]> git.lizzy.rs Git - rust.git/log
rust.git
10 years agoFix usage of libuv for windows
Alex Crichton [Sat, 9 Nov 2013 19:02:16 +0000 (11:02 -0800)]
Fix usage of libuv for windows

10 years agoCarefully destroy channels at the right time.
Alex Crichton [Sat, 9 Nov 2013 05:59:50 +0000 (21:59 -0800)]
Carefully destroy channels at the right time.

When a channel is destroyed, it may attempt scheduler operations which could
move a task off of it's I/O scheduler. This is obviously a bad interaction, and
some finesse is required to make it work (making destructors run at the right
time).

Closes #10375

10 years agoAnother round of test fixes from previous commits
Alex Crichton [Fri, 8 Nov 2013 04:13:25 +0000 (20:13 -0800)]
Another round of test fixes from previous commits

10 years agoUpdate to the latest libuv
Alex Crichton [Thu, 7 Nov 2013 23:26:47 +0000 (15:26 -0800)]
Update to the latest libuv

At this time, also point the libuv submodule to the official repo instead of my
own off to the side.

cc #10246
Closes #10329

10 years agoFall back from uv tty instances more aggressively
Alex Crichton [Thu, 7 Nov 2013 23:24:30 +0000 (15:24 -0800)]
Fall back from uv tty instances more aggressively

It appears that uv's support for interacting with a stdio stream as a tty when
it's actually a pipe is pretty problematic. To get around this, promote a check
to see if the stream is a tty to the top of the tty constructor, and bail out
quickly if it's not identified as a tty.

Closes #10237

10 years agoMake the uv bindings resilient to linked failure
Alex Crichton [Thu, 7 Nov 2013 23:13:06 +0000 (15:13 -0800)]
Make the uv bindings resilient to linked failure

In the ideal world, uv I/O could be canceled safely at any time. In reality,
however, we are unable to do this. Right now linked failure is fairly flaky as
implemented in the runtime, making it very difficult to test whether the linked
failure mechanisms inside of the uv bindings are ready for this kind of
interaction.

Right now, all constructors will execute in a task::unkillable block, and all
homing I/O operations will prevent linked failure in the duration of the homing
operation. What this means is that tasks which perform I/O are still susceptible
to linked failure, but the I/O operations themselves will never get interrupted.
Instead, the linked failure will be received at the edge of the I/O operation.

10 years agoAssorted test fixes and merge conflicts
Alex Crichton [Wed, 6 Nov 2013 07:29:11 +0000 (23:29 -0800)]
Assorted test fixes and merge conflicts

10 years agoRework the idle callback to have a safer interface
Alex Crichton [Wed, 6 Nov 2013 19:38:53 +0000 (11:38 -0800)]
Rework the idle callback to have a safer interface

It turns out that the uv implementation would cause use-after-free if the idle
callback was used after the call to `close`, and additionally nothing would ever
really work that well if `start()` were called twice. To change this, the
`start` and `close` methods were removed in favor of specifying the callback at
creation, and allowing destruction to take care of closing the watcher.

10 years agoUpdate all uv tests to pass again
Alex Crichton [Wed, 6 Nov 2013 19:03:11 +0000 (11:03 -0800)]
Update all uv tests to pass again

10 years agoWork around bugs in 32-bit enum FFI
Alex Crichton [Wed, 6 Nov 2013 03:16:48 +0000 (19:16 -0800)]
Work around bugs in 32-bit enum FFI

cc #10308

10 years agoMigrate all streams to synchronous closing
Alex Crichton [Wed, 6 Nov 2013 03:14:17 +0000 (19:14 -0800)]
Migrate all streams to synchronous closing

10 years agoAdd bindings to uv's utime function
Alex Crichton [Tue, 5 Nov 2013 23:48:27 +0000 (15:48 -0800)]
Add bindings to uv's utime function

This exposes the ability to change the modification and access times on a file.

