1 // Copyright 2014 The Rust Project Developers. See the COPYRIGHT
2 // file at the top-level directory of this distribution and at
3 // http://rust-lang.org/COPYRIGHT.
5 // Licensed under the Apache License, Version 2.0 <LICENSE-APACHE or
6 // http://www.apache.org/licenses/LICENSE-2.0> or the MIT license
7 // <LICENSE-MIT or http://opensource.org/licenses/MIT>, at your
8 // option. This file may not be copied, modified, or distributed
9 // except according to those terms.
11 // ignore-lexer-test FIXME #15679
13 //! An owned, growable string that enforces that its contents are valid UTF-8.
15 #![stable(feature = "rust1", since = "1.0.0")]
19 use core::borrow::{Cow, IntoCow};
20 use core::default::Default;
21 use core::error::Error;
24 use core::iter::FromIterator;
26 use core::ops::{self, Deref, Add, Index};
28 use core::raw::Slice as RawSlice;
29 use unicode::str as unicode_str;
30 use unicode::str::Utf16Item;
32 use str::{self, CharRange, FromStr, Utf8Error};
33 use vec::{DerefVec, Vec, as_vec};
35 /// A growable string stored as a UTF-8 encoded buffer.
36 #[derive(Clone, PartialOrd, Eq, Ord)]
37 #[stable(feature = "rust1", since = "1.0.0")]
42 /// A possible error value from the `String::from_utf8` function.
43 #[stable(feature = "rust1", since = "1.0.0")]
45 pub struct FromUtf8Error {
50 /// A possible error value from the `String::from_utf16` function.
51 #[stable(feature = "rust1", since = "1.0.0")]
52 #[allow(missing_copy_implementations)]
54 pub struct FromUtf16Error(());
57 /// Creates a new string buffer initialized with the empty string.
62 /// let mut s = String::new();
65 #[stable(feature = "rust1", since = "1.0.0")]
66 pub fn new() -> String {
72 /// Creates a new string buffer with the given capacity.
73 /// The string will be able to hold exactly `capacity` bytes without
74 /// reallocating. If `capacity` is 0, the string will not allocate.
79 /// let mut s = String::with_capacity(10);
82 #[stable(feature = "rust1", since = "1.0.0")]
83 pub fn with_capacity(capacity: uint) -> String {
85 vec: Vec::with_capacity(capacity),
89 /// Creates a new string buffer from the given string.
94 /// let s = String::from_str("hello");
95 /// assert_eq!(s.as_slice(), "hello");
98 #[unstable(feature = "collections",
99 reason = "needs investigation to see if to_string() can match perf")]
100 pub fn from_str(string: &str) -> String {
101 String { vec: ::slice::SliceExt::to_vec(string.as_bytes()) }
104 /// Returns the vector as a string buffer, if possible, taking care not to
109 /// If the given vector is not valid UTF-8, then the original vector and the
110 /// corresponding error is returned.
115 /// use std::str::Utf8Error;
117 /// let hello_vec = vec![104, 101, 108, 108, 111];
118 /// let s = String::from_utf8(hello_vec).unwrap();
119 /// assert_eq!(s, "hello");
121 /// let invalid_vec = vec![240, 144, 128];
122 /// let s = String::from_utf8(invalid_vec).err().unwrap();
123 /// assert_eq!(s.utf8_error(), Utf8Error::TooShort);
124 /// assert_eq!(s.into_bytes(), vec![240, 144, 128]);
127 #[stable(feature = "rust1", since = "1.0.0")]
128 pub fn from_utf8(vec: Vec<u8>) -> Result<String, FromUtf8Error> {
129 match str::from_utf8(vec.as_slice()) {
130 Ok(..) => Ok(String { vec: vec }),
131 Err(e) => Err(FromUtf8Error { bytes: vec, error: e })
135 /// Converts a vector of bytes to a new UTF-8 string.
136 /// Any invalid UTF-8 sequences are replaced with U+FFFD REPLACEMENT CHARACTER.
141 /// let input = b"Hello \xF0\x90\x80World";
142 /// let output = String::from_utf8_lossy(input);
143 /// assert_eq!(output.as_slice(), "Hello \u{FFFD}World");
145 #[stable(feature = "rust1", since = "1.0.0")]
146 pub fn from_utf8_lossy<'a>(v: &'a [u8]) -> CowString<'a> {
148 match str::from_utf8(v) {
149 Ok(s) => return Cow::Borrowed(s),
151 if let Utf8Error::InvalidByte(firstbad) = e {
157 static TAG_CONT_U8: u8 = 128u8;
158 static REPLACEMENT: &'static [u8] = b"\xEF\xBF\xBD"; // U+FFFD in UTF-8
160 fn unsafe_get(xs: &[u8], i: uint) -> u8 {
161 unsafe { *xs.get_unchecked(i) }
163 fn safe_get(xs: &[u8], i: uint, total: uint) -> u8 {
171 let mut res = String::with_capacity(total);
175 res.as_mut_vec().push_all(&v[..i])
179 // subseqidx is the index of the first byte of the subsequence we're looking at.
