char-ranges

Similar to the standard library's .char_indicies(), but instead of only producing the start byte position. This library implements .char_ranges(), that produce both the start and end byte positions.

Note that simply using .char_indicies() and creating a range by mapping the returned index i to i..(i + 1) is not guaranteed to be valid. Given that some UTF-8 characters can be up to 4 bytes.

Assumes encoded in UTF-8.

The implementation specializes last(), nth(), next_back(), and nth_back(). Such that the length of intermediate characters is not wastefully calculated.

Example

use char_ranges::CharRangesExt;

let text = "Hello 🗻∈🌏";

let mut chars = text.char_ranges();
assert_eq!(chars.as_str(), "Hello 🗻∈🌏");

assert_eq!(chars.next(), Some((0..1, 'H'))); // These chars are 1 byte
assert_eq!(chars.next(), Some((1..2, 'e')));
assert_eq!(chars.next(), Some((2..3, 'l')));
assert_eq!(chars.next(), Some((3..4, 'l')));
assert_eq!(chars.next(), Some((4..5, 'o')));
assert_eq!(chars.next(), Some((5..6, ' ')));

// Get the remaining substring
assert_eq!(chars.as_str(), "🗻∈🌏");

assert_eq!(chars.next(), Some((6..10, '🗻'))); // This char is 4 bytes
assert_eq!(chars.next(), Some((10..13, '∈'))); // This char is 3 bytes
assert_eq!(chars.next(), Some((13..17, '🌏'))); // This char is 4 bytes
assert_eq!(chars.next(), None);

DoubleEndedIterator

CharRanges also implements DoubleEndedIterator making it possible to iterate backwards.

use char_ranges::CharRangesExt;

let text = "ABCDE";

let mut chars = text.char_ranges();
assert_eq!(chars.as_str(), "ABCDE");

assert_eq!(chars.next(), Some((0..1, 'A')));
assert_eq!(chars.next_back(), Some((4..5, 'E')));
assert_eq!(chars.as_str(), "BCD");

assert_eq!(chars.next_back(), Some((3..4, 'D')));
assert_eq!(chars.next(), Some((1..2, 'B')));
assert_eq!(chars.as_str(), "C");

assert_eq!(chars.next(), Some((2..3, 'C')));
assert_eq!(chars.as_str(), "");

assert_eq!(chars.next(), None);

Offset Ranges

If the input text is a substring of some original text, and the produced ranges are desired to be offset in relation to the substring. Then instead of .char_ranges() use .char_ranges_offset(offset) or .char_ranges().offset(offset).

use char_ranges::CharRangesExt;

let text = "Hello 👋 World 🌏";

let start = 11; // Start index of 'W'
let text = &text[start..]; // "World 🌏"

let mut chars = text.char_ranges_offset(start);
// or
// let mut chars = text.char_ranges().offset(start);

assert_eq!(chars.next(), Some((11..12, 'W'))); // These chars are 1 byte
assert_eq!(chars.next(), Some((12..13, 'o')));
assert_eq!(chars.next(), Some((13..14, 'r')));

assert_eq!(chars.next_back(), Some((17..21, '🌏'))); // This char is 4 bytes