]> git.lizzy.rs Git - rust.git/log
rust.git
9 years agorollup merge of #21252: nikomatsakis/assoc-type-ice-hunt-take-2
Alex Crichton [Wed, 21 Jan 2015 19:51:10 +0000 (11:51 -0800)]
rollup merge of #21252: nikomatsakis/assoc-type-ice-hunt-take-2

Project region bounds out of the trait when deciding whether a projection type outlives a given regions.

Fixes #20890.
Fixes #21150.

9 years agorollup merge of #20642: michaelwoerister/sane-source-locations-pt1
Alex Crichton [Wed, 21 Jan 2015 19:50:34 +0000 (11:50 -0800)]
rollup merge of #20642: michaelwoerister/sane-source-locations-pt1

Conflicts:
src/librustc_trans/trans/debuginfo.rs

9 years agoMove regression test for #20971 into run-fail, since it panics.
Niko Matsakis [Wed, 21 Jan 2015 19:35:41 +0000 (14:35 -0500)]
Move regression test for #20971 into run-fail, since it panics.

9 years agoTest fixes and rebase conflicts
Alex Crichton [Wed, 21 Jan 2015 17:23:27 +0000 (09:23 -0800)]
Test fixes and rebase conflicts

9 years agorollup merge of #21463: sanxiyn/demut
Alex Crichton [Wed, 21 Jan 2015 17:21:09 +0000 (09:21 -0800)]
rollup merge of #21463: sanxiyn/demut

9 years agorollup merge of #21462: ahmedcharles/remove-ratchet
Alex Crichton [Wed, 21 Jan 2015 17:21:06 +0000 (09:21 -0800)]
rollup merge of #21462: ahmedcharles/remove-ratchet

Conflicts:
src/libtest/lib.rs

9 years agorollup merge of #21457: alexcrichton/issue-21436
Alex Crichton [Wed, 21 Jan 2015 17:20:35 +0000 (09:20 -0800)]
rollup merge of #21457: alexcrichton/issue-21436

Conflicts:
src/liballoc/boxed.rs
src/librustc/middle/traits/error_reporting.rs
src/libstd/sync/mpsc/mod.rs

9 years agorollup merge of #21447: cmr/master
Alex Crichton [Wed, 21 Jan 2015 17:19:03 +0000 (09:19 -0800)]
rollup merge of #21447: cmr/master

Closes #13971

9 years agorollup merge of #21446: stepancheg/boxed-test
Alex Crichton [Wed, 21 Jan 2015 17:18:56 +0000 (09:18 -0800)]
rollup merge of #21446: stepancheg/boxed-test

Conflicts:
src/liballoc/boxed.rs

9 years agorollup merge of #21444: petrochenkov/null
Alex Crichton [Wed, 21 Jan 2015 17:18:07 +0000 (09:18 -0800)]
rollup merge of #21444: petrochenkov/null

Conflicts:
src/libstd/sync/mpsc/select.rs

9 years agorollup merge of #21441: alexcrichton/rustc-opts
Alex Crichton [Wed, 21 Jan 2015 17:17:43 +0000 (09:17 -0800)]
rollup merge of #21441: alexcrichton/rustc-opts

This is a bit of cleanup work to clean out some old deprecated flags and deprecated lint names from the compiler (they've been deprecated for quite awhile now).

This also notably puts `--pretty` behind the `-Z unstable-options` flag (where it was supposed to be previously).

9 years agorollup merge of #21438: taralx/kill-racycell
Alex Crichton [Wed, 21 Jan 2015 17:17:40 +0000 (09:17 -0800)]
rollup merge of #21438: taralx/kill-racycell

Conflicts:
src/libstd/sync/mpsc/mod.rs

9 years agorollup merge of #21437: FlaPer87/snapshot
Alex Crichton [Wed, 21 Jan 2015 17:16:40 +0000 (09:16 -0800)]
rollup merge of #21437: FlaPer87/snapshot

r? @alexcrichton

9 years agorollup merge of #21433: alfie/typobook
Alex Crichton [Wed, 21 Jan 2015 17:16:38 +0000 (09:16 -0800)]
rollup merge of #21433: alfie/typobook

Tiny fix

9 years agorollup merge of #21429: GuillaumeGomez/macro-fix
Alex Crichton [Wed, 21 Jan 2015 17:16:37 +0000 (09:16 -0800)]
rollup merge of #21429: GuillaumeGomez/macro-fix

This is little clean code of this PR: #21366. I patched the same thing as aochagavia but too slowly obviously. This is a merge of our two codes, more "rust-like".