180 // It's used to copy a bunch of contiguous good codepoints at once instead of copying
182 let mut subseqidx = i;
186 let byte = unsafe_get(v, i);
189 macro_rules! error { () => ({
192 res.as_mut_vec().push_all(&v[subseqidx..i_]);
195 res.as_mut_vec().push_all(REPLACEMENT);
200 // subseqidx handles this
202 let w = unicode_str::utf8_char_width(byte);
206 if safe_get(v, i, total) & 192u8 != TAG_CONT_U8 {
213 match (byte, safe_get(v, i, total)) {
214 (0xE0 , 0xA0 ... 0xBF) => (),
215 (0xE1 ... 0xEC, 0x80 ... 0xBF) => (),
216 (0xED , 0x80 ... 0x9F) => (),
217 (0xEE ... 0xEF, 0x80 ... 0xBF) => (),
224 if safe_get(v, i, total) & 192u8 != TAG_CONT_U8 {
231 match (byte, safe_get(v, i, total)) {
232 (0xF0 , 0x90 ... 0xBF) => (),
233 (0xF1 ... 0xF3, 0x80 ... 0xBF) => (),
234 (0xF4 , 0x80 ... 0x8F) => (),
241 if safe_get(v, i, total) & 192u8 != TAG_CONT_U8 {
246 if safe_get(v, i, total) & 192u8 != TAG_CONT_U8 {
259 if subseqidx < total {
261 res.as_mut_vec().push_all(&v[subseqidx..total])
267 /// Decode a UTF-16 encoded vector `v` into a `String`, returning `None`
268 /// if `v` contains any invalid data.
274 /// let mut v = &mut [0xD834, 0xDD1E, 0x006d, 0x0075,
275 /// 0x0073, 0x0069, 0x0063];
276 /// assert_eq!(String::from_utf16(v).unwrap(),
277 /// "𝄞music".to_string());
279 /// // 𝄞mu<invalid>ic
281 /// assert!(String::from_utf16(v).is_err());
283 #[stable(feature = "rust1", since = "1.0.0")]
284 pub fn from_utf16(v: &[u16]) -> Result<String, FromUtf16Error> {
285 let mut s = String::with_capacity(v.len());
286 for c in unicode_str::utf16_items(v) {
288 Utf16Item::ScalarValue(c) => s.push(c),
289 Utf16Item::LoneSurrogate(_) => return Err(FromUtf16Error(())),
295 /// Decode a UTF-16 encoded vector `v` into a string, replacing
296 /// invalid data with the replacement character (U+FFFD).
301 /// // 𝄞mus<invalid>ic<invalid>
302 /// let v = &[0xD834, 0xDD1E, 0x006d, 0x0075,
303 /// 0x0073, 0xDD1E, 0x0069, 0x0063,
306 /// assert_eq!(String::from_utf16_lossy(v),
307 /// "𝄞mus\u{FFFD}ic\u{FFFD}".to_string());
310 #[stable(feature = "rust1", since = "1.0.0")]
311 pub fn from_utf16_lossy(v: &[u16]) -> String {
312 unicode_str::utf16_items(v).map(|c| c.to_char_lossy()).collect()
315 /// Creates a new `String` from a length, capacity, and pointer.
317 /// This is unsafe because:
318 /// * We call `Vec::from_raw_parts` to get a `Vec<u8>`;
319 /// * We assume that the `Vec` contains valid UTF-8.
321 #[stable(feature = "rust1", since = "1.0.0")]
322 pub unsafe fn from_raw_parts(buf: *mut u8, length: uint, capacity: uint) -> String {
324 vec: Vec::from_raw_parts(buf, length, capacity),
328 /// Converts a vector of bytes to a new `String` without checking if
329 /// it contains valid UTF-8. This is unsafe because it assumes that
330 /// the UTF-8-ness of the vector has already been validated.
332 #[stable(feature = "rust1", since = "1.0.0")]
333 pub unsafe fn from_utf8_unchecked(bytes: Vec<u8>) -> String {
334 String { vec: bytes }
337 /// Return the underlying byte buffer, encoded as UTF-8.
342 /// let s = String::from_str("hello");
343 /// let bytes = s.into_bytes();
344 /// assert_eq!(bytes, vec![104, 101, 108, 108, 111]);
347 #[stable(feature = "rust1", since = "1.0.0")]
348 pub fn into_bytes(self) -> Vec<u8> {
352 /// Pushes the given string onto this string buffer.
357 /// let mut s = String::from_str("foo");
358 /// s.push_str("bar");
359 /// assert_eq!(s.as_slice(), "foobar");
362 #[stable(feature = "rust1", since = "1.0.0")]
363 pub fn push_str(&mut self, string: &str) {
364 self.vec.push_all(string.as_bytes())
367 /// Returns the number of bytes that this string buffer can hold without
373 /// let s = String::with_capacity(10);
374 /// assert!(s.capacity() >= 10);
377 #[stable(feature = "rust1", since = "1.0.0")]
378 pub fn capacity(&self) -> uint {
382 /// Reserves capacity for at least `additional` more bytes to be inserted
383 /// in the given `String`. The collection may reserve more space to avoid
384 /// frequent reallocations.
388 /// Panics if the new capacity overflows `uint`.
