#[derive(Into)]

[dingfengquek]

struct Oldtype;
struct Newtype(Oldtype);

impl From<Newtype> for Oldtype { ... } // #[derive(From)]
impl From<Oldtype> for Newtype { ... } // #[derive(Into)]

Code would be very similar to #[derive(From)].
Although #[derive(FromInverse)] might be clearer, but longer.

[JelteF]

Good idea! I think whatever it would be named it it will be confusing though. I think the easiest way to avoid it would be to derive both when deriving From and possibly have long descriptive variants that only allow one way conversion.

[dingfengquek]

Some non-confusing very explicit names: #[derive(FromSelfForInner)] and #[derive(FromInnerForSelf)], although they're somewhat long.

[dingfengquek]

I'm implementing for my own use because I want it badly. Would you like a PR for it in a few week's time?

[JelteF]

Yes, sounds great! For structs it should be quite straightforward when looking at the opposite From implementation and just reversing it should be fine. I would start with those. Enums are likely much more complicated. For them you probably have to implement From for Option<InternalType>, since you don't know if it is the correct variant. Also multiple variants could implement Into for the same resulting type, they would need to be merged into a single Into impl. These are the ones that are currently being ignored when deriving From, because they would be ambiguous. On Wed, Feb 8, 2017, 17:43 Dingfeng Quek <notifications@github.com> wrote: I'm implementing for my own use because I want it badly. Would you like a PR for it in a few week's time? — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#13 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/ABG8Jof9v01Gzc1en5ef_pvdmqJkkpz_ks5rafCwgaJpZM4L4VAb> .

[JelteF]

Btw, I think calling it derive(Into) and having to derive it explicitly would make the most sense after all. As long as we make the docs clear on which does which it should be fine.

…

On Wed, Feb 8, 2017, 17:43 Dingfeng Quek ***@***.***> wrote: I'm implementing for my own use because I want it badly. Would you like a PR for it in a few week's time? — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#13 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/ABG8Jof9v01Gzc1en5ef_pvdmqJkkpz_ks5rafCwgaJpZM4L4VAb> .

[JelteF]

If you just need derive(Into) for regular or tuple structs it's also fine if you just make a pull request for those btw. I can add the rest of the support later.

[JelteF]

Can you let me know if you're already working on this. Otherwise I think I'll start on this in a couple of days.

[JelteF]

I just added an implementation of derive(Into) to the master branch (only for structs, not enums).

I will close this issue for now. Could you test if it indeed covers your use case? Please reopen the issue if this is not the case.

[JelteF]

I just released 0.6.0, this also includes the generics fixes from before.