Tracking issue for RFC 2345, "Allow panicking in constants" (const_panic) #51999

Centril · 2018-07-02T18:25:03Z

This is a tracking issue for the RFC "Allow panicking in constants" (rust-lang/rfcs#2345).

Steps:

Implement the RFC (cc @rust-lang/compiler -- can anyone write up mentoring instructions?)
Adjust documentation (see instructions on forge)
Stabilization PR (see instructions on forge)

Unresolved questions:

Should there be some additional message in the error about this being a panic turned error?
Or do we just produce the exact message the panic would produce?
This change becomes really useful if Result::unwrap and Option::unwrap become const fn, doing both in one go might be a good idea.

Blockers:

ICE: const_panic: ICE on non-&str panic payload #66693

The text was updated successfully, but these errors were encountered:

clarfonthey · 2018-07-02T19:58:09Z

Don't forget expect and all the other unwrap-like methods. :P

vi · 2018-07-04T12:44:07Z

I suppose unimplemented!() and unreachable!() should also be included?

SimonSapin · 2018-07-04T12:53:57Z

I think they would be "automatically", since macros don’t have a separation similar to fn v.s. const fn.

oli-obk · 2018-07-04T13:08:34Z

I'll add some tests to #52011, but yes, it just works.

@eddyb

…anics_constantly, r=eddyb Allow panicking with string literal messages inside constants r? @eddyb cc #51999 we can't implement things like `panic!("foo: {}", x)` right now because we can't call trait methods (most notably `Display::fmt`) inside constants. Also most of these impls probably have loops and conditions, so it's messy anyway. But hey `panic!("foo")` works at least. cc @japaric got any test ideas for `#![no_std]`?

Centril · 2018-09-15T20:24:13Z

Triage: @oli-obk so this is implemented now, right?

oli-obk · 2018-09-16T20:36:18Z

Yes

IsaacWoods · 2018-10-19T22:06:00Z

Now this is implemented, is there anything stopping Option::unwrap and friends from becoming const fns (I'm assuming with a rustc_const_unstable(feature = "??"))?

Centril · 2018-10-19T22:09:58Z

@IsaacWoods to my knowledge there shouldn't be anything in the way of that.

IsaacWoods · 2018-10-19T22:19:17Z

@Centril great! I'll start work on a PR in that case

clarfonthey · 2018-10-19T22:19:24Z

The thing in the way of that is running fmt::Debug::fmt in const context.

…oh, right, I was thinking in terms of Result::unwrap and Result::expect. Option should be fine.

IsaacWoods · 2018-10-19T22:26:37Z

On second thoughts, even Option is still blocked I think - I don't think we can use match in const contexts yet either

oli-obk · 2018-10-19T22:41:59Z

Yea, we need conditions in constants first

Centril · 2018-10-19T22:44:03Z

Right; bummer.

@oli-obk Any thoughts about the first unresolved question?

oli-obk · 2018-10-19T22:48:59Z

I'm for basically reporting it verbatim, but having the same prefix (error[0080]:) as all other errors.

Centril · 2018-10-19T22:52:22Z

@oli-obk I like that; (and it's a diagnostics issue anyways, not part of the spec, so it is up to T-compiler to decide this...).

nikomatsakis · 2021-05-11T19:38:18Z

New issue: #85194

jhpratt · 2021-07-30T05:21:14Z

Now that #86998 is merged, this should be good to stabilize, correct?

oli-obk · 2021-07-30T12:02:31Z

I'm still unhappy that in 2018 we can do panic!(some_computed_str) but in 2021 we can't do the equivalent, but it seems better to only allow literal strings for now than to not allow any custom panicking. Imo this is purely a T-libs issue, but I'll let @rust-lang/wg-const-eval weigh in on that before escalating to T-libs for stabilization.

RalfJung · 2021-07-30T13:32:09Z

it seems better to only allow literal strings for now than to not allow any custom panicking

👍 as long as this is enforced properly(i.e. everything accepted in 2018 on stable also works in 2021).

oli-obk · 2021-07-30T13:41:16Z

it is not yet. Not sure how doable that is, it may be easier to just figure out how to do panic!("{}", some_computed_str) on all editions.

abonander · 2021-08-12T21:38:42Z

For our planned usage of this feature in SQLx we would really prefer either support for formatting or being able to compute panic strings. We can sorta work around it if the panic strings need to be literals but it won't be the greatest.

clarfonthey · 2021-09-09T16:56:34Z

One thing that I noticed which kind of sucks: this being unstable means that you can't use todo!() in consts. Although I understand not wanting to stabilise that so that people don't use it instead of panic.