393 /// let mut s = String::new();
395 /// assert!(s.capacity() >= 10);
398 #[stable(feature = "rust1", since = "1.0.0")]
399 pub fn reserve(&mut self, additional: uint) {
400 self.vec.reserve(additional)
403 /// Reserves the minimum capacity for exactly `additional` more bytes to be
404 /// inserted in the given `String`. Does nothing if the capacity is already
407 /// Note that the allocator may give the collection more space than it
408 /// requests. Therefore capacity can not be relied upon to be precisely
409 /// minimal. Prefer `reserve` if future insertions are expected.
413 /// Panics if the new capacity overflows `uint`.
418 /// let mut s = String::new();
420 /// assert!(s.capacity() >= 10);
423 #[stable(feature = "rust1", since = "1.0.0")]
424 pub fn reserve_exact(&mut self, additional: uint) {
425 self.vec.reserve_exact(additional)
428 /// Shrinks the capacity of this string buffer to match its length.
433 /// let mut s = String::from_str("foo");
435 /// assert!(s.capacity() >= 100);
436 /// s.shrink_to_fit();
437 /// assert_eq!(s.capacity(), 3);
440 #[stable(feature = "rust1", since = "1.0.0")]
441 pub fn shrink_to_fit(&mut self) {
442 self.vec.shrink_to_fit()
445 /// Adds the given character to the end of the string.
450 /// let mut s = String::from_str("abc");
454 /// assert_eq!(s.as_slice(), "abc123");
457 #[stable(feature = "rust1", since = "1.0.0")]
458 pub fn push(&mut self, ch: char) {
459 if (ch as u32) < 0x80 {
460 self.vec.push(ch as u8);
464 let cur_len = self.len();
465 // This may use up to 4 bytes.
469 // Attempt to not use an intermediate buffer by just pushing bytes
470 // directly onto this string.
471 let slice = RawSlice {
472 data: self.vec.as_ptr().offset(cur_len as int),
475 let used = ch.encode_utf8(mem::transmute(slice)).unwrap_or(0);
476 self.vec.set_len(cur_len + used);
480 /// Works with the underlying buffer as a byte slice.
485 /// let s = String::from_str("hello");
486 /// let b: &[_] = &[104, 101, 108, 108, 111];
487 /// assert_eq!(s.as_bytes(), b);
490 #[stable(feature = "rust1", since = "1.0.0")]
491 pub fn as_bytes<'a>(&'a self) -> &'a [u8] {
495 /// Shortens a string to the specified length.
499 /// Panics if `new_len` > current length,
500 /// or if `new_len` is not a character boundary.
505 /// let mut s = String::from_str("hello");
507 /// assert_eq!(s.as_slice(), "he");
510 #[stable(feature = "rust1", since = "1.0.0")]
511 pub fn truncate(&mut self, new_len: uint) {
512 assert!(self.is_char_boundary(new_len));
513 self.vec.truncate(new_len)
516 /// Removes the last character from the string buffer and returns it.
517 /// Returns `None` if this string buffer is empty.
522 /// let mut s = String::from_str("foo");
523 /// assert_eq!(s.pop(), Some('o'));
524 /// assert_eq!(s.pop(), Some('o'));
525 /// assert_eq!(s.pop(), Some('f'));
526 /// assert_eq!(s.pop(), None);
529 #[stable(feature = "rust1", since = "1.0.0")]
530 pub fn pop(&mut self) -> Option<char> {
531 let len = self.len();
536 let CharRange {ch, next} = self.char_range_at_reverse(len);
538 self.vec.set_len(next);
543 /// Removes the character from the string buffer at byte position `idx` and
548 /// This is an O(n) operation as it requires copying every element in the
553 /// If `idx` does not lie on a character boundary, or if it is out of
554 /// bounds, then this function will panic.
559 /// let mut s = String::from_str("foo");
560 /// assert_eq!(s.remove(0), 'f');
561 /// assert_eq!(s.remove(1), 'o');
562 /// assert_eq!(s.remove(0), 'o');
565 #[stable(feature = "rust1", since = "1.0.0")]
566 pub fn remove(&mut self, idx: uint) -> char {
567 let len = self.len();
570 let CharRange { ch, next } = self.char_range_at(idx);
572 ptr::copy_memory(self.vec.as_mut_ptr().offset(idx as int),
573 self.vec.as_ptr().offset(next as int),
575 self.vec.set_len(len - (next - idx));
580 /// Insert a character into the string buffer at byte position `idx`.
584 /// This is an O(n) operation as it requires copying every element in the
589 /// If `idx` does not lie on a character boundary or is out of bounds, then
590 /// this function will panic.
592 #[stable(feature = "rust1", since = "1.0.0")]
593 pub fn insert(&mut self, idx: uint, ch: char) {
594 let len = self.len();
596 assert!(self.is_char_boundary(idx));
598 let mut bits = [0; 4];
599 let amt = ch.encode_utf8(&mut bits).unwrap();
602 ptr::copy_memory(self.vec.as_mut_ptr().offset((idx + amt) as int),
603 self.vec.as_ptr().offset(idx as int),
605 ptr::copy_memory(self.vec.as_mut_ptr().offset(idx as int),
608 self.vec.set_len(len + amt);
612 /// Views the string buffer as a mutable sequence of bytes.
614 /// This is unsafe because it does not check
615 /// to ensure that the resulting string will be valid UTF-8.
