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
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
1259
1260
1261
1262
1263
1264
1265
1266
1267
1268
1269
1270
1271
1272
1273
1274
1275
1276
1277
1278
1279
1280
1281
1282
1283
1284
1285
1286
1287
1288
1289
1290
1291
1292
1293
1294
1295
1296
1297
1298
1299
1300
1301
1302
1303
1304
1305
1306
1307
1308
1309
1310
1311
1312
1313
1314
1315
1316
1317
1318
1319
1320
1321
1322
1323
1324
1325
1326
1327
1328
1329
1330
1331
1332
1333
1334
1335
1336
1337
1338
1339
1340
1341
1342
1343
1344
1345
1346
1347
1348
1349
1350
1351
1352
1353
1354
1355
1356
1357
1358
1359
1360
1361
1362
1363
1364
1365
1366
1367
1368
1369
1370
1371
1372
1373
1374
1375
1376
1377
1378
1379
1380
1381
1382
1383
1384
1385
1386
1387
1388
1389
1390
1391
1392
1393
1394
1395
1396
1397
1398
1399
1400
1401
1402
1403
1404
1405
1406
1407
1408
1409
1410
1411
1412
1413
1414
1415
1416
1417
1418
1419
1420
1421
1422
1423
1424
1425
1426
1427
1428
1429
1430
1431
1432
1433
1434
1435
1436
1437
1438
1439
1440
1441
1442
1443
1444
1445
1446
1447
1448
1449
1450
1451
1452
1453
1454
1455
1456
1457
1458
1459
1460
1461
1462
1463
1464
1465
1466
1467
1468
1469
1470
1471
1472
1473
1474
1475
1476
1477
1478
1479
1480
1481
1482
1483
1484
1485
1486
1487
1488
1489
1490
1491
1492
1493
1494
1495
1496
1497
1498
1499
1500
1501
1502
1503
1504
1505
1506
1507
1508
1509
1510
1511
1512
1513
1514
1515
1516
1517
1518
1519
1520
1521
1522
1523
1524
1525
1526
1527
1528
1529
1530
1531
1532
1533
1534
1535
1536
1537
1538
1539
1540
1541
1542
1543
1544
1545
1546
1547
1548
1549
1550
1551
1552
1553
1554
1555
1556
1557
1558
1559
1560
1561
1562
1563
1564
1565
1566
1567
1568
1569
1570
1571
1572
1573
1574
1575
1576
1577
1578
1579
1580
1581
1582
1583
1584
1585
1586
1587
1588
1589
1590
1591
1592
1593
1594
1595
1596
1597
1598
1599
1600
1601
1602
1603
1604
1605
1606
1607
1608
1609
1610
1611
1612
1613
1614
1615
1616
1617
1618
1619
1620
1621
1622
1623
1624
1625
1626
1627
1628
1629
1630
1631
1632
1633
1634
1635
1636
1637
1638
1639
1640
1641
1642
1643
1644
1645
1646
1647
1648
1649
1650
1651
1652
1653
1654
1655
1656
1657
1658
1659
1660
1661
1662
1663
1664
1665
1666
1667
1668
1669
1670
1671
1672
1673
1674
1675
1676
1677
1678
1679
1680
1681
1682
1683
1684
1685
1686
1687
1688
1689
1690
1691
1692
1693
1694
1695
1696
1697
1698
1699
1700
1701
1702
1703
1704
1705
1706
1707
1708
1709
1710
1711
1712
1713
1714
1715
1716
1717
1718
1719
1720
1721
1722
1723
1724
1725
1726
1727
1728
1729
1730
1731
1732
1733
1734
1735
1736
1737
1738
1739
1740
1741
1742
1743
1744
1745
1746
1747
1748
1749
1750
1751
1752
1753
1754
1755
1756
1757
1758
1759
1760
1761
1762
1763
1764
1765
1766
1767
1768
1769
1770
1771
1772
1773
1774
1775
1776
1777
1778
1779
1780
1781
1782
1783
1784
1785
1786
1787
1788
1789
1790
1791
1792
1793
1794
1795
1796
1797
1798
1799
1800
1801
1802
1803
1804
1805
1806
1807
1808
1809
1810
1811
1812
1813
1814
1815
1816
1817
1818
1819
1820
1821
1822
1823
1824
1825
1826
1827
1828
1829
1830
1831
1832
1833
1834
1835
1836
1837
1838
1839
1840
1841
1842
1843
1844
1845
1846
1847
1848
1849
1850
1851
1852
1853
1854
1855
1856
1857
1858
1859
1860
1861
1862
1863
1864
1865
1866
1867
1868
1869
1870
1871
1872
1873
1874
1875
1876
1877
1878
1879
1880
1881
1882
1883
1884
1885
1886
1887
1888
1889
1890
1891
1892
1893
1894
1895
1896
1897
1898
1899
1900
1901
1902
1903
1904
1905
1906
1907
1908
1909
1910
1911
1912
1913
1914
1915
1916
1917
1918
1919
1920
1921
1922
1923
1924
1925
1926
1927
1928
1929
1930
1931
1932
1933
1934
1935
1936
1937
1938
1939
1940
1941
1942
1943
1944
1945
1946
1947
1948
1949
1950
1951
1952
1953
1954
1955
1956
1957
1958
1959
1960
1961
1962
1963
1964
1965
1966
1967
1968
1969
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
2031
2032
2033
2034
2035
2036
2037
2038
2039
2040
2041
2042
2043
2044
2045
2046
2047
2048
2049
2050
2051
2052
2053
2054
2055
2056
2057
2058
2059
2060
2061
2062
2063
2064
2065
2066
2067
2068
2069
2070
2071
2072
2073
2074
2075
2076
2077
2078
2079
2080
2081
2082
2083
2084
2085
2086
2087
2088
2089
2090
2091
2092
2093
2094
2095
2096
2097
2098
2099
2100
2101
2102
2103
2104
2105
2106
2107
2108
2109
2110
2111
2112
2113
2114
2115
2116
2117
2118
2119
2120
2121
2122
2123
2124
2125
2126
2127
2128
2129
2130
2131
2132
2133
2134
2135
2136
2137
2138
2139
2140
// Formatting and tools for comments.

use std::{borrow::Cow, iter};

use itertools::{multipeek, MultiPeek};
use rustc_span::Span;
use tracing::{debug, trace};

use crate::config::Config;
use crate::rewrite::RewriteContext;
use crate::shape::{Indent, Shape};
use crate::string::{rewrite_string, StringFormat};
use crate::utils::{
    count_newlines, first_line_width, last_line_width, trim_left_preserve_layout,
    trimmed_last_line_width, unicode_str_width,
};
use crate::{ErrorKind, FormattingError};

fn is_custom_comment(comment: &str) -> bool {
    if !comment.starts_with("//") {
        false
    } else if let Some(c) = comment.chars().nth(2) {
        !c.is_alphanumeric() && !c.is_whitespace()
    } else {
        false
    }
}

#[derive(Copy, Clone, PartialEq, Eq)]
pub(crate) enum CommentStyle<'a> {
    DoubleSlash,
    TripleSlash,
    Doc,
    SingleBullet,
    DoubleBullet,
    Exclamation,
    Custom(&'a str),
}

fn custom_opener(s: &str) -> &str {
    s.lines().next().map_or("", |first_line| {
        first_line
            .find(' ')
            .map_or(first_line, |space_index| &first_line[0..=space_index])
    })
}

impl<'a> CommentStyle<'a> {
    /// Returns `true` if the commenting style cannot span multiple lines.
    pub(crate) fn is_line_comment(&self) -> bool {
        matches!(
            self,
            CommentStyle::DoubleSlash
                | CommentStyle::TripleSlash
                | CommentStyle::Doc
                | CommentStyle::Custom(_)
        )
    }

    /// Returns `true` if the commenting style can span multiple lines.
    pub(crate) fn is_block_comment(&self) -> bool {
        matches!(
            self,
            CommentStyle::SingleBullet | CommentStyle::DoubleBullet | CommentStyle::Exclamation
        )
    }

    /// Returns `true` if the commenting style is for documentation.
    pub(crate) fn is_doc_comment(&self) -> bool {
        matches!(*self, CommentStyle::TripleSlash | CommentStyle::Doc)
    }

    pub(crate) fn opener(&self) -> &'a str {
        match *self {
            CommentStyle::DoubleSlash => "// ",
            CommentStyle::TripleSlash => "/// ",
            CommentStyle::Doc => "//! ",
            CommentStyle::SingleBullet => "/* ",
            CommentStyle::DoubleBullet => "/** ",
            CommentStyle::Exclamation => "/*! ",
            CommentStyle::Custom(opener) => opener,
        }
    }

    pub(crate) fn closer(&self) -> &'a str {
        match *self {
            CommentStyle::DoubleSlash
            | CommentStyle::TripleSlash
            | CommentStyle::Custom(..)
            | CommentStyle::Doc => "",
            CommentStyle::SingleBullet | CommentStyle::DoubleBullet | CommentStyle::Exclamation => {
                " */"
            }
        }
    }

    pub(crate) fn line_start(&self) -> &'a str {
        match *self {
            CommentStyle::DoubleSlash => "// ",
            CommentStyle::TripleSlash => "/// ",
            CommentStyle::Doc => "//! ",
            CommentStyle::SingleBullet | CommentStyle::DoubleBullet | CommentStyle::Exclamation => {
                " * "
            }
            CommentStyle::Custom(opener) => opener,
        }
    }

    pub(crate) fn to_str_tuplet(&self) -> (&'a str, &'a str, &'a str) {
        (self.opener(), self.closer(), self.line_start())
    }
}

pub(crate) fn comment_style(orig: &str, normalize_comments: bool) -> CommentStyle<'_> {
    if !normalize_comments {
        if orig.starts_with("/**") && !orig.starts_with("/**/") {
            CommentStyle::DoubleBullet
        } else if orig.starts_with("/*!") {
            CommentStyle::Exclamation
        } else if orig.starts_with("/*") {
            CommentStyle::SingleBullet
        } else if orig.starts_with("///") && orig.chars().nth(3).map_or(true, |c| c != '/') {
            CommentStyle::TripleSlash
        } else if orig.starts_with("//!") {
            CommentStyle::Doc
        } else if is_custom_comment(orig) {
            CommentStyle::Custom(custom_opener(orig))
        } else {
            CommentStyle::DoubleSlash
        }
    } else if (orig.starts_with("///") && orig.chars().nth(3).map_or(true, |c| c != '/'))
        || (orig.starts_with("/**") && !orig.starts_with("/**/"))
    {
        CommentStyle::TripleSlash
    } else if orig.starts_with("//!") || orig.starts_with("/*!") {
        CommentStyle::Doc
    } else if is_custom_comment(orig) {
        CommentStyle::Custom(custom_opener(orig))
    } else {
        CommentStyle::DoubleSlash
    }
}

