Alex Crichton [Tue, 6 Jan 2015 02:37:23 +0000 (18:37 -0800)]
rollup merge of #20511: csouth3/derive-lint
`#[deriving]` has been changed to `#[derive]`, so we should update this lint accordingly so that it remains consistent with the language.
Also register the rename with the LintStore.
I've changed the one reference to `raw_pointer_deriving` that occurs in the tests (as well as renamed the file appropriately), but the rest of the `raw_pointer_deriving`s in the Rust codebase will need to wait for a snapshot to be changed because stage0 doesn't know about the new lint name. I'll take care of the remaining renaming after the next snapshot.
The interface of the new `CString` is outlined in the linked RFC, the primary
changes being:
* The `ToCStr` trait is gone, meaning the `with_c_str` and `to_c_str` methods
are now gone. These two methods are replaced with a `CString::from_slice`
method.
* The `CString` type is now just a wrapper around `Vec<u8>` with a static
guarantee that there is a trailing nul byte with no internal nul bytes. This
means that `CString` now implements `Deref<Target = [c_char]>`, which is where
it gains most of its methods from. A few helper methods are added to acquire a
slice of `u8` instead of `c_char`, as well as including a slice with the
trailing nul byte if necessary.
* All usage of non-owned `CString` values is now done via two functions inside
of `std::ffi`, called `c_str_to_bytes` and `c_str_to_bytes_with_nul`. These
functions are now the one method used to convert a `*const c_char` to a Rust
slice of `u8`.
Many more details, including newly deprecated methods, can be found linked in
the RFC. This is a:
Alex Crichton [Tue, 6 Jan 2015 02:37:20 +0000 (18:37 -0800)]
rollup merge of #20488: ltratt/nondeterministic_tempdir
The previous scheme made it possible for another user/attacker to cause the
temporary directory creation scheme to panic. All you needed to know was the pid
of the process you wanted to target ('other_pid') and the suffix it was using
(let's pretend it's 'sfx') and then code such as this would, in essence, DOS it:
for i in range(0u, 1001) {
let tp = &Path::new(format!("/tmp/rs-{}-{}-sfx", other_pid, i));
match fs::mkdir(tp, io::USER_RWX) { _ => () }
}
Since the scheme only 1000 times to create a temporary directory before dying,
the next time the attacked process called TempDir::new("sfx") after that would
typically cause a panic. Of course, you don't necessarily need an attacker to
cause such a DOS: creating 1000 temporary directories without closing any of the
previous would be enough to DOS yourself.
This patch broadly follows the OpenBSD implementation of mkstemp. It uses the
operating system's random number generator to produce random directory names
that are impractical to guess (and, just in case someone manages to do that, it
retries creating the directory for a long time before giving up; OpenBSD
retries INT_MAX times, although 1<<31 seems enough to thwart even the most
patient attacker).
As a small additional change while the file name is changing, this patch also
makes the argument that TempDir::new takes a prefix rather than a suffix.
This is because 1) it more closely matches what mkstemp and friends do 2)
if you're going to have a deterministic part of a filename, you really want it at
the beginning so that shell completion is useful.
Alex Crichton [Tue, 6 Jan 2015 02:37:18 +0000 (18:37 -0800)]
rollup merge of #20483: nagisa/rng-copy
* Implement (derive) `Clone` for `ChaChaRng`, `Isaac*Rng`, `StdRng` and `ThreadRng`;
* Derive `XorShiftRng` `Clone` implementation instead of implementing it explicitly.
`OsRng` is the only Rng which does not implement `Clone` or `Copy` after this patch because of its dependence on `Reader`.
Alex Crichton [Tue, 6 Jan 2015 02:36:34 +0000 (18:36 -0800)]
rollup merge of #20465: nikomatsakis/assoc-types-regions-20303
Treat associated types the same as type parameters when it comes to region bounding. Fixes #20303.
Strictly speaking, this is a [breaking-change] (if you are using
associated types). You are no longer free to wantonly violate the type
system rules by closing associated types into objects without any form
of region bound. Instead you should add region bounds like `T::X :
'a`, just as you would with a normal type parameter.
Alex Crichton [Tue, 6 Jan 2015 02:36:30 +0000 (18:36 -0800)]
rollup merge of #20424: jroesch/tuple-struct-where-clause-fix
Fixes #17904. All the cases that I believe we should support are detailed in the test case, let me know if there is there is any more desired behavior. cc @japaric.
Alex Crichton [Tue, 6 Jan 2015 02:36:21 +0000 (18:36 -0800)]
rollup merge of #20092: barosl/rustdoc-line-number-clickable
While talking on IRC, someone wanted to post a link to the Rust source code, but while the lines of the rendered source code do have anchors (`<span id="[line number]">`), there is no convenient way to make links as they are not clickable. This PR makes them clickable.
bors [Mon, 5 Jan 2015 23:51:00 +0000 (23:51 +0000)]
auto merge of #20578 : japaric/rust/no-more-bc, r=nmatsakis
This PR removes boxed closures from the language, the closure type syntax (`let f: |int| -> bool = /* ... */`) has been obsoleted. Move all your uses of closures to the new unboxed closure system (i.e. `Fn*` traits).
