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 slice::CloneSliceExt;
31 use str::{mod, CharRange, FromStr, Utf8Error};
32 use vec::{DerefVec, Vec, as_vec};
34 /// A growable string stored as a UTF-8 encoded buffer.
35 #[deriving(Clone, PartialOrd, Eq, Ord)]
41 /// A possible error value from the `String::from_utf8` function.
43 pub struct FromUtf8Error {
48 /// A possible error value from the `String::from_utf16` function.
50 #[allow(missing_copy_implementations)]
51 pub struct FromUtf16Error(());
54 /// Creates a new string buffer initialized with the empty string.
59 /// let mut s = String::new();
63 pub fn new() -> String {
69 /// Creates a new string buffer with the given capacity.
70 /// The string will be able to hold exactly `capacity` bytes without
71 /// reallocating. If `capacity` is 0, the string will not allocate.
76 /// let mut s = String::with_capacity(10);
80 pub fn with_capacity(capacity: uint) -> String {
82 vec: Vec::with_capacity(capacity),
86 /// Creates a new string buffer from the given string.
91 /// let s = String::from_str("hello");
92 /// assert_eq!(s.as_slice(), "hello");
95 #[experimental = "needs investigation to see if to_string() can match perf"]
96 pub fn from_str(string: &str) -> String {
97 String { vec: string.as_bytes().to_vec() }
100 /// Returns the vector as a string buffer, if possible, taking care not to
105 /// If the given vector is not valid UTF-8, then the original vector and the
106 /// corresponding error is returned.
111 /// # #![allow(deprecated)]
112 /// use std::str::Utf8Error;
114 /// let hello_vec = vec![104, 101, 108, 108, 111];
115 /// let s = String::from_utf8(hello_vec).unwrap();
116 /// assert_eq!(s, "hello");
118 /// let invalid_vec = vec![240, 144, 128];
119 /// let s = String::from_utf8(invalid_vec).err().unwrap();
120 /// assert_eq!(s.utf8_error(), Utf8Error::TooShort);
121 /// assert_eq!(s.into_bytes(), vec![240, 144, 128]);
125 pub fn from_utf8(vec: Vec<u8>) -> Result<String, FromUtf8Error> {
126 match str::from_utf8(vec.as_slice()) {
127 Ok(..) => Ok(String { vec: vec }),
128 Err(e) => Err(FromUtf8Error { bytes: vec, error: e })
132 /// Converts a vector of bytes to a new UTF-8 string.
133 /// Any invalid UTF-8 sequences are replaced with U+FFFD REPLACEMENT CHARACTER.
138 /// let input = b"Hello \xF0\x90\x80World";
139 /// let output = String::from_utf8_lossy(input);
140 /// assert_eq!(output.as_slice(), "Hello \u{FFFD}World");
143 pub fn from_utf8_lossy<'a>(v: &'a [u8]) -> CowString<'a> {
144 match str::from_utf8(v) {
145 Ok(s) => return Cow::Borrowed(s),
149 static TAG_CONT_U8: u8 = 128u8;
150 static REPLACEMENT: &'static [u8] = b"\xEF\xBF\xBD"; // U+FFFD in UTF-8
153 fn unsafe_get(xs: &[u8], i: uint) -> u8 {
154 unsafe { *xs.get_unchecked(i) }
156 fn safe_get(xs: &[u8], i: uint, total: uint) -> u8 {
164 let mut res = String::with_capacity(total);
168 res.as_mut_vec().push_all(v[..i])
172 // subseqidx is the index of the first byte of the subsequence we're looking at.
173 // It's used to copy a bunch of contiguous good codepoints at once instead of copying
175 let mut subseqidx = 0;
179 let byte = unsafe_get(v, i);
182 macro_rules! error(() => ({
185 res.as_mut_vec().push_all(v[subseqidx..i_]);
188 res.as_mut_vec().push_all(REPLACEMENT);
193 // subseqidx handles this
195 let w = unicode_str::utf8_char_width(byte);
199 if safe_get(v, i, total) & 192u8 != TAG_CONT_U8 {
206 match (byte, safe_get(v, i, total)) {
207 (0xE0 , 0xA0 ... 0xBF) => (),
208 (0xE1 ... 0xEC, 0x80 ... 0xBF) => (),
209 (0xED , 0x80 ... 0x9F) => (),
210 (0xEE ... 0xEF, 0x80 ... 0xBF) => (),
217 if safe_get(v, i, total) & 192u8 != TAG_CONT_U8 {
224 match (byte, safe_get(v, i, total)) {
225 (0xF0 , 0x90 ... 0xBF) => (),
226 (0xF1 ... 0xF3, 0x80 ... 0xBF) => (),
227 (0xF4 , 0x80 ... 0x8F) => (),
234 if safe_get(v, i, total) & 192u8 != TAG_CONT_U8 {
239 if safe_get(v, i, total) & 192u8 != TAG_CONT_U8 {
252 if subseqidx < total {
254 res.as_mut_vec().push_all(v[subseqidx..total])
260 /// Decode a UTF-16 encoded vector `v` into a `String`, returning `None`
261 /// if `v` contains any invalid data.
