aboutsummaryrefslogtreecommitdiff
path: root/lib/urlapi.c
blob: 6919ff1bd7edb1f24bfd70f79062936692abedf5 (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
/***************************************************************************
 *                                  _   _ ____  _
 *  Project                     ___| | | |  _ \| |
 *                             / __| | | | |_) | |
 *                            | (__| |_| |  _ <| |___
 *                             \___|\___/|_| \_\_____|
 *
 * Copyright (C) 1998 - 2018, 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.haxx.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.
 *
 ***************************************************************************/

#include "curl_setup.h"

#include "urldata.h"
#include "urlapi-int.h"
#include "strcase.h"
#include "dotdot.h"
#include "url.h"
#include "escape.h"
#include "curl_ctype.h"

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

  /* MSDOS/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] == ':'))

  /* MSDOS/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))

/* 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 *port;
  char *path;
  char *query;
  char *fragment;

  char *scratch; /* temporary scratch area */
  long portnum; /* the numerical version */
};

#define DEFAULT_SCHEME "https"

#ifdef DEBUGBUILD
#define UNITTEST
#else
#define UNITTEST static
#endif

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->port);
  free(u->path);
  free(u->query);
  free(u->fragment);
  free(u->scratch);
}

/* move the full contents of one handle onto another and
   free the original */
static void mv_urlhandle(struct Curl_URL *from,
                         struct Curl_URL *to)
{
  free_urlhandle(to);
  *to = *from;
  free(from);
}

/*
 * Find the separator at the end of the host name, or the '?' in cases like
 * http://www.url.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;
}

/*
 * Decide in an encoding-independent manner whether a character in an
 * URL must be escaped. The same criterion must be used in strlen_url()
 * and strcpy_url().
 */
static bool urlchar_needs_escaping(int c)
{
    return !(ISCNTRL(c) || ISSPACE(c) || ISGRAPH(c));
}

/*
 * strlen_url() returns the length of the given URL if the spaces within the
 * URL were properly URL encoded.
 * URL encoding should be skipped for host names, otherwise IDN resolution
 * will fail.
 */
size_t Curl_strlen_url(const char *url, bool relative)
{
  const unsigned char *ptr;
  size_t newlen = 0;
  bool left = TRUE; /* left side of the ? */
  const unsigned char *host_sep = (const unsigned char *) url;

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

  for(ptr = (unsigned char *)url; *ptr; ptr++) {

    if(ptr < host_sep) {
      ++newlen;
      continue;
    }

    switch(*ptr) {
    case '?':
      left = FALSE;
      /* FALLTHROUGH */
    default:
      if(urlchar_needs_escaping(*ptr))
        newlen += 2;
      newlen++;
      break;
    case ' ':
      if(left)
        newlen += 3;
      else
        newlen++;
      break;
    }
  }
  return newlen;
}

/* strcpy_url() copies a url to a output buffer and URL-encodes the spaces in
 * the source URL accordingly.
 * URL encoding should be skipped for host names, otherwise IDN resolution
 * will fail.
 */
void Curl_strcpy_url(char *output, const char *url, bool relative)
{
  /* we must add this with whitespace-replacing */
  bool left = TRUE;
  const unsigned char *iptr;
  char *optr = output;
  const unsigned char *host_sep = (const unsigned char *) url;

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

  for(iptr = (unsigned char *)url;    /* read from here */
      *iptr;         /* until zero byte */
      iptr++) {

    if(iptr < host_sep) {
      *optr++ = *iptr;
      continue;
    }

    switch(*iptr) {
    case '?':
      left = FALSE;
      /* FALLTHROUGH */
    default:
      if(urlchar_needs_escaping(*iptr)) {
        msnprintf(optr, 4, "%%%02x", *iptr);
        optr += 3;
      }
      else
        *optr++=*iptr;
      break;
    case ' ':
      if(left) {
        *optr++='%'; /* add a '%' */
        *optr++='2'; /* add a '2' */
        *optr++='0'; /* add a '0' */
      }
      else
        *optr++='+'; /* add a '+' here */
      break;
    }
  }
  *optr = 0; /* zero terminate output buffer */

