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
#![doc(html_root_url = "https://docs.rs/httparse/1.3.2")]
#![cfg_attr(not(feature = "std"), no_std)]
#![cfg_attr(test, deny(warnings))]
#![deny(missing_docs)]

//! # httparse
//!
//! A push library for parsing HTTP/1.x requests and responses.
//!
//! The focus is on speed and safety. Unsafe code is used to keep parsing fast,
//! but unsafety is contained in a submodule, with invariants enforced. The
//! parsing internals use an `Iterator` instead of direct indexing, while
//! skipping bounds checks.
//!
//! With Rust 1.27.0 or later, support for SIMD is enabled automatically.
//! If building an executable to be run on multiple platforms, and thus
//! not passing `target_feature` or `target_cpu` flags to the compiler,
//! runtime detection can still detect SSE4.2 or AVX2 support to provide
//! massive wins.
//!
//! If compiling for a specific target, remembering to include
//! `-C target_cpu=native` allows the detection to become compile time checks,
//! making it *even* faster.
#[cfg(feature = "std")]
extern crate std as core;

use core::{fmt, result, str, slice};

use iter::Bytes;

mod iter;
#[macro_use] mod macros;
mod simd;

#[inline]
fn shrink<T>(slice: &mut &mut [T], len: usize) {
    debug_assert!(slice.len() >= len);
    let ptr = slice.as_mut_ptr();
    *slice = unsafe { slice::from_raw_parts_mut(ptr, len) };
}

/// Determines if byte is a token char.
///
/// > ```notrust
/// > token          = 1*tchar
/// >
/// > tchar          = "!" / "#" / "$" / "%" / "&" / "'" / "*"
/// >                / "+" / "-" / "." / "^" / "_" / "`" / "|" / "~"
/// >                / DIGIT / ALPHA
/// >                ; any VCHAR, except delimiters
/// > ```
#[inline]
fn is_token(b: u8) -> bool {
    b > 0x1F && b < 0x7F
}

// ASCII codes to accept URI string.
// i.e. A-Z a-z 0-9 !#$%&'*+-._();:@=,/?[]~
// TODO: Make a stricter checking for URI string?
static URI_MAP: [bool; 256] = byte_map![
    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
    0, 1, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
    1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 1, 0, 1,
    1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
    1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 1, 0, 1,
    0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
    1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 1, 0,
    // ====== Extended ASCII (aka. obs-text) ======
    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
];

#[inline]
fn is_uri_token(b: u8) -> bool {
    URI_MAP[b as usize]
}

static HEADER_NAME_MAP: [bool; 256] = byte_map![
    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
    0, 1, 0, 1, 1, 1, 1, 1, 0, 0, 1, 1, 0, 1, 1, 0,
    1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0,
    0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
    1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 1, 1,
    1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
    1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 1, 0, 1, 0,
    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
];

#[inline]
fn is_header_name_token(b: u8) -> bool {
    HEADER_NAME_MAP[b as usize]
}

static HEADER_VALUE_MAP: [bool; 256] = byte_map![
    0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0,
    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
    1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
    1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
    1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
    1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
    1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
    1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0,
    1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
    1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
    1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
    1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
    1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
    1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
    1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
    1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
];


#[inline]
fn is_header_value_token(b: u8) -> bool {
    HEADER_VALUE_MAP[b as usize]
}

/// An error in parsing.
#[derive(Copy, Clone, PartialEq, Eq, Debug)]
pub enum Error {
    /// Invalid byte in header name.
    HeaderName,
    /// Invalid byte in header value.
    HeaderValue,
    /// Invalid byte in new line.
    NewLine,
    /// Invalid byte in Response status.
    Status,
    /// Invalid byte where token is required.
    Token,
    /// Parsed more headers than provided buffer can contain.
    TooManyHeaders,
    /// Invalid byte in HTTP version.
    Version,
}

impl Error {
    #[inline]
    fn description_str(&self) -> &'static str {
        match *self {
            Error::HeaderName => "invalid header name",
            Error::HeaderValue => "invalid header value",
            Error::NewLine => "invalid new line",
            Error::Status => "invalid response status",
            Error::Token => "invalid token",
            Error::TooManyHeaders => "too many headers",
            Error::Version => "invalid HTTP version",
        }
    }
}

impl fmt::Display for Error {
    fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
        f.write_str(self.description_str())
    }
}

