summaryrefslogtreecommitdiff
path: root/lib/url.js
blob: 0e02dbc13121015a0814d9000a59b8c3f7fed6c9 (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
// Copyright Joyent, Inc. and other Node contributors.
//
// Permission is hereby granted, free of charge, to any person obtaining a
// copy of this software and associated documentation files (the
// "Software"), to deal in the Software without restriction, including
// without limitation the rights to use, copy, modify, merge, publish,
// distribute, sublicense, and/or sell copies of the Software, and to permit
// persons to whom the Software is furnished to do so, subject to the
// following conditions:
//
// The above copyright notice and this permission notice shall be included
// in all copies or substantial portions of the Software.
//
// THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS
// OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
// MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN
// NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM,
// DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR
// OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE
// USE OR OTHER DEALINGS IN THE SOFTWARE.

'use strict';

const { toASCII } = require('internal/idna');
const { hexTable } = require('internal/querystring');
const { SafeSet } = primordials;

const {
  ERR_INVALID_ARG_TYPE
} = require('internal/errors').codes;
const { validateString } = require('internal/validators');

// This ensures setURLConstructor() is called before the native
// URL::ToObject() method is used.
const { spliceOne } = require('internal/util');

// WHATWG URL implementation provided by internal/url
const {
  URL,
  URLSearchParams,
  domainToASCII,
  domainToUnicode,
  formatSymbol,
  encodeStr,
  pathToFileURL,
  fileURLToPath
} = require('internal/url');

// Original url.parse() API

function Url() {
  this.protocol = null;
  this.slashes = null;
  this.auth = null;
  this.host = null;
  this.port = null;
  this.hostname = null;
  this.hash = null;
  this.search = null;
  this.query = null;
  this.pathname = null;
  this.path = null;
  this.href = null;
}

// Reference: RFC 3986, RFC 1808, RFC 2396

// define these here so at least they only have to be
// compiled once on the first module load.
const protocolPattern = /^[a-z0-9.+-]+:/i;
const portPattern = /:[0-9]*$/;
const hostPattern = /^\/\/[^@/]+@[^@/]+/;

// Special case for a simple path URL
const simplePathPattern = /^(\/\/?(?!\/)[^?\s]*)(\?[^\s]*)?$/;

const hostnameMaxLen = 255;
// Protocols that can allow "unsafe" and "unwise" chars.
const unsafeProtocol = new SafeSet([
  'javascript',
  'javascript:'
]);
// protocols that never have a hostname.
const hostlessProtocol = new SafeSet([
  'javascript',
  'javascript:'
]);
// protocols that always contain a // bit.
const slashedProtocol = new SafeSet([
  'http',
  'http:',
  'https',
  'https:',
  'ftp',
  'ftp:',
  'gopher',
  'gopher:',
  'file',
  'file:'
]);
const {
  CHAR_SPACE,
  CHAR_TAB,
  CHAR_CARRIAGE_RETURN,
  CHAR_LINE_FEED,
  CHAR_FORM_FEED,
  CHAR_NO_BREAK_SPACE,
  CHAR_ZERO_WIDTH_NOBREAK_SPACE,
  CHAR_HASH,
  CHAR_FORWARD_SLASH,
  CHAR_LEFT_SQUARE_BRACKET,
  CHAR_RIGHT_SQUARE_BRACKET,
  CHAR_LEFT_ANGLE_BRACKET,
  CHAR_RIGHT_ANGLE_BRACKET,
  CHAR_LEFT_CURLY_BRACKET,
  CHAR_RIGHT_CURLY_BRACKET,
  CHAR_QUESTION_MARK,
  CHAR_LOWERCASE_A,
  CHAR_LOWERCASE_Z,
  CHAR_UPPERCASE_A,
  CHAR_UPPERCASE_Z,
  CHAR_DOT,
  CHAR_0,
  CHAR_9,
  CHAR_HYPHEN_MINUS,
  CHAR_PLUS,
  CHAR_UNDERSCORE,
  CHAR_DOUBLE_QUOTE,
  CHAR_SINGLE_QUOTE,
  CHAR_PERCENT,
  CHAR_SEMICOLON,
  CHAR_BACKWARD_SLASH,
  CHAR_CIRCUMFLEX_ACCENT,
  CHAR_GRAVE_ACCENT,
  CHAR_VERTICAL_LINE,
  CHAR_AT,
} = require('internal/constants');

