2019-04-22 14:29:04 +00:00
|
|
|
|
// ignore-tidy-filelength
|
|
|
|
|
|
2019-02-08 13:53:55 +00:00
|
|
|
|
//! String manipulation.
|
2014-05-01 06:06:36 +00:00
|
|
|
|
//!
|
2019-02-08 13:53:55 +00:00
|
|
|
|
//! For more details, see the `std::str` module.
|
2014-05-01 06:06:36 +00:00
|
|
|
|
|
2015-06-09 18:18:03 +00:00
|
|
|
|
#![stable(feature = "rust1", since = "1.0.0")]
|
2014-05-29 02:53:37 +00:00
|
|
|
|
|
2015-03-23 13:21:42 +00:00
|
|
|
|
use self::pattern::Pattern;
|
2019-12-22 22:42:04 +00:00
|
|
|
|
use self::pattern::{DoubleEndedSearcher, ReverseSearcher, SearchStep, Searcher};
|
2014-11-06 08:05:53 +00:00
|
|
|
|
|
2019-04-15 02:23:21 +00:00
|
|
|
|
use crate::char;
|
|
|
|
|
use crate::fmt::{self, Write};
|
2019-12-22 22:42:04 +00:00
|
|
|
|
use crate::iter::{Chain, FlatMap, Flatten};
|
|
|
|
|
use crate::iter::{Cloned, Filter, FusedIterator, Map, TrustedLen, TrustedRandomAccess};
|
2019-04-15 02:23:21 +00:00
|
|
|
|
use crate::mem;
|
|
|
|
|
use crate::ops::Try;
|
|
|
|
|
use crate::option;
|
2019-12-22 22:42:04 +00:00
|
|
|
|
use crate::slice::{self, SliceIndex, Split as SliceSplit};
|
2014-05-01 06:06:36 +00:00
|
|
|
|
|
2015-03-23 13:21:42 +00:00
|
|
|
|
pub mod pattern;
|
2014-12-18 01:12:53 +00:00
|
|
|
|
|
2019-12-21 11:16:18 +00:00
|
|
|
|
#[unstable(feature = "str_internals", issue = "none")]
|
2018-04-05 13:55:28 +00:00
|
|
|
|
#[allow(missing_docs)]
|
|
|
|
|
pub mod lossy;
|
|
|
|
|
|
2019-01-10 20:40:05 +00:00
|
|
|
|
/// Parse a value from a string
|
2015-11-20 08:41:10 +00:00
|
|
|
|
///
|
2017-03-12 18:04:52 +00:00
|
|
|
|
/// `FromStr`'s [`from_str`] method is often used implicitly, through
|
|
|
|
|
/// [`str`]'s [`parse`] method. See [`parse`]'s documentation for examples.
|
2015-11-20 08:41:10 +00:00
|
|
|
|
///
|
2017-03-12 18:04:52 +00:00
|
|
|
|
/// [`from_str`]: #tymethod.from_str
|
2016-03-08 07:55:52 +00:00
|
|
|
|
/// [`str`]: ../../std/primitive.str.html
|
2017-03-12 18:04:52 +00:00
|
|
|
|
/// [`parse`]: ../../std/primitive.str.html#method.parse
|
2017-03-25 15:56:52 +00:00
|
|
|
|
///
|
2019-01-10 20:40:05 +00:00
|
|
|
|
/// `FromStr` does not have a lifetime parameter, and so you can only parse types
|
|
|
|
|
/// that do not contain a lifetime parameter themselves. In other words, you can
|
|
|
|
|
/// parse an `i32` with `FromStr`, but not a `&i32`. You can parse a struct that
|
|
|
|
|
/// contains an `i32`, but not one that contains an `&i32`.
|
|
|
|
|
///
|
2017-03-25 15:56:52 +00:00
|
|
|
|
/// # Examples
|
|
|
|
|
///
|
|
|
|
|
/// Basic implementation of `FromStr` on an example `Point` type:
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// use std::str::FromStr;
|
|
|
|
|
/// use std::num::ParseIntError;
|
2017-03-25 16:22:23 +00:00
|
|
|
|
///
|
2017-03-25 15:56:52 +00:00
|
|
|
|
/// #[derive(Debug, PartialEq)]
|
|
|
|
|
/// struct Point {
|
|
|
|
|
/// x: i32,
|
|
|
|
|
/// y: i32
|
|
|
|
|
/// }
|
2017-03-25 16:22:23 +00:00
|
|
|
|
///
|
2017-03-25 15:56:52 +00:00
|
|
|
|
/// impl FromStr for Point {
|
|
|
|
|
/// type Err = ParseIntError;
|
2017-03-25 16:22:23 +00:00
|
|
|
|
///
|
2017-03-25 15:56:52 +00:00
|
|
|
|
/// fn from_str(s: &str) -> Result<Self, Self::Err> {
|
|
|
|
|
/// let coords: Vec<&str> = s.trim_matches(|p| p == '(' || p == ')' )
|
2018-08-22 10:55:47 +00:00
|
|
|
|
/// .split(',')
|
2017-03-25 15:56:52 +00:00
|
|
|
|
/// .collect();
|
2017-03-25 16:22:23 +00:00
|
|
|
|
///
|
2017-03-25 18:41:37 +00:00
|
|
|
|
/// let x_fromstr = coords[0].parse::<i32>()?;
|
|
|
|
|
/// let y_fromstr = coords[1].parse::<i32>()?;
|
2017-03-25 16:22:23 +00:00
|
|
|
|
///
|
2017-03-25 15:56:52 +00:00
|
|
|
|
/// Ok(Point { x: x_fromstr, y: y_fromstr })
|
|
|
|
|
/// }
|
|
|
|
|
/// }
|
2017-03-25 16:22:23 +00:00
|
|
|
|
///
|
2017-03-25 15:56:52 +00:00
|
|
|
|
/// let p = Point::from_str("(1,2)");
|
|
|
|
|
/// assert_eq!(p.unwrap(), Point{ x: 1, y: 2} )
|
|
|
|
|
/// ```
|
2015-01-28 06:52:32 +00:00
|
|
|
|
#[stable(feature = "rust1", since = "1.0.0")]
|
2015-08-07 13:27:27 +00:00
|
|
|
|
pub trait FromStr: Sized {
|
2015-01-28 06:52:32 +00:00
|
|
|
|
/// The associated error which can be returned from parsing.
|
|
|
|
|
#[stable(feature = "rust1", since = "1.0.0")]
|
|
|
|
|
type Err;
|
|
|
|
|
|
2015-05-10 22:22:04 +00:00
|
|
|
|
/// Parses a string `s` to return a value of this type.
|
|
|
|
|
///
|
2017-11-13 22:25:52 +00:00
|
|
|
|
/// If parsing succeeds, return the value inside [`Ok`], otherwise
|
2015-05-10 22:22:04 +00:00
|
|
|
|
/// when the string is ill-formatted return an error specific to the
|
2017-11-13 22:25:52 +00:00
|
|
|
|
/// inside [`Err`]. The error type is specific to implementation of the trait.
|
|
|
|
|
///
|
|
|
|
|
/// [`Ok`]: ../../std/result/enum.Result.html#variant.Ok
|
|
|
|
|
/// [`Err`]: ../../std/result/enum.Result.html#variant.Err
|
2015-09-30 21:42:41 +00:00
|
|
|
|
///
|
|
|
|
|
/// # Examples
|
|
|
|
|
///
|
|
|
|
|
/// Basic usage with [`i32`][ithirtytwo], a type that implements `FromStr`:
|
|
|
|
|
///
|
2016-03-08 07:55:52 +00:00
|
|
|
|
/// [ithirtytwo]: ../../std/primitive.i32.html
|
2015-09-30 21:42:41 +00:00
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// use std::str::FromStr;
|
|
|
|
|
///
|
|
|
|
|
/// let s = "5";
|
|
|
|
|
/// let x = i32::from_str(s).unwrap();
|
|
|
|
|
///
|
|
|
|
|
/// assert_eq!(5, x);
|
|
|
|
|
/// ```
|
2015-01-28 06:52:32 +00:00
|
|
|
|
#[stable(feature = "rust1", since = "1.0.0")]
|
|
|
|
|
fn from_str(s: &str) -> Result<Self, Self::Err>;
|
2014-11-15 04:52:00 +00:00
|
|
|
|
}
|
|
|
|
|
|
2015-01-28 06:52:32 +00:00
|
|
|
|
#[stable(feature = "rust1", since = "1.0.0")]
|
2014-11-15 04:52:00 +00:00
|
|
|
|
impl FromStr for bool {
|
2015-01-28 06:52:32 +00:00
|
|
|
|
type Err = ParseBoolError;
|
|
|
|
|
|
2014-11-15 04:52:00 +00:00
|
|
|
|
/// Parse a `bool` from a string.
|
|
|
|
|
///
|
2015-03-02 10:01:01 +00:00
|
|
|
|
/// Yields a `Result<bool, ParseBoolError>`, because `s` may or may not
|
|
|
|
|
/// actually be parseable.
|
2014-11-15 04:52:00 +00:00
|
|
|
|
///
|
|
|
|
|
/// # Examples
|
|
|
|
|
///
|
2015-03-13 02:42:38 +00:00
|
|
|
|
/// ```
|
2015-03-02 10:01:01 +00:00
|
|
|
|
/// use std::str::FromStr;
|
|
|
|
|
///
|
|
|
|
|
/// assert_eq!(FromStr::from_str("true"), Ok(true));
|
|
|
|
|
/// assert_eq!(FromStr::from_str("false"), Ok(false));
|
|
|
|
|
/// assert!(<bool as FromStr>::from_str("not even a boolean").is_err());
|
|
|
|
|
/// ```
|
|
|
|
|
///
|
2015-03-24 18:57:49 +00:00
|
|
|
|
/// Note, in many cases, the `.parse()` method on `str` is more proper.
|
2015-03-02 10:01:01 +00:00
|
|
|
|
///
|
2015-03-13 02:42:38 +00:00
|
|
|
|
/// ```
|
2015-01-28 06:52:32 +00:00
|
|
|
|
/// assert_eq!("true".parse(), Ok(true));
|
|
|
|
|
/// assert_eq!("false".parse(), Ok(false));
|
|
|
|
|
/// assert!("not even a boolean".parse::<bool>().is_err());
|
2014-11-15 04:52:00 +00:00
|
|
|
|
/// ```
|
|
|
|
|
#[inline]
|
2015-01-28 06:52:32 +00:00
|
|
|
|
fn from_str(s: &str) -> Result<bool, ParseBoolError> {
|
2014-11-15 04:52:00 +00:00
|
|
|
|
match s {
|
2019-12-22 22:42:04 +00:00
|
|
|
|
"true" => Ok(true),
|
2015-01-28 06:52:32 +00:00
|
|
|
|
"false" => Ok(false),
|
2019-12-22 22:42:04 +00:00
|
|
|
|
_ => Err(ParseBoolError { _priv: () }),
|
2014-11-15 04:52:00 +00:00
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
2017-03-25 10:25:08 +00:00
|
|
|
|
/// An error returned when parsing a `bool` using [`from_str`] fails
|
|
|
|
|
///
|
|
|
|
|
/// [`from_str`]: ../../std/primitive.bool.html#method.from_str
|
2016-09-12 19:37:41 +00:00
|
|
|
|
#[derive(Debug, Clone, PartialEq, Eq)]
|
2015-01-28 06:52:32 +00:00
|
|
|
|
#[stable(feature = "rust1", since = "1.0.0")]
|
2019-12-22 22:42:04 +00:00
|
|
|
|
pub struct ParseBoolError {
|
|
|
|
|
_priv: (),
|
|
|
|
|
}
|
2015-01-28 06:52:32 +00:00
|
|
|
|
|
|
|
|
|
#[stable(feature = "rust1", since = "1.0.0")]
|
|
|
|
|
impl fmt::Display for ParseBoolError {
|
2019-04-18 23:37:12 +00:00
|
|
|
|
fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
|
2015-01-28 06:52:32 +00:00
|
|
|
|
"provided string was not `true` or `false`".fmt(f)
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
2014-05-01 06:06:36 +00:00
|
|
|
|
/*
|
|
|
|
|
Section: Creating a string
|
|
|
|
|
*/
|
|
|
|
|
|
2017-04-01 03:22:22 +00:00
|
|
|
|
/// Errors which can occur when attempting to interpret a sequence of [`u8`]
|
2015-10-02 18:36:02 +00:00
|
|
|
|
/// as a string.
|
|
|
|
|
///
|
2017-04-01 03:22:22 +00:00
|
|
|
|
/// [`u8`]: ../../std/primitive.u8.html
|
|
|
|
|
///
|
|
|
|
|
/// As such, the `from_utf8` family of functions and methods for both [`String`]s
|
|
|
|
|
/// and [`&str`]s make use of this error, for example.
|
|
|
|
|
///
|
|
|
|
|
/// [`String`]: ../../std/string/struct.String.html#method.from_utf8
|
|
|
|
|
/// [`&str`]: ../../std/str/fn.from_utf8.html
|
2018-03-17 09:05:23 +00:00
|
|
|
|
///
|
|
|
|
|
/// # Examples
|
|
|
|
|
///
|
|
|
|
|
/// This error type’s methods can be used to create functionality
|
|
|
|
|
/// similar to `String::from_utf8_lossy` without allocating heap memory:
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// fn from_utf8_lossy<F>(mut input: &[u8], mut push: F) where F: FnMut(&str) {
|
|
|
|
|
/// loop {
|
2019-10-20 18:13:47 +00:00
|
|
|
|
/// match std::str::from_utf8(input) {
|
2018-03-17 09:05:23 +00:00
|
|
|
|
/// Ok(valid) => {
|
|
|
|
|
/// push(valid);
|
|
|
|
|
/// break
|
|
|
|
|
/// }
|
|
|
|
|
/// Err(error) => {
|
|
|
|
|
/// let (valid, after_valid) = input.split_at(error.valid_up_to());
|
|
|
|
|
/// unsafe {
|
2019-10-20 18:13:47 +00:00
|
|
|
|
/// push(std::str::from_utf8_unchecked(valid))
|
2018-03-17 09:05:23 +00:00
|
|
|
|
/// }
|
|
|
|
|
/// push("\u{FFFD}");
|
|
|
|
|
///
|
|
|
|
|
/// if let Some(invalid_sequence_length) = error.error_len() {
|
|
|
|
|
/// input = &after_valid[invalid_sequence_length..]
|
|
|
|
|
/// } else {
|
|
|
|
|
/// break
|
|
|
|
|
/// }
|
|
|
|
|
/// }
|
|
|
|
|
/// }
|
|
|
|
|
/// }
|
|
|
|
|
/// }
|
|
|
|
|
/// ```
|
2015-01-28 13:34:18 +00:00
|
|
|
|
#[derive(Copy, Eq, PartialEq, Clone, Debug)]
|
2015-04-10 23:05:09 +00:00
|
|
|
|
#[stable(feature = "rust1", since = "1.0.0")]
|
|
|
|
|
pub struct Utf8Error {
|
|
|
|
|
valid_up_to: usize,
|
2017-03-06 21:06:30 +00:00
|
|
|
|
error_len: Option<u8>,
|
2015-04-10 23:05:09 +00:00
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
impl Utf8Error {
|
|
|
|
|
/// Returns the index in the given string up to which valid UTF-8 was
|
|
|
|
|
/// verified.
|
std: Stabilize the std::str module
This commit starts out by consolidating all `str` extension traits into one
`StrExt` trait to be included in the prelude. This means that
`UnicodeStrPrelude`, `StrPrelude`, and `StrAllocating` have all been merged into
one `StrExt` exported by the standard library. Some functionality is currently
duplicated with the `StrExt` present in libcore.
This commit also currently avoids any methods which require any form of pattern
to operate. These functions will be stabilized via a separate RFC.
Next, stability of methods and structures are as follows:
Stable
* from_utf8_unchecked
* CowString - after moving to std::string
* StrExt::as_bytes
* StrExt::as_ptr
* StrExt::bytes/Bytes - also made a struct instead of a typedef
* StrExt::char_indices/CharIndices - CharOffsets was renamed
* StrExt::chars/Chars
* StrExt::is_empty
* StrExt::len
* StrExt::lines/Lines
* StrExt::lines_any/LinesAny
* StrExt::slice_unchecked
* StrExt::trim
* StrExt::trim_left
* StrExt::trim_right
* StrExt::words/Words - also made a struct instead of a typedef
Unstable
* from_utf8 - the error type was changed to a `Result`, but the error type has
yet to prove itself
* from_c_str - this function will be handled by the c_str RFC
* FromStr - this trait will have an associated error type eventually
* StrExt::escape_default - needs iterators at least, unsure if it should make
the cut
* StrExt::escape_unicode - needs iterators at least, unsure if it should make
the cut
* StrExt::slice_chars - this function has yet to prove itself
* StrExt::slice_shift_char - awaiting conventions about slicing and shifting
* StrExt::graphemes/Graphemes - this functionality may only be in libunicode
* StrExt::grapheme_indices/GraphemeIndices - this functionality may only be in
libunicode
* StrExt::width - this functionality may only be in libunicode
* StrExt::utf16_units - this functionality may only be in libunicode
* StrExt::nfd_chars - this functionality may only be in libunicode
* StrExt::nfkd_chars - this functionality may only be in libunicode
* StrExt::nfc_chars - this functionality may only be in libunicode
* StrExt::nfkc_chars - this functionality may only be in libunicode
* StrExt::is_char_boundary - naming is uncertain with container conventions
* StrExt::char_range_at - naming is uncertain with container conventions
* StrExt::char_range_at_reverse - naming is uncertain with container conventions
* StrExt::char_at - naming is uncertain with container conventions
* StrExt::char_at_reverse - naming is uncertain with container conventions
* StrVector::concat - this functionality may be replaced with iterators, but
it's not certain at this time
* StrVector::connect - as with concat, may be deprecated in favor of iterators
Deprecated
* StrAllocating and UnicodeStrPrelude have been merged into StrExit
* eq_slice - compiler implementation detail
* from_str - use the inherent parse() method
* is_utf8 - call from_utf8 instead
* replace - call the method instead
* truncate_utf16_at_nul - this is an implementation detail of windows and does
not need to be exposed.
* utf8_char_width - moved to libunicode
* utf16_items - moved to libunicode
* is_utf16 - moved to libunicode
* Utf16Items - moved to libunicode
* Utf16Item - moved to libunicode
* Utf16Encoder - moved to libunicode
* AnyLines - renamed to LinesAny and made a struct
* SendStr - use CowString<'static> instead
* str::raw - all functionality is deprecated
* StrExt::into_string - call to_string() instead
* StrExt::repeat - use iterators instead
* StrExt::char_len - use .chars().count() instead
* StrExt::is_alphanumeric - use .chars().all(..)
* StrExt::is_whitespace - use .chars().all(..)
Pending deprecation -- while slicing syntax is being worked out, these methods
are all #[unstable]
* Str - while currently used for generic programming, this trait will be
replaced with one of [], deref coercions, or a generic conversion trait.
* StrExt::slice - use slicing syntax instead
* StrExt::slice_to - use slicing syntax instead
* StrExt::slice_from - use slicing syntax instead
* StrExt::lev_distance - deprecated with no replacement
Awaiting stabilization due to patterns and/or matching
* StrExt::contains
* StrExt::contains_char
* StrExt::split
* StrExt::splitn
* StrExt::split_terminator
* StrExt::rsplitn
* StrExt::match_indices
* StrExt::split_str
* StrExt::starts_with
* StrExt::ends_with
* StrExt::trim_chars
* StrExt::trim_left_chars
* StrExt::trim_right_chars
* StrExt::find
* StrExt::rfind
* StrExt::find_str
* StrExt::subslice_offset
2014-12-10 17:02:31 +00:00
|
|
|
|
///
|
2017-03-02 16:27:57 +00:00
|
|
|
|
/// It is the maximum index such that `from_utf8(&input[..index])`
|
2016-09-30 04:13:14 +00:00
|
|
|
|
/// would return `Ok(_)`.
|
2015-10-02 18:36:02 +00:00
|
|
|
|
///
|
|
|
|
|
/// # Examples
|
|
|
|
|
///
|
|
|
|
|
/// Basic usage:
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// use std::str;
|
|
|
|
|
///
|
|
|
|
|
/// // some invalid bytes, in a vector
|
|
|
|
|
/// let sparkle_heart = vec![0, 159, 146, 150];
|
|
|
|
|
///
|
|
|
|
|
/// // std::str::from_utf8 returns a Utf8Error
|
|
|
|
|
/// let error = str::from_utf8(&sparkle_heart).unwrap_err();
|
|
|
|
|
///
|
2016-02-14 17:52:05 +00:00
|
|
|
|
/// // the second byte is invalid here
|
2015-10-02 18:36:02 +00:00
|
|
|
|
/// assert_eq!(1, error.valid_up_to());
|
|
|
|
|
/// ```
|
std: Stabilize library APIs for 1.5
This commit stabilizes and deprecates library APIs whose FCP has closed in the
last cycle, specifically:
Stabilized APIs:
* `fs::canonicalize`
* `Path::{metadata, symlink_metadata, canonicalize, read_link, read_dir, exists,
is_file, is_dir}` - all moved to inherent methods from the `PathExt` trait.
* `Formatter::fill`
* `Formatter::width`
* `Formatter::precision`
* `Formatter::sign_plus`
* `Formatter::sign_minus`
* `Formatter::alternate`
* `Formatter::sign_aware_zero_pad`
* `string::ParseError`
* `Utf8Error::valid_up_to`
* `Iterator::{cmp, partial_cmp, eq, ne, lt, le, gt, ge}`
* `<[T]>::split_{first,last}{,_mut}`
* `Condvar::wait_timeout` - note that `wait_timeout_ms` is not yet deprecated
but will be once 1.5 is released.
* `str::{R,}MatchIndices`
* `str::{r,}match_indices`
* `char::from_u32_unchecked`
* `VecDeque::insert`
* `VecDeque::shrink_to_fit`
* `VecDeque::as_slices`
* `VecDeque::as_mut_slices`
* `VecDeque::swap_remove_front` - (renamed from `swap_front_remove`)
* `VecDeque::swap_remove_back` - (renamed from `swap_back_remove`)
* `Vec::resize`
* `str::slice_mut_unchecked`
* `FileTypeExt`
* `FileTypeExt::{is_block_device, is_char_device, is_fifo, is_socket}`
* `BinaryHeap::from` - `from_vec` deprecated in favor of this
* `BinaryHeap::into_vec` - plus a `Into` impl
* `BinaryHeap::into_sorted_vec`
Deprecated APIs
* `slice::ref_slice`
* `slice::mut_ref_slice`
* `iter::{range_inclusive, RangeInclusive}`
* `std::dynamic_lib`
Closes #27706
Closes #27725
cc #27726 (align not stabilized yet)
Closes #27734
Closes #27737
Closes #27742
Closes #27743
Closes #27772
Closes #27774
Closes #27777
Closes #27781
cc #27788 (a few remaining methods though)
Closes #27790
Closes #27793
Closes #27796
Closes #27810
cc #28147 (not all parts stabilized)
2015-10-22 23:28:45 +00:00
|
|
|
|
#[stable(feature = "utf8_error", since = "1.5.0")]
|
2019-12-22 22:42:04 +00:00
|
|
|
|
pub fn valid_up_to(&self) -> usize {
|
|
|
|
|
self.valid_up_to
|
|
|
|
|
}
|
2017-03-02 16:27:57 +00:00
|
|
|
|
|
2019-02-09 22:16:58 +00:00
|
|
|
|
/// Provides more information about the failure:
|
2017-03-02 16:27:57 +00:00
|
|
|
|
///
|
|
|
|
|
/// * `None`: the end of the input was reached unexpectedly.
|
|
|
|
|
/// `self.valid_up_to()` is 1 to 3 bytes from the end of the input.
|
|
|
|
|
/// If a byte stream (such as a file or a network socket) is being decoded incrementally,
|
|
|
|
|
/// this could be a valid `char` whose UTF-8 byte sequence is spanning multiple chunks.
|
|
|
|
|
///
|
2017-03-06 21:06:30 +00:00
|
|
|
|
/// * `Some(len)`: an unexpected byte was encountered.
|
|
|
|
|
/// The length provided is that of the invalid byte sequence
|
|
|
|
|
/// that starts at the index given by `valid_up_to()`.
|
|
|
|
|
/// Decoding should resume after that sequence
|
2018-08-11 18:09:59 +00:00
|
|
|
|
/// (after inserting a [`U+FFFD REPLACEMENT CHARACTER`][U+FFFD]) in case of
|
|
|
|
|
/// lossy decoding.
|
|
|
|
|
///
|
|
|
|
|
/// [U+FFFD]: ../../std/char/constant.REPLACEMENT_CHARACTER.html
|
2017-07-20 22:46:46 +00:00
|
|
|
|
#[stable(feature = "utf8_error_error_len", since = "1.20.0")]
|
2017-03-06 21:06:30 +00:00
|
|
|
|
pub fn error_len(&self) -> Option<usize> {
|
|
|
|
|
self.error_len.map(|len| len as usize)
|
2017-03-02 16:27:57 +00:00
|
|
|
|
}
|
std: Stabilize the std::str module
This commit starts out by consolidating all `str` extension traits into one
`StrExt` trait to be included in the prelude. This means that
`UnicodeStrPrelude`, `StrPrelude`, and `StrAllocating` have all been merged into
one `StrExt` exported by the standard library. Some functionality is currently
duplicated with the `StrExt` present in libcore.
This commit also currently avoids any methods which require any form of pattern
to operate. These functions will be stabilized via a separate RFC.
Next, stability of methods and structures are as follows:
Stable
* from_utf8_unchecked
* CowString - after moving to std::string
* StrExt::as_bytes
* StrExt::as_ptr
* StrExt::bytes/Bytes - also made a struct instead of a typedef
* StrExt::char_indices/CharIndices - CharOffsets was renamed
* StrExt::chars/Chars
* StrExt::is_empty
* StrExt::len
* StrExt::lines/Lines
* StrExt::lines_any/LinesAny
* StrExt::slice_unchecked
* StrExt::trim
* StrExt::trim_left
* StrExt::trim_right
* StrExt::words/Words - also made a struct instead of a typedef
Unstable
* from_utf8 - the error type was changed to a `Result`, but the error type has
yet to prove itself
* from_c_str - this function will be handled by the c_str RFC
* FromStr - this trait will have an associated error type eventually
* StrExt::escape_default - needs iterators at least, unsure if it should make
the cut
* StrExt::escape_unicode - needs iterators at least, unsure if it should make
the cut
* StrExt::slice_chars - this function has yet to prove itself
* StrExt::slice_shift_char - awaiting conventions about slicing and shifting
* StrExt::graphemes/Graphemes - this functionality may only be in libunicode
* StrExt::grapheme_indices/GraphemeIndices - this functionality may only be in
libunicode
* StrExt::width - this functionality may only be in libunicode
* StrExt::utf16_units - this functionality may only be in libunicode
* StrExt::nfd_chars - this functionality may only be in libunicode
* StrExt::nfkd_chars - this functionality may only be in libunicode
* StrExt::nfc_chars - this functionality may only be in libunicode
* StrExt::nfkc_chars - this functionality may only be in libunicode
* StrExt::is_char_boundary - naming is uncertain with container conventions
* StrExt::char_range_at - naming is uncertain with container conventions
* StrExt::char_range_at_reverse - naming is uncertain with container conventions
* StrExt::char_at - naming is uncertain with container conventions
* StrExt::char_at_reverse - naming is uncertain with container conventions
* StrVector::concat - this functionality may be replaced with iterators, but
it's not certain at this time
* StrVector::connect - as with concat, may be deprecated in favor of iterators
Deprecated
* StrAllocating and UnicodeStrPrelude have been merged into StrExit
* eq_slice - compiler implementation detail
* from_str - use the inherent parse() method
* is_utf8 - call from_utf8 instead
* replace - call the method instead
* truncate_utf16_at_nul - this is an implementation detail of windows and does
not need to be exposed.
* utf8_char_width - moved to libunicode
* utf16_items - moved to libunicode
* is_utf16 - moved to libunicode
* Utf16Items - moved to libunicode
* Utf16Item - moved to libunicode
* Utf16Encoder - moved to libunicode
* AnyLines - renamed to LinesAny and made a struct
* SendStr - use CowString<'static> instead
* str::raw - all functionality is deprecated
* StrExt::into_string - call to_string() instead
* StrExt::repeat - use iterators instead
* StrExt::char_len - use .chars().count() instead
* StrExt::is_alphanumeric - use .chars().all(..)
* StrExt::is_whitespace - use .chars().all(..)
Pending deprecation -- while slicing syntax is being worked out, these methods
are all #[unstable]
* Str - while currently used for generic programming, this trait will be
replaced with one of [], deref coercions, or a generic conversion trait.
* StrExt::slice - use slicing syntax instead
* StrExt::slice_to - use slicing syntax instead
* StrExt::slice_from - use slicing syntax instead
* StrExt::lev_distance - deprecated with no replacement
Awaiting stabilization due to patterns and/or matching
* StrExt::contains
* StrExt::contains_char
* StrExt::split
* StrExt::splitn
* StrExt::split_terminator
* StrExt::rsplitn
* StrExt::match_indices
* StrExt::split_str
* StrExt::starts_with
* StrExt::ends_with
* StrExt::trim_chars
* StrExt::trim_left_chars
* StrExt::trim_right_chars
* StrExt::find
* StrExt::rfind
* StrExt::find_str
* StrExt::subslice_offset
2014-12-10 17:02:31 +00:00
|
|
|
|
}
|
|
|
|
|
|
2015-10-02 18:36:02 +00:00
|
|
|
|
/// Converts a slice of bytes to a string slice.
|
2014-05-01 06:06:36 +00:00
|
|
|
|
///
|
2017-04-01 13:56:40 +00:00
|
|
|
|
/// A string slice ([`&str`]) is made of bytes ([`u8`]), and a byte slice
|
|
|
|
|
/// ([`&[u8]`][byteslice]) is made of bytes, so this function converts between
|
|
|
|
|
/// the two. Not all byte slices are valid string slices, however: [`&str`] requires
|
|
|
|
|
/// that it is valid UTF-8. `from_utf8()` checks to ensure that the bytes are valid
|
|
|
|
|
/// UTF-8, and then does the conversion.
|
|
|
|
|
///
|
|
|
|
|
/// [`&str`]: ../../std/primitive.str.html
|
|
|
|
|
/// [`u8`]: ../../std/primitive.u8.html
|
|
|
|
|
/// [byteslice]: ../../std/primitive.slice.html
|
2015-10-02 18:36:02 +00:00
|
|
|
|
///
|
|
|
|
|
/// If you are sure that the byte slice is valid UTF-8, and you don't want to
|
|
|
|
|
/// incur the overhead of the validity check, there is an unsafe version of
|
2017-03-12 18:04:52 +00:00
|
|
|
|
/// this function, [`from_utf8_unchecked`][fromutf8u], which has the same
|
2015-10-02 18:36:02 +00:00
|
|
|
|
/// behavior but skips the check.
|
|
|
|
|
///
|
2016-02-14 18:38:37 +00:00
|
|
|
|
/// [fromutf8u]: fn.from_utf8_unchecked.html
|
2015-10-02 18:36:02 +00:00
|
|
|
|
///
|
|
|
|
|
/// If you need a `String` instead of a `&str`, consider
|
2017-03-12 18:04:52 +00:00
|
|
|
|
/// [`String::from_utf8`][string].
|
2015-10-02 18:36:02 +00:00
|
|
|
|
///
|
2016-03-08 07:55:52 +00:00
|
|
|
|
/// [string]: ../../std/string/struct.String.html#method.from_utf8
|
2015-10-02 18:36:02 +00:00
|
|
|
|
///
|
2017-04-01 13:56:40 +00:00
|
|
|
|
/// Because you can stack-allocate a `[u8; N]`, and you can take a
|
|
|
|
|
/// [`&[u8]`][byteslice] of it, this function is one way to have a
|
|
|
|
|
/// stack-allocated string. There is an example of this in the
|
|
|
|
|
/// examples section below.
|
|
|
|
|
///
|
|
|
|
|
/// [byteslice]: ../../std/primitive.slice.html
|
2014-05-01 06:06:36 +00:00
|
|
|
|
///
|
2016-02-02 02:41:29 +00:00
|
|
|
|
/// # Errors
|
std: Stabilize the std::str module
This commit starts out by consolidating all `str` extension traits into one
`StrExt` trait to be included in the prelude. This means that
`UnicodeStrPrelude`, `StrPrelude`, and `StrAllocating` have all been merged into
one `StrExt` exported by the standard library. Some functionality is currently
duplicated with the `StrExt` present in libcore.
This commit also currently avoids any methods which require any form of pattern
to operate. These functions will be stabilized via a separate RFC.
Next, stability of methods and structures are as follows:
Stable
* from_utf8_unchecked
* CowString - after moving to std::string
* StrExt::as_bytes
* StrExt::as_ptr
* StrExt::bytes/Bytes - also made a struct instead of a typedef
* StrExt::char_indices/CharIndices - CharOffsets was renamed
* StrExt::chars/Chars
* StrExt::is_empty
* StrExt::len
* StrExt::lines/Lines
* StrExt::lines_any/LinesAny
* StrExt::slice_unchecked
* StrExt::trim
* StrExt::trim_left
* StrExt::trim_right
* StrExt::words/Words - also made a struct instead of a typedef
Unstable
* from_utf8 - the error type was changed to a `Result`, but the error type has
yet to prove itself
* from_c_str - this function will be handled by the c_str RFC
* FromStr - this trait will have an associated error type eventually
* StrExt::escape_default - needs iterators at least, unsure if it should make
the cut
* StrExt::escape_unicode - needs iterators at least, unsure if it should make
the cut
* StrExt::slice_chars - this function has yet to prove itself
* StrExt::slice_shift_char - awaiting conventions about slicing and shifting
* StrExt::graphemes/Graphemes - this functionality may only be in libunicode
* StrExt::grapheme_indices/GraphemeIndices - this functionality may only be in
libunicode
* StrExt::width - this functionality may only be in libunicode
* StrExt::utf16_units - this functionality may only be in libunicode
* StrExt::nfd_chars - this functionality may only be in libunicode
* StrExt::nfkd_chars - this functionality may only be in libunicode
* StrExt::nfc_chars - this functionality may only be in libunicode
* StrExt::nfkc_chars - this functionality may only be in libunicode
* StrExt::is_char_boundary - naming is uncertain with container conventions
* StrExt::char_range_at - naming is uncertain with container conventions
* StrExt::char_range_at_reverse - naming is uncertain with container conventions
* StrExt::char_at - naming is uncertain with container conventions
* StrExt::char_at_reverse - naming is uncertain with container conventions
* StrVector::concat - this functionality may be replaced with iterators, but
it's not certain at this time
* StrVector::connect - as with concat, may be deprecated in favor of iterators
Deprecated
* StrAllocating and UnicodeStrPrelude have been merged into StrExit
* eq_slice - compiler implementation detail
* from_str - use the inherent parse() method
* is_utf8 - call from_utf8 instead
* replace - call the method instead
* truncate_utf16_at_nul - this is an implementation detail of windows and does
not need to be exposed.
* utf8_char_width - moved to libunicode
* utf16_items - moved to libunicode
* is_utf16 - moved to libunicode
* Utf16Items - moved to libunicode
* Utf16Item - moved to libunicode
* Utf16Encoder - moved to libunicode
* AnyLines - renamed to LinesAny and made a struct
* SendStr - use CowString<'static> instead
* str::raw - all functionality is deprecated
* StrExt::into_string - call to_string() instead
* StrExt::repeat - use iterators instead
* StrExt::char_len - use .chars().count() instead
* StrExt::is_alphanumeric - use .chars().all(..)
* StrExt::is_whitespace - use .chars().all(..)
Pending deprecation -- while slicing syntax is being worked out, these methods
are all #[unstable]
* Str - while currently used for generic programming, this trait will be
replaced with one of [], deref coercions, or a generic conversion trait.
* StrExt::slice - use slicing syntax instead
* StrExt::slice_to - use slicing syntax instead
* StrExt::slice_from - use slicing syntax instead
* StrExt::lev_distance - deprecated with no replacement
Awaiting stabilization due to patterns and/or matching
* StrExt::contains
* StrExt::contains_char
* StrExt::split
* StrExt::splitn
* StrExt::split_terminator
* StrExt::rsplitn
* StrExt::match_indices
* StrExt::split_str
* StrExt::starts_with
* StrExt::ends_with
* StrExt::trim_chars
* StrExt::trim_left_chars
* StrExt::trim_right_chars
* StrExt::find
* StrExt::rfind
* StrExt::find_str
* StrExt::subslice_offset
2014-12-10 17:02:31 +00:00
|
|
|
|
///
|
2015-07-24 01:04:55 +00:00
|
|
|
|
/// Returns `Err` if the slice is not UTF-8 with a description as to why the
|
|
|
|
|
/// provided slice is not UTF-8.
|
2015-10-02 18:36:02 +00:00
|
|
|
|
///
|
|
|
|
|
/// # Examples
|
|
|
|
|
///
|
|
|
|
|
/// Basic usage:
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// use std::str;
|
|
|
|
|
///
|
|
|
|
|
/// // some bytes, in a vector
|
|
|
|
|
/// let sparkle_heart = vec![240, 159, 146, 150];
|
|
|
|
|
///
|
|
|
|
|
/// // We know these bytes are valid, so just use `unwrap()`.
|
|
|
|
|
/// let sparkle_heart = str::from_utf8(&sparkle_heart).unwrap();
|
|
|
|
|
///
|
|
|
|
|
/// assert_eq!("💖", sparkle_heart);
|
|
|
|
|
/// ```
|
|
|
|
|
///
|
|
|
|
|
/// Incorrect bytes:
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// use std::str;
|
|
|
|
|
///
|
|
|
|
|
/// // some invalid bytes, in a vector
|
|
|
|
|
/// let sparkle_heart = vec![0, 159, 146, 150];
|
|
|
|
|
///
|
|
|
|
|
/// assert!(str::from_utf8(&sparkle_heart).is_err());
|
|
|
|
|
/// ```
|
|
|
|
|
///
|
|
|
|
|
/// See the docs for [`Utf8Error`][error] for more details on the kinds of
|
|
|
|
|
/// errors that can be returned.
|
|
|
|
|
///
|
|
|
|
|
/// [error]: struct.Utf8Error.html
|
|
|
|
|
///
|
|
|
|
|
/// A "stack allocated string":
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// use std::str;
|
|
|
|
|
///
|
|
|
|
|
/// // some bytes, in a stack-allocated array
|
|
|
|
|
/// let sparkle_heart = [240, 159, 146, 150];
|
|
|
|
|
///
|
|
|
|
|
/// // We know these bytes are valid, so just use `unwrap()`.
|
|
|
|
|
/// let sparkle_heart = str::from_utf8(&sparkle_heart).unwrap();
|
|
|
|
|
///
|
|
|
|
|
/// assert_eq!("💖", sparkle_heart);
|
|
|
|
|
/// ```
|
2015-01-24 05:48:20 +00:00
|
|
|
|
#[stable(feature = "rust1", since = "1.0.0")]
|
std: Stabilize the std::str module
This commit starts out by consolidating all `str` extension traits into one
`StrExt` trait to be included in the prelude. This means that
`UnicodeStrPrelude`, `StrPrelude`, and `StrAllocating` have all been merged into
one `StrExt` exported by the standard library. Some functionality is currently
duplicated with the `StrExt` present in libcore.
This commit also currently avoids any methods which require any form of pattern
to operate. These functions will be stabilized via a separate RFC.
Next, stability of methods and structures are as follows:
Stable
* from_utf8_unchecked
* CowString - after moving to std::string
* StrExt::as_bytes
* StrExt::as_ptr
* StrExt::bytes/Bytes - also made a struct instead of a typedef
* StrExt::char_indices/CharIndices - CharOffsets was renamed
* StrExt::chars/Chars
* StrExt::is_empty
* StrExt::len
* StrExt::lines/Lines
* StrExt::lines_any/LinesAny
* StrExt::slice_unchecked
* StrExt::trim
* StrExt::trim_left
* StrExt::trim_right
* StrExt::words/Words - also made a struct instead of a typedef
Unstable
* from_utf8 - the error type was changed to a `Result`, but the error type has
yet to prove itself
* from_c_str - this function will be handled by the c_str RFC
* FromStr - this trait will have an associated error type eventually
* StrExt::escape_default - needs iterators at least, unsure if it should make
the cut
* StrExt::escape_unicode - needs iterators at least, unsure if it should make
the cut
* StrExt::slice_chars - this function has yet to prove itself
* StrExt::slice_shift_char - awaiting conventions about slicing and shifting
* StrExt::graphemes/Graphemes - this functionality may only be in libunicode
* StrExt::grapheme_indices/GraphemeIndices - this functionality may only be in
libunicode
* StrExt::width - this functionality may only be in libunicode
* StrExt::utf16_units - this functionality may only be in libunicode
* StrExt::nfd_chars - this functionality may only be in libunicode
* StrExt::nfkd_chars - this functionality may only be in libunicode
* StrExt::nfc_chars - this functionality may only be in libunicode
* StrExt::nfkc_chars - this functionality may only be in libunicode
* StrExt::is_char_boundary - naming is uncertain with container conventions
* StrExt::char_range_at - naming is uncertain with container conventions
* StrExt::char_range_at_reverse - naming is uncertain with container conventions
* StrExt::char_at - naming is uncertain with container conventions
* StrExt::char_at_reverse - naming is uncertain with container conventions
* StrVector::concat - this functionality may be replaced with iterators, but
it's not certain at this time
* StrVector::connect - as with concat, may be deprecated in favor of iterators
Deprecated
* StrAllocating and UnicodeStrPrelude have been merged into StrExit
* eq_slice - compiler implementation detail
* from_str - use the inherent parse() method
* is_utf8 - call from_utf8 instead
* replace - call the method instead
* truncate_utf16_at_nul - this is an implementation detail of windows and does
not need to be exposed.
* utf8_char_width - moved to libunicode
* utf16_items - moved to libunicode
* is_utf16 - moved to libunicode
* Utf16Items - moved to libunicode
* Utf16Item - moved to libunicode
* Utf16Encoder - moved to libunicode
* AnyLines - renamed to LinesAny and made a struct
* SendStr - use CowString<'static> instead
* str::raw - all functionality is deprecated
* StrExt::into_string - call to_string() instead
* StrExt::repeat - use iterators instead
* StrExt::char_len - use .chars().count() instead
* StrExt::is_alphanumeric - use .chars().all(..)
* StrExt::is_whitespace - use .chars().all(..)
Pending deprecation -- while slicing syntax is being worked out, these methods
are all #[unstable]
* Str - while currently used for generic programming, this trait will be
replaced with one of [], deref coercions, or a generic conversion trait.
* StrExt::slice - use slicing syntax instead
* StrExt::slice_to - use slicing syntax instead
* StrExt::slice_from - use slicing syntax instead
* StrExt::lev_distance - deprecated with no replacement
Awaiting stabilization due to patterns and/or matching
* StrExt::contains
* StrExt::contains_char
* StrExt::split
* StrExt::splitn
* StrExt::split_terminator
* StrExt::rsplitn
* StrExt::match_indices
* StrExt::split_str
* StrExt::starts_with
* StrExt::ends_with
* StrExt::trim_chars
* StrExt::trim_left_chars
* StrExt::trim_right_chars
* StrExt::find
* StrExt::rfind
* StrExt::find_str
* StrExt::subslice_offset
2014-12-10 17:02:31 +00:00
|
|
|
|
pub fn from_utf8(v: &[u8]) -> Result<&str, Utf8Error> {
|
2016-03-23 03:01:37 +00:00
|
|
|
|
run_utf8_validation(v)?;
|
2019-11-20 02:28:32 +00:00
|
|
|
|
// SAFETY: just ran validation
|
std: Stabilize the std::str module
This commit starts out by consolidating all `str` extension traits into one
`StrExt` trait to be included in the prelude. This means that
`UnicodeStrPrelude`, `StrPrelude`, and `StrAllocating` have all been merged into
one `StrExt` exported by the standard library. Some functionality is currently
duplicated with the `StrExt` present in libcore.
This commit also currently avoids any methods which require any form of pattern
to operate. These functions will be stabilized via a separate RFC.
Next, stability of methods and structures are as follows:
Stable
* from_utf8_unchecked
* CowString - after moving to std::string
* StrExt::as_bytes
* StrExt::as_ptr
* StrExt::bytes/Bytes - also made a struct instead of a typedef
* StrExt::char_indices/CharIndices - CharOffsets was renamed
* StrExt::chars/Chars
* StrExt::is_empty
* StrExt::len
* StrExt::lines/Lines
* StrExt::lines_any/LinesAny
* StrExt::slice_unchecked
* StrExt::trim
* StrExt::trim_left
* StrExt::trim_right
* StrExt::words/Words - also made a struct instead of a typedef
Unstable
* from_utf8 - the error type was changed to a `Result`, but the error type has
yet to prove itself
* from_c_str - this function will be handled by the c_str RFC
* FromStr - this trait will have an associated error type eventually
* StrExt::escape_default - needs iterators at least, unsure if it should make
the cut
* StrExt::escape_unicode - needs iterators at least, unsure if it should make
the cut
* StrExt::slice_chars - this function has yet to prove itself
* StrExt::slice_shift_char - awaiting conventions about slicing and shifting
* StrExt::graphemes/Graphemes - this functionality may only be in libunicode
* StrExt::grapheme_indices/GraphemeIndices - this functionality may only be in
libunicode
* StrExt::width - this functionality may only be in libunicode
* StrExt::utf16_units - this functionality may only be in libunicode
* StrExt::nfd_chars - this functionality may only be in libunicode
* StrExt::nfkd_chars - this functionality may only be in libunicode
* StrExt::nfc_chars - this functionality may only be in libunicode
* StrExt::nfkc_chars - this functionality may only be in libunicode
* StrExt::is_char_boundary - naming is uncertain with container conventions
* StrExt::char_range_at - naming is uncertain with container conventions
* StrExt::char_range_at_reverse - naming is uncertain with container conventions
* StrExt::char_at - naming is uncertain with container conventions
* StrExt::char_at_reverse - naming is uncertain with container conventions
* StrVector::concat - this functionality may be replaced with iterators, but
it's not certain at this time
* StrVector::connect - as with concat, may be deprecated in favor of iterators
Deprecated
* StrAllocating and UnicodeStrPrelude have been merged into StrExit
* eq_slice - compiler implementation detail
* from_str - use the inherent parse() method
* is_utf8 - call from_utf8 instead
* replace - call the method instead
* truncate_utf16_at_nul - this is an implementation detail of windows and does
not need to be exposed.
* utf8_char_width - moved to libunicode
* utf16_items - moved to libunicode
* is_utf16 - moved to libunicode
* Utf16Items - moved to libunicode
* Utf16Item - moved to libunicode
* Utf16Encoder - moved to libunicode
* AnyLines - renamed to LinesAny and made a struct
* SendStr - use CowString<'static> instead
* str::raw - all functionality is deprecated
* StrExt::into_string - call to_string() instead
* StrExt::repeat - use iterators instead
* StrExt::char_len - use .chars().count() instead
* StrExt::is_alphanumeric - use .chars().all(..)
