]>
Commit | Line | Data |
---|---|---|
92a42be0 SL |
1 | //! A UTF-8 encoded, growable string. |
2 | //! | |
3 | //! This module contains the [`String`] type, a trait for converting | |
4 | //! [`ToString`]s, and several error types that may result from working with | |
5 | //! [`String`]s. | |
6 | //! | |
92a42be0 | 7 | //! [`ToString`]: trait.ToString.html |
9cc50fc6 SL |
8 | //! |
9 | //! # Examples | |
10 | //! | |
c30ab7b3 | 11 | //! There are multiple ways to create a new [`String`] from a string literal: |
9cc50fc6 | 12 | //! |
c30ab7b3 | 13 | //! ``` |
9cc50fc6 SL |
14 | //! let s = "Hello".to_string(); |
15 | //! | |
16 | //! let s = String::from("world"); | |
17 | //! let s: String = "also this".into(); | |
18 | //! ``` | |
19 | //! | |
c30ab7b3 | 20 | //! You can create a new [`String`] from an existing one by concatenating with |
9cc50fc6 SL |
21 | //! `+`: |
22 | //! | |
c30ab7b3 SL |
23 | //! [`String`]: struct.String.html |
24 | //! | |
25 | //! ``` | |
9cc50fc6 SL |
26 | //! let s = "Hello".to_string(); |
27 | //! | |
28 | //! let message = s + " world!"; | |
29 | //! ``` | |
30 | //! | |
3b2f2976 | 31 | //! If you have a vector of valid UTF-8 bytes, you can make a [`String`] out of |
9cc50fc6 SL |
32 | //! it. You can do the reverse too. |
33 | //! | |
c30ab7b3 | 34 | //! ``` |
9cc50fc6 SL |
35 | //! let sparkle_heart = vec![240, 159, 146, 150]; |
36 | //! | |
37 | //! // We know these bytes are valid, so we'll use `unwrap()`. | |
38 | //! let sparkle_heart = String::from_utf8(sparkle_heart).unwrap(); | |
39 | //! | |
40 | //! assert_eq!("💖", sparkle_heart); | |
41 | //! | |
42 | //! let bytes = sparkle_heart.into_bytes(); | |
43 | //! | |
44 | //! assert_eq!(bytes, [240, 159, 146, 150]); | |
45 | //! ``` | |
1a4d82fc | 46 | |
85aaf69f | 47 | #![stable(feature = "rust1", since = "1.0.0")] |
1a4d82fc | 48 | |
83c7162d | 49 | use core::char::{decode_utf16, REPLACEMENT_CHARACTER}; |
1a4d82fc JJ |
50 | use core::fmt; |
51 | use core::hash; | |
9e0c209e | 52 | use core::iter::{FromIterator, FusedIterator}; |
0531ce1d | 53 | use core::ops::{self, Add, AddAssign, Index, IndexMut, RangeBounds}; |
9fa01778 | 54 | use core::ops::Bound::{Excluded, Included, Unbounded}; |
1a4d82fc | 55 | use core::ptr; |
9fa01778 | 56 | use core::str::{pattern::Pattern, lossy}; |
1a4d82fc | 57 | |
9fa01778 XL |
58 | use crate::borrow::{Cow, ToOwned}; |
59 | use crate::collections::CollectionAllocErr; | |
60 | use crate::boxed::Box; | |
61 | use crate::str::{self, from_boxed_utf8_unchecked, FromStr, Utf8Error, Chars}; | |
62 | use crate::vec::Vec; | |
1a4d82fc | 63 | |
92a42be0 SL |
64 | /// A UTF-8 encoded, growable string. |
65 | /// | |
66 | /// The `String` type is the most common string type that has ownership over the | |
67 | /// contents of the string. It has a close relationship with its borrowed | |
68 | /// counterpart, the primitive [`str`]. | |
69 | /// | |
54a0048b | 70 | /// [`str`]: ../../std/primitive.str.html |
92a42be0 SL |
71 | /// |
72 | /// # Examples | |
73 | /// | |
3b2f2976 | 74 | /// You can create a `String` from a literal string with [`String::from`]: |
92a42be0 SL |
75 | /// |
76 | /// ``` | |
77 | /// let hello = String::from("Hello, world!"); | |
78 | /// ``` | |
79 | /// | |
cc61c64b XL |
80 | /// You can append a [`char`] to a `String` with the [`push`] method, and |
81 | /// append a [`&str`] with the [`push_str`] method: | |
92a42be0 SL |
82 | /// |
83 | /// ``` | |
84 | /// let mut hello = String::from("Hello, "); | |
85 | /// | |
86 | /// hello.push('w'); | |
87 | /// hello.push_str("orld!"); | |
88 | /// ``` | |
89 | /// | |
3b2f2976 | 90 | /// [`String::from`]: #method.from |
54a0048b | 91 | /// [`char`]: ../../std/primitive.char.html |
cc61c64b XL |
92 | /// [`push`]: #method.push |
93 | /// [`push_str`]: #method.push_str | |
92a42be0 SL |
94 | /// |
95 | /// If you have a vector of UTF-8 bytes, you can create a `String` from it with | |
cc61c64b | 96 | /// the [`from_utf8`] method: |
92a42be0 SL |
97 | /// |
98 | /// ``` | |
99 | /// // some bytes, in a vector | |
100 | /// let sparkle_heart = vec![240, 159, 146, 150]; | |
101 | /// | |
102 | /// // We know these bytes are valid, so we'll use `unwrap()`. | |
103 | /// let sparkle_heart = String::from_utf8(sparkle_heart).unwrap(); | |
104 | /// | |
105 | /// assert_eq!("💖", sparkle_heart); | |
106 | /// ``` | |
107 | /// | |
cc61c64b | 108 | /// [`from_utf8`]: #method.from_utf8 |
92a42be0 SL |
109 | /// |
110 | /// # UTF-8 | |
111 | /// | |
112 | /// `String`s are always valid UTF-8. This has a few implications, the first of | |
113 | /// which is that if you need a non-UTF-8 string, consider [`OsString`]. It is | |
114 | /// similar, but without the UTF-8 constraint. The second implication is that | |
115 | /// you cannot index into a `String`: | |
116 | /// | |
041b39d2 | 117 | /// ```compile_fail,E0277 |
92a42be0 SL |
118 | /// let s = "hello"; |
119 | /// | |
120 | /// println!("The first letter of s is {}", s[0]); // ERROR!!! | |
121 | /// ``` | |
122 | /// | |
54a0048b | 123 | /// [`OsString`]: ../../std/ffi/struct.OsString.html |
92a42be0 SL |
124 | /// |
125 | /// Indexing is intended to be a constant-time operation, but UTF-8 encoding | |
9e0c209e | 126 | /// does not allow us to do this. Furthermore, it's not clear what sort of |
92a42be0 | 127 | /// thing the index should return: a byte, a codepoint, or a grapheme cluster. |
cc61c64b | 128 | /// The [`bytes`] and [`chars`] methods return iterators over the first |
92a42be0 SL |
129 | /// two, respectively. |
130 | /// | |
cc61c64b XL |
131 | /// [`bytes`]: #method.bytes |
132 | /// [`chars`]: #method.chars | |
92a42be0 SL |
133 | /// |
134 | /// # Deref | |
135 | /// | |
136 | /// `String`s implement [`Deref`]`<Target=str>`, and so inherit all of [`str`]'s | |
3b2f2976 | 137 | /// methods. In addition, this means that you can pass a `String` to a |
92a42be0 SL |
138 | /// function which takes a [`&str`] by using an ampersand (`&`): |
139 | /// | |
140 | /// ``` | |
141 | /// fn takes_str(s: &str) { } | |
142 | /// | |
143 | /// let s = String::from("Hello"); | |
144 | /// | |
145 | /// takes_str(&s); | |
146 | /// ``` | |
147 | /// | |
92a42be0 SL |
148 | /// This will create a [`&str`] from the `String` and pass it in. This |
149 | /// conversion is very inexpensive, and so generally, functions will accept | |
3b2f2976 XL |
150 | /// [`&str`]s as arguments unless they need a `String` for some specific |
151 | /// reason. | |
152 | /// | |
153 | /// In certain cases Rust doesn't have enough information to make this | |
154 | /// conversion, known as [`Deref`] coercion. In the following example a string | |
155 | /// slice [`&'a str`][`&str`] implements the trait `TraitExample`, and the function | |
156 | /// `example_func` takes anything that implements the trait. In this case Rust | |
157 | /// would need to make two implicit conversions, which Rust doesn't have the | |
158 | /// means to do. For that reason, the following example will not compile. | |
159 | /// | |
160 | /// ```compile_fail,E0277 | |
161 | /// trait TraitExample {} | |
162 | /// | |
163 | /// impl<'a> TraitExample for &'a str {} | |
92a42be0 | 164 | /// |
3b2f2976 XL |
165 | /// fn example_func<A: TraitExample>(example_arg: A) {} |
166 | /// | |
167 | /// fn main() { | |
168 | /// let example_string = String::from("example_string"); | |
169 | /// example_func(&example_string); | |
170 | /// } | |
171 | /// ``` | |
172 | /// | |
173 | /// There are two options that would work instead. The first would be to | |
174 | /// change the line `example_func(&example_string);` to | |
175 | /// `example_func(example_string.as_str());`, using the method [`as_str()`] | |
176 | /// to explicitly extract the string slice containing the string. The second | |
177 | /// way changes `example_func(&example_string);` to | |
178 | /// `example_func(&*example_string);`. In this case we are dereferencing a | |
179 | /// `String` to a [`str`][`&str`], then referencing the [`str`][`&str`] back to | |
180 | /// [`&str`]. The second way is more idiomatic, however both work to do the | |
181 | /// conversion explicitly rather than relying on the implicit conversion. | |
92a42be0 SL |
182 | /// |
183 | /// # Representation | |
184 | /// | |
185 | /// A `String` is made up of three components: a pointer to some bytes, a | |
186 | /// length, and a capacity. The pointer points to an internal buffer `String` | |
187 | /// uses to store its data. The length is the number of bytes currently stored | |
188 | /// in the buffer, and the capacity is the size of the buffer in bytes. As such, | |
189 | /// the length will always be less than or equal to the capacity. | |
190 | /// | |
191 | /// This buffer is always stored on the heap. | |
192 | /// | |
cc61c64b | 193 | /// You can look at these with the [`as_ptr`], [`len`], and [`capacity`] |
92a42be0 SL |
194 | /// methods: |
195 | /// | |
196 | /// ``` | |
197 | /// use std::mem; | |
198 | /// | |
199 | /// let story = String::from("Once upon a time..."); | |
200 | /// | |
201 | /// let ptr = story.as_ptr(); | |
202 | /// let len = story.len(); | |
203 | /// let capacity = story.capacity(); | |
204 | /// | |
a7813a04 | 205 | /// // story has nineteen bytes |
92a42be0 SL |
206 | /// assert_eq!(19, len); |
207 | /// | |
208 | /// // Now that we have our parts, we throw the story away. | |
209 | /// mem::forget(story); | |
210 | /// | |
211 | /// // We can re-build a String out of ptr, len, and capacity. This is all | |
7453a54e | 212 | /// // unsafe because we are responsible for making sure the components are |
92a42be0 SL |
213 | /// // valid: |
214 | /// let s = unsafe { String::from_raw_parts(ptr as *mut _, len, capacity) } ; | |
215 | /// | |
216 | /// assert_eq!(String::from("Once upon a time..."), s); | |
217 | /// ``` | |
218 | /// | |
cc61c64b XL |
219 | /// [`as_ptr`]: #method.as_ptr |
220 | /// [`len`]: #method.len | |
221 | /// [`capacity`]: #method.capacity | |
92a42be0 SL |
222 | /// |
223 | /// If a `String` has enough capacity, adding elements to it will not | |
224 | /// re-allocate. For example, consider this program: | |
225 | /// | |
226 | /// ``` | |
227 | /// let mut s = String::new(); | |
228 | /// | |
229 | /// println!("{}", s.capacity()); | |
230 | /// | |
231 | /// for _ in 0..5 { | |
232 | /// s.push_str("hello"); | |
233 | /// println!("{}", s.capacity()); | |
234 | /// } | |
235 | /// ``` | |
236 | /// | |
237 | /// This will output the following: | |
238 | /// | |
239 | /// ```text | |
240 | /// 0 | |
241 | /// 5 | |
242 | /// 10 | |
243 | /// 20 | |
244 | /// 20 | |
245 | /// 40 | |
246 | /// ``` | |
247 | /// | |
248 | /// At first, we have no memory allocated at all, but as we append to the | |
249 | /// string, it increases its capacity appropriately. If we instead use the | |
cc61c64b | 250 | /// [`with_capacity`] method to allocate the correct capacity initially: |
92a42be0 SL |
251 | /// |
252 | /// ``` | |
253 | /// let mut s = String::with_capacity(25); | |
254 | /// | |
255 | /// println!("{}", s.capacity()); | |
256 | /// | |
257 | /// for _ in 0..5 { | |
258 | /// s.push_str("hello"); | |
259 | /// println!("{}", s.capacity()); | |
260 | /// } | |
261 | /// ``` | |
262 | /// | |
cc61c64b | 263 | /// [`with_capacity`]: #method.with_capacity |
92a42be0 SL |
264 | /// |
265 | /// We end up with a different output: | |
266 | /// | |
267 | /// ```text | |
268 | /// 25 | |
269 | /// 25 | |
270 | /// 25 | |
271 | /// 25 | |
272 | /// 25 | |
273 | /// 25 | |
274 | /// ``` | |
275 | /// | |
276 | /// Here, there's no need to allocate more memory inside the loop. | |
3b2f2976 XL |
277 | /// |
278 | /// [`&str`]: ../../std/primitive.str.html | |
279 | /// [`Deref`]: ../../std/ops/trait.Deref.html | |
280 | /// [`as_str()`]: struct.String.html#method.as_str | |
b039eaaf | 281 | #[derive(PartialOrd, Eq, Ord)] |
85aaf69f | 282 | #[stable(feature = "rust1", since = "1.0.0")] |
1a4d82fc JJ |
283 | pub struct String { |
284 | vec: Vec<u8>, | |
285 | } | |
286 | ||
92a42be0 SL |
287 | /// A possible error value when converting a `String` from a UTF-8 byte vector. |