/// Returns true if the last line of the passed string finishes with a block-comment.
pub(crate) fn is_last_comment_block(s: &str) -> bool {
    s.trim_end().ends_with("*/")
}

/// Combine `prev_str` and `next_str` into a single `String`. `span` may contain
/// comments between two strings. If there are such comments, then that will be
/// recovered. If `allow_extend` is true and there is no comment between the two
/// strings, then they will be put on a single line as long as doing so does not
/// exceed max width.
pub(crate) fn combine_strs_with_missing_comments(
    context: &RewriteContext<'_>,
    prev_str: &str,
    next_str: &str,
    span: Span,
    shape: Shape,
    allow_extend: bool,
) -> Option<String> {
    trace!(
        "combine_strs_with_missing_comments `{}` `{}` {:?} {:?}",
        prev_str, next_str, span, shape
    );

    let mut result =
        String::with_capacity(prev_str.len() + next_str.len() + shape.indent.width() + 128);
    result.push_str(prev_str);
    let mut allow_one_line = !prev_str.contains('\n') && !next_str.contains('\n');
    let first_sep =
        if prev_str.is_empty() || next_str.is_empty() || trimmed_last_line_width(prev_str) == 0 {
            ""
        } else {
            " "
        };
    let mut one_line_width =
        last_line_width(prev_str) + first_line_width(next_str) + first_sep.len();

    let config = context.config;
    let indent = shape.indent;
    let missing_comment = rewrite_missing_comment(span, shape, context)?;

    if missing_comment.is_empty() {
        if allow_extend && one_line_width <= shape.width {
            result.push_str(first_sep);
        } else if !prev_str.is_empty() {
            result.push_str(&indent.to_string_with_newline(config))
        }
        result.push_str(next_str);
        return Some(result);
    }

    // We have a missing comment between the first expression and the second expression.

    // Peek the original source code and find out whether there is a newline between the first
    // expression and the second expression or the missing comment. We will preserve the original
    // layout whenever possible.
    let original_snippet = context.snippet(span);
    let prefer_same_line = if let Some(pos) = original_snippet.find('/') {
        !original_snippet[..pos].contains('\n')
    } else {
        !original_snippet.contains('\n')
    };

    one_line_width -= first_sep.len();
    let first_sep = if prev_str.is_empty() || missing_comment.is_empty() {
        Cow::from("")
    } else {
        let one_line_width = last_line_width(prev_str) + first_line_width(&missing_comment) + 1;
        if prefer_same_line && one_line_width <= shape.width {
            Cow::from(" ")
        } else {
            indent.to_string_with_newline(config)
        }
    };
    result.push_str(&first_sep);
    result.push_str(&missing_comment);

    let second_sep = if missing_comment.is_empty() || next_str.is_empty() {
        Cow::from("")
    } else if missing_comment.starts_with("//") {
        indent.to_string_with_newline(config)
    } else {
        one_line_width += missing_comment.len() + first_sep.len() + 1;
        allow_one_line &= !missing_comment.starts_with("//") && !missing_comment.contains('\n');
        if prefer_same_line && allow_one_line && one_line_width <= shape.width {
            Cow::from(" ")
        } else {
            indent.to_string_with_newline(config)
        }
    };
    result.push_str(&second_sep);
    result.push_str(next_str);

    Some(result)
}

pub(crate) fn rewrite_doc_comment(orig: &str, shape: Shape, config: &Config) -> Option<String> {
    identify_comment(orig, false, shape, config, true)
}

pub(crate) fn rewrite_comment(
    orig: &str,
    block_style: bool,
    shape: Shape,
    config: &Config,
) -> Option<String> {
    identify_comment(orig, block_style, shape, config, false)
}

fn identify_comment(
    orig: &str,
    block_style: bool,
    shape: Shape,
    config: &Config,
    is_doc_comment: bool,
) -> Option<String> {
    let style = comment_style(orig, false);

    // Computes the byte length of line taking into account a newline if the line is part of a
    // paragraph.
    fn compute_len(orig: &str, line: &str) -> usize {
        if orig.len() > line.len() {
            if orig.as_bytes()[line.len()] == b'\r' {
                line.len() + 2
            } else {
                line.len() + 1
            }
        } else {
            line.len()
        }
    }

    // Get the first group of line comments having the same commenting style.
    //
    // Returns a tuple with:
    // - a boolean indicating if there is a blank line
    // - a number indicating the size of the first group of comments
    fn consume_same_line_comments(
        style: CommentStyle<'_>,
        orig: &str,
        line_start: &str,
    ) -> (bool, usize) {
        let mut first_group_ending = 0;
        let mut hbl = false;

        for line in orig.lines() {
            let trimmed_line = line.trim_start();
            if trimmed_line.is_empty() {
                hbl = true;
                break;
            } else if trimmed_line.starts_with(line_start)
                || comment_style(trimmed_line, false) == style
            {
                first_group_ending += compute_len(&orig[first_group_ending..], line);
            } else {
                break;
            }
        }
        (hbl, first_group_ending)
    }

    let (has_bare_lines, first_group_ending) = match style {
        CommentStyle::DoubleSlash | CommentStyle::TripleSlash | CommentStyle::Doc => {
            let line_start = style.line_start().trim_start();
            consume_same_line_comments(style, orig, line_start)
        }
        CommentStyle::Custom(opener) => {
            let trimmed_opener = opener.trim_end();
            consume_same_line_comments(style, orig, trimmed_opener)
        }
        // for a block comment, search for the closing symbol
        CommentStyle::DoubleBullet | CommentStyle::SingleBullet | CommentStyle::Exclamation => {
            let closer = style.closer().trim_start();
            let mut count = orig.matches(closer).count();
            let mut closing_symbol_offset = 0;
            let mut hbl = false;
            let mut first = true;
            for line in orig.lines() {
                closing_symbol_offset += compute_len(&orig[closing_symbol_offset..], line);
                let mut trimmed_line = line.trim_start();
                if !trimmed_line.starts_with('*')
                    && !trimmed_line.starts_with("//")
                    && !trimmed_line.starts_with("/*")
                {
                    hbl = true;
                }

                // Remove opener from consideration when searching for closer
                if first {
                    let opener = style.opener().trim_end();
                    trimmed_line = &trimmed_line[opener.len()..];
                    first = false;
                }
                if trimmed_line.ends_with(closer) {
                    count -= 1;
                    if count == 0 {
                        break;
                    }
                }
            }
            (hbl, closing_symbol_offset)
        }
    };

    let (first_group, rest) = orig.split_at(first_group_ending);
    let rewritten_first_group =
        if !config.normalize_comments() && has_bare_lines && style.is_block_comment() {
            trim_left_preserve_layout(first_group, shape.indent, config)?
        } else if !config.normalize_comments()
            && !config.wrap_comments()
            && !(
                // `format_code_in_doc_comments` should only take effect on doc comments,
                // so we only consider it when this comment block is a doc comment block.
                is_doc_comment && config.format_code_in_doc_comments()
            )
        {
            light_rewrite_comment(first_group, shape.indent, config, is_doc_comment)
        } else {
            rewrite_comment_inner(
                first_group,
                block_style,
                style,
                shape,
                config,
                is_doc_comment || style.is_doc_comment(),
            )?
        };
    if rest.is_empty() {
        Some(rewritten_first_group)
    } else {
        identify_comment(
            rest.trim_start(),
            block_style,
            shape,
            config,
            is_doc_comment,
        )
        .map(|rest_str| {
            format!(
                "{}\n{}{}{}",
                rewritten_first_group,
                // insert back the blank line
                if has_bare_lines && style.is_line_comment() {
                    "\n"
                } else {
                    ""
                },
                shape.indent.to_string(config),
                rest_str
            )
        })
    }
}

/// Enum indicating if the code block contains rust based on attributes
enum CodeBlockAttribute {
    Rust,
    NotRust,
}

impl CodeBlockAttribute {
    /// Parse comma separated attributes list. Return rust only if all
    /// attributes are valid rust attributes
    /// See <https://doc.rust-lang.org/rustdoc/print.html#attributes>
    fn new(attributes: &str) -> CodeBlockAttribute {
        for attribute in attributes.split(',') {
            match attribute.trim() {
                "" | "rust" | "should_panic" | "no_run" | "edition2015" | "edition2018"
                | "edition2021" => (),
                "ignore" | "compile_fail" | "text" => return CodeBlockAttribute::NotRust,
                _ => return CodeBlockAttribute::NotRust,
            }
        }
        CodeBlockAttribute::Rust
    }
}