#[cfg(feature = "std")]
impl std::error::Error for Error {
    fn description(&self) -> &str {
        self.description_str()
    }
}

/// An error in parsing a chunk size.
// Note: Move this into the error enum once v2.0 is released.
#[derive(Debug, PartialEq, Eq)]
pub struct InvalidChunkSize;

impl fmt::Display for InvalidChunkSize {
    fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
        f.write_str("invalid chunk size")
    }
}

/// A Result of any parsing action.
///
/// If the input is invalid, an `Error` will be returned. Note that incomplete
/// data is not considered invalid, and so will not return an error, but rather
/// a `Ok(Status::Partial)`.
pub type Result<T> = result::Result<Status<T>, Error>;

/// The result of a successful parse pass.
///
/// `Complete` is used when the buffer contained the complete value.
/// `Partial` is used when parsing did not reach the end of the expected value,
/// but no invalid data was found.
#[derive(Copy, Clone, PartialEq, Debug)]
pub enum Status<T> {
    /// The completed result.
    Complete(T),
    /// A partial result.
    Partial
}

impl<T> Status<T> {
    /// Convenience method to check if status is complete.
    #[inline]
    pub fn is_complete(&self) -> bool {
        match *self {
            Status::Complete(..) => true,
            Status::Partial => false
        }
    }

    /// Convenience method to check if status is partial.
    #[inline]
    pub fn is_partial(&self) -> bool {
        match *self {
            Status::Complete(..) => false,
            Status::Partial => true
        }
    }

    /// Convenience method to unwrap a Complete value. Panics if the status is
    /// `Partial`.
    #[inline]
    pub fn unwrap(self) -> T {
        match self {
            Status::Complete(t) => t,
            Status::Partial => panic!("Tried to unwrap Status::Partial")
        }
    }
}

/// A parsed Request.
///
/// The optional values will be `None` if a parse was not complete, and did not
/// parse the associated property. This allows you to inspect the parts that
/// could be parsed, before reading more, in case you wish to exit early.
///
/// # Example
///
/// ```no_run
/// let buf = b"GET /404 HTTP/1.1\r\nHost:";
/// let mut headers = [httparse::EMPTY_HEADER; 16];
/// let mut req = httparse::Request::new(&mut headers);
/// let res = req.parse(buf).unwrap();
/// if res.is_partial() {
///     match req.path {
///         Some(ref path) => {
///             // check router for path.
///             // /404 doesn't exist? we could stop parsing
///         },
///         None => {
///             // must read more and parse again
///         }
///     }
/// }
/// ```
#[derive(Debug, PartialEq)]
pub struct Request<'headers, 'buf: 'headers> {
    /// The request method, such as `GET`.
    pub method: Option<&'buf str>,
    /// The request path, such as `/about-us`.
    pub path: Option<&'buf str>,
    /// The request version, such as `HTTP/1.1`.
    pub version: Option<u8>,
    /// The request headers.
    pub headers: &'headers mut [Header<'buf>]
}

impl<'h, 'b> Request<'h, 'b> {
    /// Creates a new Request, using a slice of headers you allocate.
    #[inline]
    pub fn new(headers: &'h mut [Header<'b>]) -> Request<'h, 'b> {
        Request {
            method: None,
            path: None,
            version: None,
            headers: headers,
        }
    }

    /// Try to parse a buffer of bytes into the Request.
    pub fn parse(&mut self, buf: &'b [u8]) -> Result<usize> {
        let orig_len = buf.len();
        let mut bytes = Bytes::new(buf);
        complete!(skip_empty_lines(&mut bytes));
        self.method = Some(complete!(parse_token(&mut bytes)));
        self.path = Some(complete!(parse_uri(&mut bytes)));
        self.version = Some(complete!(parse_version(&mut bytes)));
        newline!(bytes);

        let len = orig_len - bytes.len();
        let headers_len = complete!(parse_headers_iter(&mut self.headers, &mut bytes));