// Lazy loaded for startup performance.
let querystring;

function urlParse(url, parseQueryString, slashesDenoteHost) {
  if (url instanceof Url) return url;

  var urlObject = new Url();
  urlObject.parse(url, parseQueryString, slashesDenoteHost);
  return urlObject;
}

Url.prototype.parse = function parse(url, parseQueryString, slashesDenoteHost) {
  validateString(url, 'url');

  // Copy chrome, IE, opera backslash-handling behavior.
  // Back slashes before the query string get converted to forward slashes
  // See: https://code.google.com/p/chromium/issues/detail?id=25916
  var hasHash = false;
  var start = -1;
  var end = -1;
  var rest = '';
  var lastPos = 0;
  var i = 0;
  for (var inWs = false, split = false; i < url.length; ++i) {
    const code = url.charCodeAt(i);

    // Find first and last non-whitespace characters for trimming
    const isWs = code === CHAR_SPACE ||
                 code === CHAR_TAB ||
                 code === CHAR_CARRIAGE_RETURN ||
                 code === CHAR_LINE_FEED ||
                 code === CHAR_FORM_FEED ||
                 code === CHAR_NO_BREAK_SPACE ||
                 code === CHAR_ZERO_WIDTH_NOBREAK_SPACE;
    if (start === -1) {
      if (isWs)
        continue;
      lastPos = start = i;
    } else if (inWs) {
      if (!isWs) {
        end = -1;
        inWs = false;
      }
    } else if (isWs) {
      end = i;
      inWs = true;
    }

    // Only convert backslashes while we haven't seen a split character
    if (!split) {
      switch (code) {
        case CHAR_HASH:
          hasHash = true;
        // Fall through
        case CHAR_QUESTION_MARK:
          split = true;
          break;
        case CHAR_BACKWARD_SLASH:
          if (i - lastPos > 0)
            rest += url.slice(lastPos, i);
          rest += '/';
          lastPos = i + 1;
          break;
      }
    } else if (!hasHash && code === CHAR_HASH) {
      hasHash = true;
    }
  }

  // Check if string was non-empty (including strings with only whitespace)
  if (start !== -1) {
    if (lastPos === start) {
      // We didn't convert any backslashes

      if (end === -1) {
        if (start === 0)
          rest = url;
        else
          rest = url.slice(start);
      } else {
        rest = url.slice(start, end);
      }
    } else if (end === -1 && lastPos < url.length) {
      // We converted some backslashes and have only part of the entire string
      rest += url.slice(lastPos);
    } else if (end !== -1 && lastPos < end) {
      // We converted some backslashes and have only part of the entire string
      rest += url.slice(lastPos, end);
    }
  }

  if (!slashesDenoteHost && !hasHash) {
    // Try fast path regexp
    const simplePath = simplePathPattern.exec(rest);
    if (simplePath) {
      this.path = rest;
      this.href = rest;
      this.pathname = simplePath[1];
      if (simplePath[2]) {
        this.search = simplePath[2];
        if (parseQueryString) {
          if (querystring === undefined) querystring = require('querystring');
          this.query = querystring.parse(this.search.slice(1));
        } else {
          this.query = this.search.slice(1);
        }
      } else if (parseQueryString) {
        this.search = null;
        this.query = Object.create(null);
      }
      return this;
    }
  }

  var proto = protocolPattern.exec(rest);
  if (proto) {
    proto = proto[0];
    var lowerProto = proto.toLowerCase();
    this.protocol = lowerProto;
    rest = rest.slice(proto.length);
  }

