summaryrefslogtreecommitdiff
path: root/scanner.py
blob: f98da00aa7b7cde784eb3b395c0132ef7ec2587b (plain)
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
# coding: utf-8

# Scanner produces tokens of the following types:
# STREAM-START
# STREAM-END
# DIRECTIVE(name, value)
# DOCUMENT-START
# DOCUMENT-END
# BLOCK-SEQUENCE-START
# BLOCK-MAPPING-START
# BLOCK-END
# FLOW-SEQUENCE-START
# FLOW-MAPPING-START
# FLOW-SEQUENCE-END
# FLOW-MAPPING-END
# BLOCK-ENTRY
# FLOW-ENTRY
# KEY
# VALUE
# ALIAS(value)
# ANCHOR(value)
# TAG(value)
# SCALAR(value, plain, style)
#
# RoundTripScanner
# COMMENT(value)
#
# Read comments in the Scanner code for more details.
#

import inspect
from ruamel.yaml.error import MarkedYAMLError, CommentMark
from ruamel.yaml.tokens import *  # NOQA
from ruamel.yaml.compat import _F, check_anchorname_char, nprint, nprintf  # NOQA

if False:  # MYPY
    from typing import Any, Dict, Optional, List, Union, Text  # NOQA
    from ruamel.yaml.compat import VersionType  # NOQA

__all__ = ['Scanner', 'RoundTripScanner', 'ScannerError']


_THE_END = '\n\0\r\x85\u2028\u2029'
_THE_END_SPACE_TAB = ' \n\0\t\r\x85\u2028\u2029'
_SPACE_TAB = ' \t'


class ScannerError(MarkedYAMLError):
    pass


class SimpleKey:
    # See below simple keys treatment.

    def __init__(self, token_number, required, index, line, column, mark):
        # type: (Any, Any, int, int, int, Any) -> None
        self.token_number = token_number
        self.required = required
        self.index = index
        self.line = line
        self.column = column
        self.mark = mark


class Scanner:
    def __init__(self, loader=None):
        # type: (Any) -> None
        """Initialize the scanner."""
        # It is assumed that Scanner and Reader will have a common descendant.
        # Reader do the dirty work of checking for BOM and converting the
        # input data to Unicode. It also adds NUL to the end.
        #
        # Reader supports the following methods
        #   self.peek(i=0)    # peek the next i-th character
        #   self.prefix(l=1)  # peek the next l characters
        #   self.forward(l=1) # read the next l characters and move the pointer

        self.loader = loader
        if self.loader is not None and getattr(self.loader, '_scanner', None) is None:
            self.loader._scanner = self
        self.reset_scanner()
        self.first_time = False
        self.yaml_version = None  # type: Any

    @property
    def flow_level(self):
        # type: () -> int
        return len(self.flow_context)

    def reset_scanner(self):
        # type: () -> None
        # Had we reached the end of the stream?
        self.done = False

        # flow_context is an expanding/shrinking list consisting of '{' and '['
        # for each unclosed flow context. If empty list that means block context
        self.flow_context = []  # type: List[Text]

        # List of processed tokens that are not yet emitted.
        self.tokens = []  # type: List[Any]

        # Add the STREAM-START token.
        self.fetch_stream_start()

        # Number of tokens that were emitted through the `get_token` method.
        self.tokens_taken = 0

        # The current indentation level.
        self.indent = -1

        # Past indentation levels.
        self.indents = []  # type: List[int]

        # Variables related to simple keys treatment.

        # A simple key is a key that is not denoted by the '?' indicator.
        # Example of simple keys:
        #   ---
        #   block simple key: value
        #   ? not a simple key:
        #   : { flow simple key: value }
        # We emit the KEY token before all keys, so when we find a potential
        # simple key, we try to locate the corresponding ':' indicator.
        # Simple keys should be limited to a single line and 1024 characters.

        # Can a simple key start at the current position? A simple key may
        # start:
        # - at the beginning of the line, not counting indentation spaces
        #       (in block context),
        # - after '{', '[', ',' (in the flow context),
        # - after '?', ':', '-' (in the block context).
        # In the block context, this flag also signifies if a block collection
        # may start at the current position.
        self.allow_simple_key = True

        # Keep track of possible simple keys. This is a dictionary. The key
        # is `flow_level`; there can be no more that one possible simple key
        # for each level. The value is a SimpleKey record:
        #   (token_number, required, index, line, column, mark)
        # A simple key may start with ALIAS, ANCHOR, TAG, SCALAR(flow),
        # '[', or '{' tokens.
        self.possible_simple_keys = {}  # type: Dict[Any, Any]

    @property
    def reader(self):
        # type: () -> Any
        try:
            return self._scanner_reader  # type: ignore
        except AttributeError:
            if hasattr(self.loader, 'typ'):
                self._scanner_reader = self.loader.reader
            else:
                self._scanner_reader = self.loader._reader
            return self._scanner_reader

    @property
    def scanner_processing_version(self):  # prefix until un-composited
        # type: () -> Any
        if hasattr(self.loader, 'typ'):
            return self.loader.resolver.processing_version
        return self.loader.processing_version

    # Public methods.

    def check_token(self, *choices):
        # type: (Any) -> bool
        # Check if the next token is one of the given types.
        while self.need_more_tokens():
            self.fetch_more_tokens()
        if bool(self.tokens):
            if not choices:
                return True
            for choice in choices:
                if isinstance(self.tokens[0], choice):
                    return True
        return False

    def peek_token(self):
        # type: () -> Any
        # Return the next token, but do not delete if from the queue.
        while self.need_more_tokens():
            self.fetch_more_tokens()
        if bool(self.tokens):
            return self.tokens[0]

    def get_token(self):
        # type: () -> Any
        # Return the next token.
        while self.need_more_tokens():
            self.fetch_more_tokens()
        if bool(self.tokens):
            self.tokens_taken += 1
            return self.tokens.pop(0)

    # Private methods.

    def need_more_tokens(self):
        # type: () -> bool
        if self.done:
            return False
        if not self.tokens:
            return True
        # The current token may be a potential simple key, so we
        # need to look further.
        self.stale_possible_simple_keys()
        if self.next_possible_simple_key() == self.tokens_taken:
            return True
        return False

    def fetch_comment(self, comment):
        # type: (Any) -> None
        raise NotImplementedError

    def fetch_more_tokens(self):
        # type: () -> Any
        # Eat whitespaces and comments until we reach the next token.
        comment = self.scan_to_next_token()
        if comment is not None:  # never happens for base scanner
            return self.fetch_comment(comment)
        # Remove obsolete possible simple keys.
        self.stale_possible_simple_keys()

        # Compare the current indentation and column. It may add some tokens
        # and decrease the current indentation level.
        self.unwind_indent(self.reader.column)

        # Peek the next character.
        ch = self.reader.peek()

        # Is it the end of stream?
        if ch == '\0':
            return self.fetch_stream_end()

        # Is it a directive?
        if ch == '%' and self.check_directive():
            return self.fetch_directive()

        # Is it the document start?
        if ch == '-' and self.check_document_start():
            return self.fetch_document_start()

        # Is it the document end?
        if ch == '.' and self.check_document_end():
            return self.fetch_document_end()

        # TODO: support for BOM within a stream.
        # if ch == '\uFEFF':
        #     return self.fetch_bom()    <-- issue BOMToken

        # Note: the order of the following checks is NOT significant.

        # Is it the flow sequence start indicator?
        if ch == '[':
            return self.fetch_flow_sequence_start()

        # Is it the flow mapping start indicator?
        if ch == '{':
            return self.fetch_flow_mapping_start()

        # Is it the flow sequence end indicator?
        if ch == ']':
            return self.fetch_flow_sequence_end()

        # Is it the flow mapping end indicator?
        if ch == '}':
            return self.fetch_flow_mapping_end()

        # Is it the flow entry indicator?
        if ch == ',':
            return self.fetch_flow_entry()

        # Is it the block entry indicator?
        if ch == '-' and self.check_block_entry():
            return self.fetch_block_entry()