}

/*
 * Returns true if the given URL is absolute (as opposed to relative) within
 * the buffer size. Returns the scheme in the buffer if TRUE and 'buf' is
 * non-NULL.
 */
bool Curl_is_absolute_url(const char *url, char *buf, size_t buflen)
{
  size_t i;
#ifdef WIN32
  if(STARTS_WITH_DRIVE_PREFIX(url))
    return FALSE;
#endif
  for(i = 0; i < buflen && url[i]; ++i) {
    char s = url[i];
    if((s == ':') && (url[i + 1] == '/')) {
      if(buf)
        buf[i] = 0;
      return TRUE;
    }
    /* RFC 3986 3.1 explains:
      scheme      = ALPHA *( ALPHA / DIGIT / "+" / "-" / "." )
    */
    else if(ISALNUM(s) || (s == '+') || (s == '-') || (s == '.') ) {
      if(buf)
        buf[i] = (char)TOLOWER(s);
    }
    else
      break;
  }
  return FALSE;
}

/*
 * 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).
 */
char *Curl_concat_url(const char *base, const char *relurl)
{
  /***
   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...
  */
  char *newest;
  char *protsep;
  char *pathsep;
  size_t newlen;
  bool host_changed = FALSE;

  const char *useurl = relurl;
  size_t urllen;

  /* we must make our own copy of the URL to play with, as it may
     point to read-only data */
  char *url_clone = strdup(base);

  if(!url_clone)
    return NULL; /* skip out of this NOW */

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

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

    /* First we need to find out if there's 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's one
       available, or if the new URL is just a query string (starts with a
       '?')  we append the new one at the end of the entire currently worked
       out URL */
    if(useurl[0] != '?') {
      pathsep = strrchr(protsep, '/');
      if(pathsep)
        *pathsep = 0;
    }

    /* Check if there's any slash after the host name, 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 {
    /* We got a new absolute path for this server */

    if((relurl[0] == '/') && (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.url.com?dir=/home/daniel" we must not use the first
           slash, if there's 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.url.com?id=2380" which doesn't
           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;
      }
    }
  }

  /* If the new part contains a space, this is a mighty stupid redirect
     but we still make an effort to do "right". To the left of a '?'
     letter we replace each space with %20 while it is replaced with '+'
     on the right side of the '?' letter.
  */
  newlen = Curl_strlen_url(useurl, !host_changed);

  urllen = strlen(url_clone);

  newest = malloc(urllen + 1 + /* possible slash */
                  newlen + 1 /* zero byte */);

  if(!newest) {
    free(url_clone); /* don't leak this */
    return NULL;
  }

  /* copy over the root url part */
  memcpy(newest, url_clone, urllen);

  /* check if we need to append a slash */
  if(('/' == useurl[0]) || (protsep && !*protsep) || ('?' == useurl[0]))
    ;
  else
    newest[urllen++]='/';

  /* then append the new piece on the right side */
  Curl_strcpy_url(&newest[urllen], useurl, !host_changed);

  free(url_clone);

  return newest;
}

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

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

  char *ptr = strchr(*hostname, '@');
  char *login = *hostname;

  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 */
  *hostname = ++ptr;

  /* 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_MALFORMED_INPUT;
    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;
    }

    u->user = userp;
  }

  if(passwdp)
    u->password = passwdp;

  if(optionsp)
    u->options = optionsp;

  return CURLUE_OK;
  out:

  free(userp);
  free(passwdp);
  free(optionsp);

  return result;
}

UNITTEST CURLUcode Curl_parse_port(struct Curl_URL *u, char *hostname)
{
  char *portptr = NULL;
  char endbracket;
  int len;

  /*
   * Find the end of an IPv6 address, either on the ']' ending bracket or
   * a percent-encoded zone index.
   */
  if(1 == sscanf(hostname, "[%*45[0123456789abcdefABCDEF:.]%c%n",
                 &endbracket, &len)) {
    if(']' == endbracket)
      portptr = &hostname[len];
    else if('%' == endbracket) {
      int zonelen = len;
      if(1 == sscanf(hostname + zonelen, "25%*[^]]]%c%n", &endbracket, &len))
        portptr = &hostname[--zonelen + len];
      else
        return CURLUE_MALFORMED_INPUT;
    }
    else
      return CURLUE_MALFORMED_INPUT;

    /* this is a RFC2732-style specified IP-address */
    if(portptr && *portptr) {
      if(*portptr != ':')
        return CURLUE_MALFORMED_INPUT;
    }
    else
      portptr = NULL;
  }
  else
    portptr = strchr(hostname, ':');

  if(portptr) {
    char *rest;
    long port;
    char portbuf[7];