* StrExt::is_whitespace - use .chars().all(..)
Pending deprecation -- while slicing syntax is being worked out, these methods
are all #[unstable]
* Str - while currently used for generic programming, this trait will be
replaced with one of [], deref coercions, or a generic conversion trait.
* StrExt::slice - use slicing syntax instead
* StrExt::slice_to - use slicing syntax instead
* StrExt::slice_from - use slicing syntax instead
* StrExt::lev_distance - deprecated with no replacement
Awaiting stabilization due to patterns and/or matching
* StrExt::contains
* StrExt::contains_char
* StrExt::split
* StrExt::splitn
* StrExt::split_terminator
* StrExt::rsplitn
* StrExt::match_indices
* StrExt::split_str
* StrExt::starts_with
* StrExt::ends_with
* StrExt::trim_chars
* StrExt::trim_left_chars
* StrExt::trim_right_chars
* StrExt::find
* StrExt::rfind
* StrExt::find_str
* StrExt::subslice_offset
2014-12-10 17:02:31 +00:00
|
|
|
|
Ok(unsafe { from_utf8_unchecked(v) })
|
2014-11-20 18:11:15 +00:00
|
|
|
|
}
|
|
|
|
|
|
2017-03-10 17:10:26 +00:00
|
|
|
|
/// Converts a mutable slice of bytes to a mutable string slice.
|
2017-09-10 03:10:19 +00:00
|
|
|
|
///
|
|
|
|
|
/// # Examples
|
|
|
|
|
///
|
|
|
|
|
/// Basic usage:
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// use std::str;
|
|
|
|
|
///
|
|
|
|
|
/// // "Hello, Rust!" as a mutable vector
|
|
|
|
|
/// let mut hellorust = vec![72, 101, 108, 108, 111, 44, 32, 82, 117, 115, 116, 33];
|
|
|
|
|
///
|
|
|
|
|
/// // As we know these bytes are valid, we can use `unwrap()`
|
|
|
|
|
/// let outstr = str::from_utf8_mut(&mut hellorust).unwrap();
|
|
|
|
|
///
|
|
|
|
|
/// assert_eq!("Hello, Rust!", outstr);
|
|
|
|
|
/// ```
|
2017-09-10 13:12:23 +00:00
|
|
|
|
///
|
2017-09-10 03:33:24 +00:00
|
|
|
|
/// Incorrect bytes:
|
2017-09-10 03:10:19 +00:00
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// use std::str;
|
2017-09-10 13:25:23 +00:00
|
|
|
|
///
|
2017-09-10 03:10:19 +00:00
|
|
|
|
/// // Some invalid bytes in a mutable vector
|
|
|
|
|
/// let mut invalid = vec![128, 223];
|
|
|
|
|
///
|
|
|
|
|
/// assert!(str::from_utf8_mut(&mut invalid).is_err());
|
|
|
|
|
/// ```
|
|
|
|
|
/// See the docs for [`Utf8Error`][error] for more details on the kinds of
|
|
|
|
|
/// errors that can be returned.
|
2017-09-10 13:12:23 +00:00
|
|
|
|
///
|
|
|
|
|
/// [error]: struct.Utf8Error.html
|
2017-07-20 22:52:12 +00:00
|
|
|
|
#[stable(feature = "str_mut_extras", since = "1.20.0")]
|
2017-03-10 17:10:26 +00:00
|
|
|
|
pub fn from_utf8_mut(v: &mut [u8]) -> Result<&mut str, Utf8Error> {
|
|
|
|
|
run_utf8_validation(v)?;
|
2019-11-20 02:28:32 +00:00
|
|
|
|
// SAFETY: just ran validation
|
2017-03-10 17:10:26 +00:00
|
|
|
|
Ok(unsafe { from_utf8_unchecked_mut(v) })
|
|
|
|
|
}
|
|
|
|
|
|
2014-11-20 18:11:15 +00:00
|
|
|
|
/// Converts a slice of bytes to a string slice without checking
|
|
|
|
|
/// that the string contains valid UTF-8.
|
2015-10-02 18:36:02 +00:00
|
|
|
|
///
|
2017-03-12 18:04:52 +00:00
|
|
|
|
/// See the safe version, [`from_utf8`][fromutf8], for more information.
|
2015-10-02 18:36:02 +00:00
|
|
|
|
///
|
|
|
|
|
/// [fromutf8]: fn.from_utf8.html
|
|
|
|
|
///
|
2015-10-23 15:42:14 +00:00
|
|
|
|
/// # Safety
|
2015-10-02 18:36:02 +00:00
|
|
|
|
///
|
|
|
|
|
/// This function is unsafe because it does not check that the bytes passed to
|
|
|
|
|
/// it are valid UTF-8. If this constraint is violated, undefined behavior
|
2017-03-29 17:21:31 +00:00
|
|
|
|
/// results, as the rest of Rust assumes that [`&str`]s are valid UTF-8.
|
|
|
|
|
///
|
|
|
|
|
/// [`&str`]: ../../std/primitive.str.html
|
2015-10-02 18:36:02 +00:00
|
|
|
|
///
|
|
|
|
|
/// # Examples
|
|
|
|
|
///
|
|
|
|
|
/// Basic usage:
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// use std::str;
|
|
|
|
|
///
|
|
|
|
|
/// // some bytes, in a vector
|
|
|
|
|
/// let sparkle_heart = vec![240, 159, 146, 150];
|
|
|
|
|
///
|
|
|
|
|
/// let sparkle_heart = unsafe {
|
|
|
|
|
/// str::from_utf8_unchecked(&sparkle_heart)
|
|
|
|
|
/// };
|
|
|
|
|
///
|
|
|
|
|
/// assert_eq!("💖", sparkle_heart);
|
|
|
|
|
/// ```
|
2017-07-20 18:14:13 +00:00
|
|
|
|
#[inline]
|
2015-01-24 05:48:20 +00:00
|
|
|
|
#[stable(feature = "rust1", since = "1.0.0")]
|
2015-09-03 09:49:08 +00:00
|
|
|
|
pub unsafe fn from_utf8_unchecked(v: &[u8]) -> &str {
|
2017-09-17 16:03:56 +00:00
|
|
|
|
&*(v as *const [u8] as *const str)
|
2014-11-20 18:11:15 +00:00
|
|
|
|
}
|
|
|
|
|
|
2017-03-10 17:10:26 +00:00
|
|
|
|
/// Converts a slice of bytes to a string slice without checking
|
|
|
|
|
/// that the string contains valid UTF-8; mutable version.
|
|
|
|
|
///
|
|
|
|
|
/// See the immutable version, [`from_utf8_unchecked()`][fromutf8], for more information.
|
|
|
|
|
///
|
|
|
|
|
/// [fromutf8]: fn.from_utf8_unchecked.html
|
2017-09-10 18:27:57 +00:00
|
|
|
|
///
|
|
|
|
|
/// # Examples
|
|
|
|
|
///
|
|
|
|
|
/// Basic usage:
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// use std::str;
|
|
|
|
|
///
|
|
|
|
|
/// let mut heart = vec![240, 159, 146, 150];
|
|
|
|
|
/// let heart = unsafe { str::from_utf8_unchecked_mut(&mut heart) };
|
|
|
|
|
///
|
|
|
|
|
/// assert_eq!("💖", heart);
|
|
|
|
|
/// ```
|
2017-07-20 18:14:13 +00:00
|
|
|
|
#[inline]
|
2017-07-20 22:52:12 +00:00
|
|
|
|
#[stable(feature = "str_mut_extras", since = "1.20.0")]
|
2017-03-10 17:10:26 +00:00
|
|
|
|
pub unsafe fn from_utf8_unchecked_mut(v: &mut [u8]) -> &mut str {
|
2017-09-17 16:11:42 +00:00
|
|
|
|
&mut *(v as *mut [u8] as *mut str)
|
2017-03-10 17:10:26 +00:00
|
|
|
|
}
|
|
|
|
|
|
2015-01-24 17:15:42 +00:00
|
|
|
|
#[stable(feature = "rust1", since = "1.0.0")]
|
2015-01-20 23:45:07 +00:00
|
|
|
|
impl fmt::Display for Utf8Error {
|
2019-04-18 23:37:12 +00:00
|
|
|
|
fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
|
2017-03-06 21:06:30 +00:00
|
|
|
|
if let Some(error_len) = self.error_len {
|
2019-12-22 22:42:04 +00:00
|
|
|
|
write!(
|
|
|
|
|
f,
|
|
|
|
|
"invalid utf-8 sequence of {} bytes from index {}",
|
|
|
|
|
error_len, self.valid_up_to
|
|
|
|
|
)
|
2017-03-02 16:27:57 +00:00
|
|
|
|
} else {
|
|
|
|
|
write!(f, "incomplete utf-8 byte sequence from index {}", self.valid_up_to)
|
|
|
|
|
}
|
2015-01-20 23:45:07 +00:00
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
2014-05-01 06:06:36 +00:00
|
|
|
|
/*
|
|
|
|
|
Section: Iterators
|
|
|
|
|
*/
|
|
|
|
|
|
2017-03-30 19:51:49 +00:00
|
|
|
|
/// An iterator over the [`char`]s of a string slice.
|
2014-07-17 17:34:07 +00:00
|
|
|
|
///
|
2017-03-30 19:51:49 +00:00
|
|
|
|
/// [`char`]: ../../std/primitive.char.html
|
2015-12-11 13:25:50 +00:00
|
|
|
|
///
|
2017-03-30 22:33:23 +00:00
|
|
|
|
/// This struct is created by the [`chars`] method on [`str`].
|
2017-03-30 19:51:49 +00:00
|
|
|
|
/// See its documentation for more.
|
2015-12-11 13:25:50 +00:00
|
|
|
|
///
|
2017-03-12 18:04:52 +00:00
|
|
|
|
/// [`chars`]: ../../std/primitive.str.html#method.chars
|
2017-03-30 19:51:49 +00:00
|
|
|
|
/// [`str`]: ../../std/primitive.str.html
|
2019-07-26 06:58:37 +00:00
|
|
|
|
#[derive(Clone)]
|
2015-01-24 05:48:20 +00:00
|
|
|
|
#[stable(feature = "rust1", since = "1.0.0")]
|
2014-05-01 06:06:36 +00:00
|
|
|
|
pub struct Chars<'a> {
|
2019-12-22 22:42:04 +00:00
|
|
|
|
iter: slice::Iter<'a, u8>,
|
2014-07-17 17:34:07 +00:00
|
|
|
|
}
|
|
|
|
|
|
2017-03-22 00:42:23 +00:00
|
|
|
|
/// Returns the initial codepoint accumulator for the first byte.
|
2015-03-10 11:06:44 +00:00
|
|
|
|
/// The first byte is special, only want bottom 5 bits for width 2, 4 bits
|
|
|
|
|
/// for width 3, and 3 bits for width 4.
|
|
|
|
|
#[inline]
|
2019-12-22 22:42:04 +00:00
|
|
|
|
fn utf8_first_byte(byte: u8, width: u32) -> u32 {
|
|
|
|
|
(byte & (0x7F >> width)) as u32
|
|
|
|
|
}
|
2014-07-17 17:34:07 +00:00
|
|
|
|
|
2017-03-22 00:42:23 +00:00
|
|
|
|
/// Returns the value of `ch` updated with continuation byte `byte`.
|
2015-03-10 11:06:44 +00:00
|
|
|
|
#[inline]
|
2019-12-22 22:42:04 +00:00
|
|
|
|
fn utf8_acc_cont_byte(ch: u32, byte: u8) -> u32 {
|
|
|
|
|
(ch << 6) | (byte & CONT_MASK) as u32
|
|
|
|
|
}
|
2014-07-17 17:34:07 +00:00
|
|
|
|
|
2018-11-27 02:59:49 +00:00
|
|
|
|
/// Checks whether the byte is a UTF-8 continuation byte (i.e., starts with the
|
2015-03-10 11:06:44 +00:00
|
|
|
|
/// bits `10`).
|
|
|
|
|
#[inline]
|
2019-12-22 22:42:04 +00:00
|
|
|
|
fn utf8_is_cont_byte(byte: u8) -> bool {
|
|
|
|
|
(byte & !CONT_MASK) == TAG_CONT_U8
|
|
|
|
|
}
|
2014-07-17 17:34:07 +00:00
|
|
|
|
|
|
|
|
|
#[inline]
|
|
|
|
|
fn unwrap_or_0(opt: Option<&u8>) -> u8 {
|
|
|
|
|
match opt {
|
|
|
|
|
Some(&byte) => byte,
|
|
|
|
|
None => 0,
|
|
|
|
|
}
|
2014-05-01 06:06:36 +00:00
|
|
|
|
}
|
|
|
|
|
|
2015-01-21 23:55:31 +00:00
|
|
|
|
/// Reads the next code point out of a byte iterator (assuming a
|
|
|
|
|
/// UTF-8-like encoding).
|
2019-12-21 11:16:18 +00:00
|
|
|
|
#[unstable(feature = "str_internals", issue = "none")]
|
2015-01-27 13:09:18 +00:00
|
|
|
|
#[inline]
|
2016-05-27 01:44:13 +00:00
|
|
|
|
pub fn next_code_point<'a, I: Iterator<Item = &'a u8>>(bytes: &mut I) -> Option<u32> {
|
2015-01-21 23:55:31 +00:00
|
|
|
|
// Decode UTF-8
|
2017-12-09 01:32:04 +00:00
|
|
|
|
let x = *bytes.next()?;
|
|
|
|
|
if x < 128 {
|
2019-12-22 22:42:04 +00:00
|
|
|
|
return Some(x as u32);
|
2017-12-09 01:32:04 +00:00
|
|
|
|
}
|
2015-01-21 23:55:31 +00:00
|
|
|
|
|
|
|
|
|
// Multibyte case follows
|
|
|
|
|
// Decode from a byte combination out of: [[[x y] z] w]
|
|
|
|
|
// NOTE: Performance is sensitive to the exact formulation here
|
2015-03-10 11:06:44 +00:00
|
|
|
|
let init = utf8_first_byte(x, 2);
|
2015-01-21 23:55:31 +00:00
|
|
|
|
let y = unwrap_or_0(bytes.next());
|
2015-03-10 11:06:44 +00:00
|
|
|
|
let mut ch = utf8_acc_cont_byte(init, y);
|
2015-01-21 23:55:31 +00:00
|
|
|
|
if x >= 0xE0 {
|
|
|
|
|
// [[x y z] w] case
|
|
|
|
|
// 5th bit in 0xE0 .. 0xEF is always clear, so `init` is still valid
|
|
|
|
|
let z = unwrap_or_0(bytes.next());
|
2015-03-10 11:06:44 +00:00
|
|
|
|
let y_z = utf8_acc_cont_byte((y & CONT_MASK) as u32, z);
|
2015-01-21 23:55:31 +00:00
|
|
|
|
ch = init << 12 | y_z;
|
|
|
|
|
if x >= 0xF0 {
|
|
|
|
|
// [x y z w] case
|
|
|
|
|
// use only the lower 3 bits of `init`
|
|
|
|
|
let w = unwrap_or_0(bytes.next());
|
2015-03-10 11:06:44 +00:00
|
|
|
|
ch = (init & 7) << 18 | utf8_acc_cont_byte(y_z, w);
|
2015-01-21 23:55:31 +00:00
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
Some(ch)
|
|
|
|
|
}
|
|
|
|
|
|
2015-01-27 13:09:18 +00:00
|
|
|
|
/// Reads the last code point out of a byte iterator (assuming a
|
|
|
|
|
/// UTF-8-like encoding).
|
|
|
|
|
#[inline]
|
2016-08-23 00:05:53 +00:00
|
|
|
|
fn next_code_point_reverse<'a, I>(bytes: &mut I) -> Option<u32>
|
2019-12-22 22:42:04 +00:00
|
|
|
|
where
|
|
|
|
|
I: DoubleEndedIterator<Item = &'a u8>,
|
2016-08-23 00:05:53 +00:00
|
|
|
|
{
|
2015-01-27 13:09:18 +00:00
|
|
|
|
// Decode UTF-8
|
2018-11-21 04:01:56 +00:00
|
|
|
|
let w = match *bytes.next_back()? {
|
|
|
|
|
next_byte if next_byte < 128 => return Some(next_byte as u32),
|
|
|
|
|
back_byte => back_byte,
|
2015-01-27 13:09:18 +00:00
|
|
|
|
};
|
|
|
|
|
|
|
|
|
|
// Multibyte case follows
|
|
|
|
|
// Decode from a byte combination out of: [x [y [z w]]]
|
|
|
|
|
let mut ch;
|
|
|
|
|
let z = unwrap_or_0(bytes.next_back());
|
2015-03-10 11:06:44 +00:00
|
|
|
|
ch = utf8_first_byte(z, 2);
|
|
|
|
|
if utf8_is_cont_byte(z) {
|
2015-01-27 13:09:18 +00:00
|
|
|
|
let y = unwrap_or_0(bytes.next_back());
|
2015-03-10 11:06:44 +00:00
|
|
|
|
ch = utf8_first_byte(y, 3);
|
|
|
|
|
if utf8_is_cont_byte(y) {
|
2015-01-27 13:09:18 +00:00
|
|
|
|
let x = unwrap_or_0(bytes.next_back());
|
2015-03-10 11:06:44 +00:00
|
|
|
|
ch = utf8_first_byte(x, 4);
|
|
|
|
|
ch = utf8_acc_cont_byte(ch, y);
|
2015-01-27 13:09:18 +00:00
|
|
|
|
}
|
2015-03-10 11:06:44 +00:00
|
|
|
|
ch = utf8_acc_cont_byte(ch, z);
|
2015-01-27 13:09:18 +00:00
|
|
|
|
}
|
2015-03-10 11:06:44 +00:00
|
|
|
|
ch = utf8_acc_cont_byte(ch, w);
|
2015-01-27 13:09:18 +00:00
|
|
|
|
|
|
|
|
|
Some(ch)
|
|
|
|
|
}
|
|
|
|
|
|
2015-01-24 05:48:20 +00:00
|
|
|
|
#[stable(feature = "rust1", since = "1.0.0")]
|
2014-12-29 21:18:41 +00:00
|
|
|
|
impl<'a> Iterator for Chars<'a> {
|
|
|
|
|
type Item = char;
|
|
|
|
|
|
2014-05-01 06:06:36 +00:00
|
|
|
|
#[inline]
|
|
|
|
|
fn next(&mut self) -> Option<char> {
|
2015-01-21 23:55:31 +00:00
|
|
|
|
next_code_point(&mut self.iter).map(|ch| {
|
2019-11-20 02:28:32 +00:00
|
|
|
|
// SAFETY: str invariant says `ch` is a valid Unicode Scalar Value
|
2019-12-22 22:42:04 +00:00
|
|
|
|
unsafe { char::from_u32_unchecked(ch) }
|
2015-01-21 23:55:31 +00:00
|
|
|
|
})
|
2014-05-01 06:06:36 +00:00
|
|
|
|
}
|
|
|
|
|
|
2016-11-19 22:18:43 +00:00
|
|
|
|
#[inline]
|
|
|
|
|
fn count(self) -> usize {
|
|
|
|
|
// length in `char` is equal to the number of non-continuation bytes
|
|
|
|
|
let bytes_len = self.iter.len();
|
|
|
|
|
let mut cont_bytes = 0;
|
|
|
|
|
for &byte in self.iter {
|
|
|
|
|
cont_bytes += utf8_is_cont_byte(byte) as usize;
|
|
|
|
|
}
|
|
|
|
|
bytes_len - cont_bytes
|
|
|
|
|
}
|
|
|
|
|
|
2014-05-01 06:06:36 +00:00
|
|
|
|
#[inline]
|
2015-01-27 13:09:18 +00:00
|
|
|
|
fn size_hint(&self) -> (usize, Option<usize>) {
|
2016-06-23 10:26:15 +00:00
|
|
|
|
let len = self.iter.len();
|
2015-03-10 13:18:24 +00:00
|
|
|
|
// `(len + 3)` can't overflow, because we know that the `slice::Iter`
|
|
|
|
|
// belongs to a slice in memory which has a maximum length of
|
|
|
|
|
// `isize::MAX` (that's well below `usize::MAX`).
|
|
|
|
|
((len + 3) / 4, Some(len))
|
2014-05-01 06:06:36 +00:00
|
|
|
|
}
|
2016-11-19 18:43:41 +00:00
|
|
|
|
|
|
|
|
|
#[inline]
|
|
|
|
|
fn last(mut self) -> Option<char> {
|
|
|
|
|
// No need to go through the entire string.
|
|
|
|
|
self.next_back()
|
|
|
|
|
}
|
2014-05-01 06:06:36 +00:00
|
|
|
|
}
|
|
|
|
|
|
2019-07-26 06:58:37 +00:00
|
|
|
|
#[stable(feature = "chars_debug_impl", since = "1.38.0")]
|
2019-07-29 16:26:59 +00:00
|
|
|
|
impl fmt::Debug for Chars<'_> {
|
2019-07-26 06:58:37 +00:00
|
|
|
|
fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
|
|
|
|
|
write!(f, "Chars(")?;
|
|
|
|
|
f.debug_list().entries(self.clone()).finish()?;
|
|
|
|
|
write!(f, ")")?;
|
|
|
|
|
Ok(())
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
2015-01-24 05:48:20 +00:00
|
|
|
|
#[stable(feature = "rust1", since = "1.0.0")]
|
2014-12-29 21:18:41 +00:00
|
|
|
|
impl<'a> DoubleEndedIterator for Chars<'a> {
|
2014-05-01 06:06:36 +00:00
|
|
|
|
#[inline]
|
|
|
|
|
fn next_back(&mut self) -> Option<char> {
|
2015-01-27 13:09:18 +00:00
|
|
|
|
next_code_point_reverse(&mut self.iter).map(|ch| {
|
2019-11-20 02:28:32 +00:00
|
|
|
|
// SAFETY: str invariant says `ch` is a valid Unicode Scalar Value
|
2019-12-22 22:42:04 +00:00
|
|
|
|
unsafe { char::from_u32_unchecked(ch) }
|
2015-01-27 13:09:18 +00:00
|
|
|
|
})
|
2014-05-01 06:06:36 +00:00
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
2018-03-03 13:15:28 +00:00
|
|
|
|
#[stable(feature = "fused", since = "1.26.0")]
|
2018-09-03 11:50:14 +00:00
|
|
|
|
impl FusedIterator for Chars<'_> {}
|
2016-08-13 18:42:36 +00:00
|
|
|
|
|
2015-08-19 15:31:46 +00:00
|
|
|
|
impl<'a> Chars<'a> {
|
2019-02-09 21:23:30 +00:00
|
|
|
|
/// Views the underlying data as a subslice of the original data.
|
2015-08-19 15:31:46 +00:00
|
|
|
|
///
|
|
|
|
|
/// This has the same lifetime as the original slice, and so the
|
|
|
|
|
/// iterator can continue to be used while this exists.
|
2016-07-27 02:52:56 +00:00
|
|
|
|
///
|
|
|
|
|
/// # Examples
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// let mut chars = "abc".chars();
|
|
|
|
|
///
|
|
|
|
|
/// assert_eq!(chars.as_str(), "abc");
|
|
|
|
|
/// chars.next();
|
|
|
|
|
/// assert_eq!(chars.as_str(), "bc");
|
|
|
|
|
/// chars.next();
|
|
|
|
|
/// chars.next();
|
|
|
|
|
/// assert_eq!(chars.as_str(), "");
|
|
|
|
|
/// ```
|
2015-09-10 20:26:44 +00:00
|
|
|
|
#[stable(feature = "iter_to_slice", since = "1.4.0")]
|
2015-08-19 15:31:46 +00:00
|
|
|
|
#[inline]
|
|
|
|
|
pub fn as_str(&self) -> &'a str {
|
2019-12-26 19:57:57 +00:00
|
|
|
|
// SAFETY: `Chars` is only made from a str, which guarantees the iter is valid utf8
|
2015-08-19 15:31:46 +00:00
|
|
|
|
unsafe { from_utf8_unchecked(self.iter.as_slice()) }
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
2017-03-30 19:46:41 +00:00
|
|
|
|
/// An iterator over the [`char`]s of a string slice, and their positions.
|
|
|
|
|
///
|
|
|
|
|
/// [`char`]: ../../std/primitive.char.html
|
|
|
|
|
///
|
2017-03-30 22:33:23 +00:00
|
|
|
|
/// This struct is created by the [`char_indices`] method on [`str`].
|
2017-03-30 19:46:41 +00:00
|
|
|
|
/// See its documentation for more.
|
|
|
|
|
///
|
2017-03-30 22:33:23 +00:00
|
|
|
|
/// [`char_indices`]: ../../std/primitive.str.html#method.char_indices
|
2017-03-30 19:46:41 +00:00
|
|
|
|
/// [`str`]: ../../std/primitive.str.html
|
2016-03-05 02:49:43 +00:00
|
|
|
|
#[derive(Clone, Debug)]
|
2015-01-24 05:48:20 +00:00
|
|
|
|
#[stable(feature = "rust1", since = "1.0.0")]
|
std: Stabilize the std::str module
This commit starts out by consolidating all `str` extension traits into one
`StrExt` trait to be included in the prelude. This means that
`UnicodeStrPrelude`, `StrPrelude`, and `StrAllocating` have all been merged into
one `StrExt` exported by the standard library. Some functionality is currently
duplicated with the `StrExt` present in libcore.
This commit also currently avoids any methods which require any form of pattern
to operate. These functions will be stabilized via a separate RFC.
Next, stability of methods and structures are as follows:
Stable
* from_utf8_unchecked
* CowString - after moving to std::string
* StrExt::as_bytes
* StrExt::as_ptr
* StrExt::bytes/Bytes - also made a struct instead of a typedef
* StrExt::char_indices/CharIndices - CharOffsets was renamed
* StrExt::chars/Chars
* StrExt::is_empty
* StrExt::len
* StrExt::lines/Lines
* StrExt::lines_any/LinesAny
* StrExt::slice_unchecked
* StrExt::trim
* StrExt::trim_left
* StrExt::trim_right
* StrExt::words/Words - also made a struct instead of a typedef
Unstable
* from_utf8 - the error type was changed to a `Result`, but the error type has
yet to prove itself
* from_c_str - this function will be handled by the c_str RFC
* FromStr - this trait will have an associated error type eventually
* StrExt::escape_default - needs iterators at least, unsure if it should make
the cut
* StrExt::escape_unicode - needs iterators at least, unsure if it should make
the cut
* StrExt::slice_chars - this function has yet to prove itself
* StrExt::slice_shift_char - awaiting conventions about slicing and shifting
* StrExt::graphemes/Graphemes - this functionality may only be in libunicode
* StrExt::grapheme_indices/GraphemeIndices - this functionality may only be in
libunicode
* StrExt::width - this functionality may only be in libunicode
* StrExt::utf16_units - this functionality may only be in libunicode
* StrExt::nfd_chars - this functionality may only be in libunicode
* StrExt::nfkd_chars - this functionality may only be in libunicode
* StrExt::nfc_chars - this functionality may only be in libunicode
* StrExt::nfkc_chars - this functionality may only be in libunicode
* StrExt::is_char_boundary - naming is uncertain with container conventions
* StrExt::char_range_at - naming is uncertain with container conventions
* StrExt::char_range_at_reverse - naming is uncertain with container conventions
* StrExt::char_at - naming is uncertain with container conventions
* StrExt::char_at_reverse - naming is uncertain with container conventions
* StrVector::concat - this functionality may be replaced with iterators, but
it's not certain at this time
* StrVector::connect - as with concat, may be deprecated in favor of iterators
Deprecated
* StrAllocating and UnicodeStrPrelude have been merged into StrExit
* eq_slice - compiler implementation detail
* from_str - use the inherent parse() method
* is_utf8 - call from_utf8 instead
* replace - call the method instead
* truncate_utf16_at_nul - this is an implementation detail of windows and does
not need to be exposed.
* utf8_char_width - moved to libunicode
* utf16_items - moved to libunicode
* is_utf16 - moved to libunicode
* Utf16Items - moved to libunicode
* Utf16Item - moved to libunicode
* Utf16Encoder - moved to libunicode
* AnyLines - renamed to LinesAny and made a struct
* SendStr - use CowString<'static> instead
* str::raw - all functionality is deprecated
* StrExt::into_string - call to_string() instead
* StrExt::repeat - use iterators instead
* StrExt::char_len - use .chars().count() instead
* StrExt::is_alphanumeric - use .chars().all(..)
* StrExt::is_whitespace - use .chars().all(..)
Pending deprecation -- while slicing syntax is being worked out, these methods
are all #[unstable]
* Str - while currently used for generic programming, this trait will be
replaced with one of [], deref coercions, or a generic conversion trait.
* StrExt::slice - use slicing syntax instead
* StrExt::slice_to - use slicing syntax instead
* StrExt::slice_from - use slicing syntax instead
* StrExt::lev_distance - deprecated with no replacement
Awaiting stabilization due to patterns and/or matching
* StrExt::contains
* StrExt::contains_char
* StrExt::split
* StrExt::splitn
* StrExt::split_terminator
* StrExt::rsplitn
* StrExt::match_indices
* StrExt::split_str
* StrExt::starts_with
* StrExt::ends_with
* StrExt::trim_chars
* StrExt::trim_left_chars
* StrExt::trim_right_chars
* StrExt::find
* StrExt::rfind
* StrExt::find_str
* StrExt::subslice_offset
2014-12-10 17:02:31 +00:00
|
|
|
|
pub struct CharIndices<'a> {
|
2015-01-27 13:09:18 +00:00
|
|
|
|
front_offset: usize,
|
2014-05-01 06:06:36 +00:00
|
|
|
|
iter: Chars<'a>,
|
|
|
|
|
}
|
|
|
|
|
|
2015-01-24 05:48:20 +00:00
|
|
|
|
#[stable(feature = "rust1", since = "1.0.0")]
|
2014-12-29 21:18:41 +00:00
|
|
|
|
impl<'a> Iterator for CharIndices<'a> {
|
2015-01-27 13:09:18 +00:00
|
|
|
|
type Item = (usize, char);
|
2014-12-29 21:18:41 +00:00
|
|
|
|
|
2014-05-01 06:06:36 +00:00
|
|
|
|
#[inline]
|
2015-01-27 13:09:18 +00:00
|
|
|
|
fn next(&mut self) -> Option<(usize, char)> {
|
2016-06-23 10:26:15 +00:00
|
|
|
|
let pre_len = self.iter.iter.len();
|
2014-07-17 17:34:07 +00:00
|
|
|
|
match self.iter.next() {
|
|
|
|
|
None => None,
|
|
|
|
|
Some(ch) => {
|
2014-07-19 13:39:02 +00:00
|
|
|
|
let index = self.front_offset;
|
2016-06-23 10:26:15 +00:00
|
|
|
|
let len = self.iter.iter.len();
|
2014-07-19 13:39:02 +00:00
|
|
|
|
self.front_offset += pre_len - len;
|
2014-07-17 17:34:07 +00:00
|
|
|
|
Some((index, ch))
|
|
|
|
|
}
|
|
|
|
|
}
|
2014-05-01 06:06:36 +00:00
|
|
|
|
}
|
|
|
|
|
|
2016-11-19 22:18:43 +00:00
|
|
|
|
#[inline]
|
|
|
|
|
fn count(self) -> usize {
|
|
|
|
|
self.iter.count()
|
|
|
|
|
}
|
|
|
|
|
|
2014-05-01 06:06:36 +00:00
|
|
|
|
#[inline]
|
2015-01-27 13:09:18 +00:00
|
|
|
|
fn size_hint(&self) -> (usize, Option<usize>) {
|
2014-05-01 06:06:36 +00:00
|
|
|
|
self.iter.size_hint()
|
|
|
|
|
}
|
2016-11-20 00:37:48 +00:00
|
|
|
|
|
|
|
|
|
#[inline]
|
|
|
|
|
fn last(mut self) -> Option<(usize, char)> {
|
|
|
|
|
// No need to go through the entire string.
|
|
|
|
|
self.next_back()
|
|
|
|
|
}
|
2014-05-01 06:06:36 +00:00
|
|
|
|
}
|
|
|
|
|
|
2015-01-24 05:48:20 +00:00
|
|
|
|
#[stable(feature = "rust1", since = "1.0.0")]
|
2014-12-29 21:18:41 +00:00
|
|
|
|
impl<'a> DoubleEndedIterator for CharIndices<'a> {
|
2014-05-01 06:06:36 +00:00
|
|
|
|
#[inline]
|
2015-01-27 13:09:18 +00:00
|
|
|
|
fn next_back(&mut self) -> Option<(usize, char)> {
|
2018-07-24 05:04:33 +00:00
|
|
|
|
self.iter.next_back().map(|ch| {
|
|
|
|
|
let index = self.front_offset + self.iter.iter.len();
|
|
|
|
|
(index, ch)
|
|
|
|
|
})
|
2014-05-01 06:06:36 +00:00
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
2018-03-03 13:15:28 +00:00
|
|
|
|
#[stable(feature = "fused", since = "1.26.0")]
|
2018-09-03 11:50:14 +00:00
|
|
|
|
impl FusedIterator for CharIndices<'_> {}
|
2016-08-13 18:42:36 +00:00
|
|
|
|
|
2015-08-19 15:31:46 +00:00
|
|
|
|
impl<'a> CharIndices<'a> {
|
2019-02-09 21:23:30 +00:00
|
|
|
|
/// Views the underlying data as a subslice of the original data.
|
2015-08-19 15:31:46 +00:00
|
|
|
|
///
|
|
|
|
|
/// This has the same lifetime as the original slice, and so the
|
|
|
|
|
/// iterator can continue to be used while this exists.
|
2015-09-10 20:26:44 +00:00
|
|
|
|
#[stable(feature = "iter_to_slice", since = "1.4.0")]
|
2015-08-19 15:31:46 +00:00
|
|
|
|
#[inline]
|
|
|
|
|
pub fn as_str(&self) -> &'a str {
|
|
|
|
|
self.iter.as_str()
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
2017-03-30 19:40:05 +00:00
|
|
|
|
/// An iterator over the bytes of a string slice.
|
2014-12-18 01:12:53 +00:00
|
|
|
|
///
|
2017-03-30 22:33:23 +00:00
|
|
|
|
/// This struct is created by the [`bytes`] method on [`str`].
|
2017-03-30 19:40:05 +00:00
|
|
|
|
/// See its documentation for more.
|
2015-12-11 13:25:50 +00:00
|
|
|
|
///
|
2017-03-12 18:04:52 +00:00
|
|
|
|
/// [`bytes`]: ../../std/primitive.str.html#method.bytes
|
2017-03-30 19:40:05 +00:00
|
|
|
|
/// [`str`]: ../../std/primitive.str.html
|
2015-01-24 05:48:20 +00:00
|
|
|
|
#[stable(feature = "rust1", since = "1.0.0")]
|
2016-03-05 02:49:43 +00:00
|
|
|
|
#[derive(Clone, Debug)]
|
2015-08-31 09:53:42 +00:00
|
|
|
|
pub struct Bytes<'a>(Cloned<slice::Iter<'a, u8>>);
|
2015-02-15 20:09:26 +00:00
|
|
|
|
|
2015-03-15 00:48:34 +00:00
|
|
|
|
#[stable(feature = "rust1", since = "1.0.0")]
|
2018-09-03 11:50:14 +00:00
|
|
|
|
impl Iterator for Bytes<'_> {
|
2015-03-15 00:48:34 +00:00
|
|
|
|
type Item = u8;
|
2014-05-01 06:06:36 +00:00
|
|
|
|
|
2015-03-15 00:48:34 +00:00
|
|
|
|
#[inline]
|
|
|
|
|
fn next(&mut self) -> Option<u8> {
|
|
|
|
|
self.0.next()
|
|
|
|
|
}
|
2014-05-01 06:06:36 +00:00
|
|
|
|
|
2015-03-15 00:48:34 +00:00
|
|
|
|
#[inline]
|
|
|
|
|
fn size_hint(&self) -> (usize, Option<usize>) {
|
|
|
|
|
self.0.size_hint()
|
|
|
|
|
}
|
2015-08-30 15:32:43 +00:00
|
|
|
|
|
|
|
|
|
#[inline]
|
|
|
|
|
fn count(self) -> usize {
|
|
|
|
|
self.0.count()
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
#[inline]
|
|
|
|
|
fn last(self) -> Option<Self::Item> {
|
|
|
|
|
self.0.last()
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
#[inline]
|
|
|
|
|
fn nth(&mut self, n: usize) -> Option<Self::Item> {
|
|
|
|
|
self.0.nth(n)
|
|
|
|
|
}
|
2017-07-05 18:21:29 +00:00
|
|
|
|
|
|
|
|
|
#[inline]
|
2019-12-22 22:42:04 +00:00
|
|
|
|
fn all<F>(&mut self, f: F) -> bool
|
|
|
|
|
where
|
|
|
|
|
F: FnMut(Self::Item) -> bool,
|
|
|
|
|
{
|
2017-07-05 18:21:29 +00:00
|
|
|
|
self.0.all(f)
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
#[inline]
|
2019-12-22 22:42:04 +00:00
|
|
|
|
fn any<F>(&mut self, f: F) -> bool
|
|
|
|
|
where
|
|
|
|
|
F: FnMut(Self::Item) -> bool,
|
|
|
|
|
{
|
2017-07-05 18:21:29 +00:00
|
|
|
|
self.0.any(f)
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
#[inline]
|
2019-12-22 22:42:04 +00:00
|
|
|
|
fn find<P>(&mut self, predicate: P) -> Option<Self::Item>
|
|
|
|
|
where
|
|
|
|
|
P: FnMut(&Self::Item) -> bool,
|
2017-07-05 18:21:29 +00:00
|
|
|
|
{
|
|
|
|
|
self.0.find(predicate)
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
#[inline]
|
2019-12-22 22:42:04 +00:00
|
|
|
|
fn position<P>(&mut self, predicate: P) -> Option<usize>
|
|
|
|
|
where
|
|
|
|
|
P: FnMut(Self::Item) -> bool,
|
2017-07-05 18:21:29 +00:00
|
|
|
|
{
|
|
|
|
|
self.0.position(predicate)
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
#[inline]
|
2019-12-22 22:42:04 +00:00
|
|
|
|
fn rposition<P>(&mut self, predicate: P) -> Option<usize>
|
|
|
|
|
where
|
|
|
|
|
P: FnMut(Self::Item) -> bool,
|
2017-07-05 18:21:29 +00:00
|
|
|
|
{
|
|
|
|
|
self.0.rposition(predicate)
|
|
|
|
|
}
|
2014-05-01 06:06:36 +00:00
|
|
|
|
}
|
|
|
|
|
|
2015-03-15 00:48:34 +00:00
|
|
|
|
#[stable(feature = "rust1", since = "1.0.0")]
|
2018-09-03 11:50:14 +00:00
|
|
|
|
impl DoubleEndedIterator for Bytes<'_> {
|
2015-03-15 00:48:34 +00:00
|
|
|
|
#[inline]
|
|
|
|
|
fn next_back(&mut self) -> Option<u8> {
|
|
|
|
|
self.0.next_back()
|
|
|
|
|
}
|
2017-07-05 18:21:29 +00:00
|
|
|
|
|
|
|
|
|
#[inline]
|
2019-04-16 18:41:23 +00:00
|
|
|
|
fn nth_back(&mut self, n: usize) -> Option<Self::Item> {
|
|
|
|
|
self.0.nth_back(n)
|
|
|
|
|
}
|
|
|
|
|
|
2017-07-05 18:21:29 +00:00
|
|
|
|
#[inline]
|
2019-12-22 22:42:04 +00:00
|
|
|
|
fn rfind<P>(&mut self, predicate: P) -> Option<Self::Item>
|
|
|
|
|
where
|
|
|
|
|
P: FnMut(&Self::Item) -> bool,
|
2017-07-05 18:21:29 +00:00
|
|
|
|
{
|
|
|
|
|
self.0.rfind(predicate)
|
|
|
|
|
}
|
2014-05-01 06:06:36 +00:00
|
|
|
|
}
|
|
|
|
|
|
2015-01-24 05:48:20 +00:00
|
|
|
|
#[stable(feature = "rust1", since = "1.0.0")]
|
2018-09-03 11:50:14 +00:00
|
|
|
|
impl ExactSizeIterator for Bytes<'_> {
|
2015-03-15 00:48:34 +00:00
|
|
|
|
#[inline]
|
|
|
|
|
fn len(&self) -> usize {
|
|
|
|
|
self.0.len()
|
|
|
|
|
}
|
2016-12-03 20:43:51 +00:00
|
|
|
|
|
|
|
|
|
#[inline]
|
|
|
|
|
fn is_empty(&self) -> bool {
|
|
|
|
|
self.0.is_empty()
|
|
|
|
|
}
|
2015-03-14 23:34:21 +00:00
|
|
|
|
}
|
|
|
|
|
|
2018-03-03 13:15:28 +00:00
|
|
|
|
#[stable(feature = "fused", since = "1.26.0")]
|
2018-09-03 11:50:14 +00:00
|
|
|
|
impl FusedIterator for Bytes<'_> {}
|
2016-08-13 18:42:36 +00:00
|
|
|
|
|
2017-09-23 16:03:24 +00:00
|
|
|
|
#[unstable(feature = "trusted_len", issue = "37572")]
|
2018-09-03 11:50:14 +00:00
|
|
|
|
unsafe impl TrustedLen for Bytes<'_> {}
|
2017-09-23 16:03:24 +00:00
|
|
|
|
|
|
|
|
|
#[doc(hidden)]
|
2019-02-18 03:42:36 +00:00
|
|
|
|
unsafe impl TrustedRandomAccess for Bytes<'_> {
|
2017-09-23 16:03:24 +00:00
|
|
|
|
unsafe fn get_unchecked(&mut self, i: usize) -> u8 {
|
|
|
|
|
self.0.get_unchecked(i)
|
|
|
|
|
}
|
2019-12-22 22:42:04 +00:00
|
|
|
|
fn may_have_side_effect() -> bool {
|
|
|
|
|
false
|
|
|
|
|
}
|
2017-09-23 16:03:24 +00:00
|
|
|
|
}
|
|
|
|
|
|
2015-04-05 16:52:14 +00:00
|
|
|
|
/// This macro generates a Clone impl for string pattern API
|
|
|
|
|
/// wrapper types of the form X<'a, P>
|
|
|
|
|
macro_rules! derive_pattern_clone {
|
|
|
|
|
(clone $t:ident with |$s:ident| $e:expr) => {
|
2019-07-31 19:00:35 +00:00
|
|
|
|
impl<'a, P> Clone for $t<'a, P>
|
|
|
|
|
where
|
|
|
|
|
P: Pattern<'a, Searcher: Clone>,
|
2015-04-05 16:52:14 +00:00
|
|
|
|
{
|
|
|
|
|
fn clone(&self) -> Self {
|
|
|
|
|
let $s = self;
|
|
|
|
|
$e
|
|
|
|
|
}
|
|
|
|
|
}
|
2019-12-22 22:42:04 +00:00
|
|
|
|
};
|
2015-03-15 00:07:13 +00:00
|
|
|
|
}
|
2015-03-14 23:34:21 +00:00
|
|
|
|
|
2015-03-15 00:48:34 +00:00
|
|
|
|
/// This macro generates two public iterator structs
|
2015-10-07 22:11:25 +00:00
|
|
|
|
/// wrapping a private internal one that makes use of the `Pattern` API.
|
2015-03-15 00:48:34 +00:00
|
|
|
|
///
|
|
|
|
|
/// For all patterns `P: Pattern<'a>` the following items will be
|
2015-05-04 17:21:27 +00:00
|
|
|
|
/// generated (generics omitted):
|
2015-03-15 00:48:34 +00:00
|
|
|
|
///
|
|
|
|
|
/// struct $forward_iterator($internal_iterator);
|
|
|
|
|
/// struct $reverse_iterator($internal_iterator);
|
|
|
|
|
///
|
|
|
|
|
/// impl Iterator for $forward_iterator
|
|
|
|
|
/// { /* internal ends up calling Searcher::next_match() */ }
|
|
|
|
|
///
|
|
|
|
|
/// impl DoubleEndedIterator for $forward_iterator
|
|
|
|
|
/// where P::Searcher: DoubleEndedSearcher
|
|
|
|
|
/// { /* internal ends up calling Searcher::next_match_back() */ }
|
|
|
|
|
///
|
|
|
|
|
/// impl Iterator for $reverse_iterator
|
|
|
|
|
/// where P::Searcher: ReverseSearcher
|
|
|
|
|
/// { /* internal ends up calling Searcher::next_match_back() */ }
|
|
|
|
|
///
|
|
|
|
|
/// impl DoubleEndedIterator for $reverse_iterator
|
|
|
|
|
/// where P::Searcher: DoubleEndedSearcher
|
|
|
|
|
/// { /* internal ends up calling Searcher::next_match() */ }
|
|
|
|
|
///
|
|
|
|
|
/// The internal one is defined outside the macro, and has almost the same
|
|
|
|
|
/// semantic as a DoubleEndedIterator by delegating to `pattern::Searcher` and
|
|
|
|
|
/// `pattern::ReverseSearcher` for both forward and reverse iteration.
|
|
|
|
|
///
|
|
|
|
|
/// "Almost", because a `Searcher` and a `ReverseSearcher` for a given
|
|
|
|
|
/// `Pattern` might not return the same elements, so actually implementing
|
|
|
|
|
/// `DoubleEndedIterator` for it would be incorrect.
|
|
|
|
|
/// (See the docs in `str::pattern` for more details)
|
|
|
|
|
///
|
|
|
|
|
/// However, the internal struct still represents a single ended iterator from
|
|
|
|
|
/// either end, and depending on pattern is also a valid double ended iterator,
|
|
|
|
|
/// so the two wrapper structs implement `Iterator`
|
|
|
|
|
/// and `DoubleEndedIterator` depending on the concrete pattern type, leading
|
|
|
|
|
/// to the complex impls seen above.