267 /// let mut v = &mut [0xD834, 0xDD1E, 0x006d, 0x0075,
268 /// 0x0073, 0x0069, 0x0063];
269 /// assert_eq!(String::from_utf16(v).unwrap(),
270 /// "𝄞music".to_string());
272 /// // 𝄞mu<invalid>ic
274 /// assert!(String::from_utf16(v).is_err());
277 pub fn from_utf16(v: &[u16]) -> Result<String, FromUtf16Error> {
278 let mut s = String::with_capacity(v.len());
279 for c in unicode_str::utf16_items(v) {
281 Utf16Item::ScalarValue(c) => s.push(c),
282 Utf16Item::LoneSurrogate(_) => return Err(FromUtf16Error(())),
288 /// Decode a UTF-16 encoded vector `v` into a string, replacing
289 /// invalid data with the replacement character (U+FFFD).
294 /// // 𝄞mus<invalid>ic<invalid>
295 /// let v = &[0xD834, 0xDD1E, 0x006d, 0x0075,
296 /// 0x0073, 0xDD1E, 0x0069, 0x0063,
299 /// assert_eq!(String::from_utf16_lossy(v),
300 /// "𝄞mus\u{FFFD}ic\u{FFFD}".to_string());
303 pub fn from_utf16_lossy(v: &[u16]) -> String {
304 unicode_str::utf16_items(v).map(|c| c.to_char_lossy()).collect()
307 /// Convert a vector of `char`s to a `String`.
312 /// # #![allow(deprecated)]
313 /// let chars = &['h', 'e', 'l', 'l', 'o'];
314 /// let s = String::from_chars(chars);
315 /// assert_eq!(s.as_slice(), "hello");
318 #[deprecated = "use .collect() instead"]
319 pub fn from_chars(chs: &[char]) -> String {
320 chs.iter().map(|c| *c).collect()
323 /// Creates a new `String` from a length, capacity, and pointer.
325 /// This is unsafe because:
326 /// * We call `Vec::from_raw_parts` to get a `Vec<u8>`;
327 /// * We assume that the `Vec` contains valid UTF-8.
330 pub unsafe fn from_raw_parts(buf: *mut u8, length: uint, capacity: uint) -> String {
332 vec: Vec::from_raw_parts(buf, length, capacity),
336 /// Creates a `String` from a null-terminated `*const u8` buffer.
338 /// This function is unsafe because we dereference memory until we find the
339 /// NUL character, which is not guaranteed to be present. Additionally, the
340 /// slice is not checked to see whether it contains valid UTF-8
341 #[unstable = "just renamed from `mod raw`"]
342 pub unsafe fn from_raw_buf(buf: *const u8) -> String {
343 String::from_str(str::from_c_str(buf as *const i8))
346 /// Creates a `String` from a `*const u8` buffer of the given length.
348 /// This function is unsafe because it blindly assumes the validity of the
349 /// pointer `buf` for `len` bytes of memory. This function will copy the
350 /// memory from `buf` into a new allocation (owned by the returned
353 /// This function is also unsafe because it does not validate that the
354 /// buffer is valid UTF-8 encoded data.
355 #[unstable = "just renamed from `mod raw`"]
356 pub unsafe fn from_raw_buf_len(buf: *const u8, len: uint) -> String {
357 String::from_utf8_unchecked(Vec::from_raw_buf(buf, len))
360 /// Converts a vector of bytes to a new `String` without checking if
361 /// it contains valid UTF-8. This is unsafe because it assumes that
362 /// the UTF-8-ness of the vector has already been validated.
365 pub unsafe fn from_utf8_unchecked(bytes: Vec<u8>) -> String {
366 String { vec: bytes }
369 /// Return the underlying byte buffer, encoded as UTF-8.
374 /// let s = String::from_str("hello");
375 /// let bytes = s.into_bytes();
376 /// assert_eq!(bytes, vec![104, 101, 108, 108, 111]);
380 pub fn into_bytes(self) -> Vec<u8> {
384 /// Creates a string buffer by repeating a character `length` times.
389 /// # #![allow(deprecated)]
390 /// let s = String::from_char(5, 'a');
391 /// assert_eq!(s.as_slice(), "aaaaa");
394 #[deprecated = "use repeat(ch).take(length).collect() instead"]
395 pub fn from_char(length: uint, ch: char) -> String {
400 let mut buf = String::new();
402 let size = buf.len() * (length - 1);
403 buf.reserve_exact(size);
404 for _ in range(1, length) {
410 /// Pushes the given string onto this string buffer.
415 /// let mut s = String::from_str("foo");
416 /// s.push_str("bar");
417 /// assert_eq!(s.as_slice(), "foobar");
421 pub fn push_str(&mut self, string: &str) {
422 self.vec.push_all(string.as_bytes())
425 /// Pushes `ch` onto the given string `count` times.
430 /// # #![allow(deprecated)]
431 /// let mut s = String::from_str("foo");
433 /// assert_eq!(s.as_slice(), "fooZZZZZ");
436 #[deprecated = "deprecated in favor of .extend(repeat(ch).take(count))"]
437 pub fn grow(&mut self, count: uint, ch: char) {
438 for _ in range(0, count) {
443 /// Returns the number of bytes that this string buffer can hold without
449 /// let s = String::with_capacity(10);
450 /// assert!(s.capacity() >= 10);
454 pub fn capacity(&self) -> uint {
458 /// Deprecated: Renamed to `reserve`.