        Ok(Status::Complete(len + headers_len))
    }
}

#[inline]
fn skip_empty_lines(bytes: &mut Bytes) -> Result<()> {
    loop {
        let b = bytes.peek();
        match b {
            Some(b'\r') => {
                // there's `\r`, so it's safe to bump 1 pos
                unsafe { bytes.bump() };
                expect!(bytes.next() == b'\n' => Err(Error::NewLine));
            },
            Some(b'\n') => {
                // there's `\n`, so it's safe to bump 1 pos
                unsafe { bytes.bump(); }
            },
            Some(..) => {
                bytes.slice();
                return Ok(Status::Complete(()));
            },
            None => return Ok(Status::Partial)
        }
    }
}

/// A parsed Response.
///
/// See `Request` docs for explanation of optional values.
#[derive(Debug, PartialEq)]
pub struct Response<'headers, 'buf: 'headers> {
    /// The response version, such as `HTTP/1.1`.
    pub version: Option<u8>,
    /// The response code, such as `200`.
    pub code: Option<u16>,
    /// The response reason-phrase, such as `OK`.
    pub reason: Option<&'buf str>,
    /// The response headers.
    pub headers: &'headers mut [Header<'buf>]
}

impl<'h, 'b> Response<'h, 'b> {
    /// Creates a new `Response` using a slice of `Header`s you have allocated.
    #[inline]
    pub fn new(headers: &'h mut [Header<'b>]) -> Response<'h, 'b> {
        Response {
            version: None,
            code: None,
            reason: None,
            headers: headers,
        }
    }

    /// Try to parse a buffer of bytes into this `Response`.
    pub fn parse(&mut self, buf: &'b [u8]) -> Result<usize> {
        let orig_len = buf.len();
        let mut bytes = Bytes::new(buf);

        complete!(skip_empty_lines(&mut bytes));
        self.version = Some(complete!(parse_version(&mut bytes)));
        space!(bytes or Error::Version);
        self.code = Some(complete!(parse_code(&mut bytes)));

        // RFC7230 says there must be 'SP' and then reason-phrase, but admits
        // its only for legacy reasons. With the reason-phrase completely
        // optional (and preferred to be omitted) in HTTP2, we'll just
        // handle any response that doesn't include a reason-phrase, because
        // it's more lenient, and we don't care anyways.
        //
        // So, a SP means parse a reason-phrase.
        // A newline means go to headers.
        // Anything else we'll say is a malformed status.
        match next!(bytes) {
            b' ' => {
                bytes.slice();
                self.reason = Some(complete!(parse_reason(&mut bytes)));
            },
            b'\r' => {
                expect!(bytes.next() == b'\n' => Err(Error::Status));
                bytes.slice();
                self.reason = Some("");
            },
            b'\n' => self.reason = Some(""),
            _ => return Err(Error::Status),
        }


        let len = orig_len - bytes.len();
        let headers_len = complete!(parse_headers_iter(&mut self.headers, &mut bytes));
        Ok(Status::Complete(len + headers_len))
    }
}

/// Represents a parsed header.
#[derive(Copy, Clone, PartialEq, Debug)]
pub struct Header<'a> {
    /// The name portion of a header.
    ///
    /// A header name must be valid ASCII-US, so it's safe to store as a `&str`.
    pub name: &'a str,
    /// The value portion of a header.
    ///
    /// While headers **should** be ASCII-US, the specification allows for
    /// values that may not be, and so the value is stored as bytes.
    pub value: &'a [u8],
}

/// An empty header, useful for constructing a `Header` array to pass in for
/// parsing.
///
/// # Example
///
/// ```
/// let headers = [httparse::EMPTY_HEADER; 64];
/// ```
pub const EMPTY_HEADER: Header<'static> = Header { name: "", value: b"" };

#[inline]
fn parse_version(bytes: &mut Bytes) -> Result<u8> {
    if let Some(mut eight) = bytes.next_8() {
        expect!(eight._0() => b'H' |? Err(Error::Version));
        expect!(eight._1() => b'T' |? Err(Error::Version));
        expect!(eight._2() => b'T' |? Err(Error::Version));
        expect!(eight._3() => b'P' |? Err(Error::Version));
        expect!(eight._4() => b'/' |? Err(Error::Version));
        expect!(eight._5() => b'1' |? Err(Error::Version));
        expect!(eight._6() => b'.' |? Err(Error::Version));
        let v = match eight._7() {
            b'0' => 0,
            b'1' => 1,
            _ => return Err(Error::Version)
        };
        return Ok(Status::Complete(v))
    }

