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.
19 use core::borrow::{Cow, IntoCow};
20 use core::default::Default;
26 use core::raw::Slice as RawSlice;
27 use unicode::str as unicode_str;
28 use unicode::str::Utf16Item;
30 use str::{mod, CharRange, FromStr, Utf8Error};
31 use vec::{DerefVec, Vec, as_vec};
33 /// A growable string stored as a UTF-8 encoded buffer.
34 #[deriving(Clone, PartialOrd, Eq, Ord)]
40 /// A possible error value from the `String::from_utf8` function.
42 pub struct FromUtf8Error {
47 /// A possible error value from the `String::from_utf16` function.
49 #[allow(missing_copy_implementations)]
50 pub struct FromUtf16Error(());
53 /// Creates a new string buffer initialized with the empty string.
58 /// let mut s = String::new();
62 pub fn new() -> String {
68 /// Creates a new string buffer with the given capacity.
69 /// The string will be able to hold exactly `capacity` bytes without
70 /// reallocating. If `capacity` is 0, the string will not allocate.
75 /// let mut s = String::with_capacity(10);
79 pub fn with_capacity(capacity: uint) -> String {
81 vec: Vec::with_capacity(capacity),
85 /// Creates a new string buffer from the given string.
90 /// let s = String::from_str("hello");
91 /// assert_eq!(s.as_slice(), "hello");
94 #[experimental = "needs investigation to see if to_string() can match perf"]
95 pub fn from_str(string: &str) -> String {
96 String { vec: ::slice::SliceExt::to_vec(string.as_bytes()) }
99 /// Returns the vector as a string buffer, if possible, taking care not to
104 /// If the given vector is not valid UTF-8, then the original vector and the
105 /// corresponding error is returned.
110 /// # #![allow(deprecated)]
111 /// use std::str::Utf8Error;
113 /// let hello_vec = vec![104, 101, 108, 108, 111];
114 /// let s = String::from_utf8(hello_vec).unwrap();
115 /// assert_eq!(s, "hello");
117 /// let invalid_vec = vec![240, 144, 128];
118 /// let s = String::from_utf8(invalid_vec).err().unwrap();
119 /// assert_eq!(s.utf8_error(), Utf8Error::TooShort);
120 /// assert_eq!(s.into_bytes(), vec![240, 144, 128]);
124 pub fn from_utf8(vec: Vec<u8>) -> Result<String, FromUtf8Error> {
125 match str::from_utf8(vec.as_slice()) {
126 Ok(..) => Ok(String { vec: vec }),
127 Err(e) => Err(FromUtf8Error { bytes: vec, error: e })
131 /// Converts a vector of bytes to a new UTF-8 string.
132 /// Any invalid UTF-8 sequences are replaced with U+FFFD REPLACEMENT CHARACTER.
137 /// let input = b"Hello \xF0\x90\x80World";
138 /// let output = String::from_utf8_lossy(input);
139 /// assert_eq!(output.as_slice(), "Hello \u{FFFD}World");
142 pub fn from_utf8_lossy<'a>(v: &'a [u8]) -> CowString<'a> {
143 match str::from_utf8(v) {
144 Ok(s) => return Cow::Borrowed(s),
148 static TAG_CONT_U8: u8 = 128u8;
149 static REPLACEMENT: &'static [u8] = b"\xEF\xBF\xBD"; // U+FFFD in UTF-8
152 fn unsafe_get(xs: &[u8], i: uint) -> u8 {
153 unsafe { *xs.get_unchecked(i) }
155 fn safe_get(xs: &[u8], i: uint, total: uint) -> u8 {
163 let mut res = String::with_capacity(total);
167 res.as_mut_vec().push_all(v[..i])
171 // subseqidx is the index of the first byte of the subsequence we're looking at.
172 // It's used to copy a bunch of contiguous good codepoints at once instead of copying
174 let mut subseqidx = 0;
178 let byte = unsafe_get(v, i);
181 macro_rules! error(() => ({
184 res.as_mut_vec().push_all(v[subseqidx..i_]);
187 res.as_mut_vec().push_all(REPLACEMENT);
192 // subseqidx handles this
194 let w = unicode_str::utf8_char_width(byte);
198 if safe_get(v, i, total) & 192u8 != TAG_CONT_U8 {
205 match (byte, safe_get(v, i, total)) {
206 (0xE0 , 0xA0 ... 0xBF) => (),
207 (0xE1 ... 0xEC, 0x80 ... 0xBF) => (),
208 (0xED , 0x80 ... 0x9F) => (),
209 (0xEE ... 0xEF, 0x80 ... 0xBF) => (),
216 if safe_get(v, i, total) & 192u8 != TAG_CONT_U8 {
223 match (byte, safe_get(v, i, total)) {
224 (0xF0 , 0x90 ... 0xBF) => (),
225 (0xF1 ... 0xF3, 0x80 ... 0xBF) => (),
226 (0xF4 , 0x80 ... 0x8F) => (),
233 if safe_get(v, i, total) & 192u8 != TAG_CONT_U8 {
238 if safe_get(v, i, total) & 192u8 != TAG_CONT_U8 {
251 if subseqidx < total {
253 res.as_mut_vec().push_all(v[subseqidx..total])
259 /// Decode a UTF-16 encoded vector `v` into a `String`, returning `None`
260 /// if `v` contains any invalid data.