459 #[deprecated = "Renamed to `reserve`"]
460 pub fn reserve_additional(&mut self, extra: uint) {
461 self.vec.reserve(extra)
464 /// Reserves capacity for at least `additional` more bytes to be inserted
465 /// in the given `String`. The collection may reserve more space to avoid
466 /// frequent reallocations.
470 /// Panics if the new capacity overflows `uint`.
475 /// let mut s = String::new();
477 /// assert!(s.capacity() >= 10);
481 pub fn reserve(&mut self, additional: uint) {
482 self.vec.reserve(additional)
485 /// Reserves the minimum capacity for exactly `additional` more bytes to be
486 /// inserted in the given `String`. Does nothing if the capacity is already
489 /// Note that the allocator may give the collection more space than it
490 /// requests. Therefore capacity can not be relied upon to be precisely
491 /// minimal. Prefer `reserve` if future insertions are expected.
495 /// Panics if the new capacity overflows `uint`.
500 /// let mut s = String::new();
502 /// assert!(s.capacity() >= 10);
506 pub fn reserve_exact(&mut self, additional: uint) {
507 self.vec.reserve_exact(additional)
510 /// Shrinks the capacity of this string buffer to match its length.
515 /// let mut s = String::from_str("foo");
517 /// assert!(s.capacity() >= 100);
518 /// s.shrink_to_fit();
519 /// assert_eq!(s.capacity(), 3);
523 pub fn shrink_to_fit(&mut self) {
524 self.vec.shrink_to_fit()
527 /// Adds the given character to the end of the string.
532 /// let mut s = String::from_str("abc");
536 /// assert_eq!(s.as_slice(), "abc123");
540 pub fn push(&mut self, ch: char) {
541 if (ch as u32) < 0x80 {
542 self.vec.push(ch as u8);
546 let cur_len = self.len();
547 // This may use up to 4 bytes.
551 // Attempt to not use an intermediate buffer by just pushing bytes
552 // directly onto this string.
553 let slice = RawSlice {
554 data: self.vec.as_ptr().offset(cur_len as int),
557 let used = ch.encode_utf8(mem::transmute(slice)).unwrap_or(0);
558 self.vec.set_len(cur_len + used);
562 /// Works with the underlying buffer as a byte slice.
567 /// let s = String::from_str("hello");
568 /// let b: &[_] = &[104, 101, 108, 108, 111];
569 /// assert_eq!(s.as_bytes(), b);
573 pub fn as_bytes<'a>(&'a self) -> &'a [u8] {
577 /// Shortens a string to the specified length.
581 /// Panics if `new_len` > current length,
582 /// or if `new_len` is not a character boundary.
587 /// let mut s = String::from_str("hello");
589 /// assert_eq!(s.as_slice(), "he");
593 pub fn truncate(&mut self, new_len: uint) {
594 assert!(self.is_char_boundary(new_len));
595 self.vec.truncate(new_len)
598 /// Removes the last character from the string buffer and returns it.
599 /// Returns `None` if this string buffer is empty.
604 /// let mut s = String::from_str("foo");
605 /// assert_eq!(s.pop(), Some('o'));
606 /// assert_eq!(s.pop(), Some('o'));
607 /// assert_eq!(s.pop(), Some('f'));
608 /// assert_eq!(s.pop(), None);
612 pub fn pop(&mut self) -> Option<char> {
613 let len = self.len();
618 let CharRange {ch, next} = self.char_range_at_reverse(len);
620 self.vec.set_len(next);
625 /// Removes the character from the string buffer at byte position `idx` and
630 /// This is an O(n) operation as it requires copying every element in the
635 /// If `idx` does not lie on a character boundary, or if it is out of
636 /// bounds, then this function will panic.
641 /// let mut s = String::from_str("foo");
642 /// assert_eq!(s.remove(0), 'f');
643 /// assert_eq!(s.remove(1), 'o');
644 /// assert_eq!(s.remove(0), 'o');
647 pub fn remove(&mut self, idx: uint) -> char {
648 let len = self.len();
651 let CharRange { ch, next } = self.char_range_at(idx);
653 ptr::copy_memory(self.vec.as_mut_ptr().offset(idx as int),
654 self.vec.as_ptr().offset(next as int),
656 self.vec.set_len(len - (next - idx));
661 /// Insert a character into the string buffer at byte position `idx`.
665 /// This is an O(n) operation as it requires copying every element in the
670 /// If `idx` does not lie on a character boundary or is out of bounds, then
671 /// this function will panic.
673 pub fn insert(&mut self, idx: uint, ch: char) {
674 let len = self.len();
676 assert!(self.is_char_boundary(idx));
678 let mut bits = [0; 4];
679 let amt = ch.encode_utf8(&mut bits).unwrap();
682 ptr::copy_memory(self.vec.as_mut_ptr().offset((idx + amt) as int),
683 self.vec.as_ptr().offset(idx as int),
685 ptr::copy_memory(self.vec.as_mut_ptr().offset(idx as int),
688 self.vec.set_len(len + amt);
692 /// Views the string buffer as a mutable sequence of bytes.
