sup-xml-xslt 1.2.0

XSLT 1.0 transformation engine built on sup-xml's XPath core
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
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
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
//! `format-number(number, picture, decimal-format?)` — XSLT 1.0 §12.3.
//!
//! Picture string grammar (XSLT 1.0 §12.3, §13.1.4):
//!
//! ```text
//! picture       := subpicture (';' subpicture)?
//! subpicture    := prefix? integer-part ('.' fraction-part)? suffix?
//! integer-part  := (digit-or-group)* (mandatory-digit)+
//! fraction-part := (mandatory-digit)* (optional-digit)*
//! ```
//!
//! Two subpictures: first is positive, second (after `;`) is
//! negative.  If no negative subpicture, the negative form prepends
//! the decimal-format's `minus-sign` to the positive form.
//!
//! The "characters" in a picture come from a [`DecimalFormat`]
//! (defaults documented in the struct).  Anything in the picture
//! that's NOT a special character is literal — copied to output as
//! a prefix or suffix.

/// Decimal-format settings, with `xsl:decimal-format` defaults.
/// Hand-wired here; named-`xsl:decimal-format` lookup from the
/// stylesheet isn't threaded through the function dispatcher yet.
#[derive(Clone, Debug, PartialEq, Eq)]
pub struct DecimalFormat {
    pub decimal_separator: char,
    pub grouping_separator: char,
    pub infinity:          String,
    pub minus_sign:        char,
    pub nan:               String,
    pub percent:           char,
    pub per_mille:         char,
    pub zero_digit:        char,
    pub digit:             char,
    pub pattern_separator: char,
}

impl Default for DecimalFormat {
    fn default() -> Self {
        DecimalFormat {
            decimal_separator: '.',
            grouping_separator: ',',
            infinity:           "Infinity".to_string(),
            minus_sign:         '-',
            nan:                "NaN".to_string(),
            percent:            '%',
            per_mille:          '\u{2030}',
            zero_digit:         '0',
            digit:              '#',
            pattern_separator:  ';',
        }
    }
}

/// Format `value` per `picture` using `df`.  Returns the formatted
/// string, or an XTDE1310 message when the picture violates the
/// rules of XSLT 2.0 §16.4.3.
pub fn format_number(value: f64, picture: &str, df: &DecimalFormat)
    -> Result<String, String>
{
    if value.is_nan() { return Ok(df.nan.clone()); }
    validate_picture(picture, df)?;

    // Split into positive / negative subpictures at the
    // pattern-separator char (default `;`).  Pattern-separator
    // chars inside a quoted literal in the picture are NOT split
    // points; XSLT picture strings don't have quoting in 1.0, so
    // we split unconditionally.
    let mut parts = picture.splitn(2, df.pattern_separator);
    let pos_pic = parts.next().unwrap_or("");
    let neg_pic = parts.next();

    let pic = if value < 0.0 && neg_pic.is_some() {
        neg_pic.unwrap()
    } else {
        pos_pic
    };
    let positive_pic = parse_subpicture(pic, df);
    let abs = if value < 0.0 { -value } else { value };

    let formatted = format_subpicture(abs, &positive_pic, df);

    // If no explicit negative subpicture, prepend the minus sign
    // for negative inputs.
    if value < 0.0 && neg_pic.is_none() {
        return Ok(format!("{}{}", df.minus_sign, formatted));
    }
    Ok(formatted)
}