        # Is it the key indicator?
        if ch == '?' and self.check_key():
            return self.fetch_key()

        # Is it the value indicator?
        if ch == ':' and self.check_value():
            return self.fetch_value()

        # Is it an alias?
        if ch == '*':
            return self.fetch_alias()

        # Is it an anchor?
        if ch == '&':
            return self.fetch_anchor()

        # Is it a tag?
        if ch == '!':
            return self.fetch_tag()

        # Is it a literal scalar?
        if ch == '|' and not self.flow_level:
            return self.fetch_literal()

        # Is it a folded scalar?
        if ch == '>' and not self.flow_level:
            return self.fetch_folded()

        # Is it a single quoted scalar?
        if ch == "'":
            return self.fetch_single()

        # Is it a double quoted scalar?
        if ch == '"':
            return self.fetch_double()

        # It must be a plain scalar then.
        if self.check_plain():
            return self.fetch_plain()

        # No? It's an error. Let's produce a nice error message.
        raise ScannerError(
            'while scanning for the next token',
            None,
            _F('found character {ch!r} that cannot start any token', ch=ch),
            self.reader.get_mark(),
        )

    # Simple keys treatment.

    def next_possible_simple_key(self):
        # type: () -> Any
        # Return the number of the nearest possible simple key. Actually we
        # don't need to loop through the whole dictionary. We may replace it
        # with the following code:
        #   if not self.possible_simple_keys:
        #       return None
        #   return self.possible_simple_keys[
        #           min(self.possible_simple_keys.keys())].token_number
        min_token_number = None
        for level in self.possible_simple_keys:
            key = self.possible_simple_keys[level]
            if min_token_number is None or key.token_number < min_token_number:
                min_token_number = key.token_number
        return min_token_number

    def stale_possible_simple_keys(self):
        # type: () -> None
        # Remove entries that are no longer possible simple keys. According to
        # the YAML specification, simple keys
        # - should be limited to a single line,
        # - should be no longer than 1024 characters.
        # Disabling this procedure will allow simple keys of any length and
        # height (may cause problems if indentation is broken though).
        for level in list(self.possible_simple_keys):
            key = self.possible_simple_keys[level]
            if key.line != self.reader.line or self.reader.index - key.index > 1024:
                if key.required:
                    raise ScannerError(
                        'while scanning a simple key',
                        key.mark,
                        "could not find expected ':'",
                        self.reader.get_mark(),
                    )
                del self.possible_simple_keys[level]

    def save_possible_simple_key(self):
        # type: () -> None
        # The next token may start a simple key. We check if it's possible
        # and save its position. This function is called for
        #   ALIAS, ANCHOR, TAG, SCALAR(flow), '[', and '{'.

        # Check if a simple key is required at the current position.
        required = not self.flow_level and self.indent == self.reader.column

        # The next token might be a simple key. Let's save it's number and
        # position.
        if self.allow_simple_key:
            self.remove_possible_simple_key()
            token_number = self.tokens_taken + len(self.tokens)
            key = SimpleKey(
                token_number,
                required,
                self.reader.index,
                self.reader.line,
                self.reader.column,
                self.reader.get_mark(),
            )
            self.possible_simple_keys[self.flow_level] = key

    def remove_possible_simple_key(self):
        # type: () -> None
        # Remove the saved possible key position at the current flow level.
        if self.flow_level in self.possible_simple_keys:
            key = self.possible_simple_keys[self.flow_level]

            if key.required:
                raise ScannerError(
                    'while scanning a simple key',
                    key.mark,
                    "could not find expected ':'",
                    self.reader.get_mark(),
                )

            del self.possible_simple_keys[self.flow_level]

    # Indentation functions.

    def unwind_indent(self, column):
        # type: (Any) -> None
        # In flow context, tokens should respect indentation.
        # Actually the condition should be `self.indent >= column` according to
        # the spec. But this condition will prohibit intuitively correct
        # constructions such as
        # key : {
        # }
        # ####
        # if self.flow_level and self.indent > column:
        #     raise ScannerError(None, None,
        #             "invalid intendation or unclosed '[' or '{'",
        #             self.reader.get_mark())

        # In the flow context, indentation is ignored. We make the scanner less
        # restrictive then specification requires.
        if bool(self.flow_level):
            return

        # In block context, we may need to issue the BLOCK-END tokens.
        while self.indent > column:
            mark = self.reader.get_mark()
            self.indent = self.indents.pop()
            self.tokens.append(BlockEndToken(mark, mark))

    def add_indent(self, column):
        # type: (int) -> bool
        # Check if we need to increase indentation.
        if self.indent < column:
            self.indents.append(self.indent)
            self.indent = column
            return True
        return False

    # Fetchers.

    def fetch_stream_start(self):
        # type: () -> None
        # We always add STREAM-START as the first token and STREAM-END as the
        # last token.
        # Read the token.
        mark = self.reader.get_mark()
        # Add STREAM-START.
        self.tokens.append(StreamStartToken(mark, mark, encoding=self.reader.encoding))

    def fetch_stream_end(self):
        # type: () -> None
        # Set the current intendation to -1.
        self.unwind_indent(-1)
        # Reset simple keys.
        self.remove_possible_simple_key()
        self.allow_simple_key = False
        self.possible_simple_keys = {}
        # Read the token.
        mark = self.reader.get_mark()
        # Add STREAM-END.
        self.tokens.append(StreamEndToken(mark, mark))
        # The steam is finished.
        self.done = True

    def fetch_directive(self):
        # type: () -> None
        # Set the current intendation to -1.
        self.unwind_indent(-1)

        # Reset simple keys.
        self.remove_possible_simple_key()
        self.allow_simple_key = False

        # Scan and add DIRECTIVE.
        self.tokens.append(self.scan_directive())

    def fetch_document_start(self):
        # type: () -> None
        self.fetch_document_indicator(DocumentStartToken)

    def fetch_document_end(self):
        # type: () -> None
        self.fetch_document_indicator(DocumentEndToken)

    def fetch_document_indicator(self, TokenClass):
        # type: (Any) -> None
        # Set the current intendation to -1.
        self.unwind_indent(-1)

        # Reset simple keys. Note that there could not be a block collection
        # after '---'.
        self.remove_possible_simple_key()
        self.allow_simple_key = False

        # Add DOCUMENT-START or DOCUMENT-END.
        start_mark = self.reader.get_mark()
        self.reader.forward(3)
        end_mark = self.reader.get_mark()
        self.tokens.append(TokenClass(start_mark, end_mark))

    def fetch_flow_sequence_start(self):
        # type: () -> None
        self.fetch_flow_collection_start(FlowSequenceStartToken, to_push='[')

    def fetch_flow_mapping_start(self):
        # type: () -> None
        self.fetch_flow_collection_start(FlowMappingStartToken, to_push='{')

    def fetch_flow_collection_start(self, TokenClass, to_push):
        # type: (Any, Text) -> None
        # '[' and '{' may start a simple key.
        self.save_possible_simple_key()
        # Increase the flow level.
        self.flow_context.append(to_push)
        # Simple keys are allowed after '[' and '{'.
        self.allow_simple_key = True
        # Add FLOW-SEQUENCE-START or FLOW-MAPPING-START.
        start_mark = self.reader.get_mark()
        self.reader.forward()
        end_mark = self.reader.get_mark()
        self.tokens.append(TokenClass(start_mark, end_mark))

    def fetch_flow_sequence_end(self):
        # type: () -> None
        self.fetch_flow_collection_end(FlowSequenceEndToken)

