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;
23 use core::iter::FromIterator;
25 use core::ops::{self, Deref, Add};
27 use core::raw::Slice as RawSlice;
28 use unicode::str as unicode_str;
29 use unicode::str::Utf16Item;
31 use str::{self, CharRange, FromStr, Utf8Error};
32 use vec::{DerefVec, Vec, as_vec};
34 /// A growable string stored as a UTF-8 encoded buffer.
35 #[derive(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: ::slice::SliceExt::to_vec(string.as_bytes()) }
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 /// 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> {
144 match str::from_utf8(v) {
145 Ok(s) => return Cow::Borrowed(s),
147 if let Utf8Error::InvalidByte(firstbad) = e {
153 static TAG_CONT_U8: u8 = 128u8;
154 static REPLACEMENT: &'static [u8] = b"\xEF\xBF\xBD"; // U+FFFD in UTF-8
156 fn unsafe_get(xs: &[u8], i: uint) -> u8 {
157 unsafe { *xs.get_unchecked(i) }
159 fn safe_get(xs: &[u8], i: uint, total: uint) -> u8 {
167 let mut res = String::with_capacity(total);
171 res.as_mut_vec().push_all(v[..i])
175 // subseqidx is the index of the first byte of the subsequence we're looking at.
176 // It's used to copy a bunch of contiguous good codepoints at once instead of copying
178 let mut subseqidx = i;
182 let byte = unsafe_get(v, i);
185 macro_rules! error(() => ({
188 res.as_mut_vec().push_all(v[subseqidx..i_]);
191 res.as_mut_vec().push_all(REPLACEMENT);
196 // subseqidx handles this
198 let w = unicode_str::utf8_char_width(byte);
202 if safe_get(v, i, total) & 192u8 != TAG_CONT_U8 {
209 match (byte, safe_get(v, i, total)) {
210 (0xE0 , 0xA0 ... 0xBF) => (),
211 (0xE1 ... 0xEC, 0x80 ... 0xBF) => (),
212 (0xED , 0x80 ... 0x9F) => (),
213 (0xEE ... 0xEF, 0x80 ... 0xBF) => (),
220 if safe_get(v, i, total) & 192u8 != TAG_CONT_U8 {
227 match (byte, safe_get(v, i, total)) {
228 (0xF0 , 0x90 ... 0xBF) => (),
229 (0xF1 ... 0xF3, 0x80 ... 0xBF) => (),
230 (0xF4 , 0x80 ... 0x8F) => (),
237 if safe_get(v, i, total) & 192u8 != TAG_CONT_U8 {
242 if safe_get(v, i, total) & 192u8 != TAG_CONT_U8 {
255 if subseqidx < total {
257 res.as_mut_vec().push_all(v[subseqidx..total])
263 /// Decode a UTF-16 encoded vector `v` into a `String`, returning `None`
264 /// if `v` contains any invalid data.
270 /// let mut v = &mut [0xD834, 0xDD1E, 0x006d, 0x0075,
271 /// 0x0073, 0x0069, 0x0063];
272 /// assert_eq!(String::from_utf16(v).unwrap(),
273 /// "𝄞music".to_string());
275 /// // 𝄞mu<invalid>ic
277 /// assert!(String::from_utf16(v).is_err());
280 pub fn from_utf16(v: &[u16]) -> Result<String, FromUtf16Error> {
281 let mut s = String::with_capacity(v.len());
282 for c in unicode_str::utf16_items(v) {
284 Utf16Item::ScalarValue(c) => s.push(c),
285 Utf16Item::LoneSurrogate(_) => return Err(FromUtf16Error(())),
291 /// Decode a UTF-16 encoded vector `v` into a string, replacing
292 /// invalid data with the replacement character (U+FFFD).
297 /// // 𝄞mus<invalid>ic<invalid>
298 /// let v = &[0xD834, 0xDD1E, 0x006d, 0x0075,
299 /// 0x0073, 0xDD1E, 0x0069, 0x0063,
302 /// assert_eq!(String::from_utf16_lossy(v),
303 /// "𝄞mus\u{FFFD}ic\u{FFFD}".to_string());
306 pub fn from_utf16_lossy(v: &[u16]) -> String {
307 unicode_str::utf16_items(v).map(|c| c.to_char_lossy()).collect()
310 /// Creates a new `String` from a length, capacity, and pointer.