    // else (but not in `else` because of borrow checker)

    // If there aren't at least 8 bytes, we still want to detect early
    // if this is a valid version or not. If it is, we'll return Partial.
    expect!(bytes.next() == b'H' => Err(Error::Version));
    expect!(bytes.next() == b'T' => Err(Error::Version));
    expect!(bytes.next() == b'T' => Err(Error::Version));
    expect!(bytes.next() == b'P' => Err(Error::Version));
    expect!(bytes.next() == b'/' => Err(Error::Version));
    expect!(bytes.next() == b'1' => Err(Error::Version));
    expect!(bytes.next() == b'.' => Err(Error::Version));
    Ok(Status::Partial)
}

/// From [RFC 7230](https://tools.ietf.org/html/rfc7230):
///
/// > ```notrust
/// > reason-phrase  = *( HTAB / SP / VCHAR / obs-text )
/// > HTAB           = %x09        ; horizontal tab
/// > VCHAR          = %x21-7E     ; visible (printing) characters
/// > obs-text       = %x80-FF
/// > ```
///
/// > A.2.  Changes from RFC 2616
/// >
/// > Non-US-ASCII content in header fields and the reason phrase
/// > has been obsoleted and made opaque (the TEXT rule was removed).
///
/// Note that the following implementation deliberately rejects the obsoleted (non-US-ASCII) text range.
///
/// The fully compliant parser should probably just return the reason-phrase as an opaque &[u8] data
/// and leave interpretation to user or specialized helpers (akin to .display() in std::path::Path)
#[inline]
fn parse_reason<'a>(bytes: &mut Bytes<'a>) -> Result<&'a str> {
    loop {
        let b = next!(bytes);
        if b == b'\r' {
            expect!(bytes.next() == b'\n' => Err(Error::Status));
            return Ok(Status::Complete(unsafe {
                // all bytes up till `i` must have been HTAB / SP / VCHAR
                str::from_utf8_unchecked(bytes.slice_skip(2))
            }));
        } else if b == b'\n' {
            return Ok(Status::Complete(unsafe {
                // all bytes up till `i` must have been HTAB / SP / VCHAR
                str::from_utf8_unchecked(bytes.slice_skip(1))
            }));
        } else if !((b >= 0x20 && b <= 0x7E) || b == b'\t') {
            return Err(Error::Status);
        }
    }
}

#[inline]
fn parse_token<'a>(bytes: &mut Bytes<'a>) -> Result<&'a str> {
    loop {
        let b = next!(bytes);
        if b == b' ' {
            return Ok(Status::Complete(unsafe {
                // all bytes up till `i` must have been `is_token`.
                str::from_utf8_unchecked(bytes.slice_skip(1))
            }));
        } else if !is_token(b) {
            return Err(Error::Token);
        }
    }
}

#[inline]
fn parse_uri<'a>(bytes: &mut Bytes<'a>) -> Result<&'a str> {
    simd::match_uri_vectored(bytes);

    loop {
        let b = next!(bytes);
        if b == b' ' {
            return Ok(Status::Complete(unsafe {
                // all bytes up till `i` must have been `is_token`.
                str::from_utf8_unchecked(bytes.slice_skip(1))
            }));
        } else if !is_uri_token(b) {
            return Err(Error::Token);
        }
    }
}


#[inline]
fn parse_code(bytes: &mut Bytes) -> Result<u16> {
    let hundreds = expect!(bytes.next() == b'0'...b'9' => Err(Error::Status));
    let tens = expect!(bytes.next() == b'0'...b'9' => Err(Error::Status));
    let ones = expect!(bytes.next() == b'0'...b'9' => Err(Error::Status));

    Ok(Status::Complete((hundreds - b'0') as u16 * 100 +
        (tens - b'0') as u16 * 10 +
        (ones - b'0') as u16))
}