288 | /// | |
cc61c64b | 289 | /// This type is the error type for the [`from_utf8`] method on [`String`]. It |
92a42be0 | 290 | /// is designed in such a way to carefully avoid reallocations: the |
cc61c64b | 291 | /// [`into_bytes`] method will give back the byte vector that was used in the |
92a42be0 SL |
292 | /// conversion attempt. |
293 | /// | |
cc61c64b | 294 | /// [`from_utf8`]: struct.String.html#method.from_utf8 |
92a42be0 | 295 | /// [`String`]: struct.String.html |
cc61c64b | 296 | /// [`into_bytes`]: struct.FromUtf8Error.html#method.into_bytes |
92a42be0 SL |
297 | /// |
298 | /// The [`Utf8Error`] type provided by [`std::str`] represents an error that may | |
299 | /// occur when converting a slice of [`u8`]s to a [`&str`]. In this sense, it's | |
300 | /// an analogue to `FromUtf8Error`, and you can get one from a `FromUtf8Error` | |
cc61c64b | 301 | /// through the [`utf8_error`] method. |
92a42be0 | 302 | /// |
54a0048b SL |
303 | /// [`Utf8Error`]: ../../std/str/struct.Utf8Error.html |
304 | /// [`std::str`]: ../../std/str/index.html | |
305 | /// [`u8`]: ../../std/primitive.u8.html | |
306 | /// [`&str`]: ../../std/primitive.str.html | |
cc61c64b | 307 | /// [`utf8_error`]: #method.utf8_error |
92a42be0 SL |
308 | /// |
309 | /// # Examples | |
310 | /// | |
311 | /// Basic usage: | |
312 | /// | |
313 | /// ``` | |
314 | /// // some invalid bytes, in a vector | |
315 | /// let bytes = vec![0, 159]; | |
316 | /// | |
317 | /// let value = String::from_utf8(bytes); | |
318 | /// | |
319 | /// assert!(value.is_err()); | |
320 | /// assert_eq!(vec![0, 159], value.unwrap_err().into_bytes()); | |
321 | /// ``` | |
85aaf69f SL |
322 | #[stable(feature = "rust1", since = "1.0.0")] |
323 | #[derive(Debug)] | |
1a4d82fc JJ |
324 | pub struct FromUtf8Error { |
325 | bytes: Vec<u8>, | |
326 | error: Utf8Error, | |
327 | } | |
328 | ||
92a42be0 SL |
329 | /// A possible error value when converting a `String` from a UTF-16 byte slice. |
330 | /// | |
cc61c64b | 331 | /// This type is the error type for the [`from_utf16`] method on [`String`]. |
92a42be0 | 332 | /// |
cc61c64b | 333 | /// [`from_utf16`]: struct.String.html#method.from_utf16 |
92a42be0 SL |
334 | /// [`String`]: struct.String.html |
335 | /// | |
336 | /// # Examples | |
337 | /// | |
338 | /// Basic usage: | |
339 | /// | |
340 | /// ``` | |
341 | /// // 𝄞mu<invalid>ic | |
342 | /// let v = &[0xD834, 0xDD1E, 0x006d, 0x0075, | |
343 | /// 0xD800, 0x0069, 0x0063]; | |
344 | /// | |
345 | /// assert!(String::from_utf16(v).is_err()); | |
346 | /// ``` | |
85aaf69f SL |
347 | #[stable(feature = "rust1", since = "1.0.0")] |
348 | #[derive(Debug)] | |
1a4d82fc JJ |
349 | pub struct FromUtf16Error(()); |
350 | ||
351 | impl String { | |
9cc50fc6 SL |
352 | /// Creates a new empty `String`. |
353 | /// | |
354 | /// Given that the `String` is empty, this will not allocate any initial | |
355 | /// buffer. While that means that this initial operation is very | |
0531ce1d | 356 | /// inexpensive, it may cause excessive allocation later when you add |
9cc50fc6 | 357 | /// data. If you have an idea of how much data the `String` will hold, |
cc61c64b | 358 | /// consider the [`with_capacity`] method to prevent excessive |
9cc50fc6 SL |
359 | /// re-allocation. |
360 | /// | |
cc61c64b | 361 | /// [`with_capacity`]: #method.with_capacity |
1a4d82fc JJ |
362 | /// |
363 | /// # Examples | |
364 | /// | |
9cc50fc6 SL |
365 | /// Basic usage: |
366 | /// | |
1a4d82fc | 367 | /// ``` |
9cc50fc6 | 368 | /// let s = String::new(); |
1a4d82fc JJ |
369 | /// ``` |
370 | #[inline] | |
85aaf69f | 371 | #[stable(feature = "rust1", since = "1.0.0")] |
94b46f34 XL |
372 | #[rustc_const_unstable(feature = "const_string_new")] |
373 | pub const fn new() -> String { | |
92a42be0 | 374 | String { vec: Vec::new() } |
1a4d82fc JJ |
375 | } |
376 | ||
9cc50fc6 SL |
377 | /// Creates a new empty `String` with a particular capacity. |
378 | /// | |
379 | /// `String`s have an internal buffer to hold their data. The capacity is | |
cc61c64b | 380 | /// the length of that buffer, and can be queried with the [`capacity`] |
9cc50fc6 SL |
381 | /// method. This method creates an empty `String`, but one with an initial |
382 | /// buffer that can hold `capacity` bytes. This is useful when you may be | |
383 | /// appending a bunch of data to the `String`, reducing the number of | |
384 | /// reallocations it needs to do. | |
385 | /// | |
cc61c64b | 386 | /// [`capacity`]: #method.capacity |
9cc50fc6 SL |
387 | /// |
388 | /// If the given capacity is `0`, no allocation will occur, and this method | |
cc61c64b | 389 | /// is identical to the [`new`] method. |
9cc50fc6 | 390 | /// |
cc61c64b | 391 | /// [`new`]: #method.new |
1a4d82fc JJ |
392 | /// |
393 | /// # Examples | |
394 | /// | |
9cc50fc6 SL |
395 | /// Basic usage: |
396 | /// | |
1a4d82fc JJ |
397 | /// ``` |
398 | /// let mut s = String::with_capacity(10); | |
92a42be0 SL |
399 | /// |
400 | /// // The String contains no chars, even though it has capacity for more | |
401 | /// assert_eq!(s.len(), 0); | |
402 | /// | |
403 | /// // These are all done without reallocating... | |
404 | /// let cap = s.capacity(); | |
a1dfa0c6 | 405 | /// for _ in 0..10 { |
92a42be0 SL |
406 | /// s.push('a'); |
407 | /// } | |
408 | /// | |
409 | /// assert_eq!(s.capacity(), cap); | |
410 | /// | |
411 | /// // ...but this may make the vector reallocate | |
412 | /// s.push('a'); | |
1a4d82fc JJ |
413 | /// ``` |
414 | #[inline] | |
85aaf69f SL |
415 | #[stable(feature = "rust1", since = "1.0.0")] |
416 | pub fn with_capacity(capacity: usize) -> String { | |
92a42be0 | 417 | String { vec: Vec::with_capacity(capacity) } |
1a4d82fc JJ |
418 | } |
419 | ||
c34b1796 AL |
420 | // HACK(japaric): with cfg(test) the inherent `[T]::to_vec` method, which is |
421 | // required for this method definition, is not available. Since we don't | |
422 | // require this method for testing purposes, I'll just stub it | |
423 | // NB see the slice::hack module in slice.rs for more information | |
424 | #[inline] | |
425 | #[cfg(test)] | |
426 | pub fn from_str(_: &str) -> String { | |
427 | panic!("not available with cfg(test)"); | |
1a4d82fc JJ |
428 | } |
429 | ||
b039eaaf SL |
430 | /// Converts a vector of bytes to a `String`. |
431 | /// | |
9cc50fc6 SL |
432 | /// A string slice ([`&str`]) is made of bytes ([`u8`]), and a vector of bytes |
433 | /// ([`Vec<u8>`]) is made of bytes, so this function converts between the | |
b039eaaf SL |
434 | /// two. Not all byte slices are valid `String`s, however: `String` |
435 | /// requires that it is valid UTF-8. `from_utf8()` checks to ensure that | |
436 | /// the bytes are valid UTF-8, and then does the conversion. | |
437 | /// | |
438 | /// If you are sure that the byte slice is valid UTF-8, and you don't want | |
439 | /// to incur the overhead of the validity check, there is an unsafe version | |
cc61c64b | 440 | /// of this function, [`from_utf8_unchecked`], which has the same behavior |
9cc50fc6 | 441 | /// but skips the check. |
b039eaaf | 442 | /// |
b039eaaf SL |
443 | /// This method will take care to not copy the vector, for efficiency's |
444 | /// sake. | |
445 | /// | |
3b2f2976 | 446 | /// If you need a [`&str`] instead of a `String`, consider |
cc61c64b | 447 | /// [`str::from_utf8`]. |
b039eaaf | 448 | /// |
8bb4bdeb XL |
449 | /// The inverse of this method is [`as_bytes`]. |
450 | /// | |
7453a54e | 451 | /// # Errors |
1a4d82fc | 452 | /// |
3b2f2976 | 453 | /// Returns [`Err`] if the slice is not UTF-8 with a description as to why the |
b039eaaf | 454 | /// provided bytes are not UTF-8. The vector you moved in is also included. |
1a4d82fc JJ |
455 | /// |
456 | /// # Examples | |
457 | /// | |
b039eaaf SL |
458 | /// Basic usage: |
459 | /// | |
c34b1796 | 460 | /// ``` |
b039eaaf SL |
461 | /// // some bytes, in a vector |
462 | /// let sparkle_heart = vec![240, 159, 146, 150]; | |
463 | /// | |
92a42be0 | 464 | /// // We know these bytes are valid, so we'll use `unwrap()`. |
b039eaaf SL |
465 | /// let sparkle_heart = String::from_utf8(sparkle_heart).unwrap(); |
466 | /// | |
467 | /// assert_eq!("💖", sparkle_heart); | |
468 | /// ``` | |
469 | /// | |
470 | /// Incorrect bytes: | |
471 | /// | |
1a4d82fc | 472 | /// ``` |
b039eaaf SL |
473 | /// // some invalid bytes, in a vector |
474 | /// let sparkle_heart = vec![0, 159, 146, 150]; | |
475 | /// | |
476 | /// assert!(String::from_utf8(sparkle_heart).is_err()); | |
477 | /// ``` | |
478 | /// | |
9cc50fc6 SL |
479 | /// See the docs for [`FromUtf8Error`] for more details on what you can do |
480 | /// with this error. | |
b039eaaf | 481 | /// |
3b2f2976 XL |
482 | /// [`from_utf8_unchecked`]: struct.String.html#method.from_utf8_unchecked |
483 | /// [`&str`]: ../../std/primitive.str.html | |
484 | /// [`u8`]: ../../std/primitive.u8.html | |
485 | /// [`Vec<u8>`]: ../../std/vec/struct.Vec.html | |
486 | /// [`str::from_utf8`]: ../../std/str/fn.from_utf8.html | |
487 | /// [`as_bytes`]: struct.String.html#method.as_bytes | |
9cc50fc6 | 488 | /// [`FromUtf8Error`]: struct.FromUtf8Error.html |
9fa01778 | 489 | /// [`Err`]: ../../std/result/enum.Result.html#variant.Err |
1a4d82fc | 490 | #[inline] |
85aaf69f | 491 | #[stable(feature = "rust1", since = "1.0.0")] |
1a4d82fc | 492 | pub fn from_utf8(vec: Vec<u8>) -> Result<String, FromUtf8Error> { |
85aaf69f | 493 | match str::from_utf8(&vec) { |
a1dfa0c6 | 494 | Ok(..) => Ok(String { vec }), |
92a42be0 SL |
495 | Err(e) => { |
496 | Err(FromUtf8Error { | |
497 | bytes: vec, | |
498 | error: e, | |
499 | }) | |
500 | } | |
1a4d82fc JJ |
501 | } |
502 | } | |
503 | ||
7453a54e | 504 | /// Converts a slice of bytes to a string, including invalid characters. |
b039eaaf | 505 | /// |
7453a54e SL |
506 | /// Strings are made of bytes ([`u8`]), and a slice of bytes |
507 | /// ([`&[u8]`][byteslice]) is made of bytes, so this function converts | |
508 | /// between the two. Not all byte slices are valid strings, however: strings | |
509 | /// are required to be valid UTF-8. During this conversion, | |
9cc50fc6 | 510 | /// `from_utf8_lossy()` will replace any invalid UTF-8 sequences with |
b7449926 | 511 | /// [`U+FFFD REPLACEMENT CHARACTER`][U+FFFD], which looks like this: � |
b039eaaf | 512 | /// |
54a0048b SL |
513 | /// [`u8`]: ../../std/primitive.u8.html |
514 | /// [byteslice]: ../../std/primitive.slice.html | |
b7449926 | 515 | /// [U+FFFD]: ../char/constant.REPLACEMENT_CHARACTER.html |
9cc50fc6 | 516 | /// |
b039eaaf SL |
517 | /// If you are sure that the byte slice is valid UTF-8, and you don't want |
518 | /// to incur the overhead of the conversion, there is an unsafe version | |
cc61c64b | 519 | /// of this function, [`from_utf8_unchecked`], which has the same behavior |
9cc50fc6 | 520 | /// but skips the checks. |
b039eaaf | 521 | /// |
cc61c64b | 522 | /// [`from_utf8_unchecked`]: struct.String.html#method.from_utf8_unchecked |
b039eaaf | 523 | /// |
7453a54e SL |
524 | /// This function returns a [`Cow<'a, str>`]. If our byte slice is invalid |
525 | /// UTF-8, then we need to insert the replacement characters, which will | |
526 | /// change the size of the string, and hence, require a `String`. But if | |
527 | /// it's already valid UTF-8, we don't need a new allocation. This return | |
528 | /// type allows us to handle both cases. | |
b039eaaf | 529 | /// |
54a0048b | 530 | /// [`Cow<'a, str>`]: ../../std/borrow/enum.Cow.html |
1a4d82fc JJ |
531 | /// |
532 | /// # Examples | |
533 | /// | |
b039eaaf SL |
534 | /// Basic usage: |
535 | /// | |
c34b1796 | 536 | /// ``` |
b039eaaf SL |
537 | /// // some bytes, in a vector |
538 | /// let sparkle_heart = vec![240, 159, 146, 150]; | |
539 | /// | |
7453a54e | 540 | /// let sparkle_heart = String::from_utf8_lossy(&sparkle_heart); |
b039eaaf SL |
541 | /// |
542 | /// assert_eq!("💖", sparkle_heart); | |
543 | /// ``` | |
544 | /// | |
545 | /// Incorrect bytes: | |
546 | /// | |
547 | /// ``` | |
548 | /// // some invalid bytes | |
1a4d82fc JJ |