266 /// let mut v = &mut [0xD834, 0xDD1E, 0x006d, 0x0075,
267 /// 0x0073, 0x0069, 0x0063];
268 /// assert_eq!(String::from_utf16(v).unwrap(),
269 /// "𝄞music".to_string());
271 /// // 𝄞mu<invalid>ic
273 /// assert!(String::from_utf16(v).is_err());
276 pub fn from_utf16(v: &[u16]) -> Result<String, FromUtf16Error> {
277 let mut s = String::with_capacity(v.len());
278 for c in unicode_str::utf16_items(v) {
280 Utf16Item::ScalarValue(c) => s.push(c),
281 Utf16Item::LoneSurrogate(_) => return Err(FromUtf16Error(())),
287 /// Decode a UTF-16 encoded vector `v` into a string, replacing
288 /// invalid data with the replacement character (U+FFFD).
293 /// // 𝄞mus<invalid>ic<invalid>
294 /// let v = &[0xD834, 0xDD1E, 0x006d, 0x0075,
295 /// 0x0073, 0xDD1E, 0x0069, 0x0063,
298 /// assert_eq!(String::from_utf16_lossy(v),
299 /// "𝄞mus\u{FFFD}ic\u{FFFD}".to_string());
302 pub fn from_utf16_lossy(v: &[u16]) -> String {
303 unicode_str::utf16_items(v).map(|c| c.to_char_lossy()).collect()
306 /// Convert a vector of `char`s to a `String`.
311 /// # #![allow(deprecated)]
312 /// let chars = &['h', 'e', 'l', 'l', 'o'];
313 /// let s = String::from_chars(chars);
314 /// assert_eq!(s.as_slice(), "hello");
317 #[deprecated = "use .collect() instead"]
318 pub fn from_chars(chs: &[char]) -> String {
319 chs.iter().map(|c| *c).collect()
322 /// Creates a new `String` from a length, capacity, and pointer.
324 /// This is unsafe because:
325 /// * We call `Vec::from_raw_parts` to get a `Vec<u8>`;
326 /// * We assume that the `Vec` contains valid UTF-8.
329 pub unsafe fn from_raw_parts(buf: *mut u8, length: uint, capacity: uint) -> String {
331 vec: Vec::from_raw_parts(buf, length, capacity),
335 /// Creates a `String` from a null-terminated `*const u8` buffer.
337 /// This function is unsafe because we dereference memory until we find the
338 /// NUL character, which is not guaranteed to be present. Additionally, the
339 /// slice is not checked to see whether it contains valid UTF-8
340 #[unstable = "just renamed from `mod raw`"]
341 pub unsafe fn from_raw_buf(buf: *const u8) -> String {
342 String::from_str(str::from_c_str(buf as *const i8))
345 /// Creates a `String` from a `*const u8` buffer of the given length.
347 /// This function is unsafe because it blindly assumes the validity of the
348 /// pointer `buf` for `len` bytes of memory. This function will copy the
349 /// memory from `buf` into a new allocation (owned by the returned
352 /// This function is also unsafe because it does not validate that the
353 /// buffer is valid UTF-8 encoded data.
354 #[unstable = "just renamed from `mod raw`"]
355 pub unsafe fn from_raw_buf_len(buf: *const u8, len: uint) -> String {
356 String::from_utf8_unchecked(Vec::from_raw_buf(buf, len))
359 /// Converts a vector of bytes to a new `String` without checking if
360 /// it contains valid UTF-8. This is unsafe because it assumes that
361 /// the UTF-8-ness of the vector has already been validated.
364 pub unsafe fn from_utf8_unchecked(bytes: Vec<u8>) -> String {
365 String { vec: bytes }
368 /// Return the underlying byte buffer, encoded as UTF-8.
373 /// let s = String::from_str("hello");
374 /// let bytes = s.into_bytes();
375 /// assert_eq!(bytes, vec![104, 101, 108, 108, 111]);
379 pub fn into_bytes(self) -> Vec<u8> {
383 /// Creates a string buffer by repeating a character `length` times.
388 /// # #![allow(deprecated)]
389 /// let s = String::from_char(5, 'a');
390 /// assert_eq!(s.as_slice(), "aaaaa");
393 #[deprecated = "use repeat(ch).take(length).collect() instead"]
394 pub fn from_char(length: uint, ch: char) -> String {
399 let mut buf = String::new();
401 let size = buf.len() * (length - 1);
402 buf.reserve_exact(size);
403 for _ in range(1, length) {
409 /// Pushes the given string onto this string buffer.
414 /// let mut s = String::from_str("foo");
415 /// s.push_str("bar");
416 /// assert_eq!(s.as_slice(), "foobar");
420 pub fn push_str(&mut self, string: &str) {
421 self.vec.push_all(string.as_bytes())
424 /// Pushes `ch` onto the given string `count` times.