/// Parse a buffer of bytes as headers.
///
/// The return value, if complete and successful, includes the index of the
/// buffer that parsing stopped at, and a sliced reference to the parsed
/// headers. The length of the slice will be equal to the number of properly
/// parsed headers.
///
/// # Example
///
/// ```
/// let buf = b"Host: foo.bar\nAccept: */*\n\nblah blah";
/// let mut headers = [httparse::EMPTY_HEADER; 4];
/// assert_eq!(httparse::parse_headers(buf, &mut headers),
///            Ok(httparse::Status::Complete((27, &[
///                httparse::Header { name: "Host", value: b"foo.bar" },
///                httparse::Header { name: "Accept", value: b"*/*" }
///            ][..]))));
/// ```
pub fn parse_headers<'b: 'h, 'h>(src: &'b [u8], mut dst: &'h mut [Header<'b>])
    -> Result<(usize, &'h [Header<'b>])> {
    let mut iter = Bytes::new(src);
    let pos = complete!(parse_headers_iter(&mut dst, &mut iter));
    Ok(Status::Complete((pos, dst)))
}

/*
#[cfg(not(any(
    feature = "httparse_simd",
    target_arch = "x86",
    target_arch = "x86_64",
    target_feature = "sse4.2"
)))]
*/


#[inline]
fn parse_headers_iter<'a, 'b>(headers: &mut &mut [Header<'a>], bytes: &'b mut Bytes<'a>)
    -> Result<usize> {
    let mut num_headers: usize = 0;
    let mut count: usize = 0;
    let mut result = Err(Error::TooManyHeaders);

    {
        let mut iter = headers.iter_mut();

        'headers: loop {
            // a newline here means the head is over!
            let b = next!(bytes);
            if b == b'\r' {
                expect!(bytes.next() == b'\n' => Err(Error::NewLine));
                result = Ok(Status::Complete(count + bytes.pos()));
                break;
            } else if b == b'\n' {
                result = Ok(Status::Complete(count + bytes.pos()));
                break;
            } else if !is_header_name_token(b) {
                return Err(Error::HeaderName);
            }

            let header = match iter.next() {
                Some(header) => header,
                None => break 'headers
            };

            num_headers += 1;
            // parse header name until colon
            'name: loop {
                let b = next!(bytes);
                if b == b':' {
                    count += bytes.pos();
                    header.name = unsafe {
                        str::from_utf8_unchecked(bytes.slice_skip(1))
                    };
                    break 'name;
                } else if !is_header_name_token(b) {
                    return Err(Error::HeaderName);
                }
            }

            let mut b;

            'value: loop {

                // eat white space between colon and value
                'whitespace: loop {
                    b = next!(bytes);
                    if b == b' ' || b == b'\t' {
                        count += bytes.pos();
                        bytes.slice();
                        continue 'whitespace;
                    } else {
                        if !is_header_value_token(b) {
                            break 'value;
                        }
                        break 'whitespace;
                    }
                }

                // parse value till EOL

                simd::match_header_value_vectored(bytes);

                macro_rules! check {
                    ($bytes:ident, $i:ident) => ({
                        b = $bytes.$i();
                        if !is_header_value_token(b) {
                            break 'value;
                        }
                    });
                    ($bytes:ident) => ({
                        check!($bytes, _0);
                        check!($bytes, _1);
                        check!($bytes, _2);
                        check!($bytes, _3);
                        check!($bytes, _4);
                        check!($bytes, _5);
                        check!($bytes, _6);
                        check!($bytes, _7);
                    })
                }
                while let Some(mut bytes8) = bytes.next_8() {
                    check!(bytes8);
                }
                loop {
                    b = next!(bytes);
                    if !is_header_value_token(b) {
                        break 'value;
                    }
                }
            }