549 | /// let input = b"Hello \xF0\x90\x80World"; |
550 | /// let output = String::from_utf8_lossy(input); | |
b039eaaf SL |
551 | /// |
552 | /// assert_eq!("Hello �World", output); | |
1a4d82fc | 553 | /// ``` |
85aaf69f SL |
554 | #[stable(feature = "rust1", since = "1.0.0")] |
555 | pub fn from_utf8_lossy<'a>(v: &'a [u8]) -> Cow<'a, str> { | |
041b39d2 | 556 | let mut iter = lossy::Utf8Lossy::from_bytes(v).chunks(); |
1a4d82fc | 557 | |
041b39d2 XL |
558 | let (first_valid, first_broken) = if let Some(chunk) = iter.next() { |
559 | let lossy::Utf8LossyChunk { valid, broken } = chunk; | |
560 | if valid.len() == v.len() { | |
561 | debug_assert!(broken.is_empty()); | |
562 | return Cow::Borrowed(valid); | |
563 | } | |
564 | (valid, broken) | |
565 | } else { | |
566 | return Cow::Borrowed(""); | |
567 | }; | |
1a4d82fc | 568 | |
0731742a | 569 | const REPLACEMENT: &str = "\u{FFFD}"; |
1a4d82fc | 570 | |
041b39d2 XL |
571 | let mut res = String::with_capacity(v.len()); |
572 | res.push_str(first_valid); | |
573 | if !first_broken.is_empty() { | |
574 | res.push_str(REPLACEMENT); | |
1a4d82fc JJ |
575 | } |
576 | ||
041b39d2 XL |
577 | for lossy::Utf8LossyChunk { valid, broken } in iter { |
578 | res.push_str(valid); | |
579 | if !broken.is_empty() { | |
580 | res.push_str(REPLACEMENT); | |
1a4d82fc JJ |
581 | } |
582 | } | |
041b39d2 | 583 | |
1a4d82fc JJ |
584 | Cow::Owned(res) |
585 | } | |
586 | ||
3b2f2976 | 587 | /// Decode a UTF-16 encoded vector `v` into a `String`, returning [`Err`] |
1a4d82fc JJ |
588 | /// if `v` contains any invalid data. |
589 | /// | |
ff7c6d11 | 590 | /// [`Err`]: ../../std/result/enum.Result.html#variant.Err |
3b2f2976 | 591 | /// |
1a4d82fc JJ |
592 | /// # Examples |
593 | /// | |
9cc50fc6 SL |
594 | /// Basic usage: |
595 | /// | |
c34b1796 | 596 | /// ``` |
1a4d82fc | 597 | /// // 𝄞music |
92a42be0 SL |
598 | /// let v = &[0xD834, 0xDD1E, 0x006d, 0x0075, |
599 | /// 0x0073, 0x0069, 0x0063]; | |
9cc50fc6 SL |
600 | /// assert_eq!(String::from("𝄞music"), |
601 | /// String::from_utf16(v).unwrap()); | |
1a4d82fc JJ |
602 | /// |
603 | /// // 𝄞mu<invalid>ic | |
92a42be0 SL |
604 | /// let v = &[0xD834, 0xDD1E, 0x006d, 0x0075, |
605 | /// 0xD800, 0x0069, 0x0063]; | |
1a4d82fc JJ |
606 | /// assert!(String::from_utf16(v).is_err()); |
607 | /// ``` | |
85aaf69f | 608 | #[stable(feature = "rust1", since = "1.0.0")] |
1a4d82fc | 609 | pub fn from_utf16(v: &[u16]) -> Result<String, FromUtf16Error> { |
a1dfa0c6 XL |
610 | // This isn't done via collect::<Result<_, _>>() for performance reasons. |
611 | // FIXME: the function can be simplified again when #48994 is closed. | |
612 | let mut ret = String::with_capacity(v.len()); | |
613 | for c in decode_utf16(v.iter().cloned()) { | |
614 | if let Ok(c) = c { | |
615 | ret.push(c); | |
616 | } else { | |
617 | return Err(FromUtf16Error(())); | |
618 | } | |
619 | } | |
620 | Ok(ret) | |
1a4d82fc JJ |
621 | } |
622 | ||
3b2f2976 | 623 | /// Decode a UTF-16 encoded slice `v` into a `String`, replacing |
b7449926 | 624 | /// invalid data with [the replacement character (`U+FFFD`)][U+FFFD]. |
1a4d82fc | 625 | /// |
ea8adc8c XL |
626 | /// Unlike [`from_utf8_lossy`] which returns a [`Cow<'a, str>`], |
627 | /// `from_utf16_lossy` returns a `String` since the UTF-16 to UTF-8 | |
628 | /// conversion requires a memory allocation. | |
629 | /// | |
630 | /// [`from_utf8_lossy`]: #method.from_utf8_lossy | |
631 | /// [`Cow<'a, str>`]: ../borrow/enum.Cow.html | |
b7449926 | 632 | /// [U+FFFD]: ../char/constant.REPLACEMENT_CHARACTER.html |
ea8adc8c | 633 | /// |
1a4d82fc JJ |
634 | /// # Examples |
635 | /// | |
9cc50fc6 SL |
636 | /// Basic usage: |
637 | /// | |
c34b1796 | 638 | /// ``` |
1a4d82fc JJ |
639 | /// // 𝄞mus<invalid>ic<invalid> |
640 | /// let v = &[0xD834, 0xDD1E, 0x006d, 0x0075, | |
641 | /// 0x0073, 0xDD1E, 0x0069, 0x0063, | |
642 | /// 0xD834]; | |
643 | /// | |
9cc50fc6 SL |
644 | /// assert_eq!(String::from("𝄞mus\u{FFFD}ic\u{FFFD}"), |
645 | /// String::from_utf16_lossy(v)); | |
1a4d82fc | 646 | /// ``` |
85aaf69f SL |
647 | #[inline] |
648 | #[stable(feature = "rust1", since = "1.0.0")] | |
1a4d82fc | 649 | pub fn from_utf16_lossy(v: &[u16]) -> String { |
e9174d1e | 650 | decode_utf16(v.iter().cloned()).map(|r| r.unwrap_or(REPLACEMENT_CHARACTER)).collect() |
1a4d82fc JJ |
651 | } |
652 | ||
653 | /// Creates a new `String` from a length, capacity, and pointer. | |
654 | /// | |
b039eaaf | 655 | /// # Safety |
c1a9b12d | 656 | /// |
9cc50fc6 SL |
657 | /// This is highly unsafe, due to the number of invariants that aren't |
658 | /// checked: | |
659 | /// | |
660 | /// * The memory at `ptr` needs to have been previously allocated by the | |
661 | /// same allocator the standard library uses. | |
662 | /// * `length` needs to be less than or equal to `capacity`. | |
663 | /// * `capacity` needs to be the correct value. | |
664 | /// | |
665 | /// Violating these may cause problems like corrupting the allocator's | |
3b2f2976 | 666 | /// internal data structures. |
9cc50fc6 | 667 | /// |
5bcae85e SL |
668 | /// The ownership of `ptr` is effectively transferred to the |
669 | /// `String` which may then deallocate, reallocate or change the | |
670 | /// contents of memory pointed to by the pointer at will. Ensure | |
671 | /// that nothing else uses the pointer after calling this | |
672 | /// function. | |
673 | /// | |
9cc50fc6 SL |
674 | /// # Examples |
675 | /// | |
676 | /// Basic usage: | |
677 | /// | |
678 | /// ``` | |
679 | /// use std::mem; | |
680 | /// | |
681 | /// unsafe { | |
682 | /// let s = String::from("hello"); | |
683 | /// let ptr = s.as_ptr(); | |
684 | /// let len = s.len(); | |
685 | /// let capacity = s.capacity(); | |
686 | /// | |
687 | /// mem::forget(s); | |
688 | /// | |
689 | /// let s = String::from_raw_parts(ptr as *mut _, len, capacity); | |
c34b1796 | 690 | /// |
9cc50fc6 SL |
691 | /// assert_eq!(String::from("hello"), s); |
692 | /// } | |
693 | /// ``` | |
1a4d82fc | 694 | #[inline] |
85aaf69f SL |
695 | #[stable(feature = "rust1", since = "1.0.0")] |
696 | pub unsafe fn from_raw_parts(buf: *mut u8, length: usize, capacity: usize) -> String { | |
92a42be0 | 697 | String { vec: Vec::from_raw_parts(buf, length, capacity) } |
1a4d82fc JJ |
698 | } |
699 | ||
b039eaaf SL |
700 | /// Converts a vector of bytes to a `String` without checking that the |
701 | /// string contains valid UTF-8. | |
702 | /// | |
cc61c64b | 703 | /// See the safe version, [`from_utf8`], for more details. |
b039eaaf | 704 | /// |
cc61c64b | 705 | /// [`from_utf8`]: struct.String.html#method.from_utf8 |
b039eaaf SL |
706 | /// |
707 | /// # Safety | |
708 | /// | |
9cc50fc6 SL |
709 | /// This function is unsafe because it does not check that the bytes passed |
710 | /// to it are valid UTF-8. If this constraint is violated, it may cause | |
711 | /// memory unsafety issues with future users of the `String`, as the rest of | |
712 | /// the standard library assumes that `String`s are valid UTF-8. | |
b039eaaf SL |
713 | /// |
714 | /// # Examples | |
715 | /// | |
716 | /// Basic usage: | |
717 | /// | |
718 | /// ``` | |
719 | /// // some bytes, in a vector | |
720 | /// let sparkle_heart = vec![240, 159, 146, 150]; | |
721 | /// | |
722 | /// let sparkle_heart = unsafe { | |
723 | /// String::from_utf8_unchecked(sparkle_heart) | |
724 | /// }; | |
725 | /// | |
726 | /// assert_eq!("💖", sparkle_heart); | |
727 | /// ``` | |
1a4d82fc | 728 | #[inline] |
85aaf69f | 729 | #[stable(feature = "rust1", since = "1.0.0")] |
1a4d82fc JJ |
730 | pub unsafe fn from_utf8_unchecked(bytes: Vec<u8>) -> String { |
731 | String { vec: bytes } | |
732 | } | |
733 | ||
9cc50fc6 SL |
734 | /// Converts a `String` into a byte vector. |
735 | /// | |
736 | /// This consumes the `String`, so we do not need to copy its contents. | |
1a4d82fc JJ |
737 | /// |
738 | /// # Examples | |
739 | /// | |
9cc50fc6 SL |
740 | /// Basic usage: |
741 | /// | |
1a4d82fc | 742 | /// ``` |
62682a34 | 743 | /// let s = String::from("hello"); |
1a4d82fc | 744 | /// let bytes = s.into_bytes(); |
9cc50fc6 SL |
745 | /// |
746 | /// assert_eq!(&[104, 101, 108, 108, 111][..], &bytes[..]); | |
1a4d82fc JJ |
747 | /// ``` |
748 | #[inline] | |
85aaf69f | 749 | #[stable(feature = "rust1", since = "1.0.0")] |
1a4d82fc JJ |
750 | pub fn into_bytes(self) -> Vec<u8> { |
751 | self.vec | |
752 | } | |
753 | ||
b7449926 | 754 | /// Extracts a string slice containing the entire `String`. |
ea8adc8c XL |
755 | /// |
756 | /// # Examples | |
757 | /// | |
758 | /// Basic usage: | |
759 | /// | |
760 | /// ``` | |
761 | /// let s = String::from("foo"); | |
762 | /// | |
763 | /// assert_eq!("foo", s.as_str()); | |
764 | /// ``` | |
c34b1796 | 765 | #[inline] |
9cc50fc6 | 766 | #[stable(feature = "string_as_str", since = "1.7.0")] |
c34b1796 AL |
767 | pub fn as_str(&self) -> &str { |
768 | self | |
769 | } | |
770 | ||
ea8adc8c XL |
771 | /// Converts a `String` into a mutable string slice. |
772 | /// | |
773 | /// # Examples | |
774 | /// | |
775 | /// Basic usage: | |
776 | /// | |
777 | /// ``` | |
ea8adc8c XL |
778 | /// let mut s = String::from("foobar"); |
779 | /// let s_mut_str = s.as_mut_str(); | |
780 | /// | |
781 | /// s_mut_str.make_ascii_uppercase(); | |
782 | /// | |
783 | /// assert_eq!("FOOBAR", s_mut_str); | |
784 | /// ``` | |
9cc50fc6 SL |
785 | #[inline] |
786 | #[stable(feature = "string_as_str", since = "1.7.0")] | |
787 | pub fn as_mut_str(&mut self) -> &mut str { | |
788 | self | |
789 | } | |
790 | ||
791 | /// Appends a given string slice onto the end of this `String`. | |
1a4d82fc JJ |
792 | /// |
793 | /// # Examples | |
794 | /// | |
9cc50fc6 SL |
795 | /// Basic usage: |
796 | /// | |
1a4d82fc | 797 | /// ``` |
62682a34 | 798 | /// let mut s = String::from("foo"); |
9cc50fc6 | 799 | /// |
1a4d82fc | 800 | /// s.push_str("bar"); |
9cc50fc6 SL |
801 | /// |
802 | /// assert_eq!("foobar", s); | |
1a4d82fc JJ |
803 | /// ``` |
804 | #[inline] | |
85aaf69f | 805 | #[stable(feature = "rust1", since = "1.0.0")] |
1a4d82fc | 806 | pub fn push_str(&mut self, string: &str) { |
92a42be0 | 807 | self.vec.extend_from_slice(string.as_bytes()) |
1a4d82fc JJ |
808 | } |
809 | ||
9cc50fc6 | 810 | /// Returns this `String`'s capacity, in bytes. |
1a4d82fc JJ |
811 | /// |
812 | /// # Examples | |
813 | /// | |
9cc50fc6 SL |
814 | /// Basic usage: |
815 | /// | |
1a4d82fc JJ |
816 | /// ``` |
817 | /// let s = String::with_capacity(10); | |
9cc50fc6 | 818 | /// |
1a4d82fc JJ |
819 | /// assert!(s.capacity() >= 10); |
820 | /// ``` | |
821 | #[inline] | |
85aaf69f SL |
822 | #[stable(feature = "rust1", since = "1.0.0")] |
823 | pub fn capacity(&self) -> usize { | |
1a4d82fc JJ |
824 | self.vec.capacity() |
825 | } | |
826 | ||
9cc50fc6 SL |
827 | /// Ensures that this `String`'s capacity is at least `additional` bytes |
828 | /// larger than its length. | |
829 | /// | |
830 | /// The capacity may be increased by more than `additional` bytes if it | |
831 | /// chooses, to prevent frequent reallocations. | |
832 | /// | |
cc61c64b | 833 | /// If you do not want this "at least" behavior, see the [`reserve_exact`] |
9cc50fc6 SL |
834 | /// method. |
835 | /// | |
1a4d82fc JJ |
836 | /// # Panics |
837 | /// | |
3b2f2976 XL |
838 | /// Panics if the new capacity overflows [`usize`]. |
839 | /// | |
840 | /// [`reserve_exact`]: struct.String.html#method.reserve_exact | |
841 | /// [`usize`]: ../../std/primitive.usize.html | |
1a4d82fc JJ |
842 | /// |
843 | /// # Examples | |
844 | /// | |
9cc50fc6 SL |
845 | /// Basic usage: |
846 | /// | |
1a4d82fc JJ |
847 | /// ``` |
848 | /// let mut s = String::new(); | |
9cc50fc6 | 849 | /// |
1a4d82fc | 850 | /// s.reserve(10); |
9cc50fc6 | 851 | /// |
1a4d82fc JJ |
852 | /// assert!(s.capacity() >= 10); |
853 | /// ``` | |
9cc50fc6 SL |
854 | /// |
855 | /// This may not actually increase the capacity: | |
856 | /// | |
857 | /// ``` | |
858 | /// let mut s = String::with_capacity(10); | |
859 | /// s.push('a'); | |
860 | /// s.push('b'); | |
861 | /// | |