/// Block that is formatted as an item.
///
/// An item starts with either a star `*`, a dash `-`, a greater-than `>`, a plus '+', or a number
/// `12.` or `34)` (with at most 2 digits). An item represents CommonMark's ["list
/// items"](https://spec.commonmark.org/0.30/#list-items) and/or ["block
/// quotes"](https://spec.commonmark.org/0.30/#block-quotes), but note that only a subset of
/// CommonMark is recognized - see the doc comment of [`ItemizedBlock::get_marker_length`] for more
/// details.
///
/// Different level of indentation are handled by shrinking the shape accordingly.
struct ItemizedBlock {
    /// the lines that are identified as part of an itemized block
    lines: Vec<String>,
    /// the number of characters (typically whitespaces) up to the item marker
    indent: usize,
    /// the string that marks the start of an item
    opener: String,
    /// sequence of characters (typically whitespaces) to prefix new lines that are part of the item
    line_start: String,
}

impl ItemizedBlock {
    /// Checks whether the `trimmed` line includes an item marker. Returns `None` if there is no
    /// marker. Returns the length of the marker (in bytes) if one is present. Note that the length
    /// includes the whitespace that follows the marker, for example the marker in `"* list item"`
    /// has the length of 2.
    ///
    /// This function recognizes item markers that correspond to CommonMark's
    /// ["bullet list marker"](https://spec.commonmark.org/0.30/#bullet-list-marker),
    /// ["block quote marker"](https://spec.commonmark.org/0.30/#block-quote-marker), and/or
    /// ["ordered list marker"](https://spec.commonmark.org/0.30/#ordered-list-marker).
    ///
    /// Compared to CommonMark specification, the number of digits that are allowed in an ["ordered
    /// list marker"](https://spec.commonmark.org/0.30/#ordered-list-marker) is more limited (to at
    /// most 2 digits). Limiting the length of the marker helps reduce the risk of recognizing
    /// arbitrary numbers as markers. See also
    /// <https://talk.commonmark.org/t/blank-lines-before-lists-revisited/1990> which gives the
    /// following example where a number (i.e. "1868") doesn't signify an ordered list:
    /// ```md
    /// The Captain died in
    /// 1868. He wes buried in...
    /// ```
    fn get_marker_length(trimmed: &str) -> Option<usize> {
        // https://spec.commonmark.org/0.30/#bullet-list-marker or
        // https://spec.commonmark.org/0.30/#block-quote-marker
        let itemized_start = ["* ", "- ", "> ", "+ "];
        if itemized_start.iter().any(|s| trimmed.starts_with(s)) {
            return Some(2); // All items in `itemized_start` have length 2.
        }

        // https://spec.commonmark.org/0.30/#ordered-list-marker, where at most 2 digits are
        // allowed.
        for suffix in [". ", ") "] {
            if let Some((prefix, _)) = trimmed.split_once(suffix) {
                let has_leading_digits = (1..=2).contains(&prefix.len())
                    && prefix.chars().all(|c| char::is_ascii_digit(&c));
                if has_leading_digits {
                    return Some(prefix.len() + suffix.len());
                }
            }
        }

        None // No markers found.
    }

    /// Creates a new `ItemizedBlock` described with the given `line`.
    /// Returns `None` if `line` doesn't start an item.
    fn new(line: &str) -> Option<ItemizedBlock> {
        let marker_length = ItemizedBlock::get_marker_length(line.trim_start())?;
        let space_to_marker = line.chars().take_while(|c| c.is_whitespace()).count();
        let mut indent = space_to_marker + marker_length;
        let mut line_start = " ".repeat(indent);

        // Markdown blockquote start with a "> "
        if line.trim_start().starts_with('>') {
            // remove the original +2 indent because there might be multiple nested block quotes
            // and it's easier to reason about the final indent by just taking the length
            // of the new line_start. We update the indent because it effects the max width
            // of each formatted line.
            line_start = itemized_block_quote_start(line, line_start, 2);
            indent = line_start.len();
        }
        Some(ItemizedBlock {
            lines: vec![line[indent..].to_string()],
            indent,
            opener: line[..indent].to_string(),
            line_start,
        })
    }

    /// Returns a `StringFormat` used for formatting the content of an item.
    fn create_string_format<'a>(&'a self, fmt: &'a StringFormat<'_>) -> StringFormat<'a> {
        StringFormat {
            opener: "",
            closer: "",
            line_start: "",
            line_end: "",
            shape: Shape::legacy(fmt.shape.width.saturating_sub(self.indent), Indent::empty()),
            trim_end: true,
            config: fmt.config,
        }
    }

    /// Returns `true` if the line is part of the current itemized block.
    /// If it is, then it is added to the internal lines list.
    fn add_line(&mut self, line: &str) -> bool {
        if ItemizedBlock::get_marker_length(line.trim_start()).is_none()
            && self.indent <= line.chars().take_while(|c| c.is_whitespace()).count()
        {
            self.lines.push(line.to_string());
            return true;
        }
        false
    }

    /// Returns the block as a string, with each line trimmed at the start.
    fn trimmed_block_as_string(&self) -> String {
        self.lines
            .iter()
            .map(|line| format!("{} ", line.trim_start()))
            .collect::<String>()
    }

    /// Returns the block as a string under its original form.
    fn original_block_as_string(&self) -> String {
        self.lines.join("\n")
    }
}

/// Determine the line_start when formatting markdown block quotes.
/// The original line_start likely contains indentation (whitespaces), which we'd like to
/// replace with '> ' characters.
fn itemized_block_quote_start(line: &str, mut line_start: String, remove_indent: usize) -> String {
    let quote_level = line
        .chars()
        .take_while(|c| !c.is_alphanumeric())
        .fold(0, |acc, c| if c == '>' { acc + 1 } else { acc });

    for _ in 0..remove_indent {
        line_start.pop();
    }

    for _ in 0..quote_level {
        line_start.push_str("> ")
    }
    line_start
}

struct CommentRewrite<'a> {
    result: String,
    code_block_buffer: String,
    is_prev_line_multi_line: bool,
    code_block_attr: Option<CodeBlockAttribute>,
    item_block: Option<ItemizedBlock>,
    comment_line_separator: String,
    indent_str: String,
    max_width: usize,
    fmt_indent: Indent,
    fmt: StringFormat<'a>,

    opener: String,
    closer: String,
    line_start: String,
    style: CommentStyle<'a>,
}

impl<'a> CommentRewrite<'a> {
    fn new(
        orig: &'a str,
        block_style: bool,
        shape: Shape,
        config: &'a Config,
    ) -> CommentRewrite<'a> {
        let ((opener, closer, line_start), style) = if block_style {
            (
                CommentStyle::SingleBullet.to_str_tuplet(),
                CommentStyle::SingleBullet,
            )
        } else {
            let style = comment_style(orig, config.normalize_comments());
            (style.to_str_tuplet(), style)
        };

        let max_width = shape
            .width
            .checked_sub(closer.len() + opener.len())
            .unwrap_or(1);
        let indent_str = shape.indent.to_string_with_newline(config).to_string();

        let mut cr = CommentRewrite {
            result: String::with_capacity(orig.len() * 2),
            code_block_buffer: String::with_capacity(128),
            is_prev_line_multi_line: false,
            code_block_attr: None,
            item_block: None,
            comment_line_separator: format!("{indent_str}{line_start}"),
            max_width,
            indent_str,
            fmt_indent: shape.indent,

            fmt: StringFormat {
                opener: "",
                closer: "",
                line_start,
                line_end: "",
                shape: Shape::legacy(max_width, shape.indent),
                trim_end: true,
                config,
            },

            opener: opener.to_owned(),
            closer: closer.to_owned(),
            line_start: line_start.to_owned(),
            style,
        };
        cr.result.push_str(opener);
        cr
    }

    fn join_block(s: &str, sep: &str) -> String {
        let mut result = String::with_capacity(s.len() + 128);
        let mut iter = s.lines().peekable();
        while let Some(line) = iter.next() {
            result.push_str(line);
            result.push_str(match iter.peek() {
                Some(&"") => sep.trim_end(),
                Some(..) => sep,
                None => "",
            });
        }
        result
    }

    /// Check if any characters were written to the result buffer after the start of the comment.
    /// when calling [`CommentRewrite::new()`] the result buffer is initialized with the opening
    /// characters for the comment.
    fn buffer_contains_comment(&self) -> bool {
        // if self.result.len() < self.opener.len() then an empty comment is in the buffer
        // if self.result.len() > self.opener.len() then a non empty comment is in the buffer
        self.result.len() != self.opener.len()
    }

    fn finish(mut self) -> String {
        if !self.code_block_buffer.is_empty() {
            // There is a code block that is not properly enclosed by backticks.
            // We will leave them untouched.
            self.result.push_str(&self.comment_line_separator);
            self.result.push_str(&Self::join_block(
                &trim_custom_comment_prefix(&self.code_block_buffer),
                &self.comment_line_separator,
            ));
        }

        if let Some(ref ib) = self.item_block {
            // the last few lines are part of an itemized block
            self.fmt.shape = Shape::legacy(self.max_width, self.fmt_indent);
            let item_fmt = ib.create_string_format(&self.fmt);