    def fetch_flow_mapping_end(self):
        # type: () -> None
        self.fetch_flow_collection_end(FlowMappingEndToken)

    def fetch_flow_collection_end(self, TokenClass):
        # type: (Any) -> None
        # Reset possible simple key on the current level.
        self.remove_possible_simple_key()
        # Decrease the flow level.
        try:
            popped = self.flow_context.pop()  # NOQA
        except IndexError:
            # We must not be in a list or object.
            # Defer error handling to the parser.
            pass
        # No simple keys after ']' or '}'.
        self.allow_simple_key = False
        # Add FLOW-SEQUENCE-END or FLOW-MAPPING-END.
        start_mark = self.reader.get_mark()
        self.reader.forward()
        end_mark = self.reader.get_mark()
        self.tokens.append(TokenClass(start_mark, end_mark))

    def fetch_flow_entry(self):
        # type: () -> None
        # Simple keys are allowed after ','.
        self.allow_simple_key = True
        # Reset possible simple key on the current level.
        self.remove_possible_simple_key()
        # Add FLOW-ENTRY.
        start_mark = self.reader.get_mark()
        self.reader.forward()
        end_mark = self.reader.get_mark()
        self.tokens.append(FlowEntryToken(start_mark, end_mark))

    def fetch_block_entry(self):
        # type: () -> None
        # Block context needs additional checks.
        if not self.flow_level:
            # Are we allowed to start a new entry?
            if not self.allow_simple_key:
                raise ScannerError(
                    None, None, 'sequence entries are not allowed here', self.reader.get_mark()
                )
            # We may need to add BLOCK-SEQUENCE-START.
            if self.add_indent(self.reader.column):
                mark = self.reader.get_mark()
                self.tokens.append(BlockSequenceStartToken(mark, mark))
        # It's an error for the block entry to occur in the flow context,
        # but we let the parser detect this.
        else:
            pass
        # Simple keys are allowed after '-'.
        self.allow_simple_key = True
        # Reset possible simple key on the current level.
        self.remove_possible_simple_key()

        # Add BLOCK-ENTRY.
        start_mark = self.reader.get_mark()
        self.reader.forward()
        end_mark = self.reader.get_mark()
        self.tokens.append(BlockEntryToken(start_mark, end_mark))

    def fetch_key(self):
        # type: () -> None
        # Block context needs additional checks.
        if not self.flow_level:

            # Are we allowed to start a key (not nessesary a simple)?
            if not self.allow_simple_key:
                raise ScannerError(
                    None, None, 'mapping keys are not allowed here', self.reader.get_mark()
                )

            # We may need to add BLOCK-MAPPING-START.
            if self.add_indent(self.reader.column):
                mark = self.reader.get_mark()
                self.tokens.append(BlockMappingStartToken(mark, mark))

        # Simple keys are allowed after '?' in the block context.
        self.allow_simple_key = not self.flow_level

        # Reset possible simple key on the current level.
        self.remove_possible_simple_key()

        # Add KEY.
        start_mark = self.reader.get_mark()
        self.reader.forward()
        end_mark = self.reader.get_mark()
        self.tokens.append(KeyToken(start_mark, end_mark))

    def fetch_value(self):
        # type: () -> None
        # Do we determine a simple key?
        if self.flow_level in self.possible_simple_keys:
            # Add KEY.
            key = self.possible_simple_keys[self.flow_level]
            del self.possible_simple_keys[self.flow_level]
            self.tokens.insert(
                key.token_number - self.tokens_taken, KeyToken(key.mark, key.mark)
            )

            # If this key starts a new block mapping, we need to add
            # BLOCK-MAPPING-START.
            if not self.flow_level:
                if self.add_indent(key.column):
                    self.tokens.insert(
                        key.token_number - self.tokens_taken,
                        BlockMappingStartToken(key.mark, key.mark),
                    )

            # There cannot be two simple keys one after another.
            self.allow_simple_key = False

        # It must be a part of a complex key.
        else:

            # Block context needs additional checks.
            # (Do we really need them? They will be caught by the parser
            # anyway.)
            if not self.flow_level:

                # We are allowed to start a complex value if and only if
                # we can start a simple key.
                if not self.allow_simple_key:
                    raise ScannerError(
                        None,
                        None,
                        'mapping values are not allowed here',
                        self.reader.get_mark(),
                    )

            # If this value starts a new block mapping, we need to add
            # BLOCK-MAPPING-START.  It will be detected as an error later by
            # the parser.
            if not self.flow_level:
                if self.add_indent(self.reader.column):
                    mark = self.reader.get_mark()
                    self.tokens.append(BlockMappingStartToken(mark, mark))

            # Simple keys are allowed after ':' in the block context.
            self.allow_simple_key = not self.flow_level

            # Reset possible simple key on the current level.
            self.remove_possible_simple_key()

        # Add VALUE.
        start_mark = self.reader.get_mark()
        self.reader.forward()
        end_mark = self.reader.get_mark()
        self.tokens.append(ValueToken(start_mark, end_mark))

    def fetch_alias(self):
        # type: () -> None
        # ALIAS could be a simple key.
        self.save_possible_simple_key()
        # No simple keys after ALIAS.
        self.allow_simple_key = False
        # Scan and add ALIAS.
        self.tokens.append(self.scan_anchor(AliasToken))

    def fetch_anchor(self):
        # type: () -> None
        # ANCHOR could start a simple key.
        self.save_possible_simple_key()
        # No simple keys after ANCHOR.
        self.allow_simple_key = False
        # Scan and add ANCHOR.
        self.tokens.append(self.scan_anchor(AnchorToken))

    def fetch_tag(self):
        # type: () -> None
        # TAG could start a simple key.
        self.save_possible_simple_key()
        # No simple keys after TAG.
        self.allow_simple_key = False
        # Scan and add TAG.
        self.tokens.append(self.scan_tag())

    def fetch_literal(self):
        # type: () -> None
        self.fetch_block_scalar(style='|')

    def fetch_folded(self):
        # type: () -> None
        self.fetch_block_scalar(style='>')

    def fetch_block_scalar(self, style):
        # type: (Any) -> None
        # A simple key may follow a block scalar.
        self.allow_simple_key = True
        # Reset possible simple key on the current level.
        self.remove_possible_simple_key()
        # Scan and add SCALAR.
        self.tokens.append(self.scan_block_scalar(style))

    def fetch_single(self):
        # type: () -> None
        self.fetch_flow_scalar(style="'")

    def fetch_double(self):
        # type: () -> None
        self.fetch_flow_scalar(style='"')

    def fetch_flow_scalar(self, style):
        # type: (Any) -> None
        # A flow scalar could be a simple key.
        self.save_possible_simple_key()
        # No simple keys after flow scalars.
        self.allow_simple_key = False
        # Scan and add SCALAR.
        self.tokens.append(self.scan_flow_scalar(style))

    def fetch_plain(self):
        # type: () -> None
        # A plain scalar could be a simple key.
        self.save_possible_simple_key()
        # No simple keys after plain scalars. But note that `scan_plain` will
        # change this flag if the scan is finished at the beginning of the
        # line.
        self.allow_simple_key = False
        # Scan and add SCALAR. May change `allow_simple_key`.
        self.tokens.append(self.scan_plain())

