1//! Ways to create a `str` from bytes slice.
2
3use crate::{mem, ptr};
4
5use super::validations::run_utf8_validation;
6use super::Utf8Error;
7
8/// Converts a slice of bytes to a string slice.
9///
10/// A string slice ([`&str`]) is made of bytes ([`u8`]), and a byte slice
11/// ([`&[u8]`][byteslice]) is made of bytes, so this function converts between
12/// the two. Not all byte slices are valid string slices, however: [`&str`] requires
13/// that it is valid UTF-8. `from_utf8()` checks to ensure that the bytes are valid
14/// UTF-8, and then does the conversion.
15///
16/// [`&str`]: str
17/// [byteslice]: slice
18///
19/// If you are sure that the byte slice is valid UTF-8, and you don't want to
20/// incur the overhead of the validity check, there is an unsafe version of
21/// this function, [`from_utf8_unchecked`], which has the same
22/// behavior but skips the check.
23///
24/// If you need a `String` instead of a `&str`, consider
25/// [`String::from_utf8`][string].
26///
27/// [string]: ../../std/string/struct.String.html#method.from_utf8
28///
29/// Because you can stack-allocate a `[u8; N]`, and you can take a
30/// [`&[u8]`][byteslice] of it, this function is one way to have a
31/// stack-allocated string. There is an example of this in the
32/// examples section below.
33///
34/// [byteslice]: slice
35///
36/// # Errors
37///
38/// Returns `Err` if the slice is not UTF-8 with a description as to why the
39/// provided slice is not UTF-8.
40///
41/// # Examples
42///
43/// Basic usage:
44///
45/// ```
46/// use std::str;
47///
48/// // some bytes, in a vector
49/// let sparkle_heart = vec![240, 159, 146, 150];
50///
51/// // We know these bytes are valid, so just use `unwrap()`.
52/// let sparkle_heart = str::from_utf8(&sparkle_heart).unwrap();
53///
54/// assert_eq!("💖", sparkle_heart);
55/// ```
56///
57/// Incorrect bytes:
58///
59/// ```
60/// use std::str;
61///
62/// // some invalid bytes, in a vector
63/// let sparkle_heart = vec![0, 159, 146, 150];
64///
65/// assert!(str::from_utf8(&sparkle_heart).is_err());
66/// ```
67///
68/// See the docs for [`Utf8Error`] for more details on the kinds of
69/// errors that can be returned.
70///
71/// A "stack allocated string":
72///
73/// ```
74/// use std::str;
75///
76/// // some bytes, in a stack-allocated array
77/// let sparkle_heart = [240, 159, 146, 150];
78///
79/// // We know these bytes are valid, so just use `unwrap()`.
80/// let sparkle_heart: &str = str::from_utf8(&sparkle_heart).unwrap();
81///
82/// assert_eq!("💖", sparkle_heart);
83/// ```
84#[stable(feature = "rust1", since = "1.0.0")]
85#[rustc_const_stable(feature = "const_str_from_utf8_shared", since = "1.63.0")]
86#[rustc_allow_const_fn_unstable(str_internals)]
87#[rustc_diagnostic_item = "str_from_utf8"]
88pub const fn from_utf8(v: &[u8]) -> Result<&str, Utf8Error> {
89 // FIXME: This should use `?` again, once it's `const`
90 match run_utf8_validation(v) {
91 Ok(_) => {
92 // SAFETY: validation succeeded.
93 Ok(unsafe { from_utf8_unchecked(v) })
94 }
95 Err(err: Utf8Error) => Err(err),
96 }
97}
98
99/// Converts a mutable slice of bytes to a mutable string slice.
100///
101/// # Examples
102///
103/// Basic usage:
104///
105/// ```
106/// use std::str;
107///
108/// // "Hello, Rust!" as a mutable vector
109/// let mut hellorust = vec![72, 101, 108, 108, 111, 44, 32, 82, 117, 115, 116, 33];
110///
111/// // As we know these bytes are valid, we can use `unwrap()`
112/// let outstr = str::from_utf8_mut(&mut hellorust).unwrap();
113///
114/// assert_eq!("Hello, Rust!", outstr);
115/// ```
116///
117/// Incorrect bytes:
118///
119/// ```
120/// use std::str;
121///
122/// // Some invalid bytes in a mutable vector
123/// let mut invalid = vec![128, 223];
124///
125/// assert!(str::from_utf8_mut(&mut invalid).is_err());
126/// ```
127/// See the docs for [`Utf8Error`] for more details on the kinds of
128/// errors that can be returned.
129#[stable(feature = "str_mut_extras", since = "1.20.0")]
130#[rustc_const_unstable(feature = "const_str_from_utf8", issue = "91006")]
131#[rustc_diagnostic_item = "str_from_utf8_mut"]
132pub const fn from_utf8_mut(v: &mut [u8]) -> Result<&mut str, Utf8Error> {
133 // This should use `?` again, once it's `const`
134 match run_utf8_validation(v) {
135 Ok(_) => {
136 // SAFETY: validation succeeded.