  // figure out if it's got a host
  // user@server is *always* interpreted as a hostname, and url
  // resolution will treat //foo/bar as host=foo,path=bar because that's
  // how the browser resolves relative URLs.
  if (slashesDenoteHost || proto || hostPattern.test(rest)) {
    var slashes = rest.charCodeAt(0) === CHAR_FORWARD_SLASH &&
                  rest.charCodeAt(1) === CHAR_FORWARD_SLASH;
    if (slashes && !(proto && hostlessProtocol.has(lowerProto))) {
      rest = rest.slice(2);
      this.slashes = true;
    }
  }

  if (!hostlessProtocol.has(lowerProto) &&
      (slashes || (proto && !slashedProtocol.has(proto)))) {

    // there's a hostname.
    // the first instance of /, ?, ;, or # ends the host.
    //
    // If there is an @ in the hostname, then non-host chars *are* allowed
    // to the left of the last @ sign, unless some host-ending character
    // comes *before* the @-sign.
    // URLs are obnoxious.
    //
    // ex:
    // http://a@b@c/ => user:a@b host:c
    // http://a@b?@c => user:a host:b path:/?@c

    var hostEnd = -1;
    var atSign = -1;
    var nonHost = -1;
    for (i = 0; i < rest.length; ++i) {
      switch (rest.charCodeAt(i)) {
        case CHAR_TAB:
        case CHAR_LINE_FEED:
        case CHAR_CARRIAGE_RETURN:
        case CHAR_SPACE:
        case CHAR_DOUBLE_QUOTE:
        case CHAR_PERCENT:
        case CHAR_SINGLE_QUOTE:
        case CHAR_SEMICOLON:
        case CHAR_LEFT_ANGLE_BRACKET:
        case CHAR_RIGHT_ANGLE_BRACKET:
        case CHAR_BACKWARD_SLASH:
        case CHAR_CIRCUMFLEX_ACCENT:
        case CHAR_GRAVE_ACCENT:
        case CHAR_LEFT_CURLY_BRACKET:
        case CHAR_VERTICAL_LINE:
        case CHAR_RIGHT_CURLY_BRACKET:
          // Characters that are never ever allowed in a hostname from RFC 2396
          if (nonHost === -1)
            nonHost = i;
          break;
        case CHAR_HASH:
        case CHAR_FORWARD_SLASH:
        case CHAR_QUESTION_MARK:
          // Find the first instance of any host-ending characters
          if (nonHost === -1)
            nonHost = i;
          hostEnd = i;
          break;
        case CHAR_AT:
          // At this point, either we have an explicit point where the
          // auth portion cannot go past, or the last @ char is the decider.
          atSign = i;
          nonHost = -1;
          break;
      }
      if (hostEnd !== -1)
        break;
    }
    start = 0;
    if (atSign !== -1) {
      this.auth = decodeURIComponent(rest.slice(0, atSign));
      start = atSign + 1;
    }
    if (nonHost === -1) {
      this.host = rest.slice(start);
      rest = '';
    } else {
      this.host = rest.slice(start, nonHost);
      rest = rest.slice(nonHost);
    }

    // pull out port.
    this.parseHost();

    // we've indicated that there is a hostname,
    // so even if it's empty, it has to be present.
    if (typeof this.hostname !== 'string')
      this.hostname = '';

    var hostname = this.hostname;

    // if hostname begins with [ and ends with ]
    // assume that it's an IPv6 address.
    var ipv6Hostname = hostname.charCodeAt(0) === CHAR_LEFT_SQUARE_BRACKET &&
      hostname.charCodeAt(hostname.length - 1) === CHAR_RIGHT_SQUARE_BRACKET;

    // validate a little.
    if (!ipv6Hostname) {
      const result = validateHostname(this, rest, hostname);
      if (result !== undefined)
        rest = result;
    }

    if (this.hostname.length > hostnameMaxLen) {
      this.hostname = '';
    } else {
      // hostnames are always lower case.
      this.hostname = this.hostname.toLowerCase();
    }

    if (!ipv6Hostname) {
      // IDNA Support: Returns a punycoded representation of "domain".
      // It only converts parts of the domain name that
      // have non-ASCII characters, i.e. it doesn't matter if
      // you call it with a domain that already is ASCII-only.