312 /// This is unsafe because:
313 /// * We call `Vec::from_raw_parts` to get a `Vec<u8>`;
314 /// * We assume that the `Vec` contains valid UTF-8.
317 pub unsafe fn from_raw_parts(buf: *mut u8, length: uint, capacity: uint) -> String {
319 vec: Vec::from_raw_parts(buf, length, capacity),
323 /// Creates a `String` from a null-terminated `*const u8` buffer.
325 /// This function is unsafe because we dereference memory until we find the
326 /// NUL character, which is not guaranteed to be present. Additionally, the
327 /// slice is not checked to see whether it contains valid UTF-8
328 #[unstable = "just renamed from `mod raw`"]
329 pub unsafe fn from_raw_buf(buf: *const u8) -> String {
330 String::from_str(str::from_c_str(buf as *const i8))
333 /// Creates a `String` from a `*const u8` buffer of the given length.
335 /// This function is unsafe because it blindly assumes the validity of the
336 /// pointer `buf` for `len` bytes of memory. This function will copy the
337 /// memory from `buf` into a new allocation (owned by the returned
340 /// This function is also unsafe because it does not validate that the
341 /// buffer is valid UTF-8 encoded data.
342 #[unstable = "just renamed from `mod raw`"]
343 pub unsafe fn from_raw_buf_len(buf: *const u8, len: uint) -> String {
344 String::from_utf8_unchecked(Vec::from_raw_buf(buf, len))
347 /// Converts a vector of bytes to a new `String` without checking if
348 /// it contains valid UTF-8. This is unsafe because it assumes that
349 /// the UTF-8-ness of the vector has already been validated.
352 pub unsafe fn from_utf8_unchecked(bytes: Vec<u8>) -> String {
353 String { vec: bytes }
356 /// Return the underlying byte buffer, encoded as UTF-8.
361 /// let s = String::from_str("hello");
362 /// let bytes = s.into_bytes();
363 /// assert_eq!(bytes, vec![104, 101, 108, 108, 111]);
367 pub fn into_bytes(self) -> Vec<u8> {
371 /// Pushes the given string onto this string buffer.
376 /// let mut s = String::from_str("foo");
377 /// s.push_str("bar");
378 /// assert_eq!(s.as_slice(), "foobar");
382 pub fn push_str(&mut self, string: &str) {
383 self.vec.push_all(string.as_bytes())
386 /// Returns the number of bytes that this string buffer can hold without
392 /// let s = String::with_capacity(10);
393 /// assert!(s.capacity() >= 10);
397 pub fn capacity(&self) -> uint {
401 /// Reserves capacity for at least `additional` more bytes to be inserted
402 /// in the given `String`. The collection may reserve more space to avoid
403 /// frequent reallocations.
407 /// Panics if the new capacity overflows `uint`.
412 /// let mut s = String::new();
414 /// assert!(s.capacity() >= 10);
418 pub fn reserve(&mut self, additional: uint) {
419 self.vec.reserve(additional)
422 /// Reserves the minimum capacity for exactly `additional` more bytes to be
423 /// inserted in the given `String`. Does nothing if the capacity is already
426 /// Note that the allocator may give the collection more space than it
427 /// requests. Therefore capacity can not be relied upon to be precisely
428 /// minimal. Prefer `reserve` if future insertions are expected.
432 /// Panics if the new capacity overflows `uint`.
437 /// let mut s = String::new();
439 /// assert!(s.capacity() >= 10);
443 pub fn reserve_exact(&mut self, additional: uint) {
444 self.vec.reserve_exact(additional)
447 /// Shrinks the capacity of this string buffer to match its length.
452 /// let mut s = String::from_str("foo");
454 /// assert!(s.capacity() >= 100);
455 /// s.shrink_to_fit();
456 /// assert_eq!(s.capacity(), 3);
460 pub fn shrink_to_fit(&mut self) {
461 self.vec.shrink_to_fit()
464 /// Adds the given character to the end of the string.
469 /// let mut s = String::from_str("abc");
473 /// assert_eq!(s.as_slice(), "abc123");
477 pub fn push(&mut self, ch: char) {
478 if (ch as u32) < 0x80 {
479 self.vec.push(ch as u8);
483 let cur_len = self.len();
484 // This may use up to 4 bytes.
488 // Attempt to not use an intermediate buffer by just pushing bytes
489 // directly onto this string.