    /* Browser behavior adaptation. If there's 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. */
    if(!portptr[1]) {
      *portptr = '\0';
      return CURLUE_OK;
    }

    if(!ISDIGIT(portptr[1]))
      return CURLUE_BAD_PORT_NUMBER;

    port = strtol(portptr + 1, &rest, 10);  /* Port number must be decimal */

    if((port <= 0) || (port > 0xffff))
      /* Single unix standard says port numbers are 16 bits long, but we don't
         treat port zero as OK. */
      return CURLUE_BAD_PORT_NUMBER;

    if(rest[0])
      return CURLUE_BAD_PORT_NUMBER;

    *portptr++ = '\0'; /* cut off the name there */
    *rest = 0;
    /* generate a new port number string to get rid of leading zeroes etc */
    msnprintf(portbuf, sizeof(portbuf), "%ld", port);
    u->portnum = port;
    u->port = strdup(portbuf);
    if(!u->port)
      return CURLUE_OUT_OF_MEMORY;
  }

  return CURLUE_OK;
}

/* scan for byte values < 31 or 127 */
static CURLUcode junkscan(char *part)
{
  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 /* zero terminate */
  };
  if(part) {
    size_t n = strlen(part);
    size_t nfine = strcspn(part, badbytes);
    if(nfine != n)
      /* since we don't know which part is scanned, return a generic error
         code */
      return CURLUE_MALFORMED_INPUT;
  }
  return CURLUE_OK;
}

static CURLUcode hostname_check(char *hostname, unsigned int flags)
{
  const char *l = NULL; /* accepted characters */
  size_t len;
  size_t hlen = strlen(hostname);
  (void)flags;

  if(hostname[0] == '[') {
    hostname++;
    l = "0123456789abcdefABCDEF::.%";
    hlen -= 2;
  }

  if(l) {
    /* only valid letters are ok */
    len = strspn(hostname, l);
    if(hlen != len)
      /* hostname with bad content */
      return CURLUE_MALFORMED_INPUT;
  }
  else {
    /* letters from the second string is not ok */
    len = strcspn(hostname, " ");
    if(hlen != len)
      /* hostname with bad content */
      return CURLUE_MALFORMED_INPUT;
  }
  return CURLUE_OK;
}

#define HOSTNAME_END(x) (((x) == '/') || ((x) == '?') || ((x) == '#'))

static CURLUcode seturl(const char *url, CURLU *u, unsigned int flags)
{
  char *path;
  bool path_alloced = FALSE;
  char *hostname;
  char *query = NULL;
  char *fragment = NULL;
  CURLUcode result;
  bool url_has_scheme = FALSE;
  char schemebuf[MAX_SCHEME_LEN];
  char *schemep = NULL;
  size_t schemelen = 0;
  size_t urllen;
  const struct Curl_handler *h = NULL;

  if(!url)
    return CURLUE_MALFORMED_INPUT;

  /*************************************************************
   * Parse the URL.
   ************************************************************/
  /* allocate scratch area */
  urllen = strlen(url);
  path = u->scratch = malloc(urllen * 2 + 2);
  if(!path)
    return CURLUE_OUT_OF_MEMORY;

  hostname = &path[urllen + 1];
  hostname[0] = 0;

  if(Curl_is_absolute_url(url, schemebuf, sizeof(schemebuf))) {
    url_has_scheme = TRUE;
    schemelen = strlen(schemebuf);
  }

  /* handle the file: scheme */
  if(url_has_scheme && strcasecompare(schemebuf, "file")) {
    /* path has been allocated large enough to hold this */
    strcpy(path, &url[5]);

    hostname = NULL; /* no host for file: URLs */
    u->scheme = strdup("file");
    if(!u->scheme)
      return CURLUE_OUT_OF_MEMORY;

    /* 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 */
      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.
       *
       * For brevity, we only consider URLs with empty, "localhost", or
       * "127.0.0.1" hostnames as local.
       *
       * 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 host name, it must match "localhost" or
           "127.0.0.1" to be valid */
        if(!checkprefix("localhost/", ptr) &&
           !checkprefix("127.0.0.1/", ptr)) {
          /* Invalid file://hostname/, expected localhost or 127.0.0.1 or
             none */
          return CURLUE_MALFORMED_INPUT;
        }
        ptr += 9; /* now points to the slash after the host */
      }

      path = ptr;
    }

#if !defined(MSDOS) && !defined(WIN32) && !defined(__CYGWIN__)
    /* Don't 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 MSDOS/Windows */
      return CURLUE_MALFORMED_INPUT;
    }
#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! */
      memmove(path, &path[1], strlen(&path[1]) + 1);
    }
#endif