      // Use lenient mode (`true`) to try to support even non-compliant
      // URLs.
      this.hostname = toASCII(this.hostname, true);
    }

    var p = this.port ? ':' + this.port : '';
    var h = this.hostname || '';
    this.host = h + p;

    // strip [ and ] from the hostname
    // the host field still retains them, though
    if (ipv6Hostname) {
      this.hostname = this.hostname.slice(1, -1);
      if (rest[0] !== '/') {
        rest = '/' + rest;
      }
    }
  }

  // now rest is set to the post-host stuff.
  // chop off any delim chars.
  if (!unsafeProtocol.has(lowerProto)) {
    // First, make 100% sure that any "autoEscape" chars get
    // escaped, even if encodeURIComponent doesn't think they
    // need to be.
    rest = autoEscapeStr(rest);
  }

  var questionIdx = -1;
  var hashIdx = -1;
  for (i = 0; i < rest.length; ++i) {
    const code = rest.charCodeAt(i);
    if (code === CHAR_HASH) {
      this.hash = rest.slice(i);
      hashIdx = i;
      break;
    } else if (code === CHAR_QUESTION_MARK && questionIdx === -1) {
      questionIdx = i;
    }
  }

  if (questionIdx !== -1) {
    if (hashIdx === -1) {
      this.search = rest.slice(questionIdx);
      this.query = rest.slice(questionIdx + 1);
    } else {
      this.search = rest.slice(questionIdx, hashIdx);
      this.query = rest.slice(questionIdx + 1, hashIdx);
    }
    if (parseQueryString) {
      if (querystring === undefined) querystring = require('querystring');
      this.query = querystring.parse(this.query);
    }
  } else if (parseQueryString) {
    // No query string, but parseQueryString still requested
    this.search = null;
    this.query = Object.create(null);
  }

  const useQuestionIdx =
    questionIdx !== -1 && (hashIdx === -1 || questionIdx < hashIdx);
  const firstIdx = useQuestionIdx ? questionIdx : hashIdx;
  if (firstIdx === -1) {
    if (rest.length > 0)
      this.pathname = rest;
  } else if (firstIdx > 0) {
    this.pathname = rest.slice(0, firstIdx);
  }
  if (slashedProtocol.has(lowerProto) &&
      this.hostname && !this.pathname) {
    this.pathname = '/';
  }

  // to support http.request
  if (this.pathname || this.search) {
    const p = this.pathname || '';
    const s = this.search || '';
    this.path = p + s;
  }

  // Finally, reconstruct the href based on what has been validated.
  this.href = this.format();
  return this;
};

function validateHostname(self, rest, hostname) {
  for (var i = 0; i < hostname.length; ++i) {
    const code = hostname.charCodeAt(i);
    const isValid = (code >= CHAR_LOWERCASE_A && code <= CHAR_LOWERCASE_Z) ||
                    code === CHAR_DOT ||
                    (code >= CHAR_UPPERCASE_A && code <= CHAR_UPPERCASE_Z) ||
                    (code >= CHAR_0 && code <= CHAR_9) ||
                    code === CHAR_HYPHEN_MINUS ||
                    code === CHAR_PLUS ||
                    code === CHAR_UNDERSCORE ||
                    code > 127;

    // Invalid host character
    if (!isValid) {
      self.hostname = hostname.slice(0, i);
      return '/' + hostname.slice(i) + rest;
    }
  }
}