694 /// This is unsafe because it does not check
695 /// to ensure that the resulting string will be valid UTF-8.
700 /// let mut s = String::from_str("hello");
702 /// let vec = s.as_mut_vec();
703 /// assert!(vec == &mut vec![104, 101, 108, 108, 111]);
706 /// assert_eq!(s.as_slice(), "olleh");
709 pub unsafe fn as_mut_vec<'a>(&'a mut self) -> &'a mut Vec<u8> {
713 /// Return the number of bytes in this string.
718 /// let a = "foo".to_string();
719 /// assert_eq!(a.len(), 3);
723 pub fn len(&self) -> uint { self.vec.len() }
725 /// Returns true if the string contains no bytes
730 /// let mut v = String::new();
731 /// assert!(v.is_empty());
733 /// assert!(!v.is_empty());
736 pub fn is_empty(&self) -> bool { self.len() == 0 }
738 /// Truncates the string, returning it to 0 length.
743 /// let mut s = "foo".to_string();
745 /// assert!(s.is_empty());
749 pub fn clear(&mut self) {
755 /// Consume this error, returning the bytes that were attempted to make a
758 pub fn into_bytes(self) -> Vec<u8> { self.bytes }
760 /// Access the underlying UTF8-error that was the cause of this error.
762 pub fn utf8_error(&self) -> Utf8Error { self.error }
765 impl fmt::Show for FromUtf8Error {
766 fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
771 impl fmt::Show for FromUtf16Error {
772 fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
773 "invalid utf-16: lone surrogate found".fmt(f)
777 #[experimental = "waiting on FromIterator stabilization"]
778 impl FromIterator<char> for String {
779 fn from_iter<I:Iterator<char>>(iterator: I) -> String {
780 let mut buf = String::new();
781 buf.extend(iterator);
786 #[experimental = "waiting on FromIterator stabilization"]
787 impl<'a> FromIterator<&'a str> for String {
788 fn from_iter<I:Iterator<&'a str>>(iterator: I) -> String {
789 let mut buf = String::new();
790 buf.extend(iterator);
795 #[experimental = "waiting on Extend stabilization"]
796 impl Extend<char> for String {
797 fn extend<I:Iterator<char>>(&mut self, mut iterator: I) {
798 let (lower_bound, _) = iterator.size_hint();
799 self.reserve(lower_bound);
806 #[experimental = "waiting on Extend stabilization"]
807 impl<'a> Extend<&'a str> for String {
808 fn extend<I: Iterator<&'a str>>(&mut self, mut iterator: I) {
809 // A guess that at least one byte per iterator element will be needed.
810 let (lower_bound, _) = iterator.size_hint();
811 self.reserve(lower_bound);
819 impl PartialEq for String {
821 fn eq(&self, other: &String) -> bool { PartialEq::eq(&**self, &**other) }
823 fn ne(&self, other: &String) -> bool { PartialEq::ne(&**self, &**other) }
826 macro_rules! impl_eq {
827 ($lhs:ty, $rhs: ty) => {
829 impl<'a> PartialEq<$rhs> for $lhs {
831 fn eq(&self, other: &$rhs) -> bool { PartialEq::eq(&**self, &**other) }
833 fn ne(&self, other: &$rhs) -> bool { PartialEq::ne(&**self, &**other) }
837 impl<'a> PartialEq<$lhs> for $rhs {
839 fn eq(&self, other: &$lhs) -> bool { PartialEq::eq(&**self, &**other) }
841 fn ne(&self, other: &$lhs) -> bool { PartialEq::ne(&**self, &**other) }
847 impl_eq! { String, &'a str }
848 impl_eq! { CowString<'a>, String }
851 impl<'a, 'b> PartialEq<&'b str> for CowString<'a> {
853 fn eq(&self, other: &&'b str) -> bool { PartialEq::eq(&**self, &**other) }
855 fn ne(&self, other: &&'b str) -> bool { PartialEq::ne(&**self, &**other) }
859 impl<'a, 'b> PartialEq<CowString<'a>> for &'b str {
861 fn eq(&self, other: &CowString<'a>) -> bool { PartialEq::eq(&**self, &**other) }
863 fn ne(&self, other: &CowString<'a>) -> bool { PartialEq::ne(&**self, &**other) }
866 #[experimental = "waiting on Str stabilization"]
868 impl Str for String {
871 fn as_slice<'a>(&'a self) -> &'a str {
872 unsafe { mem::transmute(self.vec.as_slice()) }
877 impl Default for String {
879 fn default() -> String {
884 #[experimental = "waiting on Show stabilization"]
885 impl fmt::Show for String {
886 fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
891 #[experimental = "waiting on Hash stabilization"]
892 impl<H: hash::Writer> hash::Hash<H> for String {
894 fn hash(&self, hasher: &mut H) {
895 (**self).hash(hasher)
900 #[deprecated = "Use overloaded `core::cmp::PartialEq`"]
901 impl<'a, S: Str> Equiv<S> for String {
903 fn equiv(&self, other: &S) -> bool {
904 self.as_slice() == other.as_slice()
908 #[experimental = "waiting on Add stabilization"]
909 impl<'a> Add<&'a str, String> for String {
910 fn add(mut self, other: &str) -> String {
911 self.push_str(other);
916 impl ops::Slice<uint, str> for String {
918 fn as_slice_<'a>(&'a self) -> &'a str {
919 unsafe { mem::transmute(self.vec.as_slice()) }
923 fn slice_from_or_fail<'a>(&'a self, from: &uint) -> &'a str {
928 fn slice_to_or_fail<'a>(&'a self, to: &uint) -> &'a str {
933 fn slice_or_fail<'a>(&'a self, from: &uint, to: &uint) -> &'a str {
938 #[experimental = "waiting on Deref stabilization"]
939 impl ops::Deref<str> for String {
940 fn deref<'a>(&'a self) -> &'a str {
941 unsafe { mem::transmute(self.vec[]) }
945 /// Wrapper type providing a `&String` reference via `Deref`.