137 Ok(unsafe { from_utf8_unchecked_mut(v) })
138 }
139 Err(err: Utf8Error) => Err(err),
140 }
141}
142
143/// Converts a slice of bytes to a string slice without checking
144/// that the string contains valid UTF-8.
145///
146/// See the safe version, [`from_utf8`], for more information.
147///
148/// # Safety
149///
150/// The bytes passed in must be valid UTF-8.
151///
152/// # Examples
153///
154/// Basic usage:
155///
156/// ```
157/// use std::str;
158///
159/// // some bytes, in a vector
160/// let sparkle_heart = vec![240, 159, 146, 150];
161///
162/// let sparkle_heart = unsafe {
163/// str::from_utf8_unchecked(&sparkle_heart)
164/// };
165///
166/// assert_eq!("💖", sparkle_heart);
167/// ```
168#[inline]
169#[must_use]
170#[stable(feature = "rust1", since = "1.0.0")]
171#[rustc_const_stable(feature = "const_str_from_utf8_unchecked", since = "1.55.0")]
172#[rustc_diagnostic_item = "str_from_utf8_unchecked"]
173pub const unsafe fn from_utf8_unchecked(v: &[u8]) -> &str {
174 // SAFETY: the caller must guarantee that the bytes `v` are valid UTF-8.
175 // Also relies on `&str` and `&[u8]` having the same layout.
176 unsafe { mem::transmute(src:v) }
177}
178
179/// Converts a slice of bytes to a string slice without checking
180/// that the string contains valid UTF-8; mutable version.
181///
182/// See the immutable version, [`from_utf8_unchecked()`] for more information.
183///
184/// # Examples
185///
186/// Basic usage:
187///
188/// ```
189/// use std::str;
190///
191/// let mut heart = vec![240, 159, 146, 150];
192/// let heart = unsafe { str::from_utf8_unchecked_mut(&mut heart) };
193///
194/// assert_eq!("💖", heart);
195/// ```
196#[inline]
197#[must_use]
198#[stable(feature = "str_mut_extras", since = "1.20.0")]
199#[rustc_const_unstable(feature = "const_str_from_utf8_unchecked_mut", issue = "91005")]
200#[rustc_diagnostic_item = "str_from_utf8_unchecked_mut"]
201pub const unsafe fn from_utf8_unchecked_mut(v: &mut [u8]) -> &mut str {
202 // SAFETY: the caller must guarantee that the bytes `v`
203 // are valid UTF-8, thus the cast to `*mut str` is safe.
204 // Also, the pointer dereference is safe because that pointer
205 // comes from a reference which is guaranteed to be valid for writes.
206 unsafe { &mut *(v as *mut [u8] as *mut str) }
207}
208
209/// Creates an `&str` from a pointer and a length.
210///
211/// The pointed-to bytes must be valid UTF-8.
212/// If this might not be the case, use `str::from_utf8(slice::from_raw_parts(ptr, len))`,
213/// which will return an `Err` if the data isn't valid UTF-8.
214///
215/// This function is the `str` equivalent of [`slice::from_raw_parts`](crate::slice::from_raw_parts).
216/// See that function's documentation for safety concerns and examples.
217///
218/// The mutable version of this function is [`from_raw_parts_mut`].
219#[inline]
220#[must_use]
221#[unstable(feature = "str_from_raw_parts", issue = "119206")]
222#[rustc_const_unstable(feature = "str_from_raw_parts", issue = "119206")]
223pub const unsafe fn from_raw_parts<'a>(ptr: *const u8, len: usize) -> &'a str {
224 // SAFETY: the caller must uphold the safety contract for `from_raw_parts`.
225 unsafe { &*ptr::from_raw_parts(data_pointer:ptr.cast(), metadata:len) }
226}
227
228/// Creates an `&mut str` from a pointer and a length.
229///
230/// The pointed-to bytes must be valid UTF-8.
231/// If this might not be the case, use `str::from_utf8_mut(slice::from_raw_parts_mut(ptr, len))`,
232/// which will return an `Err` if the data isn't valid UTF-8.
233///
234/// This function is the `str` equivalent of [`slice::from_raw_parts_mut`](crate::slice::from_raw_parts_mut).
235/// See that function's documentation for safety concerns and examples.
236///
237/// The immutable version of this function is [`from_raw_parts`].
238#[inline]
239#[must_use]
240#[unstable(feature = "str_from_raw_parts", issue = "119206")]
241#[rustc_const_unstable(feature = "const_str_from_raw_parts_mut", issue = "119206")]
242pub const unsafe fn from_raw_parts_mut<'a>(ptr: *mut u8, len: usize) -> &'a str {
243 // SAFETY: the caller must uphold the safety contract for `from_raw_parts_mut`.
244 unsafe { &mut *ptr::from_raw_parts_mut(data_pointer:ptr.cast(), metadata:len) }
245}
246