[breaking-change] patterns
- `lef f = || {}`
This binding used to type check to a boxed closure. Now that boxed closures are gone, you need to annotate the "kind" of the unboxed closure, i.e. you need pick one of these: `|&:| {}`, `|&mut:| {}` or `|:| {}`.
In the (near) future we'll have closure "kind" inference, so the compiler will infer which `Fn*` trait to use based on how the closure is used. Once this inference machinery is in place, we'll be able to remove the kind annotation from most closures.
- `type Alias<'a> = |int|:'a -> bool`
Use a trait object: `type Alias<'a> = Box<FnMut(int) -> bool + 'a>`. Use the `Fn*` trait that makes sense for your use case.
- `fn foo(&self, f: |uint| -> bool)`
In this case you can use either a trait object or an unboxed closure:
``` rust
fn foo(&self, f: F) where F: FnMut(uint) -> bool;
// or
fn foo(&self, f: Box<FnMut(uint) -> bool>);
```
- `struct Struct<'a> { f: |uint|:'a -> bool }`
Again, you can use either a trait object or an unboxed closure:
``` rust
struct Struct<F> where F: FnMut(uint) -> bool { f: F }
// or
struct Struct<'a> { f: Box<FnMut(uint) -> bool + 'a> }
```
- Using `|x, y| f(x, y)` for closure "borrows"
This comes up in recursive functions, consider the following (contrived) example:
``` rust
fn foo(x: uint, f: |uint| -> bool) -> bool {
//foo(x / 2, f) && f(x) // can't use this because `f` gets moved away in the `foo` call
foo(x / 2, |x| f(x)) && f(x) // instead "borrow" `f` in the `foo` call
}
```
If you attempt to do the same with unboxed closures you'll hit ""error: reached the recursion limit during monomorphization" (see #19596):
``` rust
fn foo<F>(x: uint, mut f: F) -> bool where F: FnMut(uint) -> bool {
foo(x / 2, &mut f) && f(x)
//~^ error: the trait `FnMut` is not implemented for the type `&mut F`
}
```
But as you see above `&mut F` doesn't implement the `FnMut` trait. `&mut F` *should* implement the `FnMut` and the above code *should* work, but due to a bug (see #18835) it doesn't (for now).
You can work around the issue by rewriting the function to take `&mut F` instead of `F`:
There is more cleanup to do: like renaming functions/types from `unboxed_closure` to just `closure`, removing more dead code, simplify functions which now have unused arguments, update the documentation, etc. But that can be done in another PR.
r? @nikomatsakis @aturon (You probably want to focus on the deleted/modified tests.)
cc @eddyb
Aaron Turon [Mon, 5 Jan 2015 00:35:20 +0000 (16:35 -0800)]
Stabilize collection modules
The earlier collections stabilization did not cover the modules
themselves. This commit marks as stable those modules whose types have
been stabilized.
Aaron Turon [Sun, 4 Jan 2015 23:42:51 +0000 (15:42 -0800)]
Stabilize core::ops
This commit marks as stable those parts of `core::ops` that are in their
final planned form: `Drop`, all of the mathematical operators (`Add`,
`Sub`, etc), `Deref`/`DerefMut`. It leaves the `Index*`, `Slice*` and
`Fn*` traits unstable, as they are still undergoing active changes.
Aaron Turon [Sun, 4 Jan 2015 16:43:27 +0000 (08:43 -0800)]
Final alpha stabilization of core::iter
This commit wraps up the adjustments to the iterator for recent language
changes.
* Moves `rposition` from `ExactSizeIterator` to `IteratorExt` using a
`where` clause, thereby removing the `ExactSizeIterator:
DoubleEndedIterator` constraint.
* Merges `MutableDoubleEndedIterator` into `IteratorExt`, renaming
`reverse_` to `reverse_in_place`.
* Merges `IteratorOrdExt`, `IteratorCloneExt` and `CloneIteratorExt`
into `IteratorExt` using `where` clauses.
bors [Mon, 5 Jan 2015 20:02:14 +0000 (20:02 +0000)]
auto merge of #20572 : nikomatsakis/rust/assoc-supertrait-stuff, r=brson
The first few commits in the PR are just general refactoring. I was intending them for some other code I didn't get around to writing yet, but might as well land them now.
The interface of the new `CString` is outlined in the linked RFC, the primary
changes being:
* The `ToCStr` trait is gone, meaning the `with_c_str` and `to_c_str` methods
are now gone. These two methods are replaced with a `CString::from_slice`
method.
