Commit Graph

69669 Commits

Author SHA1 Message Date
Alex Crichton
c11ac9441a rustbuild: Update the OpenSSL version to link
This updates the OpenSSL tarball download to reflect OpenSSL's newest release.
2017-11-04 08:05:46 -07:00
Marc-Antoine Perennou
784528b8da rustbuild: don't try to install rls if ToolState is not Testing
The Dist Step is not ran in that case so we would end up trying to
install something that we didn't dist.

Signed-off-by: Marc-Antoine Perennou <Marc-Antoine@Perennou.com>
2017-11-04 15:49:18 +01:00
Nadav Zingerman
2304b3a7f1 Add comment explaining the ctxt field in Span 2017-11-04 15:54:03 +02:00
bors
dcd343bfbc Auto merge of - GuillaumeGomez:search-fixes, r=QuietMisdreavus
Search fixes

Fixes .

r? @QuietMisdreavus
2017-11-04 12:54:34 +00:00
bors
a6885cb853 Auto merge of - Nashenas88:derive-newtype, r=nikomatsakis
Add derive and doc comment capabilities to newtype_index macro

This moves `RustcDecodable` and `RustcEncodable` out of the macro definition and into the macro uses. They were conflicting with `CrateNum`'s impls of `serialize::UseSpecializedEncodable` and `serialize::UseSpecializedDecodable`, and now it's not :). `CrateNum` is now defined with the `newtype_index` macro. I also added support for doc comments on constant definitions and allowed a type to remove the pub specification on the tuple param (otherwise a LOT of code would refuse to compile for `CrateNum`). I was getting dozens of errors like this if `CrateNum` was defined as `pub struct CrateNum(pub u32)`:
```
error[E0530]: match bindings cannot shadow tuple structs
   --> src/librustc/dep_graph/dep_node.rs:624:25
    |
63  | use hir::def_id::{CrateNum, DefId, DefIndex, CRATE_DEF_INDEX};
    |                   -------- a tuple struct `CrateNum` is imported here
...
624 |     [] MissingLangItems(CrateNum),
    |                         ^^^^^^^^ cannot be named the same as a tuple struct
```

I also cleaned up the formatting of the macro bodies as they were getting impossibly long. Should I go back and fix the matching rules to this style too?

I also want to see what the test results look like because `CrateNum` used to just derive `Debug`, but the `newtype_index` macro has a custom implementation. This might require further pushes.

Feel free to bikeshed on the macro language, I have no preference here.
2017-11-04 10:24:20 +00:00
topecongiro
0745733286 Fix typos 2017-11-04 18:23:54 +09:00
bors
9acc3331e1 Auto merge of - kennytm:rollup, r=kennytm
Rollup of 9 pull requests

- Successful merges: , , , , , , , , 
- Failed merges:
2017-11-04 06:59:29 +00:00
kennytm
e3b25a5942 Rollup merge of - GuillaumeGomez:methods-doc-test, r=steveklabnik
Add tests for methods listing in rust docs

r? @rust-lang/docs
2017-11-04 13:49:33 +08:00
kennytm
606e269e14 Rollup merge of - rkarp:master, r=petrochenkov
Fix libstd compile error for windows-gnu targets without `backtrace`

This is basically an addition to . Compiling `libstd` still fails when targeting `windows-gnu` with `panic = "abort"` because the items in the `...c::gnu` module are not used. They are only referenced from `backtrace_gnu.rs`, which is indirectly feature gated behind `backtrace` [here](9f3b09116b/src/libstd/sys/windows/mod.rs (L23)).
2017-11-04 13:49:33 +08:00
kennytm
7a766a4a41 Rollup merge of - mikhail-m1:improve-mir-opt-error-output, r=alexcrichton
improve compiletest output for errors from mir-opt tests

improvement:
1. Report filename against general cannot open error
2. Report current MIR block
2017-11-04 13:49:32 +08:00
kennytm
ff00a5f8fb Rollup merge of - Ljzn:patch-2, r=BurntSushi
Fix typo

`accomodate` -> `accommodate`
2017-11-04 13:49:31 +08:00
kennytm
2c293922d0 Rollup merge of - Ljzn:patch-1, r=kennytm
Fix typo.
2017-11-04 13:49:30 +08:00
kennytm
1815265510 Rollup merge of - cardoe:metadata, r=kennytm
add Cargo metadata to libstd

Add license and repository metadata to libstd
2017-11-04 13:49:29 +08:00
kennytm
ea572657da Rollup merge of - LaurentMazare:master, r=petrochenkov
Add a nicer error message for missing  in for loop, fixes .

As suggested by @estebank in issue , this works in the same way as : if the in keyword is missing, we continue parsing the expression and if this works correctly an adapted error message is produced. Otherwise we return the old error.