9 years agorollup merge of #21423: oli-obk/prettier_read_until
Alex Crichton [Wed, 21 Jan 2015 17:16:33 +0000 (09:16 -0800)]
rollup merge of #21423: oli-obk/prettier_read_until

Conflicts:
src/libstd/io/mod.rs

9 years agorollup merge of #21421: huonw/one-suggestion-per-trait
Alex Crichton [Wed, 21 Jan 2015 17:16:05 +0000 (09:16 -0800)]
rollup merge of #21421: huonw/one-suggestion-per-trait

This is clearly useless, the user doesn't need to know that they could
implement/import `foo::bar::Baz` 4 times.

Fixes #21405.

9 years agorollup merge of #21419: Toby-S/patch-1
Alex Crichton [Wed, 21 Jan 2015 17:16:03 +0000 (09:16 -0800)]
rollup merge of #21419: Toby-S/patch-1

This just corrects a couple of typos in doc comments, and changes some to conform to the Rust guidelines.

9 years agorollup merge of #21418: Aatch/assume-refcount
Alex Crichton [Wed, 21 Jan 2015 17:16:01 +0000 (09:16 -0800)]
rollup merge of #21418: Aatch/assume-refcount

The reference count can never be 0, unless we're about to drop the data
completely. Using the `assume` intrinsic allows us to inform LLVM about
that invariant, meaning it can avoid unnecessary drops.

---

Before and after IR: https://gist.github.com/Aatch/3786d20df2edaad6a0e8

Generated from the example in #13018

Fixes #13018

9 years agorollup merge of #21414: ejjeong/aarch64-linux-android
Alex Crichton [Wed, 21 Jan 2015 17:15:59 +0000 (09:15 -0800)]
rollup merge of #21414: ejjeong/aarch64-linux-android

Initial support for aarch64-linux-android (#18920)
- Add new configuration files
- Modify some options to compile & link succesfully.
  (PIE, disable tls on jemalloc, modify some external function linkage, ..)
- To build, refer to https://github.com/rust-lang/rust/wiki/Doc-building-for-android.
   (tested with platform=21 and toolchain=aarch64-linux-android-4.9)

9 years agorollup merge of #21413: ahmedcharles/remove-test-features
Alex Crichton [Wed, 21 Jan 2015 17:15:57 +0000 (09:15 -0800)]
rollup merge of #21413: ahmedcharles/remove-test-features

This isn't the entire set of changes, there are more coming. #19145

9 years agorollup merge of #21411: P1start/help-tweaks
Alex Crichton [Wed, 21 Jan 2015 17:15:54 +0000 (09:15 -0800)]
rollup merge of #21411: P1start/help-tweaks

Conflicts:
src/librustc_typeck/check/closure.rs

9 years agorollup merge of #21396: japaric/no-parens-in-range
Alex Crichton [Wed, 21 Jan 2015 17:15:15 +0000 (09:15 -0800)]
rollup merge of #21396: japaric/no-parens-in-range

Conflicts:
src/libsyntax/parse/lexer/comments.rs

9 years agorollup merge of #21394: japaric/nonono
Alex Crichton [Wed, 21 Jan 2015 17:14:42 +0000 (09:14 -0800)]
rollup merge of #21394: japaric/nonono

r? @FlaPer87

9 years agorollup merge of #21393: loganchien/fix-if-stmt-doc-title
Alex Crichton [Wed, 21 Jan 2015 17:14:41 +0000 (09:14 -0800)]
rollup merge of #21393: loganchien/fix-if-stmt-doc-title

Slightly change the title to make it look more consistent with other chapters (e.g. Match.)