947 pub struct DerefString<'a> {
951 impl<'a> Deref<String> for DerefString<'a> {
952 fn deref<'b>(&'b self) -> &'b String {
953 unsafe { mem::transmute(&*self.x) }
957 /// Convert a string slice to a wrapper type providing a `&String` reference.
962 /// use std::string::as_string;
964 /// fn string_consumer(s: String) {
965 /// assert_eq!(s, "foo".to_string());
968 /// let string = as_string("foo").clone();
969 /// string_consumer(string);
972 pub fn as_string<'a>(x: &'a str) -> DerefString<'a> {
973 DerefString { x: as_vec(x.as_bytes()) }
976 impl FromStr for String {
978 fn from_str(s: &str) -> Option<String> {
979 Some(String::from_str(s))
983 /// Trait for converting a type to a string, consuming it in the process.
984 #[deprecated = "trait will be removed"]
985 pub trait IntoString {
986 /// Consume and convert to a string.
987 fn into_string(self) -> String;
990 /// A generic trait for converting a value to a string
992 /// Converts the value of `self` to an owned string
993 fn to_string(&self) -> String;
996 impl<T: fmt::Show> ToString for T {
997 fn to_string(&self) -> String {
998 let mut buf = Vec::<u8>::new();
999 let _ = fmt::write(&mut buf, format_args!("{}", *self));
1000 String::from_utf8(buf).unwrap()
1004 impl IntoCow<'static, String, str> for String {
1005 fn into_cow(self) -> CowString<'static> {
1010 impl<'a> IntoCow<'a, String, str> for &'a str {
1011 fn into_cow(self) -> CowString<'a> {
1016 /// Unsafe operations
1022 /// Creates a new `String` from a length, capacity, and pointer.
1024 /// This is unsafe because:
1025 /// * We call `Vec::from_raw_parts` to get a `Vec<u8>`;
1026 /// * We assume that the `Vec` contains valid UTF-8.
1028 #[deprecated = "renamed to String::from_raw_parts"]
1029 pub unsafe fn from_parts(buf: *mut u8, length: uint, capacity: uint) -> String {
1030 String::from_raw_parts(buf, length, capacity)
1033 /// Creates a `String` from a `*const u8` buffer of the given length.
1035 /// This function is unsafe because of two reasons:
1037 /// * A raw pointer is dereferenced and transmuted to `&[u8]`;
1038 /// * The slice is not checked to see whether it contains valid UTF-8.
1039 #[deprecated = "renamed to String::from_raw_buf_len"]
1040 pub unsafe fn from_buf_len(buf: *const u8, len: uint) -> String {
1041 String::from_raw_buf_len(buf, len)
1044 /// Creates a `String` from a null-terminated `*const u8` buffer.
1046 /// This function is unsafe because we dereference memory until we find the NUL character,
1047 /// which is not guaranteed to be present. Additionally, the slice is not checked to see
1048 /// whether it contains valid UTF-8
1049 #[deprecated = "renamed to String::from_raw_buf"]
1050 pub unsafe fn from_buf(buf: *const u8) -> String {
1051 String::from_raw_buf(buf)
1054 /// Converts a vector of bytes to a new `String` without checking if
1055 /// it contains valid UTF-8. This is unsafe because it assumes that
1056 /// the UTF-8-ness of the vector has already been validated.