A specific test case has also been added.
This is my first PR on rust-lang/rust so any feedback is very welcome.
2017-11-04 13:49:28 +08:00
kennytm
ae512c4144 Rollup merge of - strake:atomic_from, r=nagisa
impl From<T> for AtomicT
2017-11-04 13:49:27 +08:00
kennytm
9ca6ee0299 Rollup merge of - ratmice:master, r=oli-obk
issue  don't build clippy stage 1


Wasn't sure top_stage was the right thing, but seemed to go ahead building clippy stage 2.
2017-11-04 13:49:26 +08:00
Scott McMurray
1b19e64324 Add overflow tests for next_power_of_two 2017-11-03 22:33:34 -07:00
Scott McMurray
15ea3d80da Fix : next_power_of_two should panic on overflow 2017-11-03 21:48:33 -07:00
bors
a4541525d5 Auto merge of - gnzlbg:jemalloc_realloc2, r=sfackler
[jemalloc] set correct excess in realloc_excess
2017-11-04 04:28:13 +00:00
Esteban Küber
9dc7abe06d Detect = -> : typo in let bindings
When encountering a let binding type error, attempt to parse as
initializer instead. If successful, it is likely just a typo:

```rust
fn main() {
    let x: Vec::with_capacity(10);
}
```

```
error: expected type, found `10`
 --> file.rs:3:31
  |
3 |     let x: Vec::with_capacity(10, 20);
  |         --                    ^^
  |         ||
  |         |help: did you mean assign here?: `=`
  |         while parsing the type for `x`
```
2017-11-03 17:39:16 -07:00
Vadim Petrochenkov
765076faab Give better help for identifier patterns failing exhaustiveness check 2017-11-04 03:09:47 +03:00
Vadim Petrochenkov
3386757923 Move some tests to UI 2017-11-04 03:09:47 +03:00
Vadim Petrochenkov
db0be32fec resolve: Use same rules for disambiguating fresh bindings in match and let 2017-11-04 03:09:47 +03:00
bors
95a401609f Auto merge of - mikhail-m1:mir_add_false_edges_terminator_kind, r=arielb1
add TerminatorKind::FalseEdges and use it in matches

impl  and fixes  right way.

False edges unexpectedly affects uninitialized variables analysis in MIR borrowck.
2017-11-04 00:09:14 +00:00
Lukas Kalbertodt
259c125267 Mark several ascii methods as unstable again
We don't want to stabilize them now already. The goal of this set of
commits is just to add inherent methods to the four types. Stabilizing
all of those methods can be done later.
2017-11-03 21:28:04 +01:00
Lukas Kalbertodt
da57580736 Remove unused AsciiExt imports and fix tests related to ascii methods
Many AsciiExt imports have become useless thanks to the inherent ascii
methods added in the last commits. These were removed. In some places, I
fully specified the ascii method being called to enforce usage of the
AsciiExt trait. Note that some imports are not removed but tagged with
a `#[cfg(stage0)]` attribute. This is necessary, because certain ascii
methods are not yet available in stage0. All those imports will be
removed later.

Additionally, failing tests were fixed. The test suite should exit
successfully now.
2017-11-03 21:27:40 +01:00
Lukas Kalbertodt
1916e3c4aa Copy AsciiExt methods to str directly
This is done in order to deprecate AsciiExt eventually. Note that
this commit contains a bunch of `cfg(stage0)` statements. This is
due to a new compiler feature this commit depends on: the
`slice_u8` lang item. Once this lang item is available in the
stage0 compiler, all those cfg flags (and more) can be removed.
2017-11-03 21:27:40 +01:00
Lukas Kalbertodt
5a1d11a733 Copy AsciiExt methods to [u8] directly
This is done in order to deprecate AsciiExt eventually. Note that
this commit contains a bunch of `cfg(stage0)` statements. This is
due to a new compiler feature I am using: the `slice_u8` lang item.
Once this lang item is available in the stage0 compiler, all those
cfg flags (and more) can be removed.
2017-11-03 21:27:40 +01:00
Simon Sapin
9e441c76f7 Add a lang item to allow impl [u8] {…} in the standard library 2017-11-03 21:27:40 +01:00
Lukas Kalbertodt
8a4fa742a1 Fix lists in doc comments for ascii methods of u8 and char 2017-11-03 21:27:40 +01:00
Lukas Kalbertodt
f373916cb5 Add missing space in match arm 2017-11-03 21:27:40 +01:00
Lukas Kalbertodt
3b13b66351 Tweak documentation for u8::eq_ignore_ascii_case() 2017-11-03 21:27:40 +01:00
Lukas Kalbertodt
a5277622c5 Remove examples in doc-comments of AsciiExt methods
The doc comments were incorrect before: since the inherent ascii methods
shadow the `AsciiExt` methods, the examples didn't use the `AsciiExt` at
all. Since the trait will be deprecated soon anyway, the easiest solution
was to remove the examples and already mention that the methods will be
deprecated in the near future.
2017-11-03 21:27:40 +01:00
Lukas Kalbertodt
6f5556854e Use direct implementation on u8/char to implement AsciiExt
The methods were copied to u8/char, so we can just use it in
AsciiExt impls to avoid duplicate code.
2017-11-03 21:27:40 +01:00
Lukas Kalbertodt
d425f8d226 Copy AsciiExt methods to char directly
This is done in order to deprecate AsciiExt eventually.
2017-11-03 21:27:40 +01:00
Lukas Kalbertodt
5061c9fecb Revert signature of eq_ignore_ascii_case() to original
Since the methods on u8 directly will shadow the AsciiExt methods,
we cannot change the signature without breaking everything. It
would have been nice to take `u8` as argument instead of `&u8`, but
we cannot break stuff! So this commit reverts it to the original
`&u8` version.
2017-11-03 21:27:40 +01:00
Lukas Kalbertodt
04070d1148 Make ascii methods on u8 insta-stable
Those methods will shadow the methods of `AsciiExt`, so if we don't
make them insta-stable, everyone will hitting stability errors. It
is fine adding those as stable, because they are just being moved
around [according to sfackler][1].