/// XSLT 2.0 §16.4.3 picture-string validation.  Each subpicture
/// must contain at least one digit-character (zero-digit or
/// digit-`#`), `#` digits in the integer part may not appear after
/// a zero-digit, and `0` digits in the fraction part may not
/// appear after a `#`.  Any violation is XTDE1310.
fn validate_picture(picture: &str, df: &DecimalFormat) -> Result<(), String> {
    // XSLT 2.0 §16.4.3 — at most two subpictures separated by the
    // pattern-separator.  Three or more is XTDE1310.
    let parts: Vec<&str> = picture.split(df.pattern_separator).collect();
    if parts.len() > 2 {
        return Err(format!(
            "format-number picture '{picture}' has more than two subpictures \
             (XTDE1310)"
        ));
    }
    for p in &parts { validate_subpicture(p, df)?; }
    Ok(())
}

fn validate_subpicture(s: &str, df: &DecimalFormat) -> Result<(), String> {
    if s.is_empty() {
        return Err(format!(
            "format-number picture '{s}' has an empty subpicture (XTDE1310)"
        ));
    }
    // XSLT 2.0 §16.4.3 — a subpicture may contain at most one
    // decimal-separator and at most one percent / per-mille
    // (and not both).  Multiple percent OR multiple per-mille OR
    // a mix of the two is XTDE1310.
    let dec_count    = s.matches(df.decimal_separator).count();
    let percent_n    = s.matches(df.percent).count();
    let per_mille_n  = s.matches(df.per_mille).count();
    if dec_count > 1 {
        return Err(format!(
            "format-number picture '{s}' contains more than one \
             decimal-separator (XTDE1310)"
        ));
    }
    if percent_n + per_mille_n > 1 {
        return Err(format!(
            "format-number picture '{s}' contains more than one \
             percent / per-mille marker (XTDE1310)"
        ));
    }
    // Walk the subpicture in two halves split by the decimal-
    // separator (if any), enforcing the per-side ordering rules.
    let halves: Vec<&str> = s.splitn(2, df.decimal_separator).collect();
    let int_part  = halves[0];
    let frac_part = halves.get(1).copied().unwrap_or("");
    // §16.4.3 — a grouping-separator may not appear adjacent to the
    // decimal-separator (`0,.0` is invalid even though both halves
    // contain digits).
    if int_part.ends_with(df.grouping_separator) && halves.len() > 1 {
        return Err(format!(
            "format-number picture '{s}' has a grouping-separator \
             adjacent to the decimal-separator (XTDE1310)"
        ));
    }
    if frac_part.starts_with(df.grouping_separator) {
        return Err(format!(
            "format-number picture '{s}' has a grouping-separator \
             adjacent to the decimal-separator (XTDE1310)"
        ));
    }

    // Integer side: zero-digit may not precede the optional-digit
    // sign.  Once we've seen a `0`, every subsequent digit char
    // must also be `0`.
    let mut saw_zero = false;
    let mut int_digits = 0usize;
    for c in int_part.chars() {
        if c == df.zero_digit {
            saw_zero = true;
            int_digits += 1;
        } else if c == df.digit {
            if saw_zero {
                return Err(format!(
                    "format-number picture '{s}' has '#' after '0' in the \
                     integer part (XTDE1310)"
                ));
            }
            int_digits += 1;
        }
        // Other chars (grouping-separator / prefix-suffix) are
        // legal placeholders here.
    }

    // Fraction side: `0` may not appear AFTER `#`.  Walk left-to-
    // right; once we've seen a `#`, every subsequent digit char
    // must also be `#`.
    let mut saw_optional = false;
    let mut frac_digits = 0usize;
    for c in frac_part.chars() {
        if c == df.digit {
            saw_optional = true;
            frac_digits += 1;
        } else if c == df.zero_digit {
            if saw_optional {
                return Err(format!(
                    "format-number picture '{s}' has '0' after '#' in the \
                     fraction part (XTDE1310)"
                ));
            }
            frac_digits += 1;
        }
    }