1058 #[deprecated = "renamed to String::from_utf8_unchecked"]
1059 pub unsafe fn from_utf8(bytes: Vec<u8>) -> String {
1060 String::from_utf8_unchecked(bytes)
1064 /// A clone-on-write string
1066 pub type CowString<'a> = Cow<'a, String, str>;
1068 #[allow(deprecated)]
1069 impl<'a> Str for CowString<'a> {
1071 fn as_slice<'b>(&'b self) -> &'b str {
1083 use super::as_string;
1086 fn test_as_string() {
1088 assert_eq!(x, as_string(x).as_slice());
1092 fn test_from_str() {
1093 let owned: Option<::std::string::String> = from_str("string");
1094 assert_eq!(owned.as_ref().map(|s| s.as_slice()), Some("string"));
1098 fn test_from_utf8() {
1099 let xs = b"hello".to_vec();
1100 assert_eq!(String::from_utf8(xs).unwrap(),
1101 String::from_str("hello"));
1103 let xs = "ศไทย中华Việt Nam".as_bytes().to_vec();
1104 assert_eq!(String::from_utf8(xs).unwrap(),
1105 String::from_str("ศไทย中华Việt Nam"));
1107 let xs = b"hello\xFF".to_vec();
1108 let err = String::from_utf8(xs).err().unwrap();
1109 assert_eq!(err.utf8_error(), Utf8Error::TooShort);
1110 assert_eq!(err.into_bytes(), b"hello\xff".to_vec());
1114 fn test_from_utf8_lossy() {
1116 let ys: str::CowString = "hello".into_cow();
1117 assert_eq!(String::from_utf8_lossy(xs), ys);
1119 let xs = "ศไทย中华Việt Nam".as_bytes();
1120 let ys: str::CowString = "ศไทย中华Việt Nam".into_cow();
1121 assert_eq!(String::from_utf8_lossy(xs), ys);
1123 let xs = b"Hello\xC2 There\xFF Goodbye";
1124 assert_eq!(String::from_utf8_lossy(xs),
1125 String::from_str("Hello\u{FFFD} There\u{FFFD} Goodbye").into_cow());
1127 let xs = b"Hello\xC0\x80 There\xE6\x83 Goodbye";
1128 assert_eq!(String::from_utf8_lossy(xs),
1129 String::from_str("Hello\u{FFFD}\u{FFFD} There\u{FFFD} Goodbye").into_cow());
1131 let xs = b"\xF5foo\xF5\x80bar";
1132 assert_eq!(String::from_utf8_lossy(xs),
1133 String::from_str("\u{FFFD}foo\u{FFFD}\u{FFFD}bar").into_cow());
1135 let xs = b"\xF1foo\xF1\x80bar\xF1\x80\x80baz";
1136 assert_eq!(String::from_utf8_lossy(xs),
1137 String::from_str("\u{FFFD}foo\u{FFFD}bar\u{FFFD}baz").into_cow());
1139 let xs = b"\xF4foo\xF4\x80bar\xF4\xBFbaz";
1140 assert_eq!(String::from_utf8_lossy(xs),
1141 String::from_str("\u{FFFD}foo\u{FFFD}bar\u{FFFD}\u{FFFD}baz").into_cow());
1143 let xs = b"\xF0\x80\x80\x80foo\xF0\x90\x80\x80bar";
1144 assert_eq!(String::from_utf8_lossy(xs), String::from_str("\u{FFFD}\u{FFFD}\u{FFFD}\u{FFFD}\
1145 foo\u{10000}bar").into_cow());
1148 let xs = b"\xED\xA0\x80foo\xED\xBF\xBFbar";
1149 assert_eq!(String::from_utf8_lossy(xs), String::from_str("\u{FFFD}\u{FFFD}\u{FFFD}foo\
1150 \u{FFFD}\u{FFFD}\u{FFFD}bar").into_cow());
1154 fn test_from_utf16() {
1156 [(String::from_str("𐍅𐌿𐌻𐍆𐌹𐌻𐌰\n"),
1157 vec![0xd800_u16, 0xdf45_u16, 0xd800_u16, 0xdf3f_u16,
1158 0xd800_u16, 0xdf3b_u16, 0xd800_u16, 0xdf46_u16,
1159 0xd800_u16, 0xdf39_u16, 0xd800_u16, 0xdf3b_u16,
1160 0xd800_u16, 0xdf30_u16, 0x000a_u16]),
1162 (String::from_str("𐐒𐑉𐐮𐑀𐐲𐑋 𐐏𐐲𐑍\n"),
1163 vec![0xd801_u16, 0xdc12_u16, 0xd801_u16,
1164 0xdc49_u16, 0xd801_u16, 0xdc2e_u16, 0xd801_u16,
1165 0xdc40_u16, 0xd801_u16, 0xdc32_u16, 0xd801_u16,
1166 0xdc4b_u16, 0x0020_u16, 0xd801_u16, 0xdc0f_u16,
1167 0xd801_u16, 0xdc32_u16, 0xd801_u16, 0xdc4d_u16,
1170 (String::from_str("𐌀𐌖𐌋𐌄𐌑𐌉·𐌌𐌄𐌕𐌄𐌋𐌉𐌑\n"),
1171 vec![0xd800_u16, 0xdf00_u16, 0xd800_u16, 0xdf16_u16,
1172 0xd800_u16, 0xdf0b_u16, 0xd800_u16, 0xdf04_u16,
1173 0xd800_u16, 0xdf11_u16, 0xd800_u16, 0xdf09_u16,
1174 0x00b7_u16, 0xd800_u16, 0xdf0c_u16, 0xd800_u16,
1175 0xdf04_u16, 0xd800_u16, 0xdf15_u16, 0xd800_u16,
1176 0xdf04_u16, 0xd800_u16, 0xdf0b_u16, 0xd800_u16,