    # Checkers.

    def check_directive(self):
        # type: () -> Any
        # DIRECTIVE:        ^ '%' ...
        # The '%' indicator is already checked.
        if self.reader.column == 0:
            return True
        return None

    def check_document_start(self):
        # type: () -> Any
        # DOCUMENT-START:   ^ '---' (' '|'\n')
        if self.reader.column == 0:
            if self.reader.prefix(3) == '---' and self.reader.peek(3) in _THE_END_SPACE_TAB:
                return True
        return None

    def check_document_end(self):
        # type: () -> Any
        # DOCUMENT-END:     ^ '...' (' '|'\n')
        if self.reader.column == 0:
            if self.reader.prefix(3) == '...' and self.reader.peek(3) in _THE_END_SPACE_TAB:
                return True
        return None

    def check_block_entry(self):
        # type: () -> Any
        # BLOCK-ENTRY:      '-' (' '|'\n')
        return self.reader.peek(1) in _THE_END_SPACE_TAB

    def check_key(self):
        # type: () -> Any
        # KEY(flow context):    '?'
        if bool(self.flow_level):
            return True
        # KEY(block context):   '?' (' '|'\n')
        return self.reader.peek(1) in _THE_END_SPACE_TAB

    def check_value(self):
        # type: () -> Any
        # VALUE(flow context):  ':'
        if self.scanner_processing_version == (1, 1):
            if bool(self.flow_level):
                return True
        else:
            if bool(self.flow_level):
                if self.flow_context[-1] == '[':
                    if self.reader.peek(1) not in _THE_END_SPACE_TAB:
                        return False
                elif self.tokens and isinstance(self.tokens[-1], ValueToken):
                    # mapping flow context scanning a value token
                    if self.reader.peek(1) not in _THE_END_SPACE_TAB:
                        return False
                return True
        # VALUE(block context): ':' (' '|'\n')
        return self.reader.peek(1) in _THE_END_SPACE_TAB

    def check_plain(self):
        # type: () -> Any
        # A plain scalar may start with any non-space character except:
        #   '-', '?', ':', ',', '[', ']', '{', '}',
        #   '#', '&', '*', '!', '|', '>', '\'', '\"',
        #   '%', '@', '`'.
        #
        # It may also start with
        #   '-', '?', ':'
        # if it is followed by a non-space character.
        #
        # Note that we limit the last rule to the block context (except the
        # '-' character) because we want the flow context to be space
        # independent.
        srp = self.reader.peek
        ch = srp()
        if self.scanner_processing_version == (1, 1):
            return ch not in '\0 \t\r\n\x85\u2028\u2029-?:,[]{}#&*!|>\'"%@`' or (
                srp(1) not in _THE_END_SPACE_TAB
                and (ch == '-' or (not self.flow_level and ch in '?:'))
            )
        # YAML 1.2
        if ch not in '\0 \t\r\n\x85\u2028\u2029-?:,[]{}#&*!|>\'"%@`':
            # ###################                ^ ???
            return True
        ch1 = srp(1)
        if ch == '-' and ch1 not in _THE_END_SPACE_TAB:
            return True
        if ch == ':' and bool(self.flow_level) and ch1 not in _SPACE_TAB:
            return True

        return srp(1) not in _THE_END_SPACE_TAB and (
            ch == '-' or (not self.flow_level and ch in '?:')
        )

    # Scanners.

    def scan_to_next_token(self):
        # type: () -> Any
        # We ignore spaces, line breaks and comments.
        # If we find a line break in the block context, we set the flag
        # `allow_simple_key` on.
        # The byte order mark is stripped if it's the first character in the
        # stream. We do not yet support BOM inside the stream as the
        # specification requires. Any such mark will be considered as a part
        # of the document.
        #
        # TODO: We need to make tab handling rules more sane. A good rule is
        #   Tabs cannot precede tokens
        #   BLOCK-SEQUENCE-START, BLOCK-MAPPING-START, BLOCK-END,
        #   KEY(block), VALUE(block), BLOCK-ENTRY
        # So the checking code is
        #   if <TAB>:
        #       self.allow_simple_keys = False
        # We also need to add the check for `allow_simple_keys == True` to
        # `unwind_indent` before issuing BLOCK-END.
        # Scanners for block, flow, and plain scalars need to be modified.
        srp = self.reader.peek
        srf = self.reader.forward
        if self.reader.index == 0 and srp() == '\uFEFF':
            srf()
        found = False
        _the_end = _THE_END
        while not found:
            while srp() == ' ':
                srf()
            if srp() == '#':
                while srp() not in _the_end:
                    srf()
            if self.scan_line_break():
                if not self.flow_level:
                    self.allow_simple_key = True
            else:
                found = True
        return None

    def scan_directive(self):
        # type: () -> Any
        # See the specification for details.
        srp = self.reader.peek
        srf = self.reader.forward
        start_mark = self.reader.get_mark()
        srf()
        name = self.scan_directive_name(start_mark)
        value = None
        if name == 'YAML':
            value = self.scan_yaml_directive_value(start_mark)
            end_mark = self.reader.get_mark()
        elif name == 'TAG':
            value = self.scan_tag_directive_value(start_mark)
            end_mark = self.reader.get_mark()
        else:
            end_mark = self.reader.get_mark()
            while srp() not in _THE_END:
                srf()
        self.scan_directive_ignored_line(start_mark)
        return DirectiveToken(name, value, start_mark, end_mark)

    def scan_directive_name(self, start_mark):
        # type: (Any) -> Any
        # See the specification for details.
        length = 0
        srp = self.reader.peek
        ch = srp(length)
        while '0' <= ch <= '9' or 'A' <= ch <= 'Z' or 'a' <= ch <= 'z' or ch in '-_:.':
            length += 1
            ch = srp(length)
        if not length:
            raise ScannerError(
                'while scanning a directive',
                start_mark,
                _F('expected alphabetic or numeric character, but found {ch!r}', ch=ch),
                self.reader.get_mark(),
            )
        value = self.reader.prefix(length)
        self.reader.forward(length)
        ch = srp()
        if ch not in '\0 \r\n\x85\u2028\u2029':
            raise ScannerError(
                'while scanning a directive',
                start_mark,
                _F('expected alphabetic or numeric character, but found {ch!r}', ch=ch),
                self.reader.get_mark(),
            )
        return value

    def scan_yaml_directive_value(self, start_mark):
        # type: (Any) -> Any
        # See the specification for details.
        srp = self.reader.peek
        srf = self.reader.forward
        while srp() == ' ':
            srf()
        major = self.scan_yaml_directive_number(start_mark)
        if srp() != '.':
            raise ScannerError(
                'while scanning a directive',
                start_mark,
                _F("expected a digit or '.', but found {srp_call!r}", srp_call=srp()),
                self.reader.get_mark(),
            )
        srf()
        minor = self.scan_yaml_directive_number(start_mark)
        if srp() not in '\0 \r\n\x85\u2028\u2029':
            raise ScannerError(
                'while scanning a directive',
                start_mark,
                _F("expected a digit or '.', but found {srp_call!r}", srp_call=srp()),
                self.reader.get_mark(),
            )
        self.yaml_version = (major, minor)
        return self.yaml_version

    def scan_yaml_directive_number(self, start_mark):
        # type: (Any) -> Any
        # See the specification for details.
        srp = self.reader.peek
        srf = self.reader.forward
        ch = srp()
        if not ('0' <= ch <= '9'):
            raise ScannerError(
                'while scanning a directive',
                start_mark,
                _F('expected a digit, but found {ch!r}', ch=ch),
                self.reader.get_mark(),
            )
        length = 0
        while '0' <= srp(length) <= '9':
            length += 1
        value = int(self.reader.prefix(length))
        srf(length)
        return value

    def scan_tag_directive_value(self, start_mark):
        # type: (Any) -> Any
        # See the specification for details.
        srp = self.reader.peek
        srf = self.reader.forward
        while srp() == ' ':
            srf()
        handle = self.scan_tag_directive_handle(start_mark)
        while srp() == ' ':
            srf()
        prefix = self.scan_tag_directive_prefix(start_mark)
        return (handle, prefix)

    def scan_tag_directive_handle(self, start_mark):
        # type: (Any) -> Any
        # See the specification for details.
        value = self.scan_tag_handle('directive', start_mark)
        ch = self.reader.peek()
        if ch != ' ':
            raise ScannerError(
                'while scanning a directive',
                start_mark,
                _F("expected ' ', but found {ch!r}", ch=ch),
                self.reader.get_mark(),
            )
        return value

