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
// Copyright 2013-2015 Simon Sapin.
//
// Licensed under the Apache License, Version 2.0 <LICENSE-APACHE or
// http://www.apache.org/licenses/LICENSE-2.0> or the MIT license
// <LICENSE-MIT or http://opensource.org/licenses/MIT>, at your
// option. This file may not be copied, modified, or distributed
// except according to those terms.

/*!

<a href="https://github.com/servo/rust-url"><img style="position: absolute; top: 0; left: 0; border: 0;" src="../github.png" alt="Fork me on GitHub"></a>
<style>.sidebar { margin-top: 53px }</style>

rust-url is an implementation of the [URL Standard](http://url.spec.whatwg.org/)
for the [Rust](http://rust-lang.org/) programming language.

It builds with [Cargo](http://crates.io/).
To use it in your project, add this to your `Cargo.toml` file:

```Cargo
[dependencies.url]
git = "https://github.com/servo/rust-url"
```

Supporting encodings other than UTF-8 in query strings is an optional feature
that requires [rust-encoding](https://github.com/lifthrasiir/rust-encoding)
and is off by default.
You can enable it with
[Cargo’s *features* mechanism](http://doc.crates.io/manifest.html#the-[features]-section):

```Cargo
[dependencies.url]
git = "https://github.com/servo/rust-url"
features = ["query_encoding"]
```

… or by passing `--cfg 'feature="query_encoding"'` to rustc.


# URL parsing and data structures

First, URL parsing may fail for various reasons and therefore returns a `Result`.

```
use url::{Url, ParseError};

assert!(Url::parse("http://[:::1]") == Err(ParseError::InvalidIpv6Address))
```

Let’s parse a valid URL and look at its components.

```
use url::{Url, SchemeData};

let issue_list_url = Url::parse(
    "https://github.com/rust-lang/rust/issues?labels=E-easy&state=open"
).unwrap();


assert!(issue_list_url.scheme == "https".to_string());
assert!(issue_list_url.domain() == Some("github.com"));
assert!(issue_list_url.port() == None);
assert!(issue_list_url.path() == Some(&["rust-lang".to_string(),
                                        "rust".to_string(),
                                        "issues".to_string()][..]));
assert!(issue_list_url.query == Some("labels=E-easy&state=open".to_string()));
assert!(issue_list_url.fragment == None);
match issue_list_url.scheme_data {
    SchemeData::Relative(..) => {},  // Expected
    SchemeData::NonRelative(..) => panic!(),
}
```

The `scheme`, `query`, and `fragment` are directly fields of the `Url` struct:
they apply to all URLs.
Every other components has accessors because they only apply to URLs said to be
“in a relative scheme”. `https` is a relative scheme, but `data` is not:

```
use url::{Url, SchemeData};

let data_url = Url::parse("data:text/plain,Hello#").unwrap();

assert!(data_url.scheme == "data".to_string());
assert!(data_url.scheme_data == SchemeData::NonRelative("text/plain,Hello".to_string()));
assert!(data_url.non_relative_scheme_data() == Some("text/plain,Hello"));
assert!(data_url.query == None);
assert!(data_url.fragment == Some("".to_string()));
```


# Base URL

Many contexts allow URL *references* that can be relative to a *base URL*:

```html
<link rel="stylesheet" href="../main.css">
```

Since parsed URL are absolute, giving a base is required:

```
use url::{Url, ParseError};

assert!(Url::parse("../main.css") == Err(ParseError::RelativeUrlWithoutBase))
```

`UrlParser` is a method-chaining API to provide various optional parameters
to URL parsing, including a base URL.

```
use url::{Url, UrlParser};

let this_document = Url::parse("http://servo.github.io/rust-url/url/index.html").unwrap();
let css_url = UrlParser::new().base_url(&this_document).parse("../main.css").unwrap();
assert!(css_url.serialize() == "http://servo.github.io/rust-url/main.css".to_string());
```