Closes #10266

10 years agoDon't overflow in a converting stat times to u64
Alex Crichton [Tue, 5 Nov 2013 23:30:42 +0000 (15:30 -0800)]
Don't overflow in a converting stat times to u64

Closes #10297

10 years agoClean up the remaining chunks of uv
Alex Crichton [Tue, 5 Nov 2013 19:29:45 +0000 (11:29 -0800)]
Clean up the remaining chunks of uv

10 years agoMigrate uv net bindings away from ~fn()
Alex Crichton [Tue, 5 Nov 2013 08:27:41 +0000 (00:27 -0800)]
Migrate uv net bindings away from ~fn()

10 years agoMigrate uv getaddrinfo away from ~fn()
Alex Crichton [Tue, 5 Nov 2013 06:52:33 +0000 (22:52 -0800)]
Migrate uv getaddrinfo away from ~fn()

10 years agoMigrate uv file bindings away from ~fn()
Alex Crichton [Tue, 5 Nov 2013 05:08:25 +0000 (21:08 -0800)]
Migrate uv file bindings away from ~fn()

10 years agoStart migrating stream I/O away from ~fn()
Alex Crichton [Tue, 5 Nov 2013 00:42:05 +0000 (16:42 -0800)]
Start migrating stream I/O away from ~fn()

10 years agoFixing rebase conflicts and such
Alex Crichton [Mon, 4 Nov 2013 22:03:32 +0000 (14:03 -0800)]
Fixing rebase conflicts and such

This cleans up the merging of removing ~fn() and removing C++ wrappers to a
compile-able and progress-ready state

10 years agoRemove usage of ~fn from the scheduler
Alex Crichton [Mon, 4 Nov 2013 20:45:39 +0000 (12:45 -0800)]
Remove usage of ~fn from the scheduler

10 years agoRemove usage of ~fn() from uv async/idle
Alex Crichton [Mon, 4 Nov 2013 20:45:05 +0000 (12:45 -0800)]
Remove usage of ~fn() from uv async/idle

10 years agoMigrate uv signal handling away from ~fn()
Alex Crichton [Fri, 1 Nov 2013 18:13:22 +0000 (11:13 -0700)]
Migrate uv signal handling away from ~fn()

10 years agoMigrate uv process bindings away from ~fn()
Alex Crichton [Fri, 1 Nov 2013 17:26:43 +0000 (10:26 -0700)]
Migrate uv process bindings away from ~fn()

10 years agoMigrate uv timer bindings away from ~fn()
Alex Crichton [Fri, 1 Nov 2013 16:36:21 +0000 (09:36 -0700)]
Migrate uv timer bindings away from ~fn()

10 years agouv: Remove closure-based home_for_io for raii
Alex Crichton [Sun, 3 Nov 2013 19:26:08 +0000 (11:26 -0800)]
uv: Remove closure-based home_for_io for raii

Using an raii wrapper means that there's no need for a '_self' variant and we
can greatly reduce the amount of 'self_'-named variables.

10 years agouv: Provide a helper fn to Result<(), IoError>
Alex Crichton [Sun, 3 Nov 2013 19:02:19 +0000 (11:02 -0800)]
uv: Provide a helper fn to Result<(), IoError>

10 years agouv: Remove lots of uv/C++ wrappers
Alex Crichton [Sun, 3 Nov 2013 18:39:39 +0000 (10:39 -0800)]
uv: Remove lots of uv/C++ wrappers

10 years agoauto merge of #10367 : alexcrichton/rust/system-abi, r=nikomatsakis
bors [Sat, 9 Nov 2013 20:26:12 +0000 (12:26 -0800)]
auto merge of #10367 : alexcrichton/rust/system-abi, r=nikomatsakis

This adds an other ABI option which allows a custom selection over the target
architecture and OS. The only current candidate for this change is that kernel32
on win32 uses stdcall, but on win64 it uses the cdecl calling convention.
Otherwise everywhere else this is defined as using the Cdecl calling convention.