OPEN QUESTION: this commit also stabilizes the `AsciiExt` methods
that were previously feature gated by the `ascii_ctype` feature.
Maybe we don't want to stablilize those yet.

[1]: https://github.com/rust-lang/rust/pull/44042#issuecomment-329939279
2017-11-03 21:27:40 +01:00
Lukas Kalbertodt
d3f2be4bd8 Add all methods of AsciiExt to u8 directly
This is the first step in order to deprecate AsciiExt. Since
this is a WIP commit, there is still some code duplication (notably
the static arrays) that will be removed later.
2017-11-03 21:27:17 +01:00
Vadim Petrochenkov
d588f9338d Shorten paths to auxiliary files created by tests 2017-11-03 22:41:15 +03:00
David Wood
d3babe5eaa
Added page to unstable book. 2017-11-03 19:36:19 +00:00
David Wood
059eccb07f
Implemented RFC 2008 for enums (not including variants) and structs. 2017-11-03 19:36:18 +00:00
David Wood
6c19ebe128
Added tests for RFC 2008. 2017-11-03 19:36:18 +00:00
David Wood
d51ea538e4
Added feature gate for RFC 2008. 2017-11-03 19:36:17 +00:00
bors
2278506f68 Auto merge of - leodasvacas:implement-auto-trait-syntax, r=nikomatsakis
[Syntax] Implement auto trait syntax

Implements `auto trait Send {}` as a substitute for `trait Send {} impl Send for .. {}`.

See the [internals thread](https://internals.rust-lang.org/t/pre-rfc-renaming-oibits-and-changing-their-declaration-syntax/3086) for motivation. Part of .

The first commit is just a rename moving from "default trait" to "auto trait". The rest is parser->AST->HIR work and making it the same as the current syntax for everything below HIR. It's under the `optin_builtin_traits` feature gate.

When can we remove the old syntax? Do we need to wait for a new `stage0`? We also need to formally decide for the new form (even if the keyword is not settled yet).

Observations:
- If you `auto trait Auto {}` and then `impl Auto for .. {}` that's accepted even if it's redundant.
- The new syntax is simpler internally which will allow for a net removal of code, for example well-formedness checks are effectively moved to the parser.
- Rustfmt and clippy are broken, need to fix those.
- Rustdoc just ignores it for now.

ping @petrochenkov @nikomatsakis
2017-11-03 19:07:45 +00:00
leonardo.yvens
5190abb941 Fix unsafe auto trait pretty print.
It was being printed wrong as auto unsafe trait
2017-11-03 16:13:23 -02:00
leonardo.yvens
ca26f01dd1 Update pretty test for auto trait syntax. 2017-11-03 16:13:22 -02:00
leonardo.yvens
3241f4583b rustfmt is broken which makes rls broken 2017-11-03 16:13:22 -02:00
leonardo.yvens
5e74353981 fix rebase conflict 2017-11-03 16:13:22 -02:00
leonardo.yvens
27efe126e0 Rename trait_has_auto_impl to trait_is_auto 2017-11-03 16:13:22 -02:00
leonardo.yvens
94b07a91bc update unstable book and error example 2017-11-03 16:13:22 -02:00