mrz 0.7.0

Zero-dependency ICAO 9303 Machine Readable Zone (MRZ) parser, emitter, and check-digit validator for TD1/TD2/TD3 and MRV-A/MRV-B (visa) documents.
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
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
1198
1199
1200
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
//! ICAO 9303 Machine Readable Zone parser, emitter, and check-digit validator.
//!
//! Zero runtime dependencies so it compiles to native and `wasm32-unknown-unknown`
//! targets alike. Supports:
//! - **TD3** (passports): 2 lines × 44 characters
//! - **TD2** (official travel documents / ID cards): 2 lines × 36 characters
//! - **TD1** (ID cards): 3 lines × 30 characters
//! - **MRV-A** (visas, passport-book size): 2 lines × 44 characters
//! - **MRV-B** (visas, smaller size): 2 lines × 36 characters
//!
//! Check digits use the standard 7-3-1 weighting over the value mapping
//! `0-9 → 0-9`, `A-Z → 10-35`, `< → 0`. A field checksum that validates
//! mathematically proves the OCR read is faithful to the printed document —
//! no probabilistic model involved.
//!
//! ```
//! let doc = mrz::parse_td3(
//!     "P<UTOERIKSSON<<ANNA<MARIA<<<<<<<<<<<<<<<<<<<",
//!     "L898902C36UTO7408122F1204159ZE184226B<<<<<10",
//! ).unwrap();
//!
//! assert_eq!(doc.surname, "ERIKSSON");
//! assert_eq!(doc.date_of_birth, "1974-08-12"); // expanded to ISO 8601
//!
//! // Per-field proof, not a single boolean.
//! assert!(doc.checks.document_number);
//! assert!(doc.checks.composite);
//! assert!(doc.valid()); // every check digit verified
//! ```
//!
//! `parse_td1`, `parse_td2`, `parse_mrv_a` and `parse_mrv_b` cover the other
//! formats. From there:
//!
//! - [`find_and_parse`] — scan free-form OCR text for an MRZ
//! - [`format_td3`] and its siblings — emit an MRZ, national characters and all
//! - [`MrzData::full_document_number`] — numbers too long for the printed field
//! - [`transliterate`] — Doc 9303 Part 3 §6 A (Latin), including its multi-valued cells
//! - [`transliterate_cyrillic`] — Doc 9303 Part 3 §6 B (Cyrillic), per [`CyrillicLanguage`]
//! - [`MrzData::validity`] — why a proven *read* is not a valid *document*
//! - [`Blindspot`] — the substitutions no check digit can ever catch
//!
//! The engine is split across (private) modules, whose public items are all
//! re-exported at the crate root:
//! - `checksum` — check-digit math and generic OCR-repair primitives
//! - `blindspot` — the substitutions check digits provably cannot catch
//! - `parser` — the TD1/TD2/TD3 parsers and the free-text scanner
//! - `dates` — `YYMMDD` expansion and date-plausibility checks
//! - `countries` — ICAO/ISO 3166-1 code → country name
//!
//! A valid composite check digit proves a faithful *read*; it does not prove
//! the document is in date — see [`MrzData::validity`].

#![warn(missing_docs)]

/// Compiles and runs every Rust example in `README.md` as a doctest, so a
/// README snippet can never drift from the API it demonstrates. `cfg(doctest)`
/// means this is *only* built while collecting doctests — the README is not
/// injected into the rendered crate documentation, which has its own prose
/// above.
#[cfg(doctest)]
#[doc = include_str!("../README.md")]
struct ReadmeDoctests;

#[cfg(feature = "serde")]
use serde::{Deserialize, Serialize};
#[cfg(feature = "zeroize")]
use zeroize::ZeroizeOnDrop;

mod blindspot;
mod checksum;
mod countries;
mod dates;
mod doccode;
mod emit;
mod parser;
mod repair;
mod translit;

pub use blindspot::{blindspot, class_of, collisions, Blindspot, CLASSES};
pub use checksum::{check_digit, verify};
pub use countries::{code_for_name, codes, codes_equivalent, country_name};
pub use dates::{
    date_completeness, expand_date, expand_date_with_pivot, Date, DateCompleteness, DateValidity,
    CURRENT_YY,
};
pub use doccode::{passport_type, PassportType};
pub use emit::{
    encode_name_component, format_mrv_a, format_mrv_b, format_td1, format_td2, format_td3,
    MrvAFields, MrvBFields, Td1Fields, Td2Fields, Td3Fields,
};
pub use parser::{
    find_and_parse, find_and_parse_with, parse_mrv_a, parse_mrv_a_with, parse_mrv_b,
    parse_mrv_b_with, parse_td1, parse_td1_with, parse_td2, parse_td2_with, parse_td3,
    parse_td3_with,
};
pub use repair::{
    solve_field, solve_substitution, substitution_candidates, width_candidates, FieldKind,
    Resolution, CONFUSABLES, MRZ_ALPHABET, UNKNOWN,
};
pub use translit::{
    transliterate, transliterate_char, transliterate_cyrillic, transliterate_cyrillic_char,
    transliterations, CyrillicLanguage, TransliterationStyle,
};

/// Tunables for the parsing entry points.
///
/// Every `parse_*` / [`find_and_parse`] function is the `ParseOptions::default()`
/// case of its `*_with` counterpart, so existing calls are unaffected.
#[derive(Debug, Clone, Copy, PartialEq, Eq)]
#[cfg_attr(feature = "serde", derive(Serialize, Deserialize))]
pub struct ParseOptions {
    /// Two-digit century pivot for [`expand_date_with_pivot`]. Defaults to
    /// [`CURRENT_YY`]; set it explicitly to pin behaviour instead of inheriting
    /// the constant this crate was compiled with.
    pub pivot_yy: u32,
}

impl Default for ParseOptions {
    fn default() -> Self {
        Self {
            pivot_yy: CURRENT_YY,
        }
    }
}

/// Per-field check-digit verification results.
///
/// `#[non_exhaustive]`: a future MRZ format may carry a check digit these five
/// fields don't name, and adding it should not be a breaking change. Construct
/// one from a `parse_*` function rather than by literal.
#[derive(Debug, Clone, PartialEq, Eq)]
#[cfg_attr(feature = "serde", derive(Serialize, Deserialize))]
#[non_exhaustive]
pub struct Checks {
    /// The primary document-number field.
    pub document_number: bool,
    /// The date-of-birth field.
    pub date_of_birth: bool,
    /// The date-of-expiry field.
    pub date_of_expiry: bool,
    /// TD3 only; `true` for TD1/TD2 (no such check digit exists there).
    pub personal_number: bool,
    /// The composite check digit over the whole zone.
    pub composite: bool,
}

impl Checks {
    /// All check digits valid — the MRZ read is mathematically verified.
    pub fn all_valid(&self) -> bool {
        self.document_number
            && self.date_of_birth
            && self.date_of_expiry
            && self.personal_number
            && self.composite
    }
}