|
|
|
|
|
macro_rules! generate_pattern_iterators {
|
|
|
|
|
{
|
|
|
|
|
// Forward iterator
|
|
|
|
|
forward:
|
|
|
|
|
$(#[$forward_iterator_attribute:meta])*
|
|
|
|
|
struct $forward_iterator:ident;
|
|
|
|
|
|
|
|
|
|
// Reverse iterator
|
|
|
|
|
reverse:
|
|
|
|
|
$(#[$reverse_iterator_attribute:meta])*
|
|
|
|
|
struct $reverse_iterator:ident;
|
|
|
|
|
|
|
|
|
|
// Stability of all generated items
|
|
|
|
|
stability:
|
|
|
|
|
$(#[$common_stability_attribute:meta])*
|
|
|
|
|
|
|
|
|
|
// Internal almost-iterator that is being delegated to
|
|
|
|
|
internal:
|
|
|
|
|
$internal_iterator:ident yielding ($iterty:ty);
|
|
|
|
|
|
2017-07-23 15:06:16 +00:00
|
|
|
|
// Kind of delegation - either single ended or double ended
|
2015-03-15 00:48:34 +00:00
|
|
|
|
delegate $($t:tt)*
|
|
|
|
|
} => {
|
|
|
|
|
$(#[$forward_iterator_attribute])*
|
|
|
|
|
$(#[$common_stability_attribute])*
|
|
|
|
|
pub struct $forward_iterator<'a, P: Pattern<'a>>($internal_iterator<'a, P>);
|
|
|
|
|
|
2016-03-05 02:49:43 +00:00
|
|
|
|
$(#[$common_stability_attribute])*
|
2019-07-31 19:00:35 +00:00
|
|
|
|
impl<'a, P> fmt::Debug for $forward_iterator<'a, P>
|
|
|
|
|
where
|
|
|
|
|
P: Pattern<'a, Searcher: fmt::Debug>,
|
2016-03-05 02:49:43 +00:00
|
|
|
|
{
|
2019-04-18 23:37:12 +00:00
|
|
|
|
fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
|
2016-03-05 02:49:43 +00:00
|
|
|
|
f.debug_tuple(stringify!($forward_iterator))
|
|
|
|
|
.field(&self.0)
|
|
|
|
|
.finish()
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
2015-03-15 00:48:34 +00:00
|
|
|
|
$(#[$common_stability_attribute])*
|
|
|
|
|
impl<'a, P: Pattern<'a>> Iterator for $forward_iterator<'a, P> {
|
|
|
|
|
type Item = $iterty;
|
|
|
|
|
|
|
|
|
|
#[inline]
|
|
|
|
|
fn next(&mut self) -> Option<$iterty> {
|
|
|
|
|
self.0.next()
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
2015-04-05 16:52:14 +00:00
|
|
|
|
$(#[$common_stability_attribute])*
|
2019-07-31 19:00:35 +00:00
|
|
|
|
impl<'a, P> Clone for $forward_iterator<'a, P>
|
|
|
|
|
where
|
|
|
|
|
P: Pattern<'a, Searcher: Clone>,
|
2015-04-05 16:52:14 +00:00
|
|
|
|
{
|
|
|
|
|
fn clone(&self) -> Self {
|
|
|
|
|
$forward_iterator(self.0.clone())
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
2015-03-15 00:48:34 +00:00
|
|
|
|
$(#[$reverse_iterator_attribute])*
|
|
|
|
|
$(#[$common_stability_attribute])*
|
|
|
|
|
pub struct $reverse_iterator<'a, P: Pattern<'a>>($internal_iterator<'a, P>);
|
|
|
|
|
|
2016-03-05 02:49:43 +00:00
|
|
|
|
$(#[$common_stability_attribute])*
|
2019-07-31 19:00:35 +00:00
|
|
|
|
impl<'a, P> fmt::Debug for $reverse_iterator<'a, P>
|
|
|
|
|
where
|
|
|
|
|
P: Pattern<'a, Searcher: fmt::Debug>,
|
2016-03-05 02:49:43 +00:00
|
|
|
|
{
|
2019-04-18 23:37:12 +00:00
|
|
|
|
fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
|
2016-03-05 02:49:43 +00:00
|
|
|
|
f.debug_tuple(stringify!($reverse_iterator))
|
|
|
|
|
.field(&self.0)
|
|
|
|
|
.finish()
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
2015-03-15 00:48:34 +00:00
|
|
|
|
$(#[$common_stability_attribute])*
|
2019-07-31 19:00:35 +00:00
|
|
|
|
impl<'a, P> Iterator for $reverse_iterator<'a, P>
|
|
|
|
|
where
|
|
|
|
|
P: Pattern<'a, Searcher: ReverseSearcher<'a>>,
|
2015-03-15 00:48:34 +00:00
|
|
|
|
{
|
|
|
|
|
type Item = $iterty;
|
|
|
|
|
|
|
|
|
|
#[inline]
|
|
|
|
|
fn next(&mut self) -> Option<$iterty> {
|
|
|
|
|
self.0.next_back()
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
2015-04-05 16:52:14 +00:00
|
|
|
|
$(#[$common_stability_attribute])*
|
2019-07-31 19:00:35 +00:00
|
|
|
|
impl<'a, P> Clone for $reverse_iterator<'a, P>
|
|
|
|
|
where
|
|
|
|
|
P: Pattern<'a, Searcher: Clone>,
|
2015-04-05 16:52:14 +00:00
|
|
|
|
{
|
|
|
|
|
fn clone(&self) -> Self {
|
|
|
|
|
$reverse_iterator(self.0.clone())
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
2018-03-03 13:15:28 +00:00
|
|
|
|
#[stable(feature = "fused", since = "1.26.0")]
|
2016-08-13 18:42:36 +00:00
|
|
|
|
impl<'a, P: Pattern<'a>> FusedIterator for $forward_iterator<'a, P> {}
|
|
|
|
|
|
2018-03-03 13:15:28 +00:00
|
|
|
|
#[stable(feature = "fused", since = "1.26.0")]
|
2019-07-31 19:00:35 +00:00
|
|
|
|
impl<'a, P> FusedIterator for $reverse_iterator<'a, P>
|
|
|
|
|
where
|
|
|
|
|
P: Pattern<'a, Searcher: ReverseSearcher<'a>>,
|
|
|
|
|
{}
|
2016-08-13 18:42:36 +00:00
|
|
|
|
|
2015-03-15 00:48:34 +00:00
|
|
|
|
generate_pattern_iterators!($($t)* with $(#[$common_stability_attribute])*,
|
|
|
|
|
$forward_iterator,
|
|
|
|
|
$reverse_iterator, $iterty);
|
|
|
|
|
};
|
|
|
|
|
{
|
|
|
|
|
double ended; with $(#[$common_stability_attribute:meta])*,
|
|
|
|
|
$forward_iterator:ident,
|
|
|
|
|
$reverse_iterator:ident, $iterty:ty
|
|
|
|
|
} => {
|
|
|
|
|
$(#[$common_stability_attribute])*
|
2019-07-31 19:00:35 +00:00
|
|
|
|
impl<'a, P> DoubleEndedIterator for $forward_iterator<'a, P>
|
|
|
|
|
where
|
|
|
|
|
P: Pattern<'a, Searcher: DoubleEndedSearcher<'a>>,
|
2015-03-15 00:48:34 +00:00
|
|
|
|
{
|
|
|
|
|
#[inline]
|
|
|
|
|
fn next_back(&mut self) -> Option<$iterty> {
|
|
|
|
|
self.0.next_back()
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
$(#[$common_stability_attribute])*
|
2019-07-31 19:00:35 +00:00
|
|
|
|
impl<'a, P> DoubleEndedIterator for $reverse_iterator<'a, P>
|
|
|
|
|
where
|
|
|
|
|
P: Pattern<'a, Searcher: DoubleEndedSearcher<'a>>,
|
2015-03-15 00:48:34 +00:00
|
|
|
|
{
|
|
|
|
|
#[inline]
|
|
|
|
|
fn next_back(&mut self) -> Option<$iterty> {
|
|
|
|
|
self.0.next()
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
};
|
|
|
|
|
{
|
|
|
|
|
single ended; with $(#[$common_stability_attribute:meta])*,
|
|
|
|
|
$forward_iterator:ident,
|
|
|
|
|
$reverse_iterator:ident, $iterty:ty
|
|
|
|
|
} => {}
|
std: Stabilize the std::str module
This commit starts out by consolidating all `str` extension traits into one
`StrExt` trait to be included in the prelude. This means that
`UnicodeStrPrelude`, `StrPrelude`, and `StrAllocating` have all been merged into
one `StrExt` exported by the standard library. Some functionality is currently
duplicated with the `StrExt` present in libcore.
This commit also currently avoids any methods which require any form of pattern
to operate. These functions will be stabilized via a separate RFC.
Next, stability of methods and structures are as follows:
Stable
* from_utf8_unchecked
* CowString - after moving to std::string
* StrExt::as_bytes
* StrExt::as_ptr
* StrExt::bytes/Bytes - also made a struct instead of a typedef
* StrExt::char_indices/CharIndices - CharOffsets was renamed
* StrExt::chars/Chars
* StrExt::is_empty
* StrExt::len
* StrExt::lines/Lines
* StrExt::lines_any/LinesAny
* StrExt::slice_unchecked
* StrExt::trim
* StrExt::trim_left
* StrExt::trim_right
* StrExt::words/Words - also made a struct instead of a typedef
Unstable
* from_utf8 - the error type was changed to a `Result`, but the error type has
yet to prove itself
* from_c_str - this function will be handled by the c_str RFC
* FromStr - this trait will have an associated error type eventually
* StrExt::escape_default - needs iterators at least, unsure if it should make
the cut
* StrExt::escape_unicode - needs iterators at least, unsure if it should make
the cut
* StrExt::slice_chars - this function has yet to prove itself
* StrExt::slice_shift_char - awaiting conventions about slicing and shifting
* StrExt::graphemes/Graphemes - this functionality may only be in libunicode
* StrExt::grapheme_indices/GraphemeIndices - this functionality may only be in
libunicode
* StrExt::width - this functionality may only be in libunicode
* StrExt::utf16_units - this functionality may only be in libunicode
* StrExt::nfd_chars - this functionality may only be in libunicode
* StrExt::nfkd_chars - this functionality may only be in libunicode
* StrExt::nfc_chars - this functionality may only be in libunicode
* StrExt::nfkc_chars - this functionality may only be in libunicode
* StrExt::is_char_boundary - naming is uncertain with container conventions
* StrExt::char_range_at - naming is uncertain with container conventions
* StrExt::char_range_at_reverse - naming is uncertain with container conventions
* StrExt::char_at - naming is uncertain with container conventions
* StrExt::char_at_reverse - naming is uncertain with container conventions
* StrVector::concat - this functionality may be replaced with iterators, but
it's not certain at this time
* StrVector::connect - as with concat, may be deprecated in favor of iterators
Deprecated
* StrAllocating and UnicodeStrPrelude have been merged into StrExit
* eq_slice - compiler implementation detail
* from_str - use the inherent parse() method
* is_utf8 - call from_utf8 instead
* replace - call the method instead
* truncate_utf16_at_nul - this is an implementation detail of windows and does
not need to be exposed.
* utf8_char_width - moved to libunicode
* utf16_items - moved to libunicode
* is_utf16 - moved to libunicode
* Utf16Items - moved to libunicode
* Utf16Item - moved to libunicode
* Utf16Encoder - moved to libunicode
* AnyLines - renamed to LinesAny and made a struct
* SendStr - use CowString<'static> instead
* str::raw - all functionality is deprecated
* StrExt::into_string - call to_string() instead
* StrExt::repeat - use iterators instead
* StrExt::char_len - use .chars().count() instead
* StrExt::is_alphanumeric - use .chars().all(..)
* StrExt::is_whitespace - use .chars().all(..)
Pending deprecation -- while slicing syntax is being worked out, these methods
are all #[unstable]
* Str - while currently used for generic programming, this trait will be
replaced with one of [], deref coercions, or a generic conversion trait.
* StrExt::slice - use slicing syntax instead
* StrExt::slice_to - use slicing syntax instead
* StrExt::slice_from - use slicing syntax instead
* StrExt::lev_distance - deprecated with no replacement
Awaiting stabilization due to patterns and/or matching
* StrExt::contains
* StrExt::contains_char
* StrExt::split
* StrExt::splitn
* StrExt::split_terminator
* StrExt::rsplitn
* StrExt::match_indices
* StrExt::split_str
* StrExt::starts_with
* StrExt::ends_with
* StrExt::trim_chars
* StrExt::trim_left_chars
* StrExt::trim_right_chars
* StrExt::find
* StrExt::rfind
* StrExt::find_str
* StrExt::subslice_offset
2014-12-10 17:02:31 +00:00
|
|
|
|
}
|
|
|
|
|
|
2019-12-22 22:42:04 +00:00
|
|
|
|
derive_pattern_clone! {
|
2015-04-05 16:52:14 +00:00
|
|
|
|
clone SplitInternal
|
|
|
|
|
with |s| SplitInternal { matcher: s.matcher.clone(), ..*s }
|
|
|
|
|
}
|
2016-03-05 02:49:43 +00:00
|
|
|
|
|
2015-03-15 00:48:34 +00:00
|
|
|
|
struct SplitInternal<'a, P: Pattern<'a>> {
|
|
|
|
|
start: usize,
|
|
|
|
|
end: usize,
|
|
|
|
|
matcher: P::Searcher,
|
|
|
|
|
allow_trailing_empty: bool,
|
|
|
|
|
finished: bool,
|
std: Stabilize the std::str module
This commit starts out by consolidating all `str` extension traits into one
`StrExt` trait to be included in the prelude. This means that
`UnicodeStrPrelude`, `StrPrelude`, and `StrAllocating` have all been merged into
one `StrExt` exported by the standard library. Some functionality is currently
duplicated with the `StrExt` present in libcore.
This commit also currently avoids any methods which require any form of pattern
to operate. These functions will be stabilized via a separate RFC.
Next, stability of methods and structures are as follows:
Stable
* from_utf8_unchecked
* CowString - after moving to std::string
* StrExt::as_bytes
* StrExt::as_ptr
* StrExt::bytes/Bytes - also made a struct instead of a typedef
* StrExt::char_indices/CharIndices - CharOffsets was renamed
* StrExt::chars/Chars
* StrExt::is_empty
* StrExt::len
* StrExt::lines/Lines
* StrExt::lines_any/LinesAny
* StrExt::slice_unchecked
* StrExt::trim
* StrExt::trim_left
* StrExt::trim_right
* StrExt::words/Words - also made a struct instead of a typedef
Unstable
* from_utf8 - the error type was changed to a `Result`, but the error type has
yet to prove itself
* from_c_str - this function will be handled by the c_str RFC
* FromStr - this trait will have an associated error type eventually
* StrExt::escape_default - needs iterators at least, unsure if it should make
the cut
* StrExt::escape_unicode - needs iterators at least, unsure if it should make
the cut
* StrExt::slice_chars - this function has yet to prove itself
* StrExt::slice_shift_char - awaiting conventions about slicing and shifting
* StrExt::graphemes/Graphemes - this functionality may only be in libunicode
* StrExt::grapheme_indices/GraphemeIndices - this functionality may only be in
libunicode
* StrExt::width - this functionality may only be in libunicode
* StrExt::utf16_units - this functionality may only be in libunicode
* StrExt::nfd_chars - this functionality may only be in libunicode
* StrExt::nfkd_chars - this functionality may only be in libunicode
* StrExt::nfc_chars - this functionality may only be in libunicode
* StrExt::nfkc_chars - this functionality may only be in libunicode
* StrExt::is_char_boundary - naming is uncertain with container conventions
* StrExt::char_range_at - naming is uncertain with container conventions
* StrExt::char_range_at_reverse - naming is uncertain with container conventions
* StrExt::char_at - naming is uncertain with container conventions
* StrExt::char_at_reverse - naming is uncertain with container conventions
* StrVector::concat - this functionality may be replaced with iterators, but
it's not certain at this time
* StrVector::connect - as with concat, may be deprecated in favor of iterators
Deprecated
* StrAllocating and UnicodeStrPrelude have been merged into StrExit
* eq_slice - compiler implementation detail
* from_str - use the inherent parse() method
* is_utf8 - call from_utf8 instead
* replace - call the method instead
* truncate_utf16_at_nul - this is an implementation detail of windows and does
not need to be exposed.
* utf8_char_width - moved to libunicode
* utf16_items - moved to libunicode
* is_utf16 - moved to libunicode
* Utf16Items - moved to libunicode
* Utf16Item - moved to libunicode
* Utf16Encoder - moved to libunicode
* AnyLines - renamed to LinesAny and made a struct
* SendStr - use CowString<'static> instead
* str::raw - all functionality is deprecated
* StrExt::into_string - call to_string() instead
* StrExt::repeat - use iterators instead
* StrExt::char_len - use .chars().count() instead
* StrExt::is_alphanumeric - use .chars().all(..)
* StrExt::is_whitespace - use .chars().all(..)
Pending deprecation -- while slicing syntax is being worked out, these methods
are all #[unstable]
* Str - while currently used for generic programming, this trait will be
replaced with one of [], deref coercions, or a generic conversion trait.
* StrExt::slice - use slicing syntax instead
* StrExt::slice_to - use slicing syntax instead
* StrExt::slice_from - use slicing syntax instead
* StrExt::lev_distance - deprecated with no replacement
Awaiting stabilization due to patterns and/or matching
* StrExt::contains
* StrExt::contains_char
* StrExt::split
* StrExt::splitn
* StrExt::split_terminator
* StrExt::rsplitn
* StrExt::match_indices
* StrExt::split_str
* StrExt::starts_with
* StrExt::ends_with
* StrExt::trim_chars
* StrExt::trim_left_chars
* StrExt::trim_right_chars
* StrExt::find
* StrExt::rfind
* StrExt::find_str
* StrExt::subslice_offset
2014-12-10 17:02:31 +00:00
|
|
|
|
}
|
2014-05-01 06:06:36 +00:00
|
|
|
|
|
2019-07-31 19:00:35 +00:00
|
|
|
|
impl<'a, P> fmt::Debug for SplitInternal<'a, P>
|
|
|
|
|
where
|
|
|
|
|
P: Pattern<'a, Searcher: fmt::Debug>,
|
|
|
|
|
{
|
2019-04-18 23:37:12 +00:00
|
|
|
|
fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
|
2016-03-05 02:49:43 +00:00
|
|
|
|
f.debug_struct("SplitInternal")
|
|
|
|
|
.field("start", &self.start)
|
|
|
|
|
.field("end", &self.end)
|
|
|
|
|
.field("matcher", &self.matcher)
|
|
|
|
|
.field("allow_trailing_empty", &self.allow_trailing_empty)
|
|
|
|
|
.field("finished", &self.finished)
|
|
|
|
|
.finish()
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
2015-03-15 00:48:34 +00:00
|
|
|
|
impl<'a, P: Pattern<'a>> SplitInternal<'a, P> {
|
2014-05-01 06:06:36 +00:00
|
|
|
|
#[inline]
|
|
|
|
|
fn get_end(&mut self) -> Option<&'a str> {
|
2015-01-27 13:09:18 +00:00
|
|
|
|
if !self.finished && (self.allow_trailing_empty || self.end - self.start > 0) {
|
2014-05-01 06:06:36 +00:00
|
|
|
|
self.finished = true;
|
2019-11-20 02:28:32 +00:00
|
|
|
|
// SAFETY: `self.start` and `self.end` always lie on unicode boundaries
|
2015-01-27 13:09:18 +00:00
|
|
|
|
unsafe {
|
2018-06-26 10:34:42 +00:00
|
|
|
|
let string = self.matcher.haystack().get_unchecked(self.start..self.end);
|
2015-01-27 13:09:18 +00:00
|
|
|
|
Some(string)
|
|
|
|
|
}
|
2014-05-01 06:06:36 +00:00
|
|
|
|
} else {
|
|
|
|
|
None
|
|
|
|
|
}
|
|
|
|
|
}
|
2014-12-29 21:18:41 +00:00
|
|
|
|
|
2014-05-01 06:06:36 +00:00
|
|
|
|
#[inline]
|
|
|
|
|
fn next(&mut self) -> Option<&'a str> {
|
2019-12-22 22:42:04 +00:00
|
|
|
|
if self.finished {
|
|
|
|
|
return None;
|
|
|
|
|
}
|
2014-05-01 06:06:36 +00:00
|
|
|
|
|
2015-01-27 13:09:18 +00:00
|
|
|
|
let haystack = self.matcher.haystack();
|
|
|
|
|
match self.matcher.next_match() {
|
2019-11-20 02:28:32 +00:00
|
|
|
|
// SAFETY: `Searcher` guarantees that `a` and `b` lie on unicode boundaries
|
2014-05-01 06:06:36 +00:00
|
|
|
|
Some((a, b)) => unsafe {
|
2018-06-26 10:34:42 +00:00
|
|
|
|
let elt = haystack.get_unchecked(self.start..a);
|
2015-01-27 13:09:18 +00:00
|
|
|
|
self.start = b;
|
2014-05-01 06:06:36 +00:00
|
|
|
|
Some(elt)
|
|
|
|
|
},
|
|
|
|
|
None => self.get_end(),
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
#[inline]
|
2015-03-15 00:48:34 +00:00
|
|
|
|
fn next_back(&mut self) -> Option<&'a str>
|
2019-12-22 22:42:04 +00:00
|
|
|
|
where
|
|
|
|
|
P::Searcher: ReverseSearcher<'a>,
|
2015-03-15 00:48:34 +00:00
|
|
|
|
{
|
2019-12-22 22:42:04 +00:00
|
|
|
|
if self.finished {
|
|
|
|
|
return None;
|
|
|
|
|
}
|
2014-05-01 06:06:36 +00:00
|
|
|
|
|
|
|
|
|
if !self.allow_trailing_empty {
|
|
|
|
|
self.allow_trailing_empty = true;
|
|
|
|
|
match self.next_back() {
|
|
|
|
|
Some(elt) if !elt.is_empty() => return Some(elt),
|
2019-12-22 22:42:04 +00:00
|
|
|
|
_ => {
|
|
|
|
|
if self.finished {
|
|
|
|
|
return None;
|
|
|
|
|
}
|
|
|
|
|
}
|
2014-05-01 06:06:36 +00:00
|
|
|
|
}
|
|
|
|
|
}
|
2015-01-27 13:09:18 +00:00
|
|
|
|
|
|
|
|
|
let haystack = self.matcher.haystack();
|
|
|
|
|
match self.matcher.next_match_back() {
|
2019-11-20 02:28:32 +00:00
|
|
|
|
// SAFETY: `Searcher` guarantees that `a` and `b` lie on unicode boundaries
|
2014-05-01 06:06:36 +00:00
|
|
|
|
Some((a, b)) => unsafe {
|
2018-06-26 10:34:42 +00:00
|
|
|
|
let elt = haystack.get_unchecked(b..self.end);
|
2015-01-27 13:09:18 +00:00
|
|
|
|
self.end = a;
|
2014-05-01 06:06:36 +00:00
|
|
|
|
Some(elt)
|
|
|
|
|
},
|
2019-11-20 02:28:32 +00:00
|
|
|
|
// SAFETY: `self.start` and `self.end` always lie on unicode boundaries
|
2015-01-27 13:09:18 +00:00
|
|
|
|
None => unsafe {
|
|
|
|
|
self.finished = true;
|
2018-06-26 10:34:42 +00:00
|
|
|
|
Some(haystack.get_unchecked(self.start..self.end))
|
2015-01-27 13:09:18 +00:00
|
|
|
|
},
|
2014-05-01 06:06:36 +00:00
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
2015-03-15 00:48:34 +00:00
|
|
|
|
generate_pattern_iterators! {
|
|
|
|
|
forward:
|
2017-03-12 18:04:52 +00:00
|
|
|
|
/// Created with the method [`split`].
|
2015-12-11 13:25:50 +00:00
|
|
|
|
///
|
2017-03-12 18:04:52 +00:00
|
|
|
|
/// [`split`]: ../../std/primitive.str.html#method.split
|
2015-03-15 00:48:34 +00:00
|
|
|
|
struct Split;
|
|
|
|
|
reverse:
|
2017-03-12 18:04:52 +00:00
|
|
|
|
/// Created with the method [`rsplit`].
|
2015-12-11 13:25:50 +00:00
|
|
|
|
///
|
2017-03-12 18:04:52 +00:00
|
|
|
|
/// [`rsplit`]: ../../std/primitive.str.html#method.rsplit
|
2015-03-15 00:48:34 +00:00
|
|
|
|
struct RSplit;
|
|
|
|
|
stability:
|
|
|
|
|
#[stable(feature = "rust1", since = "1.0.0")]
|
|
|
|
|
internal:
|
|
|
|
|
SplitInternal yielding (&'a str);
|
|
|
|
|
delegate double ended;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
generate_pattern_iterators! {
|
|
|
|
|
forward:
|
2017-03-12 18:04:52 +00:00
|
|
|
|
/// Created with the method [`split_terminator`].
|
2015-12-11 13:25:50 +00:00
|
|
|
|
///
|
2017-03-12 18:04:52 +00:00
|
|
|
|
/// [`split_terminator`]: ../../std/primitive.str.html#method.split_terminator
|
2015-03-15 00:48:34 +00:00
|
|
|
|
struct SplitTerminator;
|
|
|
|
|
reverse:
|
2017-03-12 18:04:52 +00:00
|
|
|
|
/// Created with the method [`rsplit_terminator`].
|
2015-12-11 13:25:50 +00:00
|
|
|
|
///
|
2017-03-12 18:04:52 +00:00
|
|
|
|
/// [`rsplit_terminator`]: ../../std/primitive.str.html#method.rsplit_terminator
|
2015-03-15 00:48:34 +00:00
|
|
|
|
struct RSplitTerminator;
|
|
|
|
|
stability:
|
|
|
|
|
#[stable(feature = "rust1", since = "1.0.0")]
|
|
|
|
|
internal:
|
|
|
|
|
SplitInternal yielding (&'a str);
|
|
|
|
|
delegate double ended;
|
|
|
|
|
}
|
2014-12-29 21:18:41 +00:00
|
|
|
|
|
2019-12-22 22:42:04 +00:00
|
|
|
|
derive_pattern_clone! {
|
2015-04-05 16:52:14 +00:00
|
|
|
|
clone SplitNInternal
|
|
|
|
|
with |s| SplitNInternal { iter: s.iter.clone(), ..*s }
|
|
|
|
|
}
|
2016-03-05 02:49:43 +00:00
|
|
|
|
|
2015-03-15 00:48:34 +00:00
|
|
|
|
struct SplitNInternal<'a, P: Pattern<'a>> {
|
|
|
|
|
iter: SplitInternal<'a, P>,
|
|
|
|
|
/// The number of splits remaining
|
|
|
|
|
count: usize,
|
|
|
|
|
}
|
2014-12-29 21:18:41 +00:00
|
|
|
|
|
2019-07-31 19:00:35 +00:00
|
|
|
|
impl<'a, P> fmt::Debug for SplitNInternal<'a, P>
|
|
|
|
|
where
|
|
|
|
|
P: Pattern<'a, Searcher: fmt::Debug>,
|
|
|
|
|
{
|
2019-04-18 23:37:12 +00:00
|
|
|
|
fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
|
2016-03-05 02:49:43 +00:00
|
|
|
|
f.debug_struct("SplitNInternal")
|
|
|
|
|
.field("iter", &self.iter)
|
|
|
|
|
.field("count", &self.count)
|
|
|
|
|
.finish()
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
2015-03-15 00:48:34 +00:00
|
|
|
|
impl<'a, P: Pattern<'a>> SplitNInternal<'a, P> {
|
2014-05-01 06:06:36 +00:00
|
|
|
|
#[inline]
|
|
|
|
|
fn next(&mut self) -> Option<&'a str> {
|
2015-04-01 18:28:34 +00:00
|
|
|
|
match self.count {
|
|
|
|
|
0 => None,
|
2019-12-22 22:42:04 +00:00
|
|
|
|
1 => {
|
|
|
|
|
self.count = 0;
|
|
|
|
|
self.iter.get_end()
|
|
|
|
|
}
|
|
|
|
|
_ => {
|
|
|
|
|
self.count -= 1;
|
|
|
|
|
self.iter.next()
|
|
|
|
|
}
|
2014-05-01 06:06:36 +00:00
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
2015-03-14 23:34:21 +00:00
|
|
|
|
#[inline]
|
2015-03-15 00:48:34 +00:00
|
|
|
|
fn next_back(&mut self) -> Option<&'a str>
|
2019-12-22 22:42:04 +00:00
|
|
|
|
where
|
|
|
|
|
P::Searcher: ReverseSearcher<'a>,
|
2015-03-15 00:48:34 +00:00
|
|
|
|
{
|
|
|
|
|
match self.count {
|
|
|
|
|
0 => None,
|
2019-12-22 22:42:04 +00:00
|
|
|
|
1 => {
|
|
|
|
|
self.count = 0;
|
|
|
|
|
self.iter.get_end()
|
|
|
|
|
}
|
|
|
|
|
_ => {
|
|
|
|
|
self.count -= 1;
|
|
|
|
|
self.iter.next_back()
|
|
|
|
|
}
|
2015-03-14 23:34:21 +00:00
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
2015-03-15 00:48:34 +00:00
|
|
|
|
generate_pattern_iterators! {
|
|
|
|
|
forward:
|
2017-03-12 18:04:52 +00:00
|
|
|
|
/// Created with the method [`splitn`].
|
2015-12-11 13:25:50 +00:00
|
|
|
|
///
|
2017-03-12 18:04:52 +00:00
|
|
|
|
/// [`splitn`]: ../../std/primitive.str.html#method.splitn
|
2015-03-15 00:48:34 +00:00
|
|
|
|
struct SplitN;
|
|
|
|
|
reverse:
|
2017-03-12 18:04:52 +00:00
|
|
|
|
/// Created with the method [`rsplitn`].
|
2015-12-11 13:25:50 +00:00
|
|
|
|
///
|
2017-03-12 18:04:52 +00:00
|
|
|
|
/// [`rsplitn`]: ../../std/primitive.str.html#method.rsplitn
|
2015-03-15 00:48:34 +00:00
|
|
|
|
struct RSplitN;
|
|
|
|
|
stability:
|
|
|
|
|
#[stable(feature = "rust1", since = "1.0.0")]
|
|
|
|
|
internal:
|
|
|
|
|
SplitNInternal yielding (&'a str);
|
|
|
|
|
delegate single ended;
|
|
|
|
|
}
|
|
|
|
|
|
2019-12-22 22:42:04 +00:00
|
|
|
|
derive_pattern_clone! {
|
2015-04-05 16:52:14 +00:00
|
|
|
|
clone MatchIndicesInternal
|
|
|
|
|
with |s| MatchIndicesInternal(s.0.clone())
|
|
|
|
|
}
|
2016-03-05 02:49:43 +00:00
|
|
|
|
|
2015-03-15 00:48:34 +00:00
|
|
|
|
struct MatchIndicesInternal<'a, P: Pattern<'a>>(P::Searcher);
|
|
|
|
|
|
2019-07-31 19:00:35 +00:00
|
|
|
|
impl<'a, P> fmt::Debug for MatchIndicesInternal<'a, P>
|
|
|
|
|
where
|
|
|
|
|
P: Pattern<'a, Searcher: fmt::Debug>,
|
|
|
|
|
{
|
2019-04-18 23:37:12 +00:00
|
|
|
|
fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
|
2019-12-22 22:42:04 +00:00
|
|
|
|
f.debug_tuple("MatchIndicesInternal").field(&self.0).finish()
|
2016-03-05 02:49:43 +00:00
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
2015-03-15 00:48:34 +00:00
|
|
|
|
impl<'a, P: Pattern<'a>> MatchIndicesInternal<'a, P> {
|
|
|
|
|
#[inline]
|
2015-09-24 15:38:48 +00:00
|
|
|
|
fn next(&mut self) -> Option<(usize, &'a str)> {
|
2019-11-20 02:28:32 +00:00
|
|
|
|
// SAFETY: `Searcher` guaratees that `start` and `end` lie on unicode boundaries
|
2019-12-22 22:42:04 +00:00
|
|
|
|
self.0
|
|
|
|
|
.next_match()
|
|
|
|
|
.map(|(start, end)| unsafe { (start, self.0.haystack().get_unchecked(start..end)) })
|
2015-03-15 00:48:34 +00:00
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
#[inline]
|
2015-09-24 15:38:48 +00:00
|
|
|
|
fn next_back(&mut self) -> Option<(usize, &'a str)>
|
2019-12-22 22:42:04 +00:00
|
|
|
|
where
|
|
|
|
|
P::Searcher: ReverseSearcher<'a>,
|
2015-03-15 00:48:34 +00:00
|
|
|
|
{
|
2019-11-20 02:28:32 +00:00
|
|
|
|
// SAFETY: `Searcher` guarantees that `start` and `end` lie on unicode boundaries
|
2019-12-22 22:42:04 +00:00
|
|
|
|
self.0
|
|
|
|
|
.next_match_back()
|
|
|
|
|
.map(|(start, end)| unsafe { (start, self.0.haystack().get_unchecked(start..end)) })
|
2015-03-15 00:48:34 +00:00
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
generate_pattern_iterators! {
|
|
|
|
|
forward:
|
2017-03-12 18:04:52 +00:00
|
|
|
|
/// Created with the method [`match_indices`].
|
2015-12-11 13:25:50 +00:00
|
|
|
|
///
|
2017-03-12 18:04:52 +00:00
|
|
|
|
/// [`match_indices`]: ../../std/primitive.str.html#method.match_indices
|
2015-03-15 00:48:34 +00:00
|
|
|
|
struct MatchIndices;
|
|
|
|
|
reverse:
|
2017-03-12 18:04:52 +00:00
|
|
|
|
/// Created with the method [`rmatch_indices`].
|
2015-12-11 13:25:50 +00:00
|
|
|
|
///
|
2017-03-12 18:04:52 +00:00
|
|
|
|
/// [`rmatch_indices`]: ../../std/primitive.str.html#method.rmatch_indices
|
2015-03-15 00:48:34 +00:00
|
|
|
|
struct RMatchIndices;
|
|
|
|
|
stability:
|
std: Stabilize library APIs for 1.5
This commit stabilizes and deprecates library APIs whose FCP has closed in the
last cycle, specifically:
Stabilized APIs:
* `fs::canonicalize`
* `Path::{metadata, symlink_metadata, canonicalize, read_link, read_dir, exists,
is_file, is_dir}` - all moved to inherent methods from the `PathExt` trait.
* `Formatter::fill`
* `Formatter::width`
* `Formatter::precision`
* `Formatter::sign_plus`
* `Formatter::sign_minus`
* `Formatter::alternate`
* `Formatter::sign_aware_zero_pad`
* `string::ParseError`
* `Utf8Error::valid_up_to`
* `Iterator::{cmp, partial_cmp, eq, ne, lt, le, gt, ge}`
* `<[T]>::split_{first,last}{,_mut}`
* `Condvar::wait_timeout` - note that `wait_timeout_ms` is not yet deprecated
but will be once 1.5 is released.
* `str::{R,}MatchIndices`
* `str::{r,}match_indices`
* `char::from_u32_unchecked`
* `VecDeque::insert`
* `VecDeque::shrink_to_fit`
* `VecDeque::as_slices`
* `VecDeque::as_mut_slices`
* `VecDeque::swap_remove_front` - (renamed from `swap_front_remove`)
* `VecDeque::swap_remove_back` - (renamed from `swap_back_remove`)
* `Vec::resize`
* `str::slice_mut_unchecked`
* `FileTypeExt`
* `FileTypeExt::{is_block_device, is_char_device, is_fifo, is_socket}`
* `BinaryHeap::from` - `from_vec` deprecated in favor of this
* `BinaryHeap::into_vec` - plus a `Into` impl
* `BinaryHeap::into_sorted_vec`
Deprecated APIs
* `slice::ref_slice`
* `slice::mut_ref_slice`
* `iter::{range_inclusive, RangeInclusive}`
* `std::dynamic_lib`
Closes #27706
Closes #27725
cc #27726 (align not stabilized yet)
Closes #27734
Closes #27737
Closes #27742
Closes #27743
Closes #27772
Closes #27774
Closes #27777
Closes #27781
cc #27788 (a few remaining methods though)
Closes #27790
Closes #27793
Closes #27796
Closes #27810
cc #28147 (not all parts stabilized)
2015-10-22 23:28:45 +00:00
|
|
|
|
#[stable(feature = "str_match_indices", since = "1.5.0")]
|
2015-03-15 00:48:34 +00:00
|
|
|
|
internal:
|
2015-09-24 15:38:48 +00:00
|
|
|
|
MatchIndicesInternal yielding ((usize, &'a str));
|
2015-03-15 00:48:34 +00:00
|
|
|
|
delegate double ended;
|
|
|
|
|
}
|
|
|
|
|
|
2019-12-22 22:42:04 +00:00
|
|
|
|
derive_pattern_clone! {
|
2015-04-05 16:52:14 +00:00
|
|
|
|
clone MatchesInternal
|
|
|
|
|
with |s| MatchesInternal(s.0.clone())
|
|
|
|
|
}
|
2016-03-05 02:49:43 +00:00
|
|
|
|
|
2015-03-15 00:48:34 +00:00
|
|
|
|
struct MatchesInternal<'a, P: Pattern<'a>>(P::Searcher);
|
|
|
|
|
|
2019-07-31 19:00:35 +00:00
|
|
|
|
impl<'a, P> fmt::Debug for MatchesInternal<'a, P>
|
|
|
|
|
where
|
|
|
|
|
P: Pattern<'a, Searcher: fmt::Debug>,
|
|
|
|
|
{
|
2019-04-18 23:37:12 +00:00
|
|
|
|
fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
|
2019-12-22 22:42:04 +00:00
|
|
|
|
f.debug_tuple("MatchesInternal").field(&self.0).finish()
|
2016-03-05 02:49:43 +00:00
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
2015-03-15 00:48:34 +00:00
|
|
|
|
impl<'a, P: Pattern<'a>> MatchesInternal<'a, P> {
|
|
|
|
|
#[inline]
|
|
|
|
|
fn next(&mut self) -> Option<&'a str> {
|
2019-11-20 02:28:32 +00:00
|
|
|
|
// SAFETY: `Searcher` guarantees that `start` and `end` lie on unicode boundaries
|
2015-03-15 00:48:34 +00:00
|
|
|
|
self.0.next_match().map(|(a, b)| unsafe {
|
|
|
|
|
// Indices are known to be on utf8 boundaries
|
2018-06-26 10:34:42 +00:00
|
|
|
|
self.0.haystack().get_unchecked(a..b)
|
2015-03-15 00:48:34 +00:00
|
|
|
|
})
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
#[inline]
|
|
|
|
|
fn next_back(&mut self) -> Option<&'a str>
|
2019-12-22 22:42:04 +00:00
|
|
|
|
where
|
|
|
|
|
P::Searcher: ReverseSearcher<'a>,
|
2015-03-15 00:48:34 +00:00
|
|
|
|
{
|
2019-11-20 02:28:32 +00:00
|
|
|
|
// SAFETY: `Searcher` guarantees that `start` and `end` lie on unicode boundaries
|
2015-03-15 00:48:34 +00:00
|
|
|
|
self.0.next_match_back().map(|(a, b)| unsafe {
|
|
|
|
|
// Indices are known to be on utf8 boundaries
|
2018-06-26 10:34:42 +00:00
|
|
|
|
self.0.haystack().get_unchecked(a..b)
|
2015-03-15 00:48:34 +00:00
|
|
|
|
})
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
generate_pattern_iterators! {
|
|
|
|
|
forward:
|
2017-03-12 18:04:52 +00:00
|
|
|
|
/// Created with the method [`matches`].
|
2015-12-11 13:25:50 +00:00
|
|
|
|
///
|
2017-03-12 18:04:52 +00:00
|
|
|
|
/// [`matches`]: ../../std/primitive.str.html#method.matches
|
2015-03-15 00:48:34 +00:00
|
|
|
|
struct Matches;
|
|
|
|
|
reverse:
|
2017-03-12 18:04:52 +00:00
|
|
|
|
/// Created with the method [`rmatches`].
|
2015-12-11 13:25:50 +00:00
|
|
|
|
///
|
2017-03-12 18:04:52 +00:00
|
|
|
|
/// [`rmatches`]: ../../std/primitive.str.html#method.rmatches
|
2015-03-15 00:48:34 +00:00
|
|
|
|
struct RMatches;
|
|
|
|
|
stability:
|
2015-06-11 01:49:26 +00:00
|
|
|
|
#[stable(feature = "str_matches", since = "1.2.0")]
|
2015-03-15 00:48:34 +00:00
|
|
|
|
internal:
|
|
|
|
|
MatchesInternal yielding (&'a str);
|
|
|
|
|
delegate double ended;
|
|
|
|
|
}
|
|
|
|
|
|
2017-03-30 19:36:50 +00:00
|
|
|
|
/// An iterator over the lines of a string, as string slices.
|
2015-12-11 13:25:50 +00:00
|
|
|
|
///
|
2017-03-30 22:33:23 +00:00
|
|
|
|
/// This struct is created with the [`lines`] method on [`str`].
|
2017-03-30 19:36:50 +00:00
|
|
|
|
/// See its documentation for more.
|
2015-12-11 13:25:50 +00:00
|
|
|
|
///
|
2017-03-12 18:04:52 +00:00
|
|
|
|
/// [`lines`]: ../../std/primitive.str.html#method.lines
|
2017-03-30 19:36:50 +00:00
|
|
|
|
/// [`str`]: ../../std/primitive.str.html
|
2015-03-15 00:48:34 +00:00
|
|
|
|
#[stable(feature = "rust1", since = "1.0.0")]
|
2016-03-05 02:49:43 +00:00
|
|
|
|
#[derive(Clone, Debug)]
|
2015-08-27 00:30:45 +00:00
|
|
|
|
pub struct Lines<'a>(Map<SplitTerminator<'a, char>, LinesAnyMap>);
|
2015-03-15 00:48:34 +00:00
|
|
|
|
|
2015-03-14 23:34:21 +00:00
|
|
|
|
#[stable(feature = "rust1", since = "1.0.0")]
|
2015-03-15 00:48:34 +00:00
|
|
|
|
impl<'a> Iterator for Lines<'a> {
|
2015-03-14 23:34:21 +00:00
|
|
|
|
type Item = &'a str;
|
|
|
|
|
|
|
|
|
|
#[inline]
|
|
|
|
|
fn next(&mut self) -> Option<&'a str> {
|
2015-03-15 00:48:34 +00:00
|
|
|
|
self.0.next()
|
|
|
|
|
}
|
2015-03-14 23:34:21 +00:00
|
|
|
|
|
2015-03-15 00:48:34 +00:00
|
|
|
|
#[inline]
|
|
|
|
|
fn size_hint(&self) -> (usize, Option<usize>) {
|
|
|
|
|
self.0.size_hint()
|
|
|
|
|
}
|
When possible without changing semantics, implement Iterator::last in terms of DoubleEndedIterator::next_back for types in liballoc and libcore.
Provided that the iterator has finite length and does not trigger user-provided code, this is safe.
What follows is a full list of the DoubleEndedIterators in liballoc/libcore and whether this optimization is safe, and if not, why not.
src/liballoc/boxed.rs
Box: Pass through to avoid defeating optimization of the underlying DoubleIterator implementation. This has no correctness impact.
src/liballoc/collections/binary_heap.rs
Iter: Pass through to avoid defeating optimizations on slice::Iter
IntoIter: Not safe, changes Drop order
Drain: Not safe, changes Drop order
src/liballoc/collections/btree/map.rs
Iter: Safe to call next_back, invokes no user defined code.
IterMut: ditto
IntoIter: Not safe, changes Drop order
Keys: Safe to call next_back, invokes no user defined code.
Values: ditto
ValuesMut: ditto
Range: ditto
RangeMut: ditto
src/liballoc/collections/btree/set.rs
Iter: Safe to call next_back, invokes no user defined code.
IntoIter: Not safe, changes Drop order
Range: Safe to call next_back, invokes no user defined code.
src/liballoc/collections/linked_list.rs
Iter: Safe to call next_back, invokes no user defined code.
IterMut: ditto
IntoIter: Not safe, changes Drop order
src/liballoc/collections/vec_deque.rs
Iter: Safe to call next_back, invokes no user defined code.
IterMut: ditto
IntoIter: Not safe, changes Drop order
Drain: ditto
src/liballoc/string.rs
Drain: Safe because return type is a primitive (char)
src/liballoc/vec.rs
IntoIter: Not safe, changes Drop order
Drain: ditto
Splice: ditto
src/libcore/ascii.rs
EscapeDefault: Safe because return type is a primitive (u8)
src/libcore/iter/adapters/chain.rs
Chain: Not safe, invokes user defined code (Iterator impl)
src/libcore/iter/adapters/flatten.rs
FlatMap: Not safe, invokes user defined code (Iterator impl)
Flatten: ditto
FlattenCompat: ditto
src/libcore/iter/adapters/mod.rs
Rev: Not safe, invokes user defined code (Iterator impl)
Copied: ditto
Cloned: Not safe, invokes user defined code (Iterator impl and T::clone)
Map: Not safe, invokes user defined code (Iterator impl + closure)
Filter: ditto
FilterMap: ditto
Enumerate: Not safe, invokes user defined code (Iterator impl)
Skip: ditto
Fuse: ditto
Inspect: ditto
src/libcore/iter/adapters/zip.rs
Zip: Not safe, invokes user defined code (Iterator impl)
src/libcore/iter/range.rs
ops::Range: Not safe, changes Drop order, but ALREADY HAS SPECIALIZATION
ops::RangeInclusive: ditto
src/libcore/iter/sources.rs
Repeat: Not safe, calling last should iloop.
Empty: No point, iterator is at most one item long.
Once: ditto
OnceWith: ditto
src/libcore/option.rs
Item: No point, iterator is at most one item long.
Iter: ditto
IterMut: ditto
IntoIter: ditto
src/libcore/result.rs
Iter: No point, iterator is at most one item long
IterMut: ditto
IntoIter: ditto
src/libcore/slice/mod.rs
Split: Not safe, invokes user defined closure
SplitMut: ditto
RSplit: ditto
RSplitMut: ditto
Windows: Safe, already has specialization
Chunks: ditto
ChunksMut: ditto
ChunksExact: ditto
ChunksExactMut: ditto
RChunks: ditto
RChunksMut: ditto
RChunksExact: ditto
RChunksExactMut: ditto
src/libcore/str/mod.rs
Chars: Safe, already has specialization
CharIndices: ditto
Bytes: ditto
Lines: Safe to call next_back, invokes no user defined code.
LinesAny: Deprecated
Everything that is generic over P: Pattern: Not safe because Pattern invokes user defined code.
SplitWhitespace: Safe to call next_back, invokes no user defined code.