862 | /// // s now has a length of 2 and a capacity of 10 | |
863 | /// assert_eq!(2, s.len()); | |
864 | /// assert_eq!(10, s.capacity()); | |
865 | /// | |
866 | /// // Since we already have an extra 8 capacity, calling this... | |
867 | /// s.reserve(8); | |
868 | /// | |
869 | /// // ... doesn't actually increase. | |
870 | /// assert_eq!(10, s.capacity()); | |
871 | /// ``` | |
1a4d82fc | 872 | #[inline] |
85aaf69f SL |
873 | #[stable(feature = "rust1", since = "1.0.0")] |
874 | pub fn reserve(&mut self, additional: usize) { | |
1a4d82fc JJ |
875 | self.vec.reserve(additional) |
876 | } | |
877 | ||
9cc50fc6 SL |
878 | /// Ensures that this `String`'s capacity is `additional` bytes |
879 | /// larger than its length. | |
1a4d82fc | 880 | /// |
cc61c64b | 881 | /// Consider using the [`reserve`] method unless you absolutely know |
9cc50fc6 SL |
882 | /// better than the allocator. |
883 | /// | |
cc61c64b | 884 | /// [`reserve`]: #method.reserve |
1a4d82fc JJ |
885 | /// |
886 | /// # Panics | |
887 | /// | |
85aaf69f | 888 | /// Panics if the new capacity overflows `usize`. |
1a4d82fc JJ |
889 | /// |
890 | /// # Examples | |
891 | /// | |
9cc50fc6 SL |
892 | /// Basic usage: |
893 | /// | |
1a4d82fc JJ |
894 | /// ``` |
895 | /// let mut s = String::new(); | |
9cc50fc6 | 896 | /// |
62682a34 | 897 | /// s.reserve_exact(10); |
9cc50fc6 | 898 | /// |
1a4d82fc JJ |
899 | /// assert!(s.capacity() >= 10); |
900 | /// ``` | |
9cc50fc6 SL |
901 | /// |
902 | /// This may not actually increase the capacity: | |
903 | /// | |
904 | /// ``` | |
905 | /// let mut s = String::with_capacity(10); | |
906 | /// s.push('a'); | |
907 | /// s.push('b'); | |
908 | /// | |
909 | /// // s now has a length of 2 and a capacity of 10 | |
910 | /// assert_eq!(2, s.len()); | |
911 | /// assert_eq!(10, s.capacity()); | |
912 | /// | |
913 | /// // Since we already have an extra 8 capacity, calling this... | |
914 | /// s.reserve_exact(8); | |
915 | /// | |
916 | /// // ... doesn't actually increase. | |
917 | /// assert_eq!(10, s.capacity()); | |
918 | /// ``` | |
1a4d82fc | 919 | #[inline] |
85aaf69f SL |
920 | #[stable(feature = "rust1", since = "1.0.0")] |
921 | pub fn reserve_exact(&mut self, additional: usize) { | |
1a4d82fc JJ |
922 | self.vec.reserve_exact(additional) |
923 | } | |
924 | ||
0531ce1d XL |
925 | /// Tries to reserve capacity for at least `additional` more elements to be inserted |
926 | /// in the given `String`. The collection may reserve more space to avoid | |
927 | /// frequent reallocations. After calling `reserve`, capacity will be | |
928 | /// greater than or equal to `self.len() + additional`. Does nothing if | |
929 | /// capacity is already sufficient. | |
930 | /// | |
931 | /// # Errors | |
932 | /// | |
933 | /// If the capacity overflows, or the allocator reports a failure, then an error | |
934 | /// is returned. | |
935 | /// | |
936 | /// # Examples | |
937 | /// | |
938 | /// ``` | |
939 | /// #![feature(try_reserve)] | |
940 | /// use std::collections::CollectionAllocErr; | |
941 | /// | |
942 | /// fn process_data(data: &str) -> Result<String, CollectionAllocErr> { | |
943 | /// let mut output = String::new(); | |
944 | /// | |
945 | /// // Pre-reserve the memory, exiting if we can't | |
946 | /// output.try_reserve(data.len())?; | |
947 | /// | |
948 | /// // Now we know this can't OOM in the middle of our complex work | |
949 | /// output.push_str(data); | |
950 | /// | |
951 | /// Ok(output) | |
952 | /// } | |
953 | /// # process_data("rust").expect("why is the test harness OOMing on 4 bytes?"); | |
954 | /// ``` | |
955 | #[unstable(feature = "try_reserve", reason = "new API", issue="48043")] | |
956 | pub fn try_reserve(&mut self, additional: usize) -> Result<(), CollectionAllocErr> { | |
957 | self.vec.try_reserve(additional) | |
958 | } | |
959 | ||
960 | /// Tries to reserves the minimum capacity for exactly `additional` more elements to | |
961 | /// be inserted in the given `String`. After calling `reserve_exact`, | |
962 | /// capacity will be greater than or equal to `self.len() + additional`. | |
963 | /// Does nothing if the capacity is already sufficient. | |
964 | /// | |
965 | /// Note that the allocator may give the collection more space than it | |
9fa01778 | 966 | /// requests. Therefore, capacity can not be relied upon to be precisely |
0531ce1d XL |
967 | /// minimal. Prefer `reserve` if future insertions are expected. |
968 | /// | |
969 | /// # Errors | |
970 | /// | |
971 | /// If the capacity overflows, or the allocator reports a failure, then an error | |
972 | /// is returned. | |
973 | /// | |
974 | /// # Examples | |
975 | /// | |
976 | /// ``` | |
977 | /// #![feature(try_reserve)] | |
978 | /// use std::collections::CollectionAllocErr; | |
979 | /// | |
980 | /// fn process_data(data: &str) -> Result<String, CollectionAllocErr> { | |
981 | /// let mut output = String::new(); | |
982 | /// | |
983 | /// // Pre-reserve the memory, exiting if we can't | |
984 | /// output.try_reserve(data.len())?; | |
985 | /// | |
986 | /// // Now we know this can't OOM in the middle of our complex work | |
987 | /// output.push_str(data); | |
988 | /// | |
989 | /// Ok(output) | |
990 | /// } | |
991 | /// # process_data("rust").expect("why is the test harness OOMing on 4 bytes?"); | |
992 | /// ``` | |
993 | #[unstable(feature = "try_reserve", reason = "new API", issue="48043")] | |
994 | pub fn try_reserve_exact(&mut self, additional: usize) -> Result<(), CollectionAllocErr> { | |
995 | self.vec.try_reserve_exact(additional) | |
996 | } | |
997 | ||
9cc50fc6 | 998 | /// Shrinks the capacity of this `String` to match its length. |
1a4d82fc JJ |
999 | /// |
1000 | /// # Examples | |
1001 | /// | |
9cc50fc6 SL |
1002 | /// Basic usage: |
1003 | /// | |
1a4d82fc | 1004 | /// ``` |
62682a34 | 1005 | /// let mut s = String::from("foo"); |
9cc50fc6 | 1006 | /// |
1a4d82fc JJ |
1007 | /// s.reserve(100); |
1008 | /// assert!(s.capacity() >= 100); | |
9cc50fc6 | 1009 | /// |
1a4d82fc | 1010 | /// s.shrink_to_fit(); |
9cc50fc6 | 1011 | /// assert_eq!(3, s.capacity()); |
1a4d82fc JJ |
1012 | /// ``` |
1013 | #[inline] | |
85aaf69f | 1014 | #[stable(feature = "rust1", since = "1.0.0")] |
1a4d82fc JJ |
1015 | pub fn shrink_to_fit(&mut self) { |
1016 | self.vec.shrink_to_fit() | |
1017 | } | |
1018 | ||
0531ce1d XL |
1019 | /// Shrinks the capacity of this `String` with a lower bound. |
1020 | /// | |
1021 | /// The capacity will remain at least as large as both the length | |
1022 | /// and the supplied value. | |
1023 | /// | |
1024 | /// Panics if the current capacity is smaller than the supplied | |
1025 | /// minimum capacity. | |
1026 | /// | |
1027 | /// # Examples | |
1028 | /// | |
1029 | /// ``` | |
1030 | /// #![feature(shrink_to)] | |
1031 | /// let mut s = String::from("foo"); | |
1032 | /// | |
1033 | /// s.reserve(100); | |
1034 | /// assert!(s.capacity() >= 100); | |
1035 | /// | |
1036 | /// s.shrink_to(10); | |
1037 | /// assert!(s.capacity() >= 10); | |
1038 | /// s.shrink_to(0); | |
1039 | /// assert!(s.capacity() >= 3); | |
1040 | /// ``` | |
1041 | #[inline] | |
a1dfa0c6 | 1042 | #[unstable(feature = "shrink_to", reason = "new API", issue="56431")] |
0531ce1d XL |
1043 | pub fn shrink_to(&mut self, min_capacity: usize) { |
1044 | self.vec.shrink_to(min_capacity) | |
1045 | } | |
1046 | ||
3b2f2976 XL |
1047 | /// Appends the given [`char`] to the end of this `String`. |
1048 | /// | |
1049 | /// [`char`]: ../../std/primitive.char.html | |
1a4d82fc JJ |
1050 | /// |
1051 | /// # Examples | |
1052 | /// | |
9cc50fc6 SL |
1053 | /// Basic usage: |
1054 | /// | |
1a4d82fc | 1055 | /// ``` |
62682a34 | 1056 | /// let mut s = String::from("abc"); |
9cc50fc6 | 1057 | /// |
1a4d82fc JJ |
1058 | /// s.push('1'); |
1059 | /// s.push('2'); | |
1060 | /// s.push('3'); | |
9cc50fc6 SL |
1061 | /// |
1062 | /// assert_eq!("abc123", s); | |
1a4d82fc JJ |
1063 | /// ``` |
1064 | #[inline] | |
85aaf69f | 1065 | #[stable(feature = "rust1", since = "1.0.0")] |
1a4d82fc | 1066 | pub fn push(&mut self, ch: char) { |
62682a34 SL |
1067 | match ch.len_utf8() { |
1068 | 1 => self.vec.push(ch as u8), | |
32a655c1 | 1069 | _ => self.vec.extend_from_slice(ch.encode_utf8(&mut [0; 4]).as_bytes()), |
1a4d82fc JJ |
1070 | } |
1071 | } | |
1072 | ||
9cc50fc6 | 1073 | /// Returns a byte slice of this `String`'s contents. |
1a4d82fc | 1074 | /// |
8bb4bdeb XL |
1075 | /// The inverse of this method is [`from_utf8`]. |
1076 | /// | |
1077 | /// [`from_utf8`]: #method.from_utf8 | |
1078 | /// | |
1a4d82fc JJ |
1079 | /// # Examples |
1080 | /// | |
9cc50fc6 SL |
1081 | /// Basic usage: |
1082 | /// | |
1a4d82fc | 1083 | /// ``` |
62682a34 | 1084 | /// let s = String::from("hello"); |
9cc50fc6 SL |
1085 | /// |
1086 | /// assert_eq!(&[104, 101, 108, 108, 111], s.as_bytes()); | |
1a4d82fc JJ |
1087 | /// ``` |
1088 | #[inline] | |
85aaf69f SL |
1089 | #[stable(feature = "rust1", since = "1.0.0")] |
1090 | pub fn as_bytes(&self) -> &[u8] { | |
1091 | &self.vec | |
1a4d82fc JJ |
1092 | } |
1093 | ||
9cc50fc6 | 1094 | /// Shortens this `String` to the specified length. |
1a4d82fc | 1095 | /// |
a7813a04 XL |
1096 | /// If `new_len` is greater than the string's current length, this has no |
1097 | /// effect. | |
1098 | /// | |
8bb4bdeb XL |
1099 | /// Note that this method has no effect on the allocated capacity |
1100 | /// of the string | |
1101 | /// | |
1a4d82fc JJ |
1102 | /// # Panics |
1103 | /// | |
a7813a04 | 1104 | /// Panics if `new_len` does not lie on a [`char`] boundary. |
9cc50fc6 | 1105 | /// |
54a0048b | 1106 | /// [`char`]: ../../std/primitive.char.html |
1a4d82fc JJ |
1107 | /// |
1108 | /// # Examples | |
1109 | /// | |
9cc50fc6 SL |
1110 | /// Basic usage: |
1111 | /// | |
1a4d82fc | 1112 | /// ``` |
62682a34 | 1113 | /// let mut s = String::from("hello"); |
9cc50fc6 | 1114 | /// |
1a4d82fc | 1115 | /// s.truncate(2); |
9cc50fc6 SL |
1116 | /// |
1117 | /// assert_eq!("he", s); | |
1a4d82fc JJ |
1118 | /// ``` |
1119 | #[inline] | |
85aaf69f SL |
1120 | #[stable(feature = "rust1", since = "1.0.0")] |
1121 | pub fn truncate(&mut self, new_len: usize) { | |
a7813a04 XL |
1122 | if new_len <= self.len() { |
1123 | assert!(self.is_char_boundary(new_len)); | |
1124 | self.vec.truncate(new_len) | |
1125 | } | |
1a4d82fc JJ |
1126 | } |
1127 | ||
1128 | /// Removes the last character from the string buffer and returns it. | |
9cc50fc6 | 1129 | /// |
3b2f2976 XL |
1130 | /// Returns [`None`] if this `String` is empty. |
1131 | /// | |
1132 | /// [`None`]: ../../std/option/enum.Option.html#variant.None | |
1a4d82fc JJ |
1133 | /// |
1134 | /// # Examples | |
1135 | /// | |
9cc50fc6 SL |
1136 | /// Basic usage: |
1137 | /// | |
1a4d82fc | 1138 | /// ``` |
62682a34 | 1139 | /// let mut s = String::from("foo"); |
9cc50fc6 | 1140 | /// |
1a4d82fc JJ |
1141 | /// assert_eq!(s.pop(), Some('o')); |
1142 | /// assert_eq!(s.pop(), Some('o')); | |
1143 | /// assert_eq!(s.pop(), Some('f')); | |
9cc50fc6 | 1144 | /// |
1a4d82fc JJ |
1145 | /// assert_eq!(s.pop(), None); |
1146 | /// ``` | |
1147 | #[inline] | |
85aaf69f | 1148 | #[stable(feature = "rust1", since = "1.0.0")] |
1a4d82fc | 1149 | pub fn pop(&mut self) -> Option<char> { |
ff7c6d11 | 1150 | let ch = self.chars().rev().next()?; |
54a0048b | 1151 | let newlen = self.len() - ch.len_utf8(); |
1a4d82fc | 1152 | unsafe { |
54a0048b | 1153 | self.vec.set_len(newlen); |
1a4d82fc JJ |
1154 | } |
1155 | Some(ch) | |
1156 | } | |
1157 | ||
3b2f2976 | 1158 | /// Removes a [`char`] from this `String` at a byte position and returns it. |
1a4d82fc | 1159 | /// |
9cc50fc6 | 1160 | /// This is an `O(n)` operation, as it requires copying every element in the |
1a4d82fc JJ |
1161 | /// buffer. |
1162 | /// | |
1163 | /// # Panics | |
1164 | /// | |
9cc50fc6 SL |
1165 | /// Panics if `idx` is larger than or equal to the `String`'s length, |
1166 | /// or if it does not lie on a [`char`] boundary. | |
1167 | /// | |
54a0048b | 1168 | /// [`char`]: ../../std/primitive.char.html |
1a4d82fc JJ |
1169 | /// |
1170 | /// # Examples | |
1171 | /// | |