// Escaped characters. Use empty strings to fill up unused entries.
// Using Array is faster than Object/Map
const escapedCodes = [
  /* 0 - 9 */ '', '', '', '', '', '', '', '', '', '%09',
  /* 10 - 19 */ '%0A', '', '', '%0D', '', '', '', '', '', '',
  /* 20 - 29 */ '', '', '', '', '', '', '', '', '', '',
  /* 30 - 39 */ '', '', '%20', '', '%22', '', '', '', '', '%27',
  /* 40 - 49 */ '', '', '', '', '', '', '', '', '', '',
  /* 50 - 59 */ '', '', '', '', '', '', '', '', '', '',
  /* 60 - 69 */ '%3C', '', '%3E', '', '', '', '', '', '', '',
  /* 70 - 79 */ '', '', '', '', '', '', '', '', '', '',
  /* 80 - 89 */ '', '', '', '', '', '', '', '', '', '',
  /* 90 - 99 */ '', '', '%5C', '', '%5E', '', '%60', '', '', '',
  /* 100 - 109 */ '', '', '', '', '', '', '', '', '', '',
  /* 110 - 119 */ '', '', '', '', '', '', '', '', '', '',
  /* 120 - 125 */ '', '', '', '%7B', '%7C', '%7D'
];

// Automatically escape all delimiters and unwise characters from RFC 2396.
// Also escape single quotes in case of an XSS attack.
// Return the escaped string.
function autoEscapeStr(rest) {
  var escaped = '';
  var lastEscapedPos = 0;
  for (var i = 0; i < rest.length; ++i) {
    // `escaped` contains substring up to the last escaped character.
    var escapedChar = escapedCodes[rest.charCodeAt(i)];
    if (escapedChar) {
      // Concat if there are ordinary characters in the middle.
      if (i > lastEscapedPos)
        escaped += rest.slice(lastEscapedPos, i);
      escaped += escapedChar;
      lastEscapedPos = i + 1;
    }
  }
  if (lastEscapedPos === 0)  // Nothing has been escaped.
    return rest;

  // There are ordinary characters at the end.
  if (lastEscapedPos < rest.length)
    escaped += rest.slice(lastEscapedPos);

  return escaped;
}

// format a parsed object into a url string
function urlFormat(urlObject, options) {
  // ensure it's an object, and not a string url.
  // If it's an object, this is a no-op.
  // this way, you can call urlParse() on strings
  // to clean up potentially wonky urls.
  if (typeof urlObject === 'string') {
    urlObject = urlParse(urlObject);
  } else if (typeof urlObject !== 'object' || urlObject === null) {
    throw new ERR_INVALID_ARG_TYPE('urlObject',
                                   ['Object', 'string'], urlObject);
  } else if (!(urlObject instanceof Url)) {
    var format = urlObject[formatSymbol];
    return format ?
      format.call(urlObject, options) :
      Url.prototype.format.call(urlObject);
  }
  return urlObject.format();
}

// These characters do not need escaping:
// ! - . _ ~
// ' ( ) * :
// digits
// alpha (uppercase)
// alpha (lowercase)
const noEscapeAuth = [
  0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, // 0x00 - 0x0F
  0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, // 0x10 - 0x1F
  0, 1, 0, 0, 0, 0, 0, 1, 1, 1, 1, 0, 0, 1, 1, 0, // 0x20 - 0x2F
  1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, // 0x30 - 0x3F
  0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, // 0x40 - 0x4F
  1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 1, // 0x50 - 0x5F
  0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, // 0x60 - 0x6F
  1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 1, 0  // 0x70 - 0x7F
];

Url.prototype.format = function format() {
  var auth = this.auth || '';
  if (auth) {
    auth = encodeStr(auth, noEscapeAuth, hexTable);
    auth += '@';
  }

  var protocol = this.protocol || '';
  var pathname = this.pathname || '';
  var hash = this.hash || '';
  var host = '';
  var query = '';

  if (this.host) {
    host = auth + this.host;
  } else if (this.hostname) {
    host = auth + (
      this.hostname.indexOf(':') === -1 ?
        this.hostname :
        '[' + this.hostname + ']'
    );
    if (this.port) {
      host += ':' + this.port;
    }
  }

  if (this.query !== null && typeof this.query === 'object') {
    if (querystring === undefined) querystring = require('querystring');
    query = querystring.stringify(this.query);
  }