SplitAsciiWhitespace: ditto
2019-07-02 20:45:29 +00:00
|
|
|
|
|
|
|
|
|
#[inline]
|
|
|
|
|
fn last(mut self) -> Option<&'a str> {
|
|
|
|
|
self.next_back()
|
|
|
|
|
}
|
2015-03-15 00:48:34 +00:00
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
#[stable(feature = "rust1", since = "1.0.0")]
|
|
|
|
|
impl<'a> DoubleEndedIterator for Lines<'a> {
|
|
|
|
|
#[inline]
|
|
|
|
|
fn next_back(&mut self) -> Option<&'a str> {
|
|
|
|
|
self.0.next_back()
|
2015-03-14 23:34:21 +00:00
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
2018-03-03 13:15:28 +00:00
|
|
|
|
#[stable(feature = "fused", since = "1.26.0")]
|
2018-09-03 11:50:14 +00:00
|
|
|
|
impl FusedIterator for Lines<'_> {}
|
2016-08-13 18:42:36 +00:00
|
|
|
|
|
2017-03-12 18:04:52 +00:00
|
|
|
|
/// Created with the method [`lines_any`].
|
2015-12-11 13:25:50 +00:00
|
|
|
|
///
|
2017-03-12 18:04:52 +00:00
|
|
|
|
/// [`lines_any`]: ../../std/primitive.str.html#method.lines_any
|
2015-03-15 00:48:34 +00:00
|
|
|
|
#[stable(feature = "rust1", since = "1.0.0")]
|
2015-11-20 13:11:20 +00:00
|
|
|
|
#[rustc_deprecated(since = "1.4.0", reason = "use lines()/Lines instead now")]
|
2016-03-05 02:49:43 +00:00
|
|
|
|
#[derive(Clone, Debug)]
|
2015-08-27 00:30:45 +00:00
|
|
|
|
#[allow(deprecated)]
|
|
|
|
|
pub struct LinesAny<'a>(Lines<'a>);
|
2015-04-05 16:52:14 +00:00
|
|
|
|
|
2019-02-01 11:43:29 +00:00
|
|
|
|
impl_fn_for_zst! {
|
|
|
|
|
/// A nameable, cloneable fn type
|
|
|
|
|
#[derive(Clone)]
|
|
|
|
|
struct LinesAnyMap impl<'a> Fn = |line: &'a str| -> &'a str {
|
2015-04-05 16:52:14 +00:00
|
|
|
|
let l = line.len();
|
|
|
|
|
if l > 0 && line.as_bytes()[l - 1] == b'\r' { &line[0 .. l - 1] }
|
|
|
|
|
else { line }
|
2019-02-01 11:43:29 +00:00
|
|
|
|
};
|
2015-03-14 23:34:21 +00:00
|
|
|
|
}
|
|
|
|
|
|
2015-03-15 00:07:13 +00:00
|
|
|
|
#[stable(feature = "rust1", since = "1.0.0")]
|
2015-08-27 00:30:45 +00:00
|
|
|
|
#[allow(deprecated)]
|
2015-03-15 00:48:34 +00:00
|
|
|
|
impl<'a> Iterator for LinesAny<'a> {
|
2015-03-15 00:07:13 +00:00
|
|
|
|
type Item = &'a str;
|
|
|
|
|
|
|
|
|
|
#[inline]
|
|
|
|
|
fn next(&mut self) -> Option<&'a str> {
|
2015-03-15 00:48:34 +00:00
|
|
|
|
self.0.next()
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
#[inline]
|
|
|
|
|
fn size_hint(&self) -> (usize, Option<usize>) {
|
|
|
|
|
self.0.size_hint()
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
#[stable(feature = "rust1", since = "1.0.0")]
|
2015-08-27 00:30:45 +00:00
|
|
|
|
#[allow(deprecated)]
|
2015-03-15 00:48:34 +00:00
|
|
|
|
impl<'a> DoubleEndedIterator for LinesAny<'a> {
|
|
|
|
|
#[inline]
|
|
|
|
|
fn next_back(&mut self) -> Option<&'a str> {
|
|
|
|
|
self.0.next_back()
|
2015-03-15 00:07:13 +00:00
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
2018-03-03 13:15:28 +00:00
|
|
|
|
#[stable(feature = "fused", since = "1.26.0")]
|
2016-08-13 18:42:36 +00:00
|
|
|
|
#[allow(deprecated)]
|
2018-09-03 11:50:14 +00:00
|
|
|
|
impl FusedIterator for LinesAny<'_> {}
|
2016-08-13 18:42:36 +00:00
|
|
|
|
|
2014-05-01 06:06:36 +00:00
|
|
|
|
/*
|
Add fast path for ASCII in UTF-8 validation
This speeds up the ascii case (and long stretches of ascii in otherwise
mixed UTF-8 data) when checking UTF-8 validity.
Benchmark results suggest that on purely ASCII input, we can improve
throughput (megabytes verified / second) by a factor of 13 to 14!
On xml and mostly english language input (en.wikipedia xml dump),
throughput increases by a factor 7.
On mostly non-ASCII input, performance increases slightly or is the
same.
The UTF-8 validation is rewritten to use indexed access; since all
access is preceded by a (mandatory for validation) length check, they
are statically elided by llvm and this formulation is in fact the best
for performance. A previous version had losses due to slice to iterator
conversions.
A large credit to Björn Steinbrink who improved this patch immensely,
writing this second version.
Benchmark results on x86-64 (Sandy Bridge) compiled with -C opt-level=3.
Old code is `regular`, this PR is called `fast`.
Datasets:
- `ascii` is just ascii (2.5 kB)
- `cyr` is cyrillic script with ascii spaces (5 kB)
- `dewik10` is 10MB of a de.wikipedia xml dump
- `enwik10` is 100MB of an en.wikipedia xml dump
- `jawik10` is 10MB of a ja.wikipedia xml dump
```
test from_utf8_ascii_fast ... bench: 140 ns/iter (+/- 4) = 18221 MB/s
test from_utf8_ascii_regular ... bench: 1,932 ns/iter (+/- 19) = 1320 MB/s
test from_utf8_cyr_fast ... bench: 10,025 ns/iter (+/- 245) = 511 MB/s
test from_utf8_cyr_regular ... bench: 12,250 ns/iter (+/- 437) = 418 MB/s
test from_utf8_dewik10_fast ... bench: 6,017,909 ns/iter (+/- 105,755) = 1740 MB/s
test from_utf8_dewik10_regular ... bench: 11,669,493 ns/iter (+/- 264,045) = 891 MB/s
test from_utf8_enwik8_fast ... bench: 14,085,692 ns/iter (+/- 1,643,316) = 7000 MB/s
test from_utf8_enwik8_regular ... bench: 93,657,410 ns/iter (+/- 5,353,353) = 1000 MB/s
test from_utf8_jawik10_fast ... bench: 29,154,073 ns/iter (+/- 4,659,534) = 340 MB/s
test from_utf8_jawik10_regular ... bench: 29,112,917 ns/iter (+/- 2,475,123) = 340 MB/s
```
Co-authored-by: Björn Steinbrink <bsteinbr@gmail.com>
2016-01-06 14:43:33 +00:00
|
|
|
|
Section: UTF-8 validation
|
2014-05-01 06:06:36 +00:00
|
|
|
|
*/
|
|
|
|
|
|
Add fast path for ASCII in UTF-8 validation
This speeds up the ascii case (and long stretches of ascii in otherwise
mixed UTF-8 data) when checking UTF-8 validity.
Benchmark results suggest that on purely ASCII input, we can improve
throughput (megabytes verified / second) by a factor of 13 to 14!
On xml and mostly english language input (en.wikipedia xml dump),
throughput increases by a factor 7.
On mostly non-ASCII input, performance increases slightly or is the
same.
The UTF-8 validation is rewritten to use indexed access; since all
access is preceded by a (mandatory for validation) length check, they
are statically elided by llvm and this formulation is in fact the best
for performance. A previous version had losses due to slice to iterator
conversions.
A large credit to Björn Steinbrink who improved this patch immensely,
writing this second version.
Benchmark results on x86-64 (Sandy Bridge) compiled with -C opt-level=3.
Old code is `regular`, this PR is called `fast`.
Datasets:
- `ascii` is just ascii (2.5 kB)
- `cyr` is cyrillic script with ascii spaces (5 kB)
- `dewik10` is 10MB of a de.wikipedia xml dump
- `enwik10` is 100MB of an en.wikipedia xml dump
- `jawik10` is 10MB of a ja.wikipedia xml dump
```
test from_utf8_ascii_fast ... bench: 140 ns/iter (+/- 4) = 18221 MB/s
test from_utf8_ascii_regular ... bench: 1,932 ns/iter (+/- 19) = 1320 MB/s
test from_utf8_cyr_fast ... bench: 10,025 ns/iter (+/- 245) = 511 MB/s
test from_utf8_cyr_regular ... bench: 12,250 ns/iter (+/- 437) = 418 MB/s
test from_utf8_dewik10_fast ... bench: 6,017,909 ns/iter (+/- 105,755) = 1740 MB/s
test from_utf8_dewik10_regular ... bench: 11,669,493 ns/iter (+/- 264,045) = 891 MB/s
test from_utf8_enwik8_fast ... bench: 14,085,692 ns/iter (+/- 1,643,316) = 7000 MB/s
test from_utf8_enwik8_regular ... bench: 93,657,410 ns/iter (+/- 5,353,353) = 1000 MB/s
test from_utf8_jawik10_fast ... bench: 29,154,073 ns/iter (+/- 4,659,534) = 340 MB/s
test from_utf8_jawik10_regular ... bench: 29,112,917 ns/iter (+/- 2,475,123) = 340 MB/s
```
Co-authored-by: Björn Steinbrink <bsteinbr@gmail.com>
2016-01-06 14:43:33 +00:00
|
|
|
|
// use truncation to fit u64 into usize
|
|
|
|
|
const NONASCII_MASK: usize = 0x80808080_80808080u64 as usize;
|
|
|
|
|
|
2017-03-22 00:42:23 +00:00
|
|
|
|
/// Returns `true` if any byte in the word `x` is nonascii (>= 128).
|
Add fast path for ASCII in UTF-8 validation
This speeds up the ascii case (and long stretches of ascii in otherwise
mixed UTF-8 data) when checking UTF-8 validity.
Benchmark results suggest that on purely ASCII input, we can improve
throughput (megabytes verified / second) by a factor of 13 to 14!
On xml and mostly english language input (en.wikipedia xml dump),
throughput increases by a factor 7.
On mostly non-ASCII input, performance increases slightly or is the
same.
The UTF-8 validation is rewritten to use indexed access; since all
access is preceded by a (mandatory for validation) length check, they
are statically elided by llvm and this formulation is in fact the best
for performance. A previous version had losses due to slice to iterator
conversions.
A large credit to Björn Steinbrink who improved this patch immensely,
writing this second version.
Benchmark results on x86-64 (Sandy Bridge) compiled with -C opt-level=3.
Old code is `regular`, this PR is called `fast`.
Datasets:
- `ascii` is just ascii (2.5 kB)
- `cyr` is cyrillic script with ascii spaces (5 kB)
- `dewik10` is 10MB of a de.wikipedia xml dump
- `enwik10` is 100MB of an en.wikipedia xml dump
- `jawik10` is 10MB of a ja.wikipedia xml dump
```
test from_utf8_ascii_fast ... bench: 140 ns/iter (+/- 4) = 18221 MB/s
test from_utf8_ascii_regular ... bench: 1,932 ns/iter (+/- 19) = 1320 MB/s
test from_utf8_cyr_fast ... bench: 10,025 ns/iter (+/- 245) = 511 MB/s
test from_utf8_cyr_regular ... bench: 12,250 ns/iter (+/- 437) = 418 MB/s
test from_utf8_dewik10_fast ... bench: 6,017,909 ns/iter (+/- 105,755) = 1740 MB/s
test from_utf8_dewik10_regular ... bench: 11,669,493 ns/iter (+/- 264,045) = 891 MB/s
test from_utf8_enwik8_fast ... bench: 14,085,692 ns/iter (+/- 1,643,316) = 7000 MB/s
test from_utf8_enwik8_regular ... bench: 93,657,410 ns/iter (+/- 5,353,353) = 1000 MB/s
test from_utf8_jawik10_fast ... bench: 29,154,073 ns/iter (+/- 4,659,534) = 340 MB/s
test from_utf8_jawik10_regular ... bench: 29,112,917 ns/iter (+/- 2,475,123) = 340 MB/s
```
Co-authored-by: Björn Steinbrink <bsteinbr@gmail.com>
2016-01-06 14:43:33 +00:00
|
|
|
|
#[inline]
|
2018-10-23 21:09:44 +00:00
|
|
|
|
fn contains_nonascii(x: usize) -> bool {
|
Add fast path for ASCII in UTF-8 validation
This speeds up the ascii case (and long stretches of ascii in otherwise
mixed UTF-8 data) when checking UTF-8 validity.
Benchmark results suggest that on purely ASCII input, we can improve
throughput (megabytes verified / second) by a factor of 13 to 14!
On xml and mostly english language input (en.wikipedia xml dump),
throughput increases by a factor 7.
On mostly non-ASCII input, performance increases slightly or is the
same.
The UTF-8 validation is rewritten to use indexed access; since all
access is preceded by a (mandatory for validation) length check, they
are statically elided by llvm and this formulation is in fact the best
for performance. A previous version had losses due to slice to iterator
conversions.
A large credit to Björn Steinbrink who improved this patch immensely,
writing this second version.
Benchmark results on x86-64 (Sandy Bridge) compiled with -C opt-level=3.
Old code is `regular`, this PR is called `fast`.
Datasets:
- `ascii` is just ascii (2.5 kB)
- `cyr` is cyrillic script with ascii spaces (5 kB)
- `dewik10` is 10MB of a de.wikipedia xml dump
- `enwik10` is 100MB of an en.wikipedia xml dump
- `jawik10` is 10MB of a ja.wikipedia xml dump
```
test from_utf8_ascii_fast ... bench: 140 ns/iter (+/- 4) = 18221 MB/s
test from_utf8_ascii_regular ... bench: 1,932 ns/iter (+/- 19) = 1320 MB/s
test from_utf8_cyr_fast ... bench: 10,025 ns/iter (+/- 245) = 511 MB/s
test from_utf8_cyr_regular ... bench: 12,250 ns/iter (+/- 437) = 418 MB/s
test from_utf8_dewik10_fast ... bench: 6,017,909 ns/iter (+/- 105,755) = 1740 MB/s
test from_utf8_dewik10_regular ... bench: 11,669,493 ns/iter (+/- 264,045) = 891 MB/s
test from_utf8_enwik8_fast ... bench: 14,085,692 ns/iter (+/- 1,643,316) = 7000 MB/s
test from_utf8_enwik8_regular ... bench: 93,657,410 ns/iter (+/- 5,353,353) = 1000 MB/s
test from_utf8_jawik10_fast ... bench: 29,154,073 ns/iter (+/- 4,659,534) = 340 MB/s
test from_utf8_jawik10_regular ... bench: 29,112,917 ns/iter (+/- 2,475,123) = 340 MB/s
```
Co-authored-by: Björn Steinbrink <bsteinbr@gmail.com>
2016-01-06 14:43:33 +00:00
|
|
|
|
(x & NONASCII_MASK) != 0
|
|
|
|
|
}
|
|
|
|
|
|
2018-11-22 14:26:16 +00:00
|
|
|
|
/// Walks through `v` checking that it's a valid UTF-8 sequence,
|
|
|
|
|
/// returning `Ok(())` in that case, or, if it is invalid, `Err(err)`.
|
2017-07-20 18:14:13 +00:00
|
|
|
|
#[inline]
|
Add fast path for ASCII in UTF-8 validation
This speeds up the ascii case (and long stretches of ascii in otherwise
mixed UTF-8 data) when checking UTF-8 validity.
Benchmark results suggest that on purely ASCII input, we can improve
throughput (megabytes verified / second) by a factor of 13 to 14!
On xml and mostly english language input (en.wikipedia xml dump),
throughput increases by a factor 7.
On mostly non-ASCII input, performance increases slightly or is the
same.
The UTF-8 validation is rewritten to use indexed access; since all
access is preceded by a (mandatory for validation) length check, they
are statically elided by llvm and this formulation is in fact the best
for performance. A previous version had losses due to slice to iterator
conversions.
A large credit to Björn Steinbrink who improved this patch immensely,
writing this second version.
Benchmark results on x86-64 (Sandy Bridge) compiled with -C opt-level=3.
Old code is `regular`, this PR is called `fast`.
Datasets:
- `ascii` is just ascii (2.5 kB)
- `cyr` is cyrillic script with ascii spaces (5 kB)
- `dewik10` is 10MB of a de.wikipedia xml dump
- `enwik10` is 100MB of an en.wikipedia xml dump
- `jawik10` is 10MB of a ja.wikipedia xml dump
```
test from_utf8_ascii_fast ... bench: 140 ns/iter (+/- 4) = 18221 MB/s
test from_utf8_ascii_regular ... bench: 1,932 ns/iter (+/- 19) = 1320 MB/s
test from_utf8_cyr_fast ... bench: 10,025 ns/iter (+/- 245) = 511 MB/s
test from_utf8_cyr_regular ... bench: 12,250 ns/iter (+/- 437) = 418 MB/s
test from_utf8_dewik10_fast ... bench: 6,017,909 ns/iter (+/- 105,755) = 1740 MB/s
test from_utf8_dewik10_regular ... bench: 11,669,493 ns/iter (+/- 264,045) = 891 MB/s
test from_utf8_enwik8_fast ... bench: 14,085,692 ns/iter (+/- 1,643,316) = 7000 MB/s
test from_utf8_enwik8_regular ... bench: 93,657,410 ns/iter (+/- 5,353,353) = 1000 MB/s
test from_utf8_jawik10_fast ... bench: 29,154,073 ns/iter (+/- 4,659,534) = 340 MB/s
test from_utf8_jawik10_regular ... bench: 29,112,917 ns/iter (+/- 2,475,123) = 340 MB/s
```
Co-authored-by: Björn Steinbrink <bsteinbr@gmail.com>
2016-01-06 14:43:33 +00:00
|
|
|
|
fn run_utf8_validation(v: &[u8]) -> Result<(), Utf8Error> {
|
2016-11-22 12:47:45 +00:00
|
|
|
|
let mut index = 0;
|
Add fast path for ASCII in UTF-8 validation
This speeds up the ascii case (and long stretches of ascii in otherwise
mixed UTF-8 data) when checking UTF-8 validity.
Benchmark results suggest that on purely ASCII input, we can improve
throughput (megabytes verified / second) by a factor of 13 to 14!
On xml and mostly english language input (en.wikipedia xml dump),
throughput increases by a factor 7.
On mostly non-ASCII input, performance increases slightly or is the
same.
The UTF-8 validation is rewritten to use indexed access; since all
access is preceded by a (mandatory for validation) length check, they
are statically elided by llvm and this formulation is in fact the best
for performance. A previous version had losses due to slice to iterator
conversions.
A large credit to Björn Steinbrink who improved this patch immensely,
writing this second version.
Benchmark results on x86-64 (Sandy Bridge) compiled with -C opt-level=3.
Old code is `regular`, this PR is called `fast`.
Datasets:
- `ascii` is just ascii (2.5 kB)
- `cyr` is cyrillic script with ascii spaces (5 kB)
- `dewik10` is 10MB of a de.wikipedia xml dump
- `enwik10` is 100MB of an en.wikipedia xml dump
- `jawik10` is 10MB of a ja.wikipedia xml dump
```
test from_utf8_ascii_fast ... bench: 140 ns/iter (+/- 4) = 18221 MB/s
test from_utf8_ascii_regular ... bench: 1,932 ns/iter (+/- 19) = 1320 MB/s
test from_utf8_cyr_fast ... bench: 10,025 ns/iter (+/- 245) = 511 MB/s
test from_utf8_cyr_regular ... bench: 12,250 ns/iter (+/- 437) = 418 MB/s
test from_utf8_dewik10_fast ... bench: 6,017,909 ns/iter (+/- 105,755) = 1740 MB/s
test from_utf8_dewik10_regular ... bench: 11,669,493 ns/iter (+/- 264,045) = 891 MB/s
test from_utf8_enwik8_fast ... bench: 14,085,692 ns/iter (+/- 1,643,316) = 7000 MB/s
test from_utf8_enwik8_regular ... bench: 93,657,410 ns/iter (+/- 5,353,353) = 1000 MB/s
test from_utf8_jawik10_fast ... bench: 29,154,073 ns/iter (+/- 4,659,534) = 340 MB/s
test from_utf8_jawik10_regular ... bench: 29,112,917 ns/iter (+/- 2,475,123) = 340 MB/s
```
Co-authored-by: Björn Steinbrink <bsteinbr@gmail.com>
2016-01-06 14:43:33 +00:00
|
|
|
|
let len = v.len();
|
2016-11-21 22:26:31 +00:00
|
|
|
|
|
|
|
|
|
let usize_bytes = mem::size_of::<usize>();
|
|
|
|
|
let ascii_block_size = 2 * usize_bytes;
|
|
|
|
|
let blocks_end = if len >= ascii_block_size { len - ascii_block_size + 1 } else { 0 };
|
Optimize pointer alignment in utf8 validation
This uses (and reuses) the u8 arrays's inherent block alignment when checking whether the current index is block aligned.
I initially thought that this would just move the expensive `align_offset` call out of the while loop and replace it with a subtraction and bitwise AND. But it appears this optimizes much better, too...
before: https://rust.godbolt.org/z/WIPvWl
after: https://rust.godbolt.org/z/-jBPoW
https://github.com/jridgewell/faster-from_utf8/tree/pointer-alignment
```
test from_utf8_2_bytes_fast ... bench: 310 ns/iter (+/- 42) = 1290 MB/s
test from_utf8_2_bytes_regular ... bench: 309 ns/iter (+/- 24) = 1294 MB/s
test from_utf8_3_bytes_fast ... bench: 1,027 ns/iter (+/- 62) = 1168 MB/s
test from_utf8_3_bytes_regular ... bench: 1,513 ns/iter (+/- 611) = 793 MB/s
test from_utf8_4_bytes_fast ... bench: 1,788 ns/iter (+/- 26) = 1342 MB/s
test from_utf8_4_bytes_regular ... bench: 1,907 ns/iter (+/- 181) = 1258 MB/s
test from_utf8_all_bytes_fast ... bench: 3,463 ns/iter (+/- 97) = 1155 MB/s
test from_utf8_all_bytes_regular ... bench: 4,083 ns/iter (+/- 89) = 979 MB/s
test from_utf8_ascii_fast ... bench: 88 ns/iter (+/- 4) = 28988 MB/s
test from_utf8_ascii_regular ... bench: 88 ns/iter (+/- 8) = 28988 MB/s
test from_utf8_cyr_fast ... bench: 7,707 ns/iter (+/- 531) = 665 MB/s
test from_utf8_cyr_regular ... bench: 8,202 ns/iter (+/- 135) = 625 MB/s
test from_utf8_enwik8_fast ... bench: 1,135,756 ns/iter (+/- 84,450) = 8804 MB/s
test from_utf8_enwik8_regular ... bench: 1,145,468 ns/iter (+/- 79,601) = 8730 MB/s
test from_utf8_jawik10_fast ... bench: 12,723,844 ns/iter (+/- 473,247) = 785 MB/s
test from_utf8_jawik10_regular ... bench: 13,384,596 ns/iter (+/- 666,997) = 747 MB/s
test from_utf8_mixed_fast ... bench: 2,321 ns/iter (+/- 123) = 2081 MB/s
test from_utf8_mixed_regular ... bench: 2,702 ns/iter (+/- 408) = 1788 MB/s
test from_utf8_mostlyasc_fast ... bench: 249 ns/iter (+/- 10) = 14666 MB/s
test from_utf8_mostlyasc_regular ... bench: 276 ns/iter (+/- 5) = 13231 MB/s
```
2019-05-30 03:33:35 +00:00
|
|
|
|
let align = v.as_ptr().align_offset(usize_bytes);
|
2016-11-21 22:26:31 +00:00
|
|
|
|
|
2016-11-22 12:47:45 +00:00
|
|
|
|
while index < len {
|
|
|
|
|
let old_offset = index;
|
2017-03-02 16:27:57 +00:00
|
|
|
|
macro_rules! err {
|
2017-03-06 21:06:30 +00:00
|
|
|
|
($error_len: expr) => {
|
2019-12-22 22:42:04 +00:00
|
|
|
|
return Err(Utf8Error { valid_up_to: old_offset, error_len: $error_len });
|
|
|
|
|
};
|
2017-03-02 16:27:57 +00:00
|
|
|
|
}
|
2015-01-02 22:44:21 +00:00
|
|
|
|
|
2019-12-22 22:42:04 +00:00
|
|
|
|
macro_rules! next {
|
|
|
|
|
() => {{
|
|
|
|
|
index += 1;
|
|
|
|
|
// we needed data, but there was none: error!
|
|
|
|
|
if index >= len {
|
|
|
|
|
err!(None)
|
|
|
|
|
}
|
|
|
|
|
v[index]
|
|
|
|
|
}};
|
|
|
|
|
}
|
2014-05-01 06:06:36 +00:00
|
|
|
|
|
2016-11-22 12:47:45 +00:00
|
|
|
|
let first = v[index];
|
2014-05-01 06:06:36 +00:00
|
|
|
|
if first >= 128 {
|
2015-03-10 11:06:44 +00:00
|
|
|
|
let w = UTF8_CHAR_WIDTH[first as usize];
|
2014-12-09 22:08:10 +00:00
|
|
|
|
// 2-byte encoding is for codepoints \u{0080} to \u{07ff}
|
2014-05-01 06:06:36 +00:00
|
|
|
|
// first C2 80 last DF BF
|
2014-12-09 22:08:10 +00:00
|
|
|
|
// 3-byte encoding is for codepoints \u{0800} to \u{ffff}
|
2014-05-01 06:06:36 +00:00
|
|
|
|
// first E0 A0 80 last EF BF BF
|
2014-12-09 22:08:10 +00:00
|
|
|
|
// excluding surrogates codepoints \u{d800} to \u{dfff}
|
2014-05-01 06:06:36 +00:00
|
|
|
|
// ED A0 80 to ED BF BF
|
2014-12-09 22:08:10 +00:00
|
|
|
|
// 4-byte encoding is for codepoints \u{1000}0 to \u{10ff}ff
|
2014-05-01 06:06:36 +00:00
|
|
|
|
// first F0 90 80 80 last F4 8F BF BF
|
|
|
|
|
//
|
|
|
|
|
// Use the UTF-8 syntax from the RFC
|
|
|
|
|
//
|
|
|
|
|
// https://tools.ietf.org/html/rfc3629
|
|
|
|
|
// UTF8-1 = %x00-7F
|
|
|
|
|
// UTF8-2 = %xC2-DF UTF8-tail
|
|
|
|
|
// UTF8-3 = %xE0 %xA0-BF UTF8-tail / %xE1-EC 2( UTF8-tail ) /
|
|
|
|
|
// %xED %x80-9F UTF8-tail / %xEE-EF 2( UTF8-tail )
|
|
|
|
|
// UTF8-4 = %xF0 %x90-BF 2( UTF8-tail ) / %xF1-F3 3( UTF8-tail ) /
|
|
|
|
|
// %xF4 %x80-8F 2( UTF8-tail )
|
|
|
|
|
match w {
|
2019-12-22 22:42:04 +00:00
|
|
|
|
2 => {
|
|
|
|
|
if next!() & !CONT_MASK != TAG_CONT_U8 {
|
|
|
|
|
err!(Some(1))
|
|
|
|
|
}
|
|
|
|
|
}
|
2014-05-01 06:06:36 +00:00
|
|
|
|
3 => {
|
2017-03-02 16:27:57 +00:00
|
|
|
|
match (first, next!()) {
|
2019-12-22 22:42:04 +00:00
|
|
|
|
(0xE0, 0xA0..=0xBF)
|
|
|
|
|
| (0xE1..=0xEC, 0x80..=0xBF)
|
|
|
|
|
| (0xED, 0x80..=0x9F)
|
|
|
|
|
| (0xEE..=0xEF, 0x80..=0xBF) => {}
|
|
|
|
|
_ => err!(Some(1)),
|
2017-03-02 16:27:57 +00:00
|
|
|
|
}
|
|
|
|
|
if next!() & !CONT_MASK != TAG_CONT_U8 {
|
|
|
|
|
err!(Some(2))
|
2014-05-01 06:06:36 +00:00
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
4 => {
|
2017-03-02 16:27:57 +00:00
|
|
|
|
match (first, next!()) {
|
2019-12-22 22:42:04 +00:00
|
|
|
|
(0xF0, 0x90..=0xBF) | (0xF1..=0xF3, 0x80..=0xBF) | (0xF4, 0x80..=0x8F) => {}
|
|
|
|
|
_ => err!(Some(1)),
|
2017-03-02 16:27:57 +00:00
|
|
|
|
}
|
|
|
|
|
if next!() & !CONT_MASK != TAG_CONT_U8 {
|
|
|
|
|
err!(Some(2))
|
|
|
|
|
}
|
|
|
|
|
if next!() & !CONT_MASK != TAG_CONT_U8 {
|
|
|
|
|
err!(Some(3))
|
2014-05-01 06:06:36 +00:00
|
|
|
|
}
|
|
|
|
|
}
|
2019-12-22 22:42:04 +00:00
|
|
|
|
_ => err!(Some(1)),
|
2014-05-01 06:06:36 +00:00
|
|
|
|
}
|
2016-11-22 12:47:45 +00:00
|
|
|
|
index += 1;
|
Add fast path for ASCII in UTF-8 validation
This speeds up the ascii case (and long stretches of ascii in otherwise
mixed UTF-8 data) when checking UTF-8 validity.
Benchmark results suggest that on purely ASCII input, we can improve
throughput (megabytes verified / second) by a factor of 13 to 14!
On xml and mostly english language input (en.wikipedia xml dump),
throughput increases by a factor 7.
On mostly non-ASCII input, performance increases slightly or is the
same.
The UTF-8 validation is rewritten to use indexed access; since all
access is preceded by a (mandatory for validation) length check, they
are statically elided by llvm and this formulation is in fact the best
for performance. A previous version had losses due to slice to iterator
conversions.
A large credit to Björn Steinbrink who improved this patch immensely,
writing this second version.
Benchmark results on x86-64 (Sandy Bridge) compiled with -C opt-level=3.
Old code is `regular`, this PR is called `fast`.
Datasets:
- `ascii` is just ascii (2.5 kB)
- `cyr` is cyrillic script with ascii spaces (5 kB)
- `dewik10` is 10MB of a de.wikipedia xml dump
- `enwik10` is 100MB of an en.wikipedia xml dump
- `jawik10` is 10MB of a ja.wikipedia xml dump
```
test from_utf8_ascii_fast ... bench: 140 ns/iter (+/- 4) = 18221 MB/s
test from_utf8_ascii_regular ... bench: 1,932 ns/iter (+/- 19) = 1320 MB/s
test from_utf8_cyr_fast ... bench: 10,025 ns/iter (+/- 245) = 511 MB/s
test from_utf8_cyr_regular ... bench: 12,250 ns/iter (+/- 437) = 418 MB/s
test from_utf8_dewik10_fast ... bench: 6,017,909 ns/iter (+/- 105,755) = 1740 MB/s
test from_utf8_dewik10_regular ... bench: 11,669,493 ns/iter (+/- 264,045) = 891 MB/s
test from_utf8_enwik8_fast ... bench: 14,085,692 ns/iter (+/- 1,643,316) = 7000 MB/s
test from_utf8_enwik8_regular ... bench: 93,657,410 ns/iter (+/- 5,353,353) = 1000 MB/s
test from_utf8_jawik10_fast ... bench: 29,154,073 ns/iter (+/- 4,659,534) = 340 MB/s
test from_utf8_jawik10_regular ... bench: 29,112,917 ns/iter (+/- 2,475,123) = 340 MB/s
```
Co-authored-by: Björn Steinbrink <bsteinbr@gmail.com>
2016-01-06 14:43:33 +00:00
|
|
|
|
} else {
|
|
|
|
|
// Ascii case, try to skip forward quickly.
|
2016-01-12 22:04:46 +00:00
|
|
|
|
// When the pointer is aligned, read 2 words of data per iteration
|
|
|
|
|
// until we find a word containing a non-ascii byte.
|
2019-07-05 20:39:56 +00:00
|
|
|
|
if align != usize::max_value() && align.wrapping_sub(index) % usize_bytes == 0 {
|
Optimize pointer alignment in utf8 validation
This uses (and reuses) the u8 arrays's inherent block alignment when checking whether the current index is block aligned.
I initially thought that this would just move the expensive `align_offset` call out of the while loop and replace it with a subtraction and bitwise AND. But it appears this optimizes much better, too...
before: https://rust.godbolt.org/z/WIPvWl
after: https://rust.godbolt.org/z/-jBPoW
https://github.com/jridgewell/faster-from_utf8/tree/pointer-alignment
```
test from_utf8_2_bytes_fast ... bench: 310 ns/iter (+/- 42) = 1290 MB/s
test from_utf8_2_bytes_regular ... bench: 309 ns/iter (+/- 24) = 1294 MB/s
test from_utf8_3_bytes_fast ... bench: 1,027 ns/iter (+/- 62) = 1168 MB/s
test from_utf8_3_bytes_regular ... bench: 1,513 ns/iter (+/- 611) = 793 MB/s
test from_utf8_4_bytes_fast ... bench: 1,788 ns/iter (+/- 26) = 1342 MB/s
test from_utf8_4_bytes_regular ... bench: 1,907 ns/iter (+/- 181) = 1258 MB/s
test from_utf8_all_bytes_fast ... bench: 3,463 ns/iter (+/- 97) = 1155 MB/s
test from_utf8_all_bytes_regular ... bench: 4,083 ns/iter (+/- 89) = 979 MB/s
test from_utf8_ascii_fast ... bench: 88 ns/iter (+/- 4) = 28988 MB/s
test from_utf8_ascii_regular ... bench: 88 ns/iter (+/- 8) = 28988 MB/s
test from_utf8_cyr_fast ... bench: 7,707 ns/iter (+/- 531) = 665 MB/s
test from_utf8_cyr_regular ... bench: 8,202 ns/iter (+/- 135) = 625 MB/s
test from_utf8_enwik8_fast ... bench: 1,135,756 ns/iter (+/- 84,450) = 8804 MB/s
test from_utf8_enwik8_regular ... bench: 1,145,468 ns/iter (+/- 79,601) = 8730 MB/s
test from_utf8_jawik10_fast ... bench: 12,723,844 ns/iter (+/- 473,247) = 785 MB/s
test from_utf8_jawik10_regular ... bench: 13,384,596 ns/iter (+/- 666,997) = 747 MB/s
test from_utf8_mixed_fast ... bench: 2,321 ns/iter (+/- 123) = 2081 MB/s
test from_utf8_mixed_regular ... bench: 2,702 ns/iter (+/- 408) = 1788 MB/s
test from_utf8_mostlyasc_fast ... bench: 249 ns/iter (+/- 10) = 14666 MB/s
test from_utf8_mostlyasc_regular ... bench: 276 ns/iter (+/- 5) = 13231 MB/s
```
2019-05-30 03:33:35 +00:00
|
|
|
|
let ptr = v.as_ptr();
|
2016-11-22 12:47:45 +00:00
|
|
|
|
while index < blocks_end {
|
2019-11-20 02:28:32 +00:00
|
|
|
|
// SAFETY: since `align - index` and `ascii_block_size` are multiples of
|
|
|
|
|
// `usize_bytes`, `ptr.add(index)` is always aligned with a `usize` so we
|
|
|
|
|
// may cast directly to a `const` pointer.
|
2016-11-21 22:26:31 +00:00
|
|
|
|
unsafe {
|
2018-08-20 02:16:22 +00:00
|
|
|
|
let block = ptr.add(index) as *const usize;
|
2016-11-21 22:26:31 +00:00
|
|
|
|
// break if there is a nonascii byte
|
2016-11-22 12:47:45 +00:00
|
|
|
|
let zu = contains_nonascii(*block);
|
|
|
|
|
let zv = contains_nonascii(*block.offset(1));
|
|
|
|
|
if zu | zv {
|
2016-11-21 22:26:31 +00:00
|
|
|
|
break;
|
Add fast path for ASCII in UTF-8 validation
This speeds up the ascii case (and long stretches of ascii in otherwise
mixed UTF-8 data) when checking UTF-8 validity.
Benchmark results suggest that on purely ASCII input, we can improve
throughput (megabytes verified / second) by a factor of 13 to 14!
On xml and mostly english language input (en.wikipedia xml dump),
throughput increases by a factor 7.
On mostly non-ASCII input, performance increases slightly or is the
same.
The UTF-8 validation is rewritten to use indexed access; since all
access is preceded by a (mandatory for validation) length check, they
are statically elided by llvm and this formulation is in fact the best
for performance. A previous version had losses due to slice to iterator
conversions.
A large credit to Björn Steinbrink who improved this patch immensely,
writing this second version.
Benchmark results on x86-64 (Sandy Bridge) compiled with -C opt-level=3.
Old code is `regular`, this PR is called `fast`.
Datasets:
- `ascii` is just ascii (2.5 kB)
- `cyr` is cyrillic script with ascii spaces (5 kB)
- `dewik10` is 10MB of a de.wikipedia xml dump
- `enwik10` is 100MB of an en.wikipedia xml dump
- `jawik10` is 10MB of a ja.wikipedia xml dump
```
test from_utf8_ascii_fast ... bench: 140 ns/iter (+/- 4) = 18221 MB/s
test from_utf8_ascii_regular ... bench: 1,932 ns/iter (+/- 19) = 1320 MB/s
test from_utf8_cyr_fast ... bench: 10,025 ns/iter (+/- 245) = 511 MB/s
test from_utf8_cyr_regular ... bench: 12,250 ns/iter (+/- 437) = 418 MB/s
test from_utf8_dewik10_fast ... bench: 6,017,909 ns/iter (+/- 105,755) = 1740 MB/s
test from_utf8_dewik10_regular ... bench: 11,669,493 ns/iter (+/- 264,045) = 891 MB/s
test from_utf8_enwik8_fast ... bench: 14,085,692 ns/iter (+/- 1,643,316) = 7000 MB/s
test from_utf8_enwik8_regular ... bench: 93,657,410 ns/iter (+/- 5,353,353) = 1000 MB/s
test from_utf8_jawik10_fast ... bench: 29,154,073 ns/iter (+/- 4,659,534) = 340 MB/s
test from_utf8_jawik10_regular ... bench: 29,112,917 ns/iter (+/- 2,475,123) = 340 MB/s
```
Co-authored-by: Björn Steinbrink <bsteinbr@gmail.com>
2016-01-06 14:43:33 +00:00
|
|
|
|
}
|
|
|
|
|
}
|
2016-11-22 12:47:45 +00:00
|
|
|
|
index += ascii_block_size;
|
Add fast path for ASCII in UTF-8 validation
This speeds up the ascii case (and long stretches of ascii in otherwise
mixed UTF-8 data) when checking UTF-8 validity.
Benchmark results suggest that on purely ASCII input, we can improve
throughput (megabytes verified / second) by a factor of 13 to 14!
On xml and mostly english language input (en.wikipedia xml dump),
throughput increases by a factor 7.
On mostly non-ASCII input, performance increases slightly or is the
same.
The UTF-8 validation is rewritten to use indexed access; since all
access is preceded by a (mandatory for validation) length check, they
are statically elided by llvm and this formulation is in fact the best
for performance. A previous version had losses due to slice to iterator
conversions.
A large credit to Björn Steinbrink who improved this patch immensely,
writing this second version.
Benchmark results on x86-64 (Sandy Bridge) compiled with -C opt-level=3.
Old code is `regular`, this PR is called `fast`.
Datasets:
- `ascii` is just ascii (2.5 kB)
- `cyr` is cyrillic script with ascii spaces (5 kB)
- `dewik10` is 10MB of a de.wikipedia xml dump
- `enwik10` is 100MB of an en.wikipedia xml dump
- `jawik10` is 10MB of a ja.wikipedia xml dump
```
test from_utf8_ascii_fast ... bench: 140 ns/iter (+/- 4) = 18221 MB/s
test from_utf8_ascii_regular ... bench: 1,932 ns/iter (+/- 19) = 1320 MB/s
test from_utf8_cyr_fast ... bench: 10,025 ns/iter (+/- 245) = 511 MB/s
test from_utf8_cyr_regular ... bench: 12,250 ns/iter (+/- 437) = 418 MB/s
test from_utf8_dewik10_fast ... bench: 6,017,909 ns/iter (+/- 105,755) = 1740 MB/s
test from_utf8_dewik10_regular ... bench: 11,669,493 ns/iter (+/- 264,045) = 891 MB/s
test from_utf8_enwik8_fast ... bench: 14,085,692 ns/iter (+/- 1,643,316) = 7000 MB/s
test from_utf8_enwik8_regular ... bench: 93,657,410 ns/iter (+/- 5,353,353) = 1000 MB/s
test from_utf8_jawik10_fast ... bench: 29,154,073 ns/iter (+/- 4,659,534) = 340 MB/s
test from_utf8_jawik10_regular ... bench: 29,112,917 ns/iter (+/- 2,475,123) = 340 MB/s
```
Co-authored-by: Björn Steinbrink <bsteinbr@gmail.com>
2016-01-06 14:43:33 +00:00
|
|
|
|
}
|
|
|
|
|
// step from the point where the wordwise loop stopped
|
2016-11-22 12:47:45 +00:00
|
|
|
|
while index < len && v[index] < 128 {
|
|
|
|
|
index += 1;
|
Add fast path for ASCII in UTF-8 validation
This speeds up the ascii case (and long stretches of ascii in otherwise
mixed UTF-8 data) when checking UTF-8 validity.
Benchmark results suggest that on purely ASCII input, we can improve
throughput (megabytes verified / second) by a factor of 13 to 14!
On xml and mostly english language input (en.wikipedia xml dump),
throughput increases by a factor 7.
On mostly non-ASCII input, performance increases slightly or is the
same.
The UTF-8 validation is rewritten to use indexed access; since all
access is preceded by a (mandatory for validation) length check, they
are statically elided by llvm and this formulation is in fact the best
for performance. A previous version had losses due to slice to iterator
conversions.
A large credit to Björn Steinbrink who improved this patch immensely,
writing this second version.
Benchmark results on x86-64 (Sandy Bridge) compiled with -C opt-level=3.
Old code is `regular`, this PR is called `fast`.
Datasets:
- `ascii` is just ascii (2.5 kB)
- `cyr` is cyrillic script with ascii spaces (5 kB)
- `dewik10` is 10MB of a de.wikipedia xml dump
- `enwik10` is 100MB of an en.wikipedia xml dump
- `jawik10` is 10MB of a ja.wikipedia xml dump
```
test from_utf8_ascii_fast ... bench: 140 ns/iter (+/- 4) = 18221 MB/s
test from_utf8_ascii_regular ... bench: 1,932 ns/iter (+/- 19) = 1320 MB/s
test from_utf8_cyr_fast ... bench: 10,025 ns/iter (+/- 245) = 511 MB/s
test from_utf8_cyr_regular ... bench: 12,250 ns/iter (+/- 437) = 418 MB/s
test from_utf8_dewik10_fast ... bench: 6,017,909 ns/iter (+/- 105,755) = 1740 MB/s
test from_utf8_dewik10_regular ... bench: 11,669,493 ns/iter (+/- 264,045) = 891 MB/s
test from_utf8_enwik8_fast ... bench: 14,085,692 ns/iter (+/- 1,643,316) = 7000 MB/s
test from_utf8_enwik8_regular ... bench: 93,657,410 ns/iter (+/- 5,353,353) = 1000 MB/s
test from_utf8_jawik10_fast ... bench: 29,154,073 ns/iter (+/- 4,659,534) = 340 MB/s
test from_utf8_jawik10_regular ... bench: 29,112,917 ns/iter (+/- 2,475,123) = 340 MB/s
```
Co-authored-by: Björn Steinbrink <bsteinbr@gmail.com>
2016-01-06 14:43:33 +00:00
|
|
|
|
}
|
|
|
|
|
} else {
|
2016-11-22 12:47:45 +00:00
|
|
|
|
index += 1;
|
Add fast path for ASCII in UTF-8 validation
This speeds up the ascii case (and long stretches of ascii in otherwise
mixed UTF-8 data) when checking UTF-8 validity.
Benchmark results suggest that on purely ASCII input, we can improve
throughput (megabytes verified / second) by a factor of 13 to 14!
On xml and mostly english language input (en.wikipedia xml dump),
throughput increases by a factor 7.
On mostly non-ASCII input, performance increases slightly or is the
same.
The UTF-8 validation is rewritten to use indexed access; since all
access is preceded by a (mandatory for validation) length check, they
are statically elided by llvm and this formulation is in fact the best
for performance. A previous version had losses due to slice to iterator
conversions.
A large credit to Björn Steinbrink who improved this patch immensely,
writing this second version.
Benchmark results on x86-64 (Sandy Bridge) compiled with -C opt-level=3.
Old code is `regular`, this PR is called `fast`.
Datasets:
- `ascii` is just ascii (2.5 kB)
- `cyr` is cyrillic script with ascii spaces (5 kB)
- `dewik10` is 10MB of a de.wikipedia xml dump
- `enwik10` is 100MB of an en.wikipedia xml dump
- `jawik10` is 10MB of a ja.wikipedia xml dump
```
test from_utf8_ascii_fast ... bench: 140 ns/iter (+/- 4) = 18221 MB/s
test from_utf8_ascii_regular ... bench: 1,932 ns/iter (+/- 19) = 1320 MB/s
test from_utf8_cyr_fast ... bench: 10,025 ns/iter (+/- 245) = 511 MB/s
test from_utf8_cyr_regular ... bench: 12,250 ns/iter (+/- 437) = 418 MB/s
test from_utf8_dewik10_fast ... bench: 6,017,909 ns/iter (+/- 105,755) = 1740 MB/s
test from_utf8_dewik10_regular ... bench: 11,669,493 ns/iter (+/- 264,045) = 891 MB/s
test from_utf8_enwik8_fast ... bench: 14,085,692 ns/iter (+/- 1,643,316) = 7000 MB/s
test from_utf8_enwik8_regular ... bench: 93,657,410 ns/iter (+/- 5,353,353) = 1000 MB/s
test from_utf8_jawik10_fast ... bench: 29,154,073 ns/iter (+/- 4,659,534) = 340 MB/s
test from_utf8_jawik10_regular ... bench: 29,112,917 ns/iter (+/- 2,475,123) = 340 MB/s
```
Co-authored-by: Björn Steinbrink <bsteinbr@gmail.com>
2016-01-06 14:43:33 +00:00
|
|
|
|
}
|
2014-05-01 06:06:36 +00:00
|
|
|
|
}
|
|
|
|
|
}
|
Add fast path for ASCII in UTF-8 validation
This speeds up the ascii case (and long stretches of ascii in otherwise
mixed UTF-8 data) when checking UTF-8 validity.
Benchmark results suggest that on purely ASCII input, we can improve
throughput (megabytes verified / second) by a factor of 13 to 14!
On xml and mostly english language input (en.wikipedia xml dump),
throughput increases by a factor 7.
On mostly non-ASCII input, performance increases slightly or is the
same.
The UTF-8 validation is rewritten to use indexed access; since all
access is preceded by a (mandatory for validation) length check, they
are statically elided by llvm and this formulation is in fact the best
for performance. A previous version had losses due to slice to iterator
conversions.
A large credit to Björn Steinbrink who improved this patch immensely,
writing this second version.
Benchmark results on x86-64 (Sandy Bridge) compiled with -C opt-level=3.
Old code is `regular`, this PR is called `fast`.