/// A single, coordinated completeness signal spanning both outcomes of a
/// [`find_and_parse`] attempt — the "several
/// separate vocabularies for 'is this record complete'" gap
/// `knowledge/MRZ_SEQUENCE_COMPLETENESS.md` chunk 5 exists to close.
///
/// Deliberately **not** a struct that duplicates [`MrzData`]'s own fields
/// (a second, independently-stale source of truth for the same facts):
/// [`Complete`](Self::Complete) is built from the values already on
/// `MrzData` via [`MrzData::sequence_completeness`], and
/// [`Partial`](Self::Partial) mirrors [`MrzError::IncompleteSequence`]'s
/// own fields exactly. This type's only job is to be the *one* thing a
/// caller can hold instead of separately branching on
/// `Result<MrzData, MrzError>`'s two differently-shaped arms —
/// [`from_parse_result`](Self::from_parse_result) is the constructor meant
/// for that.
///
/// `lines_found`/`lines_expected` never actually differ inside `Complete`:
/// an `MrzData` cannot exist at all without every line the format's
/// fixed-offset `parse_*` needs already having been found, so a *complete*
/// read has nothing left to report on that axis — which is exactly why
/// those two fields live only on `Partial`, not duplicated onto `Complete`
/// as a pair of always-equal constants.
///
/// Deliberately does **not** fold in
/// `synthpass_core::fusion::check_line1_integrity`'s verdict — that
/// heuristic layer (`document_type`/`issuing_country`/`surname`/
/// `given_names` carry no check digit at all) lives one layer up in
/// `synthpass-core` on purpose; this type unifies only the vocabularies
/// that already live inside `crates/mrz` itself.
#[derive(Debug, Clone, PartialEq, Eq)]
#[cfg_attr(feature = "serde", derive(Serialize, Deserialize))]
#[non_exhaustive]
pub enum SequenceCompleteness {
    /// Every line the format needs was found and parsed into an
    /// [`MrzData`]. Whether every *field* is proven is a separate
    /// question — see `checks`/`date_of_birth`.
    Complete {
        /// Per-field check-digit proof — see [`MrzData::checks`].
        checks: Checks,
        /// Date-of-birth field completeness — see
        /// [`MrzData::date_of_birth_completeness`]. No analogous signal
        /// exists for date-of-expiry today; that asymmetry is inherited
        /// from [`DateCompleteness`] itself, not introduced here.
        date_of_birth: DateCompleteness,
    },
    /// Fewer than `lines_expected` lines were found — mirrors
    /// [`MrzError::IncompleteSequence`] exactly.
    Partial {
        /// The format whose line 1 was recognized.
        format: Format,
        /// How many of `lines_expected` lines were actually located.
        lines_found: u8,
        /// How many lines `format` requires.
        lines_expected: u8,
    },
}

impl SequenceCompleteness {
    /// Builds the coordinated signal from either arm of a
    /// [`find_and_parse`] result. `None` when the
    /// error carries no completeness signal of its own — a bare
    /// [`MrzError::NotFound`] (nothing MRZ-shaped at all) or one of the
    /// other structural/checksum error variants, which are about a
    /// *specific* zone a caller already had in hand, not a free-text scan.
    pub fn from_parse_result(result: &Result<MrzData, MrzError>) -> Option<Self> {
        match result {
            Ok(data) => Some(data.sequence_completeness()),
            Err(MrzError::IncompleteSequence {
                format,
                lines_found,
                lines_expected,
            }) => Some(Self::Partial {
                format: *format,
                lines_found: *lines_found,
                lines_expected: *lines_expected,
            }),
            Err(_) => None,
        }
    }
}

/// `#[non_exhaustive]`: ICAO 9303 defines formats this crate does not parse yet
/// (MRP-style variants, future parts), so `match` on this must carry a `_` arm
/// and gaining a variant is not a breaking change. Adding MRV-A/MRV-B in 0.3.0
/// was breaking precisely because this attribute was missing.
#[derive(Debug, Clone, Copy, PartialEq, Eq)]
#[cfg_attr(feature = "serde", derive(Serialize, Deserialize))]
#[non_exhaustive]
pub enum Format {
    /// Passport (TD3): 2 lines × 44 characters.
    Td3,
    /// Official travel document / ID card (TD2): 2 lines × 36 characters.
    Td2,
    /// ID card (TD1): 3 lines × 30 characters.
    Td1,
    /// Machine readable visa, type A (ICAO 9303 part 7): two 44-char lines,
    /// geometry mirrors TD3 through the expiry check digit, but there is no
    /// personal-number field and no composite check digit.
    MrvA,
    /// Machine readable visa, type B (ICAO 9303 part 7): two 36-char lines,
    /// geometry mirrors TD2 through the expiry check digit, but there is no
    /// personal-number field and no composite check digit.
    MrvB,
}

/// Parsed and validated MRZ data.
///
/// The `zeroize` feature (off by default, kept off for the `wasm32-unknown-unknown`
/// browser-demo build so this crate stays zero-dependency there) derives
/// `ZeroizeOnDrop`, wiping the PII-bearing `String` fields from memory when a
/// value is dropped. `format` and `checks` carry no PII and are `Copy`, so
/// they're `#[zeroize(skip)]`.
///
/// `#[non_exhaustive]`: this struct grows as the crate decodes more of the
/// zone — `document_number_full` arrived in 0.4.0 — and that should not break
/// downstream code. Obtain one from a `parse_*` function; it is an output type
/// and there is no reason to build it by literal.
#[derive(Debug, Clone, PartialEq, Eq)]
#[cfg_attr(feature = "serde", derive(Serialize, Deserialize))]
#[cfg_attr(feature = "zeroize", derive(ZeroizeOnDrop))]
#[non_exhaustive]
pub struct MrzData {
    /// Which ICAO 9303 layout this was parsed from.
    #[cfg_attr(feature = "zeroize", zeroize(skip))]
    pub format: Format,
    /// Document code, e.g. "P" (passport), "ID"/"I" (identity card).
    pub document_type: String,
    /// Issuing state or organization (3-letter ICAO code).
    pub issuing_country: String,
    /// The 9-character document-number field as printed, truncated if the
    /// real number overflows — see [`document_number_full`](Self::document_number_full).
    pub document_number: String,
    /// The reassembled document number when it overflows the 9-character field
    /// (ICAO 9303 Part 5 note j / §4.2.4 for TD1, Part 6 note j for TD2; Part 4
    /// defines no such rule, so TD3 gets it by a deliberate extension — see
    /// `parser::read_overflow`'s doc comment. MRVs have no overflow encoding at
    /// all, per Part 7). `None` when the number fits, in which case
    /// [`document_number`](Self::document_number) is already complete.
    pub document_number_full: Option<String>,
    /// `true` when the long document number was recovered from the pre-0.6
    /// eight-character encoding this crate used to emit, rather than the
    /// Doc 9303 form (nine principal characters, filler in the check-digit
    /// position). Always `false` for conformant zones and when the number
    /// fits its field.
    #[cfg_attr(feature = "zeroize", zeroize(skip))]
    pub document_number_legacy_encoding: bool,
    /// Primary identifier / surname, as printed (uppercase, `<` runs
    /// collapsed to single spaces between components).
    pub surname: String,
    /// Secondary identifier / given names, as printed (same cleanup as
    /// [`surname`](Self::surname)).
    pub given_names: String,
    /// Nationality (3-letter ICAO code).
    pub nationality: String,
    /// ISO 8601 (`YYYY-MM-DD`), century inferred (see [`expand_date`]).
    /// Holds the raw, unexpanded `YYMMDD` field instead whenever
    /// [`date_of_birth_completeness`](Self::date_of_birth_completeness) is
    /// not [`DateCompleteness::Complete`] — see that field's doc comment.
    pub date_of_birth: String,
    /// Whether the holder's date of birth was fully known, partly filled with
    /// `<`, or entirely unknown (Doc 9303 Part 3 §4.8, `:547`). `date_of_birth`
    /// holds the raw field rather than an ISO date whenever this is not
    /// [`DateCompleteness::Complete`].
    ///
    /// An unknown or partially unknown date of birth is not a bad read: Part
    /// 3 §4.8 explicitly lets an issuer fill unknown positions with `<`, and
    /// a filler counts as zero for check-digit purposes (`:563`), so an
    /// all-filler date of birth with check digit `0` verifies. This field is
    /// how a caller distinguishes "legitimately unknown, per the issuer" from
    /// "garbage OCR" — `DateValidity::dates_well_formed` stays `false` either
    /// way, since an unknown date is genuinely not a parseable calendar date,
    /// so that struct alone cannot make the distinction.
    #[cfg_attr(feature = "zeroize", zeroize(skip))]
    pub date_of_birth_completeness: DateCompleteness,
    /// "M", "F" or "X" (unspecified).
    pub sex: String,
    /// ISO 8601 (`YYYY-MM-DD`).
    pub date_of_expiry: String,
    /// TD3: personal number field. TD1: optional data 1 + 2 joined.
    /// TD2: optional data field.
    pub personal_number: Option<String>,
    /// The raw MRZ lines, newline-joined, exactly as validated.
    pub mrz_lines: String,
    /// Per-field check-digit verification results — see [`Checks`].
    #[cfg_attr(feature = "zeroize", zeroize(skip))]
    pub checks: Checks,
}