  var search = this.search || (query && ('?' + query)) || '';

  if (protocol && protocol.charCodeAt(protocol.length - 1) !== 58/* : */)
    protocol += ':';

  var newPathname = '';
  var lastPos = 0;
  for (var i = 0; i < pathname.length; ++i) {
    switch (pathname.charCodeAt(i)) {
      case CHAR_HASH:
        if (i - lastPos > 0)
          newPathname += pathname.slice(lastPos, i);
        newPathname += '%23';
        lastPos = i + 1;
        break;
      case CHAR_QUESTION_MARK:
        if (i - lastPos > 0)
          newPathname += pathname.slice(lastPos, i);
        newPathname += '%3F';
        lastPos = i + 1;
        break;
    }
  }
  if (lastPos > 0) {
    if (lastPos !== pathname.length)
      pathname = newPathname + pathname.slice(lastPos);
    else
      pathname = newPathname;
  }

  // Only the slashedProtocols get the //.  Not mailto:, xmpp:, etc.
  // unless they had them to begin with.
  if (this.slashes || slashedProtocol.has(protocol)) {
    if (this.slashes || host) {
      if (pathname && pathname.charCodeAt(0) !== CHAR_FORWARD_SLASH)
        pathname = '/' + pathname;
      host = '//' + host;
    } else if (protocol.length >= 4 &&
               protocol.charCodeAt(0) === 102/* f */ &&
               protocol.charCodeAt(1) === 105/* i */ &&
               protocol.charCodeAt(2) === 108/* l */ &&
               protocol.charCodeAt(3) === 101/* e */) {
      host = '//';
    }
  }

  search = search.replace(/#/g, '%23');

  if (hash && hash.charCodeAt(0) !== CHAR_HASH)
    hash = '#' + hash;
  if (search && search.charCodeAt(0) !== CHAR_QUESTION_MARK)
    search = '?' + search;

  return protocol + host + pathname + search + hash;
};

function urlResolve(source, relative) {
  return urlParse(source, false, true).resolve(relative);
}

Url.prototype.resolve = function resolve(relative) {
  return this.resolveObject(urlParse(relative, false, true)).format();
};

function urlResolveObject(source, relative) {
  if (!source) return relative;
  return urlParse(source, false, true).resolveObject(relative);
}

Url.prototype.resolveObject = function resolveObject(relative) {
  if (typeof relative === 'string') {
    var rel = new Url();
    rel.parse(relative, false, true);
    relative = rel;
  }

  var result = new Url();
  var tkeys = Object.keys(this);
  for (var tk = 0; tk < tkeys.length; tk++) {
    var tkey = tkeys[tk];
    result[tkey] = this[tkey];
  }

  // hash is always overridden, no matter what.
  // even href="" will remove it.
  result.hash = relative.hash;

  // If the relative url is empty, then there's nothing left to do here.
  if (relative.href === '') {
    result.href = result.format();
    return result;
  }

  // hrefs like //foo/bar always cut to the protocol.
  if (relative.slashes && !relative.protocol) {
    // take everything except the protocol from relative
    var rkeys = Object.keys(relative);
    for (var rk = 0; rk < rkeys.length; rk++) {
      var rkey = rkeys[rk];
      if (rkey !== 'protocol')
        result[rkey] = relative[rkey];
    }

    // urlParse appends trailing / to urls like http://www.example.com
    if (slashedProtocol.has(result.protocol) &&
        result.hostname && !result.pathname) {
      result.path = result.pathname = '/';
    }

    result.href = result.format();
    return result;
  }

  if (relative.protocol && relative.protocol !== result.protocol) {
    // if it's a known url protocol, then changing
    // the protocol does weird things
    // first, if it's not file:, then we MUST have a host,
    // and if there was a path
    // to begin with, then we MUST have a path.
    // if it is file:, then the host is dropped,
    // because that's known to be hostless.
    // anything else is assumed to be absolute.
    if (!slashedProtocol.has(relative.protocol)) {
      var keys = Object.keys(relative);
      for (var v = 0; v < keys.length; v++) {
        var k = keys[v];
        result[k] = relative[k];
      }
      result.href = result.format();
      return result;
    }