For convenience, the `join` method on `Url` is also provided to achieve the same result:

```
use url::Url;

let this_document = Url::parse("http://servo.github.io/rust-url/url/index.html").unwrap();
let css_url = this_document.join("../main.css").unwrap();
assert!(&*css_url.serialize() == "http://servo.github.io/rust-url/main.css")
*/

#![cfg_attr(feature="heap_size", feature(plugin, custom_derive))]
#![cfg_attr(feature="heap_size", plugin(heapsize_plugin))]

extern crate rustc_serialize;
extern crate uuid;

#[macro_use]
extern crate matches;

#[cfg(feature="serde_serialization")]
extern crate serde;

#[cfg(feature="heap_size")]
#[macro_use] extern crate heapsize;

extern crate unicode_normalization;
extern crate unicode_bidi;

use std::fmt::{self, Formatter};
use std::str;
use std::path::{Path, PathBuf};
use std::borrow::Borrow;
use std::hash::{Hash, Hasher};
use std::cmp::Ordering;

#[cfg(feature="serde_serialization")]
use std::str::FromStr;

pub use host::Host;
pub use parser::{ErrorHandler, ParseResult, ParseError};

use percent_encoding::{percent_encode, lossy_utf8_percent_decode, DEFAULT_ENCODE_SET};

use format::{PathFormatter, UserInfoFormatter, UrlNoFragmentFormatter};
use encoding::EncodingOverride;

use uuid::Uuid;

mod encoding;
mod host;
mod parser;
pub mod urlutils;
pub mod percent_encoding;
pub mod form_urlencoded;
pub mod punycode;
pub mod format;
pub mod idna;

/// The parsed representation of an absolute URL.
#[derive(PartialEq, Eq, Clone, Debug, Hash, PartialOrd, Ord)]
#[cfg_attr(feature="heap_size", derive(HeapSizeOf))]
pub struct Url {
    /// The scheme (a.k.a. protocol) of the URL, in ASCII lower case.
    pub scheme: String,

    /// The components of the URL whose representation depends on where the scheme is *relative*.
    pub scheme_data: SchemeData,

    /// The query string of the URL.
    ///
    /// `None` if the `?` delimiter character was not part of the parsed input,
    /// otherwise a possibly empty, percent-encoded string.
    ///
    /// Percent encoded strings are within the ASCII range.
    ///
    /// See also the `query_pairs`, `set_query_from_pairs`,
    /// and `lossy_percent_decode_query` methods.
    pub query: Option<String>,

    /// The fragment identifier of the URL.
    ///
    /// `None` if the `#` delimiter character was not part of the parsed input,
    /// otherwise a possibly empty, percent-encoded string.
    ///
    /// Percent encoded strings are within the ASCII range.
    ///
    /// See also the `lossy_percent_decode_fragment` method.
    pub fragment: Option<String>,
}

/// Opaque identifier for URLs that have file or other schemes
#[derive(PartialEq, Eq, Clone, Debug)]
pub struct OpaqueOrigin(Uuid);

#[cfg(feature="heap_size")]
known_heap_size!(0, OpaqueOrigin);

impl OpaqueOrigin {
    /// Creates a new opaque origin with a random UUID.
    pub fn new() -> OpaqueOrigin {
        OpaqueOrigin(Uuid::new_v4())
    }
}

/// The origin of the URL
#[derive(PartialEq, Eq, Clone, Debug)]
#[cfg_attr(feature="heap_size", derive(HeapSizeOf))]
pub enum Origin {
    /// A globally unique identifier
    UID(OpaqueOrigin),

    /// Consists of the URL's scheme, host and port
    Tuple(String, Host, u16)
}

/// The components of the URL whose representation depends on where the scheme is *relative*.
#[derive(PartialEq, Eq, Clone, Debug, Hash, PartialOrd, Ord)]
#[cfg_attr(feature="heap_size", derive(HeapSizeOf))]
pub enum SchemeData {
    /// Components for URLs in a *relative* scheme such as HTTP.
    Relative(RelativeSchemeData),

    /// No further structure is assumed for *non-relative* schemes such as `data` and `mailto`.
    ///
    /// This is a single percent-encoded string, whose interpretation depends on the scheme.
    ///
    /// Percent encoded strings are within the ASCII range.
    NonRelative(String),
}

/// Components for URLs in a *relative* scheme such as HTTP.
#[derive(Clone, Debug)]
#[cfg_attr(feature="heap_size", derive(HeapSizeOf))]
pub struct RelativeSchemeData {
    /// The username of the URL, as a possibly empty, percent-encoded string.
    ///
    /// Percent encoded strings are within the ASCII range.
    ///
    /// See also the `lossy_percent_decode_username` method.
    pub username: String,

    /// The password of the URL.
    ///
    /// `None` if the `:` delimiter character was not part of the parsed input,
    /// otherwise a possibly empty, percent-encoded string.
    ///
    /// Percent encoded strings are within the ASCII range.
    ///
    /// See also the `lossy_percent_decode_password` method.
    pub password: Option<String>,

    /// The host of the URL, either a domain name or an IPv4 address
    pub host: Host,

    /// The port number of the URL.
    /// `None` for file-like schemes, or to indicate the default port number.
    pub port: Option<u16>,

    /// The default port number for the URL’s scheme.
    /// `None` for file-like schemes.
    pub default_port: Option<u16>,

    /// The path of the URL, as vector of percent-encoded strings.
    ///
    /// Percent encoded strings are within the ASCII range.
    ///
    /// See also the `serialize_path` method and,
    /// for URLs in the `file` scheme, the `to_file_path` method.