cc #10049
Closes #8774

10 years agoAdd a "system" ABI
Alex Crichton [Fri, 8 Nov 2013 19:06:57 +0000 (11:06 -0800)]
Add a "system" ABI

This adds an other ABI option which allows a custom selection over the target
architecture and OS. The only current candidate for this change is that kernel32
on win32 uses stdcall, but on win64 it uses the cdecl calling convention.
Otherwise everywhere else this is defined as using the Cdecl calling convention.

cc #10049
Closes #8774

10 years agoauto merge of #10153 : nikomatsakis/rust/issue-4846-multiple-lifetime-parameters...
bors [Sat, 9 Nov 2013 16:36:09 +0000 (08:36 -0800)]
auto merge of #10153 : nikomatsakis/rust/issue-4846-multiple-lifetime-parameters-7, r=pnkfelix

Fully support multiple lifetime parameters on types and elsewhere, removing special treatment for `'self`. I am submitting this a touch early in that I plan to push a new commit with more tests specifically targeting types with multiple lifetime parameters -- but the current code bootstraps and passes `make check`.

Fixes #4846

10 years agoMake irrelevant changes to regions-mock-tcx to convince the pretty printer to emit...
Niko Matsakis [Sat, 9 Nov 2013 16:34:05 +0000 (11:34 -0500)]
Make irrelevant changes to regions-mock-tcx to convince the pretty printer to emit the same thing twice in a row

10 years agoMake main public to accommodate check-fast
Niko Matsakis [Sat, 9 Nov 2013 14:13:58 +0000 (09:13 -0500)]
Make main public to accommodate check-fast

10 years agoauto merge of #10372 : catamorphism/rust/rustpkg-issue-9311, r=catamorphism
bors [Sat, 9 Nov 2013 14:11:10 +0000 (06:11 -0800)]
auto merge of #10372 : catamorphism/rust/rustpkg-issue-9311, r=catamorphism

and use opt_str instead of opt_val in rustpkg

Closes #9311

10 years agoxfail-fast regions-mock-tcx.rs
Niko Matsakis [Sat, 9 Nov 2013 12:57:10 +0000 (07:57 -0500)]
xfail-fast regions-mock-tcx.rs

10 years agoauto merge of #10371 : catamorphism/rust/rustpkg-issue-9482, r=catamorphism
bors [Sat, 9 Nov 2013 12:11:13 +0000 (04:11 -0800)]
auto merge of #10371 : catamorphism/rust/rustpkg-issue-9482, r=catamorphism

This bug was already fixed, just needed a test.

Closes #9569

10 years agoauto merge of #10368 : tautologico/rust/default_pkgid, r=catamorphism
bors [Sat, 9 Nov 2013 10:11:15 +0000 (02:11 -0800)]
auto merge of #10368 : tautologico/rust/default_pkgid, r=catamorphism

This Fixes #10265 and paves the way for fixing #9543. It works by adding a 'package_id' attribute by default for library crates that don't specify it. This is necessary to use the 'extern mod foo = "bar"' form instead of 'extern mod foo(name="bar") (as per #9543), because the former adds a required package_id when trying to link with the bar crate. I added a simple test to ensure that the default package_id value is being generated, and also added an explicit package_id in the link attribute in all rust libs to avoid getting warnings about default package_id values when building rust.

10 years agoauto merge of #10359 : tomassedovic/rust/typeid-iterbytes, r=pcwalton
bors [Sat, 9 Nov 2013 08:51:16 +0000 (00:51 -0800)]
auto merge of #10359 : tomassedovic/rust/typeid-iterbytes, r=pcwalton

I'm not sure this is something you're interested in, but I was playing around the Any trait a bit and I wanted to try it as a key in a HashMap. To do that, TypeId needs to implement IterBytes.