impl MrzData {
    /// Shorthand for `checks.all_valid()`.
    pub fn valid(&self) -> bool {
        self.checks.all_valid()
    }

    /// The complete document number: the overflow reassembly when there is
    /// one, otherwise the 9-character field as printed.
    ///
    /// When a document number exceeds the 9-character field, Doc 9303 puts the
    /// nine principal characters in the field, a filler in the check-digit
    /// position, and the remainder plus its own check digit at the front of the
    /// optional-data field. See [`MrzData::document_number_full`] for the
    /// normative scope of that rule and the TD3 caveat; when the remainder does
    /// not fit the optional field, the number truncates to nine characters
    /// instead.
    ///
    /// ```
    /// use mrz::{format_td3, Td3Fields};
    ///
    /// let lines = format_td3(&Td3Fields {
    ///     issuing_country: "UTO".into(),
    ///     document_number: "L898902C31234".into(), // 13 chars, overflows the 9-char field
    ///     surname: "ERIKSSON".into(),
    ///     given_names: "ANNA MARIA".into(),
    ///     nationality: "UTO".into(),
    ///     date_of_birth: "740812".into(),
    ///     sex: "F".into(),
    ///     date_of_expiry: "120415".into(),
    ///     ..Default::default()
    /// });
    ///
    /// let (l1, l2) = lines.split_once('\n').unwrap();
    /// let doc = mrz::parse_td3(l1, l2).unwrap();
    /// assert!(doc.valid());
    /// assert_eq!(doc.document_number, "L898902C3");            // the printed 9-char field
    /// assert_eq!(doc.full_document_number(), "L898902C31234"); // the reassembled number
    /// assert!(!doc.document_number_legacy_encoding);           // read as the Doc 9303 form
    /// ```
    pub fn full_document_number(&self) -> &str {
        self.document_number_full
            .as_deref()
            .unwrap_or(&self.document_number)
    }

    /// Human-readable name of the issuing state, if the code is recognized.
    pub fn issuing_country_name(&self) -> Option<&'static str> {
        country_name(&self.issuing_country)
    }

    /// The passport type this document's code designates, per ICAO 9303
    /// Part 4 §4.4.
    ///
    /// `None` covers three cases the caller usually wants to tell apart, and
    /// which this method deliberately does not distinguish on its own:
    ///
    /// * the document is not a passport (`Format::Td1`, `Td2`, `MrvA`, `MrvB`);
    /// * it is a passport carrying the `P<` filler form, i.e. no secondary
    ///   code — conformant today, and the most common case in real corpora;
    /// * the second character is outside the §4.4 table.
    ///
    /// Check [`format`](Self::format) and [`document_type`](Self::document_type)
    /// when the difference matters. A code outside the table never makes a
    /// document unparseable — see the [`doccode`](crate::PassportType) docs for
    /// why recognition here is deliberately not rejection.
    pub fn passport_type(&self) -> Option<PassportType> {
        passport_type(&self.document_type)
    }

    /// Human-readable name of the nationality, if the code is recognized.
    pub fn nationality_name(&self) -> Option<&'static str> {
        country_name(&self.nationality)
    }

    /// This record's [`SequenceCompleteness`] — always
    /// [`SequenceCompleteness::Complete`], computed from `self.checks`/
    /// `self.date_of_birth_completeness` rather than stored separately.
    /// [`SequenceCompleteness::Partial`] only ever comes from
    /// [`SequenceCompleteness::from_parse_result`] on the `Err` side of a
    /// [`find_and_parse`] call, since a `Partial`
    /// read has no `MrzData` to hang this method off of in the first place.
    pub fn sequence_completeness(&self) -> SequenceCompleteness {
        SequenceCompleteness::Complete {
            checks: self.checks.clone(),
            date_of_birth: self.date_of_birth_completeness,
        }
    }
}

/// Which check-digit-bearing field an error refers to.
#[derive(Debug, Clone, Copy, PartialEq, Eq)]
#[cfg_attr(feature = "serde", derive(Serialize, Deserialize))]
#[non_exhaustive]
pub enum Field {
    /// [`Checks::document_number`].
    DocumentNumber,
    /// [`Checks::date_of_birth`].
    DateOfBirth,
    /// [`Checks::date_of_expiry`].
    DateOfExpiry,
    /// [`Checks::personal_number`].
    PersonalNumber,
    /// [`Checks::composite`].
    Composite,
}

impl Field {
    /// Field name as it appears on [`Checks`].
    pub fn as_str(self) -> &'static str {
        match self {
            Self::DocumentNumber => "document_number",
            Self::DateOfBirth => "date_of_birth",
            Self::DateOfExpiry => "date_of_expiry",
            Self::PersonalNumber => "personal_number",
            Self::Composite => "composite",
        }
    }
}

impl core::fmt::Display for Field {
    fn fmt(&self, f: &mut core::fmt::Formatter<'_>) -> core::fmt::Result {
        f.write_str(self.as_str())
    }
}

impl Checks {
    /// The fields whose check digits failed, in field order. Empty when
    /// [`all_valid`](Checks::all_valid) is `true`.
    pub fn failed(&self) -> Vec<Field> {
        [
            (self.document_number, Field::DocumentNumber),
            (self.date_of_birth, Field::DateOfBirth),
            (self.date_of_expiry, Field::DateOfExpiry),
            (self.personal_number, Field::PersonalNumber),
            (self.composite, Field::Composite),
        ]
        .into_iter()
        .filter_map(|(ok, f)| (!ok).then_some(f))
        .collect()
    }

    /// How many of the five check digits validate — the ranking signal the
    /// scanner uses to pick its best-effort reading.
    pub(crate) fn score(&self) -> u8 {
        5 - self.failed().len() as u8
    }
}