1177 0xdf09_u16, 0xd800_u16, 0xdf11_u16, 0x000a_u16 ]),
1179 (String::from_str("𐒋𐒘𐒈𐒑𐒛𐒒 𐒕𐒓 𐒈𐒚𐒍 𐒏𐒜𐒒𐒖𐒆 𐒕𐒆\n"),
1180 vec![0xd801_u16, 0xdc8b_u16, 0xd801_u16, 0xdc98_u16,
1181 0xd801_u16, 0xdc88_u16, 0xd801_u16, 0xdc91_u16,
1182 0xd801_u16, 0xdc9b_u16, 0xd801_u16, 0xdc92_u16,
1183 0x0020_u16, 0xd801_u16, 0xdc95_u16, 0xd801_u16,
1184 0xdc93_u16, 0x0020_u16, 0xd801_u16, 0xdc88_u16,
1185 0xd801_u16, 0xdc9a_u16, 0xd801_u16, 0xdc8d_u16,
1186 0x0020_u16, 0xd801_u16, 0xdc8f_u16, 0xd801_u16,
1187 0xdc9c_u16, 0xd801_u16, 0xdc92_u16, 0xd801_u16,
1188 0xdc96_u16, 0xd801_u16, 0xdc86_u16, 0x0020_u16,
1189 0xd801_u16, 0xdc95_u16, 0xd801_u16, 0xdc86_u16,
1191 // Issue #12318, even-numbered non-BMP planes
1192 (String::from_str("\u{20000}"),
1193 vec![0xD840, 0xDC00])];
1195 for p in pairs.iter() {
1196 let (s, u) = (*p).clone();
1197 let s_as_utf16 = s.utf16_units().collect::<Vec<u16>>();
1198 let u_as_string = String::from_utf16(u.as_slice()).unwrap();
1200 assert!(::unicode::str::is_utf16(u.as_slice()));
1201 assert_eq!(s_as_utf16, u);
1203 assert_eq!(u_as_string, s);
1204 assert_eq!(String::from_utf16_lossy(u.as_slice()), s);
1206 assert_eq!(String::from_utf16(s_as_utf16.as_slice()).unwrap(), s);
1207 assert_eq!(u_as_string.utf16_units().collect::<Vec<u16>>(), u);
1212 fn test_utf16_invalid() {
1213 // completely positive cases tested above.
1215 assert!(String::from_utf16(&[0xD800]).is_err());
1217 assert!(String::from_utf16(&[0xD800, 0xD800]).is_err());
1220 assert!(String::from_utf16(&[0x0061, 0xDC00]).is_err());
1223 assert!(String::from_utf16(&[0xD800, 0xd801, 0xdc8b, 0xD800]).is_err());
1227 fn test_from_utf16_lossy() {
1228 // completely positive cases tested above.
1230 assert_eq!(String::from_utf16_lossy(&[0xD800]), String::from_str("\u{FFFD}"));
1232 assert_eq!(String::from_utf16_lossy(&[0xD800, 0xD800]),
1233 String::from_str("\u{FFFD}\u{FFFD}"));
1236 assert_eq!(String::from_utf16_lossy(&[0x0061, 0xDC00]), String::from_str("a\u{FFFD}"));
1239 assert_eq!(String::from_utf16_lossy(&[0xD800, 0xd801, 0xdc8b, 0xD800]),
1240 String::from_str("\u{FFFD}𐒋\u{FFFD}"));
1244 fn test_from_buf_len() {
1246 let a = vec![65u8, 65, 65, 65, 65, 65, 65, 0];
1247 assert_eq!(super::raw::from_buf_len(a.as_ptr(), 3), String::from_str("AAA"));
1252 fn test_from_buf() {
1254 let a = vec![65, 65, 65, 65, 65, 65, 65, 0];
1256 let c = super::raw::from_buf(b);
1257 assert_eq!(c, String::from_str("AAAAAAA"));
1262 fn test_push_bytes() {
1263 let mut s = String::from_str("ABC");
1265 let mv = s.as_mut_vec();
1266 mv.push_all(&[b'D']);
1268 assert_eq!(s, "ABCD");
1272 fn test_push_str() {
1273 let mut s = String::new();
1275 assert_eq!(s.slice_from(0), "");
1277 assert_eq!(s.slice_from(0), "abc");
1278 s.push_str("ประเทศไทย中华Việt Nam");
1279 assert_eq!(s.slice_from(0), "abcประเทศไทย中华Việt Nam");
1284 let mut data = String::from_str("ประเทศไทย中");
1286 data.push('b'); // 1 byte
1287 data.push('¢'); // 2 byte
1288 data.push('€'); // 3 byte
1289 data.push('𤭢'); // 4 byte
1290 assert_eq!(data, "ประเทศไทย中华b¢€𤭢");
1295 let mut data = String::from_str("ประเทศไทย中华b¢€𤭢");
1296 assert_eq!(data.pop().unwrap(), '𤭢'); // 4 bytes
1297 assert_eq!(data.pop().unwrap(), '€'); // 3 bytes
1298 assert_eq!(data.pop().unwrap(), '¢'); // 2 bytes
1299 assert_eq!(data.pop().unwrap(), 'b'); // 1 bytes
1300 assert_eq!(data.pop().unwrap(), '华');
1301 assert_eq!(data, "ประเทศไทย中");
1305 fn test_str_truncate() {
1306 let mut s = String::from_str("12345");
1308 assert_eq!(s, "12345");
1310 assert_eq!(s, "123");
1314 let mut s = String::from_str("12345");