620 /// let mut s = String::from_str("hello");
622 /// let vec = s.as_mut_vec();
623 /// assert!(vec == &mut vec![104, 101, 108, 108, 111]);
626 /// assert_eq!(s.as_slice(), "olleh");
629 #[stable(feature = "rust1", since = "1.0.0")]
630 pub unsafe fn as_mut_vec<'a>(&'a mut self) -> &'a mut Vec<u8> {
634 /// Return the number of bytes in this string.
639 /// let a = "foo".to_string();
640 /// assert_eq!(a.len(), 3);
643 #[stable(feature = "rust1", since = "1.0.0")]
644 pub fn len(&self) -> uint { self.vec.len() }
646 /// Returns true if the string contains no bytes
651 /// let mut v = String::new();
652 /// assert!(v.is_empty());
654 /// assert!(!v.is_empty());
657 #[stable(feature = "rust1", since = "1.0.0")]
658 pub fn is_empty(&self) -> bool { self.len() == 0 }
660 /// Truncates the string, returning it to 0 length.
665 /// let mut s = "foo".to_string();
667 /// assert!(s.is_empty());
670 #[stable(feature = "rust1", since = "1.0.0")]
671 pub fn clear(&mut self) {
677 /// Consume this error, returning the bytes that were attempted to make a
679 #[stable(feature = "rust1", since = "1.0.0")]
680 pub fn into_bytes(self) -> Vec<u8> { self.bytes }
682 /// Access the underlying UTF8-error that was the cause of this error.
683 #[stable(feature = "rust1", since = "1.0.0")]
684 pub fn utf8_error(&self) -> Utf8Error { self.error }
687 #[stable(feature = "rust1", since = "1.0.0")]
688 impl fmt::Display for FromUtf8Error {
689 fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
690 fmt::Display::fmt(&self.error, f)
694 #[stable(feature = "rust1", since = "1.0.0")]
695 impl Error for FromUtf8Error {
696 fn description(&self) -> &str { "invalid utf-8" }
699 #[stable(feature = "rust1", since = "1.0.0")]
700 impl fmt::Display for FromUtf16Error {
701 fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
702 fmt::Display::fmt("invalid utf-16: lone surrogate found", f)
706 #[stable(feature = "rust1", since = "1.0.0")]
707 impl Error for FromUtf16Error {
708 fn description(&self) -> &str { "invalid utf-16" }
711 #[stable(feature = "rust1", since = "1.0.0")]
712 impl FromIterator<char> for String {
713 fn from_iter<I:Iterator<Item=char>>(iterator: I) -> String {
714 let mut buf = String::new();
715 buf.extend(iterator);
720 #[stable(feature = "rust1", since = "1.0.0")]
721 impl<'a> FromIterator<&'a str> for String {
722 fn from_iter<I:Iterator<Item=&'a str>>(iterator: I) -> String {
723 let mut buf = String::new();
724 buf.extend(iterator);
729 #[unstable(feature = "collections",
730 reason = "waiting on Extend stabilization")]
731 impl Extend<char> for String {
732 fn extend<I:Iterator<Item=char>>(&mut self, iterator: I) {
733 let (lower_bound, _) = iterator.size_hint();
734 self.reserve(lower_bound);
741 #[unstable(feature = "collections",
742 reason = "waiting on Extend stabilization")]
743 impl<'a> Extend<&'a str> for String {
744 fn extend<I: Iterator<Item=&'a str>>(&mut self, iterator: I) {
745 // A guess that at least one byte per iterator element will be needed.
746 let (lower_bound, _) = iterator.size_hint();
747 self.reserve(lower_bound);
754 #[stable(feature = "rust1", since = "1.0.0")]
755 impl PartialEq for String {
757 fn eq(&self, other: &String) -> bool { PartialEq::eq(&**self, &**other) }
759 fn ne(&self, other: &String) -> bool { PartialEq::ne(&**self, &**other) }
762 macro_rules! impl_eq {
763 ($lhs:ty, $rhs: ty) => {
764 #[stable(feature = "rust1", since = "1.0.0")]
765 impl<'a> PartialEq<$rhs> for $lhs {
767 fn eq(&self, other: &$rhs) -> bool { PartialEq::eq(&**self, &**other) }
769 fn ne(&self, other: &$rhs) -> bool { PartialEq::ne(&**self, &**other) }
772 #[stable(feature = "rust1", since = "1.0.0")]
773 impl<'a> PartialEq<$lhs> for $rhs {
775 fn eq(&self, other: &$lhs) -> bool { PartialEq::eq(&**self, &**other) }