/// Why parsing an MRZ zone failed outright — distinct from a failed check
/// digit, which `parse_*` reports through [`Checks`] instead of an error.
#[derive(Debug, Clone, PartialEq, Eq)]
#[non_exhaustive]
pub enum MrzError {
    /// Line has the wrong length for the claimed format.
    BadLength {
        /// The length the claimed format requires.
        expected: usize,
        /// The length actually supplied.
        got: usize,
    },
    /// Character outside `[A-Z0-9<]`.
    BadCharacter(char),
    /// Document code not recognized for the format.
    BadDocumentCode(String),
    /// A check digit did not validate against its field.
    ///
    /// Note that the `parse_*` functions deliberately do *not* return this:
    /// they return an [`MrzData`] whose [`Checks`] report the failure, so a
    /// caller can show the user which digits disagreed. This variant exists
    /// for callers that convert a failed [`Checks`] into an error of their own.
    BadChecksum {
        /// Which check-digit-bearing field disagreed.
        field: Field,
        /// Byte offset of the check digit within the field's line.
        position: usize,
    },
    /// [`find_and_parse`]/[`find_and_parse_with`] only: a line matching one
    /// format's document-code prefix and charset was found, but no companion
    /// line ever combined with it into a full parse — distinct from
    /// [`NotFound`](Self::NotFound), which means nothing MRZ-shaped was seen
    /// at all. `lines_found` is always `1` today: only line 1's document-code
    /// prefix is a reliably shape-checkable signal on its own (a format's
    /// other lines carry no distinguishing prefix to detect in isolation).
    /// See `knowledge/MRZ_SEQUENCE_COMPLETENESS.md` chunk 4.
    IncompleteSequence {
        /// The format whose line 1 was recognized.
        format: Format,
        /// How many of `lines_expected` lines were actually located.
        lines_found: u8,
        /// How many lines `format` requires (2 for TD2/TD3/MRV-A/MRV-B, 3
        /// for TD1).
        lines_expected: u8,
    },
    /// No plausible MRZ found in the supplied text.
    NotFound,
}

impl core::fmt::Display for MrzError {
    fn fmt(&self, f: &mut core::fmt::Formatter<'_>) -> core::fmt::Result {
        match self {
            Self::BadLength { expected, got } => {
                write!(f, "bad MRZ line length: expected {expected}, got {got}")
            }
            Self::BadCharacter(c) => write!(f, "invalid MRZ character: {c:?}"),
            Self::BadDocumentCode(c) => write!(f, "unrecognized document code: {c:?}"),
            Self::BadChecksum { field, position } => {
                write!(f, "check digit failed for {field} at position {position}")
            }
            Self::IncompleteSequence {
                format,
                lines_found,
                lines_expected,
            } => {
                write!(
                    f,
                    "incomplete {format:?} sequence: found {lines_found} of {lines_expected} lines"
                )
            }
            Self::NotFound => write!(f, "no MRZ found in text"),
        }
    }
}

impl std::error::Error for MrzError {}

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

    // ICAO 9303 specimen identity (Utopia / Anna Maria Eriksson). Part 4's own
    // copy is Appendix B Figure B-1 — an image in the source PDF, not
    // extracted as text into this corpus
    // (`knowledge/docs9303/Doc_9303_Part4_Specs_for_MRPs_and_TD3_MRTDs.md:669-687`
    // is the figure caption, no MRZ text). Corroborated instead by the
    // byte-identical TD2 zone Part 6 publishes as literal text (see below,
    // `knowledge/docs9303/Doc_9303_Part6_Specs_for_TD2_MROTDs.md:487-488`).
    const TD3_L1: &str = "P<UTOERIKSSON<<ANNA<MARIA<<<<<<<<<<<<<<<<<<<";
    const TD3_L2: &str = "L898902C36UTO7408122F1204159ZE184226B<<<<<10";

    #[test]
    fn td3_specimen_fully_valid() {
        let d = parse_td3(TD3_L1, TD3_L2).unwrap();
        assert!(d.valid(), "checks: {:?}", d.checks);
        assert_eq!(d.document_type, "P");
        assert_eq!(d.issuing_country, "UTO");
        assert_eq!(d.surname, "ERIKSSON");
        assert_eq!(d.given_names, "ANNA MARIA");
        assert_eq!(d.document_number, "L898902C3");
        assert_eq!(d.nationality, "UTO");
        assert_eq!(d.date_of_birth, "1974-08-12");
        assert_eq!(d.date_of_birth_completeness, DateCompleteness::Complete);
        assert_eq!(d.sex, "F");
        assert_eq!(d.date_of_expiry, "2012-04-15");
        assert_eq!(d.personal_number.as_deref(), Some("ZE184226B"));
    }

    #[test]
    fn td3_tampered_dob_fails_checksum() {
        // Change one digit of the date of birth: 740812 → 750812.
        let tampered = TD3_L2.replacen("740812", "750812", 1);
        let d = parse_td3(TD3_L1, &tampered).unwrap();
        assert!(!d.checks.date_of_birth);
        assert!(!d.checks.composite);
        assert!(!d.valid());
    }

    #[test]
    fn td3_all_filler_dob_is_a_valid_field_not_a_bad_read() {
        // Doc 9303 Part 3 §4.8 (`:547`) lets an issuer complete an unknown
        // date of birth with filler characters, and `:563` gives a `<` the
        // value zero for check-digit purposes — so an all-filler DOB field
        // with check digit 0 is mathematically valid, not corrupt. Replace
        // TD3_L2's dob+check field (positions 14-20, 0-indexed 13..20) with
        // seven fillers followed by check digit '0'.
        assert_eq!(check_digit("<<<<<<").unwrap(), 0);
        let l2 = format!("{}{}{}", &TD3_L2[0..13], "<<<<<<0", &TD3_L2[20..]);
        assert_eq!(l2.len(), TD3_L2.len());
        let d = parse_td3(TD3_L1, &l2).unwrap();
        assert!(d.checks.date_of_birth);
        assert_eq!(d.date_of_birth, "<<<<<<"); // left unexpanded, not a lie
        assert_eq!(d.date_of_birth_completeness, DateCompleteness::Unknown);
    }

    // Mrz_Field_Layout.md §2.3: an unused TD3 personal number's check digit
    // may be issued as either '0' or '<' — both are zero-valued under the
    // ICAO 7-3-1 arithmetic (`char_value('<') == char_value('0') == 0`), so
    // the composite digit is identical either way. These two tests pin both
    // issuer-option forms independently through the real parser, not just
    // the `checksum::verify` primitive.