429 /// # #![allow(deprecated)]
430 /// let mut s = String::from_str("foo");
432 /// assert_eq!(s.as_slice(), "fooZZZZZ");
435 #[deprecated = "deprecated in favor of .extend(repeat(ch).take(count))"]
436 pub fn grow(&mut self, count: uint, ch: char) {
437 for _ in range(0, count) {
442 /// Returns the number of bytes that this string buffer can hold without
448 /// let s = String::with_capacity(10);
449 /// assert!(s.capacity() >= 10);
453 pub fn capacity(&self) -> uint {
457 /// Deprecated: Renamed to `reserve`.
458 #[deprecated = "Renamed to `reserve`"]
459 pub fn reserve_additional(&mut self, extra: uint) {
460 self.vec.reserve(extra)
463 /// Reserves capacity for at least `additional` more bytes to be inserted
464 /// in the given `String`. The collection may reserve more space to avoid
465 /// frequent reallocations.
469 /// Panics if the new capacity overflows `uint`.
474 /// let mut s = String::new();
476 /// assert!(s.capacity() >= 10);
480 pub fn reserve(&mut self, additional: uint) {
481 self.vec.reserve(additional)
484 /// Reserves the minimum capacity for exactly `additional` more bytes to be
485 /// inserted in the given `String`. Does nothing if the capacity is already
488 /// Note that the allocator may give the collection more space than it
489 /// requests. Therefore capacity can not be relied upon to be precisely
490 /// minimal. Prefer `reserve` if future insertions are expected.
494 /// Panics if the new capacity overflows `uint`.
499 /// let mut s = String::new();
501 /// assert!(s.capacity() >= 10);
505 pub fn reserve_exact(&mut self, additional: uint) {
506 self.vec.reserve_exact(additional)
509 /// Shrinks the capacity of this string buffer to match its length.
514 /// let mut s = String::from_str("foo");
516 /// assert!(s.capacity() >= 100);
517 /// s.shrink_to_fit();
518 /// assert_eq!(s.capacity(), 3);
522 pub fn shrink_to_fit(&mut self) {
523 self.vec.shrink_to_fit()
526 /// Adds the given character to the end of the string.
531 /// let mut s = String::from_str("abc");
535 /// assert_eq!(s.as_slice(), "abc123");
539 pub fn push(&mut self, ch: char) {
540 if (ch as u32) < 0x80 {
541 self.vec.push(ch as u8);
545 let cur_len = self.len();
546 // This may use up to 4 bytes.
550 // Attempt to not use an intermediate buffer by just pushing bytes
551 // directly onto this string.
552 let slice = RawSlice {
553 data: self.vec.as_ptr().offset(cur_len as int),
556 let used = ch.encode_utf8(mem::transmute(slice)).unwrap_or(0);
557 self.vec.set_len(cur_len + used);
561 /// Works with the underlying buffer as a byte slice.
566 /// let s = String::from_str("hello");
567 /// let b: &[_] = &[104, 101, 108, 108, 111];
568 /// assert_eq!(s.as_bytes(), b);
572 pub fn as_bytes<'a>(&'a self) -> &'a [u8] {
576 /// Shortens a string to the specified length.
580 /// Panics if `new_len` > current length,
581 /// or if `new_len` is not a character boundary.
586 /// let mut s = String::from_str("hello");
588 /// assert_eq!(s.as_slice(), "he");
592 pub fn truncate(&mut self, new_len: uint) {
593 assert!(self.is_char_boundary(new_len));
594 self.vec.truncate(new_len)
597 /// Removes the last character from the string buffer and returns it.
598 /// Returns `None` if this string buffer is empty.
603 /// let mut s = String::from_str("foo");
604 /// assert_eq!(s.pop(), Some('o'));
605 /// assert_eq!(s.pop(), Some('o'));
606 /// assert_eq!(s.pop(), Some('f'));
607 /// assert_eq!(s.pop(), None);
611 pub fn pop(&mut self) -> Option<char> {
612 let len = self.len();
617 let CharRange {ch, next} = self.char_range_at_reverse(len);
619 self.vec.set_len(next);
624 /// Removes the character from the string buffer at byte position `idx` and
629 /// This is an O(n) operation as it requires copying every element in the
634 /// If `idx` does not lie on a character boundary, or if it is out of
635 /// bounds, then this function will panic.
640 /// let mut s = String::from_str("foo");
641 /// assert_eq!(s.remove(0), 'f');
642 /// assert_eq!(s.remove(1), 'o');
643 /// assert_eq!(s.remove(0), 'o');
646 pub fn remove(&mut self, idx: uint) -> char {
647 let len = self.len();
650 let CharRange { ch, next } = self.char_range_at(idx);
652 ptr::copy_memory(self.vec.as_mut_ptr().offset(idx as int),
653 self.vec.as_ptr().offset(next as int),
655 self.vec.set_len(len - (next - idx));
660 /// Insert a character into the string buffer at byte position `idx`.
664 /// This is an O(n) operation as it requires copying every element in the
669 /// If `idx` does not lie on a character boundary or is out of bounds, then
670 /// this function will panic.
672 pub fn insert(&mut self, idx: uint, ch: char) {
673 let len = self.len();
675 assert!(self.is_char_boundary(idx));
677 let mut bits = [0; 4];
678 let amt = ch.encode_utf8(&mut bits).unwrap();
681 ptr::copy_memory(self.vec.as_mut_ptr().offset((idx + amt) as int),
682 self.vec.as_ptr().offset(idx as int),
684 ptr::copy_memory(self.vec.as_mut_ptr().offset(idx as int),
687 self.vec.set_len(len + amt);
691 /// Views the string buffer as a mutable sequence of bytes.
