aboutsummaryrefslogtreecommitdiffstats
path: root/contrib/libs/curl/lib/urlapi.c
blob: 3d4a3f94b85eebbd796e9e19f3cdbc90b81715b8 (plain) (blame)
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
/***************************************************************************
 *                                  _   _ ____  _
 *  Project                     ___| | | |  _ \| |
 *                             / __| | | | |_) | |
 *                            | (__| |_| |  _ <| |___
 *                             \___|\___/|_| \_\_____|
 *
 * Copyright (C) Daniel Stenberg, <daniel@haxx.se>, et al.
 *
 * This software is licensed as described in the file COPYING, which
 * you should have received as part of this distribution. The terms
 * are also available at https://curl.se/docs/copyright.html.
 *
 * You may opt to use, copy, modify, merge, publish, distribute and/or sell
 * copies of the Software, and permit persons to whom the Software is
 * furnished to do so, under the terms of the COPYING file.
 *
 * This software is distributed on an "AS IS" basis, WITHOUT WARRANTY OF ANY
 * KIND, either express or implied.
 *
 * SPDX-License-Identifier: curl
 *
 ***************************************************************************/

#include "curl_setup.h"

#include "urldata.h"
#include "urlapi-int.h"
#include "strcase.h"
#include "url.h"
#include "escape.h"
#include "curl_ctype.h"
#include "inet_pton.h"
#include "inet_ntop.h"
#include "strdup.h"
#include "idn.h"
#include "curl_memrchr.h"

/* The last 3 #include files should be in this order */
#include "curl_printf.h"
#include "curl_memory.h"
#include "memdebug.h"

  /* MS-DOS/Windows style drive prefix, eg c: in c:foo */
#define STARTS_WITH_DRIVE_PREFIX(str) \
  ((('a' <= str[0] && str[0] <= 'z') || \
    ('A' <= str[0] && str[0] <= 'Z')) && \
   (str[1] == ':'))

  /* MS-DOS/Windows style drive prefix, optionally with
   * a '|' instead of ':', followed by a slash or NUL */
#define STARTS_WITH_URL_DRIVE_PREFIX(str) \
  ((('a' <= (str)[0] && (str)[0] <= 'z') || \
    ('A' <= (str)[0] && (str)[0] <= 'Z')) && \
   ((str)[1] == ':' || (str)[1] == '|') && \
   ((str)[2] == '/' || (str)[2] == '\\' || (str)[2] == 0))

/* scheme is not URL encoded, the longest libcurl supported ones are... */
#define MAX_SCHEME_LEN 40

/*
 * If USE_IPV6 is disabled, we still want to parse IPv6 addresses, so make
 * sure we have _some_ value for AF_INET6 without polluting our fake value
 * everywhere.
 */
#if !defined(USE_IPV6) && !defined(AF_INET6)
#define AF_INET6 (AF_INET + 1)
#endif

/* Internal representation of CURLU. Point to URL-encoded strings. */
struct Curl_URL {
  char *scheme;
  char *user;
  char *password;
  char *options; /* IMAP only? */
  char *host;
  char *zoneid; /* for numerical IPv6 addresses */
  char *port;
  char *path;
  char *query;
  char *fragment;
  unsigned short portnum; /* the numerical version (if 'port' is set) */
  BIT(query_present);    /* to support blank */
  BIT(fragment_present); /* to support blank */
  BIT(guessed_scheme);   /* when a URL without scheme is parsed */
};

#define DEFAULT_SCHEME "https"

static void free_urlhandle(struct Curl_URL *u)
{
  free(u->scheme);
  free(u->user);
  free(u->password);
  free(u->options);
  free(u->host);
  free(u->zoneid);
  free(u->port);
  free(u->path);
  free(u->query);
  free(u->fragment);
}

/*
 * Find the separator at the end of the hostname, or the '?' in cases like
 * http://www.example.com?id=2380
 */
static const char *find_host_sep(const char *url)
{
  const char *sep;
  const char *query;

  /* Find the start of the hostname */
  sep = strstr(url, "//");
  if(!sep)
    sep = url;
  else
    sep += 2;

  query = strchr(sep, '?');
  sep = strchr(sep, '/');

  if(!sep)
    sep = url + strlen(url);

  if(!query)
    query = url + strlen(url);

  return sep < query ? sep : query;
}

/* convert CURLcode to CURLUcode */
#define cc2cu(x) ((x) == CURLE_TOO_LARGE ? CURLUE_TOO_LARGE :   \
                  CURLUE_OUT_OF_MEMORY)
/*
 * Decide whether a character in a URL must be escaped.
 */
#define urlchar_needs_escaping(c) (!(ISCNTRL(c) || ISSPACE(c) || ISGRAPH(c)))

static const char hexdigits[] = "0123456789abcdef";
/* urlencode_str() writes data into an output dynbuf and URL-encodes the
 * spaces in the source URL accordingly.
 *
 * URL encoding should be skipped for hostnames, otherwise IDN resolution
 * will fail.
 */
static CURLUcode urlencode_str(struct dynbuf *o, const char *url,
                               size_t len, bool relative,
                               bool query)
{
  /* we must add this with whitespace-replacing */
  bool left = !query;
  const unsigned char *iptr;
  const unsigned char *host_sep = (const unsigned char *) url;
  CURLcode result;

  if(!relative)
    host_sep = (const unsigned char *) find_host_sep(url);

  for(iptr = (unsigned char *)url;    /* read from here */
      len; iptr++, len--) {

    if(iptr < host_sep) {
      result = Curl_dyn_addn(o, iptr, 1);
      if(result)
        return cc2cu(result);
      continue;
    }

    if(*iptr == ' ') {
      if(left)
        result = Curl_dyn_addn(o, "%20", 3);
      else
        result = Curl_dyn_addn(o, "+", 1);
      if(result)
        return cc2cu(result);
      continue;
    }

    if(*iptr == '?')
      left = FALSE;

    if(urlchar_needs_escaping(*iptr)) {
      char out[3]={'%'};
      out[1] = hexdigits[*iptr>>4];
      out[2] = hexdigits[*iptr & 0xf];
      result = Curl_dyn_addn(o, out, 3);
    }
    else
      result = Curl_dyn_addn(o, iptr, 1);
    if(result)
      return cc2cu(result);
  }

  return CURLUE_OK;
}

/*
 * Returns the length of the scheme if the given URL is absolute (as opposed
 * to relative). Stores the scheme in the buffer if TRUE and 'buf' is
 * non-NULL. The buflen must be larger than MAX_SCHEME_LEN if buf is set.
 *
 * If 'guess_scheme' is TRUE, it means the URL might be provided without
 * scheme.
 */
size_t Curl_is_absolute_url(const char *url, char *buf, size_t buflen,
                            bool guess_scheme)
{
  size_t i = 0;
  DEBUGASSERT(!buf || (buflen > MAX_SCHEME_LEN));
  (void)buflen; /* only used in debug-builds */
  if(buf)
    buf[0] = 0; /* always leave a defined value in buf */
#ifdef _WIN32
  if(guess_scheme && STARTS_WITH_DRIVE_PREFIX(url))
    return 0;
#endif
  if(ISALPHA(url[0]))
    for(i = 1; i < MAX_SCHEME_LEN; ++i) {
      char s = url[i];
      if(s && (ISALNUM(s) || (s == '+') || (s == '-') || (s == '.') )) {
        /* RFC 3986 3.1 explains:
           scheme      = ALPHA *( ALPHA / DIGIT / "+" / "-" / "." )
        */
      }
      else {
        break;
      }
    }
  if(i && (url[i] == ':') && ((url[i + 1] == '/') || !guess_scheme)) {
    /* If this does not guess scheme, the scheme always ends with the colon so
       that this also detects data: URLs etc. In guessing mode, data: could
       be the hostname "data" with a specified port number. */

    /* the length of the scheme is the name part only */
    size_t len = i;
    if(buf) {
      Curl_strntolower(buf, url, i);
      buf[i] = 0;
    }
    return len;
  }
  return 0;
}