    if int_digits + frac_digits == 0 {
        return Err(format!(
            "format-number picture '{s}' contains no digit characters \
             (XTDE1310)"
        ));
    }
    // §16.4.3 — once the integer "active part" starts (first
    // digit-character), only digit-characters, the grouping-separator,
    // and the decimal-separator may appear up to the active part's
    // end (i.e. until a non-digit, non-grouping suffix character).
    // The same applies to the fraction part.  Detect a stray
    // non-placeholder mid-stream — `0$0` is the canonical failure.
    fn check_active(part: &str, df: &DecimalFormat, who: &str)
        -> Result<(), String>
    {
        let mut in_active = false;
        let mut left_active = false;
        for c in part.chars() {
            let is_digit_char = c == df.zero_digit || c == df.digit;
            let is_sep = c == df.grouping_separator;
            if !in_active && is_digit_char {
                in_active = true;
                continue;
            }
            if in_active && !left_active {
                if is_digit_char || is_sep { continue; }
                // Anything else closes the active part — but a
                // non-passive character (digit or grouping-separator)
                // after it would mean the picture splits the active
                // run with a foreign character, which §16.4.3
                // rejects as XTDE1310.
                left_active = true;
                continue;
            }
            if left_active && (is_digit_char || is_sep) {
                return Err(format!(
                    "format-number picture '{}' has a non-placeholder \
                     character splitting the {who} active part (XTDE1310)",
                    part
                ));
            }
        }
        Ok(())
    }
    check_active(int_part,  df, "integer")?;
    check_active(frac_part, df, "fraction")?;
    Ok(())
}

#[derive(Debug, Clone)]
struct SubPicture {
    prefix:           String,
    suffix:           String,
    /// Minimum digits before the decimal (count of `zero_digit` in
    /// the integer part).
    min_integer:      usize,
    /// Regular grouping interval — set when the picture's grouping
    /// separators are evenly spaced (the common `#,##0` case), so a
    /// separator repeats every `group_size` digits including for numbers
    /// wider than the picture.
    group_size:       Option<usize>,
    /// Irregular grouping positions (distance from the right edge of the
    /// integer part) for pictures like `###,##0,00` whose separators are
    /// NOT evenly spaced.  XSLT 3.0 §4.7.4: when grouping is irregular the
    /// separators appear ONLY at these positions, never repeated beyond
    /// the leftmost.  Empty when `group_size` carries a regular interval.
    group_positions:  Vec<usize>,
    /// Minimum digits after the decimal (count of `zero_digit` in
    /// the fraction part).
    min_fraction:     usize,
    /// Maximum digits after the decimal (count of `zero_digit` +
    /// `digit` in the fraction part).
    max_fraction:     usize,
    /// `*100` (percent) or `*1000` (per-mille) scale applied
    /// before formatting.  XSLT 1.0 says the suffix's % or ‰
    /// chars trigger this.
    scale:            f64,
}

fn parse_subpicture(s: &str, df: &DecimalFormat) -> SubPicture {
    let mut prefix = String::new();
    let mut suffix = String::new();
    let mut min_integer = 0;
    let mut group_size = None;
    // Left-edge positions (count of integer digits seen so far) at which
    // a grouping separator appeared, in picture order.
    let mut group_marks: Vec<usize> = Vec::new();
    let mut min_fraction = 0;
    let mut max_fraction = 0;
    let mut scale = 1.0;