10 years agoauto merge of #10358 : cmr/rust/mmap++, r=alexcrichton
bors [Sat, 9 Nov 2013 07:41:18 +0000 (23:41 -0800)]
auto merge of #10358 : cmr/rust/mmap++, r=alexcrichton

10 years agostd: os: Document MemoryMap
Corey Richardson [Fri, 8 Nov 2013 14:18:52 +0000 (09:18 -0500)]
std: os: Document MemoryMap

10 years agoauto merge of #10376 : thestinger/rust/clone, r=huonw
bors [Sat, 9 Nov 2013 04:31:23 +0000 (20:31 -0800)]
auto merge of #10376 : thestinger/rust/clone, r=huonw

Closes #10240

10 years agomake DeepClone inherit from Clone
Daniel Micay [Sat, 9 Nov 2013 04:26:53 +0000 (23:26 -0500)]
make DeepClone inherit from Clone

10 years agoadd `clone_from` and `deep_clone_from`
Daniel Micay [Sat, 9 Nov 2013 04:10:09 +0000 (23:10 -0500)]
add `clone_from` and `deep_clone_from`

Closes #10240

10 years agoFix pretty printer when there are multiple lifetime parameters
Niko Matsakis [Sat, 9 Nov 2013 03:25:22 +0000 (22:25 -0500)]
Fix pretty printer when there are multiple lifetime parameters

10 years agoMerge failures
Niko Matsakis [Sat, 9 Nov 2013 01:59:43 +0000 (20:59 -0500)]
Merge failures

10 years agoAdd new tests showing multiple lifetime parameters in use
Niko Matsakis [Fri, 8 Nov 2013 21:19:28 +0000 (16:19 -0500)]
Add new tests showing multiple lifetime parameters in use

10 years agoAddress comments from @pnkfelix (thanks for the detailed review)
Niko Matsakis [Fri, 8 Nov 2013 20:52:36 +0000 (15:52 -0500)]
Address comments from @pnkfelix (thanks for the detailed review)

10 years agoUpdate FIXMEs with issue numbers
Niko Matsakis [Tue, 29 Oct 2013 15:19:28 +0000 (11:19 -0400)]
Update FIXMEs with issue numbers

10 years agoRename and modernize region enum names
Niko Matsakis [Tue, 29 Oct 2013 14:34:11 +0000 (10:34 -0400)]
Rename and modernize region enum names

10 years agoUpdate various tests and libraries that were incorrectly
Niko Matsakis [Tue, 29 Oct 2013 10:14:59 +0000 (06:14 -0400)]
Update various tests and libraries that were incorrectly
annotated.

10 years agoImprove error reporting for region inference failures to make use of
Niko Matsakis [Tue, 29 Oct 2013 10:12:50 +0000 (06:12 -0400)]
Improve error reporting for region inference failures to make use of
all the information it has at its disposal. Unfortunately this also
reveals that we need to improve the reporting heuristics further,
as sometimes the errors it chooses to emit seem somewhat mystifying
and are not related to the actual problem.

10 years agoMove comparing of impl methods against trait from collect to check.
Niko Matsakis [Tue, 29 Oct 2013 10:11:22 +0000 (06:11 -0400)]
Move comparing of impl methods against trait from collect to check.
This code fits better in check because it is checking that the impl
matches the interface. This allows us to avoid the awkward constructions
that lazilly collect traits and so forth. It also permits us to make
use of the results of variance inference.

10 years agoIntroduce new variance inference pass that replaces (and generalizes) old
Niko Matsakis [Tue, 29 Oct 2013 10:08:34 +0000 (06:08 -0400)]
Introduce new variance inference pass that replaces (and generalizes) old
region-parameterization/variance inference. We now compute variance for
type parameters but do not make use of it (most of the way towards #3598).