777 fn ne(&self, other: &$lhs) -> bool { PartialEq::ne(&**self, &**other) }
783 impl_eq! { String, &'a str }
784 impl_eq! { CowString<'a>, String }
786 #[stable(feature = "rust1", since = "1.0.0")]
787 impl<'a, 'b> PartialEq<&'b str> for CowString<'a> {
789 fn eq(&self, other: &&'b str) -> bool { PartialEq::eq(&**self, &**other) }
791 fn ne(&self, other: &&'b str) -> bool { PartialEq::ne(&**self, &**other) }
794 #[stable(feature = "rust1", since = "1.0.0")]
795 impl<'a, 'b> PartialEq<CowString<'a>> for &'b str {
797 fn eq(&self, other: &CowString<'a>) -> bool { PartialEq::eq(&**self, &**other) }
799 fn ne(&self, other: &CowString<'a>) -> bool { PartialEq::ne(&**self, &**other) }
802 #[unstable(feature = "collections", reason = "waiting on Str stabilization")]
803 impl Str for String {
805 #[stable(feature = "rust1", since = "1.0.0")]
806 fn as_slice<'a>(&'a self) -> &'a str {
807 unsafe { mem::transmute(self.vec.as_slice()) }
811 #[stable(feature = "rust1", since = "1.0.0")]
812 impl Default for String {
814 #[stable(feature = "rust1", since = "1.0.0")]
815 fn default() -> String {
820 #[stable(feature = "rust1", since = "1.0.0")]
821 impl fmt::Display for String {
823 fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
824 fmt::Display::fmt(&**self, f)
828 #[stable(feature = "rust1", since = "1.0.0")]
829 impl fmt::Debug for String {
831 fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
832 fmt::Debug::fmt(&**self, f)
836 #[unstable(feature = "collections", reason = "waiting on Hash stabilization")]
837 impl<H: hash::Writer + hash::Hasher> hash::Hash<H> for String {
839 fn hash(&self, hasher: &mut H) {
840 (**self).hash(hasher)
844 #[unstable(feature = "collections",
845 reason = "recent addition, needs more experience")]
846 impl<'a> Add<&'a str> for String {
847 type Output = String;
850 fn add(mut self, other: &str) -> String {
851 self.push_str(other);
856 #[stable(feature = "rust1", since = "1.0.0")]
857 impl ops::Index<ops::Range<uint>> for String {
860 fn index(&self, index: &ops::Range<uint>) -> &str {
864 #[stable(feature = "rust1", since = "1.0.0")]
865 impl ops::Index<ops::RangeTo<uint>> for String {
868 fn index(&self, index: &ops::RangeTo<uint>) -> &str {
872 #[stable(feature = "rust1", since = "1.0.0")]
873 impl ops::Index<ops::RangeFrom<uint>> for String {
876 fn index(&self, index: &ops::RangeFrom<uint>) -> &str {
880 #[stable(feature = "rust1", since = "1.0.0")]
881 impl ops::Index<ops::RangeFull> for String {
884 fn index(&self, _index: &ops::RangeFull) -> &str {
885 unsafe { mem::transmute(self.vec.as_slice()) }
889 #[stable(feature = "rust1", since = "1.0.0")]
890 impl ops::Deref for String {
894 fn deref<'a>(&'a self) -> &'a str {
895 unsafe { mem::transmute(&self.vec[]) }
899 /// Wrapper type providing a `&String` reference via `Deref`.
900 #[unstable(feature = "collections")]
901 pub struct DerefString<'a> {
905 impl<'a> Deref for DerefString<'a> {
906 type Target = String;
909 fn deref<'b>(&'b self) -> &'b String {
910 unsafe { mem::transmute(&*self.x) }
914 /// Convert a string slice to a wrapper type providing a `&String` reference.
919 /// use std::string::as_string;
921 /// fn string_consumer(s: String) {
922 /// assert_eq!(s, "foo".to_string());
925 /// let string = as_string("foo").clone();
926 /// string_consumer(string);
928 #[unstable(feature = "collections")]
929 pub fn as_string<'a>(x: &'a str) -> DerefString<'a> {
930 DerefString { x: as_vec(x.as_bytes()) }
933 #[unstable(feature = "collections", reason = "associated error type may change")]
934 impl FromStr for String {
937 fn from_str(s: &str) -> Result<String, ()> {
938 Ok(String::from_str(s))
942 /// A generic trait for converting a value to a string
943 #[stable(feature = "rust1", since = "1.0.0")]