/*
 * Concatenate a relative URL to a base URL making it absolute.
 * URL-encodes any spaces.
 * The returned pointer must be freed by the caller unless NULL
 * (returns NULL on out of memory).
 *
 * Note that this function destroys the 'base' string.
 */
static CURLcode concat_url(char *base, const char *relurl, char **newurl)
{
  /***
   TRY to append this new path to the old URL
   to the right of the host part. Oh crap, this is doomed to cause
   problems in the future...
  */
  struct dynbuf newest;
  char *protsep;
  char *pathsep;
  bool host_changed = FALSE;
  const char *useurl = relurl;
  CURLcode result = CURLE_OK;
  CURLUcode uc;
  bool skip_slash = FALSE;
  *newurl = NULL;

  /* protsep points to the start of the hostname */
  protsep = strstr(base, "//");
  if(!protsep)
    protsep = base;
  else
    protsep += 2; /* pass the slashes */

  if('/' != relurl[0]) {
    int level = 0;

    /* First we need to find out if there is a ?-letter in the URL,
       and cut it and the right-side of that off */
    pathsep = strchr(protsep, '?');
    if(pathsep)
      *pathsep = 0;

    /* we have a relative path to append to the last slash if there is one
       available, or the new URL is just a query string (starts with a '?') or
       a fragment (starts with '#') we append the new one at the end of the
       current URL */
    if((useurl[0] != '?') && (useurl[0] != '#')) {
      pathsep = strrchr(protsep, '/');
      if(pathsep)
        *pathsep = 0;

      /* Check if there is any slash after the hostname, and if so, remember
         that position instead */
      pathsep = strchr(protsep, '/');
      if(pathsep)
        protsep = pathsep + 1;
      else
        protsep = NULL;

      /* now deal with one "./" or any amount of "../" in the newurl
         and act accordingly */

      if((useurl[0] == '.') && (useurl[1] == '/'))
        useurl += 2; /* just skip the "./" */

      while((useurl[0] == '.') &&
            (useurl[1] == '.') &&
            (useurl[2] == '/')) {
        level++;
        useurl += 3; /* pass the "../" */
      }

      if(protsep) {
        while(level--) {
          /* cut off one more level from the right of the original URL */
          pathsep = strrchr(protsep, '/');
          if(pathsep)
            *pathsep = 0;
          else {
            *protsep = 0;
            break;
          }
        }
      }
    }
    else
      skip_slash = TRUE;
  }
  else {
    /* We got a new absolute path for this server */

    if(relurl[1] == '/') {
      /* the new URL starts with //, just keep the protocol part from the
         original one */
      *protsep = 0;
      useurl = &relurl[2]; /* we keep the slashes from the original, so we
                              skip the new ones */
      host_changed = TRUE;
    }
    else {
      /* cut off the original URL from the first slash, or deal with URLs
         without slash */
      pathsep = strchr(protsep, '/');
      if(pathsep) {
        /* When people use badly formatted URLs, such as
           "http://www.example.com?dir=/home/daniel" we must not use the first
           slash, if there is a ?-letter before it! */
        char *sep = strchr(protsep, '?');
        if(sep && (sep < pathsep))
          pathsep = sep;
        *pathsep = 0;
      }
      else {
        /* There was no slash. Now, since we might be operating on a badly
           formatted URL, such as "http://www.example.com?id=2380" which does
           not use a slash separator as it is supposed to, we need to check
           for a ?-letter as well! */
        pathsep = strchr(protsep, '?');
        if(pathsep)
          *pathsep = 0;
      }
    }
  }

  Curl_dyn_init(&newest, CURL_MAX_INPUT_LENGTH);

  /* copy over the root URL part */
  result = Curl_dyn_add(&newest, base);
  if(result)
    return result;

  /* check if we need to append a slash */
  if(('/' == useurl[0]) || (protsep && !*protsep) || skip_slash)
    ;
  else {
    result = Curl_dyn_addn(&newest, "/", 1);
    if(result)
      return result;
  }

  /* then append the new piece on the right side */
  uc = urlencode_str(&newest, useurl, strlen(useurl), !host_changed,
                     FALSE);
  if(uc)
    return (uc == CURLUE_TOO_LARGE) ? CURLE_TOO_LARGE : CURLE_OUT_OF_MEMORY;

  *newurl = Curl_dyn_ptr(&newest);
  return CURLE_OK;
}

/* scan for byte values <= 31, 127 and sometimes space */
static CURLUcode junkscan(const char *url, size_t *urllen, unsigned int flags)
{
  static const char badbytes[]={
    /* */ 0x01, 0x02, 0x03, 0x04, 0x05, 0x06, 0x07,
    0x08, 0x09, 0x0a, 0x0b, 0x0c, 0x0d, 0x0e, 0x0f,
    0x10, 0x11, 0x12, 0x13, 0x14, 0x15, 0x16, 0x17,
    0x18, 0x19, 0x1a, 0x1b, 0x1c, 0x1d, 0x1e, 0x1f,
    0x7f, 0x00 /* null-terminate */
  };
  size_t n = strlen(url);
  size_t nfine;

  if(n > CURL_MAX_INPUT_LENGTH)
    /* excessive input length */
    return CURLUE_MALFORMED_INPUT;

  nfine = strcspn(url, badbytes);
  if((nfine != n) ||
     (!(flags & CURLU_ALLOW_SPACE) && strchr(url, ' ')))
    return CURLUE_MALFORMED_INPUT;

  *urllen = n;
  return CURLUE_OK;
}

/*
 * parse_hostname_login()
 *
 * Parse the login details (username, password and options) from the URL and
 * strip them out of the hostname
 *
 */
static CURLUcode parse_hostname_login(struct Curl_URL *u,
                                      const char *login,
                                      size_t len,
                                      unsigned int flags,
                                      size_t *offset) /* to the hostname */
{
  CURLUcode result = CURLUE_OK;
  CURLcode ccode;
  char *userp = NULL;
  char *passwdp = NULL;
  char *optionsp = NULL;
  const struct Curl_handler *h = NULL;

  /* At this point, we assume all the other special cases have been taken
   * care of, so the host is at most
   *
   *   [user[:password][;options]]@]hostname
   *
   * We need somewhere to put the embedded details, so do that first.
   */
  char *ptr;

  DEBUGASSERT(login);

  *offset = 0;
  ptr = memchr(login, '@', len);
  if(!ptr)
    goto out;

  /* We will now try to extract the
   * possible login information in a string like:
   * ftp://user:password@ftp.my.site:8021/README */
  ptr++;

  /* if this is a known scheme, get some details */
  if(u->scheme)
    h = Curl_get_scheme_handler(u->scheme);

  /* We could use the login information in the URL so extract it. Only parse
     options if the handler says we should. Note that 'h' might be NULL! */
  ccode = Curl_parse_login_details(login, ptr - login - 1,
                                   &userp, &passwdp,
                                   (h && (h->flags & PROTOPT_URLOPTIONS)) ?
                                   &optionsp:NULL);
  if(ccode) {
    result = CURLUE_BAD_LOGIN;
    goto out;
  }

  if(userp) {
    if(flags & CURLU_DISALLOW_USER) {
      /* Option DISALLOW_USER is set and URL contains username. */
      result = CURLUE_USER_NOT_ALLOWED;
      goto out;
    }
    free(u->user);
    u->user = userp;
  }

  if(passwdp) {
    free(u->password);
    u->password = passwdp;
  }

  if(optionsp) {
    free(u->options);
    u->options = optionsp;
  }

  /* the hostname starts at this offset */
  *offset = ptr - login;
  return CURLUE_OK;

out:

  free(userp);
  free(passwdp);
  free(optionsp);
  u->user = NULL;
  u->password = NULL;
  u->options = NULL;

  return result;
}

UNITTEST CURLUcode Curl_parse_port(struct Curl_URL *u, struct dynbuf *host,
                                   bool has_scheme)
{
  char *portptr;
  char *hostname = Curl_dyn_ptr(host);
  /*
   * Find the end of an IPv6 address on the ']' ending bracket.
   */
  if(hostname[0] == '[') {
    portptr = strchr(hostname, ']');
    if(!portptr)
      return CURLUE_BAD_IPV6;
    portptr++;
    /* this is a RFC2732-style specified IP-address */
    if(*portptr) {
      if(*portptr != ':')
        return CURLUE_BAD_PORT_NUMBER;
    }
    else
      portptr = NULL;
  }
  else
    portptr = strchr(hostname, ':');

  if(portptr) {
    char *rest = NULL;
    unsigned long port;
    size_t keep = portptr - hostname;

    /* Browser behavior adaptation. If there is a colon with no digits after,
       just cut off the name there which makes us ignore the colon and just
       use the default port. Firefox, Chrome and Safari all do that.