9cc50fc6 SL |
1172 | /// Basic usage: |
1173 | /// | |
1a4d82fc | 1174 | /// ``` |
62682a34 | 1175 | /// let mut s = String::from("foo"); |
9cc50fc6 | 1176 | /// |
1a4d82fc JJ |
1177 | /// assert_eq!(s.remove(0), 'f'); |
1178 | /// assert_eq!(s.remove(1), 'o'); | |
1179 | /// assert_eq!(s.remove(0), 'o'); | |
1180 | /// ``` | |
85aaf69f SL |
1181 | #[inline] |
1182 | #[stable(feature = "rust1", since = "1.0.0")] | |
1183 | pub fn remove(&mut self, idx: usize) -> char { | |
54a0048b SL |
1184 | let ch = match self[idx..].chars().next() { |
1185 | Some(ch) => ch, | |
1186 | None => panic!("cannot remove a char from the end of a string"), | |
1187 | }; | |
1a4d82fc | 1188 | |
c34b1796 | 1189 | let next = idx + ch.len_utf8(); |
54a0048b | 1190 | let len = self.len(); |
1a4d82fc | 1191 | unsafe { |
b7449926 XL |
1192 | ptr::copy(self.vec.as_ptr().add(next), |
1193 | self.vec.as_mut_ptr().add(idx), | |
c34b1796 | 1194 | len - next); |
1a4d82fc JJ |
1195 | self.vec.set_len(len - (next - idx)); |
1196 | } | |
1197 | ch | |
1198 | } | |
1199 | ||
3b2f2976 XL |
1200 | /// Retains only the characters specified by the predicate. |
1201 | /// | |
1202 | /// In other words, remove all characters `c` such that `f(c)` returns `false`. | |
48663c56 XL |
1203 | /// This method operates in place, visiting each character exactly once in the |
1204 | /// original order, and preserves the order of the retained characters. | |
3b2f2976 XL |
1205 | /// |
1206 | /// # Examples | |
1207 | /// | |
1208 | /// ``` | |
3b2f2976 XL |
1209 | /// let mut s = String::from("f_o_ob_ar"); |
1210 | /// | |
1211 | /// s.retain(|c| c != '_'); | |
1212 | /// | |
1213 | /// assert_eq!(s, "foobar"); | |
1214 | /// ``` | |
48663c56 XL |
1215 | /// |
1216 | /// The exact order may be useful for tracking external state, like an index. | |
1217 | /// | |
1218 | /// ``` | |
1219 | /// let mut s = String::from("abcde"); | |
1220 | /// let keep = [false, true, true, false, true]; | |
1221 | /// let mut i = 0; | |
1222 | /// s.retain(|_| (keep[i], i += 1).0); | |
1223 | /// assert_eq!(s, "bce"); | |
1224 | /// ``` | |
3b2f2976 | 1225 | #[inline] |
0531ce1d | 1226 | #[stable(feature = "string_retain", since = "1.26.0")] |
3b2f2976 XL |
1227 | pub fn retain<F>(&mut self, mut f: F) |
1228 | where F: FnMut(char) -> bool | |
1229 | { | |
1230 | let len = self.len(); | |
1231 | let mut del_bytes = 0; | |
1232 | let mut idx = 0; | |
1233 | ||
1234 | while idx < len { | |
1235 | let ch = unsafe { | |
8faf50e0 | 1236 | self.get_unchecked(idx..len).chars().next().unwrap() |
3b2f2976 XL |
1237 | }; |
1238 | let ch_len = ch.len_utf8(); | |
1239 | ||
1240 | if !f(ch) { | |
1241 | del_bytes += ch_len; | |
1242 | } else if del_bytes > 0 { | |
1243 | unsafe { | |
b7449926 XL |
1244 | ptr::copy(self.vec.as_ptr().add(idx), |
1245 | self.vec.as_mut_ptr().add(idx - del_bytes), | |
3b2f2976 XL |
1246 | ch_len); |
1247 | } | |
1248 | } | |
1249 | ||
1250 | // Point idx to the next char | |
1251 | idx += ch_len; | |
1252 | } | |
1253 | ||
1254 | if del_bytes > 0 { | |
1255 | unsafe { self.vec.set_len(len - del_bytes); } | |
1256 | } | |
1257 | } | |
1258 | ||
9cc50fc6 | 1259 | /// Inserts a character into this `String` at a byte position. |
1a4d82fc | 1260 | /// |
9cc50fc6 | 1261 | /// This is an `O(n)` operation as it requires copying every element in the |
1a4d82fc JJ |
1262 | /// buffer. |
1263 | /// | |
1264 | /// # Panics | |
1265 | /// | |
9cc50fc6 SL |
1266 | /// Panics if `idx` is larger than the `String`'s length, or if it does not |
1267 | /// lie on a [`char`] boundary. | |
1268 | /// | |
54a0048b | 1269 | /// [`char`]: ../../std/primitive.char.html |
9cc50fc6 SL |
1270 | /// |
1271 | /// # Examples | |
1272 | /// | |
1273 | /// Basic usage: | |
1274 | /// | |
1275 | /// ``` | |
1276 | /// let mut s = String::with_capacity(3); | |
1277 | /// | |
1278 | /// s.insert(0, 'f'); | |
1279 | /// s.insert(1, 'o'); | |
1280 | /// s.insert(2, 'o'); | |
1281 | /// | |
1282 | /// assert_eq!("foo", s); | |
1283 | /// ``` | |
85aaf69f SL |
1284 | #[inline] |
1285 | #[stable(feature = "rust1", since = "1.0.0")] | |
1286 | pub fn insert(&mut self, idx: usize, ch: char) { | |
1a4d82fc | 1287 | assert!(self.is_char_boundary(idx)); |
c30ab7b3 SL |
1288 | let mut bits = [0; 4]; |
1289 | let bits = ch.encode_utf8(&mut bits).as_bytes(); | |
5bcae85e SL |
1290 | |
1291 | unsafe { | |
c30ab7b3 | 1292 | self.insert_bytes(idx, bits); |
5bcae85e SL |
1293 | } |
1294 | } | |
1295 | ||
1296 | unsafe fn insert_bytes(&mut self, idx: usize, bytes: &[u8]) { | |
1297 | let len = self.len(); | |
1298 | let amt = bytes.len(); | |
54a0048b | 1299 | self.vec.reserve(amt); |
1a4d82fc | 1300 | |
b7449926 XL |
1301 | ptr::copy(self.vec.as_ptr().add(idx), |
1302 | self.vec.as_mut_ptr().add(idx + amt), | |
5bcae85e SL |
1303 | len - idx); |
1304 | ptr::copy(bytes.as_ptr(), | |
b7449926 | 1305 | self.vec.as_mut_ptr().add(idx), |
5bcae85e SL |
1306 | amt); |
1307 | self.vec.set_len(len + amt); | |
1308 | } | |
1309 | ||
1310 | /// Inserts a string slice into this `String` at a byte position. | |
1311 | /// | |
1312 | /// This is an `O(n)` operation as it requires copying every element in the | |
1313 | /// buffer. | |
1314 | /// | |
1315 | /// # Panics | |
1316 | /// | |
1317 | /// Panics if `idx` is larger than the `String`'s length, or if it does not | |
1318 | /// lie on a [`char`] boundary. | |
1319 | /// | |
1320 | /// [`char`]: ../../std/primitive.char.html | |
1321 | /// | |
1322 | /// # Examples | |
1323 | /// | |
1324 | /// Basic usage: | |
1325 | /// | |
1326 | /// ``` | |
5bcae85e SL |
1327 | /// let mut s = String::from("bar"); |
1328 | /// | |
1329 | /// s.insert_str(0, "foo"); | |
1330 | /// | |
1331 | /// assert_eq!("foobar", s); | |
1332 | /// ``` | |
1333 | #[inline] | |
32a655c1 | 1334 | #[stable(feature = "insert_str", since = "1.16.0")] |
5bcae85e | 1335 | pub fn insert_str(&mut self, idx: usize, string: &str) { |
5bcae85e SL |
1336 | assert!(self.is_char_boundary(idx)); |
1337 | ||
1a4d82fc | 1338 | unsafe { |
5bcae85e | 1339 | self.insert_bytes(idx, string.as_bytes()); |
1a4d82fc JJ |
1340 | } |
1341 | } | |
1342 | ||
9cc50fc6 | 1343 | /// Returns a mutable reference to the contents of this `String`. |
1a4d82fc | 1344 | /// |
9cc50fc6 SL |
1345 | /// # Safety |
1346 | /// | |
1347 | /// This function is unsafe because it does not check that the bytes passed | |
1348 | /// to it are valid UTF-8. If this constraint is violated, it may cause | |
1349 | /// memory unsafety issues with future users of the `String`, as the rest of | |
1350 | /// the standard library assumes that `String`s are valid UTF-8. | |
1a4d82fc JJ |
1351 | /// |
1352 | /// # Examples | |
1353 | /// | |
9cc50fc6 SL |
1354 | /// Basic usage: |
1355 | /// | |
1a4d82fc | 1356 | /// ``` |
62682a34 | 1357 | /// let mut s = String::from("hello"); |
9cc50fc6 | 1358 | /// |
1a4d82fc JJ |
1359 | /// unsafe { |
1360 | /// let vec = s.as_mut_vec(); | |
9cc50fc6 SL |
1361 | /// assert_eq!(&[104, 101, 108, 108, 111][..], &vec[..]); |
1362 | /// | |
1a4d82fc JJ |
1363 | /// vec.reverse(); |
1364 | /// } | |
c34b1796 | 1365 | /// assert_eq!(s, "olleh"); |
1a4d82fc | 1366 | /// ``` |
85aaf69f SL |
1367 | #[inline] |
1368 | #[stable(feature = "rust1", since = "1.0.0")] | |
1369 | pub unsafe fn as_mut_vec(&mut self) -> &mut Vec<u8> { | |
1a4d82fc JJ |
1370 | &mut self.vec |
1371 | } | |
1372 | ||
9cc50fc6 | 1373 | /// Returns the length of this `String`, in bytes. |
1a4d82fc JJ |
1374 | /// |
1375 | /// # Examples | |
1376 | /// | |
9cc50fc6 SL |
1377 | /// Basic usage: |
1378 | /// | |
1a4d82fc | 1379 | /// ``` |
9cc50fc6 SL |
1380 | /// let a = String::from("foo"); |
1381 | /// | |
1a4d82fc JJ |
1382 | /// assert_eq!(a.len(), 3); |
1383 | /// ``` | |
1384 | #[inline] | |
85aaf69f | 1385 | #[stable(feature = "rust1", since = "1.0.0")] |
92a42be0 SL |
1386 | pub fn len(&self) -> usize { |
1387 | self.vec.len() | |
1388 | } | |
1a4d82fc | 1389 | |
9fa01778 | 1390 | /// Returns `true` if this `String` has a length of zero, and `false` otherwise. |
1a4d82fc JJ |
1391 | /// |
1392 | /// # Examples | |
1393 | /// | |
9cc50fc6 SL |
1394 | /// Basic usage: |
1395 | /// | |
1a4d82fc JJ |
1396 | /// ``` |
1397 | /// let mut v = String::new(); | |
1398 | /// assert!(v.is_empty()); | |
9cc50fc6 | 1399 | /// |
1a4d82fc JJ |
1400 | /// v.push('a'); |
1401 | /// assert!(!v.is_empty()); | |
1402 | /// ``` | |
85aaf69f SL |
1403 | #[inline] |
1404 | #[stable(feature = "rust1", since = "1.0.0")] | |
92a42be0 SL |
1405 | pub fn is_empty(&self) -> bool { |
1406 | self.len() == 0 | |
1407 | } | |
1a4d82fc | 1408 | |
8bb4bdeb | 1409 | /// Splits the string into two at the given index. |
476ff2be | 1410 | /// |
8bb4bdeb XL |
1411 | /// Returns a newly allocated `String`. `self` contains bytes `[0, at)`, and |
1412 | /// the returned `String` contains bytes `[at, len)`. `at` must be on the | |
1413 | /// boundary of a UTF-8 code point. | |
476ff2be | 1414 | /// |
8bb4bdeb | 1415 | /// Note that the capacity of `self` does not change. |
476ff2be SL |
1416 | /// |
1417 | /// # Panics | |
1418 | /// | |
8bb4bdeb | 1419 | /// Panics if `at` is not on a `UTF-8` code point boundary, or if it is beyond the last |
476ff2be SL |
1420 | /// code point of the string. |
1421 | /// | |
1422 | /// # Examples | |
1423 | /// | |
1424 | /// ``` | |
476ff2be SL |
1425 | /// # fn main() { |
1426 | /// let mut hello = String::from("Hello, World!"); | |
1427 | /// let world = hello.split_off(7); | |
1428 | /// assert_eq!(hello, "Hello, "); | |
1429 | /// assert_eq!(world, "World!"); | |
1430 | /// # } | |
1431 | /// ``` | |
1432 | #[inline] | |
32a655c1 | 1433 | #[stable(feature = "string_split_off", since = "1.16.0")] |
8bb4bdeb XL |
1434 | pub fn split_off(&mut self, at: usize) -> String { |
1435 | assert!(self.is_char_boundary(at)); | |
1436 | let other = self.vec.split_off(at); | |
476ff2be SL |
1437 | unsafe { String::from_utf8_unchecked(other) } |
1438 | } | |
1439 | ||
9cc50fc6 SL |
1440 | /// Truncates this `String`, removing all contents. |
1441 | /// | |
1442 | /// While this means the `String` will have a length of zero, it does not | |
1443 | /// touch its capacity. | |
1a4d82fc JJ |
1444 | /// |
1445 | /// # Examples | |
1446 | /// | |
9cc50fc6 SL |
1447 | /// Basic usage: |
1448 | /// | |
1a4d82fc | 1449 | /// ``` |
9cc50fc6 SL |
1450 | /// let mut s = String::from("foo"); |
1451 | /// | |
1a4d82fc | 1452 | /// s.clear(); |
9cc50fc6 | 1453 | /// |
1a4d82fc | 1454 | /// assert!(s.is_empty()); |
9cc50fc6 SL |
1455 | /// assert_eq!(0, s.len()); |
1456 | /// assert_eq!(3, s.capacity()); | |
1a4d82fc JJ |
1457 | /// ``` |
1458 | #[inline] | |
85aaf69f | 1459 | #[stable(feature = "rust1", since = "1.0.0")] |
1a4d82fc JJ |
1460 | pub fn clear(&mut self) { |
1461 | self.vec.clear() | |
1462 | } | |
d9579d0f | 1463 | |
b7449926 XL |
1464 | /// Creates a draining iterator that removes the specified range in the `String` |
1465 | /// and yields the removed `chars`. | |
9cc50fc6 SL |
1466 | /// |
1467 | /// Note: The element range is removed even if the iterator is not | |
1468 | /// consumed until the end. | |
d9579d0f AL |
1469 | /// |
1470 | /// # Panics | |
1471 | /// | |
9cc50fc6 SL |
1472 | /// Panics if the starting point or end point do not lie on a [`char`] |
1473 | /// boundary, or if they're out of bounds. | |
1474 | /// | |
54a0048b | 1475 | /// [`char`]: ../../std/primitive.char.html |
d9579d0f AL |
1476 | /// |
1477 | /// # Examples | |
1478 | /// | |
9cc50fc6 SL |
1479 | /// Basic usage: |
1480 | /// | |
d9579d0f | 1481 | /// ``` |
d9579d0f AL |
1482 | /// let mut s = String::from("α is alpha, β is beta"); |
1483 | /// let beta_offset = s.find('β').unwrap_or(s.len()); | |
1484 | /// | |
1485 | /// // Remove the range up until the β from the string | |
1486 | /// let t: String = s.drain(..beta_offset).collect(); | |
1487 | /// assert_eq!(t, "α is alpha, "); | |
1488 | /// assert_eq!(s, "β is beta"); | |
1489 | /// | |
1490 | /// // A full range clears the string | |
1491 | /// s.drain(..); | |
1492 | /// assert_eq!(s, ""); | |
1493 | /// ``` | |