    pub path: Vec<String>,
}

impl RelativeSchemeData {
    fn get_identity_key(&self) -> (&String, &Option<String>, &Host, Option<u16>, Option<u16>, &Vec<String>) {
        (
            &self.username,
            &self.password,
            &self.host,
            self.port.or(self.default_port),
            self.default_port,
            &self.path
        )
    }
}


impl PartialEq for RelativeSchemeData {
    fn eq(&self, other: &RelativeSchemeData) -> bool {
        self.get_identity_key() == other.get_identity_key()
    }
}

impl Eq for RelativeSchemeData {}

impl Hash for RelativeSchemeData {
    fn hash<H: Hasher>(&self, state: &mut H) {
        self.get_identity_key().hash(state)
    }
}

impl PartialOrd for RelativeSchemeData {
    fn partial_cmp(&self, other: &RelativeSchemeData) -> Option<Ordering> {
        self.get_identity_key().partial_cmp(&other.get_identity_key())
    }
}

impl Ord for RelativeSchemeData {
    fn cmp(&self, other: &Self) -> Ordering {
        self.get_identity_key().cmp(&other.get_identity_key())
    }
}

impl str::FromStr for Url {
    type Err = ParseError;

    fn from_str(url: &str) -> ParseResult<Url> {
        Url::parse(url)
    }
}

/// A set of optional parameters for URL parsing.
pub struct UrlParser<'a> {
    base_url: Option<&'a Url>,
    query_encoding_override: EncodingOverride,
    error_handler: ErrorHandler,
    scheme_type_mapper: fn(scheme: &str) -> SchemeType,
}


/// A method-chaining API to provide a set of optional parameters for URL parsing.
impl<'a> UrlParser<'a> {
    /// Return a new UrlParser with default parameters.
    #[inline]
    pub fn new() -> UrlParser<'a> {
        fn silent_handler(_reason: ParseError) -> ParseResult<()> { Ok(()) }
        UrlParser {
            base_url: None,
            query_encoding_override: EncodingOverride::utf8(),
            error_handler: silent_handler,
            scheme_type_mapper: whatwg_scheme_type_mapper,
        }
    }

    /// Set the base URL used for resolving relative URL references, and return the `UrlParser`.
    /// The default is no base URL, so that relative URLs references fail to parse.
    #[inline]
    pub fn base_url<'b>(&'b mut self, value: &'a Url) -> &'b mut UrlParser<'a> {
        self.base_url = Some(value);
        self
    }

    /// Set the character encoding the query string is encoded as before percent-encoding,
    /// and return the `UrlParser`.
    ///
    /// This legacy quirk is only relevant to HTML.
    ///
    /// This method is only available if the `query_encoding` Cargo feature is enabled.
    #[cfg(feature = "query_encoding")]
    #[inline]
    pub fn query_encoding_override<'b>(&'b mut self, value: encoding::EncodingRef)
                                       -> &'b mut UrlParser<'a> {
        self.query_encoding_override = EncodingOverride::from_encoding(value);
        self
    }

    /// Set an error handler for non-fatal parse errors, and return the `UrlParser`.
    ///
    /// Non-fatal parse errors are normally ignored by the parser,
    /// but indicate violations of authoring requirements.
    /// An error handler can be used, for example, to log these errors in the console
    /// of a browser’s developer tools.
    ///
    /// The error handler can choose to make the error fatal by returning `Err(..)`
    #[inline]
    pub fn error_handler<'b>(&'b mut self, value: ErrorHandler) -> &'b mut UrlParser<'a> {
        self.error_handler = value;
        self
    }

    /// Set a *scheme type mapper*, and return the `UrlParser`.
    ///
    /// The URL parser behaves differently based on the `SchemeType` of the URL.
    /// See the documentation for `SchemeType` for more details.
    /// A *scheme type mapper* returns a `SchemeType`
    /// based on the scheme as an ASCII lower case string,
    /// as found in the `scheme` field of an `Url` struct.
    ///
    /// The default scheme type mapper is as follows:
    ///
    /// ```
    /// # use url::SchemeType;
    /// fn whatwg_scheme_type_mapper(scheme: &str) -> SchemeType {
    ///     match scheme {
    ///         "file" => SchemeType::FileLike,
    ///         "ftp" => SchemeType::Relative(21),
    ///         "gopher" => SchemeType::Relative(70),
    ///         "http" => SchemeType::Relative(80),
    ///         "https" => SchemeType::Relative(443),
    ///         "ws" => SchemeType::Relative(80),
    ///         "wss" => SchemeType::Relative(443),
    ///         _ => SchemeType::NonRelative,
    ///     }
    /// }
    /// ```
    ///
    /// Note that unknown schemes default to non-relative.
    /// Overriding the scheme type mapper can allow, for example,
    /// parsing URLs in the `git` or `irc` scheme as relative.
    #[inline]
    pub fn scheme_type_mapper<'b>(&'b mut self, value: fn(scheme: &str) -> SchemeType)
                       -> &'b mut UrlParser<'a> {
        self.scheme_type_mapper = value;
        self
    }

    /// Parse `input` as an URL, with all the parameters previously set in the `UrlParser`.
    #[inline]
    pub fn parse(&self, input: &str) -> ParseResult<Url> {
        parser::parse_url(input, self)
    }

    /// Parse `input` as a “standalone” URL path,
    /// with an optional query string and fragment identifier.
