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
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
// This file is part of the uutils coreutils package.
//
// For the full copyright and license information, please view the LICENSE
// file that was distributed with this source code.
use std::str::FromStr;

use crate::units::Unit;
use uucore::ranges::Range;

pub const DELIMITER: &str = "delimiter";
pub const FIELD: &str = "field";
pub const FIELD_DEFAULT: &str = "1";
pub const FORMAT: &str = "format";
pub const FROM: &str = "from";
pub const FROM_DEFAULT: &str = "none";
pub const FROM_UNIT: &str = "from-unit";
pub const FROM_UNIT_DEFAULT: &str = "1";
pub const HEADER: &str = "header";
pub const HEADER_DEFAULT: &str = "1";
pub const INVALID: &str = "invalid";
pub const NUMBER: &str = "NUMBER";
pub const PADDING: &str = "padding";
pub const ROUND: &str = "round";
pub const SUFFIX: &str = "suffix";
pub const TO: &str = "to";
pub const TO_DEFAULT: &str = "none";
pub const TO_UNIT: &str = "to-unit";
pub const TO_UNIT_DEFAULT: &str = "1";

pub struct TransformOptions {
    pub from: Unit,
    pub from_unit: usize,
    pub to: Unit,
    pub to_unit: usize,
}

#[derive(Debug, PartialEq, Eq)]
pub enum InvalidModes {
    Abort,
    Fail,
    Warn,
    Ignore,
}

pub struct NumfmtOptions {
    pub transform: TransformOptions,
    pub padding: isize,
    pub header: usize,
    pub fields: Vec<Range>,
    pub delimiter: Option<String>,
    pub round: RoundMethod,
    pub suffix: Option<String>,
    pub format: FormatOptions,
    pub invalid: InvalidModes,
}

#[derive(Clone, Copy)]
pub enum RoundMethod {
    Up,
    Down,
    FromZero,
    TowardsZero,
    Nearest,
}

impl RoundMethod {
    pub fn round(&self, f: f64) -> f64 {
        match self {
            Self::Up => f.ceil(),
            Self::Down => f.floor(),
            Self::FromZero => {
                if f < 0.0 {
                    f.floor()
                } else {
                    f.ceil()
                }
            }
            Self::TowardsZero => {
                if f < 0.0 {
                    f.ceil()
                } else {
                    f.floor()
                }
            }
            Self::Nearest => f.round(),
        }
    }
}

// Represents the options extracted from the --format argument provided by the user.
#[derive(Debug, Default, PartialEq, Eq)]
pub struct FormatOptions {
    pub grouping: bool,
    pub padding: Option<isize>,
    pub precision: Option<usize>,
    pub prefix: String,
    pub suffix: String,
    pub zero_padding: bool,
}

impl FromStr for FormatOptions {
    type Err = String;

    // The recognized format is: [PREFIX]%[0]['][-][N][.][N]f[SUFFIX]
    //
    // The format defines the printing of a floating point argument '%f'.
    // An optional quote (%'f) enables --grouping.
    // An optional width value (%10f) will pad the number.
    // An optional zero (%010f) will zero pad the number.
    // An optional negative value (%-10f) will left align.
    // An optional precision (%.1f) determines the precision of the number.
    #[allow(clippy::cognitive_complexity)]
    fn from_str(s: &str) -> Result<Self, Self::Err> {
        let mut iter = s.chars().peekable();
        let mut options = Self::default();

        let mut padding = String::new();
        let mut precision = String::new();
        let mut double_percentage_counter = 0;

        // '%' chars in the prefix, if any, must appear in blocks of even length, for example: "%%%%" and
        // "%% %%" are ok, "%%% %" is not ok. A single '%' is treated as the beginning of the
        // floating point argument.
        while let Some(c) = iter.next() {
            match c {
                '%' if iter.peek() == Some(&'%') => {
                    iter.next();
                    double_percentage_counter += 1;

                    for _ in 0..2 {
                        options.prefix.push('%');
                    }
                }
                '%' => break,
                _ => options.prefix.push(c),
            }
        }