693 /// This is unsafe because it does not check
694 /// to ensure that the resulting string will be valid UTF-8.
699 /// let mut s = String::from_str("hello");
701 /// let vec = s.as_mut_vec();
702 /// assert!(vec == &mut vec![104, 101, 108, 108, 111]);
705 /// assert_eq!(s.as_slice(), "olleh");
708 pub unsafe fn as_mut_vec<'a>(&'a mut self) -> &'a mut Vec<u8> {
712 /// Return the number of bytes in this string.
717 /// let a = "foo".to_string();
718 /// assert_eq!(a.len(), 3);
722 pub fn len(&self) -> uint { self.vec.len() }
724 /// Returns true if the string contains no bytes
729 /// let mut v = String::new();
730 /// assert!(v.is_empty());
732 /// assert!(!v.is_empty());
735 pub fn is_empty(&self) -> bool { self.len() == 0 }
737 /// Truncates the string, returning it to 0 length.
742 /// let mut s = "foo".to_string();
744 /// assert!(s.is_empty());
748 pub fn clear(&mut self) {
754 /// Consume this error, returning the bytes that were attempted to make a
757 pub fn into_bytes(self) -> Vec<u8> { self.bytes }
759 /// Access the underlying UTF8-error that was the cause of this error.
761 pub fn utf8_error(&self) -> Utf8Error { self.error }
764 impl fmt::Show for FromUtf8Error {
765 fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
770 impl fmt::Show for FromUtf16Error {
771 fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
772 "invalid utf-16: lone surrogate found".fmt(f)
776 #[experimental = "waiting on FromIterator stabilization"]
777 impl FromIterator<char> for String {
778 fn from_iter<I:Iterator<char>>(iterator: I) -> String {
779 let mut buf = String::new();
780 buf.extend(iterator);
785 #[experimental = "waiting on FromIterator stabilization"]
786 impl<'a> FromIterator<&'a str> for String {
787 fn from_iter<I:Iterator<&'a str>>(iterator: I) -> String {
788 let mut buf = String::new();
789 buf.extend(iterator);
794 #[experimental = "waiting on Extend stabilization"]
795 impl Extend<char> for String {
796 fn extend<I:Iterator<char>>(&mut self, mut iterator: I) {
797 let (lower_bound, _) = iterator.size_hint();
798 self.reserve(lower_bound);
805 #[experimental = "waiting on Extend stabilization"]
806 impl<'a> Extend<&'a str> for String {
807 fn extend<I: Iterator<&'a str>>(&mut self, mut iterator: I) {
808 // A guess that at least one byte per iterator element will be needed.
809 let (lower_bound, _) = iterator.size_hint();
810 self.reserve(lower_bound);
818 impl PartialEq for String {
820 fn eq(&self, other: &String) -> bool { PartialEq::eq(&**self, &**other) }
822 fn ne(&self, other: &String) -> bool { PartialEq::ne(&**self, &**other) }
825 macro_rules! impl_eq {
826 ($lhs:ty, $rhs: ty) => {
828 impl<'a> PartialEq<$rhs> for $lhs {
830 fn eq(&self, other: &$rhs) -> bool { PartialEq::eq(&**self, &**other) }
832 fn ne(&self, other: &$rhs) -> bool { PartialEq::ne(&**self, &**other) }
836 impl<'a> PartialEq<$lhs> for $rhs {
838 fn eq(&self, other: &$lhs) -> bool { PartialEq::eq(&**self, &**other) }
840 fn ne(&self, other: &$lhs) -> bool { PartialEq::ne(&**self, &**other) }
846 impl_eq! { String, &'a str }
847 impl_eq! { CowString<'a>, String }
850 impl<'a, 'b> PartialEq<&'b str> for CowString<'a> {
852 fn eq(&self, other: &&'b str) -> bool { PartialEq::eq(&**self, &**other) }
854 fn ne(&self, other: &&'b str) -> bool { PartialEq::ne(&**self, &**other) }
858 impl<'a, 'b> PartialEq<CowString<'a>> for &'b str {
860 fn eq(&self, other: &CowString<'a>) -> bool { PartialEq::eq(&**self, &**other) }
862 fn ne(&self, other: &CowString<'a>) -> bool { PartialEq::ne(&**self, &**other) }
865 #[experimental = "waiting on Str stabilization"]
867 impl Str for String {
870 fn as_slice<'a>(&'a self) -> &'a str {
871 unsafe { mem::transmute(self.vec.as_slice()) }
876 impl Default for String {
878 fn default() -> String {
883 #[experimental = "waiting on Show stabilization"]
884 impl fmt::Show for String {
885 fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
890 #[experimental = "waiting on Hash stabilization"]
891 impl<H: hash::Writer> hash::Hash<H> for String {
893 fn hash(&self, hasher: &mut H) {
894 (**self).hash(hasher)
899 #[deprecated = "Use overloaded `core::cmp::PartialEq`"]
900 impl<'a, S: Str> Equiv<S> for String {
902 fn equiv(&self, other: &S) -> bool {
903 self.as_slice() == other.as_slice()
907 #[experimental = "waiting on Add stabilization"]
908 impl<'a> Add<&'a str, String> for String {
909 fn add(mut self, other: &str) -> String {
910 self.push_str(other);
915 impl ops::Slice<uint, str> for String {
917 fn as_slice_<'a>(&'a self) -> &'a str {
918 unsafe { mem::transmute(self.vec.as_slice()) }
922 fn slice_from_or_fail<'a>(&'a self, from: &uint) -> &'a str {
927 fn slice_to_or_fail<'a>(&'a self, to: &uint) -> &'a str {
932 fn slice_or_fail<'a>(&'a self, from: &uint, to: &uint) -> &'a str {
937 #[experimental = "waiting on Deref stabilization"]
938 impl ops::Deref<str> for String {
939 fn deref<'a>(&'a self) -> &'a str {
940 unsafe { mem::transmute(self.vec[]) }
944 /// Wrapper type providing a `&String` reference via `Deref`.