    enum Phase { Prefix, Integer, Fraction, Suffix }
    let mut phase = Phase::Prefix;
    let mut integer_digits = 0usize;
    let mut fraction_digits = 0usize;
    for c in s.chars() {
        match phase {
            Phase::Prefix => {
                if c == df.zero_digit || c == df.digit {
                    phase = Phase::Integer;
                    integer_digits += 1;
                    if c == df.zero_digit { min_integer += 1; }
                } else if c == df.grouping_separator {
                    // A grouping separator opens the active integer part;
                    // its left position is 0 (no digits seen yet).
                    phase = Phase::Integer;
                    group_marks.push(integer_digits);
                } else {
                    prefix.push(c);
                }
            }
            Phase::Integer => {
                if c == df.zero_digit || c == df.digit {
                    integer_digits += 1;
                    if c == df.zero_digit { min_integer += 1; }
                } else if c == df.grouping_separator {
                    group_marks.push(integer_digits);
                } else if c == df.decimal_separator {
                    phase = Phase::Fraction;
                } else {
                    phase = Phase::Suffix;
                    suffix.push(c);
                    classify_suffix_char(c, df, &mut scale);
                }
            }
            Phase::Fraction => {
                if c == df.zero_digit {
                    fraction_digits += 1; min_fraction += 1; max_fraction += 1;
                } else if c == df.digit {
                    fraction_digits += 1; max_fraction += 1;
                } else {
                    phase = Phase::Suffix;
                    suffix.push(c);
                    classify_suffix_char(c, df, &mut scale);
                }
            }
            Phase::Suffix => {
                suffix.push(c);
                classify_suffix_char(c, df, &mut scale);
            }
        }
    }
    let _ = fraction_digits;
    // Convert each separator's left position to a distance from the right
    // edge of the integer part.  A trailing separator (e.g. picture `#,`)
    // sits at distance 0 and is dropped — it would otherwise mean a zero
    // group size (divide-by-zero) on malformed input.
    let mut positions: Vec<usize> = group_marks.iter()
        .map(|&m| integer_digits.saturating_sub(m))
        .filter(|&d| d > 0)
        .collect();
    positions.sort_unstable();
    positions.dedup();
    let mut group_positions = Vec::new();
    // Grouping is REGULAR when the separator positions are exactly the
    // multiples d, 2d, 3d, … of the smallest position; then a single
    // interval applies (and repeats for wider numbers).  Otherwise the
    // separators are honoured only at the positions given (XSLT 3.0
    // §4.7.4).
    if let Some(&d) = positions.first() {
        let regular = positions.iter().enumerate()
            .all(|(i, &p)| p == (i + 1) * d);
        if regular {
            group_size = Some(d);
        } else {
            group_positions = positions;
        }
    }
    SubPicture {
        prefix, suffix, min_integer, group_size, group_positions,
        min_fraction, max_fraction, scale,
    }
}

fn classify_suffix_char(c: char, df: &DecimalFormat, scale: &mut f64) {
    if c == df.percent     { *scale *= 100.0; }
    else if c == df.per_mille { *scale *= 1000.0; }
}

fn format_subpicture(abs: f64, p: &SubPicture, df: &DecimalFormat) -> String {
    if abs.is_infinite() {
        return format!("{}{}{}", p.prefix, df.infinity, p.suffix);
    }
    let scaled = abs * p.scale;
    // Round to `max_fraction` digits using XPath's half-to-even
    // behaviour — XSLT spec is silent so we match libxslt's
    // half-up.
    let factor = 10f64.powi(p.max_fraction as i32);
    let rounded = (scaled * factor).round() / factor;

    // Split int / frac as strings to preserve trailing zeros.
    let mut int_part: u64 = rounded.trunc() as u64;
    let frac_value = rounded - int_part as f64;
    let frac_str = if p.max_fraction == 0 {
        String::new()
    } else {
        let f_int = (frac_value * factor).round() as u64;
        let mut s = format!("{:0width$}", f_int, width = p.max_fraction);
        // Trim trailing zeros down to min_fraction.
        while s.len() > p.min_fraction && s.ends_with('0') {
            s.pop();
        }
        // Remap ASCII fraction digits onto the decimal-format's zero
        // digit base (matches what we do for integer digits) so locales
        // with non-ASCII digits get consistent fraction output too.
        if df.zero_digit != '0' {
            let zero_code = df.zero_digit as u32;
            s = s.chars().map(|c| match c.to_digit(10) {
                Some(d) => char::from_u32(zero_code + d).unwrap_or(c),
                None    => c,
            }).collect();
        }
        s
    };