Allow `panic!("{}", computed_str)` in const fn. Special-case `panic!("{}", arg)` and translate it to `panic_display(&arg)`. `panic_display` will behave like `panic_any` in cosnt eval and behave like `panic!(format_args!("{}", arg))` in runtime. This should bring Rust 2015 and 2021 to feature parity in terms of `const_panic`; and hopefully would unblock the stabilisation of rust-lang#51999. `@rustbot` modify labels: +T-compiler +T-libs +A-const-eval +A-const-fn r? `@oli-obk`

…shtriplett Stabilize `const_panic` Closes rust-lang#51999 FCP completed in rust-lang#89006 `@rustbot` label +A-const-eval +A-const-fn +T-lang cc `@oli-obk` for review (not `r?`'ing as not on lang team)

…shtriplett Stabilize `const_panic` Closes rust-lang#51999 FCP completed in rust-lang#89006 ``@rustbot`` label +A-const-eval +A-const-fn +T-lang cc ``@oli-obk`` for review (not `r?`'ing as not on lang team)

rodrimati1992 · 2021-10-16T08:14:38Z

I've noticed that you can use a non-static foo: &str as an argument to panic!("{}", foo), is it intentional that it doesn't require a 'static one?
https://play.rust-lang.org/?version=nightly&mode=debug&edition=2018&gist=8bd048da90965a526dbc98d187f56c56

const _: () = {
    let arr = [b'b', b'y', b'e'];
    let msg: &str = unsafe{ std::str::from_utf8_unchecked(&arr) };

    panic!("{}", msg)
};

if it is intentional, I have an idea for how to implement const panics with bounded-length formatted arguments in a library.

oli-obk · 2021-10-16T08:43:23Z

Yes this is intentional. Getting that to work was actually a blocker for stabilization.

Since const panics are now [stabilized], I figure it would be good to document some of its gotchas. I couldn't really find documents for the specifics I ran into. I am no const eval expert, and what I wrote is basically a paraphrase of Ralf's comments[^1][^2], but I hope to get the ball rolling for adding some docs. I deliberately chose to not mention the guarantee about syntactically referenced constants in functions that require code generation as it seems hard to explain completely without some ambiguity. It also seems to me that the rules are not completely stable[^3][^4] yet. [stabilized]: rust-lang/rust#89006 [^1]: rust-lang/rust#91877 (comment) [^2]: rust-lang/rust#91877 (comment) [^3]: rust-lang/rust#71800 [^4]: rust-lang/rust#51999 (comment)

tmandry · 2022-03-04T22:20:22Z

Is there an issue tracking support for non-&str panic arguments in consts?

Right now the compiler just gives an error (implemented in #80734) "argument to panic!() in a const context must have type &str", but there's no error code or other explanation text.

The RFC doesn't mention this limitation.

oli-obk · 2022-03-04T22:36:21Z

There's no special tracking for it. It's essentially just blocked on the fact that you can't implement Display or Debug in a way that is callable in a const context.

Centril added B-RFC-approved Blocker: Approved by a merged RFC but not yet implemented. T-lang Relevant to the language team, which will review and decide on the PR/issue. C-tracking-issue Category: A tracking issue for an RFC or an unstable feature. labels Jul 2, 2018

Centril mentioned this issue Jul 2, 2018

Allow panicking in constants rust-lang/rfcs#2345

Merged

oli-obk added the A-const-eval Area: constant evaluation (mir interpretation) label Jul 2, 2018

oli-obk mentioned this issue Jul 4, 2018

Allow panicking with string literal messages inside constants #52011

Merged

Centril mentioned this issue Aug 21, 2018

Tracking issue for a minimal subset of RFC 911, const fn #53555

Closed

4 tasks

Centril added B-unstable Blocker: Implemented in the nightly compiler and unstable. B-RFC-implemented Blocker: Approved by a merged RFC and implemented. labels Sep 18, 2018

This was referenced Oct 7, 2018

Static asserts rust-lang/rfcs#1146

Closed

const fn's should allow (and eval) assert! and debug_assert! statements rust-lang/rfcs#1383