9 years agorollup merge of #21392: japaric/iter
Alex Crichton [Wed, 21 Jan 2015 17:14:39 +0000 (09:14 -0800)]
rollup merge of #21392: japaric/iter

closes #20953
closes #21361

---

In the future, we will likely derive these `impl`s via syntax extensions or using compiler magic (see #20617). For the time being we can use these manual `impl`s.

r? @aturon
cc @burntsushi @Kroisse

9 years agorollup merge of #21391: klutzy/rt-time-cleanup
Alex Crichton [Wed, 21 Jan 2015 17:14:37 +0000 (09:14 -0800)]
rollup merge of #21391: klutzy/rt-time-cleanup

They are unused since libtime is gone.

cc #20861

9 years agorollup merge of #21389: retep998/timer
Alex Crichton [Wed, 21 Jan 2015 17:14:36 +0000 (09:14 -0800)]
rollup merge of #21389: retep998/timer

Fixes #20943 and adds a test for it
r? @alexcrichton

9 years agorollup merge of #21372: arielb1/remove-the-box
Alex Crichton [Wed, 21 Jan 2015 17:14:34 +0000 (09:14 -0800)]
rollup merge of #21372: arielb1/remove-the-box

It is not used anymore

9 years agorollup merge of #21368: tomjakubowski/rustdoc-miscellany
Alex Crichton [Wed, 21 Jan 2015 17:14:30 +0000 (09:14 -0800)]
rollup merge of #21368: tomjakubowski/rustdoc-miscellany

Conflicts:
src/librustdoc/clean/mod.rs

9 years agorollup merge of #21367: steveklabnik/remove_gate
Alex Crichton [Wed, 21 Jan 2015 17:13:57 +0000 (09:13 -0800)]
rollup merge of #21367: steveklabnik/remove_gate

This gate was `Accepted`, so we shouldn't need these.

9 years agorollup merge of #21354: eddyb/vec-ufcs
Alex Crichton [Wed, 21 Jan 2015 17:13:55 +0000 (09:13 -0800)]
rollup merge of #21354: eddyb/vec-ufcs

There are two limitations to the macro that this addresses:
1. the expected type is not propagated, coercions don't trigger
2. references inside element expressions don't outlive the `Vec`

Both of these limitations are caused by the block in the
macro expansion, previously needed to trigger a coercion
from `Box<[T; N]>` to `Box<[T]>`, now possible with UFCS.

9 years agorollup merge of #21342: Diggsey/issue-21310
Alex Crichton [Wed, 21 Jan 2015 17:13:53 +0000 (09:13 -0800)]
rollup merge of #21342: Diggsey/issue-21310

Fixes #21310

9 years agorollup merge of #21340: pshc/libsyntax-no-more-ints
Alex Crichton [Wed, 21 Jan 2015 17:13:51 +0000 (09:13 -0800)]
rollup merge of #21340: pshc/libsyntax-no-more-ints

Collaboration with @rylev!

I didn't change `int` in the [quasi-quoter](https://github.com/pshc/rust/blob/99ae1a30f3ca28c0f7e431620560d30e44627124/src/libsyntax/ext/quote.rs#L328), because I'm not sure if there will be adverse effects.

Addresses #21095.

9 years agorollup merge of #21333: stepancheg/trans-write-diag
Alex Crichton [Wed, 21 Jan 2015 17:13:49 +0000 (09:13 -0800)]
rollup merge of #21333: stepancheg/trans-write-diag

File cannot be written, for example, if directory does not exist.

Before this commit:

```
% rustc -o nonexistent/program program.rs
error: could not write output: No such file or directory
```

With this commit:

```
% rustc -o nonexistent/program program.rs
error: could not write output to nonexistent/program.0.o: No such file or directory
```

This is useful when full rust command is not displayed, or when last error is followed by thousands of warnings.

9 years agorollup merge of #21289: brson/errorcodes
Alex Crichton [Wed, 21 Jan 2015 17:13:46 +0000 (09:13 -0800)]
rollup merge of #21289: brson/errorcodes

This does the bare minimum to make registration of error codes work again. After this patch, every call to `span_err!` with an error code gets that error code validated against a list in that crate and a new tidy script `errorck.py` validates that no error codes are duplicated globally.

There are further improvements to be made yet, detailed in #19624.

r? @nikomatsakis

9 years agorollup merge of #21274: estsauver/21270
Alex Crichton [Wed, 21 Jan 2015 17:13:43 +0000 (09:13 -0800)]
rollup merge of #21274: estsauver/21270

The timezone of the server that builds rustc nightlies
appears to be in UTC. From what I can tell, it builds
the nightlies at 0300 UTC, which takes about ~15 minutes.
So, there were a couple options here for which offset to use.

UTC+3 would ensure that you always got the latest rust, but it
would mean the script is broken ~15/20 minutes a day. UTC+4
means users get a stale nightly for 45 minutes, but that feels
okay to me.

Ideally, buildbot would publish the "current" nightly, but
that seems unneeded relative to fixing it for all Timezones
quickly.