    #[test]
    fn td3_empty_personal_number_with_filler_check_digit_zero() {
        // Personal number all fillers, check digit '0' (value 0). Composite
        // recomputed independently of the crate (not copied from TD3_L2,
        // whose personal number differs): 8, not the '6' this fixture
        // originally carried unasserted — see the comment above.
        let l2 = "L898902C36UTO7408122F1204159<<<<<<<<<<<<<<08";
        let d = parse_td3(TD3_L1, l2).unwrap();
        assert!(d.checks.personal_number);
        assert!(d.checks.composite);
        assert_eq!(d.personal_number, None);
    }

    #[test]
    fn td3_empty_personal_number_with_filler_check_digit_filler() {
        // Same fixture, check digit '<' instead of '0' — same composite
        // digit, since both digit characters have ICAO value 0.
        let l2 = "L898902C36UTO7408122F1204159<<<<<<<<<<<<<<<8";
        let d = parse_td3(TD3_L1, l2).unwrap();
        assert!(d.checks.personal_number);
        assert!(d.checks.composite);
        assert_eq!(d.personal_number, None);
    }

    // Same Utopia/Eriksson identity as the TD3/TD2 specimens, reshaped into
    // TD1's 3-line layout. Part 5's own Appendix A/B examples are images in
    // the source PDF, not extracted as text
    // (`knowledge/docs9303/Doc_9303_Part5_Specs_for_TD1_MROTDs.md:504-534`);
    // this zone is corroborated by cross-part agreement with the literal TD2
    // specimen (same document number, dates and name), not a published Part 5
    // text specimen.
    const TD1_L1: &str = "I<UTOD231458907<<<<<<<<<<<<<<<";
    const TD1_L2: &str = "7408122F1204159UTO<<<<<<<<<<<6";
    const TD1_L3: &str = "ERIKSSON<<ANNA<MARIA<<<<<<<<<<";

    #[test]
    fn td1_specimen_fully_valid() {
        let d = parse_td1(TD1_L1, TD1_L2, TD1_L3).unwrap();
        assert!(d.valid(), "checks: {:?}", d.checks);
        assert_eq!(d.format, Format::Td1);
        assert_eq!(d.document_type, "I");
        assert_eq!(d.document_number, "D23145890");
        assert_eq!(d.surname, "ERIKSSON");
        assert_eq!(d.given_names, "ANNA MARIA");
        assert_eq!(d.date_of_birth, "1974-08-12");
        assert_eq!(d.date_of_birth_completeness, DateCompleteness::Complete);
        assert_eq!(d.date_of_expiry, "2012-04-15");
    }

    #[test]
    fn td1_all_filler_dob_reports_unknown() {
        // Same all-filler-DOB scenario as the TD3 test above, but for TD1's
        // layout, where the dob+check field is line2[0..7] rather than
        // line2[13..20] — covers a second of the five formats per the S5 plan.
        let l2 = format!("{}{}", "<<<<<<0", &TD1_L2[7..]);
        assert_eq!(l2.len(), TD1_L2.len());
        let d = parse_td1(TD1_L1, &l2, TD1_L3).unwrap();
        assert!(d.checks.date_of_birth);
        assert_eq!(d.date_of_birth, "<<<<<<");
        assert_eq!(d.date_of_birth_completeness, DateCompleteness::Unknown);
    }

    // Official ICAO 9303 part 6 TD2 specimen (Utopia / Anna Maria Eriksson),
    // published verbatim as text:
    // `knowledge/docs9303/Doc_9303_Part6_Specs_for_TD2_MROTDs.md:487-488`.
    const TD2_L1: &str = "I<UTOERIKSSON<<ANNA<MARIA<<<<<<<<<<<";
    const TD2_L2: &str = "D231458907UTO7408122F1204159<<<<<<<6";

    #[test]
    fn td2_specimen_fully_valid() {
        let d = parse_td2(TD2_L1, TD2_L2).unwrap();
        assert!(d.valid(), "checks: {:?}", d.checks);
        assert_eq!(d.format, Format::Td2);
        assert_eq!(d.document_type, "I");
        assert_eq!(d.issuing_country, "UTO");
        assert_eq!(d.document_number, "D23145890");
        assert_eq!(d.surname, "ERIKSSON");
        assert_eq!(d.given_names, "ANNA MARIA");
        assert_eq!(d.nationality, "UTO");
        assert_eq!(d.date_of_birth, "1974-08-12");
        assert_eq!(d.sex, "F");
        assert_eq!(d.date_of_expiry, "2012-04-15");
    }

    #[test]
    fn td2_tampered_expiry_fails_checksum() {
        let tampered = TD2_L2.replacen("120415", "120416", 1);
        let d = parse_td2(TD2_L1, &tampered).unwrap();
        assert!(!d.checks.date_of_expiry);
        assert!(!d.checks.composite);
        assert!(!d.valid());
    }

    #[test]
    fn td2_found_in_ocr_text() {
        let text = format!("## IDENTITY CARD\n\nnoise\n\n{TD2_L1}\n{TD2_L2}\n\nfooter");
        let d = find_and_parse(&text).unwrap();
        assert!(d.valid(), "checks: {:?}", d.checks);
        assert_eq!(d.format, Format::Td2);
        assert_eq!(d.surname, "ERIKSSON");
        assert_eq!(d.document_number, "D23145890");
    }

    #[test]
    fn find_in_ocr_noise() {
        let text = format!(
            "## REPUBLIC OF UTOPIA\n\nSome OCR noise here\n\n{}\n{}\n\nfooter",
            // OCR quirks: lowercase, stray spaces, « for <<, dropped fillers.
            "p<utoeriksson«anna<maria<<<<<<<<<<<<<<<<<",
            "L898902C36UTO7408122F1204159ZE184226B<<<<<10"
        );
        let d = find_and_parse(&text).unwrap();
        assert!(d.valid());
        assert_eq!(d.surname, "ERIKSSON");
    }

    #[test]
    fn find_html_escaped_and_merged_lines() {
        // Real docling output shape: fillers escaped as &lt; and both TD3
        // lines on one physical markdown line (Croatian specimen).
        let text = "## PUTOVNICA\n\nP&lt;HRVSPECIMEN&lt;&lt;SPECIMEN&lt;&lt;&lt;&lt;&lt;&lt;&lt;&lt;&lt;&lt;&lt;&lt;&lt;&lt;&lt;&lt;&lt;&lt;&lt;&lt;&lt; 0070070071HRV8212258F1407019&lt;&lt;&lt;&lt;&lt;&lt;&lt;&lt;&lt;&lt;&lt;&lt;&lt;&lt;06\n";
        let d = find_and_parse(text).unwrap();
        assert_eq!(d.surname, "SPECIMEN");
        assert_eq!(d.document_number, "007007007");
        assert_eq!(d.issuing_country, "HRV");
        assert!(d.checks.document_number);
        assert!(d.checks.date_of_birth);
        assert!(d.checks.date_of_expiry);
    }

    #[test]
    fn checksum_verified_ocr_repair() {
        // Verbatim tesseract.js output for the Croatian specimen at low
        // resolution: trailing fillers read as K/L runs, a hallucinated
        // leading '1' on line 2 (45 chars), and 'B' where '8' is printed.
        // The check digits prove which repaired variant is the true read.
        let text = "I 01072009 PUJZAGREB 0\n\nBIDFD WH5SS A 2\n\n01072014\nP<HRVSPECIMEN<<SPECIMEN<KLLLLLLLLLLLLLLLLLKLKL\n10070070071HRVB212258F1407019<<<<<<<<<<<<<<06\n";
        let d = find_and_parse(text).unwrap();
        assert!(d.valid(), "checks: {:?}", d.checks);
        assert_eq!(d.surname, "SPECIMEN");
        assert_eq!(d.given_names, "SPECIMEN");
        assert_eq!(d.document_number, "007007007");
        assert_eq!(d.date_of_birth, "1982-12-25");
    }

    #[test]
    fn ocr_repair_dropped_filler_mid_line() {
        // Second verbatim tesseract.js reading of the same specimen: an
        // L-run inside the personal-number field and one filler DROPPED
        // (43 chars) — the missing character must be re-inserted inside the
        // filler run, not appended, or the check digits shift.
        let text = "RF 01072009 PUZAGREB\n01072014\nP<HRVSPECIMEN<<SPECIMEN<<K<KLLLLLLLLLLLLLLLLKLKL\n0070070071HRVB212258F1407019<<<<LLLLLLL<<06\n";
        let d = find_and_parse(text).unwrap();
        assert!(d.valid(), "checks: {:?}", d.checks);
        assert_eq!(d.surname, "SPECIMEN");
        assert_eq!(d.document_number, "007007007");
        assert_eq!(d.personal_number, None);
    }