        // GNU numfmt drops a char from the prefix for every '%%' in the prefix, so we do the same
        for _ in 0..double_percentage_counter {
            options.prefix.pop();
        }

        if iter.peek().is_none() {
            return if options.prefix == s {
                Err(format!("format '{s}' has no % directive"))
            } else {
                Err(format!("format '{s}' ends in %"))
            };
        }

        // GNU numfmt allows to mix the characters " ", "'", and "0" in any way, so we do the same
        while matches!(iter.peek(), Some(' ' | '\'' | '0')) {
            match iter.next().unwrap() {
                ' ' => (),
                '\'' => options.grouping = true,
                '0' => options.zero_padding = true,
                _ => unreachable!(),
            }
        }

        if let Some('-') = iter.peek() {
            iter.next();

            match iter.peek() {
                Some(c) if c.is_ascii_digit() => padding.push('-'),
                _ => {
                    return Err(format!(
                        "invalid format '{s}', directive must be %[0]['][-][N][.][N]f"
                    ))
                }
            }
        }

        while let Some(c) = iter.peek() {
            if c.is_ascii_digit() {
                padding.push(*c);
                iter.next();
            } else {
                break;
            }
        }

        if !padding.is_empty() {
            if let Ok(p) = padding.parse() {
                options.padding = Some(p);
            } else {
                return Err(format!("invalid format '{s}' (width overflow)"));
            }
        }

        if let Some('.') = iter.peek() {
            iter.next();

            if matches!(iter.peek(), Some(' ' | '+' | '-')) {
                return Err(format!("invalid precision in format '{s}'"));
            }

            while let Some(c) = iter.peek() {
                if c.is_ascii_digit() {
                    precision.push(*c);
                    iter.next();
                } else {
                    break;
                }
            }

            if precision.is_empty() {
                options.precision = Some(0);
            } else if let Ok(p) = precision.parse() {
                options.precision = Some(p);
            } else {
                return Err(format!("invalid precision in format '{s}'"));
            }
        }

        if let Some('f') = iter.peek() {
            iter.next();
        } else {
            return Err(format!(
                "invalid format '{s}', directive must be %[0]['][-][N][.][N]f"
            ));
        }

        // '%' chars in the suffix, if any, must appear in blocks of even length, otherwise
        // it is an error. For example: "%%%%" and "%% %%" are ok, "%%% %" is not ok.
        while let Some(c) = iter.next() {
            if c != '%' {
                options.suffix.push(c);
            } else if iter.peek() == Some(&'%') {
                for _ in 0..2 {
                    options.suffix.push('%');
                }
                iter.next();
            } else {
                return Err(format!("format '{s}' has too many % directives"));
            }
        }

        Ok(options)
    }
}

impl FromStr for InvalidModes {
    type Err = String;

    fn from_str(s: &str) -> Result<Self, Self::Err> {
        match s.to_lowercase().as_str() {
            "abort" => Ok(Self::Abort),
            "fail" => Ok(Self::Fail),
            "warn" => Ok(Self::Warn),
            "ignore" => Ok(Self::Ignore),
            unknown => Err(format!("Unknown invalid mode: {unknown}")),
        }
    }
}

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

    #[test]
    fn test_parse_format() {
        assert_eq!(FormatOptions::default(), "%f".parse().unwrap());
        assert_eq!(FormatOptions::default(), "%  f".parse().unwrap());
    }

    #[test]
    #[allow(clippy::cognitive_complexity)]
    fn test_parse_format_with_invalid_formats() {
        assert!("".parse::<FormatOptions>().is_err());
        assert!("hello".parse::<FormatOptions>().is_err());
        assert!("hello%".parse::<FormatOptions>().is_err());
        assert!("%-f".parse::<FormatOptions>().is_err());
        assert!("%d".parse::<FormatOptions>().is_err());
        assert!("%4 f".parse::<FormatOptions>().is_err());
        assert!("%f%".parse::<FormatOptions>().is_err());
        assert!("%f%%%".parse::<FormatOptions>().is_err());
        assert!("%%f".parse::<FormatOptions>().is_err());
        assert!("%%%%f".parse::<FormatOptions>().is_err());
        assert!("%.-1f".parse::<FormatOptions>().is_err());
        assert!("%. 1f".parse::<FormatOptions>().is_err());
        assert!("%18446744073709551616f".parse::<FormatOptions>().is_err());
        assert!("%.18446744073709551616f".parse::<FormatOptions>().is_err());
    }