Fixes #21270

9 years agorollup merge of #21157: steveklabnik/deprecation
Alex Crichton [Wed, 21 Jan 2015 17:13:42 +0000 (09:13 -0800)]
rollup merge of #21157: steveklabnik/deprecation

9 years agorollup merge of #21138: mkpankov/master
Alex Crichton [Wed, 21 Jan 2015 17:13:40 +0000 (09:13 -0800)]
rollup merge of #21138: mkpankov/master

When combined with '--save' and '--date', it uses previously saved
tarball, making possible to re-install in offline mode.
r?

9 years agorollup merge of #21053: apasel422/exact
Alex Crichton [Wed, 21 Jan 2015 17:13:38 +0000 (09:13 -0800)]
rollup merge of #21053: apasel422/exact

9 years agorollup merge of #19913: KOMON/rust-mode-emacs-indentation
Alex Crichton [Wed, 21 Jan 2015 17:13:36 +0000 (09:13 -0800)]
rollup merge of #19913: KOMON/rust-mode-emacs-indentation

I added an option to auto-indent method chains to line up along their '.' operators. Like so:

```
let input = io::stdin().readline()
                       .ok()
                       .expect("Failed to read line");
```

The old default would indent like so:
```
let input = io::stdin().readme()
    .ok()
    .expect("Failed to read line");
```

The Rust guide explicitly condones the former, so I thought it would be nice for the emacs mode to support it. It's off by default, you have to set ```rust-indent-method-chain``` to ```t``` via your .emacs or the customize menu

9 years agoImprove quality of String#to_slice() deprecation message
Steve Klabnik [Wed, 14 Jan 2015 21:05:32 +0000 (16:05 -0500)]
Improve quality of String#to_slice() deprecation message

9 years agoDe-mut the parser
Seo Sanghyeon [Wed, 21 Jan 2015 11:44:49 +0000 (20:44 +0900)]
De-mut the parser

9 years agoAuto merge of #21242 - richo:no-perl, r=brson
bors [Wed, 21 Jan 2015 11:07:31 +0000 (11:07 +0000)]
Auto merge of #21242 - richo:no-perl, r=brson

There's only one build-critical path in which perl is used, and it was to do a text replacement trivially achievable with sed(1).

I ported the indenter script because it [appears to be used][indenter], but removed check links because it appears to be entirely out of date.

[indenter]: https://github.com/rust-lang/rust/blob/master/src/librustc/util/common.rs#L60-70

9 years agodebuginfo: Make debuginfo source location assignment more stable (Pt. 1)
Michael Woerister [Thu, 11 Dec 2014 12:53:30 +0000 (13:53 +0100)]
debuginfo: Make debuginfo source location assignment more stable (Pt. 1)

So far, the source location an LLVM instruction was linked to was controlled by
`debuginfo::set_source_location()` and `debuginfo::clear_source_location()`.
This interface mimicked how LLVM's `IRBuilder` handles debug location
assignment. While this interface has some theoretical performance benefits, it
also makes things terribly unstable: One sets some quasi-global state and then
hopes that it is still correct when a given instruction is emitted---an
assumption that has been proven to not hold a bit too often.

This patch requires the debug source location to be passed to the actual
instruction emitting function. This makes source location assignment explicit
and will prevent future changes to `trans` from accidentally breaking things in
the majority of cases.

This patch does not yet implement the new principle for all instruction kinds
but the stepping experience should have improved significantly nonetheless
already.

9 years agoAuto merge of #21227 - sellibitze:core-ops-for-references, r=aturon
bors [Wed, 21 Jan 2015 09:02:44 +0000 (09:02 +0000)]
Auto merge of #21227 - sellibitze:core-ops-for-references, r=aturon

As discussed with @aturon I added implementations of various op traits for references to built-in types which was already suggested by the ops reform RFC.

The 2nd commit updates the module documentation of core::ops to fully reflect the recent change from pass-by-reference to pass-by-value and expands on the implications for generic code.

9 years agoRemove ratchet().
Ahmed Charles [Tue, 20 Jan 2015 19:14:52 +0000 (11:14 -0800)]
Remove ratchet().

9 years agoRemove compare_to_old().
Ahmed Charles [Tue, 20 Jan 2015 17:24:06 +0000 (09:24 -0800)]
Remove compare_to_old().

9 years agoRemove write_metric_diff and supporting code.
Ahmed Charles [Mon, 19 Jan 2015 10:51:34 +0000 (02:51 -0800)]
Remove write_metric_diff and supporting code.

