raoh 0.1.0

Decoders that turn untyped JSON into typed domain values, reporting every issue with its path
Documentation 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232

//! What the operations Raoh for Java takes from the JDK mean, written once.
//!
//! Where a built-in decoder does what Raoh for Java does with a JDK method, it calls the function
//! here rather than the Rust function that looks like it: `str::len` counts UTF-8 bytes where
//! `String.length()` counts UTF-16 units, `f64`'s `Display` writes `10000000.0` where
//! `Double.toString` writes `1.0E7`, and no Rust crate reads a `.properties` file as
//! `Properties.load` does. Each function names the JDK method it follows, and the tests pin the
//! inputs on which the two differ.
//!
//! Rules Raoh for Java defines itself rather than taking from the JDK are not here: whitespace is
//! Unicode's `White_Space`, which `str::trim` already follows; the IP grammar is in `json::ip`;
//! strings sort by code point, as `str` does.

/// `String.length()`: the number of UTF-16 code units.
pub(crate) fn utf16_len(s: &str) -> usize {
    s.encode_utf16().count()
}

/// `Double.toString(double)`: the shortest decimal that reads back as `v`, written plainly with at
/// least one fraction digit when 10⁻³ ≤ |v| < 10⁷ (`100.0`, `0.001`), and as `d.dddE±n`
/// otherwise (`1.0E7`, `1.0E-4`).
pub(crate) fn double_to_string(v: f64) -> String {
    if v.is_nan() {
        return "NaN".into();
    }
    if v.is_infinite() {
        return if v > 0.0 { "Infinity" } else { "-Infinity" }.into();
    }
    let sign = if v.is_sign_negative() { "-" } else { "" };
    if v == 0.0 {
        return format!("{sign}0.0");
    }
    // Rust's `{:e}` writes the same shortest digits Java chooses, as `d.ddde±n`.
    let scientific = format!("{:e}", v.abs());
    let (mantissa, exponent) = scientific
        .split_once('e')
        .expect("`{:e}` writes an exponent");
    let exponent: i32 = exponent.parse().expect("`{:e}` writes an integer exponent");
    let digits: String = mantissa.chars().filter(char::is_ascii_digit).collect();
    let magnitude = v.abs();
    let body = if (1e-3..1e7).contains(&magnitude) {
        // The decimal point goes after `exponent + 1` digits.
        let point = exponent + 1;
        if point <= 0 {
            format!("0.{}{digits}", "0".repeat(point.unsigned_abs() as usize))
        } else {
            let point = point as usize;
            if point >= digits.len() {
                format!("{digits}{}.0", "0".repeat(point - digits.len()))
            } else {
                format!("{}.{}", &digits[..point], &digits[point..])
            }
        }
    } else {
        let (first, rest) = digits.split_at(1);
        let rest = if rest.is_empty() { "0" } else { rest };
        format!("{first}.{rest}E{exponent}")
    };
    format!("{sign}{body}")
}

/// Where [`load_properties`] stopped reading.
#[derive(Debug)]
pub(crate) struct PropertiesError {
    pub(crate) line: usize,
    pub(crate) reason: &'static str,
}

/// `Properties.load(Reader)`: the key and value pairs of a `.properties` text, in order.
///
/// A logical line continues onto the next when it ends in an odd number of backslashes, and the
/// next line's leading whitespace is skipped. Lines that are blank or whose first character other
/// than whitespace is `#` or `!` are comments. A key ends at the first `=`, `:` or whitespace not
/// escaped; whitespace around the separator is skipped. Both key and value undo the escapes `\t`,
/// `\n`, `\r`, `\f` and `\uXXXX`, and a backslash before any other character stands for that
/// character. A `\u` not followed by four hexadecimal digits is an error, as in the JDK.
pub(crate) fn load_properties(text: &str) -> Result<Vec<(String, String)>, PropertiesError> {
    let text = text.replace("\r\n", "\n").replace('\r', "\n");
    let is_space = |c: char| matches!(c, ' ' | '\t' | '\u{000C}');
    let mut pairs = Vec::new();
    let mut logical = String::new();
    let mut start = 0;
    let mut continuing = false;
    for (index, line) in text.split('\n').enumerate() {
        let line = if continuing {
            line.trim_start_matches(is_space)
        } else {
            let content = line.trim_start_matches(is_space);
            if content.is_empty() || content.starts_with('#') || content.starts_with('!') {
                continue;
            }
            start = index + 1;
            content
        };
        let trailing = line.len() - line.trim_end_matches('\\').len();
        if trailing % 2 == 1 {
            logical.push_str(&line[..line.len() - 1]);
            continuing = true;
        } else {
            logical.push_str(line);
            continuing = false;
            pairs.push(split_entry(&logical, start)?);
            logical.clear();
        }
    }
    if continuing {
        pairs.push(split_entry(&logical, start)?);
    }
    Ok(pairs)
}