    #[test]
    fn test_parse_format_with_prefix_and_suffix() {
        let formats = vec![
            ("--%f", "--", ""),
            ("%f::", "", "::"),
            ("--%f::", "--", "::"),
            ("%f%%", "", "%%"),
            ("%%%f", "%", ""),
            ("%% %f", "%%", ""),
        ];

        for (format, expected_prefix, expected_suffix) in formats {
            let options: FormatOptions = format.parse().unwrap();
            assert_eq!(expected_prefix, options.prefix);
            assert_eq!(expected_suffix, options.suffix);
        }
    }

    #[test]
    fn test_parse_format_with_padding() {
        let mut expected_options = FormatOptions::default();
        let formats = vec![("%12f", Some(12)), ("%-12f", Some(-12))];

        for (format, expected_padding) in formats {
            expected_options.padding = expected_padding;
            assert_eq!(expected_options, format.parse().unwrap());
        }
    }

    #[test]
    fn test_parse_format_with_precision() {
        let mut expected_options = FormatOptions::default();
        let formats = vec![
            ("%6.2f", Some(6), Some(2)),
            ("%6.f", Some(6), Some(0)),
            ("%.2f", None, Some(2)),
            ("%.f", None, Some(0)),
        ];

        for (format, expected_padding, expected_precision) in formats {
            expected_options.padding = expected_padding;
            expected_options.precision = expected_precision;
            assert_eq!(expected_options, format.parse().unwrap());
        }
    }

    #[test]
    fn test_parse_format_with_grouping() {
        let expected_options = FormatOptions {
            grouping: true,
            ..Default::default()
        };
        assert_eq!(expected_options, "%'f".parse().unwrap());
        assert_eq!(expected_options, "% ' f".parse().unwrap());
        assert_eq!(expected_options, "%'''''''f".parse().unwrap());
    }

    #[test]
    fn test_parse_format_with_zero_padding() {
        let expected_options = FormatOptions {
            padding: Some(10),
            zero_padding: true,
            ..Default::default()
        };
        assert_eq!(expected_options, "%010f".parse().unwrap());
        assert_eq!(expected_options, "% 0 10f".parse().unwrap());
        assert_eq!(expected_options, "%0000000010f".parse().unwrap());
    }

    #[test]
    fn test_parse_format_with_grouping_and_zero_padding() {
        let expected_options = FormatOptions {
            grouping: true,
            zero_padding: true,
            ..Default::default()
        };
        assert_eq!(expected_options, "%0'f".parse().unwrap());
        assert_eq!(expected_options, "%'0f".parse().unwrap());
        assert_eq!(expected_options, "%0'0'0'f".parse().unwrap());
        assert_eq!(expected_options, "%'0'0'0f".parse().unwrap());
    }

    #[test]
    fn test_set_invalid_mode() {
        assert_eq!(Ok(InvalidModes::Abort), InvalidModes::from_str("abort"));
        assert_eq!(Ok(InvalidModes::Abort), InvalidModes::from_str("ABORT"));

        assert_eq!(Ok(InvalidModes::Fail), InvalidModes::from_str("fail"));
        assert_eq!(Ok(InvalidModes::Fail), InvalidModes::from_str("FAIL"));

        assert_eq!(Ok(InvalidModes::Ignore), InvalidModes::from_str("ignore"));
        assert_eq!(Ok(InvalidModes::Ignore), InvalidModes::from_str("IGNORE"));

        assert_eq!(Ok(InvalidModes::Warn), InvalidModes::from_str("warn"));
        assert_eq!(Ok(InvalidModes::Warn), InvalidModes::from_str("WARN"));

        assert!(InvalidModes::from_str("something unknown").is_err());
    }
}