945 /// Converts the value of `self` to an owned string
946 #[stable(feature = "rust1", since = "1.0.0")]
947 fn to_string(&self) -> String;
950 #[stable(feature = "rust1", since = "1.0.0")]
951 impl<T: fmt::Display + ?Sized> ToString for T {
953 fn to_string(&self) -> String {
954 use core::fmt::Writer;
955 let mut buf = String::new();
956 let _ = buf.write_fmt(format_args!("{}", self));
962 impl IntoCow<'static, String, str> for String {
964 fn into_cow(self) -> CowString<'static> {
969 impl<'a> IntoCow<'a, String, str> for &'a str {
971 fn into_cow(self) -> CowString<'a> {
976 /// A clone-on-write string
977 #[stable(feature = "rust1", since = "1.0.0")]
978 pub type CowString<'a> = Cow<'a, String, str>;
980 impl<'a> Str for CowString<'a> {
982 fn as_slice<'b>(&'b self) -> &'b str {
987 #[stable(feature = "rust1", since = "1.0.0")]
988 impl fmt::Writer for String {
990 fn write_str(&mut self, s: &str) -> fmt::Result {
1002 use core::iter::repeat;
1003 use super::{as_string, CowString};
1006 fn test_as_string() {
1008 assert_eq!(x, as_string(x).as_slice());
1012 fn test_from_str() {
1013 let owned: Option<::std::string::String> = "string".parse().ok();
1014 assert_eq!(owned.as_ref().map(|s| s.as_slice()), Some("string"));
1018 fn test_unsized_to_string() {
1019 let s: &str = "abc";
1020 let _: String = (*s).to_string();
1024 fn test_from_utf8() {
1025 let xs = b"hello".to_vec();
1026 assert_eq!(String::from_utf8(xs).unwrap(),
1027 String::from_str("hello"));
1029 let xs = "ศไทย中华Việt Nam".as_bytes().to_vec();
1030 assert_eq!(String::from_utf8(xs).unwrap(),
1031 String::from_str("ศไทย中华Việt Nam"));
1033 let xs = b"hello\xFF".to_vec();
1034 let err = String::from_utf8(xs).err().unwrap();
1035 assert_eq!(err.utf8_error(), Utf8Error::TooShort);
1036 assert_eq!(err.into_bytes(), b"hello\xff".to_vec());
1040 fn test_from_utf8_lossy() {
1042 let ys: CowString = "hello".into_cow();
1043 assert_eq!(String::from_utf8_lossy(xs), ys);
1045 let xs = "ศไทย中华Việt Nam".as_bytes();
1046 let ys: CowString = "ศไทย中华Việt Nam".into_cow();
1047 assert_eq!(String::from_utf8_lossy(xs), ys);
1049 let xs = b"Hello\xC2 There\xFF Goodbye";
1050 assert_eq!(String::from_utf8_lossy(xs),
1051 String::from_str("Hello\u{FFFD} There\u{FFFD} Goodbye").into_cow());
1053 let xs = b"Hello\xC0\x80 There\xE6\x83 Goodbye";
1054 assert_eq!(String::from_utf8_lossy(xs),
1055 String::from_str("Hello\u{FFFD}\u{FFFD} There\u{FFFD} Goodbye").into_cow());
1057 let xs = b"\xF5foo\xF5\x80bar";
1058 assert_eq!(String::from_utf8_lossy(xs),
1059 String::from_str("\u{FFFD}foo\u{FFFD}\u{FFFD}bar").into_cow());
1061 let xs = b"\xF1foo\xF1\x80bar\xF1\x80\x80baz";
1062 assert_eq!(String::from_utf8_lossy(xs),
1063 String::from_str("\u{FFFD}foo\u{FFFD}bar\u{FFFD}baz").into_cow());
1065 let xs = b"\xF4foo\xF4\x80bar\xF4\xBFbaz";
1066 assert_eq!(String::from_utf8_lossy(xs),
1067 String::from_str("\u{FFFD}foo\u{FFFD}bar\u{FFFD}\u{FFFD}baz").into_cow());
1069 let xs = b"\xF0\x80\x80\x80foo\xF0\x90\x80\x80bar";
1070 assert_eq!(String::from_utf8_lossy(xs), String::from_str("\u{FFFD}\u{FFFD}\u{FFFD}\u{FFFD}\
1071 foo\u{10000}bar").into_cow());
1074 let xs = b"\xED\xA0\x80foo\xED\xBF\xBFbar";
1075 assert_eq!(String::from_utf8_lossy(xs), String::from_str("\u{FFFD}\u{FFFD}\u{FFFD}foo\
1076 \u{FFFD}\u{FFFD}\u{FFFD}bar").into_cow());
1080 fn test_from_utf16() {
1082 [(String::from_str("𐍅𐌿𐌻𐍆𐌹𐌻𐌰\n"),
1083 vec![0xd800_u16, 0xdf45_u16, 0xd800_u16, 0xdf3f_u16,
1084 0xd800_u16, 0xdf3b_u16, 0xd800_u16, 0xdf46_u16,
1085 0xd800_u16, 0xdf39_u16, 0xd800_u16, 0xdf3b_u16,
1086 0xd800_u16, 0xdf30_u16, 0x000a_u16]),
1088 (String::from_str("𐐒𐑉𐐮𐑀𐐲𐑋 𐐏𐐲𐑍\n"),
1089 vec![0xd801_u16, 0xdc12_u16, 0xd801_u16,
1090 0xdc49_u16, 0xd801_u16, 0xdc2e_u16, 0xd801_u16,
1091 0xdc40_u16, 0xd801_u16, 0xdc32_u16, 0xd801_u16,