            // only push a comment_line_separator for ItemizedBlocks if the comment is not empty
            if self.buffer_contains_comment() {
                self.result.push_str(&self.comment_line_separator);
            }

            self.result.push_str(&ib.opener);
            match rewrite_string(
                &ib.trimmed_block_as_string(),
                &item_fmt,
                self.max_width.saturating_sub(ib.indent),
            ) {
                Some(s) => self.result.push_str(&Self::join_block(
                    &s,
                    &format!("{}{}", self.comment_line_separator, ib.line_start),
                )),
                None => self.result.push_str(&Self::join_block(
                    &ib.original_block_as_string(),
                    &self.comment_line_separator,
                )),
            };
        }

        self.result.push_str(&self.closer);
        if self.result.ends_with(&self.opener) && self.opener.ends_with(' ') {
            // Trailing space.
            self.result.pop();
        }

        self.result
    }

    fn handle_line(
        &mut self,
        orig: &'a str,
        i: usize,
        line: &'a str,
        has_leading_whitespace: bool,
        is_doc_comment: bool,
    ) -> bool {
        let num_newlines = count_newlines(orig);
        let is_last = i == num_newlines;
        let needs_new_comment_line = if self.style.is_block_comment() {
            num_newlines > 0 || self.buffer_contains_comment()
        } else {
            self.buffer_contains_comment()
        };

        if let Some(ref mut ib) = self.item_block {
            if ib.add_line(line) {
                return false;
            }
            self.is_prev_line_multi_line = false;
            self.fmt.shape = Shape::legacy(self.max_width, self.fmt_indent);
            let item_fmt = ib.create_string_format(&self.fmt);

            // only push a comment_line_separator if we need to start a new comment line
            if needs_new_comment_line {
                self.result.push_str(&self.comment_line_separator);
            }

            self.result.push_str(&ib.opener);
            match rewrite_string(
                &ib.trimmed_block_as_string(),
                &item_fmt,
                self.max_width.saturating_sub(ib.indent),
            ) {
                Some(s) => self.result.push_str(&Self::join_block(
                    &s,
                    &format!("{}{}", self.comment_line_separator, ib.line_start),
                )),
                None => self.result.push_str(&Self::join_block(
                    &ib.original_block_as_string(),
                    &self.comment_line_separator,
                )),
            };
        } else if self.code_block_attr.is_some() {
            if line.starts_with("```") {
                let code_block = match self.code_block_attr.as_ref().unwrap() {
                    CodeBlockAttribute::Rust
                        if self.fmt.config.format_code_in_doc_comments()
                            && !self.code_block_buffer.trim().is_empty() =>
                    {
                        let mut config = self.fmt.config.clone();
                        config.set().wrap_comments(false);
                        let comment_max_width = config
                            .doc_comment_code_block_width()
                            .min(config.max_width());
                        config.set().max_width(comment_max_width);
                        if let Some(s) =
                            crate::format_code_block(&self.code_block_buffer, &config, false)
                        {
                            trim_custom_comment_prefix(&s.snippet)
                        } else {
                            trim_custom_comment_prefix(&self.code_block_buffer)
                        }
                    }
                    _ => trim_custom_comment_prefix(&self.code_block_buffer),
                };
                if !code_block.is_empty() {
                    self.result.push_str(&self.comment_line_separator);
                    self.result
                        .push_str(&Self::join_block(&code_block, &self.comment_line_separator));
                }
                self.code_block_buffer.clear();
                self.result.push_str(&self.comment_line_separator);
                self.result.push_str(line);
                self.code_block_attr = None;
            } else {
                self.code_block_buffer
                    .push_str(&hide_sharp_behind_comment(line));
                self.code_block_buffer.push('\n');
            }
            return false;
        }

        self.code_block_attr = None;
        self.item_block = None;
        if let Some(stripped) = line.strip_prefix("```") {
            self.code_block_attr = Some(CodeBlockAttribute::new(stripped))
        } else if self.fmt.config.wrap_comments() {
            if let Some(ib) = ItemizedBlock::new(line) {
                self.item_block = Some(ib);
                return false;
            }
        }

        if self.result == self.opener {
            let force_leading_whitespace = &self.opener == "/* " && count_newlines(orig) == 0;
            if !has_leading_whitespace && !force_leading_whitespace && self.result.ends_with(' ') {
                self.result.pop();
            }
            if line.is_empty() {
                return false;
            }
        } else if self.is_prev_line_multi_line && !line.is_empty() {
            self.result.push(' ')
        } else if is_last && line.is_empty() {
            // trailing blank lines are unwanted
            if !self.closer.is_empty() {
                self.result.push_str(&self.indent_str);
            }
            return true;
        } else {
            self.result.push_str(&self.comment_line_separator);
            if !has_leading_whitespace && self.result.ends_with(' ') {
                self.result.pop();
            }
        }

        let is_markdown_header_doc_comment = is_doc_comment && line.starts_with('#');

        // We only want to wrap the comment if:
        // 1) wrap_comments = true is configured
        // 2) The comment is not the start of a markdown header doc comment
        // 3) The comment width exceeds the shape's width
        // 4) No URLS were found in the comment
        // If this changes, the documentation in ../Configurations.md#wrap_comments
        // should be changed accordingly.
        let should_wrap_comment = self.fmt.config.wrap_comments()
            && !is_markdown_header_doc_comment
            && unicode_str_width(line) > self.fmt.shape.width
            && !has_url(line)
            && !is_table_item(line);

        if should_wrap_comment {
            match rewrite_string(line, &self.fmt, self.max_width) {
                Some(ref s) => {
                    self.is_prev_line_multi_line = s.contains('\n');
                    self.result.push_str(s);
                }
                None if self.is_prev_line_multi_line => {
                    // We failed to put the current `line` next to the previous `line`.
                    // Remove the trailing space, then start rewrite on the next line.
                    self.result.pop();
                    self.result.push_str(&self.comment_line_separator);
                    self.fmt.shape = Shape::legacy(self.max_width, self.fmt_indent);
                    match rewrite_string(line, &self.fmt, self.max_width) {
                        Some(ref s) => {
                            self.is_prev_line_multi_line = s.contains('\n');
                            self.result.push_str(s);
                        }
                        None => {
                            self.is_prev_line_multi_line = false;
                            self.result.push_str(line);
                        }
                    }
                }
                None => {
                    self.is_prev_line_multi_line = false;
                    self.result.push_str(line);
                }
            }

            self.fmt.shape = if self.is_prev_line_multi_line {
                // 1 = " "
                let offset = 1 + last_line_width(&self.result) - self.line_start.len();
                Shape {
                    width: self.max_width.saturating_sub(offset),
                    indent: self.fmt_indent,
                    offset: self.fmt.shape.offset + offset,
                }
            } else {
                Shape::legacy(self.max_width, self.fmt_indent)
            };
        } else {
            if line.is_empty() && self.result.ends_with(' ') && !is_last {
                // Remove space if this is an empty comment or a doc comment.
                self.result.pop();
            }
            self.result.push_str(line);
            self.fmt.shape = Shape::legacy(self.max_width, self.fmt_indent);
            self.is_prev_line_multi_line = false;
        }

        false
    }
}

fn rewrite_comment_inner(
    orig: &str,
    block_style: bool,
    style: CommentStyle<'_>,
    shape: Shape,
    config: &Config,
    is_doc_comment: bool,
) -> Option<String> {
    let mut rewriter = CommentRewrite::new(orig, block_style, shape, config);

    let line_breaks = count_newlines(orig.trim_end());
    let lines = orig
        .lines()
        .enumerate()
        .map(|(i, mut line)| {
            line = trim_end_unless_two_whitespaces(line.trim_start(), is_doc_comment);
            // Drop old closer.
            if i == line_breaks && line.ends_with("*/") && !line.starts_with("//") {
                line = line[..(line.len() - 2)].trim_end();
            }

            line
        })
        .map(|s| left_trim_comment_line(s, &style))
        .map(|(line, has_leading_whitespace)| {
            if orig.starts_with("/*") && line_breaks == 0 {
                (
                    line.trim_start(),
                    has_leading_whitespace || config.normalize_comments(),
                )
            } else {
                (line, has_leading_whitespace || config.normalize_comments())
            }
        });

    for (i, (line, has_leading_whitespace)) in lines.enumerate() {
        if rewriter.handle_line(orig, i, line, has_leading_whitespace, is_doc_comment) {
            break;
        }
    }

    Some(rewriter.finish())
}

const RUSTFMT_CUSTOM_COMMENT_PREFIX: &str = "//#### ";

fn hide_sharp_behind_comment(s: &str) -> Cow<'_, str> {
    let s_trimmed = s.trim();
    if s_trimmed.starts_with("# ") || s_trimmed == "#" {
        Cow::from(format!("{RUSTFMT_CUSTOM_COMMENT_PREFIX}{s}"))
    } else {
        Cow::from(s)
    }
}

fn trim_custom_comment_prefix(s: &str) -> String {
    s.lines()
        .map(|line| {
            let left_trimmed = line.trim_start();
            if left_trimmed.starts_with(RUSTFMT_CUSTOM_COMMENT_PREFIX) {
                left_trimmed.trim_start_matches(RUSTFMT_CUSTOM_COMMENT_PREFIX)
            } else {
                line
            }
        })
        .collect::<Vec<_>>()
        .join("\n")
}

/// Returns `true` if the given string MAY include URLs or alike.
fn has_url(s: &str) -> bool {
    // A regex matching reference doc links.
    //
    // ```markdown
    // /// An [example].
    // ///
    // /// [example]: this::is::a::link
    // ```
    let reference_link_url = static_regex!(r"^\[.+\]\s?:");