946 pub struct DerefString<'a> {
950 impl<'a> Deref<String> for DerefString<'a> {
951 fn deref<'b>(&'b self) -> &'b String {
952 unsafe { mem::transmute(&*self.x) }
956 /// Convert a string slice to a wrapper type providing a `&String` reference.
961 /// use std::string::as_string;
963 /// fn string_consumer(s: String) {
964 /// assert_eq!(s, "foo".to_string());
967 /// let string = as_string("foo").clone();
968 /// string_consumer(string);
971 pub fn as_string<'a>(x: &'a str) -> DerefString<'a> {
972 DerefString { x: as_vec(x.as_bytes()) }
975 impl FromStr for String {
977 fn from_str(s: &str) -> Option<String> {
978 Some(String::from_str(s))
982 /// Trait for converting a type to a string, consuming it in the process.
983 #[deprecated = "trait will be removed"]
984 pub trait IntoString {
985 /// Consume and convert to a string.
986 fn into_string(self) -> String;
989 /// A generic trait for converting a value to a string
991 /// Converts the value of `self` to an owned string
992 fn to_string(&self) -> String;
995 impl<T: fmt::Show> ToString for T {
996 fn to_string(&self) -> String {
997 let mut buf = Vec::<u8>::new();
998 let _ = fmt::write(&mut buf, format_args!("{}", *self));
999 String::from_utf8(buf).unwrap()
1003 impl IntoCow<'static, String, str> for String {
1004 fn into_cow(self) -> CowString<'static> {
1009 impl<'a> IntoCow<'a, String, str> for &'a str {
1010 fn into_cow(self) -> CowString<'a> {
1015 /// Unsafe operations
1021 /// Creates a new `String` from a length, capacity, and pointer.
1023 /// This is unsafe because:
1024 /// * We call `Vec::from_raw_parts` to get a `Vec<u8>`;
1025 /// * We assume that the `Vec` contains valid UTF-8.
1027 #[deprecated = "renamed to String::from_raw_parts"]
1028 pub unsafe fn from_parts(buf: *mut u8, length: uint, capacity: uint) -> String {
1029 String::from_raw_parts(buf, length, capacity)
1032 /// Creates a `String` from a `*const u8` buffer of the given length.
1034 /// This function is unsafe because of two reasons:
1036 /// * A raw pointer is dereferenced and transmuted to `&[u8]`;
1037 /// * The slice is not checked to see whether it contains valid UTF-8.
1038 #[deprecated = "renamed to String::from_raw_buf_len"]
1039 pub unsafe fn from_buf_len(buf: *const u8, len: uint) -> String {
1040 String::from_raw_buf_len(buf, len)
1043 /// Creates a `String` from a null-terminated `*const u8` buffer.
1045 /// This function is unsafe because we dereference memory until we find the NUL character,
1046 /// which is not guaranteed to be present. Additionally, the slice is not checked to see
1047 /// whether it contains valid UTF-8
1048 #[deprecated = "renamed to String::from_raw_buf"]
1049 pub unsafe fn from_buf(buf: *const u8) -> String {
1050 String::from_raw_buf(buf)
1053 /// Converts a vector of bytes to a new `String` without checking if
1054 /// it contains valid UTF-8. This is unsafe because it assumes that
1055 /// the UTF-8-ness of the vector has already been validated.