            //found_ctl
            if b == b'\r' {
                expect!(bytes.next() == b'\n' => Err(Error::HeaderValue));
                count += bytes.pos();
                // having just check that `\r\n` exists, it's safe to skip those 2 bytes
                unsafe {
                    header.value = bytes.slice_skip(2);
                }
            } else if b == b'\n' {
                count += bytes.pos();
                // having just check that `\r\n` exists, it's safe to skip 1 byte
                unsafe {
                    header.value = bytes.slice_skip(1);
                }
            } else {
                return Err(Error::HeaderValue);
            }
        }
    } // drop iter

    shrink(headers, num_headers);
    result
}

/// Parse a buffer of bytes as a chunk size.
///
/// The return value, if complete and successful, includes the index of the
/// buffer that parsing stopped at, and the size of the following chunk.
///
/// # Example
///
/// ```
/// let buf = b"4\r\nRust\r\n0\r\n\r\n";
/// assert_eq!(httparse::parse_chunk_size(buf),
///            Ok(httparse::Status::Complete((3, 4))));
/// ```
pub fn parse_chunk_size(buf: &[u8])
    -> result::Result<Status<(usize, u64)>, InvalidChunkSize> {
    const RADIX: u64 = 16;
    let mut bytes = Bytes::new(buf);
    let mut size = 0;
    let mut in_chunk_size = true;
    let mut in_ext = false;
    let mut count = 0;
    loop {
        let b = next!(bytes);
        match b {
            b'0' ... b'9' if in_chunk_size => {
                if count > 15 {
                    return Err(InvalidChunkSize);
                }
                count += 1;
                size *= RADIX;
                size += (b - b'0') as u64;
            },
            b'a' ... b'f' if in_chunk_size => {
                if count > 15 {
                    return Err(InvalidChunkSize);
                }
                count += 1;
                size *= RADIX;
                size += (b + 10 - b'a') as u64;
            }
            b'A' ... b'F' if in_chunk_size => {
                if count > 15 {
                    return Err(InvalidChunkSize);
                }
                count += 1;
                size *= RADIX;
                size += (b + 10 - b'A') as u64;
            }
            b'\r' => {
                match next!(bytes) {
                    b'\n' => break,
                    _ => return Err(InvalidChunkSize),
                }
            }
            // If we weren't in the extension yet, the ";" signals its start
            b';' if !in_ext => {
                in_ext = true;
                in_chunk_size = false;
            }
            // "Linear white space" is ignored between the chunk size and the
            // extension separator token (";") due to the "implied *LWS rule".
            b'\t' | b' ' if !in_ext & !in_chunk_size => {}
            // LWS can follow the chunk size, but no more digits can come
            b'\t' | b' ' if in_chunk_size => in_chunk_size = false,
            // We allow any arbitrary octet once we are in the extension, since
            // they all get ignored anyway. According to the HTTP spec, valid
            // extensions would have a more strict syntax:
            //     (token ["=" (token | quoted-string)])
            // but we gain nothing by rejecting an otherwise valid chunk size.
            _ if in_ext => {}
            // Finally, if we aren't in the extension and we're reading any
            // other octet, the chunk size line is invalid!
            _ => return Err(InvalidChunkSize),
        }
    }
    Ok(Status::Complete((bytes.pos(), size)))
}

#[cfg(test)]
mod tests {
    use super::{Request, Response, Status, EMPTY_HEADER, shrink, parse_chunk_size};

    const NUM_OF_HEADERS: usize = 4;

    #[test]
    fn test_shrink() {
        let mut arr = [EMPTY_HEADER; 16];
        {
            let slice = &mut &mut arr[..];
            assert_eq!(slice.len(), 16);
            shrink(slice, 4);
            assert_eq!(slice.len(), 4);
        }
        assert_eq!(arr.len(), 16);
    }

    macro_rules! req {
        ($name:ident, $buf:expr, |$arg:ident| $body:expr) => (
            req! {$name, $buf, Ok(Status::Complete($buf.len())), |$arg| $body }
        );
        ($name:ident, $buf:expr, $len:expr, |$arg:ident| $body:expr) => (
        #[test]
        fn $name() {
            let mut headers = [EMPTY_HEADER; NUM_OF_HEADERS];
            let mut req = Request::new(&mut headers[..]);
            let status = req.parse($buf.as_ref());
            assert_eq!(status, $len);
            closure(req);

            fn closure($arg: Request) {
                $body
            }
        }
        )
    }

