(Mini) Rollup of 15 pull requests

src/doc/intro.md

@@ -446,16 +446,16 @@ It gives us this error:
 error: aborting due to previous error
 ```
 
-It mentions that "captured outer variable in an `FnMut` closure".
-Because we declared the closure as a moving closure, and it referred
-to `numbers`, the closure will try to take ownership of the
-vector. But the closure itself is created in a loop, and hence we will
-actually create three closures, one for every iteration of the
-loop. This means that all three of those closures would try to own
-`numbers`, which is impossible -- `numbers` must have just one
-owner. Rust detects this and gives us the error: we claim that
-`numbers` has ownership, but our code tries to make three owners. This
-may cause a safety problem, so Rust disallows it.
+This is a little confusing because there are two closures here: the one passed
+to `map`, and the one passed to `thread::scoped`. In this case, the closure for
+`thread::scoped` is attempting to reference `numbers`, a `Vec<i32>`. This
+closure is a `FnOnce` closure, as that’s what `thread::scoped` takes as an
+argument. `FnOnce` closures take ownership of their environment. That’s fine,
+but there’s one detail: because of `map`, we’re going to make three of these
+closures. And since all three try to take ownership of `numbers`, that would be
+a problem. That’s what it means by ‘cannot move out of captured outer
+variable’: our `thread::scoped` closure wants to take ownership, and it can’t,
+because the closure for `map` won’t let it.
 
 What to do here? Rust has two types that helps us: `Arc<T>` and `Mutex<T>`.
 *Arc* stands for "atomically reference counted". In other words, an Arc will

src/doc/trpl/compound-data-types.md

@@ -47,7 +47,7 @@ This pattern is very powerful, and we'll see it repeated more later.
 
 There are also a few things you can do with a tuple as a whole, without
 destructuring. You can assign one tuple into another, if they have the same
-contained types and arity. Tuples have the same arity when they have the same
+contained types and [arity]. Tuples have the same arity when they have the same
 length.
 
 ```rust
@@ -196,8 +196,9 @@ Now, we have actual names, rather than positions. Good names are important,
 and with a struct, we have actual names.
 
 There _is_ one case when a tuple struct is very useful, though, and that's a
-tuple struct with only one element. We call this a *newtype*, because it lets
-you create a new type that's similar to another one:
+tuple struct with only one element. We call this the *newtype* pattern, because
+it allows you to create a new type, distinct from that of its contained value
+and expressing its own semantic meaning:
 
 ```{rust}
 struct Inches(i32);
@@ -216,7 +217,7 @@ destructuring `let`, as we discussed previously in 'tuples.' In this case, the
 
 Finally, Rust has a "sum type", an *enum*. Enums are an incredibly useful
 feature of Rust, and are used throughout the standard library. An `enum` is
-a type which ties a set of alternates to a specific name. For example, below
+a type which relates a set of alternates to a specific name. For example, below
 we define `Character` to be either a `Digit` or something else. These
 can be used via their fully scoped names: `Character::Other` (more about `::`
 below).
@@ -228,8 +229,8 @@ enum Character {
 }
 ```
 
-An `enum` variant can be defined as most normal types. Below are some example
-types which also would be allowed in an `enum`.
+Most normal types are allowed as the variant components of an `enum`. Here are
+some examples:
 
 ```rust
 struct Empty;
@@ -239,15 +240,15 @@ struct Status { Health: i32, Mana: i32, Attack: i32, Defense: i32 }
 struct HeightDatabase(Vec<i32>);
 ```
 
-So you see that depending on the sub-datastructure, the `enum` variant, same as
-a struct, may or may not hold data. That is, in `Character`, `Digit` is a name
-tied to an `i32` where `Other` is just a name. However, the fact that they are
-distinct makes this very useful.
+You see that, depending on its type, an `enum` variant may or may not hold data.
+In `Character`, for instance, `Digit` gives a meaningful name for an `i32`
+value, where `Other` is only a name. However, the fact that they represent
+distinct categories of `Character` is a very useful property.
 
-As with structures, enums don't by default have access to operators such as
-compare ( `==` and `!=`), binary operations (`*` and `+`), and order
-(`<` and `>=`). As such, using the previous `Character` type, the
-following code is invalid:
+As with structures, the variants of an enum by default are not comparable with
+equality operators (`==`, `!=`), have no ordering (`<`, `>=`, etc.), and do not
+support other binary operations such as `*` and `+`. As such, the following code
+is invalid for the example `Character` type:
 
 ```{rust,ignore}
 // These assignments both succeed
@@ -265,9 +266,10 @@ let four_equals_ten = four == ten;
 ```
 
 This may seem rather limiting, but it's a limitation which we can overcome.
-There are two ways: by implementing equality ourselves, or by using the
-[`match`][match] keyword. We don't know enough about Rust to implement equality
-yet, but we can use the `Ordering` enum from the standard library, which does:
+There are two ways: by implementing equality ourselves, or by pattern matching
+variants with [`match`][match] expressions, which you'll learn in the next
+chapter. We don't know enough about Rust to implement equality yet, but we can
+use the `Ordering` enum from the standard library, which does:
 
 ```
 enum Ordering {
@@ -277,9 +279,8 @@ enum Ordering {
 }
 ```
 