    #[test]
    fn td1_from_single_docling_line_with_k_misreads() {
        // Verbatim docling OCR of the Slovenian 2022 specimen ID card rear:
        // all three TD1 lines in ONE paragraph, `<` escaped as &lt;, and the
        // usual K-for-filler misreads (IK→I<, 145K<→145<<, VZORECKK→VZOREC<<).
        let text = "1F9874543\n\nIKSVNIE987654302806985505145K&lt; 8506287F3203282SVN&lt;&lt;&lt;&lt;&lt;&lt;&lt;&lt;&lt;&lt;&lt;2 VZORECKKJANAKKKKKKKKK&lt;&lt;KK";
        let d = find_and_parse(text).unwrap();
        assert!(d.valid(), "checks: {:?}", d.checks);
        assert_eq!(d.format, Format::Td1);
        assert_eq!(d.document_type, "I");
        assert_eq!(d.issuing_country, "SVN");
        assert_eq!(d.document_number, "IE9876543");
        assert_eq!(d.surname, "VZOREC");
        assert_eq!(d.given_names, "JANA");
        assert_eq!(d.date_of_birth, "1985-06-28");
        assert_eq!(d.date_of_expiry, "2032-03-28");
        // The trailing K in the EMŠO field is a filler misread that check
        // digits cannot catch (K ≡ < mod 10) — heuristic cleanup handles it.
        assert_eq!(d.personal_number.as_deref(), Some("2806985505145"));
    }

    #[test]
    fn ocr_repair_deeply_truncated_name_line() {
        // Verbatim ocrs output for the Croatian specimen at 600×421: line 2 is
        // read perfectly, but line 1 loses NINE trailing fillers (35/44 chars)
        // and its `<` document-code filler is misread as `K`. The name line
        // carries no check digit of its own, so padding the filler run back is
        // safe — line 2's check digits still prove the read.
        let text = "PUTOVNICA\nPKHRVSPECIMEN<<SPECIMEN<<<<<<<<<<<<\n0070070071HRV8212258F1407019<<<<<<<<<<<<<<06\n";
        let d = find_and_parse(text).unwrap();
        assert!(d.valid(), "checks: {:?}", d.checks);
        assert_eq!(d.surname, "SPECIMEN");
        assert_eq!(d.given_names, "SPECIMEN");
        assert_eq!(d.document_number, "007007007");
    }

    #[test]
    fn invalid_checksums_still_reported() {
        // A tampered MRZ parses but is flagged invalid rather than dropped.
        let tampered = TD3_L2.replacen("740812", "750812", 1);
        let text = format!("{TD3_L1}\n{tampered}");
        let d = find_and_parse(&text).unwrap();
        assert!(!d.valid());
        assert!(!d.checks.date_of_birth);
    }

    #[test]
    fn find_nothing_in_plain_text() {
        assert_eq!(
            find_and_parse("just a regular paragraph\nwith two lines"),
            Err(MrzError::NotFound)
        );
    }

    // ICAO 9303 part 7's own published MRV-A specimen, transcribed verbatim
    // from
    // `knowledge/docs9303/Doc_9303_Part7_Machine_Readable_Visas_MRVs.md:1104-1106`.
    // Unlike the hand-derived pair below, every character here comes from the
    // standard, so this is the MRV equivalent of the part 4/5/6 specimens.
    const MRV_A_ICAO_L1: &str = "V<UTOERIKSSON<<ANNA<MARIA<<<<<<<<<<<<<<<<<<<";
    const MRV_A_ICAO_L2: &str = "L898902C<3UTO6908061F9406236ZE184226B<<<<<<<";

    // ICAO 9303 part 7's own published MRV-B specimen, from the same appendix:
    // `knowledge/docs9303/Doc_9303_Part7_Machine_Readable_Visas_MRVs.md:1136-1138`.
    const MRV_B_ICAO_L1: &str = "V<UTOERIKSSON<<ANNA<MARIA<<<<<<<<<<<";
    const MRV_B_ICAO_L2: &str = "L898902C<3UTO6908061F9406236ZE184226";

    #[test]
    fn mrv_a_icao_published_specimen_fully_valid() {
        let d = parse_mrv_a(MRV_A_ICAO_L1, MRV_A_ICAO_L2).unwrap();
        assert!(d.valid(), "checks: {:?}", d.checks);
        assert_eq!(d.format, Format::MrvA);
        assert_eq!(d.document_type, "V");
        assert_eq!(d.issuing_country, "UTO");
        assert_eq!(d.surname, "ERIKSSON");
        assert_eq!(d.given_names, "ANNA MARIA");
        assert_eq!(d.nationality, "UTO");
        assert_eq!(d.sex, "F");
        // The document-number field is `L898902C<`: eight characters padded to
        // nine with a filler, check digit 3. The filler is padding, not the
        // long-number signal — that would need the *check digit* position to
        // be a filler — so no overflow is read here.
        assert_eq!(d.document_number, "L898902C");
        assert_eq!(d.document_number_full, None);
        assert_eq!(d.personal_number.as_deref(), Some("ZE184226B"));
        assert_eq!(d.date_of_birth, "1969-08-06");
        // Doc 9303 defines no century rule (part 3 §4.8 is silent), so this
        // crate's own policy applies: expiry is always read as 20xx. ICAO's
        // specimen is a 1990s document, so that policy renders 940623 as 2094
        // rather than 1994. Pinned deliberately — it is the documented
        // heuristic behaving as designed, and the clearest illustration of its
        // limit. See `dates::expand_date`.
        assert_eq!(d.date_of_expiry, "2094-06-23");
    }

    #[test]
    fn mrv_b_icao_published_specimen_fully_valid() {
        let d = parse_mrv_b(MRV_B_ICAO_L1, MRV_B_ICAO_L2).unwrap();
        assert!(d.valid(), "checks: {:?}", d.checks);
        assert_eq!(d.format, Format::MrvB);
        assert_eq!(d.document_type, "V");
        assert_eq!(d.issuing_country, "UTO");
        assert_eq!(d.surname, "ERIKSSON");
        assert_eq!(d.given_names, "ANNA MARIA");
        assert_eq!(d.nationality, "UTO");
        assert_eq!(d.document_number, "L898902C");
        assert_eq!(d.document_number_full, None);
        assert_eq!(d.personal_number.as_deref(), Some("ZE184226"));
        assert_eq!(d.date_of_birth, "1969-08-06");
        assert_eq!(d.date_of_expiry, "2094-06-23");
    }