    req! {
        test_request_simple,
        b"GET / HTTP/1.1\r\n\r\n",
        |req| {
            assert_eq!(req.method.unwrap(), "GET");
            assert_eq!(req.path.unwrap(), "/");
            assert_eq!(req.version.unwrap(), 1);
            assert_eq!(req.headers.len(), 0);
        }
    }

    req! {
        test_request_headers,
        b"GET / HTTP/1.1\r\nHost: foo.com\r\nCookie: \r\n\r\n",
        |req| {
            assert_eq!(req.method.unwrap(), "GET");
            assert_eq!(req.path.unwrap(), "/");
            assert_eq!(req.version.unwrap(), 1);
            assert_eq!(req.headers.len(), 2);
            assert_eq!(req.headers[0].name, "Host");
            assert_eq!(req.headers[0].value, b"foo.com");
            assert_eq!(req.headers[1].name, "Cookie");
            assert_eq!(req.headers[1].value, b"");
        }
    }

    req! {
        // test the scalar parsing
        test_request_header_value_htab_short,
        b"GET / HTTP/1.1\r\nUser-Agent: some\tagent\r\n\r\n",
        |req| {
            assert_eq!(req.method.unwrap(), "GET");
            assert_eq!(req.path.unwrap(), "/");
            assert_eq!(req.version.unwrap(), 1);
            assert_eq!(req.headers.len(), 1);
            assert_eq!(req.headers[0].name, "User-Agent");
            assert_eq!(req.headers[0].value, b"some\tagent");
        }
    }

    req! {
        // test the sse42 parsing
        test_request_header_value_htab_med,
        b"GET / HTTP/1.1\r\nUser-Agent: 1234567890some\tagent\r\n\r\n",
        |req| {
            assert_eq!(req.method.unwrap(), "GET");
            assert_eq!(req.path.unwrap(), "/");
            assert_eq!(req.version.unwrap(), 1);
            assert_eq!(req.headers.len(), 1);
            assert_eq!(req.headers[0].name, "User-Agent");
            assert_eq!(req.headers[0].value, b"1234567890some\tagent");
        }
    }

    req! {
        // test the avx2 parsing
        test_request_header_value_htab_long,
        b"GET / HTTP/1.1\r\nUser-Agent: 1234567890some\t1234567890agent1234567890\r\n\r\n",
        |req| {
            assert_eq!(req.method.unwrap(), "GET");
            assert_eq!(req.path.unwrap(), "/");
            assert_eq!(req.version.unwrap(), 1);
            assert_eq!(req.headers.len(), 1);
            assert_eq!(req.headers[0].name, "User-Agent");
            assert_eq!(req.headers[0].value, &b"1234567890some\t1234567890agent1234567890"[..]);
        }
    }

    req! {
        test_request_headers_max,
        b"GET / HTTP/1.1\r\nA: A\r\nB: B\r\nC: C\r\nD: D\r\n\r\n",
        |req| {
            assert_eq!(req.headers.len(), NUM_OF_HEADERS);
        }
    }

    req! {
        test_request_multibyte,
        b"GET / HTTP/1.1\r\nHost: foo.com\r\nUser-Agent: \xe3\x81\xb2\xe3/1.0\r\n\r\n",
        |req| {
            assert_eq!(req.method.unwrap(), "GET");
            assert_eq!(req.path.unwrap(), "/");
            assert_eq!(req.version.unwrap(), 1);
            assert_eq!(req.headers[0].name, "Host");
            assert_eq!(req.headers[0].value, b"foo.com");
            assert_eq!(req.headers[1].name, "User-Agent");
            assert_eq!(req.headers[1].value, b"\xe3\x81\xb2\xe3/1.0");
        }
    }


    req! {
        test_request_partial,
        b"GET / HTTP/1.1\r\n\r", Ok(Status::Partial),
        |_req| {}
    }

    req! {
        test_request_partial_version,
        b"GET / HTTP/1.", Ok(Status::Partial),
        |_req| {}
    }

    req! {
        test_request_newlines,
        b"GET / HTTP/1.1\nHost: foo.bar\n\n",
        |_r| {}
    }