* The `CString` type is now just a wrapper around `Vec<u8>` with a static
guarantee that there is a trailing nul byte with no internal nul bytes. This
means that `CString` now implements `Deref<Target = [c_char]>`, which is where
it gains most of its methods from. A few helper methods are added to acquire a
slice of `u8` instead of `c_char`, as well as including a slice with the
trailing nul byte if necessary.
* All usage of non-owned `CString` values is now done via two functions inside
of `std::ffi`, called `c_str_to_bytes` and `c_str_to_bytes_with_nul`. These
functions are now the one method used to convert a `*const c_char` to a Rust
slice of `u8`.
Many more details, including newly deprecated methods, can be found linked in
the RFC. This is a:
Niko Matsakis [Sat, 3 Jan 2015 09:40:33 +0000 (04:40 -0500)]
Treat associated types the same as type parameters when it comes to region bounding. Fixes #20303.
Strictly speaking, this is a [breaking-change] (if you are using
associated types). You are no longer free to wantonly violate the type
system rules by closing associated types into objects without any form
of region bound. Instead you should add region bounds like `T::X :
'a`, just as you would with a normal type parameter.
bors [Mon, 5 Jan 2015 14:51:03 +0000 (14:51 +0000)]
auto merge of #20514 : alexcrichton/rust/serialize-associated-type, r=aturon
This commit moves the libserialize crate (and will force the hand of the
rustc-serialize crate) to not require the `old_orphan_check` feature gate as
well as using associated types wherever possible. Concretely, the following
changes were made:
* The error type of `Encoder` and `Decoder` is now an associated type, meaning
that these traits have no type parameters.
* The `Encoder` and `Decoder` type parameters on the `Encodable` and `Decodable`
traits have moved to the corresponding method of the trait. This movement
alleviates the dependency on `old_orphan_check` but implies that
implementations can no longer be specialized for the type of encoder/decoder
being implemented.
Laurence Tratt [Sat, 3 Jan 2015 21:49:01 +0000 (21:49 +0000)]
Make temporary directory names non-deterministic.
The previous scheme made it possible for another user/attacker to cause the
temporary directory creation scheme to panic. All you needed to know was the pid
of the process you wanted to target ('other_pid') and the suffix it was using
(let's pretend it's 'sfx') and then code such as this would, in essence, DOS it:
for i in range(0u, 1001) {
let tp = &Path::new(format!("/tmp/rs-{}-{}-sfx", other_pid, i));
match fs::mkdir(tp, io::USER_RWX) { _ => () }
}
Since the scheme retried only 1000 times to create a temporary directory before
dying, the next time the attacked process called TempDir::new("sfx") after that
would typically cause a panic. Of course, you don't necessarily need an attacker
to cause such a DOS: creating 1000 temporary directories without closing any of
the previous would be enough to DOS yourself.
This patch broadly follows the OpenBSD implementation of mkstemp. It uses the
operating system's random number generator to produce random directory names
that are impractical to guess (and, just in case someone manages to do that, it
retries creating the directory for a long time before giving up; OpenBSD
retries INT_MAX times, although 1<<31 seems enough to thwart even the most
patient attacker).
As a small additional change, this patch also makes the argument that
TempDir::new takes a prefix rather than a suffix. This is because 1) it more
closely matches what mkstemp and friends do 2) if you're going to have a
deterministic part of a filename, you really want it at the beginning so that
shell completion is useful.
Alex Crichton [Sun, 4 Jan 2015 06:24:50 +0000 (22:24 -0800)]
serialize: Use assoc types + less old_orphan_check
This commit moves the libserialize crate (and will force the hand of the
rustc-serialize crate) to not require the `old_orphan_check` feature gate as
well as using associated types wherever possible. Concretely, the following
changes were made:
* The error type of `Encoder` and `Decoder` is now an associated type, meaning
that these traits have no type parameters.
* The `Encoder` and `Decoder` type parameters on the `Encodable` and `Decodable`
traits have moved to the corresponding method of the trait. This movement
alleviates the dependency on `old_orphan_check` but implies that
implementations can no longer be specialized for the type of encoder/decoder
being implemented.
bors [Mon, 5 Jan 2015 06:45:39 +0000 (06:45 +0000)]
auto merge of #20395 : huonw/rust/char-stab-2, r=aturon
cc #19260
The casing transformations are left unstable (it is highly likely to be better to adopt the proper non-1-to-1 case mappings, per #20333) as are `is_xid_*`.
I've got a little todo list in the last commit of things I thought about/was told about that I haven't yet handled (I'd also like some feedback).
Huon Wilson [Fri, 5 Dec 2014 23:56:25 +0000 (15:56 -0800)]
Change `&` pat to only work with &T, and `&mut` with &mut T.
This implements RFC 179 by making the pattern `&<pat>` require matching
against a variable of type `&T`, and introducing the pattern `&mut
<pat>` which only works with variables of type `&mut T`.
The pattern `&mut x` currently parses as `&(mut x)` i.e. a pattern match
through a `&T` or a `&mut T` that binds the variable `x` to have type
`T` and to be mutable. This should be rewritten as follows, for example,