1318 let p_ = s.as_ptr();
1324 fn test_str_truncate_invalid_len() {
1325 let mut s = String::from_str("12345");
1331 fn test_str_truncate_split_codepoint() {
1332 let mut s = String::from_str("\u{FC}"); // ü
1337 fn test_str_clear() {
1338 let mut s = String::from_str("12345");
1340 assert_eq!(s.len(), 0);
1346 let a = String::from_str("12345");
1349 assert_eq!(b.len(), 7);
1350 assert_eq!(b, "1234522");
1355 let mut s = "ศไทย中华Việt Nam; foobar".to_string();;
1356 assert_eq!(s.remove(0), 'ศ');
1357 assert_eq!(s.len(), 33);
1358 assert_eq!(s, "ไทย中华Việt Nam; foobar");
1359 assert_eq!(s.remove(17), 'ệ');
1360 assert_eq!(s, "ไทย中华Vit Nam; foobar");
1363 #[test] #[should_fail]
1365 "ศ".to_string().remove(1);
1370 let mut s = "foobar".to_string();
1372 assert_eq!(s, "ệfoobar");
1374 assert_eq!(s, "ệfooยbar");
1377 #[test] #[should_fail] fn insert_bad1() { "".to_string().insert(1, 't'); }
1378 #[test] #[should_fail] fn insert_bad2() { "ệ".to_string().insert(1, 't'); }
1382 let s = "foobar".to_string();
1383 assert_eq!("foobar", s[]);
1384 assert_eq!("foo", s[..3]);
1385 assert_eq!("bar", s[3..]);
1386 assert_eq!("oob", s[1..4]);
1390 fn test_simple_types() {
1391 assert_eq!(1i.to_string(), "1");
1392 assert_eq!((-1i).to_string(), "-1");
1393 assert_eq!(200u.to_string(), "200");
1394 assert_eq!(2u8.to_string(), "2");
1395 assert_eq!(true.to_string(), "true");
1396 assert_eq!(false.to_string(), "false");
1397 assert_eq!(().to_string(), "()");
1398 assert_eq!(("hi".to_string()).to_string(), "hi");
1403 let x: Vec<int> = vec![];
1404 assert_eq!(x.to_string(), "[]");
1405 assert_eq!((vec![1i]).to_string(), "[1]");
1406 assert_eq!((vec![1i, 2, 3]).to_string(), "[1, 2, 3]");
1407 assert!((vec![vec![], vec![1i], vec![1i, 1]]).to_string() ==
1408 "[[], [1], [1, 1]]");
1412 fn test_from_iterator() {
1413 let s = "ศไทย中华Việt Nam".to_string();
1417 let a: String = s.chars().collect();
1420 let mut b = t.to_string();
1421 b.extend(u.chars());
1424 let c: String = vec![t, u].into_iter().collect();
1427 let mut d = t.to_string();
1428 d.extend(vec![u].into_iter());
1433 fn bench_with_capacity(b: &mut Bencher) {
1435 String::with_capacity(100)
1440 fn bench_push_str(b: &mut Bencher) {
1441 let s = "ศไทย中华Việt Nam; Mary had a little lamb, Little lamb";
1443 let mut r = String::new();
1448 const REPETITIONS: u64 = 10_000;
1451 fn bench_push_str_one_byte(b: &mut Bencher) {
1452 b.bytes = REPETITIONS;
1454 let mut r = String::new();
1455 for _ in range(0, REPETITIONS) {
1462 fn bench_push_char_one_byte(b: &mut Bencher) {
1463 b.bytes = REPETITIONS;
1465 let mut r = String::new();
1466 for _ in range(0, REPETITIONS) {
1473 fn bench_push_char_two_bytes(b: &mut Bencher) {
1474 b.bytes = REPETITIONS * 2;
1476 let mut r = String::new();
1477 for _ in range(0, REPETITIONS) {
1484 fn from_utf8_lossy_100_ascii(b: &mut Bencher) {
1485 let s = b"Hello there, the quick brown fox jumped over the lazy dog! \
1486 Lorem ipsum dolor sit amet, consectetur. ";
1488 assert_eq!(100, s.len());
1490 let _ = String::from_utf8_lossy(s);
1495 fn from_utf8_lossy_100_multibyte(b: &mut Bencher) {
1496 let s = "𐌀𐌖𐌋𐌄𐌑𐌉ปรدولة الكويتทศไทย中华𐍅𐌿𐌻𐍆𐌹𐌻𐌰".as_bytes();
1497 assert_eq!(100, s.len());
1499 let _ = String::from_utf8_lossy(s);
1504 fn from_utf8_lossy_invalid(b: &mut Bencher) {
1505 let s = b"Hello\xC0\x80 There\xE6\x83 Goodbye";
1507 let _ = String::from_utf8_lossy(s);
1512 fn from_utf8_lossy_100_invalid(b: &mut Bencher) {
1513 let s = Vec::from_elem(100, 0xF5u8);
1515 let _ = String::from_utf8_lossy(s.as_slice());