       Do not do it if the URL has no scheme, to make something that looks like
       a scheme not work!
    */
    Curl_dyn_setlen(host, keep);
    portptr++;
    if(!*portptr)
      return has_scheme ? CURLUE_OK : CURLUE_BAD_PORT_NUMBER;

    if(!ISDIGIT(*portptr))
      return CURLUE_BAD_PORT_NUMBER;

    errno = 0;
    port = strtoul(portptr, &rest, 10);  /* Port number must be decimal */

    if(errno || (port > 0xffff) || *rest)
      return CURLUE_BAD_PORT_NUMBER;

    u->portnum = (unsigned short) port;
    /* generate a new port number string to get rid of leading zeroes etc */
    free(u->port);
    u->port = aprintf("%ld", port);
    if(!u->port)
      return CURLUE_OUT_OF_MEMORY;
  }

  return CURLUE_OK;
}

/* this assumes 'hostname' now starts with [ */
static CURLUcode ipv6_parse(struct Curl_URL *u, char *hostname,
                            size_t hlen) /* length of hostname */
{
  size_t len;
  DEBUGASSERT(*hostname == '[');
  if(hlen < 4) /* '[::]' is the shortest possible valid string */
    return CURLUE_BAD_IPV6;
  hostname++;
  hlen -= 2;

  /* only valid IPv6 letters are ok */
  len = strspn(hostname, "0123456789abcdefABCDEF:.");

  if(hlen != len) {
    hlen = len;
    if(hostname[len] == '%') {
      /* this could now be '%[zone id]' */
      char zoneid[16];
      int i = 0;
      char *h = &hostname[len + 1];
      /* pass '25' if present and is a URL encoded percent sign */
      if(!strncmp(h, "25", 2) && h[2] && (h[2] != ']'))
        h += 2;
      while(*h && (*h != ']') && (i < 15))
        zoneid[i++] = *h++;
      if(!i || (']' != *h))
        return CURLUE_BAD_IPV6;
      zoneid[i] = 0;
      u->zoneid = strdup(zoneid);
      if(!u->zoneid)
        return CURLUE_OUT_OF_MEMORY;
      hostname[len] = ']'; /* insert end bracket */
      hostname[len + 1] = 0; /* terminate the hostname */
    }
    else
      return CURLUE_BAD_IPV6;
    /* hostname is fine */
  }

  /* Check the IPv6 address. */
  {
    char dest[16]; /* fits a binary IPv6 address */
    char norm[MAX_IPADR_LEN];
    hostname[hlen] = 0; /* end the address there */
    if(1 != Curl_inet_pton(AF_INET6, hostname, dest))
      return CURLUE_BAD_IPV6;

    /* check if it can be done shorter */
    if(Curl_inet_ntop(AF_INET6, dest, norm, sizeof(norm)) &&
       (strlen(norm) < hlen)) {
      strcpy(hostname, norm);
      hlen = strlen(norm);
      hostname[hlen + 1] = 0;
    }
    hostname[hlen] = ']'; /* restore ending bracket */
  }
  return CURLUE_OK;
}

static CURLUcode hostname_check(struct Curl_URL *u, char *hostname,
                                size_t hlen) /* length of hostname */
{
  size_t len;
  DEBUGASSERT(hostname);

  if(!hlen)
    return CURLUE_NO_HOST;
  else if(hostname[0] == '[')
    return ipv6_parse(u, hostname, hlen);
  else {
    /* letters from the second string are not ok */
    len = strcspn(hostname, " \r\n\t/:#?!@{}[]\\$\'\"^`*<>=;,+&()%");
    if(hlen != len)
      /* hostname with bad content */
      return CURLUE_BAD_HOSTNAME;
  }
  return CURLUE_OK;
}

/*
 * Handle partial IPv4 numerical addresses and different bases, like
 * '16843009', '0x7f', '0x7f.1' '0177.1.1.1' etc.
 *
 * If the given input string is syntactically wrong IPv4 or any part for
 * example is too big, this function returns HOST_NAME.
 *
 * Output the "normalized" version of that input string in plain quad decimal
 * integers.
 *
 * Returns the host type.
 */

#define HOST_ERROR   -1 /* out of memory */

#define HOST_NAME    1
#define HOST_IPV4    2
#define HOST_IPV6    3

static int ipv4_normalize(struct dynbuf *host)
{
  bool done = FALSE;
  int n = 0;
  const char *c = Curl_dyn_ptr(host);
  unsigned long parts[4] = {0, 0, 0, 0};
  CURLcode result = CURLE_OK;

  if(*c == '[')
    return HOST_IPV6;

  errno = 0; /* for strtoul */
  while(!done) {
    char *endp = NULL;
    unsigned long l;
    if(!ISDIGIT(*c))
      /* most importantly this does not allow a leading plus or minus */
      return HOST_NAME;
    l = strtoul(c, &endp, 0);
    if(errno)
      return HOST_NAME;
#if SIZEOF_LONG > 4
    /* a value larger than 32 bits */
    if(l > UINT_MAX)
      return HOST_NAME;
#endif

    parts[n] = l;
    c = endp;

    switch(*c) {
    case '.':
      if(n == 3)
        return HOST_NAME;
      n++;
      c++;
      break;

    case '\0':
      done = TRUE;
      break;

    default:
      return HOST_NAME;
    }
  }

  switch(n) {
  case 0: /* a -- 32 bits */
    Curl_dyn_reset(host);

    result = Curl_dyn_addf(host, "%u.%u.%u.%u",
                           (unsigned int)(parts[0] >> 24),
                           (unsigned int)((parts[0] >> 16) & 0xff),
                           (unsigned int)((parts[0] >> 8) & 0xff),
                           (unsigned int)(parts[0] & 0xff));
    break;
  case 1: /* a.b -- 8.24 bits */
    if((parts[0] > 0xff) || (parts[1] > 0xffffff))
      return HOST_NAME;
    Curl_dyn_reset(host);
    result = Curl_dyn_addf(host, "%u.%u.%u.%u",
                           (unsigned int)(parts[0]),
                           (unsigned int)((parts[1] >> 16) & 0xff),
                           (unsigned int)((parts[1] >> 8) & 0xff),
                           (unsigned int)(parts[1] & 0xff));
    break;
  case 2: /* a.b.c -- 8.8.16 bits */
    if((parts[0] > 0xff) || (parts[1] > 0xff) || (parts[2] > 0xffff))
      return HOST_NAME;
    Curl_dyn_reset(host);
    result = Curl_dyn_addf(host, "%u.%u.%u.%u",
                           (unsigned int)(parts[0]),
                           (unsigned int)(parts[1]),
                           (unsigned int)((parts[2] >> 8) & 0xff),
                           (unsigned int)(parts[2] & 0xff));
    break;
  case 3: /* a.b.c.d -- 8.8.8.8 bits */
    if((parts[0] > 0xff) || (parts[1] > 0xff) || (parts[2] > 0xff) ||
       (parts[3] > 0xff))
      return HOST_NAME;
    Curl_dyn_reset(host);
    result = Curl_dyn_addf(host, "%u.%u.%u.%u",
                           (unsigned int)(parts[0]),
                           (unsigned int)(parts[1]),
                           (unsigned int)(parts[2]),
                           (unsigned int)(parts[3]));
    break;
  }
  if(result)
    return HOST_ERROR;
  return HOST_IPV4;
}