10 years agoGeneralize AST and ty::Generics to accept multiple lifetimes.
Niko Matsakis [Tue, 29 Oct 2013 10:03:32 +0000 (06:03 -0400)]
Generalize AST and ty::Generics to accept multiple lifetimes.

10 years agoextra / rustpkg: Make getopts::opt_val private again
Tim Chevalier [Sat, 9 Nov 2013 00:20:25 +0000 (16:20 -0800)]
extra / rustpkg: Make getopts::opt_val private again

and use opt_str instead of opt_val in rustpkg

Closes #9311

10 years agorustpkg: Add a test that `rustpkg install` doesn't copy a non-changed file
Tim Chevalier [Fri, 8 Nov 2013 23:42:20 +0000 (15:42 -0800)]
rustpkg: Add a test that `rustpkg install` doesn't copy a non-changed file

This bug was already fixed, just needed a test.

Closes #9569

10 years agoImplement IterBytes for TypeId
Tomas Sedovic [Fri, 8 Nov 2013 16:30:35 +0000 (17:30 +0100)]
Implement IterBytes for TypeId

10 years agoIntroduce ty_fold mechanism and port our various folders to use
Niko Matsakis [Tue, 29 Oct 2013 09:25:18 +0000 (05:25 -0400)]
Introduce ty_fold mechanism and port our various folders to use
it. This should eventually be merged with the Subst trait.

10 years agoAdded a FIXME with issue number to change package_id
Andrei Formiga [Fri, 8 Nov 2013 21:59:44 +0000 (18:59 -0300)]
Added a FIXME with issue number to change package_id

10 years agoCreate a new pass to resolve named lifetimes; rscope is not only
Niko Matsakis [Mon, 28 Oct 2013 21:37:10 +0000 (17:37 -0400)]
Create a new pass to resolve named lifetimes; rscope is not only
used to indicate when anonymous regions (i.e., &T) are permitted

10 years agoSpecify package_id for rust libs, to avoid spurious warnings
Andrei Formiga [Fri, 8 Nov 2013 20:42:29 +0000 (17:42 -0300)]
Specify package_id for rust libs, to avoid spurious warnings

10 years agoAdded tests for default generation of package_id meta attribute
Andrei Formiga [Fri, 8 Nov 2013 19:53:04 +0000 (16:53 -0300)]
Added tests for default generation of package_id meta attribute

10 years agoAdded default value for package_id attribute when encoding metadata for lib crates
Andrei Formiga [Fri, 8 Nov 2013 19:51:40 +0000 (16:51 -0300)]
Added default value for package_id attribute when encoding metadata for lib crates

10 years agoAdd default 'package_id' link meta for crates (with value equal to 'name')
Andrei Formiga [Fri, 8 Nov 2013 02:00:23 +0000 (23:00 -0300)]
Add default 'package_id' link meta for crates (with value equal to 'name')

10 years agoauto merge of #10355 : huonw/rust/deriving-field-spans, r=alexcrichton
bors [Fri, 8 Nov 2013 16:36:08 +0000 (08:36 -0800)]
auto merge of #10355 : huonw/rust/deriving-field-spans, r=alexcrichton

This rearranges the deriving code so that #[deriving] a trait on a field
that doesn't implement that trait will point to the field in question,
e.g.

    struct NotEq; // doesn't implement Eq

    #[deriving(Eq)]
    struct Foo {
        ok: int,
        also_ok: ~str,
        bad: NotEq // error points here.
    }

Unfortunately, this means the error is disconnected from the `deriving`
itself but there's no current way to pass that information through to
rustc except via the spans, at the moment.

Fixes #7724.

10 years agoauto merge of #10354 : thestinger/rust/vector, r=huonw
bors [Fri, 8 Nov 2013 13:46:04 +0000 (05:46 -0800)]
auto merge of #10354 : thestinger/rust/vector, r=huonw

This section desperately needs to be expanded, but removing the
misleading/incorrect information is a priority.