    // Hand-derived by this crate — NOT an ICAO-published specimen. Kept
    // alongside the published pair above because it exercises a different
    // nationality, dates and optional-data shape, and because the tamper
    // tests below are wired to its values. Part 7's real MRV-A worked example
    // is the one transcribed above:
    // `knowledge/docs9303/Doc_9303_Part7_Machine_Readable_Visas_MRVs.md:1104-1106`.
    // Line 2 (44 chars); check-digit arithmetic (7-3-1, values A-Z=10-35,
    // <=0), worked independently so the constant is verified rather than
    // merely asserted:
    //   doc#   XK9305487: X=33,K=20,9,3,0,5,4,8,7 * 7,3,1,7,3,1,7,3,1
    //          = 231+60+9+21+0+5+28+24+7 = 385 -> 385 mod 10 = 5
    //   DOB    850221: 8,5,0,2,2,1 * 7,3,1,7,3,1
    //          = 56+15+0+14+6+1 = 92 -> 92 mod 10 = 2
    //   expiry 270314: 2,7,0,3,1,4 * 7,3,1,7,3,1
    //          = 14+21+0+21+3+4 = 63 -> 63 mod 10 = 3
    const MRV_A_L1: &str = "V<UTOERIKSSON<<ANNA<MARIA<<<<<<<<<<<<<<<<<<<";
    const MRV_A_L2: &str = "XK93054875BRA8502212F2703143R5T6U7V8W9<<<<<<";

    #[test]
    fn mrv_a_specimen_fully_valid() {
        let d = parse_mrv_a(MRV_A_L1, MRV_A_L2).unwrap();
        assert!(d.valid(), "checks: {:?}", d.checks);
        assert_eq!(d.format, Format::MrvA);
        assert_eq!(d.document_type, "V");
        assert_eq!(d.issuing_country, "UTO");
        assert_eq!(d.surname, "ERIKSSON");
        assert_eq!(d.given_names, "ANNA MARIA");
        assert_eq!(d.nationality, "BRA");
        assert_eq!(d.date_of_birth, "1985-02-21");
        assert_eq!(d.sex, "F");
        assert_eq!(d.date_of_expiry, "2027-03-14");
        assert_eq!(d.document_number, "XK9305487");
        assert_eq!(d.personal_number.as_deref(), Some("R5T6U7V8W9"));
    }

    // Hand-derived by this crate — NOT an ICAO-published specimen. Kept for
    // the same reason as the MRV-A pair above; Part 7's real MRV-B worked
    // example is transcribed as `MRV_B_ICAO_L1`/`MRV_B_ICAO_L2`:
    // `knowledge/docs9303/Doc_9303_Part7_Machine_Readable_Visas_MRVs.md:1136-1138`.
    // Line 2 (36 chars); check-digit arithmetic (7-3-1, values A-Z=10-35,
    // <=0), worked independently so the constant is verified rather than
    // merely asserted:
    //   doc#   L23456789: L=21,2,3,4,5,6,7,8,9 * 7,3,1,7,3,1,7,3,1
    //          = 147+6+3+28+15+6+49+24+9 = 287 -> 287 mod 10 = 7
    //   DOB    920101: 9,2,0,1,0,1 * 7,3,1,7,3,1
    //          = 63+6+0+7+0+1 = 77 -> 77 mod 10 = 7
    //   expiry 270630: 2,7,0,6,3,0 * 7,3,1,7,3,1
    //          = 14+21+0+42+9+0 = 86 -> 86 mod 10 = 6
    const MRV_B_L1: &str = "V<UTOERIKSSON<<ANNA<MARIA<<<<<<<<<<<";
    const MRV_B_L2: &str = "L234567897DEU9201017F2706306QW12ER34";

    #[test]
    fn mrv_b_specimen_fully_valid() {
        let d = parse_mrv_b(MRV_B_L1, MRV_B_L2).unwrap();
        assert!(d.valid(), "checks: {:?}", d.checks);
        assert_eq!(d.format, Format::MrvB);
        assert_eq!(d.nationality, "DEU");
        assert_eq!(d.date_of_birth, "1992-01-01");
        assert_eq!(d.date_of_expiry, "2027-06-30");
        assert_eq!(d.document_number, "L23456789");
        assert_eq!(d.personal_number.as_deref(), Some("QW12ER34"));
    }

    #[test]
    fn mrv_a_tampered_dob_fails_checksum() {
        let tampered = MRV_A_L2.replacen("850221", "860221", 1);
        let d = parse_mrv_a(MRV_A_L1, &tampered).unwrap();
        assert!(!d.checks.date_of_birth);
        assert!(!d.valid());
    }

    #[test]
    fn mrv_rejects_non_v_document_code() {
        let line1 = "P<UTOERIKSSON<<ANNA<MARIA<<<<<<<<<<<<<<<<<<<";
        let result = parse_mrv_a(line1, MRV_A_L2);
        assert!(matches!(result, Err(MrzError::BadDocumentCode(_))));
    }

    #[test]
    fn mrv_a_found_in_ocr_text() {
        let text = format!("## VISA\n\nnoise\n\n{MRV_A_L1}\n{MRV_A_L2}\n\nfooter");
        let d = find_and_parse(&text).unwrap();
        assert!(d.valid(), "checks: {:?}", d.checks);
        assert_eq!(d.format, Format::MrvA);
        assert_eq!(d.surname, "ERIKSSON");
        assert_eq!(d.nationality, "BRA");
    }

    #[test]
    fn mrv_b_found_in_ocr_text() {
        let text = format!("## VISA\n\nnoise\n\n{MRV_B_L1}\n{MRV_B_L2}\n\nfooter");
        let d = find_and_parse(&text).unwrap();
        assert!(d.valid(), "checks: {:?}", d.checks);
        assert_eq!(d.format, Format::MrvB);
        assert_eq!(d.nationality, "DEU");
    }

    #[test]
    fn mrv_found_html_escaped_or_merged() {
        // HTML-escaped fillers.
        let escaped_l1 = MRV_A_L1.replace('<', "&lt;");
        let escaped_l2 = MRV_A_L2.replace('<', "&lt;");
        let text = format!("## VISA\n\n{escaped_l1}\n{escaped_l2}\n");
        let d = find_and_parse(&text).unwrap();
        assert!(d.valid(), "checks: {:?}", d.checks);
        assert_eq!(d.format, Format::MrvA);

        // Both lines merged onto one physical line with no separator at all
        // (the ~88-char merged-line case, like docling's single-paragraph
        // OCR output for a TD3 specimen).
        let merged = format!("## VISA\n\n{MRV_A_L1}{MRV_A_L2}\n");
        let d = find_and_parse(&merged).unwrap();
        assert!(d.valid(), "checks: {:?}", d.checks);
        assert_eq!(d.format, Format::MrvA);
    }

    // ---- Document-number overflow (Part 5/6 note j; TD3 by analogy — see
    // parser::read_overflow's doc comment) ----

    #[test]
    fn td3_long_document_number_round_trips() {
        let fields = Td3Fields {
            document_code: "P".into(),
            issuing_country: "UTO".into(),
            document_number: "L898902C31234".into(), // 13 chars, overflows 9
            surname: "ERIKSSON".into(),
            given_names: "ANNA MARIA".into(),
            nationality: "UTO".into(),
            date_of_birth: "740812".into(),
            sex: "F".into(),
            date_of_expiry: "120415".into(),
            personal_number: None,
        };
        let mrz = format_td3(&fields);
        let (l1, l2) = mrz.split_once('\n').unwrap();
        // Nine principal characters, then a filler where the check digit goes.
        assert_eq!(&l2[0..10], "L898902C3<");
        let d = parse_td3(l1, l2).unwrap();
        assert!(d.valid(), "checks: {:?}", d.checks);
        assert_eq!(d.document_number_full.as_deref(), Some("L898902C31234"));
        assert_eq!(d.full_document_number(), "L898902C31234");
        assert!(!d.document_number_legacy_encoding);
        // The 9-char field reading stays available and unsurprising.
        assert_eq!(d.document_number, "L898902C3");
        assert_eq!(d.personal_number, None);
    }

