1 // Copyright 2015 The Rust Project Developers. See the COPYRIGHT
2 // file at the top-level directory of this distribution and at
3 // http://rust-lang.org/COPYRIGHT.
5 // Licensed under the Apache License, Version 2.0 <LICENSE-APACHE or
6 // http://www.apache.org/licenses/LICENSE-2.0> or the MIT license
7 // <LICENSE-MIT or http://opensource.org/licenses/MIT>, at your
8 // option. This file may not be copied, modified, or distributed
9 // except according to those terms.
11 //! Windows-specific extensions to the primitives in the `std::ffi` module.
15 //! For historical reasons, the Windows API uses a form of potentially
16 //! ill-formed UTF-16 encoding for strings. Specifically, the 16-bit
17 //! code units in Windows strings may contain [isolated surrogate code
18 //! points which are not paired together][ill-formed-utf-16]. The
19 //! Unicode standard requires that surrogate code points (those in the
20 //! range U+D800 to U+DFFF) always be *paired*, because in the UTF-16
21 //! encoding a *surrogate code unit pair* is used to encode a single
22 //! character. For compatibility with code that does not enforce
23 //! these pairings, Windows does not enforce them, either.
25 //! While it is not always possible to convert such a string losslessly into
26 //! a valid UTF-16 string (or even UTF-8), it is often desirable to be
27 //! able to round-trip such a string from and to Windows APIs
28 //! losslessly. For example, some Rust code may be "bridging" some
29 //! Windows APIs together, just passing `WCHAR` strings among those
30 //! APIs without ever really looking into the strings.
32 //! If Rust code *does* need to look into those strings, it can
33 //! convert them to valid UTF-8, possibly lossily, by substituting
34 //! invalid sequences with [`U+FFFD REPLACEMENT CHARACTER`][U+FFFD], as is
35 //! conventionally done in other Rust APIs that deal with string
38 //! # `OsStringExt` and `OsStrExt`
40 //! [`OsString`] is the Rust wrapper for owned strings in the
41 //! preferred representation of the operating system. On Windows,
42 //! this struct gets augmented with an implementation of the
43 //! [`OsStringExt`] trait, which has a [`from_wide`] method. This
44 //! lets you create an [`OsString`] from a `&[u16]` slice; presumably
45 //! you get such a slice out of a `WCHAR` Windows API.
47 //! Similarly, [`OsStr`] is the Rust wrapper for borrowed strings from
48 //! preferred representation of the operating system. On Windows, the
49 //! [`OsStrExt`] trait provides the [`encode_wide`] method, which
50 //! outputs an [`EncodeWide`] iterator. You can [`collect`] this
51 //! iterator, for example, to obtain a `Vec<u16>`; you can later get a
52 //! pointer to this vector's contents and feed it to Windows APIs.
54 //! These traits, along with [`OsString`] and [`OsStr`], work in
55 //! conjunction so that it is possible to **round-trip** strings from
56 //! Windows and back, with no loss of data, even if the strings are
57 //! ill-formed UTF-16.
59 //! [ill-formed-utf-16]: https://simonsapin.github.io/wtf-8/#ill-formed-utf-16
60 //! [`OsString`]: ../../../ffi/struct.OsString.html
61 //! [`OsStr`]: ../../../ffi/struct.OsStr.html
62 //! [`OsStringExt`]: trait.OsStringExt.html
63 //! [`OsStrExt`]: trait.OsStrExt.html
64 //! [`EncodeWide`]: struct.EncodeWide.html
65 //! [`from_wide`]: trait.OsStringExt.html#tymethod.from_wide
66 //! [`encode_wide`]: trait.OsStrExt.html#tymethod.encode_wide
67 //! [`collect`]: ../../../iter/trait.Iterator.html#method.collect
68 //! [U+FFFD]: ../../../char/constant.REPLACEMENT_CHARACTER.html
70 #![stable(feature = "rust1", since = "1.0.0")]
72 use ffi::{OsString, OsStr};
74 use sys_common::wtf8::Wtf8Buf;
75 use sys_common::{FromInner, AsInner};
77 #[stable(feature = "rust1", since = "1.0.0")]
78 pub use sys_common::wtf8::EncodeWide;
80 /// Windows-specific extensions to [`OsString`].
82 /// [`OsString`]: ../../../../std/ffi/struct.OsString.html
83 #[stable(feature = "rust1", since = "1.0.0")]
84 pub trait OsStringExt {
85 /// Creates an `OsString` from a potentially ill-formed UTF-16 slice of
86 /// 16-bit code units.
88 /// This is lossless: calling [`encode_wide`] on the resulting string
89 /// will always return the original code units.
94 /// use std::ffi::OsString;
95 /// use std::os::windows::prelude::*;
97 /// // UTF-16 encoding for "Unicode".
98 /// let source = [0x0055, 0x006E, 0x0069, 0x0063, 0x006F, 0x0064, 0x0065];
100 /// let string = OsString::from_wide(&source[..]);
103 /// [`encode_wide`]: ./trait.OsStrExt.html#tymethod.encode_wide
104 #[stable(feature = "rust1", since = "1.0.0")]
105 fn from_wide(wide: &[u16]) -> Self;
108 #[stable(feature = "rust1", since = "1.0.0")]
109 impl OsStringExt for OsString {
110 fn from_wide(wide: &[u16]) -> OsString {
111 FromInner::from_inner(Buf { inner: Wtf8Buf::from_wide(wide) })
115 /// Windows-specific extensions to [`OsStr`].
117 /// [`OsStr`]: ../../../../std/ffi/struct.OsStr.html
118 #[stable(feature = "rust1", since = "1.0.0")]
120 /// Re-encodes an `OsStr` as a wide character sequence, i.e., potentially
121 /// ill-formed UTF-16.
123 /// This is lossless: calling [`OsString::from_wide`] and then
124 /// `encode_wide` on the result will yield the original code units.
125 /// Note that the encoding does not add a final null terminator.
130 /// use std::ffi::OsString;
131 /// use std::os::windows::prelude::*;
133 /// // UTF-16 encoding for "Unicode".
134 /// let source = [0x0055, 0x006E, 0x0069, 0x0063, 0x006F, 0x0064, 0x0065];
136 /// let string = OsString::from_wide(&source[..]);
138 /// let result: Vec<u16> = string.encode_wide().collect();
139 /// assert_eq!(&source[..], &result[..]);
142 /// [`OsString::from_wide`]: ./trait.OsStringExt.html#tymethod.from_wide
143 #[stable(feature = "rust1", since = "1.0.0")]
144 fn encode_wide(&self) -> EncodeWide;
147 #[stable(feature = "rust1", since = "1.0.0")]
148 impl OsStrExt for OsStr {
149 fn encode_wide(&self) -> EncodeWide {
150 self.as_inner().inner.encode_wide()