ucd_parse/
sentence_break.rs

1use std::path::Path;
2
3use crate::{
4    common::{
5        parse_break_test, parse_codepoint_association, CodepointIter,
6        Codepoints, UcdFile, UcdFileByCodepoint,
7    },
8    error::Error,
9};
10
11/// A single row in the `auxiliary/SentenceBreakProperty.txt` file.
12#[derive(Clone, Debug, Default, Eq, PartialEq)]
13pub struct SentenceBreak {
14    /// The codepoint or codepoint range for this entry.
15    pub codepoints: Codepoints,
16    /// The property value assigned to the codepoints in this entry.
17    pub value: String,
18}
19
20impl UcdFile for SentenceBreak {
21    fn relative_file_path() -> &'static Path {
22        Path::new("auxiliary/SentenceBreakProperty.txt")
23    }
24}
25
26impl UcdFileByCodepoint for SentenceBreak {
27    fn codepoints(&self) -> CodepointIter {
28        self.codepoints.into_iter()
29    }
30}
31
32impl std::str::FromStr for SentenceBreak {
33    type Err = Error;
34
35    fn from_str(line: &str) -> Result<SentenceBreak, Error> {
36        let (codepoints, value) = parse_codepoint_association(line)?;
37        Ok(SentenceBreak { codepoints, value: value.to_string() })
38    }
39}
40
41/// A single row in the `auxiliary/SentenceBreakTest.txt` file.
42///
43/// This file defines tests for the sentence break algorithm.
44#[derive(Clone, Debug, Default, Eq, PartialEq)]
45pub struct SentenceBreakTest {
46    /// Each string is a UTF-8 encoded group of codepoints that make up a
47    /// single sentence.
48    pub sentences: Vec<String>,
49    /// A human readable description of this test.
50    pub comment: String,
51}
52
53impl UcdFile for SentenceBreakTest {
54    fn relative_file_path() -> &'static Path {
55        Path::new("auxiliary/SentenceBreakTest.txt")
56    }
57}
58
59impl std::str::FromStr for SentenceBreakTest {
60    type Err = Error;
61
62    fn from_str(line: &str) -> Result<SentenceBreakTest, Error> {
63        let (groups, comment) = parse_break_test(line)?;
64        Ok(SentenceBreakTest { sentences: groups, comment })
65    }
66}
67
68#[cfg(test)]
69mod tests {
70    use super::{SentenceBreak, SentenceBreakTest};
71
72    #[test]
73    fn parse_single() {
74        let line = "11445         ; Extend # Mc       NEWA SIGN VISARGA\n";
75        let row: SentenceBreak = line.parse().unwrap();
76        assert_eq!(row.codepoints, 0x11445);
77        assert_eq!(row.value, "Extend");
78    }
79
80    #[test]
81    fn parse_range() {
82        let line = "FE31..FE32    ; SContinue # Pd   [2] PRESENTATION FORM FOR VERTICAL EM DASH..PRESENTATION FORM FOR VERTICAL EN DASH\n";
83        let row: SentenceBreak = line.parse().unwrap();
84        assert_eq!(row.codepoints, (0xFE31, 0xFE32));
85        assert_eq!(row.value, "SContinue");
86    }
87
88    #[test]
89    fn parse_test() {
90        let line = "÷ 2060 × 5B57 × 2060 × 002E × 2060 ÷ 5B57 × 2060 × 2060 ÷	#  ÷ [0.2] WORD JOINER (Format_FE) × [998.0] CJK UNIFIED IDEOGRAPH-5B57 (OLetter) × [5.0] WORD JOINER (Format_FE) × [998.0] FULL STOP (ATerm) × [5.0] WORD JOINER (Format_FE) ÷ [11.0] CJK UNIFIED IDEOGRAPH-5B57 (OLetter) × [5.0] WORD JOINER (Format_FE) × [5.0] WORD JOINER (Format_FE) ÷ [0.3]";
91
92        let row: SentenceBreakTest = line.parse().unwrap();
93        assert_eq!(
94            row.sentences,
95            vec![
96                "\u{2060}\u{5B57}\u{2060}\u{002E}\u{2060}",
97                "\u{5B57}\u{2060}\u{2060}",
98            ]
99        );
100        assert!(row.comment.contains("[5.0] WORD JOINER (Format_FE)"));
101    }
102}
ucd_parse/sentence_break.rs

ucd_parse/
sentence_break.rs