    #[test]
    fn overflow_coexists_with_personal_number() {
        let fields = Td3Fields {
            document_number: "AB1234567890".into(), // 12 chars
            personal_number: Some("ZE184".into()),
            ..Td3Fields::default()
        };
        let d = parse_td3_str(&format_td3(&fields));
        assert!(d.valid(), "checks: {:?}", d.checks);
        assert_eq!(d.document_number_full.as_deref(), Some("AB1234567890"));
        assert_eq!(d.personal_number.as_deref(), Some("ZE184"));
    }

    #[test]
    fn td2_and_td1_long_document_numbers_round_trip() {
        let td2 = Td2Fields {
            document_number: "D23145890XY".into(), // 11 chars; remainder fits 7
            date_of_birth: "740812".into(),
            date_of_expiry: "120415".into(),
            ..Td2Fields::default()
        };
        let mrz = format_td2(&td2);
        let (l1, l2) = mrz.split_once('\n').unwrap();
        let d = parse_td2(l1, l2).unwrap();
        assert!(d.valid(), "checks: {:?}", d.checks);
        assert_eq!(d.document_number_full.as_deref(), Some("D23145890XY"));

        let td1 = Td1Fields {
            document_number: "D23145890ABCDE".into(), // 14 chars; remainder fits 15
            date_of_birth: "740812".into(),
            date_of_expiry: "120415".into(),
            ..Td1Fields::default()
        };
        let mrz = format_td1(&td1);
        let mut lines = mrz.lines();
        let d = parse_td1(
            lines.next().unwrap(),
            lines.next().unwrap(),
            lines.next().unwrap(),
        )
        .unwrap();
        assert!(d.valid(), "checks: {:?}", d.checks);
        assert_eq!(d.document_number_full.as_deref(), Some("D23145890ABCDE"));
    }

    #[test]
    fn overflow_remainder_too_long_falls_back_to_truncation() {
        // TD2's optional field is 7 wide, so a remainder of 6 + check + filler
        // does not fit — the number is truncated to 9 as it always was, and the
        // ordinary (non-overflow) encoding still validates.
        let td2 = Td2Fields {
            document_number: "D23145890ABCDEF".into(), // remainder 7 → needs 9
            date_of_birth: "740812".into(),
            date_of_expiry: "120415".into(),
            ..Td2Fields::default()
        };
        let mrz = format_td2(&td2);
        let (l1, l2) = mrz.split_once('\n').unwrap();
        let d = parse_td2(l1, l2).unwrap();
        assert!(d.valid(), "checks: {:?}", d.checks);
        assert_eq!(d.document_number_full, None);
        assert_eq!(d.document_number, "D23145890");
    }

    #[test]
    fn tampered_overflow_remainder_fails_the_check_digit() {
        let fields = Td3Fields {
            document_number: "L898902C31234".into(),
            ..Td3Fields::default()
        };
        let mrz = format_td3(&fields);
        // The nine principal characters ("L898902C3") are printed in the
        // number field; only the 4-character remainder ("1234") is written
        // into the personal-number field, so corrupting the remainder means
        // corrupting "1234", not the pre-0.6 8-character-boundary "31234".
        let tampered = mrz.replacen("1234", "1235", 1);
        let d = parse_td3_str(&tampered);
        assert!(!d.checks.document_number);
        assert!(!d.valid());
        assert!(d.checks.failed().contains(&Field::DocumentNumber));
    }

    #[test]
    fn ordinary_specimens_report_no_overflow() {
        assert_eq!(
            parse_td3(TD3_L1, TD3_L2).unwrap().document_number_full,
            None
        );
        assert_eq!(
            parse_td1(TD1_L1, TD1_L2, TD1_L3)
                .unwrap()
                .document_number_full,
            None
        );
        assert_eq!(
            parse_td2(TD2_L1, TD2_L2).unwrap().document_number_full,
            None
        );
        // Empty document-number field is a blank field, not an overflow.
        let blank = "<<<<<<<<<<UTO7408122F1204159<<<<<<<<<<<<<<02";
        assert_eq!(parse_td3(TD3_L1, blank).unwrap().document_number_full, None);
    }

    fn parse_td3_str(mrz: &str) -> MrzData {
        let (l1, l2) = mrz.split_once('\n').unwrap();
        parse_td3(l1, l2).unwrap()
    }

    // ---- ParseOptions ----

    #[test]
    fn pivot_is_configurable_per_call() {
        // Birth dates land in the past relative to the pivot, expiry ahead.
        let d = parse_td3(TD3_L1, TD3_L2).unwrap();
        assert_eq!(d.date_of_birth, "1974-08-12");

        // With a pivot of 80, YY=74 reads as 2074 rather than 1974.
        let opts = ParseOptions { pivot_yy: 80 };
        let d = parse_td3_with(TD3_L1, TD3_L2, &opts).unwrap();
        assert_eq!(d.date_of_birth, "2074-08-12");
        // Check digits are untouched by the pivot — it only affects display.
        assert!(d.valid());
    }

    #[test]
    fn default_options_match_the_plain_entry_points() {
        let opts = ParseOptions::default();
        assert_eq!(opts.pivot_yy, CURRENT_YY);
        assert_eq!(
            parse_td3(TD3_L1, TD3_L2).unwrap(),
            parse_td3_with(TD3_L1, TD3_L2, &opts).unwrap()
        );
        let text = format!("## VISA\n\n{MRV_A_L1}\n{MRV_A_L2}\n");
        assert_eq!(
            find_and_parse(&text).unwrap(),
            find_and_parse_with(&text, &opts).unwrap()
        );
    }

    // ---- Checks diagnostics ----

    #[test]
    fn failed_lists_the_failing_fields() {
        let clean = parse_td3(TD3_L1, TD3_L2).unwrap();
        assert!(clean.checks.failed().is_empty());

        let tampered = TD3_L2.replacen("740812", "750812", 1);
        let d = parse_td3(TD3_L1, &tampered).unwrap();
        assert_eq!(
            d.checks.failed(),
            vec![Field::DateOfBirth, Field::Composite]
        );
        assert_eq!(Field::DateOfBirth.to_string(), "date_of_birth");
    }

    #[test]
    fn scanner_keeps_the_best_scoring_partial_read() {
        // Two check digits wrong: the returned record must be the real zone
        // with exactly those two flagged, not some worse-scoring variant.
        let tampered = TD3_L2
            .replacen("740812", "750812", 1)
            .replacen("120415", "120416", 1);
        let text = format!("noise\n\n{TD3_L1}\n{tampered}\n\nfooter");
        let d = find_and_parse(&text).unwrap();
        assert!(!d.valid());
        assert_eq!(d.surname, "ERIKSSON");
        assert_eq!(d.document_number, "L898902C3");
        assert!(d.checks.document_number);
        assert!(!d.checks.date_of_birth);
        assert!(!d.checks.date_of_expiry);
    }

    #[test]
    fn country_names_surface_on_mrzdata() {
        let d = parse_td1(TD1_L1, TD1_L2, TD1_L3).unwrap();
        assert_eq!(d.issuing_country_name(), Some("Utopia (ICAO specimen)"));
        assert_eq!(d.nationality_name(), Some("Utopia (ICAO specimen)"));
    }
}