    ///
    /// This is typically found in the start line of an HTTP header.
    ///
    /// Note that while the start line has no fragment identifier in the HTTP RFC,
    /// servers typically parse it and ignore it
    /// (rather than having it be part of the path or query string.)
    ///
    /// On success, return `(path, query_string, fragment_identifier)`
    #[inline]
    pub fn parse_path(&self, input: &str)
                      -> ParseResult<(Vec<String>, Option<String>, Option<String>)> {
        parser::parse_standalone_path(input, self)
    }
}


/// Parse `input` as a “standalone” URL path,
/// with an optional query string and fragment identifier.
///
/// This is typically found in the start line of an HTTP header.
///
/// Note that while the start line has no fragment identifier in the HTTP RFC,
/// servers typically parse it and ignore it
/// (rather than having it be part of the path or query string.)
///
/// On success, return `(path, query_string, fragment_identifier)`
///
/// ```rust
/// let (path, query, fragment) = url::parse_path("/foo/bar/../baz?q=42").unwrap();
/// assert_eq!(path, vec!["foo".to_string(), "baz".to_string()]);
/// assert_eq!(query, Some("q=42".to_string()));
/// assert_eq!(fragment, None);
/// ```
///
/// The query string returned by `url::parse_path` can be decoded with
/// `url::form_urlencoded::parse`.
#[inline]
pub fn parse_path(input: &str)
                  -> ParseResult<(Vec<String>, Option<String>, Option<String>)> {
    UrlParser::new().parse_path(input)
}


/// Private convenience methods for use in parser.rs
impl<'a> UrlParser<'a> {
    #[inline]
    fn parse_error(&self, error: ParseError) -> ParseResult<()> {
        (self.error_handler)(error)
    }

    #[inline]
    fn get_scheme_type(&self, scheme: &str) -> SchemeType {
        (self.scheme_type_mapper)(scheme)
    }
}


/// Determines the behavior of the URL parser for a given scheme.
#[derive(PartialEq, Eq, Copy, Debug, Clone, Hash, PartialOrd, Ord)]
pub enum SchemeType {
    /// Indicate that the scheme is *non-relative*.
    ///
    /// The *scheme data* of the URL
    /// (everything other than the scheme, query string, and fragment identifier)
    /// is parsed as a single percent-encoded string of which no structure is assumed.
    /// That string may need to be parsed further, per a scheme-specific format.
    NonRelative,

    /// Indicate that the scheme is *relative*, and what the default port number is.
    ///
    /// The *scheme data* is structured as
    /// *username*, *password*, *host*, *port number*, and *path*.
    /// Relative URL references are supported, if a base URL was given.
    /// The string value indicates the default port number as a string of ASCII digits,
    /// or the empty string to indicate no default port number.
    Relative(u16),

    /// Indicate a *relative* scheme similar to the *file* scheme.
    ///
    /// For example, you might want to have distinct `git+file` and `hg+file` URL schemes.
    ///
    /// This is like `Relative` except the host can be empty, there is no port number,
    /// and path parsing has (platform-independent) quirks to support Windows filenames.
    FileLike,
}

impl SchemeType {
    pub fn default_port(&self) -> Option<u16> {
        match *self {
            SchemeType::Relative(default_port) => Some(default_port),
            _ => None,
        }
    }
    pub fn same_as(&self, other: SchemeType) -> bool {
        match (self, other) {
            (&SchemeType::NonRelative, SchemeType::NonRelative) => true,
            (&SchemeType::Relative(_), SchemeType::Relative(_)) => true,
            (&SchemeType::FileLike,    SchemeType::FileLike) => true,
            _ => false
        }
    }
}

/// http://url.spec.whatwg.org/#special-scheme
pub fn whatwg_scheme_type_mapper(scheme: &str) -> SchemeType {
    match scheme {
        "file" => SchemeType::FileLike,
        "ftp" => SchemeType::Relative(21),
        "gopher" => SchemeType::Relative(70),
        "http" => SchemeType::Relative(80),
        "https" => SchemeType::Relative(443),
        "ws" => SchemeType::Relative(80),
        "wss" => SchemeType::Relative(443),
        _ => SchemeType::NonRelative,
    }
}


impl Url {
    /// Parse an URL with the default `UrlParser` parameters.
    ///
    /// In particular, relative URL references are parse errors since no base URL is provided.
    #[inline]
    pub fn parse(input: &str) -> ParseResult<Url> {
        UrlParser::new().parse(input)
    }

    /// Convert a file name as `std::path::Path` into an URL in the `file` scheme.
    ///
    /// This returns `Err` if the given path is not absolute
    /// or, with a Windows path, if the prefix is not a disk prefix (e.g. `C:`).
    pub fn from_file_path<P: AsRef<Path>>(path: P) -> Result<Url, ()> {
        let path = try!(path_to_file_url_path(path.as_ref()));
        Ok(Url::from_path_common(path))
    }

    /// Convert a directory name as `std::path::Path` into an URL in the `file` scheme.
    ///
    /// This returns `Err` if the given path is not absolute
    /// or, with a Windows path, if the prefix is not a disk prefix (e.g. `C:`).
    ///