/* if necessary, replace the host content with a URL decoded version */
static CURLUcode urldecode_host(struct dynbuf *host)
{
  char *per = NULL;
  const char *hostname = Curl_dyn_ptr(host);
  per = strchr(hostname, '%');
  if(!per)
    /* nothing to decode */
    return CURLUE_OK;
  else {
    /* encoded */
    size_t dlen;
    char *decoded;
    CURLcode result = Curl_urldecode(hostname, 0, &decoded, &dlen,
                                     REJECT_CTRL);
    if(result)
      return CURLUE_BAD_HOSTNAME;
    Curl_dyn_reset(host);
    result = Curl_dyn_addn(host, decoded, dlen);
    free(decoded);
    if(result)
      return cc2cu(result);
  }

  return CURLUE_OK;
}

static CURLUcode parse_authority(struct Curl_URL *u,
                                 const char *auth, size_t authlen,
                                 unsigned int flags,
                                 struct dynbuf *host,
                                 bool has_scheme)
{
  size_t offset;
  CURLUcode uc;
  CURLcode result;

  /*
   * Parse the login details and strip them out of the hostname.
   */
  uc = parse_hostname_login(u, auth, authlen, flags, &offset);
  if(uc)
    goto out;

  result = Curl_dyn_addn(host, auth + offset, authlen - offset);
  if(result) {
    uc = cc2cu(result);
    goto out;
  }

  uc = Curl_parse_port(u, host, has_scheme);
  if(uc)
    goto out;

  if(!Curl_dyn_len(host))
    return CURLUE_NO_HOST;

  switch(ipv4_normalize(host)) {
  case HOST_IPV4:
    break;
  case HOST_IPV6:
    uc = ipv6_parse(u, Curl_dyn_ptr(host), Curl_dyn_len(host));
    break;
  case HOST_NAME:
    uc = urldecode_host(host);
    if(!uc)
      uc = hostname_check(u, Curl_dyn_ptr(host), Curl_dyn_len(host));
    break;
  case HOST_ERROR:
    uc = CURLUE_OUT_OF_MEMORY;
    break;
  default:
    uc = CURLUE_BAD_HOSTNAME; /* Bad IPv4 address even */
    break;
  }

out:
  return uc;
}

/* used for HTTP/2 server push */
CURLUcode Curl_url_set_authority(CURLU *u, const char *authority)
{
  CURLUcode result;
  struct dynbuf host;

  DEBUGASSERT(authority);
  Curl_dyn_init(&host, CURL_MAX_INPUT_LENGTH);

  result = parse_authority(u, authority, strlen(authority),
                           CURLU_DISALLOW_USER, &host, !!u->scheme);
  if(result)
    Curl_dyn_free(&host);
  else {
    free(u->host);
    u->host = Curl_dyn_ptr(&host);
  }
  return result;
}

/*
 * "Remove Dot Segments"
 * https://datatracker.ietf.org/doc/html/rfc3986#section-5.2.4
 */

/*
 * dedotdotify()
 * @unittest: 1395
 *
 * This function gets a null-terminated path with dot and dotdot sequences
 * passed in and strips them off according to the rules in RFC 3986 section
 * 5.2.4.
 *
 * The function handles a query part ('?' + stuff) appended but it expects
 * that fragments ('#' + stuff) have already been cut off.
 *
 * RETURNS
 *
 * Zero for success and 'out' set to an allocated dedotdotified string.
 */
UNITTEST int dedotdotify(const char *input, size_t clen, char **outp);
UNITTEST int dedotdotify(const char *input, size_t clen, char **outp)
{
  char *outptr;
  const char *endp = &input[clen];
  char *out;

  *outp = NULL;
  /* the path always starts with a slash, and a slash has not dot */
  if((clen < 2) || !memchr(input, '.', clen))
    return 0;

  out = malloc(clen + 1);
  if(!out)
    return 1; /* out of memory */

  *out = 0; /* null-terminates, for inputs like "./" */
  outptr = out;

  do {
    bool dotdot = TRUE;
    if(*input == '.') {
      /*  A. If the input buffer begins with a prefix of "../" or "./", then
          remove that prefix from the input buffer; otherwise, */

      if(!strncmp("./", input, 2)) {
        input += 2;
        clen -= 2;
      }
      else if(!strncmp("../", input, 3)) {
        input += 3;
        clen -= 3;
      }
      /*  D. if the input buffer consists only of "." or "..", then remove
          that from the input buffer; otherwise, */

      else if(!strcmp(".", input) || !strcmp("..", input) ||
              !strncmp(".?", input, 2) || !strncmp("..?", input, 3)) {
        *out = 0;
        break;
      }
      else
        dotdot = FALSE;
    }
    else if(*input == '/') {
      /*  B. if the input buffer begins with a prefix of "/./" or "/.", where
          "."  is a complete path segment, then replace that prefix with "/" in
          the input buffer; otherwise, */
      if(!strncmp("/./", input, 3)) {
        input += 2;
        clen -= 2;
      }
      else if(!strcmp("/.", input) || !strncmp("/.?", input, 3)) {
        *outptr++ = '/';
        *outptr = 0;
        break;
      }

      /*  C. if the input buffer begins with a prefix of "/../" or "/..",
          where ".." is a complete path segment, then replace that prefix with
          "/" in the input buffer and remove the last segment and its
          preceding "/" (if any) from the output buffer; otherwise, */

      else if(!strncmp("/../", input, 4)) {
        input += 3;
        clen -= 3;
        /* remove the last segment from the output buffer */
        while(outptr > out) {
          outptr--;
          if(*outptr == '/')
            break;
        }
        *outptr = 0; /* null-terminate where it stops */
      }
      else if(!strcmp("/..", input) || !strncmp("/..?", input, 4)) {
        /* remove the last segment from the output buffer */
        while(outptr > out) {
          outptr--;
          if(*outptr == '/')
            break;
        }
        *outptr++ = '/';
        *outptr = 0; /* null-terminate where it stops */
        break;
      }
      else
        dotdot = FALSE;
    }
    else
      dotdot = FALSE;

    if(!dotdot) {
      /*  E. move the first path segment in the input buffer to the end of
          the output buffer, including the initial "/" character (if any) and
          any subsequent characters up to, but not including, the next "/"
          character or the end of the input buffer. */

      do {
        *outptr++ = *input++;
        clen--;
      } while(*input && (*input != '/') && (*input != '?'));
      *outptr = 0;
    }

    /* continue until end of path */
  } while(input < endp);

  *outp = out;
  return 0; /* success */
}

static CURLUcode parseurl(const char *url, CURLU *u, unsigned int flags)
{
  const char *path;
  size_t pathlen;
  char *query = NULL;
  char *fragment = NULL;
  char schemebuf[MAX_SCHEME_LEN + 1];
  size_t schemelen = 0;
  size_t urllen;
  CURLUcode result = CURLUE_OK;
  size_t fraglen = 0;
  struct dynbuf host;

  DEBUGASSERT(url);

  Curl_dyn_init(&host, CURL_MAX_INPUT_LENGTH);

  result = junkscan(url, &urllen, flags);
  if(result)
    goto fail;

  schemelen = Curl_is_absolute_url(url, schemebuf, sizeof(schemebuf),
                                   flags & (CURLU_GUESS_SCHEME|
                                            CURLU_DEFAULT_SCHEME));

  /* handle the file: scheme */
  if(schemelen && !strcmp(schemebuf, "file")) {
    bool uncpath = FALSE;
    if(urllen <= 6) {
      /* file:/ is not enough to actually be a complete file: URL */
      result = CURLUE_BAD_FILE_URL;
      goto fail;
    }