-Because we did not define `Ordering`, we must import it (from the std
-library) with the `use` keyword. Here's an example of how `Ordering` is
-used:
+Because `Ordering` has already been defined for us, we will import it with the
+`use` keyword. Here's an example of how it is used:
 
 ```{rust}
 use std::cmp::Ordering;
@@ -313,17 +314,17 @@ the standard library if you need them.
 
 Okay, let's talk about the actual code in the example. `cmp` is a function that
 compares two things, and returns an `Ordering`. We return either
-`Ordering::Less`, `Ordering::Greater`, or `Ordering::Equal`, depending on if
-the two values are less, greater, or equal. Note that each variant of the
-`enum` is namespaced under the `enum` itself: it's `Ordering::Greater` not
-`Greater`.
+`Ordering::Less`, `Ordering::Greater`, or `Ordering::Equal`, depending on
+whether the first value is less than, greater than, or equal to the second. Note
+that each variant of the `enum` is namespaced under the `enum` itself: it's
+`Ordering::Greater`, not `Greater`.
 
 The `ordering` variable has the type `Ordering`, and so contains one of the
 three values. We then do a bunch of `if`/`else` comparisons to check which
 one it is.
 
-This `Ordering::Greater` notation is too long. Let's use `use` to import the
-`enum` variants instead. This will avoid full scoping:
+This `Ordering::Greater` notation is too long. Let's use another form of `use`
+to import the `enum` variants instead. This will avoid full scoping:
 
 ```{rust}
 use std::cmp::Ordering::{self, Equal, Less, Greater};
@@ -347,16 +348,18 @@ fn main() {
 ```
 
 Importing variants is convenient and compact, but can also cause name conflicts,
-so do this with caution. It's considered good style to rarely import variants
-for this reason.
+so do this with caution. For this reason, it's normally considered better style
+to `use` an enum rather than its variants directly.
 
-As you can see, `enum`s are quite a powerful tool for data representation, and are
-even more useful when they're [generic][generics] across types. Before we
-get to generics, though, let's talk about how to use them with pattern matching, a
-tool that will let us deconstruct this sum type (the type theory term for enums)
-in a very elegant way and avoid all these messy `if`/`else`s.
+As you can see, `enum`s are quite a powerful tool for data representation, and
+are even more useful when they're [generic][generics] across types. Before we
+get to generics, though, let's talk about how to use enums with pattern
+matching, a tool that will let us deconstruct sum types (the type theory term
+for enums) like `Ordering` in a very elegant way that avoids all these messy
+and brittle `if`/`else`s.
 
 
+[arity]: ./glossary.html#arity
 [match]: ./match.html
 [game]: ./guessing-game.html#comparing-guesses
 [generics]: ./generics.html

src/doc/trpl/concurrency.md

@@ -40,14 +40,14 @@ us enforce that it can't leave the current thread.
 
 ### `Sync`
 
-The second of these two trait is called [`Sync`](../std/marker/trait.Sync.html).
+The second of these traits is called [`Sync`](../std/marker/trait.Sync.html).
 When a type `T` implements `Sync`, it indicates to the compiler that something
 of this type has no possibility of introducing memory unsafety when used from
 multiple threads concurrently.
 
 For example, sharing immutable data with an atomic reference count is
 threadsafe. Rust provides a type like this, `Arc<T>`, and it implements `Sync`,
-so that it could be safely shared between threads.
+so it is safe to share between threads.
 
 These two traits allow you to use the type system to make strong guarantees
 about the properties of your code under concurrency. Before we demonstrate
@@ -69,7 +69,7 @@ fn main() {
 }
 ```
 
-The `Thread::scoped()` method accepts a closure, which is executed in a new
+The `thread::scoped()` method accepts a closure, which is executed in a new
 thread. It's called `scoped` because this thread returns a join guard:
 
 ```
@@ -208,10 +208,10 @@ Here's the error:
 
 ```text
 <anon>:11:9: 11:22 error: the trait `core::marker::Send` is not implemented for the type `std::sync::mutex::MutexGuard<'_, collections::vec::Vec<u32>>` [E0277]
-<anon>:11         Thread::spawn(move || {
+<anon>:11         thread::spawn(move || {
                   ^~~~~~~~~~~~~
 <anon>:11:9: 11:22 note: `std::sync::mutex::MutexGuard<'_, collections::vec::Vec<u32>>` cannot be sent between threads safely
-<anon>:11         Thread::spawn(move || {
+<anon>:11         thread::spawn(move || {
                   ^~~~~~~~~~~~~
 ```
 
@@ -322,7 +322,6 @@ While this channel is just sending a generic signal, we can send any data that
 is `Send` over the channel!
 