1057 #[deprecated = "renamed to String::from_utf8_unchecked"]
1058 pub unsafe fn from_utf8(bytes: Vec<u8>) -> String {
1059 String::from_utf8_unchecked(bytes)
1063 /// A clone-on-write string
1065 pub type CowString<'a> = Cow<'a, String, str>;
1067 #[allow(deprecated)]
1068 impl<'a> Str for CowString<'a> {
1070 fn as_slice<'b>(&'b self) -> &'b str {
1082 use super::as_string;
1085 fn test_as_string() {
1087 assert_eq!(x, as_string(x).as_slice());
1091 fn test_from_str() {
1092 let owned: Option<::std::string::String> = from_str("string");
1093 assert_eq!(owned.as_ref().map(|s| s.as_slice()), Some("string"));
1097 fn test_from_utf8() {
1098 let xs = b"hello".to_vec();
1099 assert_eq!(String::from_utf8(xs).unwrap(),
1100 String::from_str("hello"));
1102 let xs = "ศไทย中华Việt Nam".as_bytes().to_vec();
1103 assert_eq!(String::from_utf8(xs).unwrap(),
1104 String::from_str("ศไทย中华Việt Nam"));
1106 let xs = b"hello\xFF".to_vec();
1107 let err = String::from_utf8(xs).err().unwrap();
1108 assert_eq!(err.utf8_error(), Utf8Error::TooShort);
1109 assert_eq!(err.into_bytes(), b"hello\xff".to_vec());
1113 fn test_from_utf8_lossy() {
1115 let ys: str::CowString = "hello".into_cow();
1116 assert_eq!(String::from_utf8_lossy(xs), ys);
1118 let xs = "ศไทย中华Việt Nam".as_bytes();
1119 let ys: str::CowString = "ศไทย中华Việt Nam".into_cow();
1120 assert_eq!(String::from_utf8_lossy(xs), ys);
1122 let xs = b"Hello\xC2 There\xFF Goodbye";
1123 assert_eq!(String::from_utf8_lossy(xs),
1124 String::from_str("Hello\u{FFFD} There\u{FFFD} Goodbye").into_cow());
1126 let xs = b"Hello\xC0\x80 There\xE6\x83 Goodbye";
1127 assert_eq!(String::from_utf8_lossy(xs),
1128 String::from_str("Hello\u{FFFD}\u{FFFD} There\u{FFFD} Goodbye").into_cow());
1130 let xs = b"\xF5foo\xF5\x80bar";
1131 assert_eq!(String::from_utf8_lossy(xs),
1132 String::from_str("\u{FFFD}foo\u{FFFD}\u{FFFD}bar").into_cow());
1134 let xs = b"\xF1foo\xF1\x80bar\xF1\x80\x80baz";
1135 assert_eq!(String::from_utf8_lossy(xs),
1136 String::from_str("\u{FFFD}foo\u{FFFD}bar\u{FFFD}baz").into_cow());
1138 let xs = b"\xF4foo\xF4\x80bar\xF4\xBFbaz";
1139 assert_eq!(String::from_utf8_lossy(xs),
1140 String::from_str("\u{FFFD}foo\u{FFFD}bar\u{FFFD}\u{FFFD}baz").into_cow());
1142 let xs = b"\xF0\x80\x80\x80foo\xF0\x90\x80\x80bar";
1143 assert_eq!(String::from_utf8_lossy(xs), String::from_str("\u{FFFD}\u{FFFD}\u{FFFD}\u{FFFD}\
1144 foo\u{10000}bar").into_cow());
1147 let xs = b"\xED\xA0\x80foo\xED\xBF\xBFbar";
1148 assert_eq!(String::from_utf8_lossy(xs), String::from_str("\u{FFFD}\u{FFFD}\u{FFFD}foo\
1149 \u{FFFD}\u{FFFD}\u{FFFD}bar").into_cow());
1153 fn test_from_utf16() {
1155 [(String::from_str("𐍅𐌿𐌻𐍆𐌹𐌻𐌰\n"),
1156 vec![0xd800_u16, 0xdf45_u16, 0xd800_u16, 0xdf3f_u16,
1157 0xd800_u16, 0xdf3b_u16, 0xd800_u16, 0xdf46_u16,
1158 0xd800_u16, 0xdf39_u16, 0xd800_u16, 0xdf3b_u16,
1159 0xd800_u16, 0xdf30_u16, 0x000a_u16]),
1161 (String::from_str("𐐒𐑉𐐮𐑀𐐲𐑋 𐐏𐐲𐑍\n"),
1162 vec![0xd801_u16, 0xdc12_u16, 0xd801_u16,
1163 0xdc49_u16, 0xd801_u16, 0xdc2e_u16, 0xd801_u16,
1164 0xdc40_u16, 0xd801_u16, 0xdc32_u16, 0xd801_u16,
1165 0xdc4b_u16, 0x0020_u16, 0xd801_u16, 0xdc0f_u16,
1166 0xd801_u16, 0xdc32_u16, 0xd801_u16, 0xdc4d_u16,
1169 (String::from_str("𐌀𐌖𐌋𐌄𐌑𐌉·𐌌𐌄𐌕𐌄𐌋𐌉𐌑\n"),
1170 vec![0xd800_u16, 0xdf00_u16, 0xd800_u16, 0xdf16_u16,
1171 0xd800_u16, 0xdf0b_u16, 0xd800_u16, 0xdf04_u16,
1172 0xd800_u16, 0xdf11_u16, 0xd800_u16, 0xdf09_u16,
1173 0x00b7_u16, 0xd800_u16, 0xdf0c_u16, 0xd800_u16,
1174 0xdf04_u16, 0xd800_u16, 0xdf15_u16, 0xd800_u16,
1175 0xdf04_u16, 0xd800_u16, 0xdf0b_u16, 0xd800_u16,