490 let slice = RawSlice {
491 data: self.vec.as_ptr().offset(cur_len as int),
494 let used = ch.encode_utf8(mem::transmute(slice)).unwrap_or(0);
495 self.vec.set_len(cur_len + used);
499 /// Works with the underlying buffer as a byte slice.
504 /// let s = String::from_str("hello");
505 /// let b: &[_] = &[104, 101, 108, 108, 111];
506 /// assert_eq!(s.as_bytes(), b);
510 pub fn as_bytes<'a>(&'a self) -> &'a [u8] {
514 /// Shortens a string to the specified length.
518 /// Panics if `new_len` > current length,
519 /// or if `new_len` is not a character boundary.
524 /// let mut s = String::from_str("hello");
526 /// assert_eq!(s.as_slice(), "he");
530 pub fn truncate(&mut self, new_len: uint) {
531 assert!(self.is_char_boundary(new_len));
532 self.vec.truncate(new_len)
535 /// Removes the last character from the string buffer and returns it.
536 /// Returns `None` if this string buffer is empty.
541 /// let mut s = String::from_str("foo");
542 /// assert_eq!(s.pop(), Some('o'));
543 /// assert_eq!(s.pop(), Some('o'));
544 /// assert_eq!(s.pop(), Some('f'));
545 /// assert_eq!(s.pop(), None);
549 pub fn pop(&mut self) -> Option<char> {
550 let len = self.len();
555 let CharRange {ch, next} = self.char_range_at_reverse(len);
557 self.vec.set_len(next);
562 /// Removes the character from the string buffer at byte position `idx` and
567 /// This is an O(n) operation as it requires copying every element in the
572 /// If `idx` does not lie on a character boundary, or if it is out of
573 /// bounds, then this function will panic.
578 /// let mut s = String::from_str("foo");
579 /// assert_eq!(s.remove(0), 'f');
580 /// assert_eq!(s.remove(1), 'o');
581 /// assert_eq!(s.remove(0), 'o');
584 pub fn remove(&mut self, idx: uint) -> char {
585 let len = self.len();
588 let CharRange { ch, next } = self.char_range_at(idx);
590 ptr::copy_memory(self.vec.as_mut_ptr().offset(idx as int),
591 self.vec.as_ptr().offset(next as int),
593 self.vec.set_len(len - (next - idx));
598 /// Insert a character into the string buffer at byte position `idx`.
602 /// This is an O(n) operation as it requires copying every element in the
607 /// If `idx` does not lie on a character boundary or is out of bounds, then
608 /// this function will panic.
610 pub fn insert(&mut self, idx: uint, ch: char) {
611 let len = self.len();
613 assert!(self.is_char_boundary(idx));
615 let mut bits = [0; 4];
616 let amt = ch.encode_utf8(&mut bits).unwrap();
619 ptr::copy_memory(self.vec.as_mut_ptr().offset((idx + amt) as int),
620 self.vec.as_ptr().offset(idx as int),
622 ptr::copy_memory(self.vec.as_mut_ptr().offset(idx as int),
625 self.vec.set_len(len + amt);
629 /// Views the string buffer as a mutable sequence of bytes.
631 /// This is unsafe because it does not check
632 /// to ensure that the resulting string will be valid UTF-8.
637 /// let mut s = String::from_str("hello");
639 /// let vec = s.as_mut_vec();
640 /// assert!(vec == &mut vec![104, 101, 108, 108, 111]);
643 /// assert_eq!(s.as_slice(), "olleh");
646 pub unsafe fn as_mut_vec<'a>(&'a mut self) -> &'a mut Vec<u8> {
650 /// Return the number of bytes in this string.
655 /// let a = "foo".to_string();
656 /// assert_eq!(a.len(), 3);
660 pub fn len(&self) -> uint { self.vec.len() }
662 /// Returns true if the string contains no bytes
667 /// let mut v = String::new();
668 /// assert!(v.is_empty());
670 /// assert!(!v.is_empty());
673 pub fn is_empty(&self) -> bool { self.len() == 0 }
675 /// Truncates the string, returning it to 0 length.
680 /// let mut s = "foo".to_string();
682 /// assert!(s.is_empty());
686 pub fn clear(&mut self) {
692 /// Consume this error, returning the bytes that were attempted to make a
695 pub fn into_bytes(self) -> Vec<u8> { self.bytes }
697 /// Access the underlying UTF8-error that was the cause of this error.