    // This function may return false positive, but should get its job done in most cases.
    s.contains("https://")
        || s.contains("http://")
        || s.contains("ftp://")
        || s.contains("file://")
        || reference_link_url.is_match(s)
}

/// Returns true if the given string may be part of a Markdown table.
fn is_table_item(mut s: &str) -> bool {
    // This function may return false positive, but should get its job done in most cases (i.e.
    // markdown tables with two column delimiters).
    s = s.trim_start();
    return s.starts_with('|')
        && match s.rfind('|') {
            Some(0) | None => false,
            _ => true,
        };
}

/// Given the span, rewrite the missing comment inside it if available.
/// Note that the given span must only include comments (or leading/trailing whitespaces).
pub(crate) fn rewrite_missing_comment(
    span: Span,
    shape: Shape,
    context: &RewriteContext<'_>,
) -> Option<String> {
    let missing_snippet = context.snippet(span);
    let trimmed_snippet = missing_snippet.trim();
    // check the span starts with a comment
    let pos = trimmed_snippet.find('/');
    if !trimmed_snippet.is_empty() && pos.is_some() {
        rewrite_comment(trimmed_snippet, false, shape, context.config)
    } else {
        Some(String::new())
    }
}

/// Recover the missing comments in the specified span, if available.
/// The layout of the comments will be preserved as long as it does not break the code
/// and its total width does not exceed the max width.
pub(crate) fn recover_missing_comment_in_span(
    span: Span,
    shape: Shape,
    context: &RewriteContext<'_>,
    used_width: usize,
) -> Option<String> {
    let missing_comment = rewrite_missing_comment(span, shape, context)?;
    if missing_comment.is_empty() {
        Some(String::new())
    } else {
        let missing_snippet = context.snippet(span);
        let pos = missing_snippet.find('/')?;
        // 1 = ` `
        let total_width = missing_comment.len() + used_width + 1;
        let force_new_line_before_comment =
            missing_snippet[..pos].contains('\n') || total_width > context.config.max_width();
        let sep = if force_new_line_before_comment {
            shape.indent.to_string_with_newline(context.config)
        } else {
            Cow::from(" ")
        };
        Some(format!("{sep}{missing_comment}"))
    }
}

/// Trim trailing whitespaces unless they consist of two or more whitespaces.
fn trim_end_unless_two_whitespaces(s: &str, is_doc_comment: bool) -> &str {
    if is_doc_comment && s.ends_with("  ") {
        s
    } else {
        s.trim_end()
    }
}

/// Trims whitespace and aligns to indent, but otherwise does not change comments.
fn light_rewrite_comment(
    orig: &str,
    offset: Indent,
    config: &Config,
    is_doc_comment: bool,
) -> String {
    let lines: Vec<&str> = orig
        .lines()
        .map(|l| {
            // This is basically just l.trim(), but in the case that a line starts
            // with `*` we want to leave one space before it, so it aligns with the
            // `*` in `/*`.
            let first_non_whitespace = l.find(|c| !char::is_whitespace(c));
            let left_trimmed = if let Some(fnw) = first_non_whitespace {
                if l.as_bytes()[fnw] == b'*' && fnw > 0 {
                    &l[fnw - 1..]
                } else {
                    &l[fnw..]
                }
            } else {
                ""
            };
            // Preserve markdown's double-space line break syntax in doc comment.
            trim_end_unless_two_whitespaces(left_trimmed, is_doc_comment)
        })
        .collect();
    lines.join(&format!("\n{}", offset.to_string(config)))
}

/// Trims comment characters and possibly a single space from the left of a string.
/// Does not trim all whitespace. If a single space is trimmed from the left of the string,
/// this function returns true.
fn left_trim_comment_line<'a>(line: &'a str, style: &CommentStyle<'_>) -> (&'a str, bool) {
    if line.starts_with("//! ")
        || line.starts_with("/// ")
        || line.starts_with("/*! ")
        || line.starts_with("/** ")
    {
        (&line[4..], true)
    } else if let CommentStyle::Custom(opener) = *style {
        if let Some(stripped) = line.strip_prefix(opener) {
            (stripped, true)
        } else {
            (&line[opener.trim_end().len()..], false)
        }
    } else if line.starts_with("/* ")
        || line.starts_with("// ")
        || line.starts_with("//!")
        || line.starts_with("///")
        || line.starts_with("** ")
        || line.starts_with("/*!")
        || (line.starts_with("/**") && !line.starts_with("/**/"))
    {
        (&line[3..], line.chars().nth(2).unwrap() == ' ')
    } else if line.starts_with("/*")
        || line.starts_with("* ")
        || line.starts_with("//")
        || line.starts_with("**")
    {
        (&line[2..], line.chars().nth(1).unwrap() == ' ')
    } else if let Some(stripped) = line.strip_prefix('*') {
        (stripped, false)
    } else {
        (line, line.starts_with(' '))
    }
}

pub(crate) trait FindUncommented {
    fn find_uncommented(&self, pat: &str) -> Option<usize>;
    fn find_last_uncommented(&self, pat: &str) -> Option<usize>;
}

impl FindUncommented for str {
    fn find_uncommented(&self, pat: &str) -> Option<usize> {
        let mut needle_iter = pat.chars();
        for (kind, (i, b)) in CharClasses::new(self.char_indices()) {
            match needle_iter.next() {
                None => {
                    return Some(i - pat.len());
                }
                Some(c) => match kind {
                    FullCodeCharKind::Normal | FullCodeCharKind::InString if b == c => {}
                    _ => {
                        needle_iter = pat.chars();
                    }
                },
            }
        }

        // Handle case where the pattern is a suffix of the search string
        match needle_iter.next() {
            Some(_) => None,
            None => Some(self.len() - pat.len()),
        }
    }

    fn find_last_uncommented(&self, pat: &str) -> Option<usize> {
        if let Some(left) = self.find_uncommented(pat) {
            let mut result = left;
            // add 1 to use find_last_uncommented for &str after pat
            while let Some(next) = self[(result + 1)..].find_last_uncommented(pat) {
                result += next + 1;
            }
            Some(result)
        } else {
            None
        }
    }
}

// Returns the first byte position after the first comment. The given string
// is expected to be prefixed by a comment, including delimiters.
// Good: `/* /* inner */ outer */ code();`
// Bad:  `code(); // hello\n world!`
pub(crate) fn find_comment_end(s: &str) -> Option<usize> {
    let mut iter = CharClasses::new(s.char_indices());
    for (kind, (i, _c)) in &mut iter {
        if kind == FullCodeCharKind::Normal || kind == FullCodeCharKind::InString {
            return Some(i);
        }
    }

    // Handle case where the comment ends at the end of `s`.
    if iter.status == CharClassesStatus::Normal {
        Some(s.len())
    } else {
        None
    }
}

/// Returns `true` if text contains any comment.
pub(crate) fn contains_comment(text: &str) -> bool {
    CharClasses::new(text.chars()).any(|(kind, _)| kind.is_comment())
}

pub(crate) struct CharClasses<T>
where
    T: Iterator,
    T::Item: RichChar,
{
    base: MultiPeek<T>,
    status: CharClassesStatus,
}

pub(crate) trait RichChar {
    fn get_char(&self) -> char;
}

impl RichChar for char {
    fn get_char(&self) -> char {
        *self
    }
}

impl RichChar for (usize, char) {
    fn get_char(&self) -> char {
        self.1
    }
}

#[derive(PartialEq, Eq, Debug, Clone, Copy)]
enum CharClassesStatus {
    Normal,
    /// Character is within a string
    LitString,
    LitStringEscape,
    /// Character is within a raw string
    LitRawString(u32),
    RawStringPrefix(u32),
    RawStringSuffix(u32),
    LitChar,
    LitCharEscape,
    /// Character inside a block comment, with the integer indicating the nesting deepness of the
    /// comment
    BlockComment(u32),
    /// Character inside a block-commented string, with the integer indicating the nesting deepness
    /// of the comment
    StringInBlockComment(u32),
    /// Status when the '/' has been consumed, but not yet the '*', deepness is
    /// the new deepness (after the comment opening).
    BlockCommentOpening(u32),
    /// Status when the '*' has been consumed, but not yet the '/', deepness is
    /// the new deepness (after the comment closing).
    BlockCommentClosing(u32),
    /// Character is within a line comment
    LineComment,
}

/// Distinguish between functional part of code and comments
#[derive(PartialEq, Eq, Debug, Clone, Copy)]
pub(crate) enum CodeCharKind {
    Normal,
    Comment,
}

/// Distinguish between functional part of code and comments,
/// describing opening and closing of comments for ease when chunking
/// code from tagged characters
#[derive(PartialEq, Eq, Debug, Clone, Copy)]
pub(crate) enum FullCodeCharKind {
    Normal,
    /// The first character of a comment, there is only one for a comment (always '/')
    StartComment,
    /// Any character inside a comment including the second character of comment
    /// marks ("//", "/*")
    InComment,
    /// Last character of a comment, '\n' for a line comment, '/' for a block comment.
    EndComment,
    /// Start of a multiline string inside a comment
    StartStringCommented,
    /// End of a multiline string inside a comment
    EndStringCommented,
    /// Inside a commented string
    InStringCommented,
    /// Start of a multiline string
    StartString,
    /// End of a multiline string
    EndString,
    /// Inside a string.
    InString,
}

impl FullCodeCharKind {
    pub(crate) fn is_comment(self) -> bool {
        match self {
            FullCodeCharKind::StartComment
            | FullCodeCharKind::InComment
            | FullCodeCharKind::EndComment
            | FullCodeCharKind::StartStringCommented
            | FullCodeCharKind::InStringCommented
            | FullCodeCharKind::EndStringCommented => true,
            _ => false,
        }
    }