 ```
-use std::sync::{Arc, Mutex};
 use std::thread;
 use std::sync::mpsc;
 

src/doc/trpl/crates-and-modules.md

@@ -430,7 +430,7 @@ fn main() {
 }
 ```
 
-But it is not idiomatic. This is significantly more likely to introducing a
+But it is not idiomatic. This is significantly more likely to introduce a
 naming conflict. In our short program, it's not a big deal, but as it grows, it
 becomes a problem. If we have conflicting names, Rust will give a compilation
 error. For example, if we made the `japanese` functions public, and tried to do

src/doc/trpl/documentation.md

@@ -333,6 +333,41 @@ By repeating all parts of the example, you can ensure that your example still
 compiles, while only showing the parts that are relevant to that part of your
 explanation.
 
+### Documenting macros
+
+Here’s an example of documenting a macro:
+
+```
+/// Panic with a given message unless an expression evaluates to true.
+///
+/// # Examples
+///
+/// ```
+/// # #[macro_use] extern crate foo;
+/// # fn main() {
+/// panic_unless!(1 + 1 == 2, “Math is broken.”);
+/// # }
+/// ```
+///
+/// ```should_fail
+/// # #[macro_use] extern crate foo;
+/// # fn main() {
+/// panic_unless!(true == false, “I’m broken.”);
+/// # }
+/// ```
+#[macro_export]
+macro_rules! panic_unless {
+    ($condition:expr, $($rest:expr),+) => ({ if ! $condition { panic!($($rest),+); } });
+} 
+```
+
+You’ll note three things: we need to add our own `extern crate` line, so that
+we can add the `#[macro_use]` attribute. Second, we’ll need to add our own
+`main()` as well. Finally, a judicious use of `#` to comment out those two
+things, so they don’t show up in the output.
+
+### Running documentation tests
+
 To run the tests, either
 
 ```bash

src/doc/trpl/more-strings.md

@@ -12,7 +12,7 @@ Additionally, strings are not null-terminated and can contain null bytes.
 
 Rust has two main types of strings: `&str` and `String`.
 
-# &str
+# `&str`
 
 The first kind is a `&str`. This is pronounced a 'string slice'.
 String literals are of the type `&str`:
@@ -36,7 +36,36 @@ Like vector slices, string slices are simply a pointer plus a length. This
 means that they're a 'view' into an already-allocated string, such as a
 string literal or a `String`.
 
-# String
+## `str`
+
+You may occasionally see references to a `str` type, without the `&`. While
+this type does exist, it’s not something you want to use yourself. Sometimes,
+people confuse `str` for `String`, and write this:
+
+```rust
+struct S {
+    s: str,
+}
+```
+
+This leads to ugly errors:
+
+```text
+error: the trait `core::marker::Sized` is not implemented for the type `str` [E0277]
+note: `str` does not have a constant size known at compile-time
+```
+
+Instead, this `struct` should be
+
+```rust
+struct S {
+    s: String,
+}
+```
+
+So let’s talk about `String`s.
+
+# `String`
 
 A `String` is a heap-allocated string. This string is growable, and is
 also guaranteed to be UTF-8. `String`s are commonly created by

src/doc/trpl/pointers.md

@@ -498,13 +498,10 @@ they go out of scope:
 However, boxes do _not_ use reference counting or garbage collection. Boxes are
 what's called an *affine type*. This means that the Rust compiler, at compile
 time, determines when the box comes into and goes out of scope, and inserts the
-appropriate calls there. Furthermore, boxes are a specific kind of affine type,
-known as a *region*. You can read more about regions [in this paper on the
-Cyclone programming
-language](http://www.cs.umd.edu/projects/cyclone/papers/cyclone-regions.pdf).
+appropriate calls there.
 
-You don't need to fully grok the theory of affine types or regions to grok
-boxes, though. As a rough approximation, you can treat this Rust code:
+You don't need to fully grok the theory of affine types to grok boxes, though.
+As a rough approximation, you can treat this Rust code:
 
 ```{rust}
 {

src/doc/trpl/standard-input.md

@@ -115,8 +115,9 @@ doesn't work, so we're okay with that. In most cases, we would want to handle
 the error case explicitly. `expect()` allows us to give an error message if
 this crash happens.
 
-We will cover the exact details of how all of this works later in the Guide.
-For now, this gives you enough of a basic understanding to work with.
+We will cover the exact details of how all of this works later in the Guide in
+[Error Handling]. For now, this gives you enough of a basic understanding to
+work with.
 
 Back to the code we were working on! Here's a refresher:
 
@@ -157,3 +158,6 @@ here.
 
 That's all you need to get basic input from the standard input! It's not too
 complicated, but there are a number of small parts.
+
+
+[Error Handling]: ./error-handling.html

src/liballoc/heap.rs

@@ -26,6 +26,9 @@ pub unsafe fn allocate(size: usize, align: usize) -> *mut u8 {
 ///
 /// On failure, return a null pointer and leave the original allocation intact.
 ///
+/// If the allocation was relocated, the memory at the passed-in pointer is
+/// undefined after the call.
+///
 /// Behavior is undefined if the requested size is 0 or the alignment is not a
 /// power of 2. The alignment must be no larger than the largest supported page
 /// size on the platform.