Managed vectors/strings are not covered, as they are feature-gated and
are only a micro-optimization to avoid double-indirection.

Closes #6882

10 years agoauto merge of #10348 : ksh8281/rust/fix_android_timegm, r=yichoi
bors [Fri, 8 Nov 2013 12:36:12 +0000 (04:36 -0800)]
auto merge of #10348 : ksh8281/rust/fix_android_timegm, r=yichoi

fix timegm bug on android
TZ restore not correctly before.
and it cause test case fail (time::tests::run_tests::test_convertions)

with @sammykim

10 years agoauto merge of #10346 : klutzy/rust/mingw64-32-fix, r=alexcrichton
bors [Fri, 8 Nov 2013 11:26:07 +0000 (03:26 -0800)]
auto merge of #10346 : klutzy/rust/mingw64-32-fix, r=alexcrichton

Fixes #10327

10 years agoauto merge of #10344 : brson/rust/atomiccopy, r=alexcrichton
bors [Fri, 8 Nov 2013 10:11:09 +0000 (02:11 -0800)]
auto merge of #10344 : brson/rust/atomiccopy, r=alexcrichton

I didn't try to add clone methods since it's not clear to me which types are appropriate to clone or what the memory ordering should be.

10 years agosyntax::ext: Make type errors in deriving point to the field itself.
Huon Wilson [Thu, 7 Nov 2013 07:49:01 +0000 (18:49 +1100)]
syntax::ext: Make type errors in deriving point to the field itself.

This rearranges the deriving code so that #[deriving] a trait on a field
that doesn't implement that trait will point to the field in question,
e.g.

    struct NotEq; // doesn't implement Eq

    #[deriving(Eq)]
    struct Foo {
        ok: int,
        also_ok: ~str,
        bad: NotEq // error points here.
    }

Unfortunately, this means the error is disconnected from the `deriving`
itself but there's no current way to pass that information through to
rustc except via the spans, at the moment.

Fixes #7724.

10 years agotutorial: rewrite the section vectors/strings
Daniel Micay [Fri, 8 Nov 2013 09:28:29 +0000 (04:28 -0500)]
tutorial: rewrite the section vectors/strings

This section desperately needs to be expanded, but removing the
misleading/incorrect information is a priority.

Managed vectors/strings are not covered, as they are feature-gated and
are only a micro-optimization to avoid double-indirection.

Closes #6882

10 years agoauto merge of #10303 : niftynif/rust/master, r=catamorphism
bors [Fri, 8 Nov 2013 08:11:07 +0000 (00:11 -0800)]
auto merge of #10303 : niftynif/rust/master, r=catamorphism

r? @catamorphism
First contribution; contains struct definitions and some useful traits.  Can possibly be improved by removing the dependence on the Clone trait.  This is intended to be the beginning of a long-term project.

10 years agoAdded proper //! documentation to the beginning of btree.rs.
niftynif [Fri, 8 Nov 2013 07:41:09 +0000 (02:41 -0500)]
Added proper //! documentation to the beginning of btree.rs.

10 years agoauto merge of #10341 : willingc/rust/minor-docfix, r=huonw
bors [Fri, 8 Nov 2013 07:01:11 +0000 (23:01 -0800)]
auto merge of #10341 : willingc/rust/minor-docfix, r=huonw

...rt::io

10 years agofix android timegm
sh8281.kim [Fri, 8 Nov 2013 06:44:36 +0000 (15:44 +0900)]
fix android timegm

10 years agoDefine WSAPOLLFD only on mingw 4.x
klutzy [Thu, 7 Nov 2013 04:37:20 +0000 (13:37 +0900)]
Define WSAPOLLFD only on mingw 4.x

Fixes #10327

10 years agoauto merge of #10281 : klutzy/rust/rt-timezone, r=alexcrichton
bors [Fri, 8 Nov 2013 05:41:07 +0000 (21:41 -0800)]
auto merge of #10281 : klutzy/rust/rt-timezone, r=alexcrichton

Previously #9418 fixed utf-8 assertion issue by wcsftime,
but the function didn't work as expected: it follows the locale
set by setlocale(), not the system code page.
This patch fixes it by manual multibyte-to-unicode conversion.

