1 | use std::path::Path; |
2 | use std::str::FromStr; |
3 | |
4 | use crate::common::{ |
5 | parse_break_test, parse_codepoint_association, CodepointIter, Codepoints, |
6 | UcdFile, UcdFileByCodepoint, |
7 | }; |
8 | use crate::error::Error; |
9 | |
10 | /// A single row in the `auxiliary/WordBreakProperty.txt` file. |
11 | #[derive (Clone, Debug, Default, Eq, PartialEq)] |
12 | pub struct WordBreak { |
13 | /// The codepoint or codepoint range for this entry. |
14 | pub codepoints: Codepoints, |
15 | /// The property value assigned to the codepoints in this entry. |
16 | pub value: String, |
17 | } |
18 | |
19 | impl UcdFile for WordBreak { |
20 | fn relative_file_path() -> &'static Path { |
21 | Path::new("auxiliary/WordBreakProperty.txt" ) |
22 | } |
23 | } |
24 | |
25 | impl UcdFileByCodepoint for WordBreak { |
26 | fn codepoints(&self) -> CodepointIter { |
27 | self.codepoints.into_iter() |
28 | } |
29 | } |
30 | |
31 | impl FromStr for WordBreak { |
32 | type Err = Error; |
33 | |
34 | fn from_str(line: &str) -> Result<WordBreak, Error> { |
35 | let (codepoints: Codepoints, value: &str) = parse_codepoint_association(line)?; |
36 | Ok(WordBreak { codepoints, value: value.to_string() }) |
37 | } |
38 | } |
39 | |
40 | /// A single row in the `auxiliary/WordBreakTest.txt` file. |
41 | /// |
42 | /// This file defines tests for the word break algorithm. |
43 | #[derive (Clone, Debug, Default, Eq, PartialEq)] |
44 | pub struct WordBreakTest { |
45 | /// Each string is a UTF-8 encoded group of codepoints that make up a |
46 | /// single word. |
47 | pub words: Vec<String>, |
48 | /// A human readable description of this test. |
49 | pub comment: String, |
50 | } |
51 | |
52 | impl UcdFile for WordBreakTest { |
53 | fn relative_file_path() -> &'static Path { |
54 | Path::new("auxiliary/WordBreakTest.txt" ) |
55 | } |
56 | } |
57 | |
58 | impl FromStr for WordBreakTest { |
59 | type Err = Error; |
60 | |
61 | fn from_str(line: &str) -> Result<WordBreakTest, Error> { |
62 | let (groups: Vec, comment: String) = parse_break_test(line)?; |
63 | Ok(WordBreakTest { words: groups, comment }) |
64 | } |
65 | } |
66 | |
67 | #[cfg (test)] |
68 | mod tests { |
69 | use super::{WordBreak, WordBreakTest}; |
70 | |
71 | #[test ] |
72 | fn parse_single() { |
73 | let line = "0A83 ; Extend # Mc GUJARATI SIGN VISARGA \n" ; |
74 | let row: WordBreak = line.parse().unwrap(); |
75 | assert_eq!(row.codepoints, 0x0A83); |
76 | assert_eq!(row.value, "Extend" ); |
77 | } |
78 | |
79 | #[test ] |
80 | fn parse_range() { |
81 | let line = "104A0..104A9 ; Numeric # Nd [10] OSMANYA DIGIT ZERO..OSMANYA DIGIT NINE \n" ; |
82 | let row: WordBreak = line.parse().unwrap(); |
83 | assert_eq!(row.codepoints, (0x104A0, 0x104A9)); |
84 | assert_eq!(row.value, "Numeric" ); |
85 | } |
86 | |
87 | #[test ] |
88 | fn parse_test() { |
89 | let line = "÷ 0031 ÷ 0027 × 0308 ÷ 0061 ÷ 0027 × 2060 ÷ # ÷ [0.2] DIGIT ONE (Numeric) ÷ [999.0] APOSTROPHE (Single_Quote) × [4.0] COMBINING DIAERESIS (Extend_FE) ÷ [999.0] LATIN SMALL LETTER A (ALetter) ÷ [999.0] APOSTROPHE (Single_Quote) × [4.0] WORD JOINER (Format_FE) ÷ [0.3]" ; |
90 | |
91 | let row: WordBreakTest = line.parse().unwrap(); |
92 | assert_eq!( |
93 | row.words, |
94 | vec![ |
95 | " \u{0031}" , |
96 | " \u{0027}\u{0308}" , |
97 | " \u{0061}" , |
98 | " \u{0027}\u{2060}" , |
99 | ] |
100 | ); |
101 | assert!(row.comment.contains("[4.0] COMBINING DIAERESIS (Extend_FE)" )); |
102 | } |
103 | } |
104 | |