Add unstable copy_from_slice

[strega-nil]

implements rust-lang/rfcs#1419

r? alexcrichton

[rust-highfive]

r? @aturon

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

[strega-nil]

I wrote that comment wrong :P

Hi @aturon :)

[bluss]

This method should be available in libcore too, so needs to be added to SliceExt, don't you agree?

[bluss]

Note, this does not fix the tracking issue -- the issue is open until the feature is stable.

[strega-nil]

@bluss alright, I'll take out that thing about "fixes".

Also, yes I do think it should go in SliceExt. I'll implement that tonight.

[alexcrichton]

Thanks @ubsan! In addition to what @bluss mentioned, could you also add a positive test which asserts that the behavior is as expected?

[strega-nil]

@alexcrichton will do :)

[bluss]

Great! Let's go..

@bors r+ 849c4ce

[bluss]

@bors r-

travis found this one

src/libcollections/lib.rs:35:12: 35:27 error: unused or unknown feature, #[deny(unused_features)] on by default

src/libcollections/lib.rs:35 #![feature(copy_from_slice)]

[bluss]

Ok that's really weird, probably bug #30209

[strega-nil]

@bluss that's weird, it worked on my end... I specifically added an unstable attribute, and then added a feature, and compiled and it worked.

Edit: and... it doesn't work for me if I take out the feature...

src/libcollections/slice.rs:859:9: 859:46 error: use of unstable library feature 'copy_from_slice' (see issue #31755)
src/libcollections/slice.rs:859         core_slice::SliceExt::copy_from_slice(self, src)

[bluss]

But then the libcollections code changed to just and UFCS trait method call, which is what the bug indicates

[bluss]

stage1 seems to have no such unused feature warning, but stage2 does when compiling libcollections.

[bluss]

When compiling libcollectionstest actually.

[alexcrichton]

I think you'll just need to add this to the top of libcollections:

#![cfg_attr(not(test), feature(...))]

[bluss]

Yeah that seems to be right. That was a bit confusing.

[bluss]

Did you mean to have the libcollectionstest tests replace the run-pass/run-fail tests? I.e. remove the other ones?

[strega-nil]

@bluss yes. Is that not what our tests should be? I am super confused about our tests :P

[bluss]

having them in just libcollectionstests seems good.

[strega-nil]

cool :)

[bluss]

Feel free to squash the commits together if you want (to create a cleaner history).

[strega-nil]

@bluss will do

[bluss]

Can we delete the two files in src/tests/ ? They are redundant now.

[bluss]

r=me without the redundant tests. I'm happy this is finally landing!

[strega-nil]

@bluss I thought I had deleted those :/

[strega-nil]

@bluss I feel like kind of a dumb right now :P

[alexcrichton]

@bors: r+ e12b1f9

Thanks @ubsan!

[bors]

⌛ Testing commit e12b1f9 with merge 0913004...

[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-mac-ios-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