Add some documentation about what as casts are legal

[sourcefrog]

Based on the description in

rust/src/librustc_typeck/check/cast.rs

Line 11 in 219eca1

//! Code for type-checking cast expressions.

and https://doc.rust-lang.org/nightly/nomicon/casts.html

[rust-highfive]

r? @nikomatsakis

(rust_highfive has picked a reviewer for you, use r? to override)

[sourcefrog]

I don't feel this is really clear enough at the level of the Rust book, yet, but it's a start.

There doesn't seem to be any definition of the concept of coercion yet.

[sourcefrog]

Thanks to Adrian in http://stackoverflow.com/a/33884290/243712 for pointing this out.

[Gankra]

Note that https://doc.rust-lang.org/nomicon/casts.html and https://doc.rust-lang.org/nomicon/coercions.html exist. I'm honestly a bit dubious of their existence, though. They were more motivated when I was trying to make the nomicon totally stand-alone and/or a substitute for the reference. Migrating all this content to The Book proper seems like a reasonable choice.

[Gankra]

r? @steveklabnik

[steveklabnik]

Part of the problem with this is that it introduces a forward reference to raw pointers, but, I'm planning on a reorganization of the book anyway, so having this text around can still be useful.

Thanks for adding this!

@bors: r+

[bors]

📌 Commit d695212 has been approved by steveklabnik

[steveklabnik]

(@gankro we should chat about your nomicon feels sometime too)

[sourcefrog]

Thanks for the speedy review. I'd like to clear up the phrasing a bit before this is actually merged.

[steveklabnik]

@bors: r-

What parts in particular?

On Nov 28, 2015, 13:34 -0500, Martin Poolnotifications@github.com, wrote:

Thanks for the speedy review. I'd like to clear up the phrasing a bit before this is actually merged.

—
Reply to this email directly orview it on GitHub(#30088 (comment)).

[sourcefrog]

Do you think it's clear enough?

It refers to some constants like Sized and C-like enum that aren't explained by this point. For Sized I could just point forward to the chapter on traits, and C-like enums are just enums whose values are all integers?

I think it's also possible to handle coercion by defining Deref?

[sourcefrog]

OK, I think this is clearer. I copied some text about the semantics of numeric casts from the nomicon, which seemed appropriate for an introduction.

If you think this is clear enough and correct, please go ahead and merge.

[steveklabnik]

It refers to some constants like Sized and C-like enum that aren't explained by this point.

Since this is getting re-arranged in the near future, I'm not as worried about it as I normally would be.

[steveklabnik]

I do still think this is great, thanks 👍

@bors: r+ rollup

[bors]

📌 Commit 797d543 has been approved by steveklabnik

[bors]

⌛ Testing commit 797d543 with merge 15ad1d1...

[bors]

☀️ Test successful - auto-linux-32-nopt-t, auto-linux-32-opt, auto-linux-64-debug-opt, auto-linux-64-nopt-t, auto-linux-64-opt, auto-linux-64-x-android-t, auto-linux-cross-opt, auto-linux-musl-64-opt, auto-mac-32-opt, auto-mac-64-nopt-t, auto-mac-64-opt, auto-win-gnu-32-nopt-t, auto-win-gnu-32-opt, auto-win-gnu-64-nopt-t, auto-win-gnu-64-opt, auto-win-msvc-32-opt, auto-win-msvc-64-opt

[Ms2ger]

... this doesn't compile.

[sourcefrog]

:/

I'll send a follow-up

[sourcefrog]

Is it considered a bug in rustbook, or in the rust tree's use of it, that it doesn't test the examples? I did run the tests on this and I guess bors did too.

[steveklabnik]

It should have. I'm not sure why this passed.

[steveklabnik]

I think it's a bug, I'm fixing it. Ugh. :(