    // Build integer string with grouping.
    let mut int_digits: Vec<char> = if int_part == 0 {
        vec![df.zero_digit]
    } else {
        let mut v = Vec::new();
        while int_part > 0 {
            v.push(char::from_digit((int_part % 10) as u32, 10).unwrap());
            int_part /= 10;
        }
        v.reverse();
        // Map ASCII digits to df.zero_digit-based digits if non-ASCII.
        if df.zero_digit != '0' {
            let zero_code = df.zero_digit as u32;
            for c in &mut v {
                let d = c.to_digit(10).unwrap();
                *c = char::from_u32(zero_code + d).unwrap_or(*c);
            }
        }
        v
    };
    // Pad to min_integer.
    while int_digits.len() < p.min_integer {
        int_digits.insert(0, df.zero_digit);
    }
    // Group.
    let mut int_str = String::new();
    let n = int_digits.len();
    if let Some(g) = p.group_size {
        for (i, c) in int_digits.iter().enumerate() {
            if i > 0 && (n - i) % g == 0 {
                int_str.push(df.grouping_separator);
            }
            int_str.push(*c);
        }
    } else if !p.group_positions.is_empty() {
        for (i, c) in int_digits.iter().enumerate() {
            if i > 0 && p.group_positions.contains(&(n - i)) {
                int_str.push(df.grouping_separator);
            }
            int_str.push(*c);
        }
    } else {
        int_str = int_digits.iter().collect();
    }

    let mut out = String::new();
    out.push_str(&p.prefix);
    out.push_str(&int_str);
    if !frac_str.is_empty() {
        out.push(df.decimal_separator);
        out.push_str(&frac_str);
    }
    out.push_str(&p.suffix);
    out
}

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

    fn df() -> DecimalFormat { DecimalFormat::default() }

    #[test]
    fn integer_with_zero_min() {
        assert_eq!(format_number(1234.0, "0", &df()).unwrap(),     "1234");
        assert_eq!(format_number(1.0,    "0000", &df()).unwrap(),  "0001");
    }

    #[test]
    fn integer_with_grouping() {
        assert_eq!(format_number(1234567.0, "#,##0", &df()).unwrap(), "1,234,567");
    }

    #[test]
    fn leading_grouping_separator_sets_interval() {
        // A grouping separator before the first digit opens the active
        // part and fixes a regular interval (group size 3 here).
        assert_eq!(format_number(1234567890.0, ",000", &df()).unwrap(),
                   "1,234,567,890");
    }

    #[test]
    fn unequal_grouping_uses_exact_positions() {
        // XSLT 3.0 §4.7.4 — irregular separators (here at 2 and 5 digits
        // from the right) appear only where the picture places them and
        // are NOT repeated beyond the leftmost.
        assert_eq!(format_number(987654321.0, "###,##0,00.00", &df()).unwrap(),
                   "9876,543,21.00");
    }

    #[test]
    fn fraction_padded_to_min() {
        assert_eq!(format_number(1.5, "0.00", &df()).unwrap(), "1.50");
    }

    #[test]
    fn fraction_trimmed_to_min() {
        assert_eq!(format_number(1.5, "0.0##", &df()).unwrap(), "1.5");
    }

    #[test]
    fn fraction_rounds_at_max() {
        assert_eq!(format_number(1.2345, "0.00", &df()).unwrap(), "1.23");
        assert_eq!(format_number(1.235,  "0.00", &df()).unwrap(), "1.24"); // half-up
    }

    #[test]
    fn negative_uses_implicit_minus_prefix() {
        assert_eq!(format_number(-42.0, "0", &df()).unwrap(), "-42");
    }

    #[test]
    fn negative_uses_explicit_subpicture() {
        assert_eq!(format_number(-42.0, "0;(0)", &df()).unwrap(), "(42)");
    }

    #[test]
    fn nan_returns_nan_literal() {
        assert_eq!(format_number(f64::NAN, "0", &df()).unwrap(), "NaN");
    }

    #[test]
    fn percent_scales_by_100() {
        assert_eq!(format_number(0.25, "0%", &df()).unwrap(), "25%");
    }
}