Fixed iOS build after oibit #19729
Fixed iOS build after oibit #19729
Conversation
Fixed iOS build after oibit Reviewed-by: alexcrichton
Fixed iOS build after oibit Reviewed-by: alexcrichton
Using a type alias for iterator implementations is fragile since this exposes the implementation to users of the iterator, and any changes could break existing code. This commit changes the iterators of `VecMap` to use proper new types, rather than type aliases. However, since it is fair-game to treat a type-alias as the aliased type, this is a: [breaking-change].
Fixed iOS build after oibit Reviewed-by: alexcrichton
| @@ -21,6 +21,7 @@ pub use self::_Unwind_Action::*; | |||
| pub use self::_Unwind_State::*; | |||
| pub use self::_Unwind_Reason_Code::*; | |||
|
|
|||
| use core::kinds::Copy; | |||
There was a problem hiding this comment.
This line caused the build to fail (unused import), so you may remove the line and use an absolute path.
There was a problem hiding this comment.
That's weird as I had no problems building it (OS X and iOS)
There was a problem hiding this comment.
I'm not sure, but maybe the warning level is different? The buildbot considers any warnings to be critical.
You can see the detailed error in here: http://buildbot.rust-lang.org/builders/auto-linux-64-x-android-t/builds/2691/steps/compile/logs/stdio
There was a problem hiding this comment.
Oh, now I see. The build fails only on Android. Your #[cfg] states "Not ARM or iOS", so it builds correctly on non-ARM targets. However it fails on ARM-based non-iOS (say, Android) targets.
There was a problem hiding this comment.
Makes sense. The fun thing is that I can't use absolute path - in order to do that I have to use core which will fail with the same reason. Got a pretty weird workaround with a module, will check how it builds now.
| @@ -33,6 +33,13 @@ pub enum _Unwind_Action { | |||
| _UA_END_OF_STACK = 16, | |||
| } | |||
|
|
|||
| #[cfg(any(not(target_arch = "arm"), target_os = "ios"))] | |||
| mod copy_hack { | |||
| use core::kinds::Copy; | |||
There was a problem hiding this comment.
😄 This looks a bit funny. Is it because we are currently forcing use directives at the top of a block? I remember having seen a suggestion to relieve this restriction, but that's another story...
There was a problem hiding this comment.
I've finally come with another way to express it, although to be honest, I can't say I like any of available options too much.
|
I think this may be able to avoid the weird inner module with |
|
@alexcrichton |
|
You'll just need to add |
|
Hmm, I've never thought about such module hacks. |
|
Does Android build run tests on device or it just builds like iOS? |
|
It runs tests in the standard emulator (does both) |
Fixed iOS build after oibit Reviewed-by: alexcrichton
No description provided.