1092 0xdc4b_u16, 0x0020_u16, 0xd801_u16, 0xdc0f_u16,
1093 0xd801_u16, 0xdc32_u16, 0xd801_u16, 0xdc4d_u16,
1096 (String::from_str("𐌀𐌖𐌋𐌄𐌑𐌉·𐌌𐌄𐌕𐌄𐌋𐌉𐌑\n"),
1097 vec![0xd800_u16, 0xdf00_u16, 0xd800_u16, 0xdf16_u16,
1098 0xd800_u16, 0xdf0b_u16, 0xd800_u16, 0xdf04_u16,
1099 0xd800_u16, 0xdf11_u16, 0xd800_u16, 0xdf09_u16,
1100 0x00b7_u16, 0xd800_u16, 0xdf0c_u16, 0xd800_u16,
1101 0xdf04_u16, 0xd800_u16, 0xdf15_u16, 0xd800_u16,
1102 0xdf04_u16, 0xd800_u16, 0xdf0b_u16, 0xd800_u16,
1103 0xdf09_u16, 0xd800_u16, 0xdf11_u16, 0x000a_u16 ]),
1105 (String::from_str("𐒋𐒘𐒈𐒑𐒛𐒒 𐒕𐒓 𐒈𐒚𐒍 𐒏𐒜𐒒𐒖𐒆 𐒕𐒆\n"),
1106 vec![0xd801_u16, 0xdc8b_u16, 0xd801_u16, 0xdc98_u16,
1107 0xd801_u16, 0xdc88_u16, 0xd801_u16, 0xdc91_u16,
1108 0xd801_u16, 0xdc9b_u16, 0xd801_u16, 0xdc92_u16,
1109 0x0020_u16, 0xd801_u16, 0xdc95_u16, 0xd801_u16,
1110 0xdc93_u16, 0x0020_u16, 0xd801_u16, 0xdc88_u16,
1111 0xd801_u16, 0xdc9a_u16, 0xd801_u16, 0xdc8d_u16,
1112 0x0020_u16, 0xd801_u16, 0xdc8f_u16, 0xd801_u16,
1113 0xdc9c_u16, 0xd801_u16, 0xdc92_u16, 0xd801_u16,
1114 0xdc96_u16, 0xd801_u16, 0xdc86_u16, 0x0020_u16,
1115 0xd801_u16, 0xdc95_u16, 0xd801_u16, 0xdc86_u16,
1117 // Issue #12318, even-numbered non-BMP planes
1118 (String::from_str("\u{20000}"),
1119 vec![0xD840, 0xDC00])];
1122 let (s, u) = (*p).clone();
1123 let s_as_utf16 = s.utf16_units().collect::<Vec<u16>>();
1124 let u_as_string = String::from_utf16(u.as_slice()).unwrap();
1126 assert!(::unicode::str::is_utf16(u.as_slice()));
1127 assert_eq!(s_as_utf16, u);
1129 assert_eq!(u_as_string, s);
1130 assert_eq!(String::from_utf16_lossy(u.as_slice()), s);
1132 assert_eq!(String::from_utf16(s_as_utf16.as_slice()).unwrap(), s);
1133 assert_eq!(u_as_string.utf16_units().collect::<Vec<u16>>(), u);
1138 fn test_utf16_invalid() {
1139 // completely positive cases tested above.
1141 assert!(String::from_utf16(&[0xD800]).is_err());
1143 assert!(String::from_utf16(&[0xD800, 0xD800]).is_err());
1146 assert!(String::from_utf16(&[0x0061, 0xDC00]).is_err());
1149 assert!(String::from_utf16(&[0xD800, 0xd801, 0xdc8b, 0xD800]).is_err());
1153 fn test_from_utf16_lossy() {
1154 // completely positive cases tested above.
1156 assert_eq!(String::from_utf16_lossy(&[0xD800]), String::from_str("\u{FFFD}"));
1158 assert_eq!(String::from_utf16_lossy(&[0xD800, 0xD800]),
1159 String::from_str("\u{FFFD}\u{FFFD}"));
1162 assert_eq!(String::from_utf16_lossy(&[0x0061, 0xDC00]), String::from_str("a\u{FFFD}"));
1165 assert_eq!(String::from_utf16_lossy(&[0xD800, 0xd801, 0xdc8b, 0xD800]),
1166 String::from_str("\u{FFFD}𐒋\u{FFFD}"));
1170 fn test_push_bytes() {
1171 let mut s = String::from_str("ABC");
1173 let mv = s.as_mut_vec();
1174 mv.push_all(&[b'D']);
1176 assert_eq!(s, "ABCD");
1180 fn test_push_str() {
1181 let mut s = String::new();
1183 assert_eq!(&s[0..], "");
1185 assert_eq!(&s[0..], "abc");
1186 s.push_str("ประเทศไทย中华Việt Nam");
1187 assert_eq!(&s[0..], "abcประเทศไทย中华Việt Nam");
1192 let mut data = String::from_str("ประเทศไทย中");
1194 data.push('b'); // 1 byte
1195 data.push('¢'); // 2 byte
1196 data.push('€'); // 3 byte
1197 data.push('𤭢'); // 4 byte
1198 assert_eq!(data, "ประเทศไทย中华b¢€𤭢");
1203 let mut data = String::from_str("ประเทศไทย中华b¢€𤭢");
1204 assert_eq!(data.pop().unwrap(), '𤭢'); // 4 bytes
1205 assert_eq!(data.pop().unwrap(), '€'); // 3 bytes
1206 assert_eq!(data.pop().unwrap(), '¢'); // 2 bytes
1207 assert_eq!(data.pop().unwrap(), 'b'); // 1 bytes
1208 assert_eq!(data.pop().unwrap(), '华');
1209 assert_eq!(data, "ประเทศไทย中");
1213 fn test_str_truncate() {
1214 let mut s = String::from_str("12345");
1216 assert_eq!(s, "12345");
1218 assert_eq!(s, "123");
1222 let mut s = String::from_str("12345");
1226 let p_ = s.as_ptr();
1232 fn test_str_truncate_invalid_len() {
1233 let mut s = String::from_str("12345");