    /// Returns true if the character is inside a comment
    pub(crate) fn inside_comment(self) -> bool {
        match self {
            FullCodeCharKind::InComment
            | FullCodeCharKind::StartStringCommented
            | FullCodeCharKind::InStringCommented
            | FullCodeCharKind::EndStringCommented => true,
            _ => false,
        }
    }

    pub(crate) fn is_string(self) -> bool {
        self == FullCodeCharKind::InString || self == FullCodeCharKind::StartString
    }

    /// Returns true if the character is within a commented string
    pub(crate) fn is_commented_string(self) -> bool {
        self == FullCodeCharKind::InStringCommented
            || self == FullCodeCharKind::StartStringCommented
    }

    fn to_codecharkind(self) -> CodeCharKind {
        if self.is_comment() {
            CodeCharKind::Comment
        } else {
            CodeCharKind::Normal
        }
    }
}

impl<T> CharClasses<T>
where
    T: Iterator,
    T::Item: RichChar,
{
    pub(crate) fn new(base: T) -> CharClasses<T> {
        CharClasses {
            base: multipeek(base),
            status: CharClassesStatus::Normal,
        }
    }
}

fn is_raw_string_suffix<T>(iter: &mut MultiPeek<T>, count: u32) -> bool
where
    T: Iterator,
    T::Item: RichChar,
{
    for _ in 0..count {
        match iter.peek() {
            Some(c) if c.get_char() == '#' => continue,
            _ => return false,
        }
    }
    true
}

impl<T> Iterator for CharClasses<T>
where
    T: Iterator,
    T::Item: RichChar,
{
    type Item = (FullCodeCharKind, T::Item);

    fn next(&mut self) -> Option<(FullCodeCharKind, T::Item)> {
        let item = self.base.next()?;
        let chr = item.get_char();
        let mut char_kind = FullCodeCharKind::Normal;
        self.status = match self.status {
            CharClassesStatus::LitRawString(sharps) => {
                char_kind = FullCodeCharKind::InString;
                match chr {
                    '"' => {
                        if sharps == 0 {
                            char_kind = FullCodeCharKind::Normal;
                            CharClassesStatus::Normal
                        } else if is_raw_string_suffix(&mut self.base, sharps) {
                            CharClassesStatus::RawStringSuffix(sharps)
                        } else {
                            CharClassesStatus::LitRawString(sharps)
                        }
                    }
                    _ => CharClassesStatus::LitRawString(sharps),
                }
            }
            CharClassesStatus::RawStringPrefix(sharps) => {
                char_kind = FullCodeCharKind::InString;
                match chr {
                    '#' => CharClassesStatus::RawStringPrefix(sharps + 1),
                    '"' => CharClassesStatus::LitRawString(sharps),
                    _ => CharClassesStatus::Normal, // Unreachable.
                }
            }
            CharClassesStatus::RawStringSuffix(sharps) => {
                match chr {
                    '#' => {
                        if sharps == 1 {
                            CharClassesStatus::Normal
                        } else {
                            char_kind = FullCodeCharKind::InString;
                            CharClassesStatus::RawStringSuffix(sharps - 1)
                        }
                    }
                    _ => CharClassesStatus::Normal, // Unreachable
                }
            }
            CharClassesStatus::LitString => {
                char_kind = FullCodeCharKind::InString;
                match chr {
                    '"' => CharClassesStatus::Normal,
                    '\\' => CharClassesStatus::LitStringEscape,
                    _ => CharClassesStatus::LitString,
                }
            }
            CharClassesStatus::LitStringEscape => {
                char_kind = FullCodeCharKind::InString;
                CharClassesStatus::LitString
            }
            CharClassesStatus::LitChar => match chr {
                '\\' => CharClassesStatus::LitCharEscape,
                '\'' => CharClassesStatus::Normal,
                _ => CharClassesStatus::LitChar,
            },
            CharClassesStatus::LitCharEscape => CharClassesStatus::LitChar,
            CharClassesStatus::Normal => match chr {
                'r' => match self.base.peek().map(RichChar::get_char) {
                    Some('#') | Some('"') => {
                        char_kind = FullCodeCharKind::InString;
                        CharClassesStatus::RawStringPrefix(0)
                    }
                    _ => CharClassesStatus::Normal,
                },
                '"' => {
                    char_kind = FullCodeCharKind::InString;
                    CharClassesStatus::LitString
                }
                '\'' => {
                    // HACK: Work around mut borrow.
                    match self.base.peek() {
                        Some(next) if next.get_char() == '\\' => {
                            self.status = CharClassesStatus::LitChar;
                            return Some((char_kind, item));
                        }
                        _ => (),
                    }

                    match self.base.peek() {
                        Some(next) if next.get_char() == '\'' => CharClassesStatus::LitChar,
                        _ => CharClassesStatus::Normal,
                    }
                }
                '/' => match self.base.peek() {
                    Some(next) if next.get_char() == '*' => {
                        self.status = CharClassesStatus::BlockCommentOpening(1);
                        return Some((FullCodeCharKind::StartComment, item));
                    }
                    Some(next) if next.get_char() == '/' => {
                        self.status = CharClassesStatus::LineComment;
                        return Some((FullCodeCharKind::StartComment, item));
                    }
                    _ => CharClassesStatus::Normal,
                },
                _ => CharClassesStatus::Normal,
            },
            CharClassesStatus::StringInBlockComment(deepness) => {
                char_kind = FullCodeCharKind::InStringCommented;
                if chr == '"' {
                    CharClassesStatus::BlockComment(deepness)
                } else if chr == '*' && self.base.peek().map(RichChar::get_char) == Some('/') {
                    char_kind = FullCodeCharKind::InComment;
                    CharClassesStatus::BlockCommentClosing(deepness - 1)
                } else {
                    CharClassesStatus::StringInBlockComment(deepness)
                }
            }
            CharClassesStatus::BlockComment(deepness) => {
                assert_ne!(deepness, 0);
                char_kind = FullCodeCharKind::InComment;
                match self.base.peek() {
                    Some(next) if next.get_char() == '/' && chr == '*' => {
                        CharClassesStatus::BlockCommentClosing(deepness - 1)
                    }
                    Some(next) if next.get_char() == '*' && chr == '/' => {
                        CharClassesStatus::BlockCommentOpening(deepness + 1)
                    }
                    _ if chr == '"' => CharClassesStatus::StringInBlockComment(deepness),
                    _ => self.status,
                }
            }
            CharClassesStatus::BlockCommentOpening(deepness) => {
                assert_eq!(chr, '*');
                self.status = CharClassesStatus::BlockComment(deepness);
                return Some((FullCodeCharKind::InComment, item));
            }
            CharClassesStatus::BlockCommentClosing(deepness) => {
                assert_eq!(chr, '/');
                if deepness == 0 {
                    self.status = CharClassesStatus::Normal;
                    return Some((FullCodeCharKind::EndComment, item));
                } else {
                    self.status = CharClassesStatus::BlockComment(deepness);
                    return Some((FullCodeCharKind::InComment, item));
                }
            }
            CharClassesStatus::LineComment => match chr {
                '\n' => {
                    self.status = CharClassesStatus::Normal;
                    return Some((FullCodeCharKind::EndComment, item));
                }
                _ => {
                    self.status = CharClassesStatus::LineComment;
                    return Some((FullCodeCharKind::InComment, item));
                }
            },
        };
        Some((char_kind, item))
    }
}

/// An iterator over the lines of a string, paired with the char kind at the
/// end of the line.
pub(crate) struct LineClasses<'a> {
    base: iter::Peekable<CharClasses<std::str::Chars<'a>>>,
    kind: FullCodeCharKind,
}

impl<'a> LineClasses<'a> {
    pub(crate) fn new(s: &'a str) -> Self {
        LineClasses {
            base: CharClasses::new(s.chars()).peekable(),
            kind: FullCodeCharKind::Normal,
        }
    }
}

impl<'a> Iterator for LineClasses<'a> {
    type Item = (FullCodeCharKind, String);

    fn next(&mut self) -> Option<Self::Item> {
        self.base.peek()?;

        let mut line = String::new();

        let start_kind = match self.base.peek() {
            Some((kind, _)) => *kind,
            None => unreachable!(),
        };

        for (kind, c) in self.base.by_ref() {
            // needed to set the kind of the ending character on the last line
            self.kind = kind;
            if c == '\n' {
                self.kind = match (start_kind, kind) {
                    (FullCodeCharKind::Normal, FullCodeCharKind::InString) => {
                        FullCodeCharKind::StartString
                    }
                    (FullCodeCharKind::InString, FullCodeCharKind::Normal) => {
                        FullCodeCharKind::EndString
                    }
                    (FullCodeCharKind::InComment, FullCodeCharKind::InStringCommented) => {
                        FullCodeCharKind::StartStringCommented
                    }
                    (FullCodeCharKind::InStringCommented, FullCodeCharKind::InComment) => {
                        FullCodeCharKind::EndStringCommented
                    }
                    _ => kind,
                };
                break;
            }
            line.push(c);
        }

        // Workaround for CRLF newline.
        if line.ends_with('\r') {
            line.pop();
        }