    /// Compared to `from_file_path`, this adds an empty component to the path
    /// (or, in terms of URL syntax, adds a trailing slash)
    /// so that the entire path is considered when using this URL as a base URL.
    ///
    /// For example:
    ///
    /// * `"index.html"` parsed with `Url::from_directory_path(Path::new("/var/www"))`
    ///   as the base URL is `file:///var/www/index.html`
    /// * `"index.html"` parsed with `Url::from_file_path(Path::new("/var/www/"))`
    ///   as the base URL is `file:///var/index.html`, which might not be what was intended.
    ///
    /// (Note that `Path::new` removes any trailing slash.)
    pub fn from_directory_path<P: AsRef<Path>>(path: P) -> Result<Url, ()> {
        let mut path = try!(path_to_file_url_path(path.as_ref()));
        // Add an empty path component (i.e. a trailing slash in serialization)
        // so that the entire path is used as a base URL.
        path.push("".to_owned());
        Ok(Url::from_path_common(path))
    }

    fn from_path_common(path: Vec<String>) -> Url {
        Url {
            scheme: "file".to_owned(),
            scheme_data: SchemeData::Relative(RelativeSchemeData {
                username: "".to_owned(),
                password: None,
                port: None,
                default_port: None,
                host: Host::Domain("".to_owned()),
                path: path,
            }),
            query: None,
            fragment: None,
        }
    }

    /// Assuming the URL is in the `file` scheme or similar,
    /// convert its path to an absolute `std::path::Path`.
    ///
    /// **Note:** This does not actually check the URL’s `scheme`,
    /// and may give nonsensical results for other schemes.
    /// It is the user’s responsibility to check the URL’s scheme before calling this.
    ///
    /// ```
    /// # use url::Url;
    /// # let url = Url::parse("file:///etc/passwd").unwrap();
    /// let path = url.to_file_path();
    /// ```
    ///
    /// Returns `Err` if the host is neither empty nor `"localhost"`,
    /// or if `Path::new_opt()` returns `None`.
    /// (That is, if the percent-decoded path contains a NUL byte or,
    /// for a Windows path, is not UTF-8.)
    #[inline]
    pub fn to_file_path(&self) -> Result<PathBuf, ()> {
        match self.scheme_data {
            SchemeData::Relative(ref scheme_data) => scheme_data.to_file_path(),
            SchemeData::NonRelative(..) => Err(()),
        }
    }

    /// Return the serialization of this URL as a string.
    pub fn serialize(&self) -> String {
        self.to_string()
    }

    /// Return the origin of this URL (https://url.spec.whatwg.org/#origin)
    pub fn origin(&self) -> Origin {
        match &*self.scheme {
            "blob" => {
                let result = Url::parse(self.non_relative_scheme_data().unwrap());
                match result {
                    Ok(ref url) => url.origin(),
                    Err(_)  => Origin::UID(OpaqueOrigin::new())
                }
            },
            "ftp" | "gopher" | "http" | "https" | "ws" | "wss" => {
                Origin::Tuple(self.scheme.clone(), self.host().unwrap().clone(),
                    self.port_or_default().unwrap())
            },
            // TODO: Figure out what to do if the scheme is a file
            "file" => Origin::UID(OpaqueOrigin::new()),
            _ => Origin::UID(OpaqueOrigin::new())
        }
    }

    /// Return the serialization of this URL, without the fragment identifier, as a string
    pub fn serialize_no_fragment(&self) -> String {
        UrlNoFragmentFormatter{ url: self }.to_string()
    }

    /// If the URL is *non-relative*, return the string scheme data.
    #[inline]
    pub fn non_relative_scheme_data(&self) -> Option<&str> {
        match self.scheme_data {
            SchemeData::Relative(..) => None,
            SchemeData::NonRelative(ref scheme_data) => Some(scheme_data),
        }
    }

    /// If the URL is *non-relative*, return a mutable reference to the string scheme data.
    #[inline]
    pub fn non_relative_scheme_data_mut(&mut self) -> Option<&mut String> {
        match self.scheme_data {
            SchemeData::Relative(..) => None,
            SchemeData::NonRelative(ref mut scheme_data) => Some(scheme_data),
        }
    }

    /// If the URL is in a *relative scheme*, return the structured scheme data.
    #[inline]
    pub fn relative_scheme_data(&self) -> Option<&RelativeSchemeData> {
        match self.scheme_data {
            SchemeData::Relative(ref scheme_data) => Some(scheme_data),
            SchemeData::NonRelative(..) => None,
        }
    }

    /// If the URL is in a *relative scheme*,
    /// return a mutable reference to the structured scheme data.
    #[inline]
    pub fn relative_scheme_data_mut(&mut self) -> Option<&mut RelativeSchemeData> {
        match self.scheme_data {
            SchemeData::Relative(ref mut scheme_data) => Some(scheme_data),
            SchemeData::NonRelative(..) => None,
        }
    }

    /// If the URL is in a *relative scheme*, return its username.
    #[inline]
    pub fn username(&self) -> Option<&str> {
        self.relative_scheme_data().map(|scheme_data| &*scheme_data.username)
    }