    def scan_tag_directive_prefix(self, start_mark):
        # type: (Any) -> Any
        # See the specification for details.
        value = self.scan_tag_uri('directive', start_mark)
        ch = self.reader.peek()
        if ch not in '\0 \r\n\x85\u2028\u2029':
            raise ScannerError(
                'while scanning a directive',
                start_mark,
                _F("expected ' ', but found {ch!r}", ch=ch),
                self.reader.get_mark(),
            )
        return value

    def scan_directive_ignored_line(self, start_mark):
        # type: (Any) -> None
        # See the specification for details.
        srp = self.reader.peek
        srf = self.reader.forward
        while srp() == ' ':
            srf()
        if srp() == '#':
            while srp() not in _THE_END:
                srf()
        ch = srp()
        if ch not in _THE_END:
            raise ScannerError(
                'while scanning a directive',
                start_mark,
                _F('expected a comment or a line break, but found {ch!r}', ch=ch),
                self.reader.get_mark(),
            )
        self.scan_line_break()

    def scan_anchor(self, TokenClass):
        # type: (Any) -> Any
        # The specification does not restrict characters for anchors and
        # aliases. This may lead to problems, for instance, the document:
        #   [ *alias, value ]
        # can be interpteted in two ways, as
        #   [ "value" ]
        # and
        #   [ *alias , "value" ]
        # Therefore we restrict aliases to numbers and ASCII letters.
        srp = self.reader.peek
        start_mark = self.reader.get_mark()
        indicator = srp()
        if indicator == '*':
            name = 'alias'
        else:
            name = 'anchor'
        self.reader.forward()
        length = 0
        ch = srp(length)
        # while '0' <= ch <= '9' or 'A' <= ch <= 'Z' or 'a' <= ch <= 'z' \
        #         or ch in '-_':
        while check_anchorname_char(ch):
            length += 1
            ch = srp(length)
        if not length:
            raise ScannerError(
                _F('while scanning an {name!s}', name=name),
                start_mark,
                _F('expected alphabetic or numeric character, but found {ch!r}', ch=ch),
                self.reader.get_mark(),
            )
        value = self.reader.prefix(length)
        self.reader.forward(length)
        # ch1 = ch
        # ch = srp()   # no need to peek, ch is already set
        # assert ch1 == ch
        if ch not in '\0 \t\r\n\x85\u2028\u2029?:,[]{}%@`':
            raise ScannerError(
                _F('while scanning an {name!s}', name=name),
                start_mark,
                _F('expected alphabetic or numeric character, but found {ch!r}', ch=ch),
                self.reader.get_mark(),
            )
        end_mark = self.reader.get_mark()
        return TokenClass(value, start_mark, end_mark)

    def scan_tag(self):
        # type: () -> Any
        # See the specification for details.
        srp = self.reader.peek
        start_mark = self.reader.get_mark()
        ch = srp(1)
        if ch == '<':
            handle = None
            self.reader.forward(2)
            suffix = self.scan_tag_uri('tag', start_mark)
            if srp() != '>':
                raise ScannerError(
                    'while parsing a tag',
                    start_mark,
                    _F("expected '>', but found {srp_call!r}", srp_call=srp()),
                    self.reader.get_mark(),
                )
            self.reader.forward()
        elif ch in _THE_END_SPACE_TAB:
            handle = None
            suffix = '!'
            self.reader.forward()
        else:
            length = 1
            use_handle = False
            while ch not in '\0 \r\n\x85\u2028\u2029':
                if ch == '!':
                    use_handle = True
                    break
                length += 1
                ch = srp(length)
            handle = '!'
            if use_handle:
                handle = self.scan_tag_handle('tag', start_mark)
            else:
                handle = '!'
                self.reader.forward()
            suffix = self.scan_tag_uri('tag', start_mark)
        ch = srp()
        if ch not in '\0 \r\n\x85\u2028\u2029':
            raise ScannerError(
                'while scanning a tag',
                start_mark,
                _F("expected ' ', but found {ch!r}", ch=ch),
                self.reader.get_mark(),
            )
        value = (handle, suffix)
        end_mark = self.reader.get_mark()
        return TagToken(value, start_mark, end_mark)

    def scan_block_scalar(self, style, rt=False):
        # type: (Any, Optional[bool]) -> Any
        # See the specification for details.
        srp = self.reader.peek
        if style == '>':
            folded = True
        else:
            folded = False

        chunks = []  # type: List[Any]
        start_mark = self.reader.get_mark()

        # Scan the header.
        self.reader.forward()
        chomping, increment = self.scan_block_scalar_indicators(start_mark)
        # block scalar comment e.g. : |+  # comment text
        block_scalar_comment = self.scan_block_scalar_ignored_line(start_mark)

        # Determine the indentation level and go to the first non-empty line.
        min_indent = self.indent + 1
        if increment is None:
            # no increment and top level, min_indent could be 0
            if min_indent < 1 and (
                style not in '|>'
                or (self.scanner_processing_version == (1, 1))
                and getattr(
                    self.loader, 'top_level_block_style_scalar_no_indent_error_1_1', False
                )
            ):
                min_indent = 1
            breaks, max_indent, end_mark = self.scan_block_scalar_indentation()
            indent = max(min_indent, max_indent)
        else:
            if min_indent < 1:
                min_indent = 1
            indent = min_indent + increment - 1
            breaks, end_mark = self.scan_block_scalar_breaks(indent)
        line_break = ""

        # Scan the inner part of the block scalar.
        while self.reader.column == indent and srp() != '\0':
            chunks.extend(breaks)
            leading_non_space = srp() not in ' \t'
            length = 0
            while srp(length) not in _THE_END:
                length += 1
            chunks.append(self.reader.prefix(length))
            self.reader.forward(length)
            line_break = self.scan_line_break()
            breaks, end_mark = self.scan_block_scalar_breaks(indent)
            if style in '|>' and min_indent == 0:
                # at the beginning of a line, if in block style see if
                # end of document/start_new_document
                if self.check_document_start() or self.check_document_end():
                    break
            if self.reader.column == indent and srp() != '\0':

                # Unfortunately, folding rules are ambiguous.
                #
                # This is the folding according to the specification:

                if rt and folded and line_break == '\n':
                    chunks.append('\a')
                if folded and line_break == '\n' and leading_non_space and srp() not in ' \t':
                    if not breaks:
                        chunks.append(' ')
                else:
                    chunks.append(line_break)

                # This is Clark Evans's interpretation (also in the spec
                # examples):
                #
                # if folded and line_break == '\n':
                #     if not breaks:
                #         if srp() not in ' \t':
                #             chunks.append(' ')
                #         else:
                #             chunks.append(line_break)
                # else:
                #     chunks.append(line_break)
            else:
                break

        # Process trailing line breaks. The 'chomping' setting determines
        # whether they are included in the value.
        trailing = []  # type: List[Any]
        if chomping in [None, True]:
            chunks.append(line_break)
        if chomping is True:
            chunks.extend(breaks)
        elif chomping in [None, False]:
            trailing.extend(breaks)

        # We are done.
        token = ScalarToken("".join(chunks), False, start_mark, end_mark, style)
        if block_scalar_comment is not None:
            token.add_pre_comments([block_scalar_comment])
        if len(trailing) > 0:
            # nprint('trailing 1', trailing)  # XXXXX
            # Eat whitespaces and comments until we reach the next token.
            comment = self.scan_to_next_token()
            while comment:
                trailing.append(' ' * comment[1].column + comment[0])
                comment = self.scan_to_next_token()