  }
  else {
    /* clear path */
    const char *p;
    const char *hostp;
    size_t len;
    path[0] = 0;

    if(url_has_scheme) {
      int i = 0;
      p = &url[schemelen + 1];
      while(p && (*p == '/') && (i < 4)) {
        p++;
        i++;
      }
      if((i < 1) || (i>3))
        /* less than one or more than three slashes */
        return CURLUE_MALFORMED_INPUT;

      schemep = schemebuf;
      if(!Curl_builtin_scheme(schemep) &&
         !(flags & CURLU_NON_SUPPORT_SCHEME))
        return CURLUE_UNSUPPORTED_SCHEME;

      if(junkscan(schemep))
        return CURLUE_MALFORMED_INPUT;
    }
    else {
      /* no scheme! */

      if(!(flags & (CURLU_DEFAULT_SCHEME|CURLU_GUESS_SCHEME)))
        return CURLUE_MALFORMED_INPUT;
      if(flags & CURLU_DEFAULT_SCHEME)
        schemep = (char *) DEFAULT_SCHEME;

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

    while(*p && !HOSTNAME_END(*p)) /* find end of host name */
      p++;

    len = p - hostp;
    if(!len)
      return CURLUE_MALFORMED_INPUT;

    memcpy(hostname, hostp, len);
    hostname[len] = 0;

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

    len = strlen(p);
    memcpy(path, p, len);
    path[len] = 0;

    u->scheme = strdup(schemep);
    if(!u->scheme)
      return CURLUE_OUT_OF_MEMORY;
  }

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

  if(junkscan(path))
    return CURLUE_MALFORMED_INPUT;

  query = strchr(path, '?');
  if(query)
    *query++ = 0;

  fragment = strchr(query?query:path, '#');
  if(fragment)
    *fragment++ = 0;

  if(!path[0])
    /* if there's no path set, unset */
    path = NULL;
  else if(!(flags & CURLU_PATH_AS_IS)) {
    /* sanitise paths and remove ../ and ./ sequences according to RFC3986 */
    char *newp = Curl_dedotdotify(path);
    if(!newp)
      return CURLUE_OUT_OF_MEMORY;

    if(strcmp(newp, path)) {
      /* if we got a new version */
      path = newp;
      path_alloced = TRUE;
    }
    else
      free(newp);
  }
  if(path) {
    u->path = path_alloced?path:strdup(path);
    if(!u->path)
      return CURLUE_OUT_OF_MEMORY;
  }

  if(hostname) {
    /*
     * Parse the login details and strip them out of the host name.
     */
    if(junkscan(hostname))
      return CURLUE_MALFORMED_INPUT;

    result = parse_hostname_login(u, h, &hostname, flags);
    if(result)
      return result;

    result = Curl_parse_port(u, hostname);
    if(result)
      return result;

    result = hostname_check(hostname, flags);
    if(result)
      return result;

    u->host = strdup(hostname);
    if(!u->host)
      return CURLUE_OUT_OF_MEMORY;
  }

  if(query) {
    u->query = strdup(query);
    if(!u->query)
      return CURLUE_OUT_OF_MEMORY;
  }
  if(fragment && fragment[0]) {
    u->fragment = strdup(fragment);
    if(!u->fragment)
      return CURLUE_OUT_OF_MEMORY;
  }

  free(u->scratch);
  u->scratch = NULL;

  return CURLUE_OK;
}

/*
 * Parse the URL and set the relevant members of the Curl_URL struct.
 */
static CURLUcode parseurl(const char *url, CURLU *u, unsigned int flags)
{
  CURLUcode result = seturl(url, u, flags);
  if(result) {
    free_urlhandle(u);
    memset(u, 0, sizeof(struct Curl_URL));
  }
  return result;
}

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

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

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

CURLU *curl_url_dup(CURLU *in)
{
  struct Curl_URL *u = calloc(sizeof(struct Curl_URL), 1);
  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);
    u->portnum = in->portnum;
  }
  return u;
  fail:
  curl_url_cleanup(u);
  return NULL;
}

