Re: [ghc-steering-committee] Proposal #631: Set program exit code by main return type, recommendation: accept something

Dear all, Shea has updated his proposal based on the committee's feedback. There seem to be two main alternatives being considered at the moment - Having a type class to compute the exit code based on the type. This is Shea's favourite. It can be done without an extension (as Shea's proposing) or with an extension. - Keep the current behaviour but emit a warning when the return type of `main` isn't `()` or `Void`. I have opinions about my preference, but I'd like to hear about everybody's thoughts first. On Thu, 7 Mar 2024 at 10:27, Adam Gundry wrote:

I've added a comment to the GitHub thread ( https://github.com/ghc-proposals/ghc-proposals/pull/631#issuecomment-1983060...)

elaborating slightly on Richard's suggestion (albeit with an effectively indefinite transition period).

Adam

On 05/03/2024 08:52, Arnaud Spiwack wrote:

...

This is Alternative 7.5 in the current version of the proposal

https://github.com/shlevy/ghc-proposals/blob/io-exitcode/proposals/0631-main... < https://github.com/shlevy/ghc-proposals/blob/io-exitcode/proposals/0631-main-return-types.rst#75require-an-exitstatus-instance> .

...

PS: I tend to agree with Richard that requiring an ExitStatus instance is the preferable option. But food for thought for the proposal thread when that conversation happens there: should that be gated behind an extension? In which case it won't become the default before the next language edition.

/Arnaud

On Mon, 4 Mar 2024 at 21:35, Simon Peyton Jones mailto:simon.peytonjones@gmail.com>

wrote:

...

I left out a key part of my last email -- apologies. I'm floating a counter-proposal where we *require* an instance of ExitStatus on the return type of `main`, with a transition period. In contrast, my understanding of the proposal written is that it would use such an instance if it exists, but issue a warning if it doesn't, in perpetuity.

Ah I had not realised that.

But why?

Rather than answer here (private to SC) why don't you put your proposal on the discussion thread, say why, and invite feedback.

Simon

On Mon, 4 Mar 2024 at 19:24, Richard Eisenberg mailto:reisenberg@janestreet.com>

wrote:

...

I left out a key part of my last email -- apologies. I'm floating a counter-proposal where we *require* an instance of ExitStatus on the return type of `main`, with a transition period. In contrast, my understanding of the proposal written is that it would use such an instance if it exists, but issue a warning if it doesn't, in perpetuity.

Richard

On Mon, Mar 4, 2024 at 6:14 AM Simon Peyton Jones mailto:simon.peytonjones@gmail.com> wrote:

I am a little worried about breaking programs that end in an innocent-looking `return 0`, just because some other languages like to end programs with that phrase

The proposal specifies that such a program returns `ExitSuccess`, but adds a warning. That seems OK to me; it does not break the program.

Oh -- maybe you mean that `return 1` means "return with exit code 1" today. Is that really true? I don't think so.

Overall this proposal seems fine to me. I'd be happy to see it done.

Simon

On Thu, 29 Feb 2024 at 12:38, Richard Eisenberg mailto:reisenberg@janestreet.com> wrote:

I haven't followed this proposal closely. But couldn't we have a transition period toward this eventual goal? That is, introduce a new warning, on by default, if `main` returns anything other than `()`. That goes for a few releases. Then we require that the return type of main has an instance of ExitStatus.

I'm not worried about changing the behavior of programs that have type IO ExitCode but expect the program to return 0 unconditionally; that's just begging for confusion. I am a little worried about breaking programs that end in an innocent-looking `return 0`, just because some other languages like to end programs with that phrase. So I'm not sure if we should have an instance ExitStatus Int (or instance ExitStatus Integer) -- but we probably should. If a program ends with `return 1`, the programmer probably wants the OS to return 1 as well.

Richard

On Thu, Feb 29, 2024 at 5:29 AM Arnaud Spiwack mailto:arnaud.spiwack@tweag.io> wrote:

Dear all,

Shea Levy proposes to do something with the values returned by `main`

https://github.com/ghc-proposals/ghc-proposals/pull/631 < https://github.com/ghc-proposals/ghc-proposals/pull/631> .

...

The problem is that `main` is allowed to be of type `IO A` for any `A`. And GHC will simply drop the value returned by `main`. Shea contends that it's surprising. I agree that dropping a value without the compiler being explicitly instructed to is surprising. But Shea says that when `A` is `ExitCode` this is even more surprising. Namely `main :: IO ExitCode; main = return $ Failure 1` actually terminates with exit code 0. And I doubt that it's what anybody expects when reading the code.

The proposal is simple, but I have a lot of comments on it. Sorry about that…

Now, this sort of proposal is tricky. When the current behaviour is confusing, we want to change the default. But putting the new default behind an extension doesn't really solve the fact that there's a trap. The extension is, therefore, unlikely to be well tested before it becomes part of the next language edition.

Shea's main proposition doesn't actually use an extension though. He adds a type class `ExitStatus`, and if `ExistStatus A`, then `main :: IO A` uses the instance to determine the exit code based on the return value.

The only change to the current behaviour is that `main :: IO ExitCode` instead of always terminating with exit code 0 when returning now terminates with the expected error code. The argument for not putting this behind an extension is that virtually anybody affected by the change will actually have the behaviour they were expecting. But maybe the argument isn't strong enough (the changes may be more “interesting” if some library exports some `ExistStatus` instance).

This design of this proposal is inspired by Rust's design. I've asked our Rust team, and they certainly seem to have internalised the idea of returning an exit code. It really seems a pretty natural feature to have. So I'm rather in favour of some flavour of the type class implementation. Though have a look at the alternatives, where you'll find other approaches such as restricting the type of `main` to unsurprising types.

One caveat with respect to the main proposal: it is proposed that when no `ExistStatus A` is found, then we drop the returned value like today. I don't know that it's quite easy to implement this behaviour. But it can be recovered by a catch-all overlapping instance, so maybe it's a better way to specify the desired behaviour.

-- Arnaud Spiwack Director, Research at https://moduscreate.com https://moduscreate.com and https://tweag.io https://tweag.io.

-- Adam Gundry, Haskell Consultant Well-Typed LLP, https://www.well-typed.com/

Registered in England & Wales, OC335890 27 Old Gloucester Street, London WC1N 3AX, England

_______________________________________________ ghc-steering-committee mailing list ghc-steering-committee@haskell.org https://mail.haskell.org/cgi-bin/mailman/listinfo/ghc-steering-committee

-- Arnaud Spiwack Director, Research at https://moduscreate.com and https://tweag.io.