            # Keep track of the trailing whitespace and following comments
            # as a comment token, if isn't all included in the actual value.
            comment_end_mark = self.reader.get_mark()
            comment = CommentToken("".join(trailing), end_mark, comment_end_mark)
            token.add_post_comment(comment)
        return token

    def scan_block_scalar_indicators(self, start_mark):
        # type: (Any) -> Any
        # See the specification for details.
        srp = self.reader.peek
        chomping = None
        increment = None
        ch = srp()
        if ch in '+-':
            if ch == '+':
                chomping = True
            else:
                chomping = False
            self.reader.forward()
            ch = srp()
            if ch in '0123456789':
                increment = int(ch)
                if increment == 0:
                    raise ScannerError(
                        'while scanning a block scalar',
                        start_mark,
                        'expected indentation indicator in the range 1-9, ' 'but found 0',
                        self.reader.get_mark(),
                    )
                self.reader.forward()
        elif ch in '0123456789':
            increment = int(ch)
            if increment == 0:
                raise ScannerError(
                    'while scanning a block scalar',
                    start_mark,
                    'expected indentation indicator in the range 1-9, ' 'but found 0',
                    self.reader.get_mark(),
                )
            self.reader.forward()
            ch = srp()
            if ch in '+-':
                if ch == '+':
                    chomping = True
                else:
                    chomping = False
                self.reader.forward()
        ch = srp()
        if ch not in '\0 \r\n\x85\u2028\u2029':
            raise ScannerError(
                'while scanning a block scalar',
                start_mark,
                _F('expected chomping or indentation indicators, but found {ch!r}', ch=ch),
                self.reader.get_mark(),
            )
        return chomping, increment

    def scan_block_scalar_ignored_line(self, start_mark):
        # type: (Any) -> Any
        # See the specification for details.
        srp = self.reader.peek
        srf = self.reader.forward
        prefix = ''
        comment = None
        while srp() == ' ':
            prefix += srp()
            srf()
        if srp() == '#':
            comment = prefix
            while srp() not in _THE_END:
                comment += srp()
                srf()
        ch = srp()
        if ch not in _THE_END:
            raise ScannerError(
                'while scanning a block scalar',
                start_mark,
                _F('expected a comment or a line break, but found {ch!r}', ch=ch),
                self.reader.get_mark(),
            )
        self.scan_line_break()
        return comment

    def scan_block_scalar_indentation(self):
        # type: () -> Any
        # See the specification for details.
        srp = self.reader.peek
        srf = self.reader.forward
        chunks = []
        max_indent = 0
        end_mark = self.reader.get_mark()
        while srp() in ' \r\n\x85\u2028\u2029':
            if srp() != ' ':
                chunks.append(self.scan_line_break())
                end_mark = self.reader.get_mark()
            else:
                srf()
                if self.reader.column > max_indent:
                    max_indent = self.reader.column
        return chunks, max_indent, end_mark

    def scan_block_scalar_breaks(self, indent):
        # type: (int) -> Any
        # See the specification for details.
        chunks = []
        srp = self.reader.peek
        srf = self.reader.forward
        end_mark = self.reader.get_mark()
        while self.reader.column < indent and srp() == ' ':
            srf()
        while srp() in '\r\n\x85\u2028\u2029':
            chunks.append(self.scan_line_break())
            end_mark = self.reader.get_mark()
            while self.reader.column < indent and srp() == ' ':
                srf()
        return chunks, end_mark

    def scan_flow_scalar(self, style):
        # type: (Any) -> Any
        # See the specification for details.
        # Note that we loose indentation rules for quoted scalars. Quoted
        # scalars don't need to adhere indentation because " and ' clearly
        # mark the beginning and the end of them. Therefore we are less
        # restrictive then the specification requires. We only need to check
        # that document separators are not included in scalars.
        if style == '"':
            double = True
        else:
            double = False
        srp = self.reader.peek
        chunks = []  # type: List[Any]
        start_mark = self.reader.get_mark()
        quote = srp()
        self.reader.forward()
        chunks.extend(self.scan_flow_scalar_non_spaces(double, start_mark))
        while srp() != quote:
            chunks.extend(self.scan_flow_scalar_spaces(double, start_mark))
            chunks.extend(self.scan_flow_scalar_non_spaces(double, start_mark))
        self.reader.forward()
        end_mark = self.reader.get_mark()
        return ScalarToken("".join(chunks), False, start_mark, end_mark, style)

    ESCAPE_REPLACEMENTS = {
        '0': '\0',
        'a': '\x07',
        'b': '\x08',
        't': '\x09',
        '\t': '\x09',
        'n': '\x0A',
        'v': '\x0B',
        'f': '\x0C',
        'r': '\x0D',
        'e': '\x1B',
        ' ': '\x20',
        '"': '"',
        '/': '/',  # as per http://www.json.org/
        '\\': '\\',
        'N': '\x85',
        '_': '\xA0',
        'L': '\u2028',
        'P': '\u2029',
    }

    ESCAPE_CODES = {'x': 2, 'u': 4, 'U': 8}

    def scan_flow_scalar_non_spaces(self, double, start_mark):
        # type: (Any, Any) -> Any
        # See the specification for details.
        chunks = []  # type: List[Any]
        srp = self.reader.peek
        srf = self.reader.forward
        while True:
            length = 0
            while srp(length) not in ' \n\'"\\\0\t\r\x85\u2028\u2029':
                length += 1
            if length != 0:
                chunks.append(self.reader.prefix(length))
                srf(length)
            ch = srp()
            if not double and ch == "'" and srp(1) == "'":
                chunks.append("'")
                srf(2)
            elif (double and ch == "'") or (not double and ch in '"\\'):
                chunks.append(ch)
                srf()
            elif double and ch == '\\':
                srf()
                ch = srp()
                if ch in self.ESCAPE_REPLACEMENTS:
                    chunks.append(self.ESCAPE_REPLACEMENTS[ch])
                    srf()
                elif ch in self.ESCAPE_CODES:
                    length = self.ESCAPE_CODES[ch]
                    srf()
                    for k in range(length):
                        if srp(k) not in '0123456789ABCDEFabcdef':
                            raise ScannerError(
                                'while scanning a double-quoted scalar',
                                start_mark,
                                _F(
                                    'expected escape sequence of {length:d} hexdecimal '
                                    'numbers, but found {srp_call!r}',
                                    length=length,
                                    srp_call=srp(k),
                                ),
                                self.reader.get_mark(),
                            )
                    code = int(self.reader.prefix(length), 16)
                    chunks.append(chr(code))
                    srf(length)
                elif ch in '\n\r\x85\u2028\u2029':
                    self.scan_line_break()
                    chunks.extend(self.scan_flow_scalar_breaks(double, start_mark))
                else:
                    raise ScannerError(
                        'while scanning a double-quoted scalar',
                        start_mark,
                        _F('found unknown escape character {ch!r}', ch=ch),
                        self.reader.get_mark(),
                    )
            else:
                return chunks

    def scan_flow_scalar_spaces(self, double, start_mark):
        # type: (Any, Any) -> Any
        # See the specification for details.
        srp = self.reader.peek
        chunks = []
        length = 0
        while srp(length) in ' \t':
            length += 1
        whitespaces = self.reader.prefix(length)
        self.reader.forward(length)
        ch = srp()
        if ch == '\0':
            raise ScannerError(
                'while scanning a quoted scalar',
                start_mark,
                'found unexpected end of stream',
                self.reader.get_mark(),
            )
        elif ch in '\r\n\x85\u2028\u2029':
            line_break = self.scan_line_break()
            breaks = self.scan_flow_scalar_breaks(double, start_mark)
            if line_break != '\n':
                chunks.append(line_break)
            elif not breaks:
                chunks.append(' ')
            chunks.extend(breaks)
        else:
            chunks.append(whitespaces)
        return chunks