1239 fn test_str_truncate_split_codepoint() {
1240 let mut s = String::from_str("\u{FC}"); // ü
1245 fn test_str_clear() {
1246 let mut s = String::from_str("12345");
1248 assert_eq!(s.len(), 0);
1254 let a = String::from_str("12345");
1257 assert_eq!(b.len(), 7);
1258 assert_eq!(b, "1234522");
1263 let mut s = "ศไทย中华Việt Nam; foobar".to_string();;
1264 assert_eq!(s.remove(0), 'ศ');
1265 assert_eq!(s.len(), 33);
1266 assert_eq!(s, "ไทย中华Việt Nam; foobar");
1267 assert_eq!(s.remove(17), 'ệ');
1268 assert_eq!(s, "ไทย中华Vit Nam; foobar");
1271 #[test] #[should_fail]
1273 "ศ".to_string().remove(1);
1278 let mut s = "foobar".to_string();
1280 assert_eq!(s, "ệfoobar");
1282 assert_eq!(s, "ệfooยbar");
1285 #[test] #[should_fail] fn insert_bad1() { "".to_string().insert(1, 't'); }
1286 #[test] #[should_fail] fn insert_bad2() { "ệ".to_string().insert(1, 't'); }
1290 let s = "foobar".to_string();
1291 assert_eq!("foobar", &s[]);
1292 assert_eq!("foo", &s[..3]);
1293 assert_eq!("bar", &s[3..]);
1294 assert_eq!("oob", &s[1..4]);
1298 fn test_simple_types() {
1299 assert_eq!(1.to_string(), "1");
1300 assert_eq!((-1).to_string(), "-1");
1301 assert_eq!(200u.to_string(), "200");
1302 assert_eq!(2u8.to_string(), "2");
1303 assert_eq!(true.to_string(), "true");
1304 assert_eq!(false.to_string(), "false");
1305 assert_eq!(("hi".to_string()).to_string(), "hi");
1310 let x: Vec<int> = vec![];
1311 assert_eq!(format!("{:?}", x), "[]");
1312 assert_eq!(format!("{:?}", vec![1]), "[1]");
1313 assert_eq!(format!("{:?}", vec![1, 2, 3]), "[1, 2, 3]");
1314 assert!(format!("{:?}", vec![vec![], vec![1], vec![1, 1]]) ==
1315 "[[], [1], [1, 1]]");
1319 fn test_from_iterator() {
1320 let s = "ศไทย中华Việt Nam".to_string();
1324 let a: String = s.chars().collect();
1327 let mut b = t.to_string();
1328 b.extend(u.chars());
1331 let c: String = vec![t, u].into_iter().collect();
1334 let mut d = t.to_string();
1335 d.extend(vec![u].into_iter());
1340 fn bench_with_capacity(b: &mut Bencher) {
1342 String::with_capacity(100)
1347 fn bench_push_str(b: &mut Bencher) {
1348 let s = "ศไทย中华Việt Nam; Mary had a little lamb, Little lamb";
1350 let mut r = String::new();
1355 const REPETITIONS: u64 = 10_000;
1358 fn bench_push_str_one_byte(b: &mut Bencher) {
1359 b.bytes = REPETITIONS;
1361 let mut r = String::new();
1362 for _ in 0..REPETITIONS {
1369 fn bench_push_char_one_byte(b: &mut Bencher) {
1370 b.bytes = REPETITIONS;
1372 let mut r = String::new();
1373 for _ in 0..REPETITIONS {
1380 fn bench_push_char_two_bytes(b: &mut Bencher) {
1381 b.bytes = REPETITIONS * 2;
1383 let mut r = String::new();
1384 for _ in 0..REPETITIONS {
1391 fn from_utf8_lossy_100_ascii(b: &mut Bencher) {
1392 let s = b"Hello there, the quick brown fox jumped over the lazy dog! \
1393 Lorem ipsum dolor sit amet, consectetur. ";
1395 assert_eq!(100, s.len());
1397 let _ = String::from_utf8_lossy(s);
1402 fn from_utf8_lossy_100_multibyte(b: &mut Bencher) {
1403 let s = "𐌀𐌖𐌋𐌄𐌑𐌉ปรدولة الكويتทศไทย中华𐍅𐌿𐌻𐍆𐌹𐌻𐌰".as_bytes();
1404 assert_eq!(100, s.len());
1406 let _ = String::from_utf8_lossy(s);
1411 fn from_utf8_lossy_invalid(b: &mut Bencher) {
1412 let s = b"Hello\xC0\x80 There\xE6\x83 Goodbye";
1414 let _ = String::from_utf8_lossy(s);
1419 fn from_utf8_lossy_100_invalid(b: &mut Bencher) {
1420 let s = repeat(0xf5u8).take(100).collect::<Vec<_>>();
1422 let _ = String::from_utf8_lossy(s.as_slice());
1427 fn bench_exact_size_shrink_to_fit(b: &mut Bencher) {
1428 let s = "Hello there, the quick brown fox jumped over the lazy dog! \
1429 Lorem ipsum dolor sit amet, consectetur. ";
1430 // ensure our operation produces an exact-size string before we benchmark it
1431 let mut r = String::with_capacity(s.len());
1433 assert_eq!(r.len(), r.capacity());
1435 let mut r = String::with_capacity(s.len());