9 years agostd: Rename Show/String to Debug/Display
Alex Crichton [Tue, 20 Jan 2015 23:45:07 +0000 (15:45 -0800)]
std: Rename Show/String to Debug/Display

This commit is an implementation of [RFC 565][rfc] which is a stabilization of
the `std::fmt` module and the implementations of various formatting traits.
Specifically, the following changes were performed:

[rfc]: https://github.com/rust-lang/rfcs/blob/master/text/0565-show-string-guidelines.md

* The `Show` trait is now deprecated, it was renamed to `Debug`
* The `String` trait is now deprecated, it was renamed to `Display`
* Many `Debug` and `Display` implementations were audited in accordance with the
  RFC and audited implementations now have the `#[stable]` attribute
  * Integers and floats no longer print a suffix
  * Smart pointers no longer print details that they are a smart pointer
  * Paths with `Debug` are now quoted and escape characters
* The `unwrap` methods on `Result` now require `Display` instead of `Debug`
* The `Error` trait no longer has a `detail` method and now requires that
  `Display` must be implemented. With the loss of `String`, this has moved into
  libcore.
* `impl<E: Error> FromError<E> for Box<Error>` now exists
* `derive(Show)` has been renamed to `derive(Debug)`. This is not currently
  warned about due to warnings being emitted on stage1+

While backwards compatibility is attempted to be maintained with a blanket
implementation of `Display` for the old `String` trait (and the same for
`Show`/`Debug`) this is still a breaking change due to primitives no longer
implementing `String` as well as modifications such as `unwrap` and the `Error`
trait. Most code is fairly straightforward to update with a rename or tweaks of
method calls.

[breaking-change]
Closes #21436

9 years agoAuto merge of #21395 - sfackler:fix-cvar-test, r=alexcrichton
bors [Wed, 21 Jan 2015 05:55:02 +0000 (05:55 +0000)]
Auto merge of #21395 - sfackler:fix-cvar-test, r=alexcrichton

r? @alexcrichton

9 years agoremove `NoSendItem` and `NoSyncItem`
Jorge Aparicio [Mon, 19 Jan 2015 15:52:11 +0000 (10:52 -0500)]
remove `NoSendItem` and `NoSyncItem`

9 years agoDisplay negative trait implementations correctly in rustdoc
Diggory Blake [Wed, 21 Jan 2015 04:35:57 +0000 (04:35 +0000)]
Display negative trait implementations correctly in rustdoc

Added doc test

9 years agoAuto merge of #21065 - ColonelJ:master, r=brson
bors [Wed, 21 Jan 2015 03:54:21 +0000 (03:54 +0000)]
Auto merge of #21065 - ColonelJ:master, r=brson

Removed use of unused LDPATH variable on Windows as is done for other platforms, and added GCC flag to ensure MINGW's ANSI compatible STDIO functions are used wherever available (required by jemalloc).

Without these changes it ends up setting the PATH twice, and the second time the PATH begins with `:` which is invalid.  Also the regular msvcrt printf-like functions would be used which don't understand stuff like %hhd and %z which jemalloc uses.

This change ought not to make any difference to the output but it fixes the build process for me since at least my build environment couldn't handle that broken path caused by LDPATH being empty.

9 years agoimplement `ExactSizeIterator` for `slice::{Chunks,ChunksMut}`
Andrew Paseltiner [Mon, 12 Jan 2015 22:08:53 +0000 (17:08 -0500)]
implement `ExactSizeIterator` for `slice::{Chunks,ChunksMut}`

9 years agotests: uncomment regression tests for 13970, 13971, 13972
Corey Richardson [Tue, 20 Jan 2015 23:43:38 +0000 (18:43 -0500)]
tests: uncomment regression tests for 13970, 13971, 13972

Closes #13971

9 years agoalso forward Iterator::size_hint()
Jorge Aparicio [Tue, 20 Jan 2015 23:15:28 +0000 (18:15 -0500)]
also forward Iterator::size_hint()

9 years agoAuto merge of #21439 - alexcrichton:rollup, r=alexcrichton
bors [Tue, 20 Jan 2015 23:03:09 +0000 (23:03 +0000)]
Auto merge of #21439 - alexcrichton:rollup, r=alexcrichton