699 pub fn utf8_error(&self) -> Utf8Error { self.error }
702 impl fmt::Show for FromUtf8Error {
703 fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
708 impl fmt::Show for FromUtf16Error {
709 fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
710 "invalid utf-16: lone surrogate found".fmt(f)
714 #[experimental = "waiting on FromIterator stabilization"]
715 impl FromIterator<char> for String {
716 fn from_iter<I:Iterator<Item=char>>(iterator: I) -> String {
717 let mut buf = String::new();
718 buf.extend(iterator);
723 #[experimental = "waiting on FromIterator stabilization"]
724 impl<'a> FromIterator<&'a str> for String {
725 fn from_iter<I:Iterator<Item=&'a str>>(iterator: I) -> String {
726 let mut buf = String::new();
727 buf.extend(iterator);
732 #[experimental = "waiting on Extend stabilization"]
733 impl Extend<char> for String {
734 fn extend<I:Iterator<Item=char>>(&mut self, mut iterator: I) {
735 let (lower_bound, _) = iterator.size_hint();
736 self.reserve(lower_bound);
743 #[experimental = "waiting on Extend stabilization"]
744 impl<'a> Extend<&'a str> for String {
745 fn extend<I: Iterator<Item=&'a str>>(&mut self, mut iterator: I) {
746 // A guess that at least one byte per iterator element will be needed.
747 let (lower_bound, _) = iterator.size_hint();
748 self.reserve(lower_bound);
756 impl PartialEq for String {
758 fn eq(&self, other: &String) -> bool { PartialEq::eq(&**self, &**other) }
760 fn ne(&self, other: &String) -> bool { PartialEq::ne(&**self, &**other) }
763 macro_rules! impl_eq {
764 ($lhs:ty, $rhs: ty) => {
766 impl<'a> PartialEq<$rhs> for $lhs {
768 fn eq(&self, other: &$rhs) -> bool { PartialEq::eq(&**self, &**other) }
770 fn ne(&self, other: &$rhs) -> bool { PartialEq::ne(&**self, &**other) }
774 impl<'a> PartialEq<$lhs> for $rhs {
776 fn eq(&self, other: &$lhs) -> bool { PartialEq::eq(&**self, &**other) }
778 fn ne(&self, other: &$lhs) -> bool { PartialEq::ne(&**self, &**other) }
784 impl_eq! { String, &'a str }
785 impl_eq! { CowString<'a>, String }
788 impl<'a, 'b> PartialEq<&'b str> for CowString<'a> {
790 fn eq(&self, other: &&'b str) -> bool { PartialEq::eq(&**self, &**other) }
792 fn ne(&self, other: &&'b str) -> bool { PartialEq::ne(&**self, &**other) }
796 impl<'a, 'b> PartialEq<CowString<'a>> for &'b str {
798 fn eq(&self, other: &CowString<'a>) -> bool { PartialEq::eq(&**self, &**other) }
800 fn ne(&self, other: &CowString<'a>) -> bool { PartialEq::ne(&**self, &**other) }
803 #[experimental = "waiting on Str stabilization"]
804 impl Str for String {
807 fn as_slice<'a>(&'a self) -> &'a str {
808 unsafe { mem::transmute(self.vec.as_slice()) }
813 impl Default for String {
815 fn default() -> String {
820 #[experimental = "waiting on Show stabilization"]
821 impl fmt::Show for String {
822 fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
827 #[experimental = "waiting on Hash stabilization"]
828 impl<H: hash::Writer> hash::Hash<H> for String {
830 fn hash(&self, hasher: &mut H) {
831 (**self).hash(hasher)
835 #[experimental = "waiting on Add stabilization"]
836 impl<'a> Add<&'a str> for String {
837 type Output = String;
839 fn add(mut self, other: &str) -> String {
840 self.push_str(other);
845 impl ops::Slice<uint, str> for String {
847 fn as_slice_<'a>(&'a self) -> &'a str {
848 unsafe { mem::transmute(self.vec.as_slice()) }
852 fn slice_from_or_fail<'a>(&'a self, from: &uint) -> &'a str {
857 fn slice_to_or_fail<'a>(&'a self, to: &uint) -> &'a str {
862 fn slice_or_fail<'a>(&'a self, from: &uint, to: &uint) -> &'a str {
867 #[experimental = "waiting on Deref stabilization"]
868 impl ops::Deref for String {
871 fn deref<'a>(&'a self) -> &'a str {
872 unsafe { mem::transmute(self.vec[]) }
876 /// Wrapper type providing a `&String` reference via `Deref`.