    def scan_flow_scalar_breaks(self, double, start_mark):
        # type: (Any, Any) -> Any
        # See the specification for details.
        chunks = []  # type: List[Any]
        srp = self.reader.peek
        srf = self.reader.forward
        while True:
            # Instead of checking indentation, we check for document
            # separators.
            prefix = self.reader.prefix(3)
            if (prefix == '---' or prefix == '...') and srp(3) in _THE_END_SPACE_TAB:
                raise ScannerError(
                    'while scanning a quoted scalar',
                    start_mark,
                    'found unexpected document separator',
                    self.reader.get_mark(),
                )
            while srp() in ' \t':
                srf()
            if srp() in '\r\n\x85\u2028\u2029':
                chunks.append(self.scan_line_break())
            else:
                return chunks

    def scan_plain(self):
        # type: () -> Any
        # See the specification for details.
        # We add an additional restriction for the flow context:
        #   plain scalars in the flow context cannot contain ',', ': '  and '?'.
        # We also keep track of the `allow_simple_key` flag here.
        # Indentation rules are loosed for the flow context.
        srp = self.reader.peek
        srf = self.reader.forward
        chunks = []  # type: List[Any]
        start_mark = self.reader.get_mark()
        end_mark = start_mark
        indent = self.indent + 1
        # We allow zero indentation for scalars, but then we need to check for
        # document separators at the beginning of the line.
        # if indent == 0:
        #     indent = 1
        spaces = []  # type: List[Any]
        while True:
            length = 0
            if srp() == '#':
                break
            while True:
                ch = srp(length)
                if ch == ':' and srp(length + 1) not in _THE_END_SPACE_TAB:
                    pass
                elif ch == '?' and self.scanner_processing_version != (1, 1):
                    pass
                elif (
                    ch in _THE_END_SPACE_TAB
                    or (
                        not self.flow_level
                        and ch == ':'
                        and srp(length + 1) in _THE_END_SPACE_TAB
                    )
                    or (self.flow_level and ch in ',:?[]{}')
                ):
                    break
                length += 1
            # It's not clear what we should do with ':' in the flow context.
            if (
                self.flow_level
                and ch == ':'
                and srp(length + 1) not in '\0 \t\r\n\x85\u2028\u2029,[]{}'
            ):
                srf(length)
                raise ScannerError(
                    'while scanning a plain scalar',
                    start_mark,
                    "found unexpected ':'",
                    self.reader.get_mark(),
                    'Please check '
                    'http://pyyaml.org/wiki/YAMLColonInFlowContext '
                    'for details.',
                )
            if length == 0:
                break
            self.allow_simple_key = False
            chunks.extend(spaces)
            chunks.append(self.reader.prefix(length))
            srf(length)
            end_mark = self.reader.get_mark()
            spaces = self.scan_plain_spaces(indent, start_mark)
            if (
                not spaces
                or srp() == '#'
                or (not self.flow_level and self.reader.column < indent)
            ):
                break

        token = ScalarToken("".join(chunks), True, start_mark, end_mark)
        if spaces and spaces[0] == '\n':
            # Create a comment token to preserve the trailing line breaks.
            comment = CommentToken("".join(spaces) + '\n', start_mark, end_mark)
            token.add_post_comment(comment)
        return token

    def scan_plain_spaces(self, indent, start_mark):
        # type: (Any, Any) -> Any
        # See the specification for details.
        # The specification is really confusing about tabs in plain scalars.
        # We just forbid them completely. Do not use tabs in YAML!
        srp = self.reader.peek
        srf = self.reader.forward
        chunks = []
        length = 0
        while srp(length) in ' ':
            length += 1
        whitespaces = self.reader.prefix(length)
        self.reader.forward(length)
        ch = srp()
        if ch in '\r\n\x85\u2028\u2029':
            line_break = self.scan_line_break()
            self.allow_simple_key = True
            prefix = self.reader.prefix(3)
            if (prefix == '---' or prefix == '...') and srp(3) in _THE_END_SPACE_TAB:
                return
            breaks = []
            while srp() in ' \r\n\x85\u2028\u2029':
                if srp() == ' ':
                    srf()
                else:
                    breaks.append(self.scan_line_break())
                    prefix = self.reader.prefix(3)
                    if (prefix == '---' or prefix == '...') and srp(3) in _THE_END_SPACE_TAB:
                        return
            if line_break != '\n':
                chunks.append(line_break)
            elif not breaks:
                chunks.append(' ')
            chunks.extend(breaks)
        elif whitespaces:
            chunks.append(whitespaces)
        return chunks

    def scan_tag_handle(self, name, start_mark):
        # type: (Any, Any) -> Any
        # See the specification for details.
        # For some strange reasons, the specification does not allow '_' in
        # tag handles. I have allowed it anyway.
        srp = self.reader.peek
        ch = srp()
        if ch != '!':
            raise ScannerError(
                _F('while scanning an {name!s}', name=name),
                start_mark,
                _F("expected '!', but found {ch!r}", ch=ch),
                self.reader.get_mark(),
            )
        length = 1
        ch = srp(length)
        if ch != ' ':
            while '0' <= ch <= '9' or 'A' <= ch <= 'Z' or 'a' <= ch <= 'z' or ch in '-_':
                length += 1
                ch = srp(length)
            if ch != '!':
                self.reader.forward(length)
                raise ScannerError(
                    _F('while scanning an {name!s}', name=name),
                    start_mark,
                    _F("expected '!', but found {ch!r}", ch=ch),
                    self.reader.get_mark(),
                )
            length += 1
        value = self.reader.prefix(length)
        self.reader.forward(length)
        return value

    def scan_tag_uri(self, name, start_mark):
        # type: (Any, Any) -> Any
        # See the specification for details.
        # Note: we do not check if URI is well-formed.
        srp = self.reader.peek
        chunks = []
        length = 0
        ch = srp(length)
        while (
            '0' <= ch <= '9'
            or 'A' <= ch <= 'Z'
            or 'a' <= ch <= 'z'
            or ch in "-;/?:@&=+$,_.!~*'()[]%"
            or ((self.scanner_processing_version > (1, 1)) and ch == '#')
        ):
            if ch == '%':
                chunks.append(self.reader.prefix(length))
                self.reader.forward(length)
                length = 0
                chunks.append(self.scan_uri_escapes(name, start_mark))
            else:
                length += 1
            ch = srp(length)
        if length != 0:
            chunks.append(self.reader.prefix(length))
            self.reader.forward(length)
            length = 0
        if not chunks:
            raise ScannerError(
                _F('while parsing an {name!s}', name=name),
                start_mark,
                _F('expected URI, but found {ch!r}', ch=ch),
                self.reader.get_mark(),
            )
        return "".join(chunks)

    def scan_uri_escapes(self, name, start_mark):
        # type: (Any, Any) -> Any
        # See the specification for details.
        srp = self.reader.peek
        srf = self.reader.forward
        code_bytes = []  # type: List[Any]
        mark = self.reader.get_mark()
        while srp() == '%':
            srf()
            for k in range(2):
                if srp(k) not in '0123456789ABCDEFabcdef':
                    raise ScannerError(
                        _F('while scanning an {name!s}', name=name),
                        start_mark,
                        _F(
                            'expected URI escape sequence of 2 hexdecimal numbers,'
                            ' but found {srp_call!r}',
                            srp_call=srp(k),
                        ),
                        self.reader.get_mark(),
                    )
            code_bytes.append(int(self.reader.prefix(2), 16))
            srf(2)
        try:
            value = bytes(code_bytes).decode('utf-8')
        except UnicodeDecodeError as exc:
            raise ScannerError(
                _F('while scanning an {name!s}', name=name), start_mark, str(exc), mark
            )
        return value