CURLUcode curl_url_get(CURLU *u, CURLUPart what,
                       char **part, unsigned int flags)
{
  char *ptr;
  CURLUcode ifmissing = CURLUE_UNKNOWN_PART;
  char portbuf[7];
  bool urldecode = (flags & CURLU_URLDECODE)?1:0;
  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 */
    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;
    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's no stored port number, but asked to deliver
         a default one for the scheme */
      const struct Curl_handler *h =
        Curl_builtin_scheme(u->scheme);
      if(h) {
        msnprintf(portbuf, sizeof(portbuf), "%ld", 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_builtin_scheme(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 = u->path = strdup("/");
      if(!u->path)
        return CURLUE_OUT_OF_MEMORY;
    }
    break;
  case CURLUPART_QUERY:
    ptr = u->query;
    ifmissing = CURLUE_NO_QUERY;
    plusdecode = urldecode;
    break;
  case CURLUPART_FRAGMENT:
    ptr = u->fragment;
    ifmissing = CURLUE_NO_FRAGMENT;
    break;
  case CURLUPART_URL: {
    char *url;
    char *scheme;
    char *options = u->options;
    char *port = u->port;
    if(u->scheme && strcasecompare("file", u->scheme)) {
      url = aprintf("file://%s%s%s",
                    u->path,
                    u->fragment? "#": "",
                    u->fragment? u->fragment : "");
    }
    else if(!u->host)
      return CURLUE_NO_HOST;
    else {
      const struct Curl_handler *h = NULL;
      if(u->scheme)
        scheme = u->scheme;
      else if(flags & CURLU_DEFAULT_SCHEME)
        scheme = (char *) DEFAULT_SCHEME;
      else
        return CURLUE_NO_SCHEME;

      if(scheme) {
        h = Curl_builtin_scheme(scheme);
        if(!port && (flags & CURLU_DEFAULT_PORT)) {
          /* there's no stored port number, but asked to deliver
             a default one for the scheme */
          if(h) {
            msnprintf(portbuf, sizeof(portbuf), "%ld", 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;

      url = aprintf("%s://%s%s%s%s%s%s%s%s%s%s%s%s%s%s%s",
                    scheme,
                    u->user ? u->user : "",
                    u->password ? ":": "",
                    u->password ? u->password : "",
                    options ? ";" : "",
                    options ? options : "",
                    (u->user || u->password || options) ? "@": "",
                    u->host,
                    port ? ":": "",
                    port ? port : "",
                    (u->path && (u->path[0] != '/')) ? "/": "",
                    u->path ? u->path : "/",
                    (u->query && u->query[0]) ? "?": "",
                    (u->query && u->query[0]) ? u->query : "",
                    u->fragment? "#": "",
                    u->fragment? u->fragment : "");
    }
    if(!url)
      return CURLUE_OUT_OF_MEMORY;
    *part = url;
    return CURLUE_OK;
    break;
  }
  default:
    ptr = NULL;
  }
  if(ptr) {
    *part = strdup(ptr);
    if(!*part)
      return CURLUE_OUT_OF_MEMORY;
    if(plusdecode) {
      /* convert + to space */
      char *plus;
      for(plus = *part; *plus; ++plus) {
        if(*plus == '+')
          *plus = ' ';
      }
    }
    if(urldecode) {
      char *decoded;
      size_t dlen;
      CURLcode res = Curl_urldecode(NULL, *part, 0, &decoded, &dlen, TRUE);
      free(*part);
      if(res) {
        *part = NULL;
        return CURLUE_URLDECODE;
      }
      *part = decoded;
    }
    return CURLUE_OK;
  }
  else
    return ifmissing;
}

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

  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;
      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_PORT:
      storep = &u->port;
      break;
    case CURLUPART_PATH:
      storep = &u->path;
      break;
    case CURLUPART_QUERY:
      storep = &u->query;
      break;
    case CURLUPART_FRAGMENT:
      storep = &u->fragment;
      break;
    default:
      return CURLUE_UNKNOWN_PART;
    }
    if(storep && *storep) {
      free(*storep);
      *storep = NULL;
    }
    return CURLUE_OK;
  }

  switch(what) {
  case CURLUPART_SCHEME:
    if(!(flags & CURLU_NON_SUPPORT_SCHEME) &&
       /* verify that it is a fine scheme */
       !Curl_builtin_scheme(part))
      return CURLUE_UNSUPPORTED_SCHEME;
    storep = &u->scheme;
    urlencode = FALSE; /* never */
    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_PORT:
    urlencode = FALSE; /* never */
    port = strtol(part, NULL, 10);  /* Port number must be decimal */
    if((port <= 0) || (port > 0xffff))
      return CURLUE_BAD_PORT_NUMBER;
    storep = &u->port;
    break;
  case CURLUPART_PATH:
    urlskipslash = TRUE;
    storep = &u->path;
    break;
  case CURLUPART_QUERY:
    plusencode = urlencode;
    appendquery = (flags & CURLU_APPENDQUERY)?1:0;
    equalsencode = appendquery;
    storep = &u->query;
    break;
  case CURLUPART_FRAGMENT:
    storep = &u->fragment;
    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.
     */
    CURLUcode result;
    char *oldurl;
    char *redired_url;
    CURLU *handle2;