1176 0xdf09_u16, 0xd800_u16, 0xdf11_u16, 0x000a_u16 ]),
1178 (String::from_str("𐒋𐒘𐒈𐒑𐒛𐒒 𐒕𐒓 𐒈𐒚𐒍 𐒏𐒜𐒒𐒖𐒆 𐒕𐒆\n"),
1179 vec![0xd801_u16, 0xdc8b_u16, 0xd801_u16, 0xdc98_u16,
1180 0xd801_u16, 0xdc88_u16, 0xd801_u16, 0xdc91_u16,
1181 0xd801_u16, 0xdc9b_u16, 0xd801_u16, 0xdc92_u16,
1182 0x0020_u16, 0xd801_u16, 0xdc95_u16, 0xd801_u16,
1183 0xdc93_u16, 0x0020_u16, 0xd801_u16, 0xdc88_u16,
1184 0xd801_u16, 0xdc9a_u16, 0xd801_u16, 0xdc8d_u16,
1185 0x0020_u16, 0xd801_u16, 0xdc8f_u16, 0xd801_u16,
1186 0xdc9c_u16, 0xd801_u16, 0xdc92_u16, 0xd801_u16,
1187 0xdc96_u16, 0xd801_u16, 0xdc86_u16, 0x0020_u16,
1188 0xd801_u16, 0xdc95_u16, 0xd801_u16, 0xdc86_u16,
1190 // Issue #12318, even-numbered non-BMP planes
1191 (String::from_str("\u{20000}"),
1192 vec![0xD840, 0xDC00])];
1194 for p in pairs.iter() {
1195 let (s, u) = (*p).clone();
1196 let s_as_utf16 = s.utf16_units().collect::<Vec<u16>>();
1197 let u_as_string = String::from_utf16(u.as_slice()).unwrap();
1199 assert!(::unicode::str::is_utf16(u.as_slice()));
1200 assert_eq!(s_as_utf16, u);
1202 assert_eq!(u_as_string, s);
1203 assert_eq!(String::from_utf16_lossy(u.as_slice()), s);
1205 assert_eq!(String::from_utf16(s_as_utf16.as_slice()).unwrap(), s);
1206 assert_eq!(u_as_string.utf16_units().collect::<Vec<u16>>(), u);
1211 fn test_utf16_invalid() {
1212 // completely positive cases tested above.
1214 assert!(String::from_utf16(&[0xD800]).is_err());
1216 assert!(String::from_utf16(&[0xD800, 0xD800]).is_err());
1219 assert!(String::from_utf16(&[0x0061, 0xDC00]).is_err());
1222 assert!(String::from_utf16(&[0xD800, 0xd801, 0xdc8b, 0xD800]).is_err());
1226 fn test_from_utf16_lossy() {
1227 // completely positive cases tested above.
1229 assert_eq!(String::from_utf16_lossy(&[0xD800]), String::from_str("\u{FFFD}"));
1231 assert_eq!(String::from_utf16_lossy(&[0xD800, 0xD800]),
1232 String::from_str("\u{FFFD}\u{FFFD}"));
1235 assert_eq!(String::from_utf16_lossy(&[0x0061, 0xDC00]), String::from_str("a\u{FFFD}"));
1238 assert_eq!(String::from_utf16_lossy(&[0xD800, 0xd801, 0xdc8b, 0xD800]),
1239 String::from_str("\u{FFFD}𐒋\u{FFFD}"));
1243 fn test_from_buf_len() {
1245 let a = vec![65u8, 65, 65, 65, 65, 65, 65, 0];
1246 assert_eq!(super::raw::from_buf_len(a.as_ptr(), 3), String::from_str("AAA"));
1251 fn test_from_buf() {
1253 let a = vec![65, 65, 65, 65, 65, 65, 65, 0];
1255 let c = super::raw::from_buf(b);
1256 assert_eq!(c, String::from_str("AAAAAAA"));
1261 fn test_push_bytes() {
1262 let mut s = String::from_str("ABC");
1264 let mv = s.as_mut_vec();
1265 mv.push_all(&[b'D']);
1267 assert_eq!(s, "ABCD");
1271 fn test_push_str() {
1272 let mut s = String::new();
1274 assert_eq!(s.slice_from(0), "");
1276 assert_eq!(s.slice_from(0), "abc");
1277 s.push_str("ประเทศไทย中华Việt Nam");
1278 assert_eq!(s.slice_from(0), "abcประเทศไทย中华Việt Nam");
1283 let mut data = String::from_str("ประเทศไทย中");
1285 data.push('b'); // 1 byte
1286 data.push('¢'); // 2 byte
1287 data.push('€'); // 3 byte
1288 data.push('𤭢'); // 4 byte
1289 assert_eq!(data, "ประเทศไทย中华b¢€𤭢");
1294 let mut data = String::from_str("ประเทศไทย中华b¢€𤭢");
1295 assert_eq!(data.pop().unwrap(), '𤭢'); // 4 bytes
1296 assert_eq!(data.pop().unwrap(), '€'); // 3 bytes
1297 assert_eq!(data.pop().unwrap(), '¢'); // 2 bytes
1298 assert_eq!(data.pop().unwrap(), 'b'); // 1 bytes
1299 assert_eq!(data.pop().unwrap(), '华');
1300 assert_eq!(data, "ประเทศไทย中");
1304 fn test_str_truncate() {
1305 let mut s = String::from_str("12345");
1307 assert_eq!(s, "12345");
1309 assert_eq!(s, "123");
1313 let mut s = String::from_str("12345");