878 pub struct DerefString<'a> {
882 impl<'a> Deref for DerefString<'a> {
883 type Target = String;
885 fn deref<'b>(&'b self) -> &'b String {
886 unsafe { mem::transmute(&*self.x) }
890 /// Convert a string slice to a wrapper type providing a `&String` reference.
895 /// use std::string::as_string;
897 /// fn string_consumer(s: String) {
898 /// assert_eq!(s, "foo".to_string());
901 /// let string = as_string("foo").clone();
902 /// string_consumer(string);
905 pub fn as_string<'a>(x: &'a str) -> DerefString<'a> {
906 DerefString { x: as_vec(x.as_bytes()) }
909 impl FromStr for String {
911 fn from_str(s: &str) -> Option<String> {
912 Some(String::from_str(s))
916 /// A generic trait for converting a value to a string
918 /// Converts the value of `self` to an owned string
919 fn to_string(&self) -> String;
922 impl<T: fmt::Show> ToString for T {
923 fn to_string(&self) -> String {
924 use core::fmt::Writer;
925 let mut buf = String::new();
926 let _ = buf.write_fmt(format_args!("{}", self));
932 impl IntoCow<'static, String, str> for String {
933 fn into_cow(self) -> CowString<'static> {
938 impl<'a> IntoCow<'a, String, str> for &'a str {
939 fn into_cow(self) -> CowString<'a> {
944 /// A clone-on-write string
946 pub type CowString<'a> = Cow<'a, String, str>;
948 impl<'a> Str for CowString<'a> {
950 fn as_slice<'b>(&'b self) -> &'b str {
955 impl fmt::Writer for String {
956 fn write_str(&mut self, s: &str) -> fmt::Result {
968 use core::iter::repeat;
969 use super::{as_string, CowString};
972 fn test_as_string() {
974 assert_eq!(x, as_string(x).as_slice());
979 let owned: Option<::std::string::String> = "string".parse();
980 assert_eq!(owned.as_ref().map(|s| s.as_slice()), Some("string"));
984 fn test_from_utf8() {
985 let xs = b"hello".to_vec();
986 assert_eq!(String::from_utf8(xs).unwrap(),
987 String::from_str("hello"));
989 let xs = "ศไทย中华Việt Nam".as_bytes().to_vec();
990 assert_eq!(String::from_utf8(xs).unwrap(),
991 String::from_str("ศไทย中华Việt Nam"));
993 let xs = b"hello\xFF".to_vec();
994 let err = String::from_utf8(xs).err().unwrap();
995 assert_eq!(err.utf8_error(), Utf8Error::TooShort);
996 assert_eq!(err.into_bytes(), b"hello\xff".to_vec());
1000 fn test_from_utf8_lossy() {
1002 let ys: CowString = "hello".into_cow();
1003 assert_eq!(String::from_utf8_lossy(xs), ys);
1005 let xs = "ศไทย中华Việt Nam".as_bytes();
1006 let ys: CowString = "ศไทย中华Việt Nam".into_cow();
1007 assert_eq!(String::from_utf8_lossy(xs), ys);
1009 let xs = b"Hello\xC2 There\xFF Goodbye";
1010 assert_eq!(String::from_utf8_lossy(xs),
1011 String::from_str("Hello\u{FFFD} There\u{FFFD} Goodbye").into_cow());
1013 let xs = b"Hello\xC0\x80 There\xE6\x83 Goodbye";
1014 assert_eq!(String::from_utf8_lossy(xs),
1015 String::from_str("Hello\u{FFFD}\u{FFFD} There\u{FFFD} Goodbye").into_cow());
1017 let xs = b"\xF5foo\xF5\x80bar";
1018 assert_eq!(String::from_utf8_lossy(xs),
1019 String::from_str("\u{FFFD}foo\u{FFFD}\u{FFFD}bar").into_cow());