Continuation of https://github.com/rust-lang/rust/pull/21428

9 years agoMake multiline errors work with codes
Brian Anderson [Tue, 20 Jan 2015 22:18:35 +0000 (14:18 -0800)]
Make multiline errors work with codes

9 years agoChanged rust-mode code for tabs -> spaces
Daniel Raloff [Tue, 20 Jan 2015 22:07:10 +0000 (14:07 -0800)]
Changed rust-mode code for tabs -> spaces

9 years agoadd stability text like suggested in discussion
Sebastian Gesemann [Tue, 20 Jan 2015 21:35:42 +0000 (22:35 +0100)]
add stability text like suggested in discussion

9 years agoRegister snapshot for 9006c3c
Flavio Percoco [Tue, 20 Jan 2015 14:51:09 +0000 (15:51 +0100)]
Register snapshot for 9006c3c

9 years agoalloc::boxed: enable test
Stepan Koltsov [Tue, 20 Jan 2015 20:57:56 +0000 (23:57 +0300)]
alloc::boxed: enable test

Previously test was disabled due to `#[cfg(test)]` before `mod boxed`.

9 years agoAdd more explanation for why the assumes are there
James Miller [Tue, 20 Jan 2015 20:56:59 +0000 (09:56 +1300)]
Add more explanation for why the assumes are there

9 years agorustc: Remove deprecated lint names
Alex Crichton [Tue, 20 Jan 2015 19:20:22 +0000 (11:20 -0800)]
rustc: Remove deprecated lint names

These were all renamed quite some time ago, so remove their old names from the
compiler.

9 years agoTest fixes and rebase conflicts
Alex Crichton [Tue, 20 Jan 2015 20:35:51 +0000 (12:35 -0800)]
Test fixes and rebase conflicts

9 years agoAdd assumptions that the pointer is non-null
James Miller [Tue, 20 Jan 2015 20:35:24 +0000 (09:35 +1300)]
Add assumptions that the pointer is non-null

9 years agoKill RacyCell in favor of marking SyncSender explicitly Send.
JP Sugarbroad [Tue, 20 Jan 2015 20:00:12 +0000 (12:00 -0800)]
Kill RacyCell in favor of marking SyncSender explicitly Send.

9 years agoAuto merge of #20919 - fay-jai:update, r=brson
bors [Tue, 20 Jan 2015 19:56:43 +0000 (19:56 +0000)]
Auto merge of #20919 - fay-jai:update, r=brson

Only made 2 changes:
1) Update the year to 2015 in LICENSE-MIT
2) Update the year in COPYRIGHT

No other changes were made.

9 years agoAdd some extended errors.
Michael Sproul [Sun, 18 Jan 2015 01:20:24 +0000 (17:20 -0800)]
Add some extended errors.

9 years agoAdd error codes to rustc
Brian Anderson [Mon, 19 Jan 2015 00:58:25 +0000 (16:58 -0800)]
Add error codes to rustc

9 years agoAdd error codes to resolve
Brian Anderson [Sun, 18 Jan 2015 21:46:57 +0000 (13:46 -0800)]
Add error codes to resolve

9 years agoMake fatal errors work with codes, add to typeck
Brian Anderson [Sun, 18 Jan 2015 21:39:18 +0000 (13:39 -0800)]
Make fatal errors work with codes, add to typeck

9 years agoAdd error codes to rustc_typeck
Brian Anderson [Sat, 17 Jan 2015 21:36:59 +0000 (13:36 -0800)]
Add error codes to rustc_typeck

9 years agoMake error code registration work again. #19624
Brian Anderson [Fri, 16 Jan 2015 23:54:58 +0000 (15:54 -0800)]
Make error code registration work again. #19624

9 years agorustc: Remove deprecated flags
Alex Crichton [Tue, 20 Jan 2015 18:57:10 +0000 (10:57 -0800)]
rustc: Remove deprecated flags

This commit removes a number of deprecated flags from the compiler:

* opt-level => -C opt-level
* debuginfo => -C debuginfo
* print-crate-name => --print crate-name
* print-file-name => --print file-names
* no-trans => -Z no-trans
* no-analysis => -Z no-analysis
* parse-only => -Z parse-only
* dep-info => --emit dep-info

This commit also moves the --pretty flag behind `-Z unstable-options` as the
pretty printer will likely not be stable for 1.0

cc #19051

9 years agodocs: typo
Alfie John [Tue, 20 Jan 2015 19:02:09 +0000 (19:02 +0000)]
docs: typo

9 years agoRollup merge of #20991 - mneumann:llvm-dragonfly-take2, r=alexcrichton
Barosl LEE [Tue, 20 Jan 2015 17:16:53 +0000 (02:16 +0900)]
Rollup merge of #20991 - mneumann:llvm-dragonfly-take2, r=alexcrichton

It got accidentially reverted in 44440e5.

