Copyedit FFI tutorial

doc/tutorial-ffi.md

@@ -2,17 +2,15 @@
 
 # Introduction
 
-One of Rust's aims, as a system programming language, is to
+Because Rust is a systems programming language, one of its goals is to
 interoperate well with C code.
 
-We'll start with an example. It's a bit bigger than usual, and
+We'll start with an example, which is a bit bigger than usual. We'll
-contains a number of new concepts. We'll go over it one piece at a
+go over it one piece at a time. This is a program that uses OpenSSL's
-time.
+`SHA1` function to compute the hash of its first command-line
-
+argument, which it then converts to a hexadecimal string and prints to
-This is a program that uses OpenSSL's `SHA1` function to compute the
+standard output. If you have the OpenSSL libraries installed, it
-hash of its first command-line argument, which it then converts to a
+should compile and run without any extra effort.
-hexadecimal string and prints to standard output. If you have the
-OpenSSL libraries installed, it should 'just work'.
 
 ~~~~ {.xfail-test}
 extern mod std;
@@ -32,7 +30,7 @@ fn sha1(data: ~str) -> ~str unsafe {
     let bytes = str::to_bytes(data);
     let hash = crypto::SHA1(vec::raw::to_ptr(bytes),
                             vec::len(bytes) as c_uint, ptr::null());
-    return as_hex(vec::raw::from_buf(hash, 20u));
+    return as_hex(vec::raw::from_buf(hash, 20));
 }
 
 fn main(args: ~[~str]) {
@@ -42,26 +40,27 @@ fn main(args: ~[~str]) {
 
 # Foreign modules
 
-Before we can call `SHA1`, we have to declare it. That is what this
+Before we can call the `SHA1` function defined in the OpenSSL library, we have
-part of the program is responsible for:
+to declare it. That is what this part of the program does:
 
 ~~~~ {.xfail-test}
 extern mod crypto {
-    fn SHA1(src: *u8, sz: uint, out: *u8) -> *u8;
+    fn SHA1(src: *u8, sz: uint, out: *u8) -> *u8; }
-}
 ~~~~
 
-An `extern` module declaration containing function signatures introduces
+An `extern` module declaration containing function signatures introduces the
-the functions listed as _foreign functions_, that are implemented in some
+functions listed as _foreign functions_. Foreign functions differ from regular
-other language (usually C) and accessed through Rust's foreign function
+Rust functions in that they are implemented in some other language (usually C)
-interface (FFI). An extern module like this is called a foreign module, and
+and called through Rust's foreign function interface (FFI). An extern module
-implicitly tells the compiler to link with a library with the same name as
+like this is called a foreign module, and implicitly tells the compiler to
-the module, and that it will find the foreign functions in that library.
+link with a library that contains the listed foreign functions, and has the
+same name as the module.
 
-In this case, it'll change the name `crypto` to a shared library name
+In this case, the Rust compiler changes the name `crypto` to a shared library
-in a platform-specific way (`libcrypto.so` on Linux, for example), and
+name in a platform-specific way (`libcrypto.so` on Linux, for example),
-link that in. If you want the module to have a different name from the
+searches for the shared library with that name, and links the library into the
-actual library, you can use the `"link_name"` attribute, like:
+program. If you want the module to have a different name from the actual
+library, you can use the `"link_name"` attribute, like:
 
 ~~~~ {.xfail-test}
 #[link_name = "crypto"]
@@ -72,11 +71,11 @@ extern mod something {
 
 # Foreign calling conventions
 
-Most foreign code will be C code, which usually uses the `cdecl` calling
+Most foreign code is C code, which usually uses the `cdecl` calling
 convention, so that is what Rust uses by default when calling foreign
 functions. Some foreign functions, most notably the Windows API, use other
-calling conventions, so Rust provides a way to hint to the compiler which
+calling conventions. Rust provides the `"abi"` attribute as a way to hint to
-is expected by using the `"abi"` attribute:
+the compiler which calling convention to use:
 
 ~~~~
 #[cfg(target_os = "win32")]
@@ -86,14 +85,14 @@ extern mod kernel32 {
 }
 ~~~~
 
-The `"abi"` attribute applies to a foreign module (it can not be applied
+The `"abi"` attribute applies to a foreign module (it cannot be applied
 to a single function within a module), and must be either `"cdecl"`
-or `"stdcall"`. Other conventions may be defined in the future.
+or `"stdcall"`. We may extend the compiler in the future to support other
+calling conventions.
 