Datasets:
- `ascii` is just ascii (2.5 kB)
- `cyr` is cyrillic script with ascii spaces (5 kB)
- `dewik10` is 10MB of a de.wikipedia xml dump
- `enwik10` is 100MB of an en.wikipedia xml dump
- `jawik10` is 10MB of a ja.wikipedia xml dump
```
test from_utf8_ascii_fast ... bench: 140 ns/iter (+/- 4) = 18221 MB/s
test from_utf8_ascii_regular ... bench: 1,932 ns/iter (+/- 19) = 1320 MB/s
test from_utf8_cyr_fast ... bench: 10,025 ns/iter (+/- 245) = 511 MB/s
test from_utf8_cyr_regular ... bench: 12,250 ns/iter (+/- 437) = 418 MB/s
test from_utf8_dewik10_fast ... bench: 6,017,909 ns/iter (+/- 105,755) = 1740 MB/s
test from_utf8_dewik10_regular ... bench: 11,669,493 ns/iter (+/- 264,045) = 891 MB/s
test from_utf8_enwik8_fast ... bench: 14,085,692 ns/iter (+/- 1,643,316) = 7000 MB/s
test from_utf8_enwik8_regular ... bench: 93,657,410 ns/iter (+/- 5,353,353) = 1000 MB/s
test from_utf8_jawik10_fast ... bench: 29,154,073 ns/iter (+/- 4,659,534) = 340 MB/s
test from_utf8_jawik10_regular ... bench: 29,112,917 ns/iter (+/- 2,475,123) = 340 MB/s
```
Co-authored-by: Björn Steinbrink <bsteinbr@gmail.com>
2016-01-06 14:43:33 +00:00
|
|
|
|
|
|
|
|
|
Ok(())
|
2014-05-01 06:06:36 +00:00
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
// https://tools.ietf.org/html/rfc3629
|
2014-12-30 08:19:41 +00:00
|
|
|
|
static UTF8_CHAR_WIDTH: [u8; 256] = [
|
2019-12-22 22:42:04 +00:00
|
|
|
|
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
|
|
|
|
|
1, // 0x1F
|
|
|
|
|
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
|
|
|
|
|
1, // 0x3F
|
|
|
|
|
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
|
|
|
|
|
1, // 0x5F
|
|
|
|
|
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
|
|
|
|
|
1, // 0x7F
|
|
|
|
|
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
|
|
|
|
|
0, // 0x9F
|
|
|
|
|
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
|
|
|
|
|
0, // 0xBF
|
|
|
|
|
0, 0, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
|
|
|
|
|
2, // 0xDF
|
|
|
|
|
3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, // 0xEF
|
|
|
|
|
4, 4, 4, 4, 4, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, // 0xFF
|
2014-05-01 06:06:36 +00:00
|
|
|
|
];
|
|
|
|
|
|
2017-03-22 00:42:23 +00:00
|
|
|
|
/// Given a first byte, determines how many bytes are in this UTF-8 character.
|
2019-12-21 11:16:18 +00:00
|
|
|
|
#[unstable(feature = "str_internals", issue = "none")]
|
2017-03-01 21:41:44 +00:00
|
|
|
|
#[inline]
|
|
|
|
|
pub fn utf8_char_width(b: u8) -> usize {
|
2018-08-04 12:31:03 +00:00
|
|
|
|
UTF8_CHAR_WIDTH[b as usize] as usize
|
2017-03-01 21:41:44 +00:00
|
|
|
|
}
|
|
|
|
|
|
2017-03-22 00:42:23 +00:00
|
|
|
|
/// Mask of the value bits of a continuation byte.
|
2015-03-03 08:42:26 +00:00
|
|
|
|
const CONT_MASK: u8 = 0b0011_1111;
|
2017-03-22 00:42:23 +00:00
|
|
|
|
/// Value of the tag bits (tag mask is !CONT_MASK) of a continuation byte.
|
2015-03-03 08:42:26 +00:00
|
|
|
|
const TAG_CONT_U8: u8 = 0b1000_0000;
|
2014-05-01 06:06:36 +00:00
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
Section: Trait implementations
|
|
|
|
|
*/
|
|
|
|
|
|
2015-01-05 07:32:20 +00:00
|
|
|
|
mod traits {
|
2019-04-15 02:23:21 +00:00
|
|
|
|
use crate::cmp::Ordering;
|
|
|
|
|
use crate::ops;
|
|
|
|
|
use crate::slice::{self, SliceIndex};
|
2014-05-01 06:06:36 +00:00
|
|
|
|
|
2017-02-16 17:18:18 +00:00
|
|
|
|
/// Implements ordering of strings.
|
|
|
|
|
///
|
2019-02-09 21:23:30 +00:00
|
|
|
|
/// Strings are ordered lexicographically by their byte values. This orders Unicode code
|
|
|
|
|
/// points based on their positions in the code charts. This is not necessarily the same as
|
|
|
|
|
/// "alphabetical" order, which varies by language and locale. Sorting strings according to
|
2017-02-16 17:18:18 +00:00
|
|
|
|
/// culturally-accepted standards requires locale-specific data that is outside the scope of
|
|
|
|
|
/// the `str` type.
|
2015-01-24 05:48:20 +00:00
|
|
|
|
#[stable(feature = "rust1", since = "1.0.0")]
|
2014-10-30 01:11:16 +00:00
|
|
|
|
impl Ord for str {
|
|
|
|
|
#[inline]
|
|
|
|
|
fn cmp(&self, other: &str) -> Ordering {
|
2016-04-05 12:06:20 +00:00
|
|
|
|
self.as_bytes().cmp(other.as_bytes())
|
2014-10-30 01:11:16 +00:00
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
2015-01-24 05:48:20 +00:00
|
|
|
|
#[stable(feature = "rust1", since = "1.0.0")]
|
2014-10-30 01:11:16 +00:00
|
|
|
|
impl PartialEq for str {
|
|
|
|
|
#[inline]
|
|
|
|
|
fn eq(&self, other: &str) -> bool {
|
2017-10-03 01:04:23 +00:00
|
|
|
|
self.as_bytes() == other.as_bytes()
|
2014-10-30 01:11:16 +00:00
|
|
|
|
}
|
|
|
|
|
#[inline]
|
2019-12-22 22:42:04 +00:00
|
|
|
|
fn ne(&self, other: &str) -> bool {
|
|
|
|
|
!(*self).eq(other)
|
|
|
|
|
}
|
2014-10-30 01:11:16 +00:00
|
|
|
|
}
|
|
|
|
|
|
2015-01-24 05:48:20 +00:00
|
|
|
|
#[stable(feature = "rust1", since = "1.0.0")]
|
2014-10-30 01:11:16 +00:00
|
|
|
|
impl Eq for str {}
|
|
|
|
|
|
2017-02-16 17:18:18 +00:00
|
|
|
|
/// Implements comparison operations on strings.
|
|
|
|
|
///
|
2019-02-09 21:23:30 +00:00
|
|
|
|
/// Strings are compared lexicographically by their byte values. This compares Unicode code
|
|
|
|
|
/// points based on their positions in the code charts. This is not necessarily the same as
|
|
|
|
|
/// "alphabetical" order, which varies by language and locale. Comparing strings according to
|
2017-02-16 17:18:18 +00:00
|
|
|
|
/// culturally-accepted standards requires locale-specific data that is outside the scope of
|
|
|
|
|
/// the `str` type.
|
2015-01-24 05:48:20 +00:00
|
|
|
|
#[stable(feature = "rust1", since = "1.0.0")]
|
2014-10-30 01:11:16 +00:00
|
|
|
|
impl PartialOrd for str {
|
|
|
|
|
#[inline]
|
|
|
|
|
fn partial_cmp(&self, other: &str) -> Option<Ordering> {
|
|
|
|
|
Some(self.cmp(other))
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
2015-03-21 23:33:27 +00:00
|
|
|
|
#[stable(feature = "rust1", since = "1.0.0")]
|
2019-01-14 13:10:45 +00:00
|
|
|
|
impl<I> ops::Index<I> for str
|
|
|
|
|
where
|
|
|
|
|
I: SliceIndex<str>,
|
|
|
|
|
{
|
|
|
|
|
type Output = I::Output;
|
2015-03-21 23:33:27 +00:00
|
|
|
|
|
|
|
|
|
#[inline]
|
2019-01-14 13:10:45 +00:00
|
|
|
|
fn index(&self, index: I) -> &I::Output {
|
2017-06-04 08:47:21 +00:00
|
|
|
|
index.index(self)
|
2015-03-21 23:33:27 +00:00
|
|
|
|
}
|
2014-12-31 07:20:40 +00:00
|
|
|
|
}
|
2015-01-18 00:15:47 +00:00
|
|
|
|
|
2015-01-28 04:06:46 +00:00
|
|
|
|
#[stable(feature = "rust1", since = "1.0.0")]
|
2019-01-14 13:10:45 +00:00
|
|
|
|
impl<I> ops::IndexMut<I> for str
|
|
|
|
|
where
|
|
|
|
|
I: SliceIndex<str>,
|
|
|
|
|
{
|
2016-01-28 16:20:48 +00:00
|
|
|
|
#[inline]
|
2019-01-14 13:10:45 +00:00
|
|
|
|
fn index_mut(&mut self, index: I) -> &mut I::Output {
|
2017-06-04 08:47:21 +00:00
|
|
|
|
index.index_mut(self)
|
2016-01-28 16:20:48 +00:00
|
|
|
|
}
|
|
|
|
|
}
|
2017-03-22 16:36:43 +00:00
|
|
|
|
|
2018-04-30 11:36:27 +00:00
|
|
|
|
#[inline(never)]
|
|
|
|
|
#[cold]
|
|
|
|
|
fn str_index_overflow_fail() -> ! {
|
|
|
|
|
panic!("attempted to index str up to maximum usize");
|
|
|
|
|
}
|
|
|
|
|
|
2019-01-14 13:10:45 +00:00
|
|
|
|
/// Implements substring slicing with syntax `&self[..]` or `&mut self[..]`.
|
|
|
|
|
///
|
|
|
|
|
/// Returns a slice of the whole string, i.e., returns `&self` or `&mut
|
|
|
|
|
/// self`. Equivalent to `&self[0 .. len]` or `&mut self[0 .. len]`. Unlike
|
|
|
|
|
/// other indexing operations, this can never panic.
|
|
|
|
|
///
|
|
|
|
|
/// This operation is `O(1)`.
|
|
|
|
|
///
|
|
|
|
|
/// Prior to 1.20.0, these indexing operations were still supported by
|
|
|
|
|
/// direct implementation of `Index` and `IndexMut`.
|
|
|
|
|
///
|
|
|
|
|
/// Equivalent to `&self[0 .. len]` or `&mut self[0 .. len]`.
|
2017-07-20 22:42:14 +00:00
|
|
|
|
#[stable(feature = "str_checked_slicing", since = "1.20.0")]
|
2017-03-22 16:36:43 +00:00
|
|
|
|
impl SliceIndex<str> for ops::RangeFull {
|
|
|
|
|
type Output = str;
|
|
|
|
|
#[inline]
|
|
|
|
|
fn get(self, slice: &str) -> Option<&Self::Output> {
|
|
|
|
|
Some(slice)
|
|
|
|
|
}
|
|
|
|
|
#[inline]
|
|
|
|
|
fn get_mut(self, slice: &mut str) -> Option<&mut Self::Output> {
|
|
|
|
|
Some(slice)
|
|
|
|
|
}
|
|
|
|
|
#[inline]
|
|
|
|
|
unsafe fn get_unchecked(self, slice: &str) -> &Self::Output {
|
|
|
|
|
slice
|
|
|
|
|
}
|
|
|
|
|
#[inline]
|
|
|
|
|
unsafe fn get_unchecked_mut(self, slice: &mut str) -> &mut Self::Output {
|
|
|
|
|
slice
|
|
|
|
|
}
|
|
|
|
|
#[inline]
|
|
|
|
|
fn index(self, slice: &str) -> &Self::Output {
|
|
|
|
|
slice
|
|
|
|
|
}
|
|
|
|
|
#[inline]
|
|
|
|
|
fn index_mut(self, slice: &mut str) -> &mut Self::Output {
|
|
|
|
|
slice
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
2019-01-14 13:10:45 +00:00
|
|
|
|
/// Implements substring slicing with syntax `&self[begin .. end]` or `&mut
|
|
|
|
|
/// self[begin .. end]`.
|
|
|
|
|
///
|
|
|
|
|
/// Returns a slice of the given string from the byte range
|
|
|
|
|
/// [`begin`, `end`).
|
|
|
|
|
///
|
|
|
|
|
/// This operation is `O(1)`.
|
|
|
|
|
///
|
|
|
|
|
/// Prior to 1.20.0, these indexing operations were still supported by
|
|
|
|
|
/// direct implementation of `Index` and `IndexMut`.
|
|
|
|
|
///
|
|
|
|
|
/// # Panics
|
|
|
|
|
///
|
|
|
|
|
/// Panics if `begin` or `end` does not point to the starting byte offset of
|
|
|
|
|
/// a character (as defined by `is_char_boundary`), if `begin > end`, or if
|
|
|
|
|
/// `end > len`.
|
|
|
|
|
///
|
|
|
|
|
/// # Examples
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// let s = "Löwe 老虎 Léopard";
|
|
|
|
|
/// assert_eq!(&s[0 .. 1], "L");
|
|
|
|
|
///
|
|
|
|
|
/// assert_eq!(&s[1 .. 9], "öwe 老");
|
|
|
|
|
///
|
|
|
|
|
/// // these will panic:
|
|
|
|
|
/// // byte 2 lies within `ö`:
|
|
|
|
|
/// // &s[2 ..3];
|
|
|
|
|
///
|
|
|
|
|
/// // byte 8 lies within `老`
|
|
|
|
|
/// // &s[1 .. 8];
|
|
|
|
|
///
|
|
|
|
|
/// // byte 100 is outside the string
|
|
|
|
|
/// // &s[3 .. 100];
|
|
|
|
|
/// ```
|
2017-07-20 22:42:14 +00:00
|
|
|
|
#[stable(feature = "str_checked_slicing", since = "1.20.0")]
|
2017-03-22 16:36:43 +00:00
|
|
|
|
impl SliceIndex<str> for ops::Range<usize> {
|
|
|
|
|
type Output = str;
|
|
|
|
|
#[inline]
|
|
|
|
|
fn get(self, slice: &str) -> Option<&Self::Output> {
|
2019-12-22 22:42:04 +00:00
|
|
|
|
if self.start <= self.end
|
|
|
|
|
&& slice.is_char_boundary(self.start)
|
|
|
|
|
&& slice.is_char_boundary(self.end)
|
|
|
|
|
{
|
2019-11-20 02:28:32 +00:00
|
|
|
|
// SAFETY: just checked that `start` and `end` are on a char boundary
|
2017-03-22 16:36:43 +00:00
|
|
|
|
Some(unsafe { self.get_unchecked(slice) })
|
|
|
|
|
} else {
|
|
|
|
|
None
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
#[inline]
|
|
|
|
|
fn get_mut(self, slice: &mut str) -> Option<&mut Self::Output> {
|
2019-12-22 22:42:04 +00:00
|
|
|
|
if self.start <= self.end
|
|
|
|
|
&& slice.is_char_boundary(self.start)
|
|
|
|
|
&& slice.is_char_boundary(self.end)
|
|
|
|
|
{
|
2019-11-20 02:28:32 +00:00
|
|
|
|
// SAFETY: just checked that `start` and `end` are on a char boundary
|
2017-03-22 16:36:43 +00:00
|
|
|
|
Some(unsafe { self.get_unchecked_mut(slice) })
|
|
|
|
|
} else {
|
|
|
|
|
None
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
#[inline]
|
|
|
|
|
unsafe fn get_unchecked(self, slice: &str) -> &Self::Output {
|
2018-08-20 02:16:22 +00:00
|
|
|
|
let ptr = slice.as_ptr().add(self.start);
|
2017-03-22 16:36:43 +00:00
|
|
|
|
let len = self.end - self.start;
|
|
|
|
|
super::from_utf8_unchecked(slice::from_raw_parts(ptr, len))
|
|
|
|
|
}
|
|
|
|
|
#[inline]
|
|
|
|
|
unsafe fn get_unchecked_mut(self, slice: &mut str) -> &mut Self::Output {
|
2019-02-05 19:07:45 +00:00
|
|
|
|
let ptr = slice.as_mut_ptr().add(self.start);
|
2017-03-22 16:36:43 +00:00
|
|
|
|
let len = self.end - self.start;
|
2019-02-06 10:23:10 +00:00
|
|
|
|
super::from_utf8_unchecked_mut(slice::from_raw_parts_mut(ptr, len))
|
2017-03-22 16:36:43 +00:00
|
|
|
|
}
|
|
|
|
|
#[inline]
|
|
|
|
|
fn index(self, slice: &str) -> &Self::Output {
|
|
|
|
|
let (start, end) = (self.start, self.end);
|
|
|
|
|
self.get(slice).unwrap_or_else(|| super::slice_error_fail(slice, start, end))
|
|
|
|
|
}
|
|
|
|
|
#[inline]
|
|
|
|
|
fn index_mut(self, slice: &mut str) -> &mut Self::Output {
|
|
|
|
|
// is_char_boundary checks that the index is in [0, .len()]
|
2018-11-11 13:52:36 +00:00
|
|
|
|
// cannot reuse `get` as above, because of NLL trouble
|
2019-12-22 22:42:04 +00:00
|
|
|
|
if self.start <= self.end
|
|
|
|
|
&& slice.is_char_boundary(self.start)
|
|
|
|
|
&& slice.is_char_boundary(self.end)
|
|
|
|
|
{
|
2019-11-20 02:28:32 +00:00
|
|
|
|
// SAFETY: just checked that `start` and `end` are on a char boundary
|
2017-03-22 16:36:43 +00:00
|
|
|
|
unsafe { self.get_unchecked_mut(slice) }
|
|
|
|
|
} else {
|
|
|
|
|
super::slice_error_fail(slice, self.start, self.end)
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
2019-01-14 13:10:45 +00:00
|
|
|
|
/// Implements substring slicing with syntax `&self[.. end]` or `&mut
|
|
|
|
|
/// self[.. end]`.
|
|
|
|
|
///
|
|
|
|
|
/// Returns a slice of the given string from the byte range [`0`, `end`).
|
|
|
|
|
/// Equivalent to `&self[0 .. end]` or `&mut self[0 .. end]`.
|
|
|
|
|
///
|
|
|
|
|
/// This operation is `O(1)`.
|
|
|
|
|
///
|
|
|
|
|
/// Prior to 1.20.0, these indexing operations were still supported by
|
|
|
|
|
/// direct implementation of `Index` and `IndexMut`.
|
|
|
|
|
///
|
|
|
|
|
/// # Panics
|
|
|
|
|
///
|
|
|
|
|
/// Panics if `end` does not point to the starting byte offset of a
|
|
|
|
|
/// character (as defined by `is_char_boundary`), or if `end > len`.
|
2017-07-20 22:42:14 +00:00
|
|
|
|
#[stable(feature = "str_checked_slicing", since = "1.20.0")]
|
2017-03-22 16:36:43 +00:00
|
|
|
|
impl SliceIndex<str> for ops::RangeTo<usize> {
|
|
|
|
|
type Output = str;
|
|
|
|
|
#[inline]
|
|
|
|
|
fn get(self, slice: &str) -> Option<&Self::Output> {
|
|
|
|
|
if slice.is_char_boundary(self.end) {
|
2019-11-20 02:28:32 +00:00
|
|
|
|
// SAFETY: just checked that `end` is on a char boundary
|
2017-03-22 16:36:43 +00:00
|
|
|
|
Some(unsafe { self.get_unchecked(slice) })
|
|
|
|
|
} else {
|
|
|
|
|
None
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
#[inline]
|
|
|
|
|
fn get_mut(self, slice: &mut str) -> Option<&mut Self::Output> {
|
|
|
|
|
if slice.is_char_boundary(self.end) {
|
2019-11-20 02:28:32 +00:00
|
|
|
|
// SAFETY: just checked that `end` is on a char boundary
|
2017-03-22 16:36:43 +00:00
|
|
|
|
Some(unsafe { self.get_unchecked_mut(slice) })
|
|
|
|
|
} else {
|
|
|
|
|
None
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
#[inline]
|
|
|
|
|
unsafe fn get_unchecked(self, slice: &str) -> &Self::Output {
|
|
|
|
|
let ptr = slice.as_ptr();
|
|
|
|
|
super::from_utf8_unchecked(slice::from_raw_parts(ptr, self.end))
|
|
|
|
|
}
|
|
|
|
|
#[inline]
|
|
|
|
|
unsafe fn get_unchecked_mut(self, slice: &mut str) -> &mut Self::Output {
|
2019-02-05 19:07:45 +00:00
|
|
|
|
let ptr = slice.as_mut_ptr();
|
2019-02-06 10:23:10 +00:00
|
|
|
|
super::from_utf8_unchecked_mut(slice::from_raw_parts_mut(ptr, self.end))
|
2017-03-22 16:36:43 +00:00
|
|
|
|
}
|
|
|
|
|
#[inline]
|
|
|
|
|
fn index(self, slice: &str) -> &Self::Output {
|
|
|
|
|
let end = self.end;
|
|
|
|
|
self.get(slice).unwrap_or_else(|| super::slice_error_fail(slice, 0, end))
|
|
|
|
|
}
|
|
|
|
|
#[inline]
|
|
|
|
|
fn index_mut(self, slice: &mut str) -> &mut Self::Output {
|
|
|
|
|
if slice.is_char_boundary(self.end) {
|
2019-11-20 02:28:32 +00:00
|
|
|
|
// SAFETY: just checked that `end` is on a char boundary
|
2017-03-22 16:36:43 +00:00
|
|
|
|
unsafe { self.get_unchecked_mut(slice) }
|
|
|
|
|
} else {
|
|
|
|
|
super::slice_error_fail(slice, 0, self.end)
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
2019-01-14 13:10:45 +00:00
|
|
|
|
/// Implements substring slicing with syntax `&self[begin ..]` or `&mut
|
|
|
|
|
/// self[begin ..]`.
|
|
|
|
|
///
|
|
|
|
|
/// Returns a slice of the given string from the byte range [`begin`,
|
|
|
|
|
/// `len`). Equivalent to `&self[begin .. len]` or `&mut self[begin ..
|
|
|
|
|
/// len]`.
|
|
|
|
|
///
|
|
|
|
|
/// This operation is `O(1)`.
|
|
|
|
|
///
|
|
|
|
|
/// Prior to 1.20.0, these indexing operations were still supported by
|
|
|
|
|
/// direct implementation of `Index` and `IndexMut`.
|
|
|
|
|
///
|
|
|
|
|
/// # Panics
|
|
|
|
|
///
|
|
|
|
|
/// Panics if `begin` does not point to the starting byte offset of
|
|
|
|
|
/// a character (as defined by `is_char_boundary`), or if `begin >= len`.
|
2017-07-20 22:42:14 +00:00
|
|
|
|
#[stable(feature = "str_checked_slicing", since = "1.20.0")]
|
2017-03-22 16:36:43 +00:00
|
|
|
|
impl SliceIndex<str> for ops::RangeFrom<usize> {
|
|
|
|
|
type Output = str;
|
|
|
|
|
#[inline]
|
|
|
|
|
fn get(self, slice: &str) -> Option<&Self::Output> {
|
|
|
|
|
if slice.is_char_boundary(self.start) {
|
2019-11-20 02:28:32 +00:00
|
|
|
|
// SAFETY: just checked that `start` is on a char boundary
|
2017-03-22 16:36:43 +00:00
|
|
|
|
Some(unsafe { self.get_unchecked(slice) })
|
|
|
|
|
} else {
|
|
|
|
|
None
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
#[inline]
|
|
|
|
|
fn get_mut(self, slice: &mut str) -> Option<&mut Self::Output> {
|
|
|
|
|
if slice.is_char_boundary(self.start) {
|
2019-11-20 02:28:32 +00:00
|
|
|
|
// SAFETY: just checked that `start` is on a char boundary
|
2017-03-22 16:36:43 +00:00
|
|
|
|
Some(unsafe { self.get_unchecked_mut(slice) })
|
|
|
|
|
} else {
|
|
|
|
|
None
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
#[inline]
|
|
|
|
|
unsafe fn get_unchecked(self, slice: &str) -> &Self::Output {
|
2018-08-20 02:16:22 +00:00
|
|
|
|
let ptr = slice.as_ptr().add(self.start);
|
2017-03-22 16:36:43 +00:00
|
|
|
|
let len = slice.len() - self.start;
|
|
|
|
|
super::from_utf8_unchecked(slice::from_raw_parts(ptr, len))
|
|
|
|
|
}
|
|
|
|
|
#[inline]
|
|
|
|
|
unsafe fn get_unchecked_mut(self, slice: &mut str) -> &mut Self::Output {
|
2019-02-05 19:07:45 +00:00
|
|
|
|
let ptr = slice.as_mut_ptr().add(self.start);
|
2017-03-22 16:36:43 +00:00
|
|
|
|
let len = slice.len() - self.start;
|
2019-02-06 10:23:10 +00:00
|
|
|
|
super::from_utf8_unchecked_mut(slice::from_raw_parts_mut(ptr, len))
|
2017-03-22 16:36:43 +00:00
|
|
|
|
}
|
|
|
|
|
#[inline]
|
|
|
|
|
fn index(self, slice: &str) -> &Self::Output {
|
|
|
|
|
let (start, end) = (self.start, slice.len());
|
|
|
|
|
self.get(slice).unwrap_or_else(|| super::slice_error_fail(slice, start, end))
|
|
|
|
|
}
|
|
|
|
|
#[inline]
|
|
|
|
|
fn index_mut(self, slice: &mut str) -> &mut Self::Output {
|
|
|
|
|
if slice.is_char_boundary(self.start) {
|
2019-11-20 02:28:32 +00:00
|
|
|
|
// SAFETY: just checked that `start` is on a char boundary
|
2017-03-22 16:36:43 +00:00
|
|
|
|
unsafe { self.get_unchecked_mut(slice) }
|
|
|
|
|
} else {
|
|
|
|
|
super::slice_error_fail(slice, self.start, slice.len())
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
2019-01-14 13:10:45 +00:00
|
|
|
|
/// Implements substring slicing with syntax `&self[begin ..= end]` or `&mut
|
|
|
|
|
/// self[begin ..= end]`.
|
|
|
|
|
///
|
|
|
|
|
/// Returns a slice of the given string from the byte range
|
|
|
|
|
/// [`begin`, `end`]. Equivalent to `&self [begin .. end + 1]` or `&mut
|
|
|
|
|
/// self[begin .. end + 1]`, except if `end` has the maximum value for
|
|
|
|
|
/// `usize`.
|
|
|
|
|
///
|
|
|
|
|
/// This operation is `O(1)`.
|
|
|
|
|
///
|
|
|
|
|
/// # Panics
|
|
|
|
|
///
|
|
|
|
|
/// Panics if `begin` does not point to the starting byte offset of
|
|
|
|
|
/// a character (as defined by `is_char_boundary`), if `end` does not point
|
|
|
|
|
/// to the ending byte offset of a character (`end + 1` is either a starting
|
|
|
|
|
/// byte offset or equal to `len`), if `begin > end`, or if `end >= len`.
|
2018-01-27 19:09:36 +00:00
|
|
|
|
#[stable(feature = "inclusive_range", since = "1.26.0")]
|
2017-03-22 16:36:43 +00:00
|
|
|
|
impl SliceIndex<str> for ops::RangeInclusive<usize> {
|
|
|
|
|
type Output = str;
|
|
|
|
|
#[inline]
|
|
|
|
|
fn get(self, slice: &str) -> Option<&Self::Output> {
|
2019-12-22 22:42:04 +00:00
|
|
|
|
if *self.end() == usize::max_value() {
|
|
|
|
|
None
|
|
|
|
|
} else {
|
|
|
|
|
(*self.start()..self.end() + 1).get(slice)
|
|
|
|
|
}
|
2017-03-22 16:36:43 +00:00
|
|
|
|
}
|
|
|
|
|
#[inline]
|
|
|
|
|
fn get_mut(self, slice: &mut str) -> Option<&mut Self::Output> {
|
2019-12-22 22:42:04 +00:00
|
|
|
|
if *self.end() == usize::max_value() {
|
|
|
|
|
None
|
|
|
|
|
} else {
|
|
|
|
|
(*self.start()..self.end() + 1).get_mut(slice)
|
|
|
|
|
}
|
2017-03-22 16:36:43 +00:00
|
|
|
|
}
|
|
|
|
|
#[inline]
|
|
|
|
|
unsafe fn get_unchecked(self, slice: &str) -> &Self::Output {
|
2019-12-22 22:42:04 +00:00
|
|
|
|
(*self.start()..self.end() + 1).get_unchecked(slice)
|
2017-03-22 16:36:43 +00:00
|
|
|
|
}
|
|
|
|
|
#[inline]
|
|
|
|
|
unsafe fn get_unchecked_mut(self, slice: &mut str) -> &mut Self::Output {
|
2019-12-22 22:42:04 +00:00
|
|
|
|
(*self.start()..self.end() + 1).get_unchecked_mut(slice)
|
2017-03-22 16:36:43 +00:00
|
|
|
|
}
|
|
|
|
|
#[inline]
|
|
|
|
|
fn index(self, slice: &str) -> &Self::Output {
|
2019-12-22 22:42:04 +00:00
|
|
|
|
if *self.end() == usize::max_value() {
|
|
|
|
|
str_index_overflow_fail();
|
|
|
|
|
}
|
|
|
|
|
(*self.start()..self.end() + 1).index(slice)
|
2017-03-22 16:36:43 +00:00
|
|
|
|
}
|
|
|
|
|
#[inline]
|
|
|
|
|
fn index_mut(self, slice: &mut str) -> &mut Self::Output {
|
2019-12-22 22:42:04 +00:00
|
|
|
|
if *self.end() == usize::max_value() {
|
|
|
|
|
str_index_overflow_fail();
|
|
|
|
|
}
|
|
|
|
|
(*self.start()..self.end() + 1).index_mut(slice)
|
2017-03-22 16:36:43 +00:00
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
2019-01-14 13:10:45 +00:00
|
|
|
|
/// Implements substring slicing with syntax `&self[..= end]` or `&mut
|
|
|
|
|
/// self[..= end]`.
|
|
|
|
|
///
|
|
|
|
|
/// Returns a slice of the given string from the byte range [0, `end`].
|
|
|
|
|
/// Equivalent to `&self [0 .. end + 1]`, except if `end` has the maximum
|
|
|
|
|
/// value for `usize`.
|
|
|
|
|
///
|
|
|
|
|
/// This operation is `O(1)`.
|
|
|
|
|
///
|
|
|
|
|
/// # Panics
|
|
|
|
|
///
|
|
|
|
|
/// Panics if `end` does not point to the ending byte offset of a character
|
|
|
|
|
/// (`end + 1` is either a starting byte offset as defined by
|
|
|
|
|
/// `is_char_boundary`, or equal to `len`), or if `end >= len`.
|
2018-01-27 19:09:36 +00:00
|
|
|
|
#[stable(feature = "inclusive_range", since = "1.26.0")]
|
2017-03-22 16:36:43 +00:00
|
|
|
|
impl SliceIndex<str> for ops::RangeToInclusive<usize> {
|
|
|
|
|
type Output = str;
|
|
|
|
|
#[inline]
|
|
|
|
|
fn get(self, slice: &str) -> Option<&Self::Output> {
|
2019-12-22 22:42:04 +00:00
|
|
|
|
if self.end == usize::max_value() { None } else { (..self.end + 1).get(slice) }
|
2017-03-22 16:36:43 +00:00
|
|
|
|
}
|
|
|
|
|
#[inline]
|
|
|
|
|
fn get_mut(self, slice: &mut str) -> Option<&mut Self::Output> {
|
2019-12-22 22:42:04 +00:00
|
|
|
|
if self.end == usize::max_value() { None } else { (..self.end + 1).get_mut(slice) }
|
2017-03-22 16:36:43 +00:00
|
|
|
|
}
|
|
|
|
|
#[inline]
|
|
|
|
|
unsafe fn get_unchecked(self, slice: &str) -> &Self::Output {
|
2019-12-22 22:42:04 +00:00
|
|
|
|
(..self.end + 1).get_unchecked(slice)
|
2017-03-22 16:36:43 +00:00
|
|
|
|
}
|
|
|
|
|
#[inline]
|
|
|
|
|
unsafe fn get_unchecked_mut(self, slice: &mut str) -> &mut Self::Output {
|
2019-12-22 22:42:04 +00:00
|
|
|
|
(..self.end + 1).get_unchecked_mut(slice)
|
2017-03-22 16:36:43 +00:00
|
|
|
|
}
|
|
|
|
|
#[inline]
|
|
|
|
|
fn index(self, slice: &str) -> &Self::Output {
|
2019-12-22 22:42:04 +00:00
|
|
|
|
if self.end == usize::max_value() {
|
|
|
|
|
str_index_overflow_fail();
|
|
|
|
|
}
|
|
|
|
|
(..self.end + 1).index(slice)
|
2017-03-22 16:36:43 +00:00
|
|
|
|
}
|
|
|
|
|
#[inline]
|
|
|
|
|
fn index_mut(self, slice: &mut str) -> &mut Self::Output {
|
2019-12-22 22:42:04 +00:00
|
|
|
|
if self.end == usize::max_value() {
|
|
|
|
|
str_index_overflow_fail();
|
|
|
|
|
}
|
|
|
|
|
(..self.end + 1).index_mut(slice)
|
2017-03-22 16:36:43 +00:00
|
|
|
|
}
|
|
|
|
|
}
|
2018-04-12 06:36:31 +00:00
|
|
|
|
}
|
2014-05-01 06:06:36 +00:00
|
|
|
|
|
2016-03-05 17:09:14 +00:00
|
|
|
|
// truncate `&str` to length at most equal to `max`
|
|
|
|
|
// return `true` if it were truncated, and the new str.
|
|
|
|
|
fn truncate_to_char_boundary(s: &str, mut max: usize) -> (bool, &str) {
|
|
|
|
|
if max >= s.len() {
|
|
|
|
|
(false, s)
|
|
|
|
|
} else {
|
|
|
|
|
while !s.is_char_boundary(max) {
|
|
|
|
|
max -= 1;
|
|
|
|
|
}
|
|
|
|
|
(true, &s[..max])
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
2014-08-23 10:30:08 +00:00
|
|
|
|
#[inline(never)]
|
2015-11-23 04:04:01 +00:00
|
|
|
|
#[cold]
|
2015-01-27 13:09:18 +00:00
|
|
|
|
fn slice_error_fail(s: &str, begin: usize, end: usize) -> ! {
|
2016-03-05 17:09:14 +00:00
|
|
|
|
const MAX_DISPLAY_LENGTH: usize = 256;
|
2016-11-29 03:11:12 +00:00
|
|
|
|
let (truncated, s_trunc) = truncate_to_char_boundary(s, MAX_DISPLAY_LENGTH);
|
2016-03-05 17:09:14 +00:00
|
|
|
|
let ellipsis = if truncated { "[...]" } else { "" };
|
|
|
|
|
|
2016-11-29 03:11:12 +00:00
|
|
|
|
// 1. out of bounds
|
|
|
|
|
if begin > s.len() || end > s.len() {
|
|
|
|
|
let oob_index = if begin > s.len() { begin } else { end };
|
|
|
|
|
panic!("byte index {} is out of bounds of `{}`{}", oob_index, s_trunc, ellipsis);
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
// 2. begin <= end
|
2019-12-22 22:42:04 +00:00
|
|
|
|
assert!(
|
|
|
|
|
begin <= end,
|
|
|
|
|
"begin <= end ({} <= {}) when slicing `{}`{}",
|
|
|
|
|
begin,
|
|
|
|
|
end,
|
|
|
|
|
s_trunc,
|
|
|
|
|
ellipsis
|
|
|
|
|
);
|
2016-11-29 03:11:12 +00:00
|
|
|
|
|
|
|
|
|
// 3. character boundary
|
|
|
|
|
let index = if !s.is_char_boundary(begin) { begin } else { end };
|
|
|
|
|
// find the character
|
|
|
|
|
let mut char_start = index;
|
|
|
|
|
while !s.is_char_boundary(char_start) {
|
|
|
|
|
char_start -= 1;
|
|
|
|
|
}
|
|
|
|
|
// `char_start` must be less than len and a char boundary
|
|
|
|
|
let ch = s[char_start..].chars().next().unwrap();
|
2019-12-22 22:42:04 +00:00
|
|
|
|
let char_range = char_start..char_start + ch.len_utf8();
|
|
|
|
|
panic!(
|
|
|
|
|
"byte index {} is not a char boundary; it is inside {:?} (bytes {:?}) of `{}`{}",
|
|
|
|
|
index, ch, char_range, s_trunc, ellipsis
|
|
|
|
|
);
|
2014-08-23 10:30:08 +00:00
|
|
|
|
}
|
|
|
|
|
|
2018-05-10 18:02:19 +00:00
|
|
|
|
#[lang = "str"]
|
|
|
|
|
#[cfg(not(test))]
|
|
|
|
|
impl str {
|
2018-04-07 19:56:02 +00:00
|
|
|
|
/// Returns the length of `self`.
|
|
|
|
|
///
|
|
|
|
|
/// This length is in bytes, not [`char`]s or graphemes. In other words,
|
|
|
|
|
/// it may not be what a human considers the length of the string.
|
|
|
|
|
///
|
|
|
|
|
/// # Examples
|
|
|
|
|
///
|
|
|
|
|
/// Basic usage:
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// let len = "foo".len();
|
|
|
|
|
/// assert_eq!(3, len);
|
|
|
|
|
///
|
2019-11-02 02:42:33 +00:00
|
|
|
|
/// assert_eq!("ƒoo".len(), 4); // fancy f!
|
|
|
|
|
/// assert_eq!("ƒoo".chars().count(), 3);
|
2018-04-07 19:56:02 +00:00
|
|
|
|
/// ```
|
|
|
|
|
#[stable(feature = "rust1", since = "1.0.0")]
|
2019-12-18 17:00:59 +00:00
|
|
|
|
#[rustc_const_stable(feature = "const_str_len", since = "1.32.0")]
|
2015-05-06 22:53:34 +00:00
|
|
|
|
#[inline]
|
2018-05-18 13:16:57 +00:00
|
|
|
|
pub const fn len(&self) -> usize {
|
2018-05-10 18:02:19 +00:00
|
|
|
|
self.as_bytes().len()
|
2015-05-06 22:53:34 +00:00
|
|
|
|
}
|
|
|
|
|
|
2018-04-07 19:56:02 +00:00
|
|
|
|
/// Returns `true` if `self` has a length of zero bytes.
|
|
|
|
|
///
|
|
|
|
|
/// # Examples
|
|
|
|
|
///
|
|
|
|
|
/// Basic usage:
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// let s = "";
|
|
|
|
|
/// assert!(s.is_empty());
|
|
|
|
|
///
|
|
|
|
|
/// let s = "not empty";
|
|
|
|
|
/// assert!(!s.is_empty());
|
|
|
|
|
/// ```
|
|
|
|
|
#[inline]
|
|
|
|
|
#[stable(feature = "rust1", since = "1.0.0")]
|
2019-12-18 17:00:59 +00:00
|
|
|
|
#[rustc_const_stable(feature = "const_str_is_empty", since = "1.32.0")]
|
2018-05-18 13:16:57 +00:00
|
|
|
|
pub const fn is_empty(&self) -> bool {
|
2018-05-10 18:02:19 +00:00
|
|
|
|
self.len() == 0
|
2018-04-07 19:56:02 +00:00
|
|
|
|
}
|
2018-04-05 17:00:48 +00:00
|
|
|
|
|
2018-04-07 19:56:02 +00:00
|
|
|
|
/// Checks that `index`-th byte lies at the start and/or end of a
|
|
|
|
|
/// UTF-8 code point sequence.
|
|
|
|
|
///
|
|
|
|
|
/// The start and end of the string (when `index == self.len()`) are
|
|
|
|
|
/// considered to be
|
|
|
|
|
/// boundaries.
|
|
|
|
|
///
|
|
|
|
|
/// Returns `false` if `index` is greater than `self.len()`.
|
|
|
|
|
///
|
|
|
|
|
/// # Examples
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// let s = "Löwe 老虎 Léopard";
|
|
|
|
|
/// assert!(s.is_char_boundary(0));
|
|
|
|
|
/// // start of `老`
|
|
|
|
|
/// assert!(s.is_char_boundary(6));
|
|
|
|
|
/// assert!(s.is_char_boundary(s.len()));
|
|
|
|
|
///
|
|
|
|
|
/// // second byte of `ö`
|
|
|
|
|
/// assert!(!s.is_char_boundary(2));
|
|
|
|
|
///
|
|
|
|
|
/// // third byte of `老`
|
|
|
|
|
/// assert!(!s.is_char_boundary(8));
|
|
|
|
|
/// ```
|
|
|
|
|
#[stable(feature = "is_char_boundary", since = "1.9.0")]
|
|
|
|
|
#[inline]
|
|
|
|
|
pub fn is_char_boundary(&self, index: usize) -> bool {
|
2018-05-10 18:02:19 +00:00
|
|
|
|
// 0 and len are always ok.
|
|
|
|
|
// Test for 0 explicitly so that it can optimize out the check
|
|
|
|
|
// easily and skip reading string data for that case.
|
2019-12-22 22:42:04 +00:00
|
|
|
|
if index == 0 || index == self.len() {
|
|
|
|
|
return true;
|
|
|
|
|
}
|
2018-05-10 18:02:19 +00:00
|
|
|
|
match self.as_bytes().get(index) {
|
|
|
|
|
None => false,
|
|
|
|
|
// This is bit magic equivalent to: b < 128 || b >= 192
|
|
|
|
|
Some(&b) => (b as i8) >= -0x40,
|
|
|
|
|
}
|
2018-04-07 19:56:02 +00:00
|
|
|
|
}
|
2018-04-05 17:00:48 +00:00
|
|
|
|
|
2018-04-07 19:56:02 +00:00
|
|
|
|
/// Converts a string slice to a byte slice. To convert the byte slice back
|
|
|
|
|
/// into a string slice, use the [`str::from_utf8`] function.
|
|
|
|
|
///
|
|
|
|
|
/// [`str::from_utf8`]: ./str/fn.from_utf8.html
|
|
|
|
|
///
|
|
|
|
|
/// # Examples
|
|
|
|
|
///
|
|
|
|
|
/// Basic usage:
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// let bytes = "bors".as_bytes();
|
|
|
|
|
/// assert_eq!(b"bors", bytes);
|
|
|
|
|
/// ```
|
|
|
|
|
#[stable(feature = "rust1", since = "1.0.0")]
|
2019-12-18 17:00:59 +00:00
|
|
|
|
#[rustc_const_stable(feature = "str_as_bytes", since = "1.32.0")]
|
2018-04-07 19:56:02 +00:00
|
|
|
|
#[inline(always)]
|
2019-10-04 14:43:15 +00:00
|
|
|
|
#[allow(unused_attributes)]
|
2019-09-25 12:42:46 +00:00
|
|
|
|
#[allow_internal_unstable(const_fn_union)]
|
2018-05-18 13:16:57 +00:00
|
|
|
|
pub const fn as_bytes(&self) -> &[u8] {
|
2019-08-28 15:38:24 +00:00
|
|
|
|
#[repr(C)]
|
2018-05-23 14:01:11 +00:00
|
|
|
|
union Slices<'a> {
|
|
|
|
|
str: &'a str,
|
|
|
|
|
slice: &'a [u8],
|
2018-05-18 13:16:57 +00:00
|
|
|
|
}
|
2019-11-20 02:28:32 +00:00
|
|
|
|
// SAFETY: const sound because we transmute two types with the same layout
|
2018-05-23 14:01:11 +00:00
|
|
|
|
unsafe { Slices { str: self }.slice }
|
2018-04-07 19:56:02 +00:00
|
|
|
|
}
|
2018-04-05 17:00:48 +00:00
|
|
|
|
|
2018-04-07 19:56:02 +00:00
|
|
|
|
/// Converts a mutable string slice to a mutable byte slice. To convert the
|
|
|
|
|
/// mutable byte slice back into a mutable string slice, use the
|
|
|
|
|
/// [`str::from_utf8_mut`] function.
|
|
|
|
|
///
|
|
|
|
|
/// [`str::from_utf8_mut`]: ./str/fn.from_utf8_mut.html
|
|
|
|
|
///
|
|
|
|
|
/// # Examples
|
|
|
|
|
///
|
|
|
|
|
/// Basic usage:
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// let mut s = String::from("Hello");
|
|
|
|
|
/// let bytes = unsafe { s.as_bytes_mut() };
|
|
|
|
|
///
|
|
|
|
|
/// assert_eq!(b"Hello", bytes);
|
|
|
|
|
/// ```
|
|
|
|
|
///
|
|
|
|
|
/// Mutability:
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// let mut s = String::from("🗻∈🌏");
|
|
|
|
|
///
|
|
|
|
|
/// unsafe {
|
|
|
|
|
/// let bytes = s.as_bytes_mut();
|
|
|
|
|
///
|
|
|
|
|
/// bytes[0] = 0xF0;
|
|
|
|
|
/// bytes[1] = 0x9F;
|
|
|
|
|
/// bytes[2] = 0x8D;
|
|
|
|
|
/// bytes[3] = 0x94;
|
|
|
|
|
/// }
|
|
|
|
|
///
|
|
|
|
|
/// assert_eq!("🍔∈🌏", s);
|
|
|
|
|
/// ```
|
|
|
|
|
#[stable(feature = "str_mut_extras", since = "1.20.0")]
|
|
|
|
|
#[inline(always)]
|
|
|
|
|
pub unsafe fn as_bytes_mut(&mut self) -> &mut [u8] {
|
2018-05-10 18:02:19 +00:00
|
|
|
|
&mut *(self as *mut str as *mut [u8])
|
2018-04-07 19:56:02 +00:00
|
|
|
|
}
|
2018-04-05 17:00:48 +00:00
|
|
|
|
|
2018-04-07 19:56:02 +00:00
|
|
|
|
/// Converts a string slice to a raw pointer.
|
|
|
|
|
///
|
|
|
|
|
/// As string slices are a slice of bytes, the raw pointer points to a
|
|
|
|
|
/// [`u8`]. This pointer will be pointing to the first byte of the string
|
|
|
|
|
/// slice.
|
|
|
|
|
///
|
2019-05-02 11:36:30 +00:00
|
|
|
|
/// The caller must ensure that the returned pointer is never written to.
|
|
|
|
|
/// If you need to mutate the contents of the string slice, use [`as_mut_ptr`].
|
2019-05-01 15:59:48 +00:00
|
|
|
|
///
|
2018-04-07 19:56:02 +00:00
|
|
|
|
/// [`u8`]: primitive.u8.html
|
2019-05-01 15:59:48 +00:00
|
|
|
|
/// [`as_mut_ptr`]: #method.as_mut_ptr
|
2018-04-07 19:56:02 +00:00
|
|
|
|
///
|
|
|
|
|
/// # Examples
|
|
|
|
|
///
|
|
|
|
|
/// Basic usage:
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// let s = "Hello";
|
|
|
|
|
/// let ptr = s.as_ptr();
|
|
|
|
|
/// ```
|
|
|
|
|
#[stable(feature = "rust1", since = "1.0.0")]
|
2019-12-18 17:00:59 +00:00
|
|
|
|
#[rustc_const_stable(feature = "rustc_str_as_ptr", since = "1.32.0")]
|
2018-04-05 17:00:48 +00:00
|
|
|
|
#[inline]
|
2018-05-18 13:16:57 +00:00
|
|
|
|
pub const fn as_ptr(&self) -> *const u8 {
|
2018-05-10 18:02:19 +00:00
|
|
|
|
self as *const str as *const u8
|
2018-04-05 17:00:48 +00:00
|
|
|
|
}
|
|
|
|
|
|
2019-02-05 19:07:45 +00:00
|
|
|
|
/// Converts a mutable string slice to a raw pointer.
|
|
|
|
|
///
|
|
|
|
|
/// As string slices are a slice of bytes, the raw pointer points to a
|
|
|
|
|
/// [`u8`]. This pointer will be pointing to the first byte of the string
|
|
|
|
|
/// slice.
|
|
|
|
|
///
|
|
|
|
|
/// It is your responsibility to make sure that the string slice only gets
|
|
|
|
|
/// modified in a way that it remains valid UTF-8.
