Fix for bad get unwrap suggestion

[ms2300]

Fix for issue outlined in #3006, get() returns either a reference or a slice and therefore "&" should only be added when get is returning a reference ( ie .get(1) instead of .get(0..1) )

[ms2300]

Access denied in dogfood test?

[flip1995]

That is unrelated: #3118

But it seems this PR produces ICEs in cargo, rls and hyper. Could you look into this? If you have any questions feel free to ask, we're happy to help!

[ms2300]

I should have that issue fixed ... thanks for pointing that out

Edit: I rebased and am now failing tests on different part of code?

[mati865]

Master fails with the same error(outdated test):

diff of stderr:
 error: casting from `*const u8` to a more-strictly-aligned pointer (`*const u16`)
   --> $DIR/cast_alignment.rs:15:5
    |
 15 |     (&1u8 as *const u8) as *const u16;
    |     ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
    |
    = note: `-D clippy::cast-ptr-alignment` implied by `-D warnings`
 error: casting from `*mut u8` to a more-strictly-aligned pointer (`*mut u16`)
   --> $DIR/cast_alignment.rs:16:5
    |
 16 |     (&mut 1u8 as *mut u8) as *mut u16;
    |     ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
-error: aborting due to 2 previous errors
+error: casting from `*const std::ffi::c_void` to a more-strictly-aligned pointer (`*const u32`)
+  --> $DIR/cast_alignment.rs:26:5
+   |
+26 |     (&1u32 as *const u32 as *const std::os::raw::c_void) as *const u32;
+   |     ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+error: aborting due to 3 previous errors

[flip1995]

Fixed in #3194. You can rebase again and then travis should pass.

[phansch]

thanks!

[flip1995]

Could you leave the requested comment in #3230 ?

[Manishearth]

Ping @ms2300 could you look at #3230 and leave the requested relicensing comment?