    /// If the URL is in a *relative scheme*, return a mutable reference to its username.
    #[inline]
    pub fn username_mut(&mut self) -> Option<&mut String> {
        self.relative_scheme_data_mut().map(|scheme_data| &mut scheme_data.username)
    }

    /// Percent-decode the URL’s username, if any.
    ///
    /// This is “lossy”: invalid UTF-8 percent-encoded byte sequences
    /// will be replaced � U+FFFD, the replacement character.
    #[inline]
    pub fn lossy_percent_decode_username(&self) -> Option<String> {
        self.relative_scheme_data().map(|scheme_data| scheme_data.lossy_percent_decode_username())
    }

    /// If the URL is in a *relative scheme*, return its password, if any.
    #[inline]
    pub fn password(&self) -> Option<&str> {
        self.relative_scheme_data().and_then(|scheme_data|
            scheme_data.password.as_ref().map(|password| password as &str))
    }

    /// If the URL is in a *relative scheme*, return a mutable reference to its password, if any.
    #[inline]
    pub fn password_mut(&mut self) -> Option<&mut String> {
        self.relative_scheme_data_mut().and_then(|scheme_data| scheme_data.password.as_mut())
    }

    /// Percent-decode the URL’s password, if any.
    ///
    /// This is “lossy”: invalid UTF-8 percent-encoded byte sequences
    /// will be replaced � U+FFFD, the replacement character.
    #[inline]
    pub fn lossy_percent_decode_password(&self) -> Option<String> {
        self.relative_scheme_data().and_then(|scheme_data|
            scheme_data.lossy_percent_decode_password())
    }

    /// Serialize the URL's username and password, if any.
    ///
    /// Format: "<username>:<password>@"
    #[inline]
    pub fn serialize_userinfo(&mut self) -> Option<String> {
        self.relative_scheme_data().map(|scheme_data| scheme_data.serialize_userinfo())
    }

    /// If the URL is in a *relative scheme*, return its structured host.
    #[inline]
    pub fn host(&self) -> Option<&Host> {
        self.relative_scheme_data().map(|scheme_data| &scheme_data.host)
    }

    /// If the URL is in a *relative scheme*, return a mutable reference to its structured host.
    #[inline]
    pub fn host_mut(&mut self) -> Option<&mut Host> {
        self.relative_scheme_data_mut().map(|scheme_data| &mut scheme_data.host)
    }

    /// If the URL is in a *relative scheme* and its host is a domain,
    /// return the domain as a string.
    #[inline]
    pub fn domain(&self) -> Option<&str> {
        self.relative_scheme_data().and_then(|scheme_data| scheme_data.domain())
    }

    /// If the URL is in a *relative scheme* and its host is a domain,
    /// return a mutable reference to the domain string.
    #[inline]
    pub fn domain_mut(&mut self) -> Option<&mut String> {
        self.relative_scheme_data_mut().and_then(|scheme_data| scheme_data.domain_mut())
    }

    /// If the URL is in a *relative scheme*, serialize its host as a string.
    ///
    /// A domain a returned as-is, an IPv6 address between [] square brackets.
    #[inline]
    pub fn serialize_host(&self) -> Option<String> {
        self.relative_scheme_data().map(|scheme_data| scheme_data.host.serialize())
    }

    /// If the URL is in a *relative scheme* and has a port number, return it.
    #[inline]
    pub fn port(&self) -> Option<u16> {
        self.relative_scheme_data().and_then(|scheme_data| scheme_data.port)
    }

    /// If the URL is in a *relative scheme*, return a mutable reference to its port.
    #[inline]
    pub fn port_mut(&mut self) -> Option<&mut Option<u16>> {
        self.relative_scheme_data_mut().map(|scheme_data| &mut scheme_data.port)
    }

    /// If the URL is in a *relative scheme* that is not a file-like,
    /// return its port number, even if it is the default.
    #[inline]
    pub fn port_or_default(&self) -> Option<u16> {
        self.relative_scheme_data().and_then(|scheme_data| scheme_data.port_or_default())
    }

    /// If the URL is in a *relative scheme*, return its path components.
    #[inline]
    pub fn path(&self) -> Option<&[String]> {
        self.relative_scheme_data().map(|scheme_data| &*scheme_data.path)
    }

    /// If the URL is in a *relative scheme*, return a mutable reference to its path components.
    #[inline]
    pub fn path_mut(&mut self) -> Option<&mut Vec<String>> {
        self.relative_scheme_data_mut().map(|scheme_data| &mut scheme_data.path)
    }

    /// If the URL is in a *relative scheme*, serialize its path as a string.
    ///
    /// The returned string starts with a "/" slash, and components are separated by slashes.
    /// A trailing slash represents an empty last component.
    #[inline]
    pub fn serialize_path(&self) -> Option<String> {
        self.relative_scheme_data().map(|scheme_data| scheme_data.serialize_path())
    }

    /// Parse the URL’s query string, if any, as `application/x-www-form-urlencoded`
    /// and return a vector of (key, value) pairs.
    #[inline]
    pub fn query_pairs(&self) -> Option<Vec<(String, String)>> {
        self.query.as_ref().map(|query| form_urlencoded::parse(query.as_bytes()))
    }

    /// Serialize an iterator of (key, value) pairs as `application/x-www-form-urlencoded`
