comfy_table/utils/formatting/content_split/normal.rs
1use unicode_segmentation::UnicodeSegmentation;
2use unicode_width::UnicodeWidthStr;
3
4/// returns printed length of string
5/// if ansi feature enabled, takes into account escape codes
6#[inline(always)]
7pub fn measure_text_width(s: &str) -> usize {
8 s.width()
9}
10
11/// Split a line into its individual parts along the given delimiter.
12pub fn split_line_by_delimiter(line: &str, delimiter: char) -> Vec<String> {
13 line.split(delimiter)
14 .map(ToString::to_string)
15 .collect::<Vec<String>>()
16}
17
18/// Splits a long word at a given character width.
19/// This needs some special logic, as we have to take multi-character UTF-8 symbols into account.
20/// When simply splitting at a certain char position, we might end up with a string that's has a
21/// wider display width than allowed.
22pub fn split_long_word(allowed_width: usize, word: &str) -> (String, String) {
23 let mut current_width = 0;
24 let mut parts = String::new();
25
26 let mut graphmes = word.graphemes(true).peekable();
27
28 // Check if the string might be too long, one Unicode grapheme at a time.
29 // Peek into the next grapheme and check the exit condition.
30 //
31 // This code uses graphemes to handle both zero-width joiner[0] UTF-8 chars, which
32 // combine multiple UTF-8 chars into a single grapheme, and variant selectors [1],
33 // which pick a certain variant of the preceding char.
34 //
35 // [0]: https://en.wikipedia.org/wiki/Zero-width_joiner
36 // [1]: https://en.wikipedia.org/wiki/Variation_Selectors_(Unicode_block)
37 while let Some(c) = graphmes.peek() {
38 if (current_width + c.width()) > allowed_width {
39 break;
40 }
41
42 // We can unwrap, as we just checked that a suitable grapheme is next in line.
43 let c = graphmes.next().unwrap();
44
45 let character_width = c.width();
46 current_width += character_width;
47 parts.push_str(c);
48 }
49
50 // Collect the remaining characters.
51 let remaining = graphmes.collect();
52 (parts, remaining)
53}