1317 let p_ = s.as_ptr();
1323 fn test_str_truncate_invalid_len() {
1324 let mut s = String::from_str("12345");
1330 fn test_str_truncate_split_codepoint() {
1331 let mut s = String::from_str("\u{FC}"); // ü
1336 fn test_str_clear() {
1337 let mut s = String::from_str("12345");
1339 assert_eq!(s.len(), 0);
1345 let a = String::from_str("12345");
1348 assert_eq!(b.len(), 7);
1349 assert_eq!(b, "1234522");
1354 let mut s = "ศไทย中华Việt Nam; foobar".to_string();;
1355 assert_eq!(s.remove(0), 'ศ');
1356 assert_eq!(s.len(), 33);
1357 assert_eq!(s, "ไทย中华Việt Nam; foobar");
1358 assert_eq!(s.remove(17), 'ệ');
1359 assert_eq!(s, "ไทย中华Vit Nam; foobar");
1362 #[test] #[should_fail]
1364 "ศ".to_string().remove(1);
1369 let mut s = "foobar".to_string();
1371 assert_eq!(s, "ệfoobar");
1373 assert_eq!(s, "ệfooยbar");
1376 #[test] #[should_fail] fn insert_bad1() { "".to_string().insert(1, 't'); }
1377 #[test] #[should_fail] fn insert_bad2() { "ệ".to_string().insert(1, 't'); }
1381 let s = "foobar".to_string();
1382 assert_eq!("foobar", s[]);
1383 assert_eq!("foo", s[..3]);
1384 assert_eq!("bar", s[3..]);
1385 assert_eq!("oob", s[1..4]);
1389 fn test_simple_types() {
1390 assert_eq!(1i.to_string(), "1");
1391 assert_eq!((-1i).to_string(), "-1");
1392 assert_eq!(200u.to_string(), "200");
1393 assert_eq!(2u8.to_string(), "2");
1394 assert_eq!(true.to_string(), "true");
1395 assert_eq!(false.to_string(), "false");
1396 assert_eq!(().to_string(), "()");
1397 assert_eq!(("hi".to_string()).to_string(), "hi");
1402 let x: Vec<int> = vec![];
1403 assert_eq!(x.to_string(), "[]");
1404 assert_eq!((vec![1i]).to_string(), "[1]");
1405 assert_eq!((vec![1i, 2, 3]).to_string(), "[1, 2, 3]");
1406 assert!((vec![vec![], vec![1i], vec![1i, 1]]).to_string() ==
1407 "[[], [1], [1, 1]]");
1411 fn test_from_iterator() {
1412 let s = "ศไทย中华Việt Nam".to_string();
1416 let a: String = s.chars().collect();
1419 let mut b = t.to_string();
1420 b.extend(u.chars());
1423 let c: String = vec![t, u].into_iter().collect();
1426 let mut d = t.to_string();
1427 d.extend(vec![u].into_iter());
1432 fn bench_with_capacity(b: &mut Bencher) {
1434 String::with_capacity(100)
1439 fn bench_push_str(b: &mut Bencher) {
1440 let s = "ศไทย中华Việt Nam; Mary had a little lamb, Little lamb";
1442 let mut r = String::new();
1447 const REPETITIONS: u64 = 10_000;
1450 fn bench_push_str_one_byte(b: &mut Bencher) {
1451 b.bytes = REPETITIONS;
1453 let mut r = String::new();
1454 for _ in range(0, REPETITIONS) {
1461 fn bench_push_char_one_byte(b: &mut Bencher) {
1462 b.bytes = REPETITIONS;
1464 let mut r = String::new();
1465 for _ in range(0, REPETITIONS) {
1472 fn bench_push_char_two_bytes(b: &mut Bencher) {
1473 b.bytes = REPETITIONS * 2;
1475 let mut r = String::new();
1476 for _ in range(0, REPETITIONS) {
1483 fn from_utf8_lossy_100_ascii(b: &mut Bencher) {
1484 let s = b"Hello there, the quick brown fox jumped over the lazy dog! \
1485 Lorem ipsum dolor sit amet, consectetur. ";
1487 assert_eq!(100, s.len());
1489 let _ = String::from_utf8_lossy(s);
1494 fn from_utf8_lossy_100_multibyte(b: &mut Bencher) {
1495 let s = "𐌀𐌖𐌋𐌄𐌑𐌉ปรدولة الكويتทศไทย中华𐍅𐌿𐌻𐍆𐌹𐌻𐌰".as_bytes();
1496 assert_eq!(100, s.len());
1498 let _ = String::from_utf8_lossy(s);
1503 fn from_utf8_lossy_invalid(b: &mut Bencher) {
1504 let s = b"Hello\xC0\x80 There\xE6\x83 Goodbye";
1506 let _ = String::from_utf8_lossy(s);
1511 fn from_utf8_lossy_100_invalid(b: &mut Bencher) {
1512 let s = Vec::from_elem(100, 0xF5u8);
1514 let _ = String::from_utf8_lossy(s.as_slice());