        Some((self.kind, line))
    }
}

/// Iterator over functional and commented parts of a string. Any part of a string is either
/// functional code, either *one* block comment, either *one* line comment. Whitespace between
/// comments is functional code. Line comments contain their ending newlines.
struct UngroupedCommentCodeSlices<'a> {
    slice: &'a str,
    iter: iter::Peekable<CharClasses<std::str::CharIndices<'a>>>,
}

impl<'a> UngroupedCommentCodeSlices<'a> {
    fn new(code: &'a str) -> UngroupedCommentCodeSlices<'a> {
        UngroupedCommentCodeSlices {
            slice: code,
            iter: CharClasses::new(code.char_indices()).peekable(),
        }
    }
}

impl<'a> Iterator for UngroupedCommentCodeSlices<'a> {
    type Item = (CodeCharKind, usize, &'a str);

    fn next(&mut self) -> Option<Self::Item> {
        let (kind, (start_idx, _)) = self.iter.next()?;
        match kind {
            FullCodeCharKind::Normal | FullCodeCharKind::InString => {
                // Consume all the Normal code
                while let Some(&(char_kind, _)) = self.iter.peek() {
                    if char_kind.is_comment() {
                        break;
                    }
                    let _ = self.iter.next();
                }
            }
            FullCodeCharKind::StartComment => {
                // Consume the whole comment
                loop {
                    match self.iter.next() {
                        Some((kind, ..)) if kind.inside_comment() => continue,
                        _ => break,
                    }
                }
            }
            _ => panic!(),
        }
        let slice = match self.iter.peek() {
            Some(&(_, (end_idx, _))) => &self.slice[start_idx..end_idx],
            None => &self.slice[start_idx..],
        };
        Some((
            if kind.is_comment() {
                CodeCharKind::Comment
            } else {
                CodeCharKind::Normal
            },
            start_idx,
            slice,
        ))
    }
}

/// Iterator over an alternating sequence of functional and commented parts of
/// a string. The first item is always a, possibly zero length, subslice of
/// functional text. Line style comments contain their ending newlines.
pub(crate) struct CommentCodeSlices<'a> {
    slice: &'a str,
    last_slice_kind: CodeCharKind,
    last_slice_end: usize,
}

impl<'a> CommentCodeSlices<'a> {
    pub(crate) fn new(slice: &'a str) -> CommentCodeSlices<'a> {
        CommentCodeSlices {
            slice,
            last_slice_kind: CodeCharKind::Comment,
            last_slice_end: 0,
        }
    }
}

impl<'a> Iterator for CommentCodeSlices<'a> {
    type Item = (CodeCharKind, usize, &'a str);

    fn next(&mut self) -> Option<Self::Item> {
        if self.last_slice_end == self.slice.len() {
            return None;
        }

        let mut sub_slice_end = self.last_slice_end;
        let mut first_whitespace = None;
        let subslice = &self.slice[self.last_slice_end..];
        let mut iter = CharClasses::new(subslice.char_indices());

        for (kind, (i, c)) in &mut iter {
            let is_comment_connector = self.last_slice_kind == CodeCharKind::Normal
                && &subslice[..2] == "//"
                && [' ', '\t'].contains(&c);

            if is_comment_connector && first_whitespace.is_none() {
                first_whitespace = Some(i);
            }

            if kind.to_codecharkind() == self.last_slice_kind && !is_comment_connector {
                let last_index = match first_whitespace {
                    Some(j) => j,
                    None => i,
                };
                sub_slice_end = self.last_slice_end + last_index;
                break;
            }

            if !is_comment_connector {
                first_whitespace = None;
            }
        }

        if let (None, true) = (iter.next(), sub_slice_end == self.last_slice_end) {
            // This was the last subslice.
            sub_slice_end = match first_whitespace {
                Some(i) => self.last_slice_end + i,
                None => self.slice.len(),
            };
        }

        let kind = match self.last_slice_kind {
            CodeCharKind::Comment => CodeCharKind::Normal,
            CodeCharKind::Normal => CodeCharKind::Comment,
        };
        let res = (
            kind,
            self.last_slice_end,
            &self.slice[self.last_slice_end..sub_slice_end],
        );
        self.last_slice_end = sub_slice_end;
        self.last_slice_kind = kind;

        Some(res)
    }
}

/// Checks is `new` didn't miss any comment from `span`, if it removed any, return previous text
/// (if it fits in the width/offset, else return `None`), else return `new`
pub(crate) fn recover_comment_removed(
    new: String,
    span: Span,
    context: &RewriteContext<'_>,
) -> Option<String> {
    let snippet = context.snippet(span);
    if snippet != new && changed_comment_content(snippet, &new) {
        // We missed some comments. Warn and keep the original text.
        if context.config.error_on_unformatted() {
            context.report.append(
                context.psess.span_to_filename(span),
                vec![FormattingError::from_span(
                    span,
                    context.psess,
                    ErrorKind::LostComment,
                )],
            );
        }
        Some(snippet.to_owned())
    } else {
        Some(new)
    }
}

pub(crate) fn filter_normal_code(code: &str) -> String {
    let mut buffer = String::with_capacity(code.len());
    LineClasses::new(code).for_each(|(kind, line)| match kind {
        FullCodeCharKind::Normal
        | FullCodeCharKind::StartString
        | FullCodeCharKind::InString
        | FullCodeCharKind::EndString => {
            buffer.push_str(&line);
            buffer.push('\n');
        }
        _ => (),
    });
    if !code.ends_with('\n') && buffer.ends_with('\n') {
        buffer.pop();
    }
    buffer
}

/// Returns `true` if the two strings of code have the same payload of comments.
/// The payload of comments is everything in the string except:
/// - actual code (not comments),
/// - comment start/end marks,
/// - whitespace,
/// - '*' at the beginning of lines in block comments.
fn changed_comment_content(orig: &str, new: &str) -> bool {
    // Cannot write this as a fn since we cannot return types containing closures.
    let code_comment_content = |code| {
        let slices = UngroupedCommentCodeSlices::new(code);
        slices
            .filter(|(kind, _, _)| *kind == CodeCharKind::Comment)
            .flat_map(|(_, _, s)| CommentReducer::new(s))
    };
    let res = code_comment_content(orig).ne(code_comment_content(new));
    debug!(
        "comment::changed_comment_content: {}\norig: '{}'\nnew: '{}'\nraw_old: {}\nraw_new: {}",
        res,
        orig,
        new,
        code_comment_content(orig).collect::<String>(),
        code_comment_content(new).collect::<String>()
    );
    res
}

/// Iterator over the 'payload' characters of a comment.
/// It skips whitespace, comment start/end marks, and '*' at the beginning of lines.
/// The comment must be one comment, ie not more than one start mark (no multiple line comments,
/// for example).
struct CommentReducer<'a> {
    is_block: bool,
    at_start_line: bool,
    iter: std::str::Chars<'a>,
}

impl<'a> CommentReducer<'a> {
    fn new(comment: &'a str) -> CommentReducer<'a> {
        let is_block = comment.starts_with("/*");
        let comment = remove_comment_header(comment);
        CommentReducer {
            is_block,
            // There are no supplementary '*' on the first line.
            at_start_line: false,
            iter: comment.chars(),
        }
    }
}

impl<'a> Iterator for CommentReducer<'a> {
    type Item = char;

    fn next(&mut self) -> Option<Self::Item> {
        loop {
            let mut c = self.iter.next()?;
            if self.is_block && self.at_start_line {
                while c.is_whitespace() {
                    c = self.iter.next()?;
                }
                // Ignore leading '*'.
                if c == '*' {
                    c = self.iter.next()?;
                }
            } else if c == '\n' {
                self.at_start_line = true;
            }
            if !c.is_whitespace() {
                return Some(c);
            }
        }
    }
}

fn remove_comment_header(comment: &str) -> &str {
    if comment.starts_with("///") || comment.starts_with("//!") {
        &comment[3..]
    } else if let Some(stripped) = comment.strip_prefix("//") {
        stripped
    } else if (comment.starts_with("/**") && !comment.starts_with("/**/"))
        || comment.starts_with("/*!")
    {
        &comment[3..comment.len() - 2]
    } else {
        assert!(
            comment.starts_with("/*"),
            "string '{comment}' is not a comment"
        );
        &comment[2..comment.len() - 2]
    }
}

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

    #[test]
    fn char_classes() {
        let mut iter = CharClasses::new("//\n\n".chars());

        assert_eq!((FullCodeCharKind::StartComment, '/'), iter.next().unwrap());
        assert_eq!((FullCodeCharKind::InComment, '/'), iter.next().unwrap());
        assert_eq!((FullCodeCharKind::EndComment, '\n'), iter.next().unwrap());
        assert_eq!((FullCodeCharKind::Normal, '\n'), iter.next().unwrap());
        assert_eq!(None, iter.next());
    }

    #[test]
    fn comment_code_slices() {
        let input = "code(); /* test */ 1 + 1";
        let mut iter = CommentCodeSlices::new(input);

        assert_eq!((CodeCharKind::Normal, 0, "code(); "), iter.next().unwrap());
        assert_eq!(
            (CodeCharKind::Comment, 8, "/* test */"),
            iter.next().unwrap()
        );
        assert_eq!((CodeCharKind::Normal, 18, " 1 + 1"), iter.next().unwrap());
        assert_eq!(None, iter.next());
    }