10 years agoextra::time: Fix test on Windows
klutzy [Wed, 6 Nov 2013 00:55:31 +0000 (09:55 +0900)]
extra::time: Fix test on Windows

Closes #10307

10 years agoauto merge of #10340 : eholk/rust/url, r=brson
bors [Fri, 8 Nov 2013 04:26:10 +0000 (20:26 -0800)]
auto merge of #10340 : eholk/rust/url, r=brson

According to http://tools.ietf.org/html/rfc3986, ~ is a legal character in URLs.

10 years agoauto merge of #10333 : huonw/rust/ascii, r=alexcrichton
bors [Fri, 8 Nov 2013 03:16:11 +0000 (19:16 -0800)]
auto merge of #10333 : huonw/rust/ascii, r=alexcrichton

This renames to_str_ascii to as_str_ascii and makes it non-copying,
which is possible now that strings no longer have a hidden extra
byte/null terminator.

Fixes #6120.

10 years agostd: Make atomic types non-copyable. Closes #8380
Brian Anderson [Fri, 8 Nov 2013 01:36:17 +0000 (17:36 -0800)]
std: Make atomic types non-copyable. Closes #8380

10 years agoauto merge of #10243 : mattcarberry/rust/master, r=brson
bors [Fri, 8 Nov 2013 01:26:12 +0000 (17:26 -0800)]
auto merge of #10243 : mattcarberry/rust/master, r=brson

Associated with Issue #6563.

Useful for Apollo Guidance Computer simulation, Unix file system permissions, and maybe one or two other things.

10 years agoauto merge of #10335 : ksh8281/rust/update_for_ndk_r9b, r=brson
bors [Thu, 7 Nov 2013 23:46:13 +0000 (15:46 -0800)]
auto merge of #10335 : ksh8281/rust/update_for_ndk_r9b, r=brson

update for ndk r9b (#10323)

````
Android NDK, Revision 9b (October 2013)
Important changes:
Updated include/android/*h and math.h for all Android API levels up to 18,
including the addition of levels 13, 15, 16 and 17.
For information on added APIs,
see commit messages for Changes 68012 and 68014. (Issues 47150, 58528, and 38423)
````

https://android-review.googlesource.com/#/c/68014/

10 years agoEdited comment for docs to show std::io is deleted and replaced by std::rt::io
Carol Willing [Thu, 7 Nov 2013 23:22:36 +0000 (15:22 -0800)]
Edited comment for docs to show std::io is deleted and replaced by std::rt::io

10 years agostd::ascii: Provide a copyless [Ascii] -> str method.
Huon Wilson [Thu, 7 Nov 2013 07:59:40 +0000 (18:59 +1100)]
std::ascii: Provide a copyless [Ascii] -> str method.

This renames to_str_ascii to as_str_ascii and makes it non-copying,
which is possible now that strings no longer have a hidden extra
byte/null terminator.

Fixes #6120.

10 years agoAdd ~ to the list of allowable URL characters.
Eric Holk [Thu, 7 Nov 2013 21:16:15 +0000 (16:16 -0500)]
Add ~ to the list of allowable URL characters.

10 years agoAddressed style concerns in btree.rs and added it to the lib file.
niftynif [Thu, 7 Nov 2013 18:34:20 +0000 (13:34 -0500)]
Addressed style concerns in btree.rs and added it to the lib file.
Added documentation within btree.rs.