 # Unsafe pointers
 
-The foreign `SHA1` function is declared to take three arguments, and
+The foreign `SHA1` function takes three arguments, and returns a pointer.
-return a pointer.
 
 ~~~~ {.xfail-test}
 # extern mod crypto {
@@ -104,21 +103,20 @@ fn SHA1(src: *u8, sz: libc::c_uint, out: *u8) -> *u8;
 When declaring the argument types to a foreign function, the Rust
 compiler has no way to check whether your declaration is correct, so
 you have to be careful. If you get the number or types of the
-arguments wrong, you're likely to get a segmentation fault. Or,
+arguments wrong, you're likely to cause a segmentation fault. Or,
 probably even worse, your code will work on one platform, but break on
 another.
 
-In this case, `SHA1` is defined as taking two `unsigned char*`
+In this case, we declare that `SHA1` takes two `unsigned char*`
-arguments and one `unsigned long`. The rust equivalents are `*u8`
+arguments and one `unsigned long`. The Rust equivalents are `*u8`
 unsafe pointers and an `uint` (which, like `unsigned long`, is a
 machine-word-sized type).
 
-Unsafe pointers can be created through various functions in the
+The standard library provides various functions to create unsafe pointers,
-standard lib, usually with `unsafe` somewhere in their name. You can
+such as those in `core::cast`. Most of these functions have `unsafe` in their
-dereference an unsafe pointer with `*` operator, but use
+name.  You can dereference an unsafe pointer with the `*` operator, but use
-caution—unlike Rust's other pointer types, unsafe pointers are
+caution: unlike Rust's other pointer types, unsafe pointers are completely
-completely unmanaged, so they might point at invalid memory, or be
+unmanaged, so they might point at invalid memory, or be null pointers.
-null pointers.
 
 # Unsafe blocks
 
@@ -134,12 +132,12 @@ fn sha1(data: ~str) -> ~str {
         let bytes = str::to_bytes(data);
         let hash = crypto::SHA1(vec::raw::to_ptr(bytes),
                                 vec::len(bytes), ptr::null());
-        return as_hex(vec::raw::from_buf(hash, 20u));
+        return as_hex(vec::raw::from_buf(hash, 20));
     }
 }
 ~~~~
 
-Firstly, what does the `unsafe` keyword at the top of the function
+First, what does the `unsafe` keyword at the top of the function
 mean? `unsafe` is a block modifier—it declares the block following it
 to be known to be unsafe.
 
@@ -158,8 +156,8 @@ advertise it to the world. An unsafe function is written like this:
 unsafe fn kaboom() { ~"I'm harmless!"; }
 ~~~~
 
-This function can only be called from an unsafe block or another
+This function can only be called from an `unsafe` block or another
-unsafe function.
+`unsafe` function.
 
 # Pointer fiddling
 
@@ -179,35 +177,36 @@ Let's look at our `sha1` function again.
 let bytes = str::to_bytes(data);
 let hash = crypto::SHA1(vec::raw::to_ptr(bytes),
                         vec::len(bytes), ptr::null());
-return as_hex(vec::raw::from_buf(hash, 20u));
+return as_hex(vec::raw::from_buf(hash, 20));
 # }
 # }
 ~~~~
 
-The `str::to_bytes` function is perfectly safe: it converts a string to
+The `str::to_bytes` function is perfectly safe: it converts a string to a
-a `[u8]`. This byte array is then fed to `vec::raw::to_ptr`, which
+`~[u8]`. The program then feeds this byte array to `vec::raw::to_ptr`, which
 returns an unsafe pointer to its contents.
 
-This pointer will become invalid as soon as the vector it points into
+This pointer will become invalid at the end of the scope in which the vector
-is cleaned up, so you should be very careful how you use it. In this
+it points to (`bytes`) is valid, so you should be very careful how you use
-case, the local variable `bytes` outlives the pointer, so we're good.
+it. In this case, the local variable `bytes` outlives the pointer, so we're
+good.
 
 Passing a null pointer as the third argument to `SHA1` makes it use a
 static buffer, and thus save us the effort of allocating memory
-ourselves. `ptr::null` is a generic function that will return an
+ourselves. `ptr::null` is a generic function that, in this case, returns an
-unsafe null pointer of the correct type (Rust generics are awesome
+unsafe null pointer of type `*u8`. (Rust generics are awesome
-like that—they can take the right form depending on the type that they
+like that: they can take the right form depending on the type that they
-are expected to return).
+are expected to return.)
 
-Finally, `vec::raw::from_buf` builds up a new `[u8]` from the
+Finally, `vec::raw::from_buf` builds up a new `~[u8]` from the
-unsafe pointer that was returned by `SHA1`. SHA1 digests are always
+unsafe pointer that `SHA1` returned. SHA1 digests are always
-twenty bytes long, so we can pass `20u` for the length of the new
+twenty bytes long, so we can pass `20` for the length of the new
 vector.
 
 # Passing structures
 
 C functions often take pointers to structs as arguments. Since Rust
-structs are binary-compatible with C structs, Rust programs can call
+`struct`s are binary-compatible with C structs, Rust programs can call
 such functions directly.
 
 This program uses the POSIX function `gettimeofday` to get a
@@ -241,12 +240,12 @@ fn unix_time_in_microseconds() -> u64 unsafe {
 The `#[nolink]` attribute indicates that there's no foreign library to
 link in. The standard C library is already linked with Rust programs.
 
-A `timeval`, in C, is a struct with two 32-bit integers. Thus, we
+In C, a `timeval` is a struct with two 32-bit integer fields. Thus, we
-define a struct type with the same contents, and declare
+define a `struct` type with the same contents, and declare
-`gettimeofday` to take a pointer to such a struct.
+`gettimeofday` to take a pointer to such a `struct`.
 
-The second argument to `gettimeofday` (the time zone) is not used by
+This program does not use the second argument to `gettimeofday` (the time
-this program, so it simply declares it to be a pointer to the nil
+ zone), so the `extern mod` declaration for it simply declares this argument
-type. Since all null pointers have the same representation regardless of
+ to be a pointer to the unit type (written `()`). Since all null pointers have
-their referent type, this is safe.
+ the same representation regardless of their referent type, this is safe.