    if(Curl_is_absolute_url(part, NULL, MAX_SCHEME_LEN)) {
      handle2 = curl_url();
      if(!handle2)
        return CURLUE_OUT_OF_MEMORY;
      result = parseurl(part, handle2, flags);
      if(!result)
        mv_urlhandle(handle2, u);
      else
        curl_url_cleanup(handle2);
      return result;
    }
    /* extract the full "old" URL to do the redirect on */
    result = curl_url_get(u, CURLUPART_URL, &oldurl, flags);
    if(result) {
      /* couldn't get the old URL, just use the new! */
      handle2 = curl_url();
      if(!handle2)
        return CURLUE_OUT_OF_MEMORY;
      result = parseurl(part, handle2, flags);
      if(!result)
        mv_urlhandle(handle2, u);
      else
        curl_url_cleanup(handle2);
      return result;
    }

    /* apply the relative part to create a new URL */
    redired_url = Curl_concat_url(oldurl, part);
    free(oldurl);
    if(!redired_url)
      return CURLUE_OUT_OF_MEMORY;

    /* now parse the new URL */
    handle2 = curl_url();
    if(!handle2) {
      free(redired_url);
      return CURLUE_OUT_OF_MEMORY;
    }
    result = parseurl(redired_url, handle2, flags);
    free(redired_url);
    if(!result)
      mv_urlhandle(handle2, u);
    else
      curl_url_cleanup(handle2);
    return result;
  }
  default:
    return CURLUE_UNKNOWN_PART;
  }
  if(storep) {
    const char *newp = part;
    size_t nalloc = strlen(part);

    if(urlencode) {
      const char *i;
      char *o;
      bool free_part = FALSE;
      char *enc = malloc(nalloc * 3 + 1); /* for worst case! */
      if(!enc)
        return CURLUE_OUT_OF_MEMORY;
      if(plusencode) {
        /* space to plus */
        i = part;
        for(o = enc; *i; ++o, ++i)
          *o = (*i == ' ') ? '+' : *i;
        *o = 0; /* zero terminate */
        part = strdup(enc);
        if(!part) {
          free(enc);
          return CURLUE_OUT_OF_MEMORY;
        }
        free_part = TRUE;
      }
      for(i = part, o = enc; *i; i++) {
        if(Curl_isunreserved(*i) ||
           ((*i == '/') && urlskipslash) ||
           ((*i == '=') && equalsencode) ||
           ((*i == '+') && plusencode)) {
          if((*i == '=') && equalsencode)
            /* only skip the first equals sign */
            equalsencode = FALSE;
          *o = *i;
          o++;
        }
        else {
          msnprintf(o, 4, "%%%02x", *i);
          o += 3;
        }
      }
      *o = 0; /* zero terminate */
      newp = enc;
      if(free_part)
        free((char *)part);
    }
    else {
      char *p;
      newp = strdup(part);
      if(!newp)
        return CURLUE_OUT_OF_MEMORY;
      p = (char *)newp;
      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] = (char)TOLOWER(p[1]);
          p[2] = (char)TOLOWER(p[2]);
          p += 3;
        }
        else
          p++;
      }
    }

    if(appendquery) {
      /* Append the string onto the old query. Add a '&' separator if none is
         present at the end of the exsting query already */
      size_t querylen = u->query ? strlen(u->query) : 0;
      bool addamperand = querylen && (u->query[querylen -1] != '&');
      if(querylen) {
        size_t newplen = strlen(newp);
        char *p = malloc(querylen + addamperand + newplen + 1);
        if(!p) {
          free((char *)newp);
          return CURLUE_OUT_OF_MEMORY;
        }
        strcpy(p, u->query); /* original query */
        if(addamperand)
          p[querylen] = '&'; /* ampersand */
        strcpy(&p[querylen + addamperand], newp); /* new suffix */
        free((char *)newp);
        free(*storep);
        *storep = p;
        return CURLUE_OK;
      }
    }

    free(*storep);
    *storep = (char *)newp;
  }
  /* set after the string, to make it not assigned if the allocation above
     fails */
  if(port)
    u->portnum = port;
  return CURLUE_OK;
}