    def scan_line_break(self):
        # type: () -> Any
        # Transforms:
        #   '\r\n'      :   '\n'
        #   '\r'        :   '\n'
        #   '\n'        :   '\n'
        #   '\x85'      :   '\n'
        #   '\u2028'    :   '\u2028'
        #   '\u2029     :   '\u2029'
        #   default     :   ''
        ch = self.reader.peek()
        if ch in '\r\n\x85':
            if self.reader.prefix(2) == '\r\n':
                self.reader.forward(2)
            else:
                self.reader.forward()
            return '\n'
        elif ch in '\u2028\u2029':
            self.reader.forward()
            return ch
        return ""


class RoundTripScanner(Scanner):
    def check_token(self, *choices):
        # type: (Any) -> bool
        # Check if the next token is one of the given types.
        while self.need_more_tokens():
            self.fetch_more_tokens()
        self._gather_comments()
        if bool(self.tokens):
            if not choices:
                return True
            for choice in choices:
                if isinstance(self.tokens[0], choice):
                    return True
        return False

    def peek_token(self):
        # type: () -> Any
        # Return the next token, but do not delete if from the queue.
        while self.need_more_tokens():
            self.fetch_more_tokens()
        self._gather_comments()
        if bool(self.tokens):
            return self.tokens[0]
        return None

    def _gather_comments(self):
        # type: () -> Any
        """combine multiple comment lines"""
        comments = []  # type: List[Any]
        if not self.tokens:
            return comments
        if isinstance(self.tokens[0], CommentToken):
            comment = self.tokens.pop(0)
            self.tokens_taken += 1
            comments.append(comment)
        while self.need_more_tokens():
            self.fetch_more_tokens()
            if not self.tokens:
                return comments
            if isinstance(self.tokens[0], CommentToken):
                self.tokens_taken += 1
                comment = self.tokens.pop(0)
                # nprint('dropping2', comment)
                comments.append(comment)
        if len(comments) >= 1:
            self.tokens[0].add_pre_comments(comments)
        # pull in post comment on e.g. ':'
        if not self.done and len(self.tokens) < 2:
            self.fetch_more_tokens()

    def get_token(self):
        # type: () -> Any
        # Return the next token.
        while self.need_more_tokens():
            self.fetch_more_tokens()
        self._gather_comments()
        if bool(self.tokens):
            # nprint('tk', self.tokens)
            # only add post comment to single line tokens:
            # scalar, value token. FlowXEndToken, otherwise
            # hidden streamtokens could get them (leave them and they will be
            # pre comments for the next map/seq
            if (
                len(self.tokens) > 1
                and isinstance(
                    self.tokens[0],
                    (ScalarToken, ValueToken, FlowSequenceEndToken, FlowMappingEndToken),
                )
                and isinstance(self.tokens[1], CommentToken)
                and self.tokens[0].end_mark.line == self.tokens[1].start_mark.line
            ):
                self.tokens_taken += 1
                c = self.tokens.pop(1)
                self.fetch_more_tokens()
                while len(self.tokens) > 1 and isinstance(self.tokens[1], CommentToken):
                    self.tokens_taken += 1
                    c1 = self.tokens.pop(1)
                    c.value = c.value + (' ' * c1.start_mark.column) + c1.value
                    self.fetch_more_tokens()
                self.tokens[0].add_post_comment(c)
            elif (
                len(self.tokens) > 1
                and isinstance(self.tokens[0], ScalarToken)
                and isinstance(self.tokens[1], CommentToken)
                and self.tokens[0].end_mark.line != self.tokens[1].start_mark.line
            ):
                self.tokens_taken += 1
                c = self.tokens.pop(1)
                c.value = (
                    '\n' * (c.start_mark.line - self.tokens[0].end_mark.line)
                    + (' ' * c.start_mark.column)
                    + c.value
                )
                self.tokens[0].add_post_comment(c)
                self.fetch_more_tokens()
                while len(self.tokens) > 1 and isinstance(self.tokens[1], CommentToken):
                    self.tokens_taken += 1
                    c1 = self.tokens.pop(1)
                    c.value = c.value + (' ' * c1.start_mark.column) + c1.value
                    self.fetch_more_tokens()
            self.tokens_taken += 1
            return self.tokens.pop(0)
        return None

    def fetch_comment(self, comment):
        # type: (Any) -> None
        value, start_mark, end_mark = comment
        while value and value[-1] == ' ':
            # empty line within indented key context
            # no need to update end-mark, that is not used
            value = value[:-1]
        self.tokens.append(CommentToken(value, start_mark, end_mark))

    # scanner

    def scan_to_next_token(self):
        # type: () -> Any
        # We ignore spaces, line breaks and comments.
        # If we find a line break in the block context, we set the flag
        # `allow_simple_key` on.
        # The byte order mark is stripped if it's the first character in the
        # stream. We do not yet support BOM inside the stream as the
        # specification requires. Any such mark will be considered as a part
        # of the document.
        #
        # TODO: We need to make tab handling rules more sane. A good rule is
        #   Tabs cannot precede tokens
        #   BLOCK-SEQUENCE-START, BLOCK-MAPPING-START, BLOCK-END,
        #   KEY(block), VALUE(block), BLOCK-ENTRY
        # So the checking code is
        #   if <TAB>:
        #       self.allow_simple_keys = False
        # We also need to add the check for `allow_simple_keys == True` to
        # `unwind_indent` before issuing BLOCK-END.
        # Scanners for block, flow, and plain scalars need to be modified.

        srp = self.reader.peek
        srf = self.reader.forward
        if self.reader.index == 0 and srp() == '\uFEFF':
            srf()
        found = False
        while not found:
            while srp() == ' ':
                srf()
            ch = srp()
            if ch == '#':
                start_mark = self.reader.get_mark()
                comment = ch
                srf()
                while ch not in _THE_END:
                    ch = srp()
                    if ch == '\0':  # don't gobble the end-of-stream character
                        # but add an explicit newline as "YAML processors should terminate
                        # the stream with an explicit line break
                        # https://yaml.org/spec/1.2/spec.html#id2780069
                        comment += '\n'
                        break
                    comment += ch
                    srf()
                # gather any blank lines following the comment too
                ch = self.scan_line_break()
                while len(ch) > 0:
                    comment += ch
                    ch = self.scan_line_break()
                end_mark = self.reader.get_mark()
                if not self.flow_level:
                    self.allow_simple_key = True
                return comment, start_mark, end_mark
            if bool(self.scan_line_break()):
                start_mark = self.reader.get_mark()
                if not self.flow_level:
                    self.allow_simple_key = True
                ch = srp()
                if ch == '\n':  # empty toplevel lines
                    start_mark = self.reader.get_mark()
                    comment = ""
                    while ch:
                        ch = self.scan_line_break(empty_line=True)
                        comment += ch
                    if srp() == '#':
                        # empty line followed by indented real comment
                        comment = comment.rsplit('\n', 1)[0] + '\n'
                    end_mark = self.reader.get_mark()
                    return comment, start_mark, end_mark
            else:
                found = True
        return None

    def scan_line_break(self, empty_line=False):
        # type: (bool) -> Text
        # Transforms:
        #   '\r\n'      :   '\n'
        #   '\r'        :   '\n'
        #   '\n'        :   '\n'
        #   '\x85'      :   '\n'
        #   '\u2028'    :   '\u2028'
        #   '\u2029     :   '\u2029'
        #   default     :   ''
        ch = self.reader.peek()  # type: Text
        if ch in '\r\n\x85':
            if self.reader.prefix(2) == '\r\n':
                self.reader.forward(2)
            else:
                self.reader.forward()
            return '\n'
        elif ch in '\u2028\u2029':
            self.reader.forward()
            return ch
        elif empty_line and ch in '\t ':
            self.reader.forward()
            return ch
        return ""

    def scan_block_scalar(self, style, rt=True):
        # type: (Any, Optional[bool]) -> Any
        return Scanner.scan_block_scalar(self, style, rt=rt)