]>
Commit | Line | Data |
---|---|---|
1b1a35ee XL |
1 | //! Ways to create a `str` from bytes slice. |
2 | ||
3 | use crate::mem; | |
4 | ||
5 | use super::validations::run_utf8_validation; | |
6 | use 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 | |
6a06907d | 17 | /// [byteslice]: slice |
1b1a35ee XL |
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 | /// | |
6a06907d | 34 | /// [byteslice]: slice |
1b1a35ee XL |
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()`. | |
487cf647 | 80 | /// let sparkle_heart: &str = str::from_utf8(&sparkle_heart).unwrap(); |
1b1a35ee XL |
81 | /// |
82 | /// assert_eq!("💖", sparkle_heart); | |
83 | /// ``` | |
84 | #[stable(feature = "rust1", since = "1.0.0")] | |
923072b8 FG |
85 | #[rustc_const_stable(feature = "const_str_from_utf8_shared", since = "1.63.0")] |
86 | #[rustc_allow_const_fn_unstable(str_internals)] | |
3c0e092e | 87 | pub const fn from_utf8(v: &[u8]) -> Result<&str, Utf8Error> { |
923072b8 | 88 | // FIXME: This should use `?` again, once it's `const` |
3c0e092e XL |
89 | match run_utf8_validation(v) { |
90 | Ok(_) => { | |
91 | // SAFETY: validation succeeded. | |
92 | Ok(unsafe { from_utf8_unchecked(v) }) | |
93 | } | |
94 | Err(err) => Err(err), | |
95 | } | |
1b1a35ee XL |
96 | } |
97 | ||
98 | /// Converts a mutable slice of bytes to a mutable string slice. | |
99 | /// | |
100 | /// # Examples | |
101 | /// | |
102 | /// Basic usage: | |
103 | /// | |
104 | /// ``` | |
105 | /// use std::str; | |
106 | /// | |
107 | /// // "Hello, Rust!" as a mutable vector | |
108 | /// let mut hellorust = vec![72, 101, 108, 108, 111, 44, 32, 82, 117, 115, 116, 33]; | |
109 | /// | |
110 | /// // As we know these bytes are valid, we can use `unwrap()` | |
111 | /// let outstr = str::from_utf8_mut(&mut hellorust).unwrap(); | |
112 | /// | |
113 | /// assert_eq!("Hello, Rust!", outstr); | |
114 | /// ``` | |
115 | /// | |
116 | /// Incorrect bytes: | |
117 | /// | |
118 | /// ``` | |
119 | /// use std::str; | |
120 | /// | |
121 | /// // Some invalid bytes in a mutable vector | |
122 | /// let mut invalid = vec![128, 223]; | |
123 | /// | |
124 | /// assert!(str::from_utf8_mut(&mut invalid).is_err()); | |
125 | /// ``` | |
126 | /// See the docs for [`Utf8Error`] for more details on the kinds of | |
127 | /// errors that can be returned. | |
128 | #[stable(feature = "str_mut_extras", since = "1.20.0")] | |
3c0e092e XL |
129 | #[rustc_const_unstable(feature = "const_str_from_utf8", issue = "91006")] |
130 | pub const fn from_utf8_mut(v: &mut [u8]) -> Result<&mut str, Utf8Error> { | |
131 | // This should use `?` again, once it's `const` | |
132 | match run_utf8_validation(v) { | |
133 | Ok(_) => { | |
134 | // SAFETY: validation succeeded. | |
135 | Ok(unsafe { from_utf8_unchecked_mut(v) }) | |
136 | } | |
137 | Err(err) => Err(err), | |
138 | } | |
1b1a35ee XL |
139 | } |
140 | ||
141 | /// Converts a slice of bytes to a string slice without checking | |
142 | /// that the string contains valid UTF-8. | |
143 | /// | |
144 | /// See the safe version, [`from_utf8`], for more information. | |
145 | /// | |
146 | /// # Safety | |
147 | /// | |
04454e1e | 148 | /// The bytes passed in must be valid UTF-8. |
1b1a35ee XL |
149 | /// |
150 | /// # Examples | |
151 | /// | |
152 | /// Basic usage: | |
153 | /// | |
154 | /// ``` | |
155 | /// use std::str; | |
156 | /// | |
157 | /// // some bytes, in a vector | |
158 | /// let sparkle_heart = vec![240, 159, 146, 150]; | |
159 | /// | |
160 | /// let sparkle_heart = unsafe { | |
161 | /// str::from_utf8_unchecked(&sparkle_heart) | |
162 | /// }; | |
163 | /// | |
164 | /// assert_eq!("💖", sparkle_heart); | |
165 | /// ``` | |
166 | #[inline] | |
c295e0f8 | 167 | #[must_use] |
1b1a35ee | 168 | #[stable(feature = "rust1", since = "1.0.0")] |
136023e0 | 169 | #[rustc_const_stable(feature = "const_str_from_utf8_unchecked", since = "1.55.0")] |
1b1a35ee XL |
170 | pub const unsafe fn from_utf8_unchecked(v: &[u8]) -> &str { |
171 | // SAFETY: the caller must guarantee that the bytes `v` are valid UTF-8. | |
172 | // Also relies on `&str` and `&[u8]` having the same layout. | |
173 | unsafe { mem::transmute(v) } | |
174 | } | |
175 | ||
176 | /// Converts a slice of bytes to a string slice without checking | |
177 | /// that the string contains valid UTF-8; mutable version. | |
178 | /// | |
179 | /// See the immutable version, [`from_utf8_unchecked()`] for more information. | |
180 | /// | |
181 | /// # Examples | |
182 | /// | |
183 | /// Basic usage: | |
184 | /// | |
185 | /// ``` | |
186 | /// use std::str; | |
187 | /// | |
188 | /// let mut heart = vec![240, 159, 146, 150]; | |
189 | /// let heart = unsafe { str::from_utf8_unchecked_mut(&mut heart) }; | |
190 | /// | |
191 | /// assert_eq!("💖", heart); | |
192 | /// ``` | |
193 | #[inline] | |
c295e0f8 | 194 | #[must_use] |
1b1a35ee | 195 | #[stable(feature = "str_mut_extras", since = "1.20.0")] |
3c0e092e XL |
196 | #[rustc_const_unstable(feature = "const_str_from_utf8_unchecked_mut", issue = "91005")] |
197 | pub const unsafe fn from_utf8_unchecked_mut(v: &mut [u8]) -> &mut str { | |
1b1a35ee XL |
198 | // SAFETY: the caller must guarantee that the bytes `v` |
199 | // are valid UTF-8, thus the cast to `*mut str` is safe. | |
200 | // Also, the pointer dereference is safe because that pointer | |
201 | // comes from a reference which is guaranteed to be valid for writes. | |
202 | unsafe { &mut *(v as *mut [u8] as *mut str) } | |
203 | } |