92a42be0 | 1494 | #[stable(feature = "drain", since = "1.6.0")] |
9fa01778 | 1495 | pub fn drain<R>(&mut self, range: R) -> Drain<'_> |
0531ce1d | 1496 | where R: RangeBounds<usize> |
92a42be0 | 1497 | { |
d9579d0f AL |
1498 | // Memory safety |
1499 | // | |
1500 | // The String version of Drain does not have the memory safety issues | |
1501 | // of the vector version. The data is just plain bytes. | |
1502 | // Because the range removal happens in Drop, if the Drain iterator is leaked, | |
1503 | // the removal will not happen. | |
1504 | let len = self.len(); | |
94b46f34 | 1505 | let start = match range.start_bound() { |
32a655c1 SL |
1506 | Included(&n) => n, |
1507 | Excluded(&n) => n + 1, | |
1508 | Unbounded => 0, | |
1509 | }; | |
94b46f34 | 1510 | let end = match range.end_bound() { |
32a655c1 SL |
1511 | Included(&n) => n + 1, |
1512 | Excluded(&n) => n, | |
1513 | Unbounded => len, | |
1514 | }; | |
d9579d0f AL |
1515 | |
1516 | // Take out two simultaneous borrows. The &mut String won't be accessed | |
1517 | // until iteration is over, in Drop. | |
1518 | let self_ptr = self as *mut _; | |
1519 | // slicing does the appropriate bounds checks | |
1520 | let chars_iter = self[start..end].chars(); | |
1521 | ||
1522 | Drain { | |
3b2f2976 XL |
1523 | start, |
1524 | end, | |
d9579d0f AL |
1525 | iter: chars_iter, |
1526 | string: self_ptr, | |
1527 | } | |
1528 | } | |
c1a9b12d | 1529 | |
83c7162d | 1530 | /// Removes the specified range in the string, |
ea8adc8c XL |
1531 | /// and replaces it with the given string. |
1532 | /// The given string doesn't need to be the same length as the range. | |
7cac9316 | 1533 | /// |
7cac9316 XL |
1534 | /// # Panics |
1535 | /// | |
1536 | /// Panics if the starting point or end point do not lie on a [`char`] | |
1537 | /// boundary, or if they're out of bounds. | |
1538 | /// | |
1539 | /// [`char`]: ../../std/primitive.char.html | |
ea8adc8c | 1540 | /// [`Vec::splice`]: ../../std/vec/struct.Vec.html#method.splice |
7cac9316 XL |
1541 | /// |
1542 | /// # Examples | |
1543 | /// | |
1544 | /// Basic usage: | |
1545 | /// | |
1546 | /// ``` | |
7cac9316 XL |
1547 | /// let mut s = String::from("α is alpha, β is beta"); |
1548 | /// let beta_offset = s.find('β').unwrap_or(s.len()); | |
1549 | /// | |
1550 | /// // Replace the range up until the β from the string | |
83c7162d | 1551 | /// s.replace_range(..beta_offset, "Α is capital alpha; "); |
7cac9316 XL |
1552 | /// assert_eq!(s, "Α is capital alpha; β is beta"); |
1553 | /// ``` | |
83c7162d XL |
1554 | #[stable(feature = "splice", since = "1.27.0")] |
1555 | pub fn replace_range<R>(&mut self, range: R, replace_with: &str) | |
0531ce1d | 1556 | where R: RangeBounds<usize> |
7cac9316 XL |
1557 | { |
1558 | // Memory safety | |
1559 | // | |
83c7162d | 1560 | // Replace_range does not have the memory safety issues of a vector Splice. |
7cac9316 | 1561 | // of the vector version. The data is just plain bytes. |
ea8adc8c | 1562 | |
94b46f34 | 1563 | match range.start_bound() { |
ea8adc8c XL |
1564 | Included(&n) => assert!(self.is_char_boundary(n)), |
1565 | Excluded(&n) => assert!(self.is_char_boundary(n + 1)), | |
1566 | Unbounded => {}, | |
7cac9316 | 1567 | }; |
94b46f34 | 1568 | match range.end_bound() { |
ea8adc8c XL |
1569 | Included(&n) => assert!(self.is_char_boundary(n + 1)), |
1570 | Excluded(&n) => assert!(self.is_char_boundary(n)), | |
1571 | Unbounded => {}, | |
7cac9316 XL |
1572 | }; |
1573 | ||
ea8adc8c XL |
1574 | unsafe { |
1575 | self.as_mut_vec() | |
1576 | }.splice(range, replace_with.bytes()); | |
7cac9316 XL |
1577 | } |
1578 | ||
3b2f2976 | 1579 | /// Converts this `String` into a [`Box`]`<`[`str`]`>`. |
9cc50fc6 SL |
1580 | /// |
1581 | /// This will drop any excess capacity. | |
1582 | /// | |
3b2f2976 XL |
1583 | /// [`Box`]: ../../std/boxed/struct.Box.html |
1584 | /// [`str`]: ../../std/primitive.str.html | |
1585 | /// | |
9cc50fc6 SL |
1586 | /// # Examples |
1587 | /// | |
1588 | /// Basic usage: | |
1589 | /// | |
1590 | /// ``` | |
1591 | /// let s = String::from("hello"); | |
c1a9b12d | 1592 | /// |
9cc50fc6 SL |
1593 | /// let b = s.into_boxed_str(); |
1594 | /// ``` | |
e9174d1e | 1595 | #[stable(feature = "box_str", since = "1.4.0")] |
83c7162d | 1596 | #[inline] |
e9174d1e | 1597 | pub fn into_boxed_str(self) -> Box<str> { |
c1a9b12d | 1598 | let slice = self.vec.into_boxed_slice(); |
041b39d2 | 1599 | unsafe { from_boxed_utf8_unchecked(slice) } |
c1a9b12d | 1600 | } |
1a4d82fc JJ |
1601 | } |
1602 | ||
1603 | impl FromUtf8Error { | |
cc61c64b XL |
1604 | /// Returns a slice of [`u8`]s bytes that were attempted to convert to a `String`. |
1605 | /// | |
1606 | /// # Examples | |
1607 | /// | |
1608 | /// Basic usage: | |
1609 | /// | |
1610 | /// ``` | |
cc61c64b XL |
1611 | /// // some invalid bytes, in a vector |
1612 | /// let bytes = vec![0, 159]; | |
1613 | /// | |
1614 | /// let value = String::from_utf8(bytes); | |
1615 | /// | |
1616 | /// assert_eq!(&[0, 159], value.unwrap_err().as_bytes()); | |
1617 | /// ``` | |
0531ce1d | 1618 | #[stable(feature = "from_utf8_error_as_bytes", since = "1.26.0")] |
cc61c64b XL |
1619 | pub fn as_bytes(&self) -> &[u8] { |
1620 | &self.bytes[..] | |
1621 | } | |
1622 | ||
92a42be0 SL |
1623 | /// Returns the bytes that were attempted to convert to a `String`. |
1624 | /// | |
1625 | /// This method is carefully constructed to avoid allocation. It will | |
1626 | /// consume the error, moving out the bytes, so that a copy of the bytes | |
1627 | /// does not need to be made. | |
1628 | /// | |
1629 | /// # Examples | |
1630 | /// | |
1631 | /// Basic usage: | |
1632 | /// | |
1633 | /// ``` | |
1634 | /// // some invalid bytes, in a vector | |
1635 | /// let bytes = vec![0, 159]; | |
1636 | /// | |
1637 | /// let value = String::from_utf8(bytes); | |
1638 | /// | |
1639 | /// assert_eq!(vec![0, 159], value.unwrap_err().into_bytes()); | |
1640 | /// ``` | |
85aaf69f | 1641 | #[stable(feature = "rust1", since = "1.0.0")] |
92a42be0 SL |
1642 | pub fn into_bytes(self) -> Vec<u8> { |
1643 | self.bytes | |
1644 | } | |
1a4d82fc | 1645 | |
92a42be0 SL |
1646 | /// Fetch a `Utf8Error` to get more details about the conversion failure. |
1647 | /// | |
1648 | /// The [`Utf8Error`] type provided by [`std::str`] represents an error that may | |
1649 | /// occur when converting a slice of [`u8`]s to a [`&str`]. In this sense, it's | |
1650 | /// an analogue to `FromUtf8Error`. See its documentation for more details | |
1651 | /// on using it. | |
1652 | /// | |
54a0048b SL |
1653 | /// [`Utf8Error`]: ../../std/str/struct.Utf8Error.html |
1654 | /// [`std::str`]: ../../std/str/index.html | |
1655 | /// [`u8`]: ../../std/primitive.u8.html | |
1656 | /// [`&str`]: ../../std/primitive.str.html | |
92a42be0 SL |
1657 | /// |
1658 | /// # Examples | |
1659 | /// | |
1660 | /// Basic usage: | |
1661 | /// | |
1662 | /// ``` | |
1663 | /// // some invalid bytes, in a vector | |
1664 | /// let bytes = vec![0, 159]; | |
1665 | /// | |
1666 | /// let error = String::from_utf8(bytes).unwrap_err().utf8_error(); | |
1667 | /// | |
1668 | /// // the first byte is invalid here | |
1669 | /// assert_eq!(1, error.valid_up_to()); | |
1670 | /// ``` | |
85aaf69f | 1671 | #[stable(feature = "rust1", since = "1.0.0")] |
92a42be0 SL |
1672 | pub fn utf8_error(&self) -> Utf8Error { |
1673 | self.error | |
1674 | } | |
1a4d82fc JJ |
1675 | } |
1676 | ||
85aaf69f SL |
1677 | #[stable(feature = "rust1", since = "1.0.0")] |
1678 | impl fmt::Display for FromUtf8Error { | |
9fa01778 | 1679 | fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result { |
85aaf69f | 1680 | fmt::Display::fmt(&self.error, f) |
1a4d82fc JJ |
1681 | } |
1682 | } | |
1683 | ||
85aaf69f SL |
1684 | #[stable(feature = "rust1", since = "1.0.0")] |
1685 | impl fmt::Display for FromUtf16Error { | |
9fa01778 | 1686 | fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result { |
85aaf69f | 1687 | fmt::Display::fmt("invalid utf-16: lone surrogate found", f) |
1a4d82fc JJ |
1688 | } |
1689 | } | |
1690 | ||
b039eaaf SL |
1691 | #[stable(feature = "rust1", since = "1.0.0")] |
1692 | impl Clone for String { | |
1693 | fn clone(&self) -> Self { | |
1694 | String { vec: self.vec.clone() } | |
1695 | } | |
1696 | ||
1697 | fn clone_from(&mut self, source: &Self) { | |
1698 | self.vec.clone_from(&source.vec); | |
1699 | } | |
1700 | } | |
1701 | ||
85aaf69f | 1702 | #[stable(feature = "rust1", since = "1.0.0")] |
1a4d82fc | 1703 | impl FromIterator<char> for String { |
54a0048b | 1704 | fn from_iter<I: IntoIterator<Item = char>>(iter: I) -> String { |
1a4d82fc | 1705 | let mut buf = String::new(); |
54a0048b | 1706 | buf.extend(iter); |
1a4d82fc JJ |
1707 | buf |
1708 | } | |
1709 | } | |
1710 | ||
8bb4bdeb XL |
1711 | #[stable(feature = "string_from_iter_by_ref", since = "1.17.0")] |
1712 | impl<'a> FromIterator<&'a char> for String { | |
1713 | fn from_iter<I: IntoIterator<Item = &'a char>>(iter: I) -> String { | |
1714 | let mut buf = String::new(); | |
1715 | buf.extend(iter); | |
1716 | buf | |
1717 | } | |
1718 | } | |
1719 | ||
85aaf69f | 1720 | #[stable(feature = "rust1", since = "1.0.0")] |
1a4d82fc | 1721 | impl<'a> FromIterator<&'a str> for String { |
54a0048b | 1722 | fn from_iter<I: IntoIterator<Item = &'a str>>(iter: I) -> String { |
e9174d1e | 1723 | let mut buf = String::new(); |
54a0048b | 1724 | buf.extend(iter); |
e9174d1e SL |
1725 | buf |
1726 | } | |
1727 | } | |
1728 | ||
1729 | #[stable(feature = "extend_string", since = "1.4.0")] | |
1730 | impl FromIterator<String> for String { | |
54a0048b | 1731 | fn from_iter<I: IntoIterator<Item = String>>(iter: I) -> String { |
0731742a XL |
1732 | let mut iterator = iter.into_iter(); |
1733 | ||
1734 | // Because we're iterating over `String`s, we can avoid at least | |
1735 | // one allocation by getting the first string from the iterator | |
1736 | // and appending to it all the subsequent strings. | |
1737 | match iterator.next() { | |
1738 | None => String::new(), | |
1739 | Some(mut buf) => { | |
1740 | buf.extend(iterator); | |
1741 | buf | |
1742 | } | |
1743 | } | |
1a4d82fc JJ |
1744 | } |
1745 | } | |
1746 | ||
7cac9316 XL |
1747 | #[stable(feature = "herd_cows", since = "1.19.0")] |
1748 | impl<'a> FromIterator<Cow<'a, str>> for String { | |
1749 | fn from_iter<I: IntoIterator<Item = Cow<'a, str>>>(iter: I) -> String { | |
0731742a XL |
1750 | let mut iterator = iter.into_iter(); |
1751 | ||
1752 | // Because we're iterating over CoWs, we can (potentially) avoid at least | |
1753 | // one allocation by getting the first item and appending to it all the | |
1754 | // subsequent items. | |
1755 | match iterator.next() { | |
1756 | None => String::new(), | |
1757 | Some(cow) => { | |
1758 | let mut buf = cow.into_owned(); | |
1759 | buf.extend(iterator); | |
1760 | buf | |
1761 | } | |
1762 | } | |
7cac9316 XL |
1763 | } |
1764 | } | |
1765 | ||
bd371182 | 1766 | #[stable(feature = "rust1", since = "1.0.0")] |
1a4d82fc | 1767 | impl Extend<char> for String { |
54a0048b SL |
1768 | fn extend<I: IntoIterator<Item = char>>(&mut self, iter: I) { |
1769 | let iterator = iter.into_iter(); | |
1a4d82fc JJ |
1770 | let (lower_bound, _) = iterator.size_hint(); |
1771 | self.reserve(lower_bound); | |
0731742a | 1772 | iterator.for_each(move |c| self.push(c)); |
1a4d82fc JJ |
1773 | } |
1774 | } | |
1775 | ||
62682a34 SL |
1776 | #[stable(feature = "extend_ref", since = "1.2.0")] |
1777 | impl<'a> Extend<&'a char> for String { | |
54a0048b SL |
1778 | fn extend<I: IntoIterator<Item = &'a char>>(&mut self, iter: I) { |
1779 | self.extend(iter.into_iter().cloned()); | |
62682a34 SL |
1780 | } |
1781 | } | |
1782 | ||
bd371182 | 1783 | #[stable(feature = "rust1", since = "1.0.0")] |
1a4d82fc | 1784 | impl<'a> Extend<&'a str> for String { |
54a0048b | 1785 | fn extend<I: IntoIterator<Item = &'a str>>(&mut self, iter: I) { |
0731742a | 1786 | iter.into_iter().for_each(move |s| self.push_str(s)); |
1a4d82fc JJ |
1787 | } |
1788 | } | |
1789 | ||
e9174d1e SL |
1790 | #[stable(feature = "extend_string", since = "1.4.0")] |