Closed

Compile-time checked version of unimplemented!() rust-lang/rfcs#1911

Open

RalfJung changed the title ~~Tracking issue for RFC 2345, "Allow panicking in constants"~~ Tracking issue for RFC 2345, "Allow panicking in constants" (const_panic) May 15, 2021

nikomatsakis removed the I-nominated label May 18, 2021

c410-f3r mentioned this issue Jul 3, 2021

Guidance for a Rust newbie - Decimal constant? paupino/rust-decimal#400

Closed

konsumlamm mentioned this issue Aug 19, 2021

Consider using const_panic in the future. nvzqz/static-assertions#43

Open

nbdd0121 mentioned this issue Sep 15, 2021

Allow panic!("{}", computed_str) in const fn. #88954

Merged

oli-obk mentioned this issue Sep 16, 2021

Stabilize RFC 2345: Allow panicking in constants #89006

Closed

jhpratt mentioned this issue Oct 4, 2021

Stabilize const_panic #89508

Merged

bors closed this as completed in 9866b09 Oct 4, 2021

DanieleDiBenedetto mentioned this issue Oct 21, 2021

Endo HorizenOfficial/ginger-lib#128

Merged

gaojiaqi7 mentioned this issue Dec 20, 2021

Latest version "0.14.7" of crate x86_64 fails to build confidential-containers/td-shim#13

Closed

XrXr mentioned this issue Dec 24, 2021

Note difference in CTFE timing between associated and free constants rust-lang/reference#1120

Merged

npmccallum mentioned this issue Mar 24, 2022

feat: prepare for enarx enarx/mmledger#12

Closed

soqb mentioned this issue Oct 17, 2022

stable reflect type name bevyengine/bevy#5805

Closed

5 tasks

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Tracking issue for RFC 2345, "Allow panicking in constants" (const_panic) #51999

Tracking issue for RFC 2345, "Allow panicking in constants" (const_panic) #51999

Centril commented Jul 2, 2018 •

edited by RalfJung

Loading

clarfonthey commented Jul 2, 2018

vi commented Jul 4, 2018

SimonSapin commented Jul 4, 2018

oli-obk commented Jul 4, 2018

Centril commented Sep 15, 2018

oli-obk commented Sep 16, 2018

IsaacWoods commented Oct 19, 2018

Centril commented Oct 19, 2018

IsaacWoods commented Oct 19, 2018

clarfonthey commented Oct 19, 2018 •

edited

Loading

IsaacWoods commented Oct 19, 2018

oli-obk commented Oct 19, 2018

Centril commented Oct 19, 2018

oli-obk commented Oct 19, 2018

Centril commented Oct 19, 2018

nikomatsakis commented May 11, 2021

jhpratt commented Jul 30, 2021

oli-obk commented Jul 30, 2021

RalfJung commented Jul 30, 2021

oli-obk commented Jul 30, 2021

abonander commented Aug 12, 2021

clarfonthey commented Sep 9, 2021

rodrimati1992 commented Oct 16, 2021 •

edited

Loading

oli-obk commented Oct 16, 2021

tmandry commented Mar 4, 2022

oli-obk commented Mar 4, 2022

Tracking issue for RFC 2345, "Allow panicking in constants" (const_panic) #51999

Tracking issue for RFC 2345, "Allow panicking in constants" (const_panic) #51999

Comments

Centril commented Jul 2, 2018 • edited by RalfJung Loading

clarfonthey commented Jul 2, 2018

vi commented Jul 4, 2018

SimonSapin commented Jul 4, 2018

oli-obk commented Jul 4, 2018

Centril commented Sep 15, 2018

oli-obk commented Sep 16, 2018

IsaacWoods commented Oct 19, 2018

Centril commented Oct 19, 2018

IsaacWoods commented Oct 19, 2018

clarfonthey commented Oct 19, 2018 • edited Loading

IsaacWoods commented Oct 19, 2018

oli-obk commented Oct 19, 2018

Centril commented Oct 19, 2018

oli-obk commented Oct 19, 2018

Centril commented Oct 19, 2018

nikomatsakis commented May 11, 2021

jhpratt commented Jul 30, 2021

oli-obk commented Jul 30, 2021

RalfJung commented Jul 30, 2021

oli-obk commented Jul 30, 2021

abonander commented Aug 12, 2021

clarfonthey commented Sep 9, 2021

rodrimati1992 commented Oct 16, 2021 • edited Loading

oli-obk commented Oct 16, 2021

tmandry commented Mar 4, 2022

oli-obk commented Mar 4, 2022

Centril commented Jul 2, 2018 •

edited by RalfJung

Loading

clarfonthey commented Oct 19, 2018 •

edited

Loading

rodrimati1992 commented Oct 16, 2021 •

edited

Loading