9 years agoRollup merge of #21179 - nodakai:reference-mod-to-self, r=huonw
Barosl LEE [Tue, 20 Jan 2015 17:16:53 +0000 (02:16 +0900)]
Rollup merge of #21179 - nodakai:reference-mod-to-self, r=huonw

This should have been done together with 56dcbd17fdad5d39b7b02e22a7490d2468718d08 for rust-lang/rust#20361

9 years agoRollup merge of #21427 - steveklabnik:generics_fix, r=alexcrichton
Barosl LEE [Tue, 20 Jan 2015 17:16:52 +0000 (02:16 +0900)]
Rollup merge of #21427 - steveklabnik:generics_fix, r=alexcrichton

Multiple people have asked me if this is a reference to Hacker News, and
I _certainly_ don't want to give them that impression.

9 years agoRollup merge of #21424 - sanxiyn:coerce-mut, r=nikomatsakis
Barosl LEE [Tue, 20 Jan 2015 17:16:52 +0000 (02:16 +0900)]
Rollup merge of #21424 - sanxiyn:coerce-mut, r=nikomatsakis

This is caught in borrowck now, but catching in typeck is faster and improves diagnostics.

CC #17561.

r? @nikomatsakis

9 years agoRollup merge of #21404 - japaric:hash, r=alexcrichton
Barosl LEE [Tue, 20 Jan 2015 17:16:52 +0000 (02:16 +0900)]
Rollup merge of #21404 - japaric:hash, r=alexcrichton

closes #21402
cc #15294

r? @alexcrichton or @aturon
cc @ExpHP (btw, this only covers arrays with arity up to 32)

9 years agoRollup merge of #21399 - kballard:fix-PLEASE_BENCH, r=Gankro
Barosl LEE [Tue, 20 Jan 2015 17:16:52 +0000 (02:16 +0900)]
Rollup merge of #21399 - kballard:fix-PLEASE_BENCH, r=Gankro

611ef49f2fa573edf9cff4442eddb8ee7e48878d removed all the metrics stuff
from tests.mk, but this meant that `PLEASE_BENCH=1` no longer did
anything.

Fixes #21324.

9 years agoRollup merge of #21388 - aochagavia:collections, r=Gankro
Barosl LEE [Tue, 20 Jan 2015 17:16:51 +0000 (02:16 +0900)]
Rollup merge of #21388 - aochagavia:collections, r=Gankro

**Breaking change**: `VecMap::into_iter` now consumes the `VecMap`. To fix it you should pass the VecMap by value instead of by reference.

[breaking-change]

r? @Gankro

9 years agoRollup merge of #21387 - retep998:hmodule, r=alexcrichton
Barosl LEE [Tue, 20 Jan 2015 17:16:51 +0000 (02:16 +0900)]
Rollup merge of #21387 - retep998:hmodule, r=alexcrichton

r? @alexcrichton

9 years agoRollup merge of #21386 - Diggsey:issue-21384, r=alexcrichton
Barosl LEE [Tue, 20 Jan 2015 17:16:51 +0000 (02:16 +0900)]
Rollup merge of #21386 - Diggsey:issue-21384, r=alexcrichton

Fixes #21384

9 years agoRollup merge of #21380 - tshepang:patch-3, r=steveklabnik
Barosl LEE [Tue, 20 Jan 2015 17:16:51 +0000 (02:16 +0900)]
Rollup merge of #21380 - tshepang:patch-3, r=steveklabnik

9 years agoRollup merge of #21123 - visualfc:master, r=alexcrichton
Barosl LEE [Tue, 20 Jan 2015 17:16:51 +0000 (02:16 +0900)]
Rollup merge of #21123 - visualfc:master, r=alexcrichton

example:
let m = "hello \
           world";

9 years agoRollup merge of #21377 - iKevinY:speedy-tidy, r=huonw
Barosl LEE [Tue, 20 Jan 2015 17:16:50 +0000 (02:16 +0900)]
Rollup merge of #21377 - iKevinY:speedy-tidy, r=huonw