    /* path has been allocated large enough to hold this */
    path = (char *)&url[5];
    pathlen = urllen - 5;

    u->scheme = strdup("file");
    if(!u->scheme) {
      result = CURLUE_OUT_OF_MEMORY;
      goto fail;
    }

    /* Extra handling URLs with an authority component (i.e. that start with
     * "file://")
     *
     * We allow omitted hostname (e.g. file:/<path>) -- valid according to
     * RFC 8089, but not the (current) WHAT-WG URL spec.
     */
    if(path[0] == '/' && path[1] == '/') {
      /* swallow the two slashes */
      const char *ptr = &path[2];

      /*
       * According to RFC 8089, a file: URL can be reliably dereferenced if:
       *
       *  o it has no/blank hostname, or
       *
       *  o the hostname matches "localhost" (case-insensitively), or
       *
       *  o the hostname is a FQDN that resolves to this machine, or
       *
       *  o it is an UNC String transformed to an URI (Windows only, RFC 8089
       *    Appendix E.3).
       *
       * For brevity, we only consider URLs with empty, "localhost", or
       * "127.0.0.1" hostnames as local, otherwise as an UNC String.
       *
       * Additionally, there is an exception for URLs with a Windows drive
       * letter in the authority (which was accidentally omitted from RFC 8089
       * Appendix E, but believe me, it was meant to be there. --MK)
       */
      if(ptr[0] != '/' && !STARTS_WITH_URL_DRIVE_PREFIX(ptr)) {
        /* the URL includes a hostname, it must match "localhost" or
           "127.0.0.1" to be valid */
        if(checkprefix("localhost/", ptr) ||
           checkprefix("127.0.0.1/", ptr)) {
          ptr += 9; /* now points to the slash after the host */
        }
        else {
#if defined(_WIN32)
          size_t len;

          /* the hostname, NetBIOS computer name, can not contain disallowed
             chars, and the delimiting slash character must be appended to the
             hostname */
          path = strpbrk(ptr, "/\\:*?\"<>|");
          if(!path || *path != '/') {
            result = CURLUE_BAD_FILE_URL;
            goto fail;
          }

          len = path - ptr;
          if(len) {
            CURLcode code = Curl_dyn_addn(&host, ptr, len);
            if(code) {
              result = cc2cu(code);
              goto fail;
            }
            uncpath = TRUE;
          }

          ptr -= 2; /* now points to the // before the host in UNC */
#else
          /* Invalid file://hostname/, expected localhost or 127.0.0.1 or
             none */
          result = CURLUE_BAD_FILE_URL;
          goto fail;
#endif
        }
      }

      path = ptr;
      pathlen = urllen - (ptr - url);
    }

    if(!uncpath)
      /* no host for file: URLs by default */
      Curl_dyn_reset(&host);

#if !defined(_WIN32) && !defined(MSDOS) && !defined(__CYGWIN__)
    /* Do not allow Windows drive letters when not in Windows.
     * This catches both "file:/c:" and "file:c:" */
    if(('/' == path[0] && STARTS_WITH_URL_DRIVE_PREFIX(&path[1])) ||
       STARTS_WITH_URL_DRIVE_PREFIX(path)) {
      /* File drive letters are only accepted in MS-DOS/Windows */
      result = CURLUE_BAD_FILE_URL;
      goto fail;
    }
#else
    /* If the path starts with a slash and a drive letter, ditch the slash */
    if('/' == path[0] && STARTS_WITH_URL_DRIVE_PREFIX(&path[1])) {
      /* This cannot be done with strcpy, as the memory chunks overlap! */
      path++;
      pathlen--;
    }
#endif

  }
  else {
    /* clear path */
    const char *schemep = NULL;
    const char *hostp;
    size_t hostlen;

    if(schemelen) {
      int i = 0;
      const char *p = &url[schemelen + 1];
      while((*p == '/') && (i < 4)) {
        p++;
        i++;
      }

      schemep = schemebuf;
      if(!Curl_get_scheme_handler(schemep) &&
         !(flags & CURLU_NON_SUPPORT_SCHEME)) {
        result = CURLUE_UNSUPPORTED_SCHEME;
        goto fail;
      }

      if((i < 1) || (i > 3)) {
        /* less than one or more than three slashes */
        result = CURLUE_BAD_SLASHES;
        goto fail;
      }
      hostp = p; /* hostname starts here */
    }
    else {
      /* no scheme! */

      if(!(flags & (CURLU_DEFAULT_SCHEME|CURLU_GUESS_SCHEME))) {
        result = CURLUE_BAD_SCHEME;
        goto fail;
      }
      if(flags & CURLU_DEFAULT_SCHEME)
        schemep = DEFAULT_SCHEME;

      /*
       * The URL was badly formatted, let's try without scheme specified.
       */
      hostp = url;
    }

    if(schemep) {
      u->scheme = strdup(schemep);
      if(!u->scheme) {
        result = CURLUE_OUT_OF_MEMORY;
        goto fail;
      }
    }

    /* find the end of the hostname + port number */
    hostlen = strcspn(hostp, "/?#");
    path = &hostp[hostlen];

    /* this pathlen also contains the query and the fragment */
    pathlen = urllen - (path - url);
    if(hostlen) {

      result = parse_authority(u, hostp, hostlen, flags, &host, schemelen);
      if(result)
        goto fail;

      if((flags & CURLU_GUESS_SCHEME) && !schemep) {
        const char *hostname = Curl_dyn_ptr(&host);
        /* legacy curl-style guess based on hostname */
        if(checkprefix("ftp.", hostname))
          schemep = "ftp";
        else if(checkprefix("dict.", hostname))
          schemep = "dict";
        else if(checkprefix("ldap.", hostname))
          schemep = "ldap";
        else if(checkprefix("imap.", hostname))
          schemep = "imap";
        else if(checkprefix("smtp.", hostname))
          schemep = "smtp";
        else if(checkprefix("pop3.", hostname))
          schemep = "pop3";
        else
          schemep = "http";

        u->scheme = strdup(schemep);
        if(!u->scheme) {
          result = CURLUE_OUT_OF_MEMORY;
          goto fail;
        }
        u->guessed_scheme = TRUE;
      }
    }
    else if(flags & CURLU_NO_AUTHORITY) {
      /* allowed to be empty. */
      if(Curl_dyn_add(&host, "")) {
        result = CURLUE_OUT_OF_MEMORY;
        goto fail;
      }
    }
    else {
      result = CURLUE_NO_HOST;
      goto fail;
    }
  }

  fragment = strchr(path, '#');
  if(fragment) {
    fraglen = pathlen - (fragment - path);
    u->fragment_present = TRUE;
    if(fraglen > 1) {
      /* skip the leading '#' in the copy but include the terminating null */
      if(flags & CURLU_URLENCODE) {
        struct dynbuf enc;
        Curl_dyn_init(&enc, CURL_MAX_INPUT_LENGTH);
        result = urlencode_str(&enc, fragment + 1, fraglen - 1, TRUE, FALSE);
        if(result)
          goto fail;
        u->fragment = Curl_dyn_ptr(&enc);
      }
      else {
        u->fragment = Curl_memdup0(fragment + 1, fraglen - 1);
        if(!u->fragment) {
          result = CURLUE_OUT_OF_MEMORY;
          goto fail;
        }
      }
    }
    /* after this, pathlen still contains the query */
    pathlen -= fraglen;
  }