1021 let xs = b"\xF1foo\xF1\x80bar\xF1\x80\x80baz";
1022 assert_eq!(String::from_utf8_lossy(xs),
1023 String::from_str("\u{FFFD}foo\u{FFFD}bar\u{FFFD}baz").into_cow());
1025 let xs = b"\xF4foo\xF4\x80bar\xF4\xBFbaz";
1026 assert_eq!(String::from_utf8_lossy(xs),
1027 String::from_str("\u{FFFD}foo\u{FFFD}bar\u{FFFD}\u{FFFD}baz").into_cow());
1029 let xs = b"\xF0\x80\x80\x80foo\xF0\x90\x80\x80bar";
1030 assert_eq!(String::from_utf8_lossy(xs), String::from_str("\u{FFFD}\u{FFFD}\u{FFFD}\u{FFFD}\
1031 foo\u{10000}bar").into_cow());
1034 let xs = b"\xED\xA0\x80foo\xED\xBF\xBFbar";
1035 assert_eq!(String::from_utf8_lossy(xs), String::from_str("\u{FFFD}\u{FFFD}\u{FFFD}foo\
1036 \u{FFFD}\u{FFFD}\u{FFFD}bar").into_cow());
1040 fn test_from_utf16() {
1042 [(String::from_str("𐍅𐌿𐌻𐍆𐌹𐌻𐌰\n"),
1043 vec![0xd800_u16, 0xdf45_u16, 0xd800_u16, 0xdf3f_u16,
1044 0xd800_u16, 0xdf3b_u16, 0xd800_u16, 0xdf46_u16,
1045 0xd800_u16, 0xdf39_u16, 0xd800_u16, 0xdf3b_u16,
1046 0xd800_u16, 0xdf30_u16, 0x000a_u16]),
1048 (String::from_str("𐐒𐑉𐐮𐑀𐐲𐑋 𐐏𐐲𐑍\n"),
1049 vec![0xd801_u16, 0xdc12_u16, 0xd801_u16,
1050 0xdc49_u16, 0xd801_u16, 0xdc2e_u16, 0xd801_u16,
1051 0xdc40_u16, 0xd801_u16, 0xdc32_u16, 0xd801_u16,
1052 0xdc4b_u16, 0x0020_u16, 0xd801_u16, 0xdc0f_u16,
1053 0xd801_u16, 0xdc32_u16, 0xd801_u16, 0xdc4d_u16,
1056 (String::from_str("𐌀𐌖𐌋𐌄𐌑𐌉·𐌌𐌄𐌕𐌄𐌋𐌉𐌑\n"),
1057 vec![0xd800_u16, 0xdf00_u16, 0xd800_u16, 0xdf16_u16,
1058 0xd800_u16, 0xdf0b_u16, 0xd800_u16, 0xdf04_u16,
1059 0xd800_u16, 0xdf11_u16, 0xd800_u16, 0xdf09_u16,
1060 0x00b7_u16, 0xd800_u16, 0xdf0c_u16, 0xd800_u16,
1061 0xdf04_u16, 0xd800_u16, 0xdf15_u16, 0xd800_u16,
1062 0xdf04_u16, 0xd800_u16, 0xdf0b_u16, 0xd800_u16,
1063 0xdf09_u16, 0xd800_u16, 0xdf11_u16, 0x000a_u16 ]),
1065 (String::from_str("𐒋𐒘𐒈𐒑𐒛𐒒 𐒕𐒓 𐒈𐒚𐒍 𐒏𐒜𐒒𐒖𐒆 𐒕𐒆\n"),
1066 vec![0xd801_u16, 0xdc8b_u16, 0xd801_u16, 0xdc98_u16,
1067 0xd801_u16, 0xdc88_u16, 0xd801_u16, 0xdc91_u16,
1068 0xd801_u16, 0xdc9b_u16, 0xd801_u16, 0xdc92_u16,
1069 0x0020_u16, 0xd801_u16, 0xdc95_u16, 0xd801_u16,
1070 0xdc93_u16, 0x0020_u16, 0xd801_u16, 0xdc88_u16,
1071 0xd801_u16, 0xdc9a_u16, 0xd801_u16, 0xdc8d_u16,
1072 0x0020_u16, 0xd801_u16, 0xdc8f_u16, 0xd801_u16,
1073 0xdc9c_u16, 0xd801_u16, 0xdc92_u16, 0xd801_u16,
1074 0xdc96_u16, 0xd801_u16, 0xdc86_u16, 0x0020_u16,
1075 0xd801_u16, 0xdc95_u16, 0xd801_u16, 0xdc86_u16,
1077 // Issue #12318, even-numbered non-BMP planes
1078 (String::from_str("\u{20000}"),
1079 vec![0xD840, 0xDC00])];
1081 for p in pairs.iter() {
1082 let (s, u) = (*p).clone();
1083 let s_as_utf16 = s.utf16_units().collect::<Vec<u16>>();
1084 let u_as_string = String::from_utf16(u.as_slice()).unwrap();
1086 assert!(::unicode::str::is_utf16(u.as_slice()));
1087 assert_eq!(s_as_utf16, u);
1089 assert_eq!(u_as_string, s);
1090 assert_eq!(String::from_utf16_lossy(u.as_slice()), s);
1092 assert_eq!(String::from_utf16(s_as_utf16.as_slice()).unwrap(), s);
1093 assert_eq!(u_as_string.utf16_units().collect::<Vec<u16>>(), u);
1098 fn test_utf16_invalid() {
1099 // completely positive cases tested above.
