this will enter its final comment period. If you spot a major issue that hasn't been raised at any point in this process, please speak up!

See this document for info about what commands tagged team members can give me.

[sfackler]

It seems like the function signature is missing a 'static bound?

[m-ou-se]

It seems like the function signature is missing a 'static bound?

Any: 'static already requires that. I can add it in the stabilization PR to make it explicit.

[sfackler]

Oh right, I think I asked this same question when this function was added! Yeah, might be best to add the duplicate bound in.