  query = memchr(path, '?', pathlen);
  if(query) {
    size_t qlen = fragment ? (size_t)(fragment - query) :
      pathlen - (query - path);
    pathlen -= qlen;
    u->query_present = TRUE;
    if(qlen > 1) {
      if(flags & CURLU_URLENCODE) {
        struct dynbuf enc;
        Curl_dyn_init(&enc, CURL_MAX_INPUT_LENGTH);
        /* skip the leading question mark */
        result = urlencode_str(&enc, query + 1, qlen - 1, TRUE, TRUE);
        if(result)
          goto fail;
        u->query = Curl_dyn_ptr(&enc);
      }
      else {
        u->query = Curl_memdup0(query + 1, qlen - 1);
        if(!u->query) {
          result = CURLUE_OUT_OF_MEMORY;
          goto fail;
        }
      }
    }
    else {
      /* single byte query */
      u->query = strdup("");
      if(!u->query) {
        result = CURLUE_OUT_OF_MEMORY;
        goto fail;
      }
    }
  }

  if(pathlen && (flags & CURLU_URLENCODE)) {
    struct dynbuf enc;
    Curl_dyn_init(&enc, CURL_MAX_INPUT_LENGTH);
    result = urlencode_str(&enc, path, pathlen, TRUE, FALSE);
    if(result)
      goto fail;
    pathlen = Curl_dyn_len(&enc);
    path = u->path = Curl_dyn_ptr(&enc);
  }

  if(pathlen <= 1) {
    /* there is no path left or just the slash, unset */
    path = NULL;
  }
  else {
    if(!u->path) {
      u->path = Curl_memdup0(path, pathlen);
      if(!u->path) {
        result = CURLUE_OUT_OF_MEMORY;
        goto fail;
      }
      path = u->path;
    }
    else if(flags & CURLU_URLENCODE)
      /* it might have encoded more than just the path so cut it */
      u->path[pathlen] = 0;

    if(!(flags & CURLU_PATH_AS_IS)) {
      /* remove ../ and ./ sequences according to RFC3986 */
      char *dedot;
      int err = dedotdotify((char *)path, pathlen, &dedot);
      if(err) {
        result = CURLUE_OUT_OF_MEMORY;
        goto fail;
      }
      if(dedot) {
        free(u->path);
        u->path = dedot;
      }
    }
  }

  u->host = Curl_dyn_ptr(&host);

  return result;
fail:
  Curl_dyn_free(&host);
  free_urlhandle(u);
  return result;
}

/*
 * Parse the URL and, if successful, replace everything in the Curl_URL struct.
 */
static CURLUcode parseurl_and_replace(const char *url, CURLU *u,
                                      unsigned int flags)
{
  CURLUcode result;
  CURLU tmpurl;
  memset(&tmpurl, 0, sizeof(tmpurl));
  result = parseurl(url, &tmpurl, flags);
  if(!result) {
    free_urlhandle(u);
    *u = tmpurl;
  }
  return result;
}

/*
 */
CURLU *curl_url(void)
{
  return calloc(1, sizeof(struct Curl_URL));
}

void curl_url_cleanup(CURLU *u)
{
  if(u) {
    free_urlhandle(u);
    free(u);
  }
}

#define DUP(dest, src, name)                    \
  do {                                          \
    if(src->name) {                             \
      dest->name = strdup(src->name);           \
      if(!dest->name)                           \
        goto fail;                              \
    }                                           \
  } while(0)

CURLU *curl_url_dup(const CURLU *in)
{
  struct Curl_URL *u = calloc(1, sizeof(struct Curl_URL));
  if(u) {
    DUP(u, in, scheme);
    DUP(u, in, user);
    DUP(u, in, password);
    DUP(u, in, options);
    DUP(u, in, host);
    DUP(u, in, port);
    DUP(u, in, path);
    DUP(u, in, query);
    DUP(u, in, fragment);
    DUP(u, in, zoneid);
    u->portnum = in->portnum;
    u->fragment_present = in->fragment_present;
    u->query_present = in->query_present;
  }
  return u;
fail:
  curl_url_cleanup(u);
  return NULL;
}

CURLUcode curl_url_get(const CURLU *u, CURLUPart what,
                       char **part, unsigned int flags)
{
  const char *ptr;
  CURLUcode ifmissing = CURLUE_UNKNOWN_PART;
  char portbuf[7];
  bool urldecode = (flags & CURLU_URLDECODE)?1:0;
  bool urlencode = (flags & CURLU_URLENCODE)?1:0;
  bool punycode = FALSE;
  bool depunyfy = FALSE;
  bool plusdecode = FALSE;
  (void)flags;
  if(!u)
    return CURLUE_BAD_HANDLE;
  if(!part)
    return CURLUE_BAD_PARTPOINTER;
  *part = NULL;

  switch(what) {
  case CURLUPART_SCHEME:
    ptr = u->scheme;
    ifmissing = CURLUE_NO_SCHEME;
    urldecode = FALSE; /* never for schemes */
    if((flags & CURLU_NO_GUESS_SCHEME) && u->guessed_scheme)
      return CURLUE_NO_SCHEME;
    break;
  case CURLUPART_USER:
    ptr = u->user;
    ifmissing = CURLUE_NO_USER;
    break;
  case CURLUPART_PASSWORD:
    ptr = u->password;
    ifmissing = CURLUE_NO_PASSWORD;
    break;
  case CURLUPART_OPTIONS:
    ptr = u->options;
    ifmissing = CURLUE_NO_OPTIONS;
    break;
  case CURLUPART_HOST:
    ptr = u->host;
    ifmissing = CURLUE_NO_HOST;
    punycode = (flags & CURLU_PUNYCODE)?1:0;
    depunyfy = (flags & CURLU_PUNY2IDN)?1:0;
    break;
  case CURLUPART_ZONEID:
    ptr = u->zoneid;
    ifmissing = CURLUE_NO_ZONEID;
    break;
  case CURLUPART_PORT:
    ptr = u->port;
    ifmissing = CURLUE_NO_PORT;
    urldecode = FALSE; /* never for port */
    if(!ptr && (flags & CURLU_DEFAULT_PORT) && u->scheme) {
      /* there is no stored port number, but asked to deliver
         a default one for the scheme */
      const struct Curl_handler *h = Curl_get_scheme_handler(u->scheme);
      if(h) {
        msnprintf(portbuf, sizeof(portbuf), "%u", h->defport);
        ptr = portbuf;
      }
    }
    else if(ptr && u->scheme) {
      /* there is a stored port number, but ask to inhibit if
         it matches the default one for the scheme */
      const struct Curl_handler *h = Curl_get_scheme_handler(u->scheme);
      if(h && (h->defport == u->portnum) &&
         (flags & CURLU_NO_DEFAULT_PORT))
        ptr = NULL;
    }
    break;
  case CURLUPART_PATH:
    ptr = u->path;
    if(!ptr)
      ptr = "/";
    break;
  case CURLUPART_QUERY:
    ptr = u->query;
    ifmissing = CURLUE_NO_QUERY;
    plusdecode = urldecode;
    if(ptr && !ptr[0] && !(flags & CURLU_GET_EMPTY))
      /* there was a blank query and the user do not ask for it */
      ptr = NULL;
    break;
  case CURLUPART_FRAGMENT:
    ptr = u->fragment;
    ifmissing = CURLUE_NO_FRAGMENT;
    if(!ptr && u->fragment_present && flags & CURLU_GET_EMPTY)
      /* there was a blank fragment and the user asks for it */
      ptr = "";
    break;
  case CURLUPART_URL: {
    char *url;
    char *scheme;
    char *options = u->options;
    char *port = u->port;
    char *allochost = NULL;
    bool show_fragment =
      u->fragment || (u->fragment_present && flags & CURLU_GET_EMPTY);
    bool show_query =
      (u->query && u->query[0]) ||
      (u->query_present && flags & CURLU_GET_EMPTY);
    punycode = (flags & CURLU_PUNYCODE)?1:0;
    depunyfy = (flags & CURLU_PUNY2IDN)?1:0;
    if(u->scheme && strcasecompare("file", u->scheme)) {
      url = aprintf("file://%s%s%s",
                    u->path,
                    show_fragment ? "#": "",
                    u->fragment ? u->fragment : "");
    }
    else if(!u->host)
      return CURLUE_NO_HOST;
    else {
      const struct Curl_handler *h = NULL;
      char schemebuf[MAX_SCHEME_LEN + 5];
      if(u->scheme)
        scheme = u->scheme;
      else if(flags & CURLU_DEFAULT_SCHEME)
        scheme = (char *) DEFAULT_SCHEME;
      else
        return CURLUE_NO_SCHEME;

