TRPL: deref coercions

[steveklabnik]

r? @alexcrichton

[alexcrichton]

s/convert/coerce/

I think the distinction may matter in some places? I'm not super sure though

[alexcrichton]

Nice! Could this also mention somewhere that Vec<T> will coerce to [T]? I think that's the only other major coercion that may need to be mentioned.

[alexcrichton]

r=me with those nits otherwise though!

[Manishearth]

Can we also mention (or link to) autoderef here somewhere? I'd like all of the deref magic to be mentioned in one place.

[steveklabnik]

issues fixed, and @Manishearth 's section added 😄

[alexcrichton]

Did this mean to use & instead of *?

[steveklabnik]

Yes. eff.

[steveklabnik]

well, i guess it actually meant to use & and * alternating, since that's the behavior, right?

[alexcrichton]

I think this means to say that a value of type &&&&&&&&&&&&&&&&Foo can still have methods defined on Foo called as the compiler will insert as many * operations as necessary (basically).

[steveklabnik]

I'm thinking of http://stackoverflow.com/questions/28519997/what-are-rusts-exact-auto-dereferencing-rules/28552082#28552082

[alexcrichton]

@bors: r+ d70a994 rollup

[steveklabnik]

adding to #24786

[frewsxcv]

Which 'two things' is this referring to?