|
|
|
|
|
///
|
|
|
|
|
/// [`u8`]: primitive.u8.html
|
2019-04-28 17:33:50 +00:00
|
|
|
|
#[stable(feature = "str_as_mut_ptr", since = "1.36.0")]
|
2019-02-05 19:07:45 +00:00
|
|
|
|
#[inline]
|
|
|
|
|
pub fn as_mut_ptr(&mut self) -> *mut u8 {
|
|
|
|
|
self as *mut str as *mut u8
|
|
|
|
|
}
|
|
|
|
|
|
2018-04-07 19:56:02 +00:00
|
|
|
|
/// Returns a subslice of `str`.
|
|
|
|
|
///
|
|
|
|
|
/// This is the non-panicking alternative to indexing the `str`. Returns
|
|
|
|
|
/// [`None`] whenever equivalent indexing operation would panic.
|
|
|
|
|
///
|
|
|
|
|
/// [`None`]: option/enum.Option.html#variant.None
|
|
|
|
|
///
|
|
|
|
|
/// # Examples
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// let v = String::from("🗻∈🌏");
|
|
|
|
|
///
|
|
|
|
|
/// assert_eq!(Some("🗻"), v.get(0..4));
|
|
|
|
|
///
|
|
|
|
|
/// // indices not on UTF-8 sequence boundaries
|
|
|
|
|
/// assert!(v.get(1..).is_none());
|
|
|
|
|
/// assert!(v.get(..8).is_none());
|
|
|
|
|
///
|
|
|
|
|
/// // out of bounds
|
|
|
|
|
/// assert!(v.get(..42).is_none());
|
|
|
|
|
/// ```
|
|
|
|
|
#[stable(feature = "str_checked_slicing", since = "1.20.0")]
|
2018-04-05 17:00:48 +00:00
|
|
|
|
#[inline]
|
2018-04-07 19:56:02 +00:00
|
|
|
|
pub fn get<I: SliceIndex<str>>(&self, i: I) -> Option<&I::Output> {
|
2018-05-10 18:02:19 +00:00
|
|
|
|
i.get(self)
|
2018-04-05 17:00:48 +00:00
|
|
|
|
}
|
|
|
|
|
|
2018-04-07 19:56:02 +00:00
|
|
|
|
/// Returns a mutable subslice of `str`.
|
|
|
|
|
///
|
|
|
|
|
/// This is the non-panicking alternative to indexing the `str`. Returns
|
|
|
|
|
/// [`None`] whenever equivalent indexing operation would panic.
|
|
|
|
|
///
|
|
|
|
|
/// [`None`]: option/enum.Option.html#variant.None
|
|
|
|
|
///
|
|
|
|
|
/// # Examples
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// let mut v = String::from("hello");
|
|
|
|
|
/// // correct length
|
|
|
|
|
/// assert!(v.get_mut(0..5).is_some());
|
|
|
|
|
/// // out of bounds
|
|
|
|
|
/// assert!(v.get_mut(..42).is_none());
|
|
|
|
|
/// assert_eq!(Some("he"), v.get_mut(0..2).map(|v| &*v));
|
|
|
|
|
///
|
|
|
|
|
/// assert_eq!("hello", v);
|
|
|
|
|
/// {
|
|
|
|
|
/// let s = v.get_mut(0..2);
|
|
|
|
|
/// let s = s.map(|s| {
|
|
|
|
|
/// s.make_ascii_uppercase();
|
|
|
|
|
/// &*s
|
|
|
|
|
/// });
|
|
|
|
|
/// assert_eq!(Some("HE"), s);
|
|
|
|
|
/// }
|
|
|
|
|
/// assert_eq!("HEllo", v);
|
|
|
|
|
/// ```
|
|
|
|
|
#[stable(feature = "str_checked_slicing", since = "1.20.0")]
|
|
|
|
|
#[inline]
|
|
|
|
|
pub fn get_mut<I: SliceIndex<str>>(&mut self, i: I) -> Option<&mut I::Output> {
|
2018-05-10 18:02:19 +00:00
|
|
|
|
i.get_mut(self)
|
2018-04-07 19:56:02 +00:00
|
|
|
|
}
|
2018-04-05 17:00:48 +00:00
|
|
|
|
|
2019-12-21 10:12:46 +00:00
|
|
|
|
/// Returns an unchecked subslice of `str`.
|
2018-04-07 19:56:02 +00:00
|
|
|
|
///
|
|
|
|
|
/// This is the unchecked alternative to indexing the `str`.
|
|
|
|
|
///
|
|
|
|
|
/// # Safety
|
|
|
|
|
///
|
|
|
|
|
/// Callers of this function are responsible that these preconditions are
|
|
|
|
|
/// satisfied:
|
|
|
|
|
///
|
|
|
|
|
/// * The starting index must come before the ending index;
|
|
|
|
|
/// * Indexes must be within bounds of the original slice;
|
|
|
|
|
/// * Indexes must lie on UTF-8 sequence boundaries.
|
|
|
|
|
///
|
|
|
|
|
/// Failing that, the returned string slice may reference invalid memory or
|
|
|
|
|
/// violate the invariants communicated by the `str` type.
|
|
|
|
|
///
|
|
|
|
|
/// # Examples
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// let v = "🗻∈🌏";
|
|
|
|
|
/// unsafe {
|
|
|
|
|
/// assert_eq!("🗻", v.get_unchecked(0..4));
|
|
|
|
|
/// assert_eq!("∈", v.get_unchecked(4..7));
|
|
|
|
|
/// assert_eq!("🌏", v.get_unchecked(7..11));
|
|
|
|
|
/// }
|
|
|
|
|
/// ```
|
|
|
|
|
#[stable(feature = "str_checked_slicing", since = "1.20.0")]
|
2018-04-05 17:00:48 +00:00
|
|
|
|
#[inline]
|
2018-04-07 19:56:02 +00:00
|
|
|
|
pub unsafe fn get_unchecked<I: SliceIndex<str>>(&self, i: I) -> &I::Output {
|
2018-05-10 18:02:19 +00:00
|
|
|
|
i.get_unchecked(self)
|
2018-04-05 17:00:48 +00:00
|
|
|
|
}
|
|
|
|
|
|
2018-04-07 19:56:02 +00:00
|
|
|
|
/// Returns a mutable, unchecked subslice of `str`.
|
|
|
|
|
///
|
|
|
|
|
/// This is the unchecked alternative to indexing the `str`.
|
|
|
|
|
///
|
|
|
|
|
/// # Safety
|
|
|
|
|
///
|
|
|
|
|
/// Callers of this function are responsible that these preconditions are
|
|
|
|
|
/// satisfied:
|
|
|
|
|
///
|
|
|
|
|
/// * The starting index must come before the ending index;
|
|
|
|
|
/// * Indexes must be within bounds of the original slice;
|
|
|
|
|
/// * Indexes must lie on UTF-8 sequence boundaries.
|
|
|
|
|
///
|
|
|
|
|
/// Failing that, the returned string slice may reference invalid memory or
|
|
|
|
|
/// violate the invariants communicated by the `str` type.
|
|
|
|
|
///
|
|
|
|
|
/// # Examples
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// let mut v = String::from("🗻∈🌏");
|
|
|
|
|
/// unsafe {
|
|
|
|
|
/// assert_eq!("🗻", v.get_unchecked_mut(0..4));
|
|
|
|
|
/// assert_eq!("∈", v.get_unchecked_mut(4..7));
|
|
|
|
|
/// assert_eq!("🌏", v.get_unchecked_mut(7..11));
|
|
|
|
|
/// }
|
|
|
|
|
/// ```
|
|
|
|
|
#[stable(feature = "str_checked_slicing", since = "1.20.0")]
|
2018-04-05 17:00:48 +00:00
|
|
|
|
#[inline]
|
2018-04-07 19:56:02 +00:00
|
|
|
|
pub unsafe fn get_unchecked_mut<I: SliceIndex<str>>(&mut self, i: I) -> &mut I::Output {
|
2018-05-10 18:02:19 +00:00
|
|
|
|
i.get_unchecked_mut(self)
|
2018-04-05 17:00:48 +00:00
|
|
|
|
}
|
|
|
|
|
|
2018-04-07 19:56:02 +00:00
|
|
|
|
/// Creates a string slice from another string slice, bypassing safety
|
|
|
|
|
/// checks.
|
|
|
|
|
///
|
|
|
|
|
/// This is generally not recommended, use with caution! For a safe
|
|
|
|
|
/// alternative see [`str`] and [`Index`].
|
|
|
|
|
///
|
|
|
|
|
/// [`str`]: primitive.str.html
|
|
|
|
|
/// [`Index`]: ops/trait.Index.html
|
|
|
|
|
///
|
|
|
|
|
/// This new slice goes from `begin` to `end`, including `begin` but
|
|
|
|
|
/// excluding `end`.
|
|
|
|
|
///
|
|
|
|
|
/// To get a mutable string slice instead, see the
|
|
|
|
|
/// [`slice_mut_unchecked`] method.
|
|
|
|
|
///
|
|
|
|
|
/// [`slice_mut_unchecked`]: #method.slice_mut_unchecked
|
|
|
|
|
///
|
|
|
|
|
/// # Safety
|
|
|
|
|
///
|
|
|
|
|
/// Callers of this function are responsible that three preconditions are
|
|
|
|
|
/// satisfied:
|
|
|
|
|
///
|
|
|
|
|
/// * `begin` must come before `end`.
|
|
|
|
|
/// * `begin` and `end` must be byte positions within the string slice.
|
|
|
|
|
/// * `begin` and `end` must lie on UTF-8 sequence boundaries.
|
|
|
|
|
///
|
|
|
|
|
/// # Examples
|
|
|
|
|
///
|
|
|
|
|
/// Basic usage:
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// let s = "Löwe 老虎 Léopard";
|
|
|
|
|
///
|
|
|
|
|
/// unsafe {
|
|
|
|
|
/// assert_eq!("Löwe 老虎 Léopard", s.slice_unchecked(0, 21));
|
|
|
|
|
/// }
|
|
|
|
|
///
|
|
|
|
|
/// let s = "Hello, world!";
|
|
|
|
|
///
|
|
|
|
|
/// unsafe {
|
|
|
|
|
/// assert_eq!("world", s.slice_unchecked(7, 12));
|
|
|
|
|
/// }
|
|
|
|
|
/// ```
|
|
|
|
|
#[stable(feature = "rust1", since = "1.0.0")]
|
2018-07-12 16:49:55 +00:00
|
|
|
|
#[rustc_deprecated(since = "1.29.0", reason = "use `get_unchecked(begin..end)` instead")]
|
2018-04-07 19:56:02 +00:00
|
|
|
|
#[inline]
|
|
|
|
|
pub unsafe fn slice_unchecked(&self, begin: usize, end: usize) -> &str {
|
2018-05-10 18:02:19 +00:00
|
|
|
|
(begin..end).get_unchecked(self)
|
2018-04-07 19:56:02 +00:00
|
|
|
|
}
|
2018-04-05 17:00:48 +00:00
|
|
|
|
|
2018-04-07 19:56:02 +00:00
|
|
|
|
/// Creates a string slice from another string slice, bypassing safety
|
|
|
|
|
/// checks.
|
|
|
|
|
/// This is generally not recommended, use with caution! For a safe
|
|
|
|
|
/// alternative see [`str`] and [`IndexMut`].
|
|
|
|
|
///
|
|
|
|
|
/// [`str`]: primitive.str.html
|
|
|
|
|
/// [`IndexMut`]: ops/trait.IndexMut.html
|
|
|
|
|
///
|
|
|
|
|
/// This new slice goes from `begin` to `end`, including `begin` but
|
|
|
|
|
/// excluding `end`.
|
|
|
|
|
///
|
|
|
|
|
/// To get an immutable string slice instead, see the
|
|
|
|
|
/// [`slice_unchecked`] method.
|
|
|
|
|
///
|
|
|
|
|
/// [`slice_unchecked`]: #method.slice_unchecked
|
|
|
|
|
///
|
|
|
|
|
/// # Safety
|
|
|
|
|
///
|
|
|
|
|
/// Callers of this function are responsible that three preconditions are
|
|
|
|
|
/// satisfied:
|
|
|
|
|
///
|
|
|
|
|
/// * `begin` must come before `end`.
|
|
|
|
|
/// * `begin` and `end` must be byte positions within the string slice.
|
|
|
|
|
/// * `begin` and `end` must lie on UTF-8 sequence boundaries.
|
|
|
|
|
#[stable(feature = "str_slice_mut", since = "1.5.0")]
|
2018-07-12 16:49:55 +00:00
|
|
|
|
#[rustc_deprecated(since = "1.29.0", reason = "use `get_unchecked_mut(begin..end)` instead")]
|
2018-04-07 19:56:02 +00:00
|
|
|
|
#[inline]
|
|
|
|
|
pub unsafe fn slice_mut_unchecked(&mut self, begin: usize, end: usize) -> &mut str {
|
2018-05-10 18:02:19 +00:00
|
|
|
|
(begin..end).get_unchecked_mut(self)
|
2018-04-07 19:56:02 +00:00
|
|
|
|
}
|
2018-04-05 17:00:48 +00:00
|
|
|
|
|
2018-04-07 19:56:02 +00:00
|
|
|
|
/// Divide one string slice into two at an index.
|
|
|
|
|
///
|
|
|
|
|
/// The argument, `mid`, should be a byte offset from the start of the
|
|
|
|
|
/// string. It must also be on the boundary of a UTF-8 code point.
|
|
|
|
|
///
|
|
|
|
|
/// The two slices returned go from the start of the string slice to `mid`,
|
|
|
|
|
/// and from `mid` to the end of the string slice.
|
|
|
|
|
///
|
|
|
|
|
/// To get mutable string slices instead, see the [`split_at_mut`]
|
|
|
|
|
/// method.
|
|
|
|
|
///
|
|
|
|
|
/// [`split_at_mut`]: #method.split_at_mut
|
|
|
|
|
///
|
|
|
|
|
/// # Panics
|
|
|
|
|
///
|
|
|
|
|
/// Panics if `mid` is not on a UTF-8 code point boundary, or if it is
|
|
|
|
|
/// beyond the last code point of the string slice.
|
|
|
|
|
///
|
|
|
|
|
/// # Examples
|
|
|
|
|
///
|
|
|
|
|
/// Basic usage:
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// let s = "Per Martin-Löf";
|
|
|
|
|
///
|
|
|
|
|
/// let (first, last) = s.split_at(3);
|
|
|
|
|
///
|
|
|
|
|
/// assert_eq!("Per", first);
|
|
|
|
|
/// assert_eq!(" Martin-Löf", last);
|
|
|
|
|
/// ```
|
|
|
|
|
#[inline]
|
|
|
|
|
#[stable(feature = "str_split_at", since = "1.4.0")]
|
|
|
|
|
pub fn split_at(&self, mid: usize) -> (&str, &str) {
|
2018-05-10 18:02:19 +00:00
|
|
|
|
// is_char_boundary checks that the index is in [0, .len()]
|
|
|
|
|
if self.is_char_boundary(mid) {
|
2019-11-20 02:28:32 +00:00
|
|
|
|
// SAFETY: just checked that `mid` is on a char boundary
|
2019-12-22 22:42:04 +00:00
|
|
|
|
unsafe { (self.get_unchecked(0..mid), self.get_unchecked(mid..self.len())) }
|
2018-05-10 18:02:19 +00:00
|
|
|
|
} else {
|
|
|
|
|
slice_error_fail(self, 0, mid)
|
|
|
|
|
}
|
2018-04-05 17:00:48 +00:00
|
|
|
|
}
|
|
|
|
|
|
2018-04-07 19:56:02 +00:00
|
|
|
|
/// Divide one mutable string slice into two at an index.
|
|
|
|
|
///
|
|
|
|
|
/// The argument, `mid`, should be a byte offset from the start of the
|
|
|
|
|
/// string. It must also be on the boundary of a UTF-8 code point.
|
|
|
|
|
///
|
|
|
|
|
/// The two slices returned go from the start of the string slice to `mid`,
|
|
|
|
|
/// and from `mid` to the end of the string slice.
|
|
|
|
|
///
|
|
|
|
|
/// To get immutable string slices instead, see the [`split_at`] method.
|
|
|
|
|
///
|
|
|
|
|
/// [`split_at`]: #method.split_at
|
|
|
|
|
///
|
|
|
|
|
/// # Panics
|
|
|
|
|
///
|
|
|
|
|
/// Panics if `mid` is not on a UTF-8 code point boundary, or if it is
|
|
|
|
|
/// beyond the last code point of the string slice.
|
|
|
|
|
///
|
|
|
|
|
/// # Examples
|
|
|
|
|
///
|
|
|
|
|
/// Basic usage:
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// let mut s = "Per Martin-Löf".to_string();
|
|
|
|
|
/// {
|
|
|
|
|
/// let (first, last) = s.split_at_mut(3);
|
|
|
|
|
/// first.make_ascii_uppercase();
|
|
|
|
|
/// assert_eq!("PER", first);
|
|
|
|
|
/// assert_eq!(" Martin-Löf", last);
|
|
|
|
|
/// }
|
|
|
|
|
/// assert_eq!("PER Martin-Löf", s);
|
|
|
|
|
/// ```
|
|
|
|
|
#[inline]
|
|
|
|
|
#[stable(feature = "str_split_at", since = "1.4.0")]
|
|
|
|
|
pub fn split_at_mut(&mut self, mid: usize) -> (&mut str, &mut str) {
|
2018-05-10 18:02:19 +00:00
|
|
|
|
// is_char_boundary checks that the index is in [0, .len()]
|
|
|
|
|
if self.is_char_boundary(mid) {
|
|
|
|
|
let len = self.len();
|
2019-02-07 16:55:25 +00:00
|
|
|
|
let ptr = self.as_mut_ptr();
|
2019-11-20 02:28:32 +00:00
|
|
|
|
// SAFETY: just checked that `mid` is on a char boundary
|
2018-05-10 18:02:19 +00:00
|
|
|
|
unsafe {
|
2019-12-22 22:42:04 +00:00
|
|
|
|
(
|
|
|
|
|
from_utf8_unchecked_mut(slice::from_raw_parts_mut(ptr, mid)),
|
|
|
|
|
from_utf8_unchecked_mut(slice::from_raw_parts_mut(ptr.add(mid), len - mid)),
|
|
|
|
|
)
|
2018-05-10 18:02:19 +00:00
|
|
|
|
}
|
|
|
|
|
} else {
|
|
|
|
|
slice_error_fail(self, 0, mid)
|
|
|
|
|
}
|
2018-04-05 17:00:48 +00:00
|
|
|
|
}
|
|
|
|
|
|
2018-04-07 19:56:02 +00:00
|
|
|
|
/// Returns an iterator over the [`char`]s of a string slice.
|
|
|
|
|
///
|
|
|
|
|
/// As a string slice consists of valid UTF-8, we can iterate through a
|
|
|
|
|
/// string slice by [`char`]. This method returns such an iterator.
|
|
|
|
|
///
|
|
|
|
|
/// It's important to remember that [`char`] represents a Unicode Scalar
|
|
|
|
|
/// Value, and may not match your idea of what a 'character' is. Iteration
|
|
|
|
|
/// over grapheme clusters may be what you actually want.
|
|
|
|
|
///
|
|
|
|
|
/// # Examples
|
|
|
|
|
///
|
|
|
|
|
/// Basic usage:
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// let word = "goodbye";
|
|
|
|
|
///
|
|
|
|
|
/// let count = word.chars().count();
|
|
|
|
|
/// assert_eq!(7, count);
|
|
|
|
|
///
|
|
|
|
|
/// let mut chars = word.chars();
|
|
|
|
|
///
|
|
|
|
|
/// assert_eq!(Some('g'), chars.next());
|
|
|
|
|
/// assert_eq!(Some('o'), chars.next());
|
|
|
|
|
/// assert_eq!(Some('o'), chars.next());
|
|
|
|
|
/// assert_eq!(Some('d'), chars.next());
|
|
|
|
|
/// assert_eq!(Some('b'), chars.next());
|
|
|
|
|
/// assert_eq!(Some('y'), chars.next());
|
|
|
|
|
/// assert_eq!(Some('e'), chars.next());
|
|
|
|
|
///
|
|
|
|
|
/// assert_eq!(None, chars.next());
|
|
|
|
|
/// ```
|
|
|
|
|
///
|
|
|
|
|
/// Remember, [`char`]s may not match your human intuition about characters:
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// let y = "y̆";
|
|
|
|
|
///
|
|
|
|
|
/// let mut chars = y.chars();
|
|
|
|
|
///
|
|
|
|
|
/// assert_eq!(Some('y'), chars.next()); // not 'y̆'
|
|
|
|
|
/// assert_eq!(Some('\u{0306}'), chars.next());
|
|
|
|
|
///
|
|
|
|
|
/// assert_eq!(None, chars.next());
|
|
|
|
|
/// ```
|
|
|
|
|
#[stable(feature = "rust1", since = "1.0.0")]
|
|
|
|
|
#[inline]
|
2019-04-18 23:37:12 +00:00
|
|
|
|
pub fn chars(&self) -> Chars<'_> {
|
2019-12-22 22:42:04 +00:00
|
|
|
|
Chars { iter: self.as_bytes().iter() }
|
2018-04-07 19:56:02 +00:00
|
|
|
|
}
|
2018-05-10 18:02:19 +00:00
|
|
|
|
|
2018-04-07 19:56:02 +00:00
|
|
|
|
/// Returns an iterator over the [`char`]s of a string slice, and their
|
|
|
|
|
/// positions.
|
|
|
|
|
///
|
|
|
|
|
/// As a string slice consists of valid UTF-8, we can iterate through a
|
|
|
|
|
/// string slice by [`char`]. This method returns an iterator of both
|
|
|
|
|
/// these [`char`]s, as well as their byte positions.
|
|
|
|
|
///
|
|
|
|
|
/// The iterator yields tuples. The position is first, the [`char`] is
|
|
|
|
|
/// second.
|
|
|
|
|
///
|
|
|
|
|
/// # Examples
|
|
|
|
|
///
|
|
|
|
|
/// Basic usage:
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// let word = "goodbye";
|
|
|
|
|
///
|
|
|
|
|
/// let count = word.char_indices().count();
|
|
|
|
|
/// assert_eq!(7, count);
|
|
|
|
|
///
|
|
|
|
|
/// let mut char_indices = word.char_indices();
|
|
|
|
|
///
|
|
|
|
|
/// assert_eq!(Some((0, 'g')), char_indices.next());
|
|
|
|
|
/// assert_eq!(Some((1, 'o')), char_indices.next());
|
|
|
|
|
/// assert_eq!(Some((2, 'o')), char_indices.next());
|
|
|
|
|
/// assert_eq!(Some((3, 'd')), char_indices.next());
|
|
|
|
|
/// assert_eq!(Some((4, 'b')), char_indices.next());
|
|
|
|
|
/// assert_eq!(Some((5, 'y')), char_indices.next());
|
|
|
|
|
/// assert_eq!(Some((6, 'e')), char_indices.next());
|
|
|
|
|
///
|
|
|
|
|
/// assert_eq!(None, char_indices.next());
|
|
|
|
|
/// ```
|
|
|
|
|
///
|
|
|
|
|
/// Remember, [`char`]s may not match your human intuition about characters:
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// let yes = "y̆es";
|
|
|
|
|
///
|
|
|
|
|
/// let mut char_indices = yes.char_indices();
|
|
|
|
|
///
|
|
|
|
|
/// assert_eq!(Some((0, 'y')), char_indices.next()); // not (0, 'y̆')
|
|
|
|
|
/// assert_eq!(Some((1, '\u{0306}')), char_indices.next());
|
|
|
|
|
///
|
|
|
|
|
/// // note the 3 here - the last character took up two bytes
|
|
|
|
|
/// assert_eq!(Some((3, 'e')), char_indices.next());
|
|
|
|
|
/// assert_eq!(Some((4, 's')), char_indices.next());
|
|
|
|
|
///
|
|
|
|
|
/// assert_eq!(None, char_indices.next());
|
|
|
|
|
/// ```
|
|
|
|
|
#[stable(feature = "rust1", since = "1.0.0")]
|
|
|
|
|
#[inline]
|
2019-04-18 23:37:12 +00:00
|
|
|
|
pub fn char_indices(&self) -> CharIndices<'_> {
|
2018-05-10 18:02:19 +00:00
|
|
|
|
CharIndices { front_offset: 0, iter: self.chars() }
|
2018-04-07 19:56:02 +00:00
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/// An iterator over the bytes of a string slice.
|
|
|
|
|
///
|
|
|
|
|
/// As a string slice consists of a sequence of bytes, we can iterate
|
|
|
|
|
/// through a string slice by byte. This method returns such an iterator.
|
|
|
|
|
///
|
|
|
|
|
/// # Examples
|
|
|
|
|
///
|
|
|
|
|
/// Basic usage:
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// let mut bytes = "bors".bytes();
|
|
|
|
|
///
|
|
|
|
|
/// assert_eq!(Some(b'b'), bytes.next());
|
|
|
|
|
/// assert_eq!(Some(b'o'), bytes.next());
|
|
|
|
|
/// assert_eq!(Some(b'r'), bytes.next());
|
|
|
|
|
/// assert_eq!(Some(b's'), bytes.next());
|
|
|
|
|
///
|
|
|
|
|
/// assert_eq!(None, bytes.next());
|
|
|
|
|
/// ```
|
|
|
|
|
#[stable(feature = "rust1", since = "1.0.0")]
|
|
|
|
|
#[inline]
|
2019-04-18 23:37:12 +00:00
|
|
|
|
pub fn bytes(&self) -> Bytes<'_> {
|
2018-05-10 18:02:19 +00:00
|
|
|
|
Bytes(self.as_bytes().iter().cloned())
|
2018-04-07 19:56:02 +00:00
|
|
|
|
}
|
|
|
|
|
|
2019-02-09 21:23:30 +00:00
|
|
|
|
/// Splits a string slice by whitespace.
|
2018-04-07 19:56:02 +00:00
|
|
|
|
///
|
|
|
|
|
/// The iterator returned will return string slices that are sub-slices of
|
|
|
|
|
/// the original string slice, separated by any amount of whitespace.
|
|
|
|
|
///
|
|
|
|
|
/// 'Whitespace' is defined according to the terms of the Unicode Derived
|
2018-05-05 04:33:20 +00:00
|
|
|
|
/// Core Property `White_Space`. If you only want to split on ASCII whitespace
|
|
|
|
|
/// instead, use [`split_ascii_whitespace`].
|
|
|
|
|
///
|
|
|
|
|
/// [`split_ascii_whitespace`]: #method.split_ascii_whitespace
|
2018-04-07 19:56:02 +00:00
|
|
|
|
///
|
|
|
|
|
/// # Examples
|
|
|
|
|
///
|
|
|
|
|
/// Basic usage:
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// let mut iter = "A few words".split_whitespace();
|
|
|
|
|
///
|
|
|
|
|
/// assert_eq!(Some("A"), iter.next());
|
|
|
|
|
/// assert_eq!(Some("few"), iter.next());
|
|
|
|
|
/// assert_eq!(Some("words"), iter.next());
|
|
|
|
|
///
|
|
|
|
|
/// assert_eq!(None, iter.next());
|
|
|
|
|
/// ```
|
|
|
|
|
///
|
|
|
|
|
/// All kinds of whitespace are considered:
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// let mut iter = " Mary had\ta\u{2009}little \n\t lamb".split_whitespace();
|
|
|
|
|
/// assert_eq!(Some("Mary"), iter.next());
|
|
|
|
|
/// assert_eq!(Some("had"), iter.next());
|
|
|
|
|
/// assert_eq!(Some("a"), iter.next());
|
|
|
|
|
/// assert_eq!(Some("little"), iter.next());
|
|
|
|
|
/// assert_eq!(Some("lamb"), iter.next());
|
|
|
|
|
///
|
|
|
|
|
/// assert_eq!(None, iter.next());
|
|
|
|
|
/// ```
|
|
|
|
|
#[stable(feature = "split_whitespace", since = "1.1.0")]
|
|
|
|
|
#[inline]
|
2019-04-18 23:37:12 +00:00
|
|
|
|
pub fn split_whitespace(&self) -> SplitWhitespace<'_> {
|
2018-05-10 18:02:19 +00:00
|
|
|
|
SplitWhitespace { inner: self.split(IsWhitespace).filter(IsNotEmpty) }
|
2018-04-07 19:56:02 +00:00
|
|
|
|
}
|
|
|
|
|
|
2019-02-09 21:23:30 +00:00
|
|
|
|
/// Splits a string slice by ASCII whitespace.
|
2018-05-05 04:33:20 +00:00
|
|
|
|
///
|
|
|
|
|
/// The iterator returned will return string slices that are sub-slices of
|
|
|
|
|
/// the original string slice, separated by any amount of ASCII whitespace.
|
|
|
|
|
///
|
|
|
|
|
/// To split by Unicode `Whitespace` instead, use [`split_whitespace`].
|
|
|
|
|
///
|
|
|
|
|
/// [`split_whitespace`]: #method.split_whitespace
|
|
|
|
|
///
|
|
|
|
|
/// # Examples
|
|
|
|
|
///
|
|
|
|
|
/// Basic usage:
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// let mut iter = "A few words".split_ascii_whitespace();
|
|
|
|
|
///
|
|
|
|
|
/// assert_eq!(Some("A"), iter.next());
|
|
|
|
|
/// assert_eq!(Some("few"), iter.next());
|
|
|
|
|
/// assert_eq!(Some("words"), iter.next());
|
|
|
|
|
///
|
|
|
|
|
/// assert_eq!(None, iter.next());
|
|
|
|
|
/// ```
|
|
|
|
|
///
|
|
|
|
|
/// All kinds of ASCII whitespace are considered:
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
2019-04-12 19:57:42 +00:00
|
|
|
|
/// let mut iter = " Mary had\ta little \n\t lamb".split_ascii_whitespace();
|
2018-05-05 04:33:20 +00:00
|
|
|
|
/// assert_eq!(Some("Mary"), iter.next());
|
|
|
|
|
/// assert_eq!(Some("had"), iter.next());
|
|
|
|
|
/// assert_eq!(Some("a"), iter.next());
|
|
|
|
|
/// assert_eq!(Some("little"), iter.next());
|
|
|
|
|
/// assert_eq!(Some("lamb"), iter.next());
|
|
|
|
|
///
|
|
|
|
|
/// assert_eq!(None, iter.next());
|
|
|
|
|
/// ```
|
2019-02-01 10:24:12 +00:00
|
|
|
|
#[stable(feature = "split_ascii_whitespace", since = "1.34.0")]
|
2018-05-05 04:33:20 +00:00
|
|
|
|
#[inline]
|
2019-04-18 23:37:12 +00:00
|
|
|
|
pub fn split_ascii_whitespace(&self) -> SplitAsciiWhitespace<'_> {
|
2019-12-22 22:42:04 +00:00
|
|
|
|
let inner =
|
|
|
|
|
self.as_bytes().split(IsAsciiWhitespace).filter(BytesIsNotEmpty).map(UnsafeBytesToStr);
|
2018-05-05 04:33:20 +00:00
|
|
|
|
SplitAsciiWhitespace { inner }
|
|
|
|
|
}
|
|
|
|
|
|
2018-04-07 19:56:02 +00:00
|
|
|
|
/// An iterator over the lines of a string, as string slices.
|
|
|
|
|
///
|
|
|
|
|
/// Lines are ended with either a newline (`\n`) or a carriage return with
|
|
|
|
|
/// a line feed (`\r\n`).
|
|
|
|
|
///
|
|
|
|
|
/// The final line ending is optional.
|
|
|
|
|
///
|
|
|
|
|
/// # Examples
|
|
|
|
|
///
|
|
|
|
|
/// Basic usage:
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// let text = "foo\r\nbar\n\nbaz\n";
|
|
|
|
|
/// let mut lines = text.lines();
|
|
|
|
|
///
|
|
|
|
|
/// assert_eq!(Some("foo"), lines.next());
|
|
|
|
|
/// assert_eq!(Some("bar"), lines.next());
|
|
|
|
|
/// assert_eq!(Some(""), lines.next());
|
|
|
|
|
/// assert_eq!(Some("baz"), lines.next());
|
|
|
|
|
///
|
|
|
|
|
/// assert_eq!(None, lines.next());
|
|
|
|
|
/// ```
|
|
|
|
|
///
|
|
|
|
|
/// The final line ending isn't required:
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// let text = "foo\nbar\n\r\nbaz";
|
|
|
|
|
/// let mut lines = text.lines();
|
|
|
|
|
///
|
|
|
|
|
/// assert_eq!(Some("foo"), lines.next());
|
|
|
|
|
/// assert_eq!(Some("bar"), lines.next());
|
|
|
|
|
/// assert_eq!(Some(""), lines.next());
|
|
|
|
|
/// assert_eq!(Some("baz"), lines.next());
|
|
|
|
|
///
|
|
|
|
|
/// assert_eq!(None, lines.next());
|
|
|
|
|
/// ```
|
|
|
|
|
#[stable(feature = "rust1", since = "1.0.0")]
|
|
|
|
|
#[inline]
|
2019-04-18 23:37:12 +00:00
|
|
|
|
pub fn lines(&self) -> Lines<'_> {
|
2018-05-10 18:02:19 +00:00
|
|
|
|
Lines(self.split_terminator('\n').map(LinesAnyMap))
|
2018-04-07 19:56:02 +00:00
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/// An iterator over the lines of a string.
|
|
|
|
|
#[stable(feature = "rust1", since = "1.0.0")]
|
|
|
|
|
#[rustc_deprecated(since = "1.4.0", reason = "use lines() instead now")]
|
|
|
|
|
#[inline]
|
|
|
|
|
#[allow(deprecated)]
|
2019-04-18 23:37:12 +00:00
|
|
|
|
pub fn lines_any(&self) -> LinesAny<'_> {
|
2018-05-10 18:02:19 +00:00
|
|
|
|
LinesAny(self.lines())
|
2018-04-07 19:56:02 +00:00
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/// Returns an iterator of `u16` over the string encoded as UTF-16.
|
|
|
|
|
///
|
|
|
|
|
/// # Examples
|
|
|
|
|
///
|
|
|
|
|
/// Basic usage:
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// let text = "Zażółć gęślą jaźń";
|
|
|
|
|
///
|
|
|
|
|
/// let utf8_len = text.len();
|
|
|
|
|
/// let utf16_len = text.encode_utf16().count();
|
|
|
|
|
///
|
|
|
|
|
/// assert!(utf16_len <= utf8_len);
|
|
|
|
|
/// ```
|
|
|
|
|
#[stable(feature = "encode_utf16", since = "1.8.0")]
|
2019-04-18 23:37:12 +00:00
|
|
|
|
pub fn encode_utf16(&self) -> EncodeUtf16<'_> {
|
2018-05-10 18:02:19 +00:00
|
|
|
|
EncodeUtf16 { chars: self.chars(), extra: 0 }
|
2018-04-07 19:56:02 +00:00
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/// Returns `true` if the given pattern matches a sub-slice of
|
|
|
|
|
/// this string slice.
|
|
|
|
|
///
|
|
|
|
|
/// Returns `false` if it does not.
|
|
|
|
|
///
|
|
|
|
|
/// # Examples
|
|
|
|
|
///
|
|
|
|
|
/// Basic usage:
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// let bananas = "bananas";
|
|
|
|
|
///
|
|
|
|
|
/// assert!(bananas.contains("nana"));
|
|
|
|
|
/// assert!(!bananas.contains("apples"));
|
|
|
|
|
/// ```
|
|
|
|
|
#[stable(feature = "rust1", since = "1.0.0")]
|
|
|
|
|
#[inline]
|
|
|
|
|
pub fn contains<'a, P: Pattern<'a>>(&'a self, pat: P) -> bool {
|
2018-05-10 18:02:19 +00:00
|
|
|
|
pat.is_contained_in(self)
|
2018-04-07 19:56:02 +00:00
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/// Returns `true` if the given pattern matches a prefix of this
|
|
|
|
|
/// string slice.
|
|
|
|
|
///
|
|
|
|
|
/// Returns `false` if it does not.
|
|
|
|
|
///
|
|
|
|
|
/// # Examples
|
|
|
|
|
///
|
|
|
|
|
/// Basic usage:
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// let bananas = "bananas";
|
|
|
|
|
///
|
|
|
|
|
/// assert!(bananas.starts_with("bana"));
|
|
|
|
|
/// assert!(!bananas.starts_with("nana"));
|
|
|
|
|
/// ```
|
|
|
|
|
#[stable(feature = "rust1", since = "1.0.0")]
|
|
|
|
|
pub fn starts_with<'a, P: Pattern<'a>>(&'a self, pat: P) -> bool {
|
2018-05-10 18:02:19 +00:00
|
|
|
|
pat.is_prefix_of(self)
|
2018-04-07 19:56:02 +00:00
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/// Returns `true` if the given pattern matches a suffix of this
|
|
|
|
|
/// string slice.
|
|
|
|
|
///
|
|
|
|
|
/// Returns `false` if it does not.
|
|
|
|
|
///
|
|
|
|
|
/// # Examples
|
|
|
|
|
///
|
|
|
|
|
/// Basic usage:
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// let bananas = "bananas";
|
|
|
|
|
///
|
|
|
|
|
/// assert!(bananas.ends_with("anas"));
|
|
|
|
|
/// assert!(!bananas.ends_with("nana"));
|
|
|
|
|
/// ```
|
|
|
|
|
#[stable(feature = "rust1", since = "1.0.0")]
|
2019-07-31 19:00:35 +00:00
|
|
|
|
pub fn ends_with<'a, P>(&'a self, pat: P) -> bool
|
|
|
|
|
where
|
|
|
|
|
P: Pattern<'a, Searcher: ReverseSearcher<'a>>,
|
2018-04-07 19:56:02 +00:00
|
|
|
|
{
|
2018-05-10 18:02:19 +00:00
|
|
|
|
pat.is_suffix_of(self)
|
2018-04-07 19:56:02 +00:00
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/// Returns the byte index of the first character of this string slice that
|
|
|
|
|
/// matches the pattern.
|
|
|
|
|
///
|
|
|
|
|
/// Returns [`None`] if the pattern doesn't match.
|
|
|
|
|
///
|
|
|
|
|
/// The pattern can be a `&str`, [`char`], or a closure that determines if
|
|
|
|
|
/// a character matches.
|
|
|
|
|
///
|
|
|
|
|
/// [`None`]: option/enum.Option.html#variant.None
|
|
|
|
|
///
|
|
|
|
|
/// # Examples
|
|
|
|
|
///
|
|
|
|
|
/// Simple patterns:
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// let s = "Löwe 老虎 Léopard";
|
|
|
|
|
///
|
|
|
|
|
/// assert_eq!(s.find('L'), Some(0));
|
|
|
|
|
/// assert_eq!(s.find('é'), Some(14));
|
|
|
|
|
/// assert_eq!(s.find("Léopard"), Some(13));
|
|
|
|
|
/// ```
|
|
|
|
|
///
|
|
|
|
|
/// More complex patterns using point-free style and closures:
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// let s = "Löwe 老虎 Léopard";
|
|
|
|
|
///
|
|
|
|
|
/// assert_eq!(s.find(char::is_whitespace), Some(5));
|
|
|
|
|
/// assert_eq!(s.find(char::is_lowercase), Some(1));
|
|
|
|
|
/// assert_eq!(s.find(|c: char| c.is_whitespace() || c.is_lowercase()), Some(1));
|
|
|
|
|
/// assert_eq!(s.find(|c: char| (c < 'o') && (c > 'a')), Some(4));
|
|
|
|
|
/// ```
|
|
|
|
|
///
|
|
|
|
|
/// Not finding the pattern:
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// let s = "Löwe 老虎 Léopard";
|
|
|
|
|
/// let x: &[_] = &['1', '2'];
|
|
|
|
|
///
|
|
|
|
|
/// assert_eq!(s.find(x), None);
|
|
|
|
|
/// ```
|
|
|
|
|
#[stable(feature = "rust1", since = "1.0.0")]
|
|
|
|
|
#[inline]
|
|
|
|
|
pub fn find<'a, P: Pattern<'a>>(&'a self, pat: P) -> Option<usize> {
|
2018-05-10 18:02:19 +00:00
|
|
|
|
pat.into_searcher(self).next_match().map(|(i, _)| i)
|
2018-04-07 19:56:02 +00:00
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/// Returns the byte index of the last character of this string slice that
|
|
|
|
|
/// matches the pattern.
|
|
|
|
|
///
|
|
|
|
|
/// Returns [`None`] if the pattern doesn't match.
|
|
|
|
|
///
|
|
|
|
|
/// The pattern can be a `&str`, [`char`], or a closure that determines if
|
|
|
|
|
/// a character matches.
|
|
|
|
|
///
|
|
|
|
|
/// [`None`]: option/enum.Option.html#variant.None
|
|
|
|
|
///
|
|
|
|
|
/// # Examples
|
|
|
|
|
///
|
|
|
|
|
/// Simple patterns:
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// let s = "Löwe 老虎 Léopard";
|
|
|
|
|
///
|
|
|
|
|
/// assert_eq!(s.rfind('L'), Some(13));
|
|
|
|
|
/// assert_eq!(s.rfind('é'), Some(14));
|
|
|
|
|
/// ```
|
|
|
|
|
///
|
|
|
|
|
/// More complex patterns with closures:
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// let s = "Löwe 老虎 Léopard";
|
|
|
|
|
///
|
|
|
|
|
/// assert_eq!(s.rfind(char::is_whitespace), Some(12));
|
|
|
|
|
/// assert_eq!(s.rfind(char::is_lowercase), Some(20));
|
|
|
|
|
/// ```
|
|
|
|
|
///
|
|
|
|
|
/// Not finding the pattern:
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// let s = "Löwe 老虎 Léopard";
|
|
|
|
|
/// let x: &[_] = &['1', '2'];
|
|
|
|
|
///
|
|
|
|
|
/// assert_eq!(s.rfind(x), None);
|
|
|
|
|
/// ```
|
|
|
|
|
#[stable(feature = "rust1", since = "1.0.0")]
|
|
|
|
|
#[inline]
|
2019-07-31 19:00:35 +00:00
|
|
|
|
pub fn rfind<'a, P>(&'a self, pat: P) -> Option<usize>
|
|
|
|
|
where
|
|
|
|
|
P: Pattern<'a, Searcher: ReverseSearcher<'a>>,
|
2018-04-07 19:56:02 +00:00
|
|
|
|
{
|
2018-05-10 18:02:19 +00:00
|
|
|
|
pat.into_searcher(self).next_match_back().map(|(i, _)| i)
|
2018-04-07 19:56:02 +00:00
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/// An iterator over substrings of this string slice, separated by
|
|
|
|
|
/// characters matched by a pattern.
|
|
|
|
|
///
|
2019-02-04 23:26:33 +00:00
|
|
|
|
/// The pattern can be any type that implements the Pattern trait. Notable
|
|
|
|
|
/// examples are `&str`, [`char`], and closures that determines the split.
|
2018-04-07 19:56:02 +00:00
|
|
|
|
///
|
|
|
|
|
/// # Iterator behavior
|
|
|
|
|
///
|
|
|
|
|
/// The returned iterator will be a [`DoubleEndedIterator`] if the pattern
|
|
|
|
|
/// allows a reverse search and forward/reverse search yields the same
|
2019-02-09 22:16:58 +00:00
|
|
|
|
/// elements. This is true for, e.g., [`char`], but not for `&str`.
|
2018-04-07 19:56:02 +00:00
|
|
|
|
///
|
|
|
|
|
/// [`DoubleEndedIterator`]: iter/trait.DoubleEndedIterator.html
|
|
|
|
|
///
|
|
|
|
|
/// If the pattern allows a reverse search but its results might differ
|
|
|
|
|
/// from a forward search, the [`rsplit`] method can be used.
|
|
|
|
|
///
|
|
|
|
|
/// [`rsplit`]: #method.rsplit
|
|
|
|
|
///
|
|
|
|
|
/// # Examples
|
|
|
|
|
///
|
|
|
|
|
/// Simple patterns:
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// let v: Vec<&str> = "Mary had a little lamb".split(' ').collect();
|
|
|
|
|
/// assert_eq!(v, ["Mary", "had", "a", "little", "lamb"]);
|
|
|
|
|
///
|
|
|
|
|
/// let v: Vec<&str> = "".split('X').collect();
|
|
|
|
|
/// assert_eq!(v, [""]);
|
|
|
|
|
///
|
|
|
|
|
/// let v: Vec<&str> = "lionXXtigerXleopard".split('X').collect();
|
|
|
|
|
/// assert_eq!(v, ["lion", "", "tiger", "leopard"]);
|
|
|
|
|
///
|
|
|
|
|
/// let v: Vec<&str> = "lion::tiger::leopard".split("::").collect();
|
|
|
|
|
/// assert_eq!(v, ["lion", "tiger", "leopard"]);
|
|
|
|
|
///
|
|
|
|
|
/// let v: Vec<&str> = "abc1def2ghi".split(char::is_numeric).collect();
|
|
|
|
|
/// assert_eq!(v, ["abc", "def", "ghi"]);
|
|
|
|
|
///
|
|
|
|
|
/// let v: Vec<&str> = "lionXtigerXleopard".split(char::is_uppercase).collect();
|
|
|
|
|
/// assert_eq!(v, ["lion", "tiger", "leopard"]);
|
|
|
|
|
/// ```
|
|
|
|
|
///
|
|
|
|
|
/// A more complex pattern, using a closure:
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// let v: Vec<&str> = "abc1defXghi".split(|c| c == '1' || c == 'X').collect();
|
|
|
|
|
/// assert_eq!(v, ["abc", "def", "ghi"]);
|
|
|
|
|
/// ```
|
|
|
|
|
///
|
|
|
|
|
/// If a string contains multiple contiguous separators, you will end up
|
|
|
|
|
/// with empty strings in the output:
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// let x = "||||a||b|c".to_string();
|
|
|
|
|
/// let d: Vec<_> = x.split('|').collect();
|
|
|
|
|
///
|
|
|
|
|
/// assert_eq!(d, &["", "", "", "", "a", "", "b", "c"]);
|
|
|
|
|
/// ```
|
|
|
|
|
///
|
|
|
|
|
/// Contiguous separators are separated by the empty string.
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// let x = "(///)".to_string();
|
|
|
|
|
/// let d: Vec<_> = x.split('/').collect();
|
|
|
|
|
///
|
|
|
|
|
/// assert_eq!(d, &["(", "", "", ")"]);
|
|
|
|
|
/// ```
|
|
|
|
|
///
|
|
|
|
|
/// Separators at the start or end of a string are neighbored
|
|
|
|
|
/// by empty strings.
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// let d: Vec<_> = "010".split("0").collect();
|
|
|
|
|
/// assert_eq!(d, &["", "1", ""]);
|
|
|
|
|
/// ```
|
|
|
|
|
///
|
|
|
|
|
/// When the empty string is used as a separator, it separates
|
|
|
|
|
/// every character in the string, along with the beginning
|
|
|
|
|
/// and end of the string.