      h = Curl_get_scheme_handler(scheme);
      if(!port && (flags & CURLU_DEFAULT_PORT)) {
        /* there is no stored port number, but asked to deliver
           a default one for the scheme */
        if(h) {
          msnprintf(portbuf, sizeof(portbuf), "%u", h->defport);
          port = portbuf;
        }
      }
      else if(port) {
        /* there is a stored port number, but asked to inhibit if it matches
           the default one for the scheme */
        if(h && (h->defport == u->portnum) &&
           (flags & CURLU_NO_DEFAULT_PORT))
          port = NULL;
      }

      if(h && !(h->flags & PROTOPT_URLOPTIONS))
        options = NULL;

      if(u->host[0] == '[') {
        if(u->zoneid) {
          /* make it '[ host %25 zoneid ]' */
          struct dynbuf enc;
          size_t hostlen = strlen(u->host);
          Curl_dyn_init(&enc, CURL_MAX_INPUT_LENGTH);
          if(Curl_dyn_addf(&enc, "%.*s%%25%s]", (int)hostlen - 1, u->host,
                           u->zoneid))
            return CURLUE_OUT_OF_MEMORY;
          allochost = Curl_dyn_ptr(&enc);
        }
      }
      else if(urlencode) {
        allochost = curl_easy_escape(NULL, u->host, 0);
        if(!allochost)
          return CURLUE_OUT_OF_MEMORY;
      }
      else if(punycode) {
        if(!Curl_is_ASCII_name(u->host)) {
#ifndef USE_IDN
          return CURLUE_LACKS_IDN;
#else
          CURLcode result = Curl_idn_decode(u->host, &allochost);
          if(result)
            return (result == CURLE_OUT_OF_MEMORY) ?
              CURLUE_OUT_OF_MEMORY : CURLUE_BAD_HOSTNAME;
#endif
        }
      }
      else if(depunyfy) {
        if(Curl_is_ASCII_name(u->host) && !strncmp("xn--", u->host, 4)) {
#ifndef USE_IDN
          return CURLUE_LACKS_IDN;
#else
          CURLcode result = Curl_idn_encode(u->host, &allochost);
          if(result)
            /* this is the most likely error */
            return (result == CURLE_OUT_OF_MEMORY) ?
              CURLUE_OUT_OF_MEMORY : CURLUE_BAD_HOSTNAME;
#endif
        }
      }

      if(!(flags & CURLU_NO_GUESS_SCHEME) || !u->guessed_scheme)
        msnprintf(schemebuf, sizeof(schemebuf), "%s://", scheme);
      else
        schemebuf[0] = 0;

      url = aprintf("%s%s%s%s%s%s%s%s%s%s%s%s%s%s%s",
                    schemebuf,
                    u->user ? u->user : "",
                    u->password ? ":": "",
                    u->password ? u->password : "",
                    options ? ";" : "",
                    options ? options : "",
                    (u->user || u->password || options) ? "@": "",
                    allochost ? allochost : u->host,
                    port ? ":": "",
                    port ? port : "",
                    u->path ? u->path : "/",
                    show_query ? "?": "",
                    u->query ? u->query : "",
                    show_fragment ? "#": "",
                    u->fragment? u->fragment : "");
      free(allochost);
    }
    if(!url)
      return CURLUE_OUT_OF_MEMORY;
    *part = url;
    return CURLUE_OK;
  }
  default:
    ptr = NULL;
    break;
  }
  if(ptr) {
    size_t partlen = strlen(ptr);
    size_t i = 0;
    *part = Curl_memdup0(ptr, partlen);
    if(!*part)
      return CURLUE_OUT_OF_MEMORY;
    if(plusdecode) {
      /* convert + to space */
      char *plus = *part;
      for(i = 0; i < partlen; ++plus, i++) {
        if(*plus == '+')
          *plus = ' ';
      }
    }
    if(urldecode) {
      char *decoded;
      size_t dlen;
      /* this unconditional rejection of control bytes is documented
         API behavior */
      CURLcode res = Curl_urldecode(*part, 0, &decoded, &dlen, REJECT_CTRL);
      free(*part);
      if(res) {
        *part = NULL;
        return CURLUE_URLDECODE;
      }
      *part = decoded;
      partlen = dlen;
    }
    if(urlencode) {
      struct dynbuf enc;
      CURLUcode uc;
      Curl_dyn_init(&enc, CURL_MAX_INPUT_LENGTH);
      uc = urlencode_str(&enc, *part, partlen, TRUE, what == CURLUPART_QUERY);
      if(uc)
        return uc;
      free(*part);
      *part = Curl_dyn_ptr(&enc);
    }
    else if(punycode) {
      if(!Curl_is_ASCII_name(u->host)) {
#ifndef USE_IDN
        return CURLUE_LACKS_IDN;
#else
        char *allochost;
        CURLcode result = Curl_idn_decode(*part, &allochost);
        if(result)
          return (result == CURLE_OUT_OF_MEMORY) ?
            CURLUE_OUT_OF_MEMORY : CURLUE_BAD_HOSTNAME;
        free(*part);
        *part = allochost;
#endif
      }
    }
    else if(depunyfy) {
      if(Curl_is_ASCII_name(u->host)  && !strncmp("xn--", u->host, 4)) {
#ifndef USE_IDN
        return CURLUE_LACKS_IDN;
#else
        char *allochost;
        CURLcode result = Curl_idn_encode(*part, &allochost);
        if(result)
          return (result == CURLE_OUT_OF_MEMORY) ?
            CURLUE_OUT_OF_MEMORY : CURLUE_BAD_HOSTNAME;
        free(*part);
        *part = allochost;
#endif
      }
    }

    return CURLUE_OK;
  }
  else
    return ifmissing;
}

CURLUcode curl_url_set(CURLU *u, CURLUPart what,
                       const char *part, unsigned int flags)
{
  char **storep = NULL;
  bool urlencode = (flags & CURLU_URLENCODE)? 1 : 0;
  bool plusencode = FALSE;
  bool urlskipslash = FALSE;
  bool leadingslash = FALSE;
  bool appendquery = FALSE;
  bool equalsencode = FALSE;
  size_t nalloc;

  if(!u)
    return CURLUE_BAD_HANDLE;
  if(!part) {
    /* setting a part to NULL clears it */
    switch(what) {
    case CURLUPART_URL:
      break;
    case CURLUPART_SCHEME:
      storep = &u->scheme;
      u->guessed_scheme = FALSE;
      break;
    case CURLUPART_USER:
      storep = &u->user;
      break;
    case CURLUPART_PASSWORD:
      storep = &u->password;
      break;
    case CURLUPART_OPTIONS:
      storep = &u->options;
      break;
    case CURLUPART_HOST:
      storep = &u->host;
      break;
    case CURLUPART_ZONEID:
      storep = &u->zoneid;
      break;
    case CURLUPART_PORT:
      u->portnum = 0;
      storep = &u->port;
      break;
    case CURLUPART_PATH:
      storep = &u->path;
      break;
    case CURLUPART_QUERY:
      storep = &u->query;
      u->query_present = FALSE;
      break;
    case CURLUPART_FRAGMENT:
      storep = &u->fragment;
      u->fragment_present = FALSE;
      break;
    default:
      return CURLUE_UNKNOWN_PART;
    }
    if(storep && *storep) {
      Curl_safefree(*storep);
    }
    else if(!storep) {
      free_urlhandle(u);
      memset(u, 0, sizeof(struct Curl_URL));
    }
    return CURLUE_OK;
  }

  nalloc = strlen(part);
  if(nalloc > CURL_MAX_INPUT_LENGTH)
    /* excessive input length */
    return CURLUE_MALFORMED_INPUT;