1101 assert!(String::from_utf16(&[0xD800]).is_err());
1103 assert!(String::from_utf16(&[0xD800, 0xD800]).is_err());
1106 assert!(String::from_utf16(&[0x0061, 0xDC00]).is_err());
1109 assert!(String::from_utf16(&[0xD800, 0xd801, 0xdc8b, 0xD800]).is_err());
1113 fn test_from_utf16_lossy() {
1114 // completely positive cases tested above.
1116 assert_eq!(String::from_utf16_lossy(&[0xD800]), String::from_str("\u{FFFD}"));
1118 assert_eq!(String::from_utf16_lossy(&[0xD800, 0xD800]),
1119 String::from_str("\u{FFFD}\u{FFFD}"));
1122 assert_eq!(String::from_utf16_lossy(&[0x0061, 0xDC00]), String::from_str("a\u{FFFD}"));
1125 assert_eq!(String::from_utf16_lossy(&[0xD800, 0xd801, 0xdc8b, 0xD800]),
1126 String::from_str("\u{FFFD}𐒋\u{FFFD}"));
1130 fn test_from_buf_len() {
1132 let a = vec![65u8, 65, 65, 65, 65, 65, 65, 0];
1133 assert_eq!(String::from_raw_buf_len(a.as_ptr(), 3), String::from_str("AAA"));
1138 fn test_from_buf() {
1140 let a = vec![65, 65, 65, 65, 65, 65, 65, 0];
1142 let c = String::from_raw_buf(b);
1143 assert_eq!(c, String::from_str("AAAAAAA"));
1148 fn test_push_bytes() {
1149 let mut s = String::from_str("ABC");
1151 let mv = s.as_mut_vec();
1152 mv.push_all(&[b'D']);
1154 assert_eq!(s, "ABCD");
1158 fn test_push_str() {
1159 let mut s = String::new();
1161 assert_eq!(s.slice_from(0), "");
1163 assert_eq!(s.slice_from(0), "abc");
1164 s.push_str("ประเทศไทย中华Việt Nam");
1165 assert_eq!(s.slice_from(0), "abcประเทศไทย中华Việt Nam");
1170 let mut data = String::from_str("ประเทศไทย中");
1172 data.push('b'); // 1 byte
1173 data.push('¢'); // 2 byte
1174 data.push('€'); // 3 byte
1175 data.push('𤭢'); // 4 byte
1176 assert_eq!(data, "ประเทศไทย中华b¢€𤭢");
1181 let mut data = String::from_str("ประเทศไทย中华b¢€𤭢");
1182 assert_eq!(data.pop().unwrap(), '𤭢'); // 4 bytes
1183 assert_eq!(data.pop().unwrap(), '€'); // 3 bytes
1184 assert_eq!(data.pop().unwrap(), '¢'); // 2 bytes
1185 assert_eq!(data.pop().unwrap(), 'b'); // 1 bytes
1186 assert_eq!(data.pop().unwrap(), '华');
1187 assert_eq!(data, "ประเทศไทย中");
1191 fn test_str_truncate() {
1192 let mut s = String::from_str("12345");
1194 assert_eq!(s, "12345");
1196 assert_eq!(s, "123");
1200 let mut s = String::from_str("12345");
1204 let p_ = s.as_ptr();
1210 fn test_str_truncate_invalid_len() {
1211 let mut s = String::from_str("12345");
1217 fn test_str_truncate_split_codepoint() {
1218 let mut s = String::from_str("\u{FC}"); // ü
1223 fn test_str_clear() {
1224 let mut s = String::from_str("12345");
1226 assert_eq!(s.len(), 0);
1232 let a = String::from_str("12345");
1235 assert_eq!(b.len(), 7);
1236 assert_eq!(b, "1234522");
1241 let mut s = "ศไทย中华Việt Nam; foobar".to_string();;
1242 assert_eq!(s.remove(0), 'ศ');
1243 assert_eq!(s.len(), 33);
1244 assert_eq!(s, "ไทย中华Việt Nam; foobar");
1245 assert_eq!(s.remove(17), 'ệ');
1246 assert_eq!(s, "ไทย中华Vit Nam; foobar");
1249 #[test] #[should_fail]