    /// and set it as the URL’s query string.
    #[inline]
    pub fn set_query_from_pairs<I, K, V>(&mut self, pairs: I)
    where I: IntoIterator, I::Item: Borrow<(K, V)>, K: AsRef<str>, V: AsRef<str> {
        self.query = Some(form_urlencoded::serialize(pairs));
    }

    /// Percent-decode the URL’s query string, if any.
    ///
    /// This is “lossy”: invalid UTF-8 percent-encoded byte sequences
    /// will be replaced � U+FFFD, the replacement character.
    #[inline]
    pub fn lossy_percent_decode_query(&self) -> Option<String> {
        self.query.as_ref().map(|value| lossy_utf8_percent_decode(value.as_bytes()))
    }

    /// Percent-decode the URL’s fragment identifier, if any.
    ///
    /// This is “lossy”: invalid UTF-8 percent-encoded byte sequences
    /// will be replaced � U+FFFD, the replacement character.
    #[inline]
    pub fn lossy_percent_decode_fragment(&self) -> Option<String> {
        self.fragment.as_ref().map(|value| lossy_utf8_percent_decode(value.as_bytes()))
    }

    /// Join a path with a base URL.
    ///
    /// Corresponds to the basic URL parser where `self` is the given base URL.
    #[inline]
    pub fn join(&self, input: &str) -> ParseResult<Url> {
        UrlParser::new().base_url(self).parse(input)
    }
}


impl rustc_serialize::Encodable for Url {
    fn encode<S: rustc_serialize::Encoder>(&self, encoder: &mut S) -> Result<(), S::Error> {
        encoder.emit_str(&self.to_string())
    }
}


impl rustc_serialize::Decodable for Url {
    fn decode<D: rustc_serialize::Decoder>(decoder: &mut D) -> Result<Url, D::Error> {
        Url::parse(&*try!(decoder.read_str())).map_err(|error| {
            decoder.error(&format!("URL parsing error: {}", error))
        })
    }
}

/// Serializes this URL into a `serde` stream.
///
/// This implementation is only available if the `serde_serialization` Cargo feature is enabled.
#[cfg(feature="serde_serialization")]
impl serde::Serialize for Url {
    fn serialize<S>(&self, serializer: &mut S) -> Result<(), S::Error> where S: serde::Serializer {
        format!("{}", self).serialize(serializer)
    }
}

/// Deserializes this URL from a `serde` stream.
///
/// This implementation is only available if the `serde_serialization` Cargo feature is enabled.
#[cfg(feature="serde_serialization")]
impl serde::Deserialize for Url {
    fn deserialize<D>(deserializer: &mut D) -> Result<Url, D::Error> where D: serde::Deserializer {
        let string_representation: String = try!(serde::Deserialize::deserialize(deserializer));
        Ok(FromStr::from_str(&string_representation[..]).unwrap())
    }
}

impl fmt::Display for Url {
    fn fmt(&self, formatter: &mut Formatter) -> fmt::Result {
        try!(UrlNoFragmentFormatter{ url: self }.fmt(formatter));
        if let Some(ref fragment) = self.fragment {
            try!(formatter.write_str("#"));
            try!(formatter.write_str(fragment));
        }
        Ok(())
    }
}


impl fmt::Display for SchemeData {
    fn fmt(&self, formatter: &mut Formatter) -> fmt::Result {
        match *self {
            SchemeData::Relative(ref scheme_data) => scheme_data.fmt(formatter),
            SchemeData::NonRelative(ref scheme_data) => scheme_data.fmt(formatter),
        }
    }
}


impl RelativeSchemeData {
    /// Percent-decode the URL’s username.
    ///
    /// This is “lossy”: invalid UTF-8 percent-encoded byte sequences
    /// will be replaced � U+FFFD, the replacement character.
    #[inline]
    pub fn lossy_percent_decode_username(&self) -> String {
        lossy_utf8_percent_decode(self.username.as_bytes())
    }

    /// Percent-decode the URL’s password, if any.
    ///
    /// This is “lossy”: invalid UTF-8 percent-encoded byte sequences
    /// will be replaced � U+FFFD, the replacement character.
    #[inline]
    pub fn lossy_percent_decode_password(&self) -> Option<String> {
        self.password.as_ref().map(|value| lossy_utf8_percent_decode(value.as_bytes()))
    }

    /// Assuming the URL is in the `file` scheme or similar,
    /// convert its path to an absolute `std::path::Path`.
    ///
    /// **Note:** This does not actually check the URL’s `scheme`,
    /// and may give nonsensical results for other schemes.
    /// It is the user’s responsibility to check the URL’s scheme before calling this.
    ///
    /// ```
    /// # use url::Url;
    /// # let url = Url::parse("file:///etc/passwd").unwrap();
    /// let path = url.to_file_path();
    /// ```
    ///
    /// Returns `Err` if the host is neither empty nor `"localhost"`,
    /// or if `Path::new_opt()` returns `None`.
    /// (That is, if the percent-decoded path contains a NUL byte or,
    /// for a Windows path, is not UTF-8.)
    #[inline]
    pub fn to_file_path(&self) -> Result<PathBuf, ()> {
        // FIXME: Figure out what to do w.r.t host.
        if !matches!(self.domain(), Some("") | Some("localhost")) {
            return Err(())
        }
        file_url_path_to_pathbuf(&self.path)
    }

    /// If the host is a domain, return the domain as a string.
    #[inline]
    pub fn domain(&self) -> Option<&str> {
        match self.host {
            Host::Domain(ref domain) => Some(domain),
            _ => None,
        }
    }