    #[test]
    fn comment_code_slices_two() {
        let input = "// comment\n    test();";
        let mut iter = CommentCodeSlices::new(input);

        assert_eq!((CodeCharKind::Normal, 0, ""), iter.next().unwrap());
        assert_eq!(
            (CodeCharKind::Comment, 0, "// comment\n"),
            iter.next().unwrap()
        );
        assert_eq!(
            (CodeCharKind::Normal, 11, "    test();"),
            iter.next().unwrap()
        );
        assert_eq!(None, iter.next());
    }

    #[test]
    fn comment_code_slices_three() {
        let input = "1 // comment\n    // comment2\n\n";
        let mut iter = CommentCodeSlices::new(input);

        assert_eq!((CodeCharKind::Normal, 0, "1 "), iter.next().unwrap());
        assert_eq!(
            (CodeCharKind::Comment, 2, "// comment\n    // comment2\n"),
            iter.next().unwrap()
        );
        assert_eq!((CodeCharKind::Normal, 29, "\n"), iter.next().unwrap());
        assert_eq!(None, iter.next());
    }

    #[test]
    #[rustfmt::skip]
    fn format_doc_comments() {
        let mut wrap_normalize_config: crate::config::Config = Default::default();
        wrap_normalize_config.set().wrap_comments(true);
        wrap_normalize_config.set().normalize_comments(true);

        let mut wrap_config: crate::config::Config = Default::default();
        wrap_config.set().wrap_comments(true);

        let comment = rewrite_comment(" //test",
                                      true,
                                      Shape::legacy(100, Indent::new(0, 100)),
                                      &wrap_normalize_config).unwrap();
        assert_eq!("/* test */", comment);

        let comment = rewrite_comment("// comment on a",
                                      false,
                                      Shape::legacy(10, Indent::empty()),
                                      &wrap_normalize_config).unwrap();
        assert_eq!("// comment\n// on a", comment);

        let comment = rewrite_comment("//  A multi line comment\n             // between args.",
                                      false,
                                      Shape::legacy(60, Indent::new(0, 12)),
                                      &wrap_normalize_config).unwrap();
        assert_eq!("//  A multi line comment\n            // between args.", comment);

        let input = "// comment";
        let expected =
            "/* comment */";
        let comment = rewrite_comment(input,
                                      true,
                                      Shape::legacy(9, Indent::new(0, 69)),
                                      &wrap_normalize_config).unwrap();
        assert_eq!(expected, comment);

        let comment = rewrite_comment("/*   trimmed    */",
                                      true,
                                      Shape::legacy(100, Indent::new(0, 100)),
                                      &wrap_normalize_config).unwrap();
        assert_eq!("/* trimmed */", comment);

        // Check that different comment style are properly recognised.
        let comment = rewrite_comment(r#"/// test1
                                         /// test2
                                         /*
                                          * test3
                                          */"#,
                                      false,
                                      Shape::legacy(100, Indent::new(0, 0)),
                                      &wrap_normalize_config).unwrap();
        assert_eq!("/// test1\n/// test2\n// test3", comment);

        // Check that the blank line marks the end of a commented paragraph.
        let comment = rewrite_comment(r#"// test1

                                         // test2"#,
                                      false,
                                      Shape::legacy(100, Indent::new(0, 0)),
                                      &wrap_normalize_config).unwrap();
        assert_eq!("// test1\n\n// test2", comment);

        // Check that the blank line marks the end of a custom-commented paragraph.
        let comment = rewrite_comment(r#"//@ test1

                                         //@ test2"#,
                                      false,
                                      Shape::legacy(100, Indent::new(0, 0)),
                                      &wrap_normalize_config).unwrap();
        assert_eq!("//@ test1\n\n//@ test2", comment);

        // Check that bare lines are just indented but otherwise left unchanged.
        let comment = rewrite_comment(r#"// test1
                                         /*
                                           a bare line!

                                                another bare line!
                                          */"#,
                                      false,
                                      Shape::legacy(100, Indent::new(0, 0)),
                                      &wrap_config).unwrap();
        assert_eq!("// test1\n/*\n a bare line!\n\n      another bare line!\n*/", comment);
    }

    // This is probably intended to be a non-test fn, but it is not used.
    // We should keep this around unless it helps us test stuff to remove it.
    fn uncommented(text: &str) -> String {
        CharClasses::new(text.chars())
            .filter_map(|(s, c)| match s {
                FullCodeCharKind::Normal | FullCodeCharKind::InString => Some(c),
                _ => None,
            })
            .collect()
    }

    #[test]
    fn test_uncommented() {
        assert_eq!(&uncommented("abc/*...*/"), "abc");
        assert_eq!(
            &uncommented("// .... /* \n../* /* *** / */ */a/* // */c\n"),
            "..ac\n"
        );
        assert_eq!(&uncommented("abc \" /* */\" qsdf"), "abc \" /* */\" qsdf");
    }

    #[test]
    fn test_contains_comment() {
        assert_eq!(contains_comment("abc"), false);
        assert_eq!(contains_comment("abc // qsdf"), true);
        assert_eq!(contains_comment("abc /* kqsdf"), true);
        assert_eq!(contains_comment("abc \" /* */\" qsdf"), false);
    }

    #[test]
    fn test_find_uncommented() {
        fn check(haystack: &str, needle: &str, expected: Option<usize>) {
            assert_eq!(expected, haystack.find_uncommented(needle));
        }

        check("/*/ */test", "test", Some(6));
        check("//test\ntest", "test", Some(7));
        check("/* comment only */", "whatever", None);
        check(
            "/* comment */ some text /* more commentary */ result",
            "result",
            Some(46),
        );
        check("sup // sup", "p", Some(2));
        check("sup", "x", None);
        check(r#"π? /**/ π is nice!"#, r#"π is nice"#, Some(9));
        check("/*sup yo? \n sup*/ sup", "p", Some(20));
        check("hel/*lohello*/lo", "hello", None);
        check("acb", "ab", None);
        check(",/*A*/ ", ",", Some(0));
        check("abc", "abc", Some(0));
        check("/* abc */", "abc", None);
        check("/**/abc/* */", "abc", Some(4));
        check("\"/* abc */\"", "abc", Some(4));
        check("\"/* abc", "abc", Some(4));
    }

    #[test]
    fn test_filter_normal_code() {
        let s = r#"
fn main() {
    println!("hello, world");
}
"#;
        assert_eq!(s, filter_normal_code(s));
        let s_with_comment = r#"
fn main() {
    // hello, world
    println!("hello, world");
}
"#;
        assert_eq!(s, filter_normal_code(s_with_comment));
    }

    #[test]
    fn test_itemized_block_first_line_handling() {
        fn run_test(
            test_input: &str,
            expected_line: &str,
            expected_indent: usize,
            expected_opener: &str,
            expected_line_start: &str,
        ) {
            let block = ItemizedBlock::new(test_input).unwrap();
            assert_eq!(1, block.lines.len(), "test_input: {test_input:?}");
            assert_eq!(expected_line, &block.lines[0], "test_input: {test_input:?}");
            assert_eq!(expected_indent, block.indent, "test_input: {test_input:?}");
            assert_eq!(expected_opener, &block.opener, "test_input: {test_input:?}");
            assert_eq!(
                expected_line_start, &block.line_start,
                "test_input: {test_input:?}"
            );
        }

        run_test("- foo", "foo", 2, "- ", "  ");
        run_test("* foo", "foo", 2, "* ", "  ");
        run_test("> foo", "foo", 2, "> ", "> ");

        run_test("1. foo", "foo", 3, "1. ", "   ");
        run_test("12. foo", "foo", 4, "12. ", "    ");
        run_test("1) foo", "foo", 3, "1) ", "   ");
        run_test("12) foo", "foo", 4, "12) ", "    ");

        run_test("    - foo", "foo", 6, "    - ", "      ");

        // https://spec.commonmark.org/0.30 says: "A start number may begin with 0s":
        run_test("0. foo", "foo", 3, "0. ", "   ");
        run_test("01. foo", "foo", 4, "01. ", "    ");
    }

    #[test]
    fn test_itemized_block_nonobvious_markers_are_rejected() {
        let test_inputs = vec![
            // Non-numeric item markers (e.g. `a.` or `iv.`) are not allowed by
            // https://spec.commonmark.org/0.30/#ordered-list-marker. We also note that allowing
            // them would risk misidentifying regular words as item markers. See also the
            // discussion in https://talk.commonmark.org/t/blank-lines-before-lists-revisited/1990
            "word.  rest of the paragraph.",
            "a.  maybe this is a list item?  maybe not?",
            "iv.  maybe this is a list item?  maybe not?",
            // Numbers with 3 or more digits are not recognized as item markers, to avoid
            // formatting the following example as a list:
            //
            // ```
            // The Captain died in
            // 1868. He was buried in...
            // ```
            "123.  only 2-digit numbers are recognized as item markers.",
            // Parens:
            "123)  giving some coverage to parens as well.",
            "a)  giving some coverage to parens as well.",
            // https://spec.commonmark.org/0.30 says that "at least one space or tab is needed
            // between the list marker and any following content":
            "1.Not a list item.",
            "1.2.3. Not a list item.",
            "1)Not a list item.",
            "-Not a list item.",
            "+Not a list item.",
            "+1 not a list item.",
            // https://spec.commonmark.org/0.30 says: "A start number may not be negative":
            "-1. Not a list item.",
            "-1 Not a list item.",
            // Marker without prefix are not recognized as item markers:
            ".   Not a list item.",
            ")   Not a list item.",
        ];
        for line in test_inputs.iter() {
            let maybe_block = ItemizedBlock::new(line);
            assert!(
                maybe_block.is_none(),
                "The following line shouldn't be classified as a list item: {line}"
            );
        }
    }
}