fn split_entry(line: &str, number: usize) -> Result<(String, String), PropertiesError> {
    let is_space = |c: char| matches!(c, ' ' | '\t' | '\u{000C}');
    let mut key_end = line.len();
    let mut escaped = false;
    for (i, c) in line.char_indices() {
        if escaped {
            escaped = false;
        } else if c == '\\' {
            escaped = true;
        } else if c == '=' || c == ':' || is_space(c) {
            key_end = i;
            break;
        }
    }
    let rest = line[key_end..].trim_start_matches(is_space);
    let rest = rest
        .strip_prefix('=')
        .or_else(|| rest.strip_prefix(':'))
        .unwrap_or(rest)
        .trim_start_matches(is_space);
    Ok((unescape(&line[..key_end], number)?, unescape(rest, number)?))
}

fn unescape(raw: &str, number: usize) -> Result<String, PropertiesError> {
    let mut units: Vec<u16> = Vec::with_capacity(raw.len());
    let mut chars = raw.chars();
    while let Some(c) = chars.next() {
        if c != '\\' {
            let mut buffer = [0; 2];
            units.extend_from_slice(c.encode_utf16(&mut buffer));
            continue;
        }
        match chars.next() {
            Some('t') => units.push(u16::from(b'\t')),
            Some('n') => units.push(u16::from(b'\n')),
            Some('r') => units.push(u16::from(b'\r')),
            Some('f') => units.push(0x0C),
            Some('u') => {
                let hex: String = chars.by_ref().take(4).collect();
                let unit = (hex.len() == 4)
                    .then(|| u16::from_str_radix(&hex, 16).ok())
                    .flatten()
                    .ok_or(PropertiesError {
                        line: number,
                        reason: "malformed \\uXXXX encoding",
                    })?;
                units.push(unit);
            }
            Some(other) => {
                let mut buffer = [0; 2];
                units.extend_from_slice(other.encode_utf16(&mut buffer));
            }
            None => {}
        }
    }
    Ok(String::from_utf16_lossy(&units))
}

#[cfg(test)]
mod tests {
    use super::*;

    #[test]
    fn doubles_are_written_as_java_writes_them() {
        let cases = [
            (0.0, "0.0"),
            (-0.0, "-0.0"),
            (1.0, "1.0"),
            (100.0, "100.0"),
            (0.5, "0.5"),
            (0.001, "0.001"),
            (0.0001, "1.0E-4"),
            (1234567.0, "1234567.0"),
            (1e7, "1.0E7"),
            (1.25e7, "1.25E7"),
            (-1.5e-5, "-1.5E-5"),
            (1e21, "1.0E21"),
            (123.456, "123.456"),
            (f64::MAX, "1.7976931348623157E308"),
            (f64::MIN_POSITIVE, "2.2250738585072014E-308"),
        ];
        for (v, java) in cases {
            assert_eq!(double_to_string(v), java, "{v:e}");
        }
    }

    #[test]
    fn properties_follow_properties_load() {
        let text = "# comment\n  ! also a comment\n\
                    a=1\n\
                    b : 2\n\
                    c 3\n\
                    d=\\u5fc5\\u9808\n\
                    e=one \\\n     two\n\
                    f\\=g=h\\:i\n\
                    emoji=\\ud83d\\ude00\n\
                    empty\n";
        let pairs = load_properties(text).unwrap();
        let get = |k: &str| {
            pairs
                .iter()
                .find(|(key, _)| key == k)
                .map(|(_, v)| v.as_str())
        };
        assert_eq!(get("a"), Some("1"));
        assert_eq!(get("b"), Some("2"));
        assert_eq!(get("c"), Some("3"));
        assert_eq!(get("d"), Some("必須"));
        assert_eq!(get("e"), Some("one two"));
        assert_eq!(get("f=g"), Some("h:i"));
        assert_eq!(get("emoji"), Some("😀"));
        assert_eq!(get("empty"), Some(""));
        assert_eq!(pairs.len(), 8);
    }

    #[test]
    fn a_malformed_unicode_escape_is_an_error() {
        let error = load_properties("ok=1\nbad=\\u12").unwrap_err();
        assert_eq!(error.line, 2);
    }
}