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// let f: Vec<_> = "rust".split("").collect();
|
|
|
|
|
/// assert_eq!(f, &["", "r", "u", "s", "t", ""]);
|
|
|
|
|
/// ```
|
|
|
|
|
///
|
|
|
|
|
/// Contiguous separators can lead to possibly surprising behavior
|
|
|
|
|
/// when whitespace is used as the separator. This code is correct:
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// let x = " a b c".to_string();
|
|
|
|
|
/// let d: Vec<_> = x.split(' ').collect();
|
|
|
|
|
///
|
|
|
|
|
/// assert_eq!(d, &["", "", "", "", "a", "", "b", "c"]);
|
|
|
|
|
/// ```
|
|
|
|
|
///
|
|
|
|
|
/// It does _not_ give you:
|
|
|
|
|
///
|
|
|
|
|
/// ```,ignore
|
|
|
|
|
/// assert_eq!(d, &["a", "b", "c"]);
|
|
|
|
|
/// ```
|
|
|
|
|
///
|
|
|
|
|
/// Use [`split_whitespace`] for this behavior.
|
|
|
|
|
///
|
|
|
|
|
/// [`split_whitespace`]: #method.split_whitespace
|
|
|
|
|
#[stable(feature = "rust1", since = "1.0.0")]
|
|
|
|
|
#[inline]
|
|
|
|
|
pub fn split<'a, P: Pattern<'a>>(&'a self, pat: P) -> Split<'a, P> {
|
2018-05-10 18:02:19 +00:00
|
|
|
|
Split(SplitInternal {
|
|
|
|
|
start: 0,
|
|
|
|
|
end: self.len(),
|
|
|
|
|
matcher: pat.into_searcher(self),
|
|
|
|
|
allow_trailing_empty: true,
|
|
|
|
|
finished: false,
|
|
|
|
|
})
|
2018-04-07 19:56:02 +00:00
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/// An iterator over substrings of the given string slice, separated by
|
|
|
|
|
/// characters matched by a pattern and yielded in reverse order.
|
|
|
|
|
///
|
2019-02-04 23:26:33 +00:00
|
|
|
|
/// The pattern can be any type that implements the Pattern trait. Notable
|
|
|
|
|
/// examples are `&str`, [`char`], and closures that determines the split.
|
2018-04-07 19:56:02 +00:00
|
|
|
|
///
|
|
|
|
|
/// # Iterator behavior
|
|
|
|
|
///
|
|
|
|
|
/// The returned iterator requires that the pattern supports a reverse
|
|
|
|
|
/// search, and it will be a [`DoubleEndedIterator`] if a forward/reverse
|
|
|
|
|
/// search yields the same elements.
|
|
|
|
|
///
|
|
|
|
|
/// [`DoubleEndedIterator`]: iter/trait.DoubleEndedIterator.html
|
|
|
|
|
///
|
|
|
|
|
/// For iterating from the front, the [`split`] method can be used.
|
|
|
|
|
///
|
|
|
|
|
/// [`split`]: #method.split
|
|
|
|
|
///
|
|
|
|
|
/// # Examples
|
|
|
|
|
///
|
|
|
|
|
/// Simple patterns:
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// let v: Vec<&str> = "Mary had a little lamb".rsplit(' ').collect();
|
|
|
|
|
/// assert_eq!(v, ["lamb", "little", "a", "had", "Mary"]);
|
|
|
|
|
///
|
|
|
|
|
/// let v: Vec<&str> = "".rsplit('X').collect();
|
|
|
|
|
/// assert_eq!(v, [""]);
|
|
|
|
|
///
|
|
|
|
|
/// let v: Vec<&str> = "lionXXtigerXleopard".rsplit('X').collect();
|
|
|
|
|
/// assert_eq!(v, ["leopard", "tiger", "", "lion"]);
|
|
|
|
|
///
|
|
|
|
|
/// let v: Vec<&str> = "lion::tiger::leopard".rsplit("::").collect();
|
|
|
|
|
/// assert_eq!(v, ["leopard", "tiger", "lion"]);
|
|
|
|
|
/// ```
|
|
|
|
|
///
|
|
|
|
|
/// A more complex pattern, using a closure:
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// let v: Vec<&str> = "abc1defXghi".rsplit(|c| c == '1' || c == 'X').collect();
|
|
|
|
|
/// assert_eq!(v, ["ghi", "def", "abc"]);
|
|
|
|
|
/// ```
|
|
|
|
|
#[stable(feature = "rust1", since = "1.0.0")]
|
|
|
|
|
#[inline]
|
2019-07-31 19:00:35 +00:00
|
|
|
|
pub fn rsplit<'a, P>(&'a self, pat: P) -> RSplit<'a, P>
|
|
|
|
|
where
|
|
|
|
|
P: Pattern<'a, Searcher: ReverseSearcher<'a>>,
|
2018-04-07 19:56:02 +00:00
|
|
|
|
{
|
2018-05-10 18:02:19 +00:00
|
|
|
|
RSplit(self.split(pat).0)
|
2018-04-07 19:56:02 +00:00
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/// An iterator over substrings of the given string slice, separated by
|
|
|
|
|
/// characters matched by a pattern.
|
|
|
|
|
///
|
2019-02-04 23:26:33 +00:00
|
|
|
|
/// The pattern can be any type that implements the Pattern trait. Notable
|
|
|
|
|
/// examples are `&str`, [`char`], and closures that determines the split.
|
2018-04-07 19:56:02 +00:00
|
|
|
|
///
|
|
|
|
|
/// Equivalent to [`split`], except that the trailing substring
|
|
|
|
|
/// is skipped if empty.
|
|
|
|
|
///
|
|
|
|
|
/// [`split`]: #method.split
|
|
|
|
|
///
|
|
|
|
|
/// This method can be used for string data that is _terminated_,
|
|
|
|
|
/// rather than _separated_ by a pattern.
|
|
|
|
|
///
|
|
|
|
|
/// # Iterator behavior
|
|
|
|
|
///
|
|
|
|
|
/// The returned iterator will be a [`DoubleEndedIterator`] if the pattern
|
|
|
|
|
/// allows a reverse search and forward/reverse search yields the same
|
2019-02-09 22:16:58 +00:00
|
|
|
|
/// elements. This is true for, e.g., [`char`], but not for `&str`.
|
2018-04-07 19:56:02 +00:00
|
|
|
|
///
|
|
|
|
|
/// [`DoubleEndedIterator`]: iter/trait.DoubleEndedIterator.html
|
|
|
|
|
///
|
|
|
|
|
/// If the pattern allows a reverse search but its results might differ
|
|
|
|
|
/// from a forward search, the [`rsplit_terminator`] method can be used.
|
|
|
|
|
///
|
|
|
|
|
/// [`rsplit_terminator`]: #method.rsplit_terminator
|
|
|
|
|
///
|
|
|
|
|
/// # Examples
|
|
|
|
|
///
|
|
|
|
|
/// Basic usage:
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// let v: Vec<&str> = "A.B.".split_terminator('.').collect();
|
|
|
|
|
/// assert_eq!(v, ["A", "B"]);
|
|
|
|
|
///
|
|
|
|
|
/// let v: Vec<&str> = "A..B..".split_terminator(".").collect();
|
|
|
|
|
/// assert_eq!(v, ["A", "", "B", ""]);
|
|
|
|
|
/// ```
|
|
|
|
|
#[stable(feature = "rust1", since = "1.0.0")]
|
|
|
|
|
#[inline]
|
|
|
|
|
pub fn split_terminator<'a, P: Pattern<'a>>(&'a self, pat: P) -> SplitTerminator<'a, P> {
|
2019-12-22 22:42:04 +00:00
|
|
|
|
SplitTerminator(SplitInternal { allow_trailing_empty: false, ..self.split(pat).0 })
|
2018-04-07 19:56:02 +00:00
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/// An iterator over substrings of `self`, separated by characters
|
|
|
|
|
/// matched by a pattern and yielded in reverse order.
|
|
|
|
|
///
|
2019-02-04 23:26:33 +00:00
|
|
|
|
/// The pattern can be any type that implements the Pattern trait. Notable
|
|
|
|
|
/// examples are `&str`, [`char`], and closures that determines the split.
|
2018-04-07 19:56:02 +00:00
|
|
|
|
/// Additional libraries might provide more complex patterns like
|
|
|
|
|
/// regular expressions.
|
|
|
|
|
///
|
|
|
|
|
/// Equivalent to [`split`], except that the trailing substring is
|
|
|
|
|
/// skipped if empty.
|
|
|
|
|
///
|
|
|
|
|
/// [`split`]: #method.split
|
|
|
|
|
///
|
|
|
|
|
/// This method can be used for string data that is _terminated_,
|
|
|
|
|
/// rather than _separated_ by a pattern.
|
|
|
|
|
///
|
|
|
|
|
/// # Iterator behavior
|
|
|
|
|
///
|
|
|
|
|
/// The returned iterator requires that the pattern supports a
|
|
|
|
|
/// reverse search, and it will be double ended if a forward/reverse
|
|
|
|
|
/// search yields the same elements.
|
|
|
|
|
///
|
|
|
|
|
/// For iterating from the front, the [`split_terminator`] method can be
|
|
|
|
|
/// used.
|
|
|
|
|
///
|
|
|
|
|
/// [`split_terminator`]: #method.split_terminator
|
|
|
|
|
///
|
|
|
|
|
/// # Examples
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// let v: Vec<&str> = "A.B.".rsplit_terminator('.').collect();
|
|
|
|
|
/// assert_eq!(v, ["B", "A"]);
|
|
|
|
|
///
|
|
|
|
|
/// let v: Vec<&str> = "A..B..".rsplit_terminator(".").collect();
|
|
|
|
|
/// assert_eq!(v, ["", "B", "", "A"]);
|
|
|
|
|
/// ```
|
|
|
|
|
#[stable(feature = "rust1", since = "1.0.0")]
|
|
|
|
|
#[inline]
|
2019-07-31 19:00:35 +00:00
|
|
|
|
pub fn rsplit_terminator<'a, P>(&'a self, pat: P) -> RSplitTerminator<'a, P>
|
|
|
|
|
where
|
|
|
|
|
P: Pattern<'a, Searcher: ReverseSearcher<'a>>,
|
2018-04-07 19:56:02 +00:00
|
|
|
|
{
|
2018-05-10 18:02:19 +00:00
|
|
|
|
RSplitTerminator(self.split_terminator(pat).0)
|
2018-04-07 19:56:02 +00:00
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/// An iterator over substrings of the given string slice, separated by a
|
|
|
|
|
/// pattern, restricted to returning at most `n` items.
|
|
|
|
|
///
|
|
|
|
|
/// If `n` substrings are returned, the last substring (the `n`th substring)
|
|
|
|
|
/// will contain the remainder of the string.
|
|
|
|
|
///
|
2019-02-04 23:26:33 +00:00
|
|
|
|
/// The pattern can be any type that implements the Pattern trait. Notable
|
|
|
|
|
/// examples are `&str`, [`char`], and closures that determines the split.
|
2018-04-07 19:56:02 +00:00
|
|
|
|
///
|
|
|
|
|
/// # Iterator behavior
|
|
|
|
|
///
|
|
|
|
|
/// The returned iterator will not be double ended, because it is
|
|
|
|
|
/// not efficient to support.
|
|
|
|
|
///
|
|
|
|
|
/// If the pattern allows a reverse search, the [`rsplitn`] method can be
|
|
|
|
|
/// used.
|
|
|
|
|
///
|
|
|
|
|
/// [`rsplitn`]: #method.rsplitn
|
|
|
|
|
///
|
|
|
|
|
/// # Examples
|
|
|
|
|
///
|
|
|
|
|
/// Simple patterns:
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// let v: Vec<&str> = "Mary had a little lambda".splitn(3, ' ').collect();
|
|
|
|
|
/// assert_eq!(v, ["Mary", "had", "a little lambda"]);
|
|
|
|
|
///
|
|
|
|
|
/// let v: Vec<&str> = "lionXXtigerXleopard".splitn(3, "X").collect();
|
|
|
|
|
/// assert_eq!(v, ["lion", "", "tigerXleopard"]);
|
|
|
|
|
///
|
|
|
|
|
/// let v: Vec<&str> = "abcXdef".splitn(1, 'X').collect();
|
|
|
|
|
/// assert_eq!(v, ["abcXdef"]);
|
|
|
|
|
///
|
|
|
|
|
/// let v: Vec<&str> = "".splitn(1, 'X').collect();
|
|
|
|
|
/// assert_eq!(v, [""]);
|
|
|
|
|
/// ```
|
|
|
|
|
///
|
|
|
|
|
/// A more complex pattern, using a closure:
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// let v: Vec<&str> = "abc1defXghi".splitn(2, |c| c == '1' || c == 'X').collect();
|
|
|
|
|
/// assert_eq!(v, ["abc", "defXghi"]);
|
|
|
|
|
/// ```
|
|
|
|
|
#[stable(feature = "rust1", since = "1.0.0")]
|
|
|
|
|
#[inline]
|
|
|
|
|
pub fn splitn<'a, P: Pattern<'a>>(&'a self, n: usize, pat: P) -> SplitN<'a, P> {
|
2019-12-22 22:42:04 +00:00
|
|
|
|
SplitN(SplitNInternal { iter: self.split(pat).0, count: n })
|
2018-04-07 19:56:02 +00:00
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/// An iterator over substrings of this string slice, separated by a
|
|
|
|
|
/// pattern, starting from the end of the string, restricted to returning
|
|
|
|
|
/// at most `n` items.
|
|
|
|
|
///
|
|
|
|
|
/// If `n` substrings are returned, the last substring (the `n`th substring)
|
|
|
|
|
/// will contain the remainder of the string.
|
|
|
|
|
///
|
2019-02-04 23:26:33 +00:00
|
|
|
|
/// The pattern can be any type that implements the Pattern trait. Notable
|
|
|
|
|
/// examples are `&str`, [`char`], and closures that determines the split.
|
2018-04-07 19:56:02 +00:00
|
|
|
|
///
|
|
|
|
|
/// # Iterator behavior
|
|
|
|
|
///
|
|
|
|
|
/// The returned iterator will not be double ended, because it is not
|
|
|
|
|
/// efficient to support.
|
|
|
|
|
///
|
|
|
|
|
/// For splitting from the front, the [`splitn`] method can be used.
|
|
|
|
|
///
|
|
|
|
|
/// [`splitn`]: #method.splitn
|
|
|
|
|
///
|
|
|
|
|
/// # Examples
|
|
|
|
|
///
|
|
|
|
|
/// Simple patterns:
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// let v: Vec<&str> = "Mary had a little lamb".rsplitn(3, ' ').collect();
|
|
|
|
|
/// assert_eq!(v, ["lamb", "little", "Mary had a"]);
|
|
|
|
|
///
|
|
|
|
|
/// let v: Vec<&str> = "lionXXtigerXleopard".rsplitn(3, 'X').collect();
|
|
|
|
|
/// assert_eq!(v, ["leopard", "tiger", "lionX"]);
|
|
|
|
|
///
|
|
|
|
|
/// let v: Vec<&str> = "lion::tiger::leopard".rsplitn(2, "::").collect();
|
|
|
|
|
/// assert_eq!(v, ["leopard", "lion::tiger"]);
|
|
|
|
|
/// ```
|
|
|
|
|
///
|
|
|
|
|
/// A more complex pattern, using a closure:
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// let v: Vec<&str> = "abc1defXghi".rsplitn(2, |c| c == '1' || c == 'X').collect();
|
|
|
|
|
/// assert_eq!(v, ["ghi", "abc1def"]);
|
|
|
|
|
/// ```
|
|
|
|
|
#[stable(feature = "rust1", since = "1.0.0")]
|
|
|
|
|
#[inline]
|
2019-07-31 19:00:35 +00:00
|
|
|
|
pub fn rsplitn<'a, P>(&'a self, n: usize, pat: P) -> RSplitN<'a, P>
|
|
|
|
|
where
|
|
|
|
|
P: Pattern<'a, Searcher: ReverseSearcher<'a>>,
|
2018-04-07 19:56:02 +00:00
|
|
|
|
{
|
2018-05-10 18:02:19 +00:00
|
|
|
|
RSplitN(self.splitn(n, pat).0)
|
2018-04-07 19:56:02 +00:00
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/// An iterator over the disjoint matches of a pattern within the given string
|
|
|
|
|
/// slice.
|
|
|
|
|
///
|
2019-12-03 22:31:41 +00:00
|
|
|
|
/// The pattern can be a `&str`, [`char`], or a closure that determines if
|
|
|
|
|
/// a character matches.
|
2018-04-07 19:56:02 +00:00
|
|
|
|
///
|
|
|
|
|
/// # Iterator behavior
|
|
|
|
|
///
|
|
|
|
|
/// The returned iterator will be a [`DoubleEndedIterator`] if the pattern
|
|
|
|
|
/// allows a reverse search and forward/reverse search yields the same
|
2019-02-09 22:16:58 +00:00
|
|
|
|
/// elements. This is true for, e.g., [`char`], but not for `&str`.
|
2018-04-07 19:56:02 +00:00
|
|
|
|
///
|
|
|
|
|
/// [`DoubleEndedIterator`]: iter/trait.DoubleEndedIterator.html
|
|
|
|
|
///
|
|
|
|
|
/// If the pattern allows a reverse search but its results might differ
|
|
|
|
|
/// from a forward search, the [`rmatches`] method can be used.
|
|
|
|
|
///
|
|
|
|
|
/// [`rmatches`]: #method.rmatches
|
|
|
|
|
///
|
|
|
|
|
/// # Examples
|
|
|
|
|
///
|
|
|
|
|
/// Basic usage:
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// let v: Vec<&str> = "abcXXXabcYYYabc".matches("abc").collect();
|
|
|
|
|
/// assert_eq!(v, ["abc", "abc", "abc"]);
|
|
|
|
|
///
|
|
|
|
|
/// let v: Vec<&str> = "1abc2abc3".matches(char::is_numeric).collect();
|
|
|
|
|
/// assert_eq!(v, ["1", "2", "3"]);
|
|
|
|
|
/// ```
|
|
|
|
|
#[stable(feature = "str_matches", since = "1.2.0")]
|
|
|
|
|
#[inline]
|
|
|
|
|
pub fn matches<'a, P: Pattern<'a>>(&'a self, pat: P) -> Matches<'a, P> {
|
2018-05-10 18:02:19 +00:00
|
|
|
|
Matches(MatchesInternal(pat.into_searcher(self)))
|
2018-04-05 17:00:48 +00:00
|
|
|
|
}
|
|
|
|
|
|
2018-04-07 19:56:02 +00:00
|
|
|
|
/// An iterator over the disjoint matches of a pattern within this string slice,
|
|
|
|
|
/// yielded in reverse order.
|
|
|
|
|
///
|
|
|
|
|
/// The pattern can be a `&str`, [`char`], or a closure that determines if
|
|
|
|
|
/// a character matches.
|
|
|
|
|
///
|
|
|
|
|
/// # Iterator behavior
|
|
|
|
|
///
|
|
|
|
|
/// The returned iterator requires that the pattern supports a reverse
|
|
|
|
|
/// search, and it will be a [`DoubleEndedIterator`] if a forward/reverse
|
|
|
|
|
/// search yields the same elements.
|
|
|
|
|
///
|
|
|
|
|
/// [`DoubleEndedIterator`]: iter/trait.DoubleEndedIterator.html
|
|
|
|
|
///
|
|
|
|
|
/// For iterating from the front, the [`matches`] method can be used.
|
|
|
|
|
///
|
|
|
|
|
/// [`matches`]: #method.matches
|
|
|
|
|
///
|
|
|
|
|
/// # Examples
|
|
|
|
|
///
|
|
|
|
|
/// Basic usage:
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// let v: Vec<&str> = "abcXXXabcYYYabc".rmatches("abc").collect();
|
|
|
|
|
/// assert_eq!(v, ["abc", "abc", "abc"]);
|
|
|
|
|
///
|
|
|
|
|
/// let v: Vec<&str> = "1abc2abc3".rmatches(char::is_numeric).collect();
|
|
|
|
|
/// assert_eq!(v, ["3", "2", "1"]);
|
|
|
|
|
/// ```
|
|
|
|
|
#[stable(feature = "str_matches", since = "1.2.0")]
|
2018-04-05 17:00:48 +00:00
|
|
|
|
#[inline]
|
2019-07-31 19:00:35 +00:00
|
|
|
|
pub fn rmatches<'a, P>(&'a self, pat: P) -> RMatches<'a, P>
|
|
|
|
|
where
|
|
|
|
|
P: Pattern<'a, Searcher: ReverseSearcher<'a>>,
|
2018-04-07 19:56:02 +00:00
|
|
|
|
{
|
2018-05-10 18:02:19 +00:00
|
|
|
|
RMatches(self.matches(pat).0)
|
2018-04-05 17:00:48 +00:00
|
|
|
|
}
|
|
|
|
|
|
2018-04-07 19:56:02 +00:00
|
|
|
|
/// An iterator over the disjoint matches of a pattern within this string
|
|
|
|
|
/// slice as well as the index that the match starts at.
|
|
|
|
|
///
|
|
|
|
|
/// For matches of `pat` within `self` that overlap, only the indices
|
|
|
|
|
/// corresponding to the first match are returned.
|
|
|
|
|
///
|
|
|
|
|
/// The pattern can be a `&str`, [`char`], or a closure that determines
|
|
|
|
|
/// if a character matches.
|
|
|
|
|
///
|
|
|
|
|
/// # Iterator behavior
|
|
|
|
|
///
|
|
|
|
|
/// The returned iterator will be a [`DoubleEndedIterator`] if the pattern
|
|
|
|
|
/// allows a reverse search and forward/reverse search yields the same
|
2019-02-09 22:16:58 +00:00
|
|
|
|
/// elements. This is true for, e.g., [`char`], but not for `&str`.
|
2018-04-07 19:56:02 +00:00
|
|
|
|
///
|
|
|
|
|
/// [`DoubleEndedIterator`]: iter/trait.DoubleEndedIterator.html
|
|
|
|
|
///
|
|
|
|
|
/// If the pattern allows a reverse search but its results might differ
|
|
|
|
|
/// from a forward search, the [`rmatch_indices`] method can be used.
|
|
|
|
|
///
|
|
|
|
|
/// [`rmatch_indices`]: #method.rmatch_indices
|
|
|
|
|
///
|
|
|
|
|
/// # Examples
|
|
|
|
|
///
|
|
|
|
|
/// Basic usage:
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// let v: Vec<_> = "abcXXXabcYYYabc".match_indices("abc").collect();
|
|
|
|
|
/// assert_eq!(v, [(0, "abc"), (6, "abc"), (12, "abc")]);
|
|
|
|
|
///
|
|
|
|
|
/// let v: Vec<_> = "1abcabc2".match_indices("abc").collect();
|
|
|
|
|
/// assert_eq!(v, [(1, "abc"), (4, "abc")]);
|
|
|
|
|
///
|
|
|
|
|
/// let v: Vec<_> = "ababa".match_indices("aba").collect();
|
|
|
|
|
/// assert_eq!(v, [(0, "aba")]); // only the first `aba`
|
|
|
|
|
/// ```
|
|
|
|
|
#[stable(feature = "str_match_indices", since = "1.5.0")]
|
2018-04-05 17:00:48 +00:00
|
|
|
|
#[inline]
|
2018-04-07 19:56:02 +00:00
|
|
|
|
pub fn match_indices<'a, P: Pattern<'a>>(&'a self, pat: P) -> MatchIndices<'a, P> {
|
2018-05-10 18:02:19 +00:00
|
|
|
|
MatchIndices(MatchIndicesInternal(pat.into_searcher(self)))
|
2018-04-05 17:00:48 +00:00
|
|
|
|
}
|
|
|
|
|
|
2018-04-07 19:56:02 +00:00
|
|
|
|
/// An iterator over the disjoint matches of a pattern within `self`,
|
|
|
|
|
/// yielded in reverse order along with the index of the match.
|
|
|
|
|
///
|
|
|
|
|
/// For matches of `pat` within `self` that overlap, only the indices
|
|
|
|
|
/// corresponding to the last match are returned.
|
|
|
|
|
///
|
|
|
|
|
/// The pattern can be a `&str`, [`char`], or a closure that determines if a
|
|
|
|
|
/// character matches.
|
|
|
|
|
///
|
|
|
|
|
/// # Iterator behavior
|
|
|
|
|
///
|
|
|
|
|
/// The returned iterator requires that the pattern supports a reverse
|
|
|
|
|
/// search, and it will be a [`DoubleEndedIterator`] if a forward/reverse
|
|
|
|
|
/// search yields the same elements.
|
|
|
|
|
///
|
|
|
|
|
/// [`DoubleEndedIterator`]: iter/trait.DoubleEndedIterator.html
|
|
|
|
|
///
|
|
|
|
|
/// For iterating from the front, the [`match_indices`] method can be used.
|
|
|
|
|
///
|
|
|
|
|
/// [`match_indices`]: #method.match_indices
|
|
|
|
|
///
|
|
|
|
|
/// # Examples
|
|
|
|
|
///
|
|
|
|
|
/// Basic usage:
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// let v: Vec<_> = "abcXXXabcYYYabc".rmatch_indices("abc").collect();
|
|
|
|
|
/// assert_eq!(v, [(12, "abc"), (6, "abc"), (0, "abc")]);
|
|
|
|
|
///
|
|
|
|
|
/// let v: Vec<_> = "1abcabc2".rmatch_indices("abc").collect();
|
|
|
|
|
/// assert_eq!(v, [(4, "abc"), (1, "abc")]);
|
|
|
|
|
///
|
|
|
|
|
/// let v: Vec<_> = "ababa".rmatch_indices("aba").collect();
|
|
|
|
|
/// assert_eq!(v, [(2, "aba")]); // only the last `aba`
|
|
|
|
|
/// ```
|
|
|
|
|
#[stable(feature = "str_match_indices", since = "1.5.0")]
|
2018-04-05 17:00:48 +00:00
|
|
|
|
#[inline]
|
2019-07-31 19:00:35 +00:00
|
|
|
|
pub fn rmatch_indices<'a, P>(&'a self, pat: P) -> RMatchIndices<'a, P>
|
|
|
|
|
where
|
|
|
|
|
P: Pattern<'a, Searcher: ReverseSearcher<'a>>,
|
2018-04-07 19:56:02 +00:00
|
|
|
|
{
|
2018-05-10 18:02:19 +00:00
|
|
|
|
RMatchIndices(self.match_indices(pat).0)
|
2018-04-07 19:56:02 +00:00
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/// Returns a string slice with leading and trailing whitespace removed.
|
|
|
|
|
///
|
|
|
|
|
/// 'Whitespace' is defined according to the terms of the Unicode Derived
|
|
|
|
|
/// Core Property `White_Space`.
|
|
|
|
|
///
|
|
|
|
|
/// # Examples
|
|
|
|
|
///
|
|
|
|
|
/// Basic usage:
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// let s = " Hello\tworld\t";
|
|
|
|
|
///
|
|
|
|
|
/// assert_eq!("Hello\tworld", s.trim());
|
|
|
|
|
/// ```
|
2018-12-25 19:28:20 +00:00
|
|
|
|
#[must_use = "this returns the trimmed string as a slice, \
|
2018-12-24 20:43:51 +00:00
|
|
|
|
without modifying the original"]
|
2018-04-07 19:56:02 +00:00
|
|
|
|
#[stable(feature = "rust1", since = "1.0.0")]
|
|
|
|
|
pub fn trim(&self) -> &str {
|
2018-05-10 18:02:19 +00:00
|
|
|
|
self.trim_matches(|c: char| c.is_whitespace())
|
2018-04-07 19:56:02 +00:00
|
|
|
|
}
|
|
|
|
|
|
2018-06-21 13:04:53 +00:00
|
|
|
|
/// Returns a string slice with leading whitespace removed.
|
|
|
|
|
///
|
|
|
|
|
/// 'Whitespace' is defined according to the terms of the Unicode Derived
|
|
|
|
|
/// Core Property `White_Space`.
|
|
|
|
|
///
|
|
|
|
|
/// # Text directionality
|
|
|
|
|
///
|
|
|
|
|
/// A string is a sequence of bytes. `start` in this context means the first
|
|
|
|
|
/// position of that byte string; for a left-to-right language like English or
|
2019-02-09 22:16:58 +00:00
|
|
|
|
/// Russian, this will be left side, and for right-to-left languages like
|
2019-08-31 16:36:55 +00:00
|
|
|
|
/// Arabic or Hebrew, this will be the right side.
|
2018-06-21 13:04:53 +00:00
|
|
|
|
///
|
|
|
|
|
/// # Examples
|
|
|
|
|
///
|
|
|
|
|
/// Basic usage:
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
2018-08-02 21:00:13 +00:00
|
|
|
|
/// let s = " Hello\tworld\t";
|
2018-06-21 13:04:53 +00:00
|
|
|
|
/// assert_eq!("Hello\tworld\t", s.trim_start());
|
|
|
|
|
/// ```
|
|
|
|
|
///
|
|
|
|
|
/// Directionality:
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
2018-08-05 20:51:40 +00:00
|
|
|
|
/// let s = " English ";
|
2018-06-21 13:04:53 +00:00
|
|
|
|
/// assert!(Some('E') == s.trim_start().chars().next());
|
|
|
|
|
///
|
2018-08-05 20:51:40 +00:00
|
|
|
|
/// let s = " עברית ";
|
2018-06-21 13:04:53 +00:00
|
|
|
|
/// assert!(Some('ע') == s.trim_start().chars().next());
|
|
|
|
|
/// ```
|
2018-12-26 21:03:04 +00:00
|
|
|
|
#[must_use = "this returns the trimmed string as a new slice, \
|
2018-12-24 20:43:51 +00:00
|
|
|
|
without modifying the original"]
|
2018-08-05 20:51:40 +00:00
|
|
|
|
#[stable(feature = "trim_direction", since = "1.30.0")]
|
2018-06-21 13:04:53 +00:00
|
|
|
|
pub fn trim_start(&self) -> &str {
|
|
|
|
|
self.trim_start_matches(|c: char| c.is_whitespace())
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/// Returns a string slice with trailing whitespace removed.
|
|
|
|
|
///
|
|
|
|
|
/// 'Whitespace' is defined according to the terms of the Unicode Derived
|
|
|
|
|
/// Core Property `White_Space`.
|
|
|
|
|
///
|
|
|
|
|
/// # Text directionality
|
|
|
|
|
///
|
|
|
|
|
/// A string is a sequence of bytes. `end` in this context means the last
|
|
|
|
|
/// position of that byte string; for a left-to-right language like English or
|
2019-02-09 22:16:58 +00:00
|
|
|
|
/// Russian, this will be right side, and for right-to-left languages like
|
2019-08-31 16:36:55 +00:00
|
|
|
|
/// Arabic or Hebrew, this will be the left side.
|
2018-06-21 13:04:53 +00:00
|
|
|
|
///
|
|
|
|
|
/// # Examples
|
|
|
|
|
///
|
|
|
|
|
/// Basic usage:
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
2018-08-02 21:00:13 +00:00
|
|
|
|
/// let s = " Hello\tworld\t";
|
2018-06-21 13:04:53 +00:00
|
|
|
|
/// assert_eq!(" Hello\tworld", s.trim_end());
|
|
|
|
|
/// ```
|
|
|
|
|
///
|
|
|
|
|
/// Directionality:
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
2018-08-05 20:51:40 +00:00
|
|
|
|
/// let s = " English ";
|
2018-06-21 13:04:53 +00:00
|
|
|
|
/// assert!(Some('h') == s.trim_end().chars().rev().next());
|
|
|
|
|
///
|
2018-08-05 20:51:40 +00:00
|
|
|
|
/// let s = " עברית ";
|
2018-06-21 13:04:53 +00:00
|
|
|
|
/// assert!(Some('ת') == s.trim_end().chars().rev().next());
|
|
|
|
|
/// ```
|
2018-12-26 21:03:04 +00:00
|
|
|
|
#[must_use = "this returns the trimmed string as a new slice, \
|
2018-12-24 20:43:51 +00:00
|
|
|
|
without modifying the original"]
|
2018-08-05 20:51:40 +00:00
|
|
|
|
#[stable(feature = "trim_direction", since = "1.30.0")]
|
2018-06-21 13:04:53 +00:00
|
|
|
|
pub fn trim_end(&self) -> &str {
|
|
|
|
|
self.trim_end_matches(|c: char| c.is_whitespace())
|
|
|
|
|
}
|
|
|
|
|
|
2018-04-07 19:56:02 +00:00
|
|
|
|
/// Returns a string slice with leading whitespace removed.
|
|
|
|
|
///
|
|
|
|
|
/// 'Whitespace' is defined according to the terms of the Unicode Derived
|
|
|
|
|
/// Core Property `White_Space`.
|
|
|
|
|
///
|
|
|
|
|
/// # Text directionality
|
|
|
|
|
///
|
|
|
|
|
/// A string is a sequence of bytes. 'Left' in this context means the first
|
|
|
|
|
/// position of that byte string; for a language like Arabic or Hebrew
|
|
|
|
|
/// which are 'right to left' rather than 'left to right', this will be
|
|
|
|
|
/// the _right_ side, not the left.
|
|
|
|
|
///
|
|
|
|
|
/// # Examples
|
|
|
|
|
///
|
|
|
|
|
/// Basic usage:
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// let s = " Hello\tworld\t";
|
|
|
|
|
///
|
|
|
|
|
/// assert_eq!("Hello\tworld\t", s.trim_left());
|
|
|
|
|
/// ```
|
|
|
|
|
///
|
|
|
|
|
/// Directionality:
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// let s = " English";
|
|
|
|
|
/// assert!(Some('E') == s.trim_left().chars().next());
|
|
|
|
|
///
|
|
|
|
|
/// let s = " עברית";
|
|
|
|
|
/// assert!(Some('ע') == s.trim_left().chars().next());
|
|
|
|
|
/// ```
|
|
|
|
|
#[stable(feature = "rust1", since = "1.0.0")]
|
2019-03-22 18:56:08 +00:00
|
|
|
|
#[rustc_deprecated(
|
|
|
|
|
since = "1.33.0",
|
|
|
|
|
reason = "superseded by `trim_start`",
|
2019-12-22 22:42:04 +00:00
|
|
|
|
suggestion = "trim_start"
|
2019-03-22 18:56:08 +00:00
|
|
|
|
)]
|
2018-04-07 19:56:02 +00:00
|
|
|
|
pub fn trim_left(&self) -> &str {
|
2018-06-21 13:04:53 +00:00
|
|
|
|
self.trim_start()
|
2018-04-07 19:56:02 +00:00
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/// Returns a string slice with trailing whitespace removed.
|
|
|
|
|
///
|
|
|
|
|
/// 'Whitespace' is defined according to the terms of the Unicode Derived
|
|
|
|
|
/// Core Property `White_Space`.
|
|
|
|
|
///
|
|
|
|
|
/// # Text directionality
|
|
|
|
|
///
|
|
|
|
|
/// A string is a sequence of bytes. 'Right' in this context means the last
|
|
|
|
|
/// position of that byte string; for a language like Arabic or Hebrew
|
|
|
|
|
/// which are 'right to left' rather than 'left to right', this will be
|
|
|
|
|
/// the _left_ side, not the right.
|
|
|
|
|
///
|
|
|
|
|
/// # Examples
|
|
|
|
|
///
|
|
|
|
|
/// Basic usage:
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// let s = " Hello\tworld\t";
|
|
|
|
|
///
|
|
|
|
|
/// assert_eq!(" Hello\tworld", s.trim_right());
|
|
|
|
|
/// ```
|
|
|
|
|
///
|
|
|
|
|
/// Directionality:
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// let s = "English ";
|
|
|
|
|
/// assert!(Some('h') == s.trim_right().chars().rev().next());
|
|
|
|
|
///
|
|
|
|
|
/// let s = "עברית ";
|
|
|
|
|
/// assert!(Some('ת') == s.trim_right().chars().rev().next());
|
|
|
|
|
/// ```
|
|
|
|
|
#[stable(feature = "rust1", since = "1.0.0")]
|
2019-03-22 18:56:08 +00:00
|
|
|
|
#[rustc_deprecated(
|
|
|
|
|
since = "1.33.0",
|
|
|
|
|
reason = "superseded by `trim_end`",
|
2019-12-22 22:42:04 +00:00
|
|
|
|
suggestion = "trim_end"
|
2019-03-22 18:56:08 +00:00
|
|
|
|
)]
|
2018-04-07 19:56:02 +00:00
|
|
|
|
pub fn trim_right(&self) -> &str {
|
2018-06-21 13:04:53 +00:00
|
|
|
|
self.trim_end()
|
2018-04-07 19:56:02 +00:00
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/// Returns a string slice with all prefixes and suffixes that match a
|
|
|
|
|
/// pattern repeatedly removed.
|
|
|
|
|
///
|
|
|
|
|
/// The pattern can be a [`char`] or a closure that determines if a
|
|
|
|
|
/// character matches.
|
|
|
|
|
///
|
|
|
|
|
/// # Examples
|
|
|
|
|
///
|
|
|
|
|
/// Simple patterns:
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// assert_eq!("11foo1bar11".trim_matches('1'), "foo1bar");
|
|
|
|
|
/// assert_eq!("123foo1bar123".trim_matches(char::is_numeric), "foo1bar");
|
|
|
|
|
///
|
|
|
|
|
/// let x: &[_] = &['1', '2'];
|
|
|
|
|
/// assert_eq!("12foo1bar12".trim_matches(x), "foo1bar");
|
|
|
|
|
/// ```
|
|
|
|
|
///
|
|
|
|
|
/// A more complex pattern, using a closure:
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// assert_eq!("1foo1barXX".trim_matches(|c| c == '1' || c == 'X'), "foo1bar");
|
|
|
|
|
/// ```
|
2018-12-26 21:03:04 +00:00
|
|
|
|
#[must_use = "this returns the trimmed string as a new slice, \
|
2018-12-24 20:43:51 +00:00
|
|
|
|
without modifying the original"]
|
2018-04-07 19:56:02 +00:00
|
|
|
|
#[stable(feature = "rust1", since = "1.0.0")]
|
2019-07-31 19:00:35 +00:00
|
|
|
|
pub fn trim_matches<'a, P>(&'a self, pat: P) -> &'a str
|
|
|
|
|
where
|
|
|
|
|
P: Pattern<'a, Searcher: DoubleEndedSearcher<'a>>,
|
2018-04-07 19:56:02 +00:00
|
|
|
|
{
|
2018-05-10 18:02:19 +00:00
|
|
|
|
let mut i = 0;
|
|
|
|
|
let mut j = 0;
|
|
|
|
|
let mut matcher = pat.into_searcher(self);
|
|
|
|
|
if let Some((a, b)) = matcher.next_reject() {
|
|
|
|
|
i = a;
|
|
|
|
|
j = b; // Remember earliest known match, correct it below if
|
2019-12-22 22:42:04 +00:00
|
|
|
|
// last match is different
|
2018-05-10 18:02:19 +00:00
|
|
|
|
}
|
|
|
|
|
if let Some((_, b)) = matcher.next_reject_back() {
|
|
|
|
|
j = b;
|
|
|
|
|
}
|
2019-11-20 02:28:32 +00:00
|
|
|
|
// SAFETY: `Searcher` is known to return valid indices
|
2018-05-10 18:02:19 +00:00
|
|
|
|
unsafe {
|
2018-06-26 10:34:42 +00:00
|
|
|
|
self.get_unchecked(i..j)
|
2018-05-10 18:02:19 +00:00
|
|
|
|
}
|
2018-04-07 19:56:02 +00:00
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/// Returns a string slice with all prefixes that match a pattern
|
|
|
|
|
/// repeatedly removed.
|
|
|
|
|
///
|
|
|
|
|
/// The pattern can be a `&str`, [`char`], or a closure that determines if
|
|
|
|
|
/// a character matches.
|
|
|
|
|
///
|
|
|
|
|
/// # Text directionality
|
|
|
|
|
///
|
2019-07-03 06:59:40 +00:00
|
|
|
|
/// A string is a sequence of bytes. `start` in this context means the first
|
|
|
|
|
/// position of that byte string; for a left-to-right language like English or
|
|
|
|
|
/// Russian, this will be left side, and for right-to-left languages like
|
2019-08-31 16:36:55 +00:00
|
|
|
|
/// Arabic or Hebrew, this will be the right side.
|
2018-04-07 19:56:02 +00:00
|
|
|
|
///
|
|
|
|
|
/// # Examples
|
|
|
|
|
///
|
|
|
|
|
/// Basic usage:
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
2018-06-21 13:04:53 +00:00
|
|
|
|
/// assert_eq!("11foo1bar11".trim_start_matches('1'), "foo1bar11");
|
|
|
|
|
/// assert_eq!("123foo1bar123".trim_start_matches(char::is_numeric), "foo1bar123");
|
2018-04-07 19:56:02 +00:00
|
|
|
|
///
|
|
|
|
|
/// let x: &[_] = &['1', '2'];
|
2018-06-21 13:04:53 +00:00
|
|
|
|
/// assert_eq!("12foo1bar12".trim_start_matches(x), "foo1bar12");
|
2018-04-07 19:56:02 +00:00
|
|
|
|
/// ```
|
2018-12-26 21:03:04 +00:00
|
|
|
|
#[must_use = "this returns the trimmed string as a new slice, \
|
2018-12-24 20:43:51 +00:00
|
|
|
|
without modifying the original"]
|
2018-08-05 20:51:40 +00:00
|
|
|
|
#[stable(feature = "trim_direction", since = "1.30.0")]
|
2018-06-21 13:04:53 +00:00
|
|
|
|
pub fn trim_start_matches<'a, P: Pattern<'a>>(&'a self, pat: P) -> &'a str {
|
2018-05-10 18:02:19 +00:00
|
|
|
|
let mut i = self.len();
|
|
|
|
|
let mut matcher = pat.into_searcher(self);
|
|
|
|
|
if let Some((a, _)) = matcher.next_reject() {
|
|
|
|
|
i = a;
|
|
|
|
|
}
|
2019-11-20 02:28:32 +00:00
|
|
|
|
// SAFETY: `Searcher` is known to return valid indices
|
2018-05-10 18:02:19 +00:00
|
|
|
|
unsafe {
|
2018-06-26 10:34:42 +00:00
|
|
|
|
self.get_unchecked(i..self.len())
|
2018-05-10 18:02:19 +00:00
|
|
|
|
}
|
2018-04-07 19:56:02 +00:00
|
|
|
|
}
|
|
|
|
|
|
2019-11-25 11:36:47 +00:00
|
|
|
|
/// Returns a string slice with the prefix removed.
|
|
|
|
|
///
|
|
|
|
|
/// If the string starts with the pattern `prefix`, `Some` is returned with the substring where
|
|
|
|
|
/// the prefix is removed. Unlike `trim_start_matches`, this method removes the prefix exactly
|
|
|
|
|
/// once.
|
|
|
|
|
///
|
2019-11-26 09:42:43 +00:00
|
|
|
|
/// If the string does not start with `prefix`, `None` is returned.
|
2019-11-25 11:36:47 +00:00
|
|
|
|
///
|
|
|
|
|
/// # Examples
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// #![feature(str_strip)]
|
|
|
|
|
///
|
|
|
|
|
/// assert_eq!("foobar".strip_prefix("foo"), Some("bar"));
|
|
|
|
|
/// assert_eq!("foobar".strip_prefix("bar"), None);
|
|
|
|
|
/// assert_eq!("foofoo".strip_prefix("foo"), Some("foo"));
|
|
|
|
|
/// ```
|
|
|
|
|
#[must_use = "this returns the remaining substring as a new slice, \
|
|
|
|
|
without modifying the original"]
|
2019-12-15 09:07:57 +00:00
|
|
|
|
#[unstable(feature = "str_strip", reason = "newly added", issue = "67302")]
|
2019-11-25 11:36:47 +00:00
|
|
|
|
pub fn strip_prefix<'a, P: Pattern<'a>>(&'a self, prefix: P) -> Option<&'a str> {
|
|
|
|
|
let mut matcher = prefix.into_searcher(self);
|
|
|
|
|
if let SearchStep::Match(start, len) = matcher.next() {
|
2019-12-22 22:42:04 +00:00
|
|
|
|
debug_assert_eq!(
|
|
|
|
|
start, 0,
|
|
|
|
|
"The first search step from Searcher \
|
|
|
|
|
must include the first character"
|
|
|
|
|
);
|
2019-11-25 11:36:47 +00:00
|
|
|
|
unsafe {
|
2019-11-26 09:42:43 +00:00
|
|
|
|
// Searcher is known to return valid indices.
|
2019-11-25 11:36:47 +00:00
|
|
|
|
Some(self.get_unchecked(len..))
|
|
|
|
|
}
|
|
|
|
|
} else {
|
|
|
|
|
None
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/// Returns a string slice with the suffix removed.
|
|
|
|
|
///
|
2019-11-26 09:42:43 +00:00
|
|
|
|
/// If the string ends with the pattern `suffix`, `Some` is returned with the substring where
|
2019-11-25 11:36:47 +00:00
|
|
|
|
/// the suffix is removed. Unlike `trim_end_matches`, this method removes the suffix exactly
|
|
|
|
|
/// once.
|
|
|
|
|
///
|
2019-11-26 09:42:43 +00:00
|
|
|
|
/// If the string does not end with `suffix`, `None` is returned.
|
2019-11-25 11:36:47 +00:00
|
|
|
|
///
|
|
|
|
|
/// # Examples
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// #![feature(str_strip)]
|
|
|
|
|
/// assert_eq!("barfoo".strip_suffix("foo"), Some("bar"));
|
|
|
|
|
/// assert_eq!("barfoo".strip_suffix("bar"), None);
|
|
|
|
|
/// assert_eq!("foofoo".strip_suffix("foo"), Some("foo"));
|
|
|
|
|
/// ```
|
|
|
|
|
#[must_use = "this returns the remaining substring as a new slice, \
|
|
|
|
|
without modifying the original"]
|
2019-12-15 09:07:57 +00:00
|
|
|
|
#[unstable(feature = "str_strip", reason = "newly added", issue = "67302")]
|
2019-11-25 11:36:47 +00:00
|
|
|
|
pub fn strip_suffix<'a, P>(&'a self, suffix: P) -> Option<&'a str>
|
|
|
|
|
where
|
|
|
|
|
P: Pattern<'a>,
|
|
|
|
|
<P as Pattern<'a>>::Searcher: ReverseSearcher<'a>,
|
|
|
|
|
{
|
|
|
|
|
let mut matcher = suffix.into_searcher(self);
|
|
|
|
|
if let SearchStep::Match(start, end) = matcher.next_back() {
|
2019-12-22 22:42:04 +00:00
|
|
|
|
debug_assert_eq!(
|
|
|
|
|
end,
|
|
|
|
|
self.len(),
|
|
|
|
|
"The first search step from ReverseSearcher \
|
|
|
|
|
must include the last character"
|
|
|
|
|
);
|
2019-11-25 11:36:47 +00:00
|
|
|
|
unsafe {
|
2019-11-26 09:42:43 +00:00
|
|
|
|
// Searcher is known to return valid indices.
|
2019-11-25 11:36:47 +00:00
|
|
|
|
Some(self.get_unchecked(..start))
|
|
|
|
|
}
|
|
|
|
|
} else {
|
|
|
|
|
None
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
2018-04-07 19:56:02 +00:00
|
|
|
|
/// Returns a string slice with all suffixes that match a pattern
|
|
|
|
|
/// repeatedly removed.