    result.protocol = relative.protocol;
    if (!relative.host &&
        !/^file:?$/.test(relative.protocol) &&
        !hostlessProtocol.has(relative.protocol)) {
      const relPath = (relative.pathname || '').split('/');
      while (relPath.length && !(relative.host = relPath.shift()));
      if (!relative.host) relative.host = '';
      if (!relative.hostname) relative.hostname = '';
      if (relPath[0] !== '') relPath.unshift('');
      if (relPath.length < 2) relPath.unshift('');
      result.pathname = relPath.join('/');
    } else {
      result.pathname = relative.pathname;
    }
    result.search = relative.search;
    result.query = relative.query;
    result.host = relative.host || '';
    result.auth = relative.auth;
    result.hostname = relative.hostname || relative.host;
    result.port = relative.port;
    // to support http.request
    if (result.pathname || result.search) {
      var p = result.pathname || '';
      var s = result.search || '';
      result.path = p + s;
    }
    result.slashes = result.slashes || relative.slashes;
    result.href = result.format();
    return result;
  }

  var isSourceAbs = (result.pathname && result.pathname.charAt(0) === '/');
  var isRelAbs = (
    relative.host || relative.pathname && relative.pathname.charAt(0) === '/'
  );
  var mustEndAbs = (isRelAbs || isSourceAbs ||
                    (result.host && relative.pathname));
  var removeAllDots = mustEndAbs;
  var srcPath = result.pathname && result.pathname.split('/') || [];
  var relPath = relative.pathname && relative.pathname.split('/') || [];
  var noLeadingSlashes = result.protocol &&
      !slashedProtocol.has(result.protocol);

  // if the url is a non-slashed url, then relative
  // links like ../.. should be able
  // to crawl up to the hostname, as well.  This is strange.
  // result.protocol has already been set by now.
  // Later on, put the first path part into the host field.
  if (noLeadingSlashes) {
    result.hostname = '';
    result.port = null;
    if (result.host) {
      if (srcPath[0] === '') srcPath[0] = result.host;
      else srcPath.unshift(result.host);
    }
    result.host = '';
    if (relative.protocol) {
      relative.hostname = null;
      relative.port = null;
      result.auth = null;
      if (relative.host) {
        if (relPath[0] === '') relPath[0] = relative.host;
        else relPath.unshift(relative.host);
      }
      relative.host = null;
    }
    mustEndAbs = mustEndAbs && (relPath[0] === '' || srcPath[0] === '');
  }

  if (isRelAbs) {
    // it's absolute.
    if (relative.host || relative.host === '') {
      if (result.host !== relative.host) result.auth = null;
      result.host = relative.host;
      result.port = relative.port;
    }
    if (relative.hostname || relative.hostname === '') {
      if (result.hostname !== relative.hostname) result.auth = null;
      result.hostname = relative.hostname;
    }
    result.search = relative.search;
    result.query = relative.query;
    srcPath = relPath;
    // fall through to the dot-handling below.
  } else if (relPath.length) {
    // it's relative
    // throw away the existing file, and take the new path instead.
    if (!srcPath) srcPath = [];
    srcPath.pop();
    srcPath = srcPath.concat(relPath);
    result.search = relative.search;
    result.query = relative.query;
  } else if (relative.search !== null && relative.search !== undefined) {
    // just pull out the search.
    // like href='?foo'.
    // Put this after the other two cases because it simplifies the booleans
    if (noLeadingSlashes) {
      result.hostname = result.host = srcPath.shift();
      // Occasionally the auth can get stuck only in host.
      // This especially happens in cases like
      // url.resolveObject('mailto:local1@domain1', 'local2@domain2')
      const authInHost =
        result.host && result.host.indexOf('@') > 0 && result.host.split('@');
      if (authInHost) {
        result.auth = authInHost.shift();
        result.host = result.hostname = authInHost.shift();
      }
    }
    result.search = relative.search;
    result.query = relative.query;
    // To support http.request
    if (result.pathname !== null || result.search !== null) {
      result.path = (result.pathname ? result.pathname : '') +
                    (result.search ? result.search : '');
    }
    result.href = result.format();
    return result;
  }