1791 | impl Extend<String> for String { | |
54a0048b | 1792 | fn extend<I: IntoIterator<Item = String>>(&mut self, iter: I) { |
0731742a | 1793 | iter.into_iter().for_each(move |s| self.push_str(&s)); |
e9174d1e SL |
1794 | } |
1795 | } | |
1796 | ||
7cac9316 XL |
1797 | #[stable(feature = "herd_cows", since = "1.19.0")] |
1798 | impl<'a> Extend<Cow<'a, str>> for String { | |
1799 | fn extend<I: IntoIterator<Item = Cow<'a, str>>>(&mut self, iter: I) { | |
0731742a | 1800 | iter.into_iter().for_each(move |s| self.push_str(&s)); |
7cac9316 XL |
1801 | } |
1802 | } | |
1803 | ||
c34b1796 | 1804 | /// A convenience impl that delegates to the impl for `&str` |
92a42be0 SL |
1805 | #[unstable(feature = "pattern", |
1806 | reason = "API not fully fleshed out and ready to be stabilized", | |
1807 | issue = "27721")] | |
c34b1796 AL |
1808 | impl<'a, 'b> Pattern<'a> for &'b String { |
1809 | type Searcher = <&'b str as Pattern<'a>>::Searcher; | |
1810 | ||
1811 | fn into_searcher(self, haystack: &'a str) -> <&'b str as Pattern<'a>>::Searcher { | |
1812 | self[..].into_searcher(haystack) | |
1813 | } | |
1814 | ||
1815 | #[inline] | |
1816 | fn is_contained_in(self, haystack: &'a str) -> bool { | |
1817 | self[..].is_contained_in(haystack) | |
1818 | } | |
1819 | ||
1820 | #[inline] | |
1821 | fn is_prefix_of(self, haystack: &'a str) -> bool { | |
1822 | self[..].is_prefix_of(haystack) | |
1823 | } | |
1824 | } | |
1825 | ||
85aaf69f | 1826 | #[stable(feature = "rust1", since = "1.0.0")] |
1a4d82fc JJ |
1827 | impl PartialEq for String { |
1828 | #[inline] | |
92a42be0 SL |
1829 | fn eq(&self, other: &String) -> bool { |
1830 | PartialEq::eq(&self[..], &other[..]) | |
1831 | } | |
1a4d82fc | 1832 | #[inline] |
92a42be0 SL |
1833 | fn ne(&self, other: &String) -> bool { |
1834 | PartialEq::ne(&self[..], &other[..]) | |
1835 | } | |
1a4d82fc JJ |
1836 | } |
1837 | ||
1838 | macro_rules! impl_eq { | |
1839 | ($lhs:ty, $rhs: ty) => { | |
85aaf69f | 1840 | #[stable(feature = "rust1", since = "1.0.0")] |
92a42be0 | 1841 | impl<'a, 'b> PartialEq<$rhs> for $lhs { |
1a4d82fc | 1842 | #[inline] |
9346a6ac | 1843 | fn eq(&self, other: &$rhs) -> bool { PartialEq::eq(&self[..], &other[..]) } |
1a4d82fc | 1844 | #[inline] |
9346a6ac | 1845 | fn ne(&self, other: &$rhs) -> bool { PartialEq::ne(&self[..], &other[..]) } |
1a4d82fc JJ |
1846 | } |
1847 | ||
85aaf69f | 1848 | #[stable(feature = "rust1", since = "1.0.0")] |
92a42be0 | 1849 | impl<'a, 'b> PartialEq<$lhs> for $rhs { |
1a4d82fc | 1850 | #[inline] |
9346a6ac | 1851 | fn eq(&self, other: &$lhs) -> bool { PartialEq::eq(&self[..], &other[..]) } |
1a4d82fc | 1852 | #[inline] |
9346a6ac | 1853 | fn ne(&self, other: &$lhs) -> bool { PartialEq::ne(&self[..], &other[..]) } |
1a4d82fc JJ |
1854 | } |
1855 | ||
1856 | } | |
1857 | } | |
1858 | ||
9346a6ac | 1859 | impl_eq! { String, str } |
1a4d82fc | 1860 | impl_eq! { String, &'a str } |
9346a6ac | 1861 | impl_eq! { Cow<'a, str>, str } |
92a42be0 | 1862 | impl_eq! { Cow<'a, str>, &'b str } |
85aaf69f | 1863 | impl_eq! { Cow<'a, str>, String } |
1a4d82fc | 1864 | |
85aaf69f | 1865 | #[stable(feature = "rust1", since = "1.0.0")] |
1a4d82fc | 1866 | impl Default for String { |
9e0c209e | 1867 | /// Creates an empty `String`. |
85aaf69f | 1868 | #[inline] |
1a4d82fc JJ |
1869 | fn default() -> String { |
1870 | String::new() | |
1871 | } | |
1872 | } | |
1873 | ||
85aaf69f SL |
1874 | #[stable(feature = "rust1", since = "1.0.0")] |
1875 | impl fmt::Display for String { | |
1876 | #[inline] | |
9fa01778 | 1877 | fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result { |
85aaf69f | 1878 | fmt::Display::fmt(&**self, f) |
1a4d82fc JJ |
1879 | } |
1880 | } | |
1881 | ||
85aaf69f SL |
1882 | #[stable(feature = "rust1", since = "1.0.0")] |
1883 | impl fmt::Debug for String { | |
1884 | #[inline] | |
9fa01778 | 1885 | fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result { |
85aaf69f | 1886 | fmt::Debug::fmt(&**self, f) |
1a4d82fc JJ |
1887 | } |
1888 | } | |
1889 | ||
85aaf69f | 1890 | #[stable(feature = "rust1", since = "1.0.0")] |
85aaf69f | 1891 | impl hash::Hash for String { |
1a4d82fc | 1892 | #[inline] |
85aaf69f | 1893 | fn hash<H: hash::Hasher>(&self, hasher: &mut H) { |
1a4d82fc JJ |
1894 | (**self).hash(hasher) |
1895 | } | |
1896 | } | |
1897 | ||
8bb4bdeb XL |
1898 | /// Implements the `+` operator for concatenating two strings. |
1899 | /// | |
1900 | /// This consumes the `String` on the left-hand side and re-uses its buffer (growing it if | |
1901 | /// necessary). This is done to avoid allocating a new `String` and copying the entire contents on | |
1902 | /// every operation, which would lead to `O(n^2)` running time when building an `n`-byte string by | |
1903 | /// repeated concatenation. | |
1904 | /// | |
1905 | /// The string on the right-hand side is only borrowed; its contents are copied into the returned | |
1906 | /// `String`. | |
1907 | /// | |
1908 | /// # Examples | |
1909 | /// | |
1910 | /// Concatenating two `String`s takes the first by value and borrows the second: | |
1911 | /// | |
1912 | /// ``` | |
1913 | /// let a = String::from("hello"); | |
1914 | /// let b = String::from(" world"); | |
1915 | /// let c = a + &b; | |
1916 | /// // `a` is moved and can no longer be used here. | |
1917 | /// ``` | |
1918 | /// | |
1919 | /// If you want to keep using the first `String`, you can clone it and append to the clone instead: | |
1920 | /// | |
1921 | /// ``` | |
1922 | /// let a = String::from("hello"); | |
1923 | /// let b = String::from(" world"); | |
1924 | /// let c = a.clone() + &b; | |
1925 | /// // `a` is still valid here. | |
1926 | /// ``` | |
1927 | /// | |
1928 | /// Concatenating `&str` slices can be done by converting the first to a `String`: | |
1929 | /// | |
1930 | /// ``` | |
1931 | /// let a = "hello"; | |
1932 | /// let b = " world"; | |
1933 | /// let c = a.to_string() + b; | |
1934 | /// ``` | |
bd371182 | 1935 | #[stable(feature = "rust1", since = "1.0.0")] |
9fa01778 | 1936 | impl Add<&str> for String { |
1a4d82fc JJ |
1937 | type Output = String; |
1938 | ||
85aaf69f | 1939 | #[inline] |
1a4d82fc JJ |
1940 | fn add(mut self, other: &str) -> String { |
1941 | self.push_str(other); | |
1942 | self | |
1943 | } | |
1944 | } | |
1945 | ||
8bb4bdeb XL |
1946 | /// Implements the `+=` operator for appending to a `String`. |
1947 | /// | |
b7449926 | 1948 | /// This has the same behavior as the [`push_str`][String::push_str] method. |
5bcae85e | 1949 | #[stable(feature = "stringaddassign", since = "1.12.0")] |
9fa01778 | 1950 | impl AddAssign<&str> for String { |
5bcae85e SL |
1951 | #[inline] |
1952 | fn add_assign(&mut self, other: &str) { | |
1953 | self.push_str(other); | |
1954 | } | |
1955 | } | |
1956 | ||
85aaf69f SL |
1957 | #[stable(feature = "rust1", since = "1.0.0")] |
1958 | impl ops::Index<ops::Range<usize>> for String { | |
1a4d82fc | 1959 | type Output = str; |
c34b1796 | 1960 | |
1a4d82fc | 1961 | #[inline] |
c34b1796 AL |
1962 | fn index(&self, index: ops::Range<usize>) -> &str { |
1963 | &self[..][index] | |
1a4d82fc JJ |
1964 | } |
1965 | } | |
85aaf69f SL |
1966 | #[stable(feature = "rust1", since = "1.0.0")] |
1967 | impl ops::Index<ops::RangeTo<usize>> for String { | |
1a4d82fc | 1968 | type Output = str; |
c34b1796 | 1969 | |
1a4d82fc | 1970 | #[inline] |
c34b1796 AL |
1971 | fn index(&self, index: ops::RangeTo<usize>) -> &str { |
1972 | &self[..][index] | |
1a4d82fc JJ |
1973 | } |
1974 | } | |
85aaf69f SL |
1975 | #[stable(feature = "rust1", since = "1.0.0")] |
1976 | impl ops::Index<ops::RangeFrom<usize>> for String { | |
1a4d82fc | 1977 | type Output = str; |
c34b1796 | 1978 | |
1a4d82fc | 1979 | #[inline] |
c34b1796 AL |
1980 | fn index(&self, index: ops::RangeFrom<usize>) -> &str { |
1981 | &self[..][index] | |
1a4d82fc JJ |
1982 | } |
1983 | } | |
85aaf69f SL |
1984 | #[stable(feature = "rust1", since = "1.0.0")] |
1985 | impl ops::Index<ops::RangeFull> for String { | |
1a4d82fc | 1986 | type Output = str; |
c34b1796 | 1987 | |
1a4d82fc | 1988 | #[inline] |
c34b1796 | 1989 | fn index(&self, _index: ops::RangeFull) -> &str { |
e9174d1e | 1990 | unsafe { str::from_utf8_unchecked(&self.vec) } |
1a4d82fc JJ |
1991 | } |
1992 | } | |
0531ce1d | 1993 | #[stable(feature = "inclusive_range", since = "1.26.0")] |
54a0048b SL |
1994 | impl ops::Index<ops::RangeInclusive<usize>> for String { |
1995 | type Output = str; | |
1996 | ||
1997 | #[inline] | |
1998 | fn index(&self, index: ops::RangeInclusive<usize>) -> &str { | |
1999 | Index::index(&**self, index) | |
2000 | } | |
2001 | } | |
0531ce1d | 2002 | #[stable(feature = "inclusive_range", since = "1.26.0")] |
54a0048b SL |
2003 | impl ops::Index<ops::RangeToInclusive<usize>> for String { |
2004 | type Output = str; | |
2005 | ||
2006 | #[inline] | |
2007 | fn index(&self, index: ops::RangeToInclusive<usize>) -> &str { | |
2008 | Index::index(&**self, index) | |
2009 | } | |
2010 | } | |
1a4d82fc | 2011 | |
7cac9316 | 2012 | #[stable(feature = "derefmut_for_string", since = "1.3.0")] |
c1a9b12d SL |
2013 | impl ops::IndexMut<ops::Range<usize>> for String { |
2014 | #[inline] | |
2015 | fn index_mut(&mut self, index: ops::Range<usize>) -> &mut str { | |
2016 | &mut self[..][index] | |
2017 | } | |
2018 | } | |
7cac9316 | 2019 | #[stable(feature = "derefmut_for_string", since = "1.3.0")] |
c1a9b12d SL |
2020 | impl ops::IndexMut<ops::RangeTo<usize>> for String { |
2021 | #[inline] | |
2022 | fn index_mut(&mut self, index: ops::RangeTo<usize>) -> &mut str { | |
2023 | &mut self[..][index] | |
2024 | } | |
2025 | } | |
7cac9316 | 2026 | #[stable(feature = "derefmut_for_string", since = "1.3.0")] |
c1a9b12d SL |
2027 | impl ops::IndexMut<ops::RangeFrom<usize>> for String { |
2028 | #[inline] | |
2029 | fn index_mut(&mut self, index: ops::RangeFrom<usize>) -> &mut str { | |
2030 | &mut self[..][index] | |
2031 | } | |
2032 | } | |
7cac9316 | 2033 | #[stable(feature = "derefmut_for_string", since = "1.3.0")] |
c1a9b12d SL |
2034 | impl ops::IndexMut<ops::RangeFull> for String { |
2035 | #[inline] | |
2036 | fn index_mut(&mut self, _index: ops::RangeFull) -> &mut str { | |
cc61c64b | 2037 | unsafe { str::from_utf8_unchecked_mut(&mut *self.vec) } |
c1a9b12d SL |
2038 | } |
2039 | } | |
0531ce1d | 2040 | #[stable(feature = "inclusive_range", since = "1.26.0")] |
54a0048b SL |
2041 | impl ops::IndexMut<ops::RangeInclusive<usize>> for String { |
2042 | #[inline] | |
2043 | fn index_mut(&mut self, index: ops::RangeInclusive<usize>) -> &mut str { | |
2044 | IndexMut::index_mut(&mut **self, index) | |
2045 | } | |
2046 | } | |
0531ce1d | 2047 | #[stable(feature = "inclusive_range", since = "1.26.0")] |
54a0048b SL |
2048 | impl ops::IndexMut<ops::RangeToInclusive<usize>> for String { |
2049 | #[inline] | |
2050 | fn index_mut(&mut self, index: ops::RangeToInclusive<usize>) -> &mut str { | |
2051 | IndexMut::index_mut(&mut **self, index) | |
2052 | } | |
2053 | } | |
c1a9b12d | 2054 | |
85aaf69f | 2055 | #[stable(feature = "rust1", since = "1.0.0")] |
1a4d82fc JJ |
2056 | impl ops::Deref for String { |
2057 | type Target = str; | |
2058 | ||
85aaf69f SL |
2059 | #[inline] |
2060 | fn deref(&self) -> &str { | |
e9174d1e | 2061 | unsafe { str::from_utf8_unchecked(&self.vec) } |
1a4d82fc JJ |
2062 | } |
2063 | } | |
2064 | ||
7cac9316 | 2065 | #[stable(feature = "derefmut_for_string", since = "1.3.0")] |
c1a9b12d | 2066 | impl ops::DerefMut for String { |
85aaf69f | 2067 | #[inline] |
c1a9b12d | 2068 | fn deref_mut(&mut self) -> &mut str { |
cc61c64b | 2069 | unsafe { str::from_utf8_unchecked_mut(&mut *self.vec) } |
1a4d82fc JJ |
2070 | } |
2071 | } | |
2072 | ||
92a42be0 SL |
2073 | /// An error when parsing a `String`. |
2074 | /// | |
2075 | /// This `enum` is slightly awkward: it will never actually exist. This error is | |
2076 | /// part of the type signature of the implementation of [`FromStr`] on | |
cc61c64b | 2077 | /// [`String`]. The return type of [`from_str`], requires that an error be |
92a42be0 SL |
2078 | /// defined, but, given that a [`String`] can always be made into a new |