    /// If the host is a domain, return a mutable reference to the domain string.
    #[inline]
    pub fn domain_mut(&mut self) -> Option<&mut String> {
        match self.host {
            Host::Domain(ref mut domain) => Some(domain),
            _ => None,
        }
    }

    /// Return the port number of the URL, even if it is the default.
    /// Return `None` for file-like URLs.
    #[inline]
    pub fn port_or_default(&self) -> Option<u16> {
        self.port.or(self.default_port)
    }

    /// Serialize the path as a string.
    ///
    /// The returned string starts with a "/" slash, and components are separated by slashes.
    /// A trailing slash represents an empty last component.
    pub fn serialize_path(&self) -> String {
        PathFormatter {
            path: &self.path
        }.to_string()
    }

    /// Serialize the userinfo as a string.
    ///
    /// Format: "<username>:<password>@".
    pub fn serialize_userinfo(&self) -> String {
        UserInfoFormatter {
            username: &self.username,
            password: self.password.as_ref().map(|s| s as &str)
        }.to_string()
    }
}


impl fmt::Display for RelativeSchemeData {
    fn fmt(&self, formatter: &mut Formatter) -> fmt::Result {
        // Write the scheme-trailing double slashes.
        try!(formatter.write_str("//"));

        // Write the user info.
        try!(UserInfoFormatter {
            username: &self.username,
            password: self.password.as_ref().map(|s| s as &str)
        }.fmt(formatter));

        // Write the host.
        try!(self.host.fmt(formatter));

        // Write the port.
        match self.port {
            Some(port) => {
                try!(write!(formatter, ":{}", port));
            },
            None => {}
        }

        // Write the path.
        PathFormatter {
            path: &self.path
        }.fmt(formatter)
    }
}


#[cfg(unix)]
fn path_to_file_url_path(path: &Path) -> Result<Vec<String>, ()> {
    use std::os::unix::prelude::OsStrExt;
    if !path.is_absolute() {
        return Err(())
    }
    // skip the root component
    Ok(path.components().skip(1).map(|c| {
        percent_encode(c.as_os_str().as_bytes(), DEFAULT_ENCODE_SET)
    }).collect())
}

#[cfg(windows)]
fn path_to_file_url_path(path: &Path) -> Result<Vec<String>, ()> {
    path_to_file_url_path_windows(path)
}

// Build this unconditionally to alleviate https://github.com/servo/rust-url/issues/102
#[cfg_attr(not(windows), allow(dead_code))]
fn path_to_file_url_path_windows(path: &Path) -> Result<Vec<String>, ()> {
    use std::path::{Prefix, Component};
    if !path.is_absolute() {
        return Err(())
    }
    let mut components = path.components();
    let disk = match components.next() {
        Some(Component::Prefix(ref p)) => match p.kind() {
            Prefix::Disk(byte) => byte,
            _ => return Err(()),
        },

        // FIXME: do something with UNC and other prefixes?
        _ => return Err(())
    };

    // Start with the prefix, e.g. "C:"
    let mut path = vec![format!("{}:", disk as char)];

    for component in components {
        if component == Component::RootDir { continue }
        // FIXME: somehow work with non-unicode?
        let part = match component.as_os_str().to_str() {
            Some(s) => s,
            None => return Err(()),
        };
        path.push(percent_encode(part.as_bytes(), DEFAULT_ENCODE_SET));
    }
    Ok(path)
}

#[cfg(unix)]
fn file_url_path_to_pathbuf(path: &[String]) -> Result<PathBuf, ()> {
    use std::ffi::OsStr;
    use std::os::unix::prelude::OsStrExt;
    use std::path::PathBuf;

    use percent_encoding::percent_decode_to;

    if path.is_empty() {
        return Ok(PathBuf::from("/"))
    }
    let mut bytes = Vec::new();
    for path_part in path {
        bytes.push(b'/');
        percent_decode_to(path_part.as_bytes(), &mut bytes);
    }
    let os_str = OsStr::from_bytes(&bytes);
    let path = PathBuf::from(os_str);
    debug_assert!(path.is_absolute(),
                  "to_file_path() failed to produce an absolute Path");
    Ok(path)
}

#[cfg(windows)]
fn file_url_path_to_pathbuf(path: &[String]) -> Result<PathBuf, ()> {
    file_url_path_to_pathbuf_windows(path)
}

// Build this unconditionally to alleviate https://github.com/servo/rust-url/issues/102
#[cfg_attr(not(windows), allow(dead_code))]
fn file_url_path_to_pathbuf_windows(path: &[String]) -> Result<PathBuf, ()> {
    use percent_encoding::percent_decode;

    if path.is_empty() {
        return Err(())
    }
    let prefix = &*path[0];
    if prefix.len() != 2 || !parser::starts_with_ascii_alpha(prefix)
            || prefix.as_bytes()[1] != b':' {
        return Err(())
    }
    let mut string = prefix.to_owned();
    for path_part in &path[1..] {
        string.push('\\');

        // Currently non-unicode windows paths cannot be represented
        match String::from_utf8(percent_decode(path_part.as_bytes())) {
            Ok(s) => string.push_str(&s),
            Err(..) => return Err(()),
        }
    }
    let path = PathBuf::from(string);
    debug_assert!(path.is_absolute(),
                  "to_file_path() failed to produce an absolute Path");
    Ok(path)
}