  switch(what) {
  case CURLUPART_SCHEME: {
    size_t plen = strlen(part);
    const char *s = part;
    if((plen > MAX_SCHEME_LEN) || (plen < 1))
      /* too long or too short */
      return CURLUE_BAD_SCHEME;
   /* verify that it is a fine scheme */
    if(!(flags & CURLU_NON_SUPPORT_SCHEME) && !Curl_get_scheme_handler(part))
      return CURLUE_UNSUPPORTED_SCHEME;
    storep = &u->scheme;
    urlencode = FALSE; /* never */
    if(ISALPHA(*s)) {
      /* ALPHA *( ALPHA / DIGIT / "+" / "-" / "." ) */
      while(--plen) {
        if(ISALNUM(*s) || (*s == '+') || (*s == '-') || (*s == '.'))
          s++; /* fine */
        else
          return CURLUE_BAD_SCHEME;
      }
    }
    else
      return CURLUE_BAD_SCHEME;
    u->guessed_scheme = FALSE;
    break;
  }
  case CURLUPART_USER:
    storep = &u->user;
    break;
  case CURLUPART_PASSWORD:
    storep = &u->password;
    break;
  case CURLUPART_OPTIONS:
    storep = &u->options;
    break;
  case CURLUPART_HOST:
    storep = &u->host;
    Curl_safefree(u->zoneid);
    break;
  case CURLUPART_ZONEID:
    storep = &u->zoneid;
    break;
  case CURLUPART_PORT:
    if(!ISDIGIT(part[0]))
      /* not a number */
      return CURLUE_BAD_PORT_NUMBER;
    else {
      char *tmp;
      char *endp;
      unsigned long port;
      errno = 0;
      port = strtoul(part, &endp, 10);  /* must be decimal */
      if(errno || (port > 0xffff) || *endp)
        /* weirdly provided number, not good! */
        return CURLUE_BAD_PORT_NUMBER;
      tmp = strdup(part);
      if(!tmp)
        return CURLUE_OUT_OF_MEMORY;
      free(u->port);
      u->port = tmp;
      u->portnum = (unsigned short)port;
      return CURLUE_OK;
    }
  case CURLUPART_PATH:
    urlskipslash = TRUE;
    leadingslash = TRUE; /* enforce */
    storep = &u->path;
    break;
  case CURLUPART_QUERY:
    plusencode = urlencode;
    appendquery = (flags & CURLU_APPENDQUERY)?1:0;
    equalsencode = appendquery;
    storep = &u->query;
    u->query_present = TRUE;
    break;
  case CURLUPART_FRAGMENT:
    storep = &u->fragment;
    u->fragment_present = TRUE;
    break;
  case CURLUPART_URL: {
    /*
     * Allow a new URL to replace the existing (if any) contents.
     *
     * If the existing contents is enough for a URL, allow a relative URL to
     * replace it.
     */
    CURLcode result;
    CURLUcode uc;
    char *oldurl;
    char *redired_url;

    if(!nalloc)
      /* a blank URL is not a valid URL */
      return CURLUE_MALFORMED_INPUT;

    /* if the new thing is absolute or the old one is not
     * (we could not get an absolute URL in 'oldurl'),
     * then replace the existing with the new. */
    if(Curl_is_absolute_url(part, NULL, 0,
                            flags & (CURLU_GUESS_SCHEME|
                                     CURLU_DEFAULT_SCHEME))
       || curl_url_get(u, CURLUPART_URL, &oldurl, flags)) {
      return parseurl_and_replace(part, u, flags);
    }

    /* apply the relative part to create a new URL
     * and replace the existing one with it. */
    result = concat_url(oldurl, part, &redired_url);
    free(oldurl);
    if(result)
      return cc2cu(result);

    uc = parseurl_and_replace(redired_url, u, flags);
    free(redired_url);
    return uc;
  }
  default:
    return CURLUE_UNKNOWN_PART;
  }
  DEBUGASSERT(storep);
  {
    const char *newp;
    struct dynbuf enc;
    Curl_dyn_init(&enc, nalloc * 3 + 1 + leadingslash);

    if(leadingslash && (part[0] != '/')) {
      CURLcode result = Curl_dyn_addn(&enc, "/", 1);
      if(result)
        return cc2cu(result);
    }
    if(urlencode) {
      const unsigned char *i;

      for(i = (const unsigned char *)part; *i; i++) {
        CURLcode result;
        if((*i == ' ') && plusencode) {
          result = Curl_dyn_addn(&enc, "+", 1);
          if(result)
            return CURLUE_OUT_OF_MEMORY;
        }
        else if(ISUNRESERVED(*i) ||
                ((*i == '/') && urlskipslash) ||
                ((*i == '=') && equalsencode)) {
          if((*i == '=') && equalsencode)
            /* only skip the first equals sign */
            equalsencode = FALSE;
          result = Curl_dyn_addn(&enc, i, 1);
          if(result)
            return cc2cu(result);
        }
        else {
          char out[3]={'%'};
          out[1] = hexdigits[*i>>4];
          out[2] = hexdigits[*i & 0xf];
          result = Curl_dyn_addn(&enc, out, 3);
          if(result)
            return cc2cu(result);
        }
      }
    }
    else {
      char *p;
      CURLcode result = Curl_dyn_add(&enc, part);
      if(result)
        return cc2cu(result);
      p = Curl_dyn_ptr(&enc);
      while(*p) {
        /* make sure percent encoded are lower case */
        if((*p == '%') && ISXDIGIT(p[1]) && ISXDIGIT(p[2]) &&
           (ISUPPER(p[1]) || ISUPPER(p[2]))) {
          p[1] = Curl_raw_tolower(p[1]);
          p[2] = Curl_raw_tolower(p[2]);
          p += 3;
        }
        else
          p++;
      }
    }
    newp = Curl_dyn_ptr(&enc);

    if(appendquery && newp) {
      /* Append the 'newp' string onto the old query. Add a '&' separator if
         none is present at the end of the existing query already */

      size_t querylen = u->query ? strlen(u->query) : 0;
      bool addamperand = querylen && (u->query[querylen -1] != '&');
      if(querylen) {
        struct dynbuf qbuf;
        Curl_dyn_init(&qbuf, CURL_MAX_INPUT_LENGTH);

        if(Curl_dyn_addn(&qbuf, u->query, querylen)) /* add original query */
          goto nomem;

        if(addamperand) {
          if(Curl_dyn_addn(&qbuf, "&", 1))
            goto nomem;
        }
        if(Curl_dyn_add(&qbuf, newp))
          goto nomem;
        Curl_dyn_free(&enc);
        free(*storep);
        *storep = Curl_dyn_ptr(&qbuf);
        return CURLUE_OK;
nomem:
        Curl_dyn_free(&enc);
        return CURLUE_OUT_OF_MEMORY;
      }
    }

    else if(what == CURLUPART_HOST) {
      size_t n = Curl_dyn_len(&enc);
      if(!n && (flags & CURLU_NO_AUTHORITY)) {
        /* Skip hostname check, it is allowed to be empty. */
      }
      else {
        bool bad = FALSE;
        if(!n)
          bad = TRUE; /* empty hostname is not okay */
        else if(!urlencode) {
          /* if the host name part was not URL encoded here, it was set ready
             URL encoded so we need to decode it to check */
          size_t dlen;
          char *decoded = NULL;
          CURLcode result =
            Curl_urldecode(newp, n, &decoded, &dlen, REJECT_CTRL);
          if(result || hostname_check(u, decoded, dlen))
            bad = TRUE;
          free(decoded);
        }
        else if(hostname_check(u, (char *)newp, n))
          bad = TRUE;
        if(bad) {
          Curl_dyn_free(&enc);
          return CURLUE_BAD_HOSTNAME;
        }
      }
    }

    free(*storep);
    *storep = (char *)newp;
  }
  return CURLUE_OK;
}