1 | use std::fmt; |
2 | use std::path::Path; |
3 | use std::str::FromStr; |
4 | |
5 | use once_cell::sync::Lazy; |
6 | use regex::Regex; |
7 | |
8 | use crate::common::{Codepoint, CodepointIter, UcdFile, UcdFileByCodepoint}; |
9 | use crate::error::Error; |
10 | |
11 | /// Represents a single row in the `BidiMirroring.txt` file. |
12 | /// |
13 | /// The field names were taken from the header of BidiMirroring.txt. |
14 | #[derive (Clone, Debug, Default, Eq, PartialEq)] |
15 | pub struct BidiMirroring { |
16 | /// The codepoint corresponding to this row. |
17 | pub codepoint: Codepoint, |
18 | /// The codepoint that has typically has a glyph that is the mirror image |
19 | /// of `codepoint`. |
20 | pub bidi_mirroring_glyph: Codepoint, |
21 | } |
22 | |
23 | impl UcdFile for BidiMirroring { |
24 | fn relative_file_path() -> &'static Path { |
25 | Path::new("BidiMirroring.txt" ) |
26 | } |
27 | } |
28 | |
29 | impl UcdFileByCodepoint for BidiMirroring { |
30 | fn codepoints(&self) -> CodepointIter { |
31 | self.codepoint.into_iter() |
32 | } |
33 | } |
34 | |
35 | impl FromStr for BidiMirroring { |
36 | type Err = Error; |
37 | |
38 | fn from_str(line: &str) -> Result<BidiMirroring, Error> { |
39 | static PARTS: Lazy<Regex> = Lazy::new(|| { |
40 | Regex::new( |
41 | r"(?x) |
42 | ^ |
43 | \s*(?P<codepoint>[A-F0-9]+)\s*; |
44 | \s*(?P<substitute_codepoint>[A-F0-9]+) |
45 | \s+ |
46 | \#(?:.+) |
47 | $ |
48 | " , |
49 | ) |
50 | .unwrap() |
51 | }); |
52 | let caps = match PARTS.captures(line.trim()) { |
53 | Some(caps) => caps, |
54 | None => return err!("invalid BidiMirroring line" ), |
55 | }; |
56 | |
57 | Ok(BidiMirroring { |
58 | codepoint: caps["codepoint" ].parse()?, |
59 | bidi_mirroring_glyph: caps["substitute_codepoint" ].parse()?, |
60 | }) |
61 | } |
62 | } |
63 | |
64 | impl fmt::Display for BidiMirroring { |
65 | fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result { |
66 | write!(f, " {};" , self.codepoint)?; |
67 | write!(f, " {};" , self.bidi_mirroring_glyph)?; |
68 | Ok(()) |
69 | } |
70 | } |
71 | |
72 | #[cfg (test)] |
73 | mod tests { |
74 | use crate::common::Codepoint; |
75 | |
76 | use super::BidiMirroring; |
77 | |
78 | fn codepoint(n: u32) -> Codepoint { |
79 | Codepoint::from_u32(n).unwrap() |
80 | } |
81 | |
82 | #[test ] |
83 | fn parse() { |
84 | let line = "0028; 0029 # LEFT PARENTHESIS \n" ; |
85 | let data: BidiMirroring = line.parse().unwrap(); |
86 | assert_eq!( |
87 | data, |
88 | BidiMirroring { |
89 | codepoint: codepoint(0x0028), |
90 | bidi_mirroring_glyph: codepoint(0x0029), |
91 | } |
92 | ); |
93 | } |
94 | |
95 | #[test ] |
96 | fn parse_best_fit() { |
97 | let line = "228A; 228B # [BEST FIT] SUBSET OF WITH NOT EQUAL TO \n" ; |
98 | let data: BidiMirroring = line.parse().unwrap(); |
99 | assert_eq!( |
100 | data, |
101 | BidiMirroring { |
102 | codepoint: codepoint(0x228A), |
103 | bidi_mirroring_glyph: codepoint(0x228B), |
104 | } |
105 | ); |
106 | } |
107 | } |
108 | |