|
|
|
|
|
///
|
|
|
|
|
/// The pattern can be a `&str`, [`char`], or a closure that
|
|
|
|
|
/// determines if a character matches.
|
|
|
|
|
///
|
|
|
|
|
/// # Text directionality
|
|
|
|
|
///
|
2019-07-03 06:59:40 +00:00
|
|
|
|
/// A string is a sequence of bytes. `end` in this context means the last
|
|
|
|
|
/// position of that byte string; for a left-to-right language like English or
|
|
|
|
|
/// Russian, this will be right side, and for right-to-left languages like
|
2019-08-31 16:36:55 +00:00
|
|
|
|
/// Arabic or Hebrew, this will be the left side.
|
2018-04-07 19:56:02 +00:00
|
|
|
|
///
|
|
|
|
|
/// # Examples
|
|
|
|
|
///
|
|
|
|
|
/// Simple patterns:
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
2018-06-21 13:04:53 +00:00
|
|
|
|
/// assert_eq!("11foo1bar11".trim_end_matches('1'), "11foo1bar");
|
|
|
|
|
/// assert_eq!("123foo1bar123".trim_end_matches(char::is_numeric), "123foo1bar");
|
2018-04-07 19:56:02 +00:00
|
|
|
|
///
|
|
|
|
|
/// let x: &[_] = &['1', '2'];
|
2018-06-21 13:04:53 +00:00
|
|
|
|
/// assert_eq!("12foo1bar12".trim_end_matches(x), "12foo1bar");
|
2018-04-07 19:56:02 +00:00
|
|
|
|
/// ```
|
|
|
|
|
///
|
|
|
|
|
/// A more complex pattern, using a closure:
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
2018-06-21 13:04:53 +00:00
|
|
|
|
/// assert_eq!("1fooX".trim_end_matches(|c| c == '1' || c == 'X'), "1foo");
|
2018-04-07 19:56:02 +00:00
|
|
|
|
/// ```
|
2018-12-26 21:03:04 +00:00
|
|
|
|
#[must_use = "this returns the trimmed string as a new slice, \
|
2018-12-24 20:43:51 +00:00
|
|
|
|
without modifying the original"]
|
2018-08-05 20:51:40 +00:00
|
|
|
|
#[stable(feature = "trim_direction", since = "1.30.0")]
|
2019-07-31 19:00:35 +00:00
|
|
|
|
pub fn trim_end_matches<'a, P>(&'a self, pat: P) -> &'a str
|
|
|
|
|
where
|
|
|
|
|
P: Pattern<'a, Searcher: ReverseSearcher<'a>>,
|
2018-04-07 19:56:02 +00:00
|
|
|
|
{
|
2018-05-10 18:02:19 +00:00
|
|
|
|
let mut j = 0;
|
|
|
|
|
let mut matcher = pat.into_searcher(self);
|
|
|
|
|
if let Some((_, b)) = matcher.next_reject_back() {
|
|
|
|
|
j = b;
|
|
|
|
|
}
|
2019-11-20 02:28:32 +00:00
|
|
|
|
// SAFETY: `Searcher` is known to return valid indices
|
2018-05-10 18:02:19 +00:00
|
|
|
|
unsafe {
|
2018-06-26 10:34:42 +00:00
|
|
|
|
self.get_unchecked(0..j)
|
2018-05-10 18:02:19 +00:00
|
|
|
|
}
|
2018-04-07 19:56:02 +00:00
|
|
|
|
}
|
|
|
|
|
|
2018-06-21 13:04:53 +00:00
|
|
|
|
/// Returns a string slice with all prefixes that match a pattern
|
|
|
|
|
/// repeatedly removed.
|
|
|
|
|
///
|
|
|
|
|
/// The pattern can be a `&str`, [`char`], or a closure that determines if
|
|
|
|
|
/// a character matches.
|
|
|
|
|
///
|
|
|
|
|
/// [`char`]: primitive.char.html
|
|
|
|
|
///
|
|
|
|
|
/// # Text directionality
|
|
|
|
|
///
|
2019-07-03 06:59:40 +00:00
|
|
|
|
/// A string is a sequence of bytes. 'Left' in this context means the first
|
|
|
|
|
/// position of that byte string; for a language like Arabic or Hebrew
|
|
|
|
|
/// which are 'right to left' rather than 'left to right', this will be
|
|
|
|
|
/// the _right_ side, not the left.
|
2018-06-21 13:04:53 +00:00
|
|
|
|
///
|
|
|
|
|
/// # Examples
|
|
|
|
|
///
|
|
|
|
|
/// Basic usage:
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// assert_eq!("11foo1bar11".trim_left_matches('1'), "foo1bar11");
|
|
|
|
|
/// assert_eq!("123foo1bar123".trim_left_matches(char::is_numeric), "foo1bar123");
|
|
|
|
|
///
|
|
|
|
|
/// let x: &[_] = &['1', '2'];
|
|
|
|
|
/// assert_eq!("12foo1bar12".trim_left_matches(x), "foo1bar12");
|
|
|
|
|
/// ```
|
|
|
|
|
#[stable(feature = "rust1", since = "1.0.0")]
|
2019-03-22 18:56:08 +00:00
|
|
|
|
#[rustc_deprecated(
|
|
|
|
|
since = "1.33.0",
|
|
|
|
|
reason = "superseded by `trim_start_matches`",
|
2019-12-22 22:42:04 +00:00
|
|
|
|
suggestion = "trim_start_matches"
|
2019-03-22 18:56:08 +00:00
|
|
|
|
)]
|
2018-06-21 13:04:53 +00:00
|
|
|
|
pub fn trim_left_matches<'a, P: Pattern<'a>>(&'a self, pat: P) -> &'a str {
|
|
|
|
|
self.trim_start_matches(pat)
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/// Returns a string slice with all suffixes that match a pattern
|
|
|
|
|
/// repeatedly removed.
|
|
|
|
|
///
|
|
|
|
|
/// The pattern can be a `&str`, [`char`], or a closure that
|
|
|
|
|
/// determines if a character matches.
|
|
|
|
|
///
|
|
|
|
|
/// [`char`]: primitive.char.html
|
|
|
|
|
///
|
|
|
|
|
/// # Text directionality
|
|
|
|
|
///
|
2019-07-03 06:59:40 +00:00
|
|
|
|
/// A string is a sequence of bytes. 'Right' in this context means the last
|
|
|
|
|
/// position of that byte string; for a language like Arabic or Hebrew
|
|
|
|
|
/// which are 'right to left' rather than 'left to right', this will be
|
|
|
|
|
/// the _left_ side, not the right.
|
2018-06-21 13:04:53 +00:00
|
|
|
|
///
|
|
|
|
|
/// # Examples
|
|
|
|
|
///
|
|
|
|
|
/// Simple patterns:
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// assert_eq!("11foo1bar11".trim_right_matches('1'), "11foo1bar");
|
|
|
|
|
/// assert_eq!("123foo1bar123".trim_right_matches(char::is_numeric), "123foo1bar");
|
|
|
|
|
///
|
|
|
|
|
/// let x: &[_] = &['1', '2'];
|
|
|
|
|
/// assert_eq!("12foo1bar12".trim_right_matches(x), "12foo1bar");
|
|
|
|
|
/// ```
|
|
|
|
|
///
|
|
|
|
|
/// A more complex pattern, using a closure:
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// assert_eq!("1fooX".trim_right_matches(|c| c == '1' || c == 'X'), "1foo");
|
|
|
|
|
/// ```
|
|
|
|
|
#[stable(feature = "rust1", since = "1.0.0")]
|
2019-03-22 18:56:08 +00:00
|
|
|
|
#[rustc_deprecated(
|
|
|
|
|
since = "1.33.0",
|
|
|
|
|
reason = "superseded by `trim_end_matches`",
|
2019-12-22 22:42:04 +00:00
|
|
|
|
suggestion = "trim_end_matches"
|
2019-03-22 18:56:08 +00:00
|
|
|
|
)]
|
2019-07-31 19:00:35 +00:00
|
|
|
|
pub fn trim_right_matches<'a, P>(&'a self, pat: P) -> &'a str
|
|
|
|
|
where
|
|
|
|
|
P: Pattern<'a, Searcher: ReverseSearcher<'a>>,
|
2018-06-21 13:04:53 +00:00
|
|
|
|
{
|
|
|
|
|
self.trim_end_matches(pat)
|
|
|
|
|
}
|
|
|
|
|
|
2018-04-07 19:56:02 +00:00
|
|
|
|
/// Parses this string slice into another type.
|
|
|
|
|
///
|
|
|
|
|
/// Because `parse` is so general, it can cause problems with type
|
|
|
|
|
/// inference. As such, `parse` is one of the few times you'll see
|
|
|
|
|
/// the syntax affectionately known as the 'turbofish': `::<>`. This
|
|
|
|
|
/// helps the inference algorithm understand specifically which type
|
|
|
|
|
/// you're trying to parse into.
|
|
|
|
|
///
|
|
|
|
|
/// `parse` can parse any type that implements the [`FromStr`] trait.
|
|
|
|
|
///
|
|
|
|
|
/// [`FromStr`]: str/trait.FromStr.html
|
|
|
|
|
///
|
|
|
|
|
/// # Errors
|
|
|
|
|
///
|
|
|
|
|
/// Will return [`Err`] if it's not possible to parse this string slice into
|
|
|
|
|
/// the desired type.
|
|
|
|
|
///
|
|
|
|
|
/// [`Err`]: str/trait.FromStr.html#associatedtype.Err
|
|
|
|
|
///
|
|
|
|
|
/// # Examples
|
|
|
|
|
///
|
|
|
|
|
/// Basic usage
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// let four: u32 = "4".parse().unwrap();
|
|
|
|
|
///
|
|
|
|
|
/// assert_eq!(4, four);
|
|
|
|
|
/// ```
|
|
|
|
|
///
|
|
|
|
|
/// Using the 'turbofish' instead of annotating `four`:
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// let four = "4".parse::<u32>();
|
|
|
|
|
///
|
|
|
|
|
/// assert_eq!(Ok(4), four);
|
|
|
|
|
/// ```
|
|
|
|
|
///
|
|
|
|
|
/// Failing to parse:
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// let nope = "j".parse::<u32>();
|
|
|
|
|
///
|
|
|
|
|
/// assert!(nope.is_err());
|
|
|
|
|
/// ```
|
|
|
|
|
#[inline]
|
|
|
|
|
#[stable(feature = "rust1", since = "1.0.0")]
|
|
|
|
|
pub fn parse<F: FromStr>(&self) -> Result<F, F::Err> {
|
2018-05-10 18:02:19 +00:00
|
|
|
|
FromStr::from_str(self)
|
2018-04-05 17:00:48 +00:00
|
|
|
|
}
|
|
|
|
|
|
2018-04-07 19:56:02 +00:00
|
|
|
|
/// Checks if all characters in this string are within the ASCII range.
|
|
|
|
|
///
|
|
|
|
|
/// # Examples
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// let ascii = "hello!\n";
|
|
|
|
|
/// let non_ascii = "Grüße, Jürgen ❤";
|
|
|
|
|
///
|
|
|
|
|
/// assert!(ascii.is_ascii());
|
|
|
|
|
/// assert!(!non_ascii.is_ascii());
|
|
|
|
|
/// ```
|
|
|
|
|
#[stable(feature = "ascii_methods_on_intrinsics", since = "1.23.0")]
|
2018-04-05 17:00:48 +00:00
|
|
|
|
#[inline]
|
2018-04-07 19:56:02 +00:00
|
|
|
|
pub fn is_ascii(&self) -> bool {
|
|
|
|
|
// We can treat each byte as character here: all multibyte characters
|
|
|
|
|
// start with a byte that is not in the ascii range, so we will stop
|
|
|
|
|
// there already.
|
|
|
|
|
self.bytes().all(|b| b.is_ascii())
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/// Checks that two strings are an ASCII case-insensitive match.
|
|
|
|
|
///
|
|
|
|
|
/// Same as `to_ascii_lowercase(a) == to_ascii_lowercase(b)`,
|
|
|
|
|
/// but without allocating and copying temporaries.
|
|
|
|
|
///
|
|
|
|
|
/// # Examples
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// assert!("Ferris".eq_ignore_ascii_case("FERRIS"));
|
|
|
|
|
/// assert!("Ferrös".eq_ignore_ascii_case("FERRöS"));
|
|
|
|
|
/// assert!(!"Ferrös".eq_ignore_ascii_case("FERRÖS"));
|
|
|
|
|
/// ```
|
|
|
|
|
#[stable(feature = "ascii_methods_on_intrinsics", since = "1.23.0")]
|
|
|
|
|
#[inline]
|
|
|
|
|
pub fn eq_ignore_ascii_case(&self, other: &str) -> bool {
|
|
|
|
|
self.as_bytes().eq_ignore_ascii_case(other.as_bytes())
|
2018-04-05 17:00:48 +00:00
|
|
|
|
}
|
2018-04-07 19:56:02 +00:00
|
|
|
|
|
|
|
|
|
/// Converts this string to its ASCII upper case equivalent in-place.
|
|
|
|
|
///
|
|
|
|
|
/// ASCII letters 'a' to 'z' are mapped to 'A' to 'Z',
|
|
|
|
|
/// but non-ASCII letters are unchanged.
|
|
|
|
|
///
|
|
|
|
|
/// To return a new uppercased value without modifying the existing one, use
|
|
|
|
|
/// [`to_ascii_uppercase`].
|
|
|
|
|
///
|
|
|
|
|
/// [`to_ascii_uppercase`]: #method.to_ascii_uppercase
|
2019-06-09 00:28:29 +00:00
|
|
|
|
///
|
|
|
|
|
/// # Examples
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// let mut s = String::from("Grüße, Jürgen ❤");
|
|
|
|
|
///
|
|
|
|
|
/// s.make_ascii_uppercase();
|
|
|
|
|
///
|
|
|
|
|
/// assert_eq!("GRüßE, JüRGEN ❤", s);
|
|
|
|
|
/// ```
|
2018-04-07 19:56:02 +00:00
|
|
|
|
#[stable(feature = "ascii_methods_on_intrinsics", since = "1.23.0")]
|
|
|
|
|
pub fn make_ascii_uppercase(&mut self) {
|
2019-11-20 02:28:32 +00:00
|
|
|
|
// SAFETY: safe because we transmute two types with the same layout
|
2018-04-07 19:56:02 +00:00
|
|
|
|
let me = unsafe { self.as_bytes_mut() };
|
|
|
|
|
me.make_ascii_uppercase()
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/// Converts this string to its ASCII lower case equivalent in-place.
|
|
|
|
|
///
|
|
|
|
|
/// ASCII letters 'A' to 'Z' are mapped to 'a' to 'z',
|
|
|
|
|
/// but non-ASCII letters are unchanged.
|
|
|
|
|
///
|
|
|
|
|
/// To return a new lowercased value without modifying the existing one, use
|
|
|
|
|
/// [`to_ascii_lowercase`].
|
|
|
|
|
///
|
|
|
|
|
/// [`to_ascii_lowercase`]: #method.to_ascii_lowercase
|
2019-06-09 00:28:29 +00:00
|
|
|
|
///
|
|
|
|
|
/// # Examples
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
2019-06-10 10:42:43 +00:00
|
|
|
|
/// let mut s = String::from("GRÜßE, JÜRGEN ❤");
|
2019-06-09 00:28:29 +00:00
|
|
|
|
///
|
|
|
|
|
/// s.make_ascii_lowercase();
|
|
|
|
|
///
|
2019-06-10 10:42:43 +00:00
|
|
|
|
/// assert_eq!("grÜße, jÜrgen ❤", s);
|
2019-06-09 00:28:29 +00:00
|
|
|
|
/// ```
|
2018-04-07 19:56:02 +00:00
|
|
|
|
#[stable(feature = "ascii_methods_on_intrinsics", since = "1.23.0")]
|
|
|
|
|
pub fn make_ascii_lowercase(&mut self) {
|
2019-11-20 02:28:32 +00:00
|
|
|
|
// SAFETY: safe because we transmute two types with the same layout
|
2018-04-07 19:56:02 +00:00
|
|
|
|
let me = unsafe { self.as_bytes_mut() };
|
|
|
|
|
me.make_ascii_lowercase()
|
|
|
|
|
}
|
2019-02-02 09:34:36 +00:00
|
|
|
|
|
2019-02-20 14:11:22 +00:00
|
|
|
|
/// Return an iterator that escapes each char in `self` with [`char::escape_debug`].
|
2019-02-02 09:34:36 +00:00
|
|
|
|
///
|
|
|
|
|
/// Note: only extended grapheme codepoints that begin the string will be
|
|
|
|
|
/// escaped.
|
|
|
|
|
///
|
|
|
|
|
/// [`char::escape_debug`]: ../std/primitive.char.html#method.escape_debug
|
2019-02-02 10:25:46 +00:00
|
|
|
|
///
|
|
|
|
|
/// # Examples
|
|
|
|
|
///
|
|
|
|
|
/// As an iterator:
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// for c in "❤\n!".escape_debug() {
|
|
|
|
|
/// print!("{}", c);
|
|
|
|
|
/// }
|
|
|
|
|
/// println!();
|
|
|
|
|
/// ```
|
|
|
|
|
///
|
|
|
|
|
/// Using `println!` directly:
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// println!("{}", "❤\n!".escape_debug());
|
|
|
|
|
/// ```
|
|
|
|
|
///
|
|
|
|
|
///
|
|
|
|
|
/// Both are equivalent to:
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// println!("❤\\n!");
|
|
|
|
|
/// ```
|
|
|
|
|
///
|
|
|
|
|
/// Using `to_string`:
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// assert_eq!("❤\n!".escape_debug().to_string(), "❤\\n!");
|
|
|
|
|
/// ```
|
2019-02-02 09:34:36 +00:00
|
|
|
|
#[stable(feature = "str_escape", since = "1.34.0")]
|
2019-04-18 23:37:12 +00:00
|
|
|
|
pub fn escape_debug(&self) -> EscapeDebug<'_> {
|
2019-02-02 09:34:36 +00:00
|
|
|
|
let mut chars = self.chars();
|
|
|
|
|
EscapeDebug {
|
2019-12-22 22:42:04 +00:00
|
|
|
|
inner: chars
|
|
|
|
|
.next()
|
2019-02-02 09:34:36 +00:00
|
|
|
|
.map(|first| first.escape_debug_ext(true))
|
|
|
|
|
.into_iter()
|
|
|
|
|
.flatten()
|
2019-12-22 22:42:04 +00:00
|
|
|
|
.chain(chars.flat_map(CharEscapeDebugContinue)),
|
2019-02-02 09:34:36 +00:00
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
2019-02-20 14:11:22 +00:00
|
|
|
|
/// Return an iterator that escapes each char in `self` with [`char::escape_default`].
|
2019-02-02 09:34:36 +00:00
|
|
|
|
///
|
|
|
|
|
/// [`char::escape_default`]: ../std/primitive.char.html#method.escape_default
|
2019-02-02 10:25:46 +00:00
|
|
|
|
///
|
|
|
|
|
/// # Examples
|
|
|
|
|
///
|
|
|
|
|
/// As an iterator:
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// for c in "❤\n!".escape_default() {
|
|
|
|
|
/// print!("{}", c);
|
|
|
|
|
/// }
|
|
|
|
|
/// println!();
|
|
|
|
|
/// ```
|
|
|
|
|
///
|
|
|
|
|
/// Using `println!` directly:
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// println!("{}", "❤\n!".escape_default());
|
|
|
|
|
/// ```
|
|
|
|
|
///
|
|
|
|
|
///
|
|
|
|
|
/// Both are equivalent to:
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
2019-04-13 19:44:21 +00:00
|
|
|
|
/// println!("\\u{{2764}}\\n!");
|
2019-02-02 10:25:46 +00:00
|
|
|
|
/// ```
|
|
|
|
|
///
|
|
|
|
|
/// Using `to_string`:
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// assert_eq!("❤\n!".escape_default().to_string(), "\\u{2764}\\n!");
|
|
|
|
|
/// ```
|
2019-02-02 09:34:36 +00:00
|
|
|
|
#[stable(feature = "str_escape", since = "1.34.0")]
|
2019-04-18 23:37:12 +00:00
|
|
|
|
pub fn escape_default(&self) -> EscapeDefault<'_> {
|
2019-02-02 09:34:36 +00:00
|
|
|
|
EscapeDefault { inner: self.chars().flat_map(CharEscapeDefault) }
|
|
|
|
|
}
|
|
|
|
|
|
2019-02-20 14:11:22 +00:00
|
|
|
|
/// Return an iterator that escapes each char in `self` with [`char::escape_unicode`].
|
2019-02-02 09:34:36 +00:00
|
|
|
|
///
|
|
|
|
|
/// [`char::escape_unicode`]: ../std/primitive.char.html#method.escape_unicode
|
2019-02-02 10:25:46 +00:00
|
|
|
|
///
|
|
|
|
|
/// # Examples
|
|
|
|
|
///
|
|
|
|
|
/// As an iterator:
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// for c in "❤\n!".escape_unicode() {
|
|
|
|
|
/// print!("{}", c);
|
|
|
|
|
/// }
|
|
|
|
|
/// println!();
|
|
|
|
|
/// ```
|
|
|
|
|
///
|
|
|
|
|
/// Using `println!` directly:
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// println!("{}", "❤\n!".escape_unicode());
|
|
|
|
|
/// ```
|
|
|
|
|
///
|
|
|
|
|
///
|
|
|
|
|
/// Both are equivalent to:
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// println!("\\u{{2764}}\\u{{a}}\\u{{21}}");
|
|
|
|
|
/// ```
|
|
|
|
|
///
|
|
|
|
|
/// Using `to_string`:
|
|
|
|
|
///
|
|
|
|
|
/// ```
|
|
|
|
|
/// assert_eq!("❤\n!".escape_unicode().to_string(), "\\u{2764}\\u{a}\\u{21}");
|
|
|
|
|
/// ```
|
2019-02-02 09:34:36 +00:00
|
|
|
|
#[stable(feature = "str_escape", since = "1.34.0")]
|
2019-04-18 23:37:12 +00:00
|
|
|
|
pub fn escape_unicode(&self) -> EscapeUnicode<'_> {
|
2019-02-02 09:34:36 +00:00
|
|
|
|
EscapeUnicode { inner: self.chars().flat_map(CharEscapeUnicode) }
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
impl_fn_for_zst! {
|
|
|
|
|
#[derive(Clone)]
|
|
|
|
|
struct CharEscapeDebugContinue impl Fn = |c: char| -> char::EscapeDebug {
|
|
|
|
|
c.escape_debug_ext(false)
|
|
|
|
|
};
|
|
|
|
|
|
|
|
|
|
#[derive(Clone)]
|
|
|
|
|
struct CharEscapeUnicode impl Fn = |c: char| -> char::EscapeUnicode {
|
|
|
|
|
c.escape_unicode()
|
|
|
|
|
};
|
|
|
|
|
#[derive(Clone)]
|
|
|
|
|
struct CharEscapeDefault impl Fn = |c: char| -> char::EscapeDefault {
|
|
|
|
|
c.escape_default()
|
|
|
|
|
};
|
2014-05-01 06:06:36 +00:00
|
|
|
|
}
|
|
|
|
|
|
2015-05-06 22:53:34 +00:00
|
|
|
|
#[stable(feature = "rust1", since = "1.0.0")]
|
|
|
|
|
impl AsRef<[u8]> for str {
|
|
|
|
|
#[inline]
|
|
|
|
|
fn as_ref(&self) -> &[u8] {
|
|
|
|
|
self.as_bytes()
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
2015-01-24 05:48:20 +00:00
|
|
|
|
#[stable(feature = "rust1", since = "1.0.0")]
|
2018-09-03 11:50:14 +00:00
|
|
|
|
impl Default for &str {
|
2016-09-11 11:30:09 +00:00
|
|
|
|
/// Creates an empty str
|
2019-12-22 22:42:04 +00:00
|
|
|
|
fn default() -> Self {
|
|
|
|
|
""
|
|
|
|
|
}
|
2014-05-01 06:06:36 +00:00
|
|
|
|
}
|
2018-04-05 17:00:48 +00:00
|
|
|
|
|
2018-06-02 16:29:50 +00:00
|
|
|
|
#[stable(feature = "default_mut_str", since = "1.28.0")]
|
2018-09-03 11:50:14 +00:00
|
|
|
|
impl Default for &mut str {
|
2018-06-02 16:29:50 +00:00
|
|
|
|
/// Creates an empty mutable str
|
2019-11-20 02:28:32 +00:00
|
|
|
|
// SAFETY: `str` is guranteed to be UTF-8
|
2019-12-22 22:42:04 +00:00
|
|
|
|
fn default() -> Self {
|
|
|
|
|
unsafe { from_utf8_unchecked_mut(&mut []) }
|
|
|
|
|
}
|
2018-06-02 16:29:50 +00:00
|
|
|
|
}
|
|
|
|
|
|
2018-04-05 17:00:48 +00:00
|
|
|
|
/// An iterator over the non-whitespace substrings of a string,
|
|
|
|
|
/// separated by any amount of whitespace.
|
|
|
|
|
///
|
|
|
|
|
/// This struct is created by the [`split_whitespace`] method on [`str`].
|
|
|
|
|
/// See its documentation for more.
|
|
|
|
|
///
|
|
|
|
|
/// [`split_whitespace`]: ../../std/primitive.str.html#method.split_whitespace
|
|
|
|
|
/// [`str`]: ../../std/primitive.str.html
|
|
|
|
|
#[stable(feature = "split_whitespace", since = "1.1.0")]
|
|
|
|
|
#[derive(Clone, Debug)]
|
|
|
|
|
pub struct SplitWhitespace<'a> {
|
|
|
|
|
inner: Filter<Split<'a, IsWhitespace>, IsNotEmpty>,
|
|
|
|
|
}
|
|
|
|
|
|
2018-05-05 04:33:20 +00:00
|
|
|
|
/// An iterator over the non-ASCII-whitespace substrings of a string,
|
|
|
|
|
/// separated by any amount of ASCII whitespace.
|
|
|
|
|
///
|
|
|
|
|
/// This struct is created by the [`split_ascii_whitespace`] method on [`str`].
|
|
|
|
|
/// See its documentation for more.
|
|
|
|
|
///
|
|
|
|
|
/// [`split_ascii_whitespace`]: ../../std/primitive.str.html#method.split_ascii_whitespace
|
|
|
|
|
/// [`str`]: ../../std/primitive.str.html
|
2019-02-01 10:24:12 +00:00
|
|
|
|
#[stable(feature = "split_ascii_whitespace", since = "1.34.0")]
|
2018-05-05 04:33:20 +00:00
|
|
|
|
#[derive(Clone, Debug)]
|
|
|
|
|
pub struct SplitAsciiWhitespace<'a> {
|
2019-02-01 11:43:29 +00:00
|
|
|
|
inner: Map<Filter<SliceSplit<'a, u8, IsAsciiWhitespace>, BytesIsNotEmpty>, UnsafeBytesToStr>,
|
2018-05-05 04:33:20 +00:00
|
|
|
|
}
|
|
|
|
|
|
2019-02-01 11:43:29 +00:00
|
|
|
|
impl_fn_for_zst! {
|
|
|
|
|
#[derive(Clone)]
|
|
|
|
|
struct IsWhitespace impl Fn = |c: char| -> bool {
|
|
|
|
|
c.is_whitespace()
|
|
|
|
|
};
|
2018-04-05 17:00:48 +00:00
|
|
|
|
|
2019-02-01 11:43:29 +00:00
|
|
|
|
#[derive(Clone)]
|
|
|
|
|
struct IsAsciiWhitespace impl Fn = |byte: &u8| -> bool {
|
|
|
|
|
byte.is_ascii_whitespace()
|
|
|
|
|
};
|
2018-05-05 04:33:20 +00:00
|
|
|
|
|
2019-02-01 11:43:29 +00:00
|
|
|
|
#[derive(Clone)]
|
|
|
|
|
struct IsNotEmpty impl<'a, 'b> Fn = |s: &'a &'b str| -> bool {
|
|
|
|
|
!s.is_empty()
|
|
|
|
|
};
|
2018-05-05 04:33:20 +00:00
|
|
|
|
|
2019-02-01 11:43:29 +00:00
|
|
|
|
#[derive(Clone)]
|
|
|
|
|
struct BytesIsNotEmpty impl<'a, 'b> Fn = |s: &'a &'b [u8]| -> bool {
|
|
|
|
|
!s.is_empty()
|
|
|
|
|
};
|
2018-05-05 04:33:20 +00:00
|
|
|
|
|
2019-02-01 11:43:29 +00:00
|
|
|
|
#[derive(Clone)]
|
|
|
|
|
struct UnsafeBytesToStr impl<'a> Fn = |bytes: &'a [u8]| -> &'a str {
|
2019-11-20 02:28:32 +00:00
|
|
|
|
// SAFETY: not safe
|
2019-02-01 11:43:29 +00:00
|
|
|
|
unsafe { from_utf8_unchecked(bytes) }
|
|
|
|
|
};
|
2018-05-05 04:33:20 +00:00
|
|
|
|
}
|
|
|
|
|
|
2018-04-05 17:00:48 +00:00
|
|
|
|
#[stable(feature = "split_whitespace", since = "1.1.0")]
|
|
|
|
|
impl<'a> Iterator for SplitWhitespace<'a> {
|
|
|
|
|
type Item = &'a str;
|
|
|
|
|
|
2018-05-05 04:33:20 +00:00
|
|
|
|
#[inline]
|
2018-04-05 17:00:48 +00:00
|
|
|
|
fn next(&mut self) -> Option<&'a str> {
|
|
|
|
|
self.inner.next()
|
|
|
|
|
}
|
2018-05-05 04:33:20 +00:00
|
|
|
|
|
|
|
|
|
#[inline]
|
|
|
|
|
fn size_hint(&self) -> (usize, Option<usize>) {
|
|
|
|
|
self.inner.size_hint()
|
|
|
|
|
}
|
When possible without changing semantics, implement Iterator::last in terms of DoubleEndedIterator::next_back for types in liballoc and libcore.
Provided that the iterator has finite length and does not trigger user-provided code, this is safe.
What follows is a full list of the DoubleEndedIterators in liballoc/libcore and whether this optimization is safe, and if not, why not.
src/liballoc/boxed.rs
Box: Pass through to avoid defeating optimization of the underlying DoubleIterator implementation. This has no correctness impact.
src/liballoc/collections/binary_heap.rs
Iter: Pass through to avoid defeating optimizations on slice::Iter
IntoIter: Not safe, changes Drop order
Drain: Not safe, changes Drop order
src/liballoc/collections/btree/map.rs
Iter: Safe to call next_back, invokes no user defined code.
IterMut: ditto
IntoIter: Not safe, changes Drop order
Keys: Safe to call next_back, invokes no user defined code.
Values: ditto
ValuesMut: ditto
Range: ditto
RangeMut: ditto
src/liballoc/collections/btree/set.rs
Iter: Safe to call next_back, invokes no user defined code.
IntoIter: Not safe, changes Drop order
Range: Safe to call next_back, invokes no user defined code.
src/liballoc/collections/linked_list.rs
Iter: Safe to call next_back, invokes no user defined code.
IterMut: ditto
IntoIter: Not safe, changes Drop order
src/liballoc/collections/vec_deque.rs
Iter: Safe to call next_back, invokes no user defined code.
IterMut: ditto
IntoIter: Not safe, changes Drop order
Drain: ditto
src/liballoc/string.rs
Drain: Safe because return type is a primitive (char)
src/liballoc/vec.rs
IntoIter: Not safe, changes Drop order
Drain: ditto
Splice: ditto
src/libcore/ascii.rs
EscapeDefault: Safe because return type is a primitive (u8)
src/libcore/iter/adapters/chain.rs
Chain: Not safe, invokes user defined code (Iterator impl)
src/libcore/iter/adapters/flatten.rs
FlatMap: Not safe, invokes user defined code (Iterator impl)
Flatten: ditto
FlattenCompat: ditto
src/libcore/iter/adapters/mod.rs
Rev: Not safe, invokes user defined code (Iterator impl)
Copied: ditto
Cloned: Not safe, invokes user defined code (Iterator impl and T::clone)
Map: Not safe, invokes user defined code (Iterator impl + closure)
Filter: ditto
FilterMap: ditto
Enumerate: Not safe, invokes user defined code (Iterator impl)
Skip: ditto
Fuse: ditto
Inspect: ditto
src/libcore/iter/adapters/zip.rs
Zip: Not safe, invokes user defined code (Iterator impl)
src/libcore/iter/range.rs
ops::Range: Not safe, changes Drop order, but ALREADY HAS SPECIALIZATION
ops::RangeInclusive: ditto
src/libcore/iter/sources.rs
Repeat: Not safe, calling last should iloop.
Empty: No point, iterator is at most one item long.
Once: ditto
OnceWith: ditto
src/libcore/option.rs
Item: No point, iterator is at most one item long.
Iter: ditto
IterMut: ditto
IntoIter: ditto
src/libcore/result.rs
Iter: No point, iterator is at most one item long
IterMut: ditto
IntoIter: ditto
src/libcore/slice/mod.rs
Split: Not safe, invokes user defined closure
SplitMut: ditto
RSplit: ditto
RSplitMut: ditto
Windows: Safe, already has specialization
Chunks: ditto
ChunksMut: ditto
ChunksExact: ditto
ChunksExactMut: ditto
RChunks: ditto
RChunksMut: ditto
RChunksExact: ditto
RChunksExactMut: ditto
src/libcore/str/mod.rs
Chars: Safe, already has specialization
CharIndices: ditto
Bytes: ditto
Lines: Safe to call next_back, invokes no user defined code.
LinesAny: Deprecated
Everything that is generic over P: Pattern: Not safe because Pattern invokes user defined code.
SplitWhitespace: Safe to call next_back, invokes no user defined code.
SplitAsciiWhitespace: ditto
2019-07-02 20:45:29 +00:00
|
|
|
|
|
|
|
|
|
#[inline]
|
|
|
|
|
fn last(mut self) -> Option<&'a str> {
|
|
|
|
|
self.next_back()
|
|
|
|
|
}
|
2018-04-05 17:00:48 +00:00
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
#[stable(feature = "split_whitespace", since = "1.1.0")]
|
|
|
|
|
impl<'a> DoubleEndedIterator for SplitWhitespace<'a> {
|
2018-05-05 04:33:20 +00:00
|
|
|
|
#[inline]
|
2018-04-05 17:00:48 +00:00
|
|
|
|
fn next_back(&mut self) -> Option<&'a str> {
|
|
|
|
|
self.inner.next_back()
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
#[stable(feature = "fused", since = "1.26.0")]
|
2018-09-03 11:50:14 +00:00
|
|
|
|
impl FusedIterator for SplitWhitespace<'_> {}
|
2018-04-07 19:56:02 +00:00
|
|
|
|
|
2019-02-01 10:24:12 +00:00
|
|
|
|
#[stable(feature = "split_ascii_whitespace", since = "1.34.0")]
|
2018-05-05 04:33:20 +00:00
|
|
|
|
impl<'a> Iterator for SplitAsciiWhitespace<'a> {
|
|
|
|
|
type Item = &'a str;
|
|
|
|
|
|
|
|
|
|
#[inline]
|
|
|
|
|
fn next(&mut self) -> Option<&'a str> {
|
|
|
|
|
self.inner.next()
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
#[inline]
|
|
|
|
|
fn size_hint(&self) -> (usize, Option<usize>) {
|
|
|
|
|
self.inner.size_hint()
|
|
|
|
|
}
|
When possible without changing semantics, implement Iterator::last in terms of DoubleEndedIterator::next_back for types in liballoc and libcore.
Provided that the iterator has finite length and does not trigger user-provided code, this is safe.
What follows is a full list of the DoubleEndedIterators in liballoc/libcore and whether this optimization is safe, and if not, why not.
src/liballoc/boxed.rs
Box: Pass through to avoid defeating optimization of the underlying DoubleIterator implementation. This has no correctness impact.
src/liballoc/collections/binary_heap.rs
Iter: Pass through to avoid defeating optimizations on slice::Iter
IntoIter: Not safe, changes Drop order
Drain: Not safe, changes Drop order
src/liballoc/collections/btree/map.rs
Iter: Safe to call next_back, invokes no user defined code.
IterMut: ditto
IntoIter: Not safe, changes Drop order
Keys: Safe to call next_back, invokes no user defined code.
Values: ditto
ValuesMut: ditto
Range: ditto
RangeMut: ditto
src/liballoc/collections/btree/set.rs
Iter: Safe to call next_back, invokes no user defined code.
IntoIter: Not safe, changes Drop order
Range: Safe to call next_back, invokes no user defined code.
src/liballoc/collections/linked_list.rs
Iter: Safe to call next_back, invokes no user defined code.
IterMut: ditto
IntoIter: Not safe, changes Drop order
src/liballoc/collections/vec_deque.rs
Iter: Safe to call next_back, invokes no user defined code.
IterMut: ditto
IntoIter: Not safe, changes Drop order
Drain: ditto
src/liballoc/string.rs
Drain: Safe because return type is a primitive (char)
src/liballoc/vec.rs
IntoIter: Not safe, changes Drop order
Drain: ditto
Splice: ditto
src/libcore/ascii.rs
EscapeDefault: Safe because return type is a primitive (u8)
src/libcore/iter/adapters/chain.rs
Chain: Not safe, invokes user defined code (Iterator impl)
src/libcore/iter/adapters/flatten.rs
FlatMap: Not safe, invokes user defined code (Iterator impl)
Flatten: ditto
FlattenCompat: ditto
src/libcore/iter/adapters/mod.rs
Rev: Not safe, invokes user defined code (Iterator impl)
Copied: ditto
Cloned: Not safe, invokes user defined code (Iterator impl and T::clone)
Map: Not safe, invokes user defined code (Iterator impl + closure)
Filter: ditto
FilterMap: ditto
Enumerate: Not safe, invokes user defined code (Iterator impl)
Skip: ditto
Fuse: ditto
Inspect: ditto
src/libcore/iter/adapters/zip.rs
Zip: Not safe, invokes user defined code (Iterator impl)
src/libcore/iter/range.rs
ops::Range: Not safe, changes Drop order, but ALREADY HAS SPECIALIZATION
ops::RangeInclusive: ditto
src/libcore/iter/sources.rs
Repeat: Not safe, calling last should iloop.
Empty: No point, iterator is at most one item long.
Once: ditto
OnceWith: ditto
src/libcore/option.rs
Item: No point, iterator is at most one item long.
Iter: ditto
IterMut: ditto
IntoIter: ditto
src/libcore/result.rs
Iter: No point, iterator is at most one item long
IterMut: ditto
IntoIter: ditto
src/libcore/slice/mod.rs
Split: Not safe, invokes user defined closure
SplitMut: ditto
RSplit: ditto
RSplitMut: ditto
Windows: Safe, already has specialization
Chunks: ditto
ChunksMut: ditto
ChunksExact: ditto
ChunksExactMut: ditto
RChunks: ditto
RChunksMut: ditto
RChunksExact: ditto
RChunksExactMut: ditto
src/libcore/str/mod.rs
Chars: Safe, already has specialization
CharIndices: ditto
Bytes: ditto
Lines: Safe to call next_back, invokes no user defined code.
LinesAny: Deprecated
Everything that is generic over P: Pattern: Not safe because Pattern invokes user defined code.
SplitWhitespace: Safe to call next_back, invokes no user defined code.
SplitAsciiWhitespace: ditto
2019-07-02 20:45:29 +00:00
|
|
|
|
|
|
|
|
|
#[inline]
|
|
|
|
|
fn last(mut self) -> Option<&'a str> {
|
|
|
|
|
self.next_back()
|
|
|
|
|
}
|
2018-05-05 04:33:20 +00:00
|
|
|
|
}
|
|
|
|
|
|
2019-02-01 10:24:12 +00:00
|
|
|
|
#[stable(feature = "split_ascii_whitespace", since = "1.34.0")]
|
2018-05-05 04:33:20 +00:00
|
|
|
|
impl<'a> DoubleEndedIterator for SplitAsciiWhitespace<'a> {
|
|
|
|
|
#[inline]
|
|
|
|
|
fn next_back(&mut self) -> Option<&'a str> {
|
|
|
|
|
self.inner.next_back()
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
2019-02-01 10:24:12 +00:00
|
|
|
|
#[stable(feature = "split_ascii_whitespace", since = "1.34.0")]
|
2018-09-03 11:50:14 +00:00
|
|
|
|
impl FusedIterator for SplitAsciiWhitespace<'_> {}
|
2018-05-05 04:33:20 +00:00
|
|
|
|
|
2018-04-07 19:56:02 +00:00
|
|
|
|
/// An iterator of [`u16`] over the string encoded as UTF-16.
|
|
|
|
|
///
|
|
|
|
|
/// [`u16`]: ../../std/primitive.u16.html
|
|
|
|
|
///
|
|
|
|
|
/// This struct is created by the [`encode_utf16`] method on [`str`].
|
|
|
|
|
/// See its documentation for more.
|
|
|
|
|
///
|
|
|
|
|
/// [`encode_utf16`]: ../../std/primitive.str.html#method.encode_utf16
|
|
|
|
|
/// [`str`]: ../../std/primitive.str.html
|
|
|
|
|
#[derive(Clone)]
|
|
|
|
|
#[stable(feature = "encode_utf16", since = "1.8.0")]
|
|
|
|
|
pub struct EncodeUtf16<'a> {
|
|
|
|
|
chars: Chars<'a>,
|
|
|
|
|
extra: u16,
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
#[stable(feature = "collection_debug", since = "1.17.0")]
|
2018-09-03 11:50:14 +00:00
|
|
|
|
impl fmt::Debug for EncodeUtf16<'_> {
|
2019-04-18 23:37:12 +00:00
|
|
|
|
fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
|
2018-04-07 19:56:02 +00:00
|
|
|
|
f.pad("EncodeUtf16 { .. }")
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
#[stable(feature = "encode_utf16", since = "1.8.0")]
|
|
|
|
|
impl<'a> Iterator for EncodeUtf16<'a> {
|
|
|
|
|
type Item = u16;
|
|
|
|
|
|
|
|
|
|
#[inline]
|
|
|
|
|
fn next(&mut self) -> Option<u16> {
|
|
|
|
|
if self.extra != 0 {
|
|
|
|
|
let tmp = self.extra;
|
|
|
|
|
self.extra = 0;
|
|
|
|
|
return Some(tmp);
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
let mut buf = [0; 2];
|
|
|
|
|
self.chars.next().map(|ch| {
|
|
|
|
|
let n = ch.encode_utf16(&mut buf).len();
|
|
|
|
|
if n == 2 {
|
|
|
|
|
self.extra = buf[1];
|
|
|
|
|
}
|
|
|
|
|
buf[0]
|
|
|
|
|
})
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
#[inline]
|
|
|
|
|
fn size_hint(&self) -> (usize, Option<usize>) {
|
|
|
|
|
let (low, high) = self.chars.size_hint();
|
|
|
|
|
// every char gets either one u16 or two u16,
|
|
|
|
|
// so this iterator is between 1 or 2 times as
|
|
|
|
|
// long as the underlying iterator.
|
|
|
|
|
(low, high.and_then(|n| n.checked_mul(2)))
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
#[stable(feature = "fused", since = "1.26.0")]
|
2018-09-03 11:50:14 +00:00
|
|
|
|
impl FusedIterator for EncodeUtf16<'_> {}
|
2019-02-02 09:34:36 +00:00
|
|
|
|
|
|
|
|
|
/// The return type of [`str::escape_debug`].
|
|
|
|
|
///
|
|
|
|
|
/// [`str::escape_debug`]: ../../std/primitive.str.html#method.escape_debug
|
|
|
|
|
#[stable(feature = "str_escape", since = "1.34.0")]
|
|
|
|
|
#[derive(Clone, Debug)]
|
|
|
|
|
pub struct EscapeDebug<'a> {
|
|
|
|
|
inner: Chain<
|
|
|
|
|
Flatten<option::IntoIter<char::EscapeDebug>>,
|
2019-12-22 22:42:04 +00:00
|
|
|
|
FlatMap<Chars<'a>, char::EscapeDebug, CharEscapeDebugContinue>,
|
2019-02-02 09:34:36 +00:00
|
|
|
|
>,
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/// The return type of [`str::escape_default`].
|
|
|
|
|
///
|
|
|
|
|
/// [`str::escape_default`]: ../../std/primitive.str.html#method.escape_default
|
|
|
|
|
#[stable(feature = "str_escape", since = "1.34.0")]
|
|
|
|
|
#[derive(Clone, Debug)]
|
|
|
|
|
pub struct EscapeDefault<'a> {
|
|
|
|
|
inner: FlatMap<Chars<'a>, char::EscapeDefault, CharEscapeDefault>,
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/// The return type of [`str::escape_unicode`].
|
|
|
|
|
///
|
|
|
|
|
/// [`str::escape_unicode`]: ../../std/primitive.str.html#method.escape_unicode
|
|
|
|
|
#[stable(feature = "str_escape", since = "1.34.0")]
|
|
|
|
|
#[derive(Clone, Debug)]
|
|
|
|
|
pub struct EscapeUnicode<'a> {
|
|
|
|
|
inner: FlatMap<Chars<'a>, char::EscapeUnicode, CharEscapeUnicode>,
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
macro_rules! escape_types_impls {
|
|
|
|
|
($( $Name: ident ),+) => {$(
|
|
|
|
|
#[stable(feature = "str_escape", since = "1.34.0")]
|
|
|
|
|
impl<'a> fmt::Display for $Name<'a> {
|
2019-04-18 23:37:12 +00:00
|
|
|
|
fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
|
2019-02-02 09:34:36 +00:00
|
|
|
|
self.clone().try_for_each(|c| f.write_char(c))
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
#[stable(feature = "str_escape", since = "1.34.0")]
|
|
|
|
|
impl<'a> Iterator for $Name<'a> {
|
|
|
|
|
type Item = char;
|
|
|
|
|
|
|
|
|
|
#[inline]
|
|
|
|
|
fn next(&mut self) -> Option<char> { self.inner.next() }
|
|
|
|
|
|
|
|
|
|
#[inline]
|
|
|
|
|
fn size_hint(&self) -> (usize, Option<usize>) { self.inner.size_hint() }
|
|
|
|
|
|
|
|
|
|
#[inline]
|
|
|
|
|
fn try_fold<Acc, Fold, R>(&mut self, init: Acc, fold: Fold) -> R where
|
|
|
|
|
Self: Sized, Fold: FnMut(Acc, Self::Item) -> R, R: Try<Ok=Acc>
|
|
|
|
|
{
|
|
|
|
|
self.inner.try_fold(init, fold)
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
#[inline]
|
|
|
|
|
fn fold<Acc, Fold>(self, init: Acc, fold: Fold) -> Acc
|
|
|
|
|
where Fold: FnMut(Acc, Self::Item) -> Acc,
|
|
|
|
|
{
|
|
|
|
|
self.inner.fold(init, fold)
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
#[stable(feature = "str_escape", since = "1.34.0")]
|
|
|
|
|
impl<'a> FusedIterator for $Name<'a> {}
|
|
|
|
|
)+}
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
escape_types_impls!(EscapeDebug, EscapeDefault, EscapeUnicode);
|