10 years agoauto merge of #10336 : ksh8281/rust/fix_rpass_core-run-destroy_for_android, r=yichoi
bors [Thu, 7 Nov 2013 10:41:10 +0000 (02:41 -0800)]
auto merge of #10336 : ksh8281/rust/fix_rpass_core-run-destroy_for_android, r=yichoi

fix rpass core-run-destroy for android

10 years agofix rpass core-run-destroy for android
sh8281.kim [Thu, 7 Nov 2013 10:22:18 +0000 (19:22 +0900)]
fix rpass core-run-destroy for android

10 years agoupdate for ndk r9b
sh8281.kim [Thu, 7 Nov 2013 09:47:56 +0000 (18:47 +0900)]
update for ndk r9b

10 years agoauto merge of #10322 : thestinger/rust/no_freeze, r=alexcrichton
bors [Thu, 7 Nov 2013 07:36:06 +0000 (23:36 -0800)]
auto merge of #10322 : thestinger/rust/no_freeze, r=alexcrichton

10 years agoadd `from_send` to Rc, since #9509 is fixed
Daniel Micay [Wed, 6 Nov 2013 22:56:22 +0000 (17:56 -0500)]
add `from_send` to Rc, since #9509 is fixed

10 years agoauto merge of #10328 : alexcrichton/rust/snapshots, r=thestinger
bors [Thu, 7 Nov 2013 04:31:08 +0000 (20:31 -0800)]
auto merge of #10328 : alexcrichton/rust/snapshots, r=thestinger

10 years agoRegister new snapshots
Alex Crichton [Tue, 5 Nov 2013 16:25:32 +0000 (08:25 -0800)]
Register new snapshots

10 years agoAdded positive tests for integer literal base interpretation.
Matt Carberry [Wed, 6 Nov 2013 23:30:12 +0000 (15:30 -0800)]
Added positive tests for integer literal base interpretation.

10 years agoauto merge of #10227 : kud1ing/rust/ios, r=alexcrichton
bors [Wed, 6 Nov 2013 22:01:14 +0000 (14:01 -0800)]
auto merge of #10227 : kud1ing/rust/ios, r=alexcrichton

10 years agoFixes for compilation to iOS:
kud1ing [Fri, 1 Nov 2013 18:39:35 +0000 (19:39 +0100)]
Fixes for compilation to iOS:

- remove /usr/include from the include path since the iOS SDK provides the correct version
- `_NSGetEnviron()` is private and not available on iOS
- `.align` without an argument is not allowed with the Apple tools. 2^2 should be the default alignment
- ignore error messages for XCode < 5
- pass include path to libuv

10 years agoauto merge of #10289 : nikomatsakis/rust/issue-10157-TypeContents-refactor, r=pcwalton
bors [Wed, 6 Nov 2013 19:51:09 +0000 (11:51 -0800)]
auto merge of #10289 : nikomatsakis/rust/issue-10157-TypeContents-refactor, r=pcwalton

See #10157. This compiles now.

Fixes #10278.

10 years agoauto merge of #10314 : ksh8281/rust/fix_some_cfail_test_cases_for_arm, r=yichoi
bors [Wed, 6 Nov 2013 08:56:08 +0000 (00:56 -0800)]
auto merge of #10314 : ksh8281/rust/fix_some_cfail_test_cases_for_arm, r=yichoi

adding #[cfg(target_arch = "arm")] for asm of cfail test cases

10 years agoauto merge of #10306 : alexcrichton/rust/issue-9970-better, r=huonw
bors [Wed, 6 Nov 2013 07:36:06 +0000 (23:36 -0800)]
auto merge of #10306 : alexcrichton/rust/issue-9970-better, r=huonw

There were a few ambiguous error messages which look like they could have
cropped up from either the rust compiler for the format string parser. To
differentiate, the prefix 'invalid format string' is now added in front of all
format string errors.

cc #9970

10 years agofix some cfail test cases for arm
sh8281.kim [Wed, 6 Nov 2013 06:45:37 +0000 (15:45 +0900)]
fix some cfail test cases for arm