2079 | /// [`String`] without error, this type will never actually be returned. As | |
2080 | /// such, it is only here to satisfy said signature, and is useless otherwise. | |
2081 | /// | |
54a0048b | 2082 | /// [`FromStr`]: ../../std/str/trait.FromStr.html |
92a42be0 | 2083 | /// [`String`]: struct.String.html |
cc61c64b | 2084 | /// [`from_str`]: ../../std/str/trait.FromStr.html#tymethod.from_str |
b039eaaf | 2085 | #[stable(feature = "str_parse_error", since = "1.5.0")] |
9fa01778 | 2086 | pub type ParseError = core::convert::Infallible; |
bd371182 AL |
2087 | |
2088 | #[stable(feature = "rust1", since = "1.0.0")] | |
1a4d82fc | 2089 | impl FromStr for String { |
9fa01778 | 2090 | type Err = core::convert::Infallible; |
1a4d82fc | 2091 | #[inline] |
bd371182 | 2092 | fn from_str(s: &str) -> Result<String, ParseError> { |
62682a34 | 2093 | Ok(String::from(s)) |
1a4d82fc JJ |
2094 | } |
2095 | } | |
2096 | ||
b039eaaf | 2097 | |
92a42be0 SL |
2098 | /// A trait for converting a value to a `String`. |
2099 | /// | |
2100 | /// This trait is automatically implemented for any type which implements the | |
2101 | /// [`Display`] trait. As such, `ToString` shouldn't be implemented directly: | |
2102 | /// [`Display`] should be implemented instead, and you get the `ToString` | |
2103 | /// implementation for free. | |
2104 | /// | |
54a0048b | 2105 | /// [`Display`]: ../../std/fmt/trait.Display.html |
85aaf69f | 2106 | #[stable(feature = "rust1", since = "1.0.0")] |
1a4d82fc | 2107 | pub trait ToString { |
92a42be0 SL |
2108 | /// Converts the given value to a `String`. |
2109 | /// | |
2110 | /// # Examples | |
2111 | /// | |
2112 | /// Basic usage: | |
2113 | /// | |
2114 | /// ``` | |
2115 | /// let i = 5; | |
2116 | /// let five = String::from("5"); | |
2117 | /// | |
2118 | /// assert_eq!(five, i.to_string()); | |
2119 | /// ``` | |
2c00a5a8 | 2120 | #[rustc_conversion_suggestion] |
85aaf69f | 2121 | #[stable(feature = "rust1", since = "1.0.0")] |
1a4d82fc JJ |
2122 | fn to_string(&self) -> String; |
2123 | } | |
2124 | ||
8bb4bdeb XL |
2125 | /// # Panics |
2126 | /// | |
2127 | /// In this implementation, the `to_string` method panics | |
2128 | /// if the `Display` implementation returns an error. | |
2129 | /// This indicates an incorrect `Display` implementation | |
2130 | /// since `fmt::Write for String` never returns an error itself. | |
85aaf69f SL |
2131 | #[stable(feature = "rust1", since = "1.0.0")] |
2132 | impl<T: fmt::Display + ?Sized> ToString for T { | |
2133 | #[inline] | |
54a0048b | 2134 | default fn to_string(&self) -> String { |
9fa01778 | 2135 | use fmt::Write; |
1a4d82fc | 2136 | let mut buf = String::new(); |
8bb4bdeb | 2137 | buf.write_fmt(format_args!("{}", self)) |
0731742a | 2138 | .expect("a Display implementation returned an error unexpectedly"); |
1a4d82fc JJ |
2139 | buf.shrink_to_fit(); |
2140 | buf | |
2141 | } | |
2142 | } | |
2143 | ||
54a0048b SL |
2144 | #[stable(feature = "str_to_string_specialization", since = "1.9.0")] |
2145 | impl ToString for str { | |
2146 | #[inline] | |
2147 | fn to_string(&self) -> String { | |
2148 | String::from(self) | |
2149 | } | |
2150 | } | |
2151 | ||
8bb4bdeb | 2152 | #[stable(feature = "cow_str_to_string_specialization", since = "1.17.0")] |
9fa01778 | 2153 | impl ToString for Cow<'_, str> { |
8bb4bdeb XL |
2154 | #[inline] |
2155 | fn to_string(&self) -> String { | |
2156 | self[..].to_owned() | |
2157 | } | |
2158 | } | |
2159 | ||
2160 | #[stable(feature = "string_to_string_specialization", since = "1.17.0")] | |
2161 | impl ToString for String { | |
2162 | #[inline] | |
2163 | fn to_string(&self) -> String { | |
2164 | self.to_owned() | |
2165 | } | |
2166 | } | |
2167 | ||
85aaf69f | 2168 | #[stable(feature = "rust1", since = "1.0.0")] |
c34b1796 | 2169 | impl AsRef<str> for String { |
d9579d0f | 2170 | #[inline] |
c34b1796 AL |
2171 | fn as_ref(&self) -> &str { |
2172 | self | |
2173 | } | |
2174 | } | |
2175 | ||
bd371182 AL |
2176 | #[stable(feature = "rust1", since = "1.0.0")] |
2177 | impl AsRef<[u8]> for String { | |
2178 | #[inline] | |
2179 | fn as_ref(&self) -> &[u8] { | |
2180 | self.as_bytes() | |
2181 | } | |
2182 | } | |
2183 | ||
c34b1796 | 2184 | #[stable(feature = "rust1", since = "1.0.0")] |
532ac7d7 | 2185 | impl From<&str> for String { |
0bf4aa26 | 2186 | #[inline] |
532ac7d7 | 2187 | fn from(s: &str) -> String { |
54a0048b | 2188 | s.to_owned() |
c34b1796 AL |
2189 | } |
2190 | } | |
2191 | ||
48663c56 XL |
2192 | #[stable(feature = "from_ref_string", since = "1.35.0")] |
2193 | impl From<&String> for String { | |
2194 | #[inline] | |
2195 | fn from(s: &String) -> String { | |
2196 | s.clone() | |
2197 | } | |
2198 | } | |
2199 | ||
cc61c64b XL |
2200 | // note: test pulls in libstd, which causes errors here |
2201 | #[cfg(not(test))] | |
7cac9316 | 2202 | #[stable(feature = "string_from_box", since = "1.18.0")] |
cc61c64b | 2203 | impl From<Box<str>> for String { |
a1dfa0c6 XL |
2204 | /// Converts the given boxed `str` slice to a `String`. |
2205 | /// It is notable that the `str` slice is owned. | |
2206 | /// | |
2207 | /// # Examples | |
2208 | /// | |
2209 | /// Basic usage: | |
2210 | /// | |
2211 | /// ``` | |
2212 | /// let s1: String = String::from("hello world"); | |
2213 | /// let s2: Box<str> = s1.into_boxed_str(); | |
2214 | /// let s3: String = String::from(s2); | |
2215 | /// | |
2216 | /// assert_eq!("hello world", s3) | |
2217 | /// ``` | |
cc61c64b XL |
2218 | fn from(s: Box<str>) -> String { |
2219 | s.into_string() | |
2220 | } | |
2221 | } | |
2222 | ||
041b39d2 XL |
2223 | #[stable(feature = "box_from_str", since = "1.20.0")] |
2224 | impl From<String> for Box<str> { | |
a1dfa0c6 XL |
2225 | /// Converts the given `String` to a boxed `str` slice that is owned. |
2226 | /// | |
2227 | /// # Examples | |
2228 | /// | |
2229 | /// Basic usage: | |
2230 | /// | |
2231 | /// ``` | |
2232 | /// let s1: String = String::from("hello world"); | |
2233 | /// let s2: Box<str> = Box::from(s1); | |
2234 | /// let s3: String = String::from(s2); | |
2235 | /// | |
2236 | /// assert_eq!("hello world", s3) | |
2237 | /// ``` | |
041b39d2 XL |
2238 | fn from(s: String) -> Box<str> { |
2239 | s.into_boxed_str() | |
cc61c64b XL |
2240 | } |
2241 | } | |
2242 | ||
c30ab7b3 SL |
2243 | #[stable(feature = "string_from_cow_str", since = "1.14.0")] |
2244 | impl<'a> From<Cow<'a, str>> for String { | |
2245 | fn from(s: Cow<'a, str>) -> String { | |
2246 | s.into_owned() | |
2247 | } | |
2248 | } | |
2249 | ||
c34b1796 AL |
2250 | #[stable(feature = "rust1", since = "1.0.0")] |
2251 | impl<'a> From<&'a str> for Cow<'a, str> { | |
2252 | #[inline] | |
2253 | fn from(s: &'a str) -> Cow<'a, str> { | |
2254 | Cow::Borrowed(s) | |
2255 | } | |
2256 | } | |
2257 | ||
2258 | #[stable(feature = "rust1", since = "1.0.0")] | |
2259 | impl<'a> From<String> for Cow<'a, str> { | |
2260 | #[inline] | |
2261 | fn from(s: String) -> Cow<'a, str> { | |
2262 | Cow::Owned(s) | |
2263 | } | |
2264 | } | |
2265 | ||
94b46f34 XL |
2266 | #[stable(feature = "cow_from_string_ref", since = "1.28.0")] |
2267 | impl<'a> From<&'a String> for Cow<'a, str> { | |
2268 | #[inline] | |
2269 | fn from(s: &'a String) -> Cow<'a, str> { | |
2270 | Cow::Borrowed(s.as_str()) | |
2271 | } | |
2272 | } | |
2273 | ||
5bcae85e SL |
2274 | #[stable(feature = "cow_str_from_iter", since = "1.12.0")] |
2275 | impl<'a> FromIterator<char> for Cow<'a, str> { | |
2276 | fn from_iter<I: IntoIterator<Item = char>>(it: I) -> Cow<'a, str> { | |
2277 | Cow::Owned(FromIterator::from_iter(it)) | |
2278 | } | |
2279 | } | |
2280 | ||
2281 | #[stable(feature = "cow_str_from_iter", since = "1.12.0")] | |
2282 | impl<'a, 'b> FromIterator<&'b str> for Cow<'a, str> { | |
2283 | fn from_iter<I: IntoIterator<Item = &'b str>>(it: I) -> Cow<'a, str> { | |
2284 | Cow::Owned(FromIterator::from_iter(it)) | |
2285 | } | |
2286 | } | |
2287 | ||
2288 | #[stable(feature = "cow_str_from_iter", since = "1.12.0")] | |
2289 | impl<'a> FromIterator<String> for Cow<'a, str> { | |
2290 | fn from_iter<I: IntoIterator<Item = String>>(it: I) -> Cow<'a, str> { | |
2291 | Cow::Owned(FromIterator::from_iter(it)) | |
2292 | } | |
2293 | } | |
2294 | ||
c30ab7b3 SL |
2295 | #[stable(feature = "from_string_for_vec_u8", since = "1.14.0")] |
2296 | impl From<String> for Vec<u8> { | |
a1dfa0c6 XL |
2297 | /// Converts the given `String` to a vector `Vec` that holds values of type `u8`. |
2298 | /// | |
2299 | /// # Examples | |
2300 | /// | |
2301 | /// Basic usage: | |
2302 | /// | |
2303 | /// ``` | |
2304 | /// let s1 = String::from("hello world"); | |
2305 | /// let v1 = Vec::from(s1); | |
2306 | /// | |
2307 | /// for b in v1 { | |
2308 | /// println!("{}", b); | |
2309 | /// } | |
2310 | /// ``` | |
32a655c1 | 2311 | fn from(string: String) -> Vec<u8> { |
c30ab7b3 | 2312 | string.into_bytes() |
c34b1796 AL |
2313 | } |
2314 | } | |
2315 | ||
85aaf69f SL |
2316 | #[stable(feature = "rust1", since = "1.0.0")] |
2317 | impl fmt::Write for String { | |
2318 | #[inline] | |
1a4d82fc JJ |
2319 | fn write_str(&mut self, s: &str) -> fmt::Result { |
2320 | self.push_str(s); | |
2321 | Ok(()) | |
2322 | } | |
d9579d0f AL |
2323 | |
2324 | #[inline] | |
2325 | fn write_char(&mut self, c: char) -> fmt::Result { | |
2326 | self.push(c); | |
2327 | Ok(()) | |
2328 | } | |
2329 | } | |
2330 | ||
2331 | /// A draining iterator for `String`. | |
7453a54e | 2332 | /// |
cc61c64b | 2333 | /// This struct is created by the [`drain`] method on [`String`]. See its |
7453a54e SL |
2334 | /// documentation for more. |
2335 | /// | |
cc61c64b | 2336 | /// [`drain`]: struct.String.html#method.drain |
7453a54e | 2337 | /// [`String`]: struct.String.html |
92a42be0 | 2338 | #[stable(feature = "drain", since = "1.6.0")] |
d9579d0f AL |
2339 | pub struct Drain<'a> { |
2340 | /// Will be used as &'a mut String in the destructor | |
2341 | string: *mut String, | |
2342 | /// Start of part to remove | |
2343 | start: usize, | |
2344 | /// End of part to remove | |
2345 | end: usize, | |
2346 | /// Current remaining range to remove | |
2347 | iter: Chars<'a>, | |
2348 | } | |
2349 | ||
8bb4bdeb | 2350 | #[stable(feature = "collection_debug", since = "1.17.0")] |
9fa01778 XL |
2351 | impl fmt::Debug for Drain<'_> { |
2352 | fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result { | |
8bb4bdeb XL |
2353 | f.pad("Drain { .. }") |
2354 | } | |
2355 | } | |
2356 | ||
92a42be0 | 2357 | #[stable(feature = "drain", since = "1.6.0")] |
9fa01778 | 2358 | unsafe impl Sync for Drain<'_> {} |
92a42be0 | 2359 | #[stable(feature = "drain", since = "1.6.0")] |
9fa01778 | 2360 | unsafe impl Send for Drain<'_> {} |
d9579d0f | 2361 | |
92a42be0 | 2362 | #[stable(feature = "drain", since = "1.6.0")] |
9fa01778 | 2363 | impl Drop for Drain<'_> { |
d9579d0f AL |
2364 | fn drop(&mut self) { |
2365 | unsafe { | |
2366 | // Use Vec::drain. "Reaffirm" the bounds checks to avoid | |
2367 | // panic code being inserted again. | |
2368 | let self_vec = (*self.string).as_mut_vec(); | |
2369 | if self.start <= self.end && self.end <= self_vec.len() { | |
2370 | self_vec.drain(self.start..self.end); | |
2371 | } | |
2372 | } | |
2373 | } | |
2374 | } | |
2375 | ||
92a42be0 | 2376 | #[stable(feature = "drain", since = "1.6.0")] |
9fa01778 | 2377 | impl Iterator for Drain<'_> { |
d9579d0f AL |
2378 | type Item = char; |
2379 | ||
2380 | #[inline] | |
2381 | fn next(&mut self) -> Option<char> { | |
2382 | self.iter.next() | |
2383 | } | |
2384 | ||
2385 | fn size_hint(&self) -> (usize, Option<usize>) { | |
2386 | self.iter.size_hint() | |
2387 | } | |
2388 | } | |
2389 | ||
92a42be0 | 2390 | #[stable(feature = "drain", since = "1.6.0")] |
9fa01778 | 2391 | impl DoubleEndedIterator for Drain<'_> { |
d9579d0f AL |
2392 | #[inline] |
2393 | fn next_back(&mut self) -> Option<char> { | |
2394 | self.iter.next_back() | |
2395 | } | |
1a4d82fc | 2396 | } |
9e0c209e | 2397 | |
0531ce1d | 2398 | #[stable(feature = "fused", since = "1.26.0")] |
9fa01778 | 2399 | impl FusedIterator for Drain<'_> {} |