`x in y` is more Pythonic than `y.find(x) != -1`. I believe it runs quite a bit faster as well (though it's probably not a bottleneck of the Travis builds):

```bash
$ python -m timeit '"abc".find("a") != -1'
1000000 loops, best of 3: 0.218 usec per loop
$ python -m timeit '"a" in "abc"'
10000000 loops, best of 3: 0.0343 usec per loop
```

9 years agoRollup merge of #21375 - petrochenkov:ssbsl, r=alexcrichton
Barosl LEE [Tue, 20 Jan 2015 17:16:50 +0000 (02:16 +0900)]
Rollup merge of #21375 - petrochenkov:ssbsl, r=alexcrichton

After PR #19766 added implicit coersions `*mut T -> *const T`, the explicit casts can be removed.
(The number of such casts turned out to be relatively small).

9 years agoRollup merge of #21369 - iKevinY:no-travis-notes, r=sanxiyn
Barosl LEE [Tue, 20 Jan 2015 17:16:49 +0000 (02:16 +0900)]
Rollup merge of #21369 - iKevinY:no-travis-notes, r=sanxiyn

Updated `tidy.py` to skip printing NOTEs if the [`TRAVIS`](http://docs.travis-ci.com/user/ci-environment/#Environment-variables) environment variable is set.

9 years agoRollup merge of #21359 - WiSaGaN:bugfix/fix_marker, r=alexcrichton
Barosl LEE [Tue, 20 Jan 2015 17:16:49 +0000 (02:16 +0900)]
Rollup merge of #21359 - WiSaGaN:bugfix/fix_marker, r=alexcrichton

From std::markers to std::marker.

9 years agoRollup merge of #21358 - glacjay:patch-2, r=alexcrichton
Barosl LEE [Tue, 20 Jan 2015 17:16:49 +0000 (02:16 +0900)]
Rollup merge of #21358 - glacjay:patch-2, r=alexcrichton

9 years agoRollup merge of #21357 - kimroen:patch-1, r=sanxiyn
Barosl LEE [Tue, 20 Jan 2015 17:16:49 +0000 (02:16 +0900)]
Rollup merge of #21357 - kimroen:patch-1, r=sanxiyn

Having both "Right now" and "at the moment" in the same statement is redundant.

9 years agoRollup merge of #21100 - tstorch:small_readability_update, r=alexcrichton
Barosl LEE [Tue, 20 Jan 2015 17:16:48 +0000 (02:16 +0900)]
Rollup merge of #21100 - tstorch:small_readability_update, r=alexcrichton

Why not use what is there?

9 years agoRollup merge of #21345 - glacjay:patch-1, r=alexcrichton
Barosl LEE [Tue, 20 Jan 2015 17:16:48 +0000 (02:16 +0900)]
Rollup merge of #21345 - glacjay:patch-1, r=alexcrichton

The reference should be `x`, not `FOO` itself.

9 years agoRollup merge of #21339 - thorncp:api-docs-search, r=alexcrichton
Barosl LEE [Tue, 20 Jan 2015 17:16:48 +0000 (02:16 +0900)]
Rollup merge of #21339 - thorncp:api-docs-search, r=alexcrichton

Increases the delay of the search box to 500ms after key up. I tried
adding a three character minimum for setting the delay, but didn't find
it very useful.

Should close #20095

@Jurily, your input is welcome!

9 years agoRollup merge of #21336 - rylev:better-nofile-error, r=brson
Barosl LEE [Tue, 20 Jan 2015 17:16:48 +0000 (02:16 +0900)]
Rollup merge of #21336 - rylev:better-nofile-error, r=brson

Contribution from @look!

Addresses https://github.com/rust-lang/rust/issues/21329

9 years agoRollup merge of #21331 - michaelsproul:sync-error-impls, r=alexcrichton
Barosl LEE [Tue, 20 Jan 2015 17:16:47 +0000 (02:16 +0900)]
Rollup merge of #21331 - michaelsproul:sync-error-impls, r=alexcrichton

Two errors in `std::sync` are currently missing implementations of the standard error trait because they contain types which aren't `Send`.

This PR therefore requires #21312.

9 years agoRollup merge of #21048 - aroben:patch-1, r=steveklabnik
Barosl LEE [Tue, 20 Jan 2015 17:16:47 +0000 (02:16 +0900)]
Rollup merge of #21048 - aroben:patch-1, r=steveklabnik

Now both the enum values and the prose describing them mention the values in the same order.