1251 "ศ".to_string().remove(1);
1256 let mut s = "foobar".to_string();
1258 assert_eq!(s, "ệfoobar");
1260 assert_eq!(s, "ệfooยbar");
1263 #[test] #[should_fail] fn insert_bad1() { "".to_string().insert(1, 't'); }
1264 #[test] #[should_fail] fn insert_bad2() { "ệ".to_string().insert(1, 't'); }
1268 let s = "foobar".to_string();
1269 assert_eq!("foobar", s[]);
1270 assert_eq!("foo", s[..3]);
1271 assert_eq!("bar", s[3..]);
1272 assert_eq!("oob", s[1..4]);
1276 fn test_simple_types() {
1277 assert_eq!(1i.to_string(), "1");
1278 assert_eq!((-1i).to_string(), "-1");
1279 assert_eq!(200u.to_string(), "200");
1280 assert_eq!(2u8.to_string(), "2");
1281 assert_eq!(true.to_string(), "true");
1282 assert_eq!(false.to_string(), "false");
1283 assert_eq!(().to_string(), "()");
1284 assert_eq!(("hi".to_string()).to_string(), "hi");
1289 let x: Vec<int> = vec![];
1290 assert_eq!(x.to_string(), "[]");
1291 assert_eq!((vec![1i]).to_string(), "[1]");
1292 assert_eq!((vec![1i, 2, 3]).to_string(), "[1, 2, 3]");
1293 assert!((vec![vec![], vec![1i], vec![1i, 1]]).to_string() ==
1294 "[[], [1], [1, 1]]");
1298 fn test_from_iterator() {
1299 let s = "ศไทย中华Việt Nam".to_string();
1303 let a: String = s.chars().collect();
1306 let mut b = t.to_string();
1307 b.extend(u.chars());
1310 let c: String = vec![t, u].into_iter().collect();
1313 let mut d = t.to_string();
1314 d.extend(vec![u].into_iter());
1319 fn bench_with_capacity(b: &mut Bencher) {
1321 String::with_capacity(100)
1326 fn bench_push_str(b: &mut Bencher) {
1327 let s = "ศไทย中华Việt Nam; Mary had a little lamb, Little lamb";
1329 let mut r = String::new();
1334 const REPETITIONS: u64 = 10_000;
1337 fn bench_push_str_one_byte(b: &mut Bencher) {
1338 b.bytes = REPETITIONS;
1340 let mut r = String::new();
1341 for _ in range(0, REPETITIONS) {
1348 fn bench_push_char_one_byte(b: &mut Bencher) {
1349 b.bytes = REPETITIONS;
1351 let mut r = String::new();
1352 for _ in range(0, REPETITIONS) {
1359 fn bench_push_char_two_bytes(b: &mut Bencher) {
1360 b.bytes = REPETITIONS * 2;
1362 let mut r = String::new();
1363 for _ in range(0, REPETITIONS) {
1370 fn from_utf8_lossy_100_ascii(b: &mut Bencher) {
1371 let s = b"Hello there, the quick brown fox jumped over the lazy dog! \
1372 Lorem ipsum dolor sit amet, consectetur. ";
1374 assert_eq!(100, s.len());
1376 let _ = String::from_utf8_lossy(s);
1381 fn from_utf8_lossy_100_multibyte(b: &mut Bencher) {
1382 let s = "𐌀𐌖𐌋𐌄𐌑𐌉ปรدولة الكويتทศไทย中华𐍅𐌿𐌻𐍆𐌹𐌻𐌰".as_bytes();
1383 assert_eq!(100, s.len());
1385 let _ = String::from_utf8_lossy(s);
1390 fn from_utf8_lossy_invalid(b: &mut Bencher) {
1391 let s = b"Hello\xC0\x80 There\xE6\x83 Goodbye";
1393 let _ = String::from_utf8_lossy(s);
1398 fn from_utf8_lossy_100_invalid(b: &mut Bencher) {
1399 let s = repeat(0xf5u8).take(100).collect::<Vec<_>>();
1401 let _ = String::from_utf8_lossy(s.as_slice());