  if (!srcPath.length) {
    // no path at all.  easy.
    // we've already handled the other stuff above.
    result.pathname = null;
    // To support http.request
    if (result.search) {
      result.path = '/' + result.search;
    } else {
      result.path = null;
    }
    result.href = result.format();
    return result;
  }

  // If a url ENDs in . or .., then it must get a trailing slash.
  // however, if it ends in anything else non-slashy,
  // then it must NOT get a trailing slash.
  var last = srcPath.slice(-1)[0];
  var hasTrailingSlash = (
    (result.host || relative.host || srcPath.length > 1) &&
    (last === '.' || last === '..') || last === '');

  // Strip single dots, resolve double dots to parent dir
  // if the path tries to go above the root, `up` ends up > 0
  var up = 0;
  for (var i = srcPath.length - 1; i >= 0; i--) {
    last = srcPath[i];
    if (last === '.') {
      spliceOne(srcPath, i);
    } else if (last === '..') {
      spliceOne(srcPath, i);
      up++;
    } else if (up) {
      spliceOne(srcPath, i);
      up--;
    }
  }

  // If the path is allowed to go above the root, restore leading ..s
  if (!mustEndAbs && !removeAllDots) {
    while (up--) {
      srcPath.unshift('..');
    }
  }

  if (mustEndAbs && srcPath[0] !== '' &&
      (!srcPath[0] || srcPath[0].charAt(0) !== '/')) {
    srcPath.unshift('');
  }

  if (hasTrailingSlash && (srcPath.join('/').substr(-1) !== '/')) {
    srcPath.push('');
  }

  var isAbsolute = srcPath[0] === '' ||
      (srcPath[0] && srcPath[0].charAt(0) === '/');

  // put the host back
  if (noLeadingSlashes) {
    result.hostname =
      result.host = isAbsolute ? '' : srcPath.length ? srcPath.shift() : '';
    // Occasionally the auth can get stuck only in host.
    // This especially happens in cases like
    // url.resolveObject('mailto:local1@domain1', 'local2@domain2')
    const authInHost = result.host && result.host.indexOf('@') > 0 ?
      result.host.split('@') : false;
    if (authInHost) {
      result.auth = authInHost.shift();
      result.host = result.hostname = authInHost.shift();
    }
  }

  mustEndAbs = mustEndAbs || (result.host && srcPath.length);

  if (mustEndAbs && !isAbsolute) {
    srcPath.unshift('');
  }

  if (!srcPath.length) {
    result.pathname = null;
    result.path = null;
  } else {
    result.pathname = srcPath.join('/');
  }

  // To support request.http
  if (result.pathname !== null || result.search !== null) {
    result.path = (result.pathname ? result.pathname : '') +
                  (result.search ? result.search : '');
  }
  result.auth = relative.auth || result.auth;
  result.slashes = result.slashes || relative.slashes;
  result.href = result.format();
  return result;
};

Url.prototype.parseHost = function parseHost() {
  var host = this.host;
  var port = portPattern.exec(host);
  if (port) {
    port = port[0];
    if (port !== ':') {
      this.port = port.slice(1);
    }
    host = host.slice(0, host.length - port.length);
  }
  if (host) this.hostname = host;
};

module.exports = {
  // Original API
  Url,
  parse: urlParse,
  resolve: urlResolve,
  resolveObject: urlResolveObject,
  format: urlFormat,

  // WHATWG API
  URL,
  URLSearchParams,
  domainToASCII,
  domainToUnicode,

  // Utilities
  pathToFileURL,
  fileURLToPath
};