    req! {
        test_request_empty_lines_prefix,
        b"\r\n\r\nGET / HTTP/1.1\r\n\r\n",
        |req| {
            assert_eq!(req.method.unwrap(), "GET");
            assert_eq!(req.path.unwrap(), "/");
            assert_eq!(req.version.unwrap(), 1);
            assert_eq!(req.headers.len(), 0);
        }
    }

    req! {
        test_request_empty_lines_prefix_lf_only,
        b"\n\nGET / HTTP/1.1\n\n",
        |req| {
            assert_eq!(req.method.unwrap(), "GET");
            assert_eq!(req.path.unwrap(), "/");
            assert_eq!(req.version.unwrap(), 1);
            assert_eq!(req.headers.len(), 0);
        }
    }

    req! {
        test_request_with_invalid_token_delimiter,
        b"GET\n/ HTTP/1.1\r\nHost: foo.bar\r\n\r\n",
        Err(::Error::Token),
        |_r| {}
    }


    req! {
        test_request_with_invalid_but_short_version,
        b"GET / HTTP/1!",
        Err(::Error::Version),
        |_r| {}
    }

    macro_rules! res {
        ($name:ident, $buf:expr, |$arg:ident| $body:expr) => (
            res! {$name, $buf, Ok(Status::Complete($buf.len())), |$arg| $body }
        );
        ($name:ident, $buf:expr, $len:expr, |$arg:ident| $body:expr) => (
        #[test]
        fn $name() {
            let mut headers = [EMPTY_HEADER; NUM_OF_HEADERS];
            let mut res = Response::new(&mut headers[..]);
            let status = res.parse($buf.as_ref());
            assert_eq!(status, $len);
            closure(res);

            fn closure($arg: Response) {
                $body
            }
        }
        )
    }

    res! {
        test_response_simple,
        b"HTTP/1.1 200 OK\r\n\r\n",
        |res| {
            assert_eq!(res.version.unwrap(), 1);
            assert_eq!(res.code.unwrap(), 200);
            assert_eq!(res.reason.unwrap(), "OK");
        }
    }

    res! {
        test_response_newlines,
        b"HTTP/1.0 403 Forbidden\nServer: foo.bar\n\n",
        |_r| {}
    }

    res! {
        test_response_reason_missing,
        b"HTTP/1.1 200 \r\n\r\n",
        |res| {
            assert_eq!(res.version.unwrap(), 1);
            assert_eq!(res.code.unwrap(), 200);
            assert_eq!(res.reason.unwrap(), "");
        }
    }

    res! {
        test_response_reason_missing_no_space,
        b"HTTP/1.1 200\r\n\r\n",
        |res| {
            assert_eq!(res.version.unwrap(), 1);
            assert_eq!(res.code.unwrap(), 200);
            assert_eq!(res.reason.unwrap(), "");
        }
    }

    res! {
        test_response_reason_missing_no_space_with_headers,
        b"HTTP/1.1 200\r\nFoo: bar\r\n\r\n",
        |res| {
            assert_eq!(res.version.unwrap(), 1);
            assert_eq!(res.code.unwrap(), 200);
            assert_eq!(res.reason.unwrap(), "");
            assert_eq!(res.headers.len(), 1);
            assert_eq!(res.headers[0].name, "Foo");
            assert_eq!(res.headers[0].value, b"bar");
        }
    }

    res! {
        test_response_reason_with_space_and_tab,
        b"HTTP/1.1 101 Switching Protocols\t\r\n\r\n",
        |res| {
            assert_eq!(res.version.unwrap(), 1);
            assert_eq!(res.code.unwrap(), 101);
            assert_eq!(res.reason.unwrap(), "Switching Protocols\t");
        }
    }

    static RESPONSE_REASON_WITH_OBS_TEXT_BYTE: &'static [u8] = b"HTTP/1.1 200 X\xFFZ\r\n\r\n";
    res! {
        test_response_reason_with_obsolete_text_byte,
        RESPONSE_REASON_WITH_OBS_TEXT_BYTE,
        Err(::Error::Status),
        |_res| {}
    }

    res! {
        test_response_reason_with_nul_byte,
        b"HTTP/1.1 200 \x00\r\n\r\n",
        Err(::Error::Status),
        |_res| {}
    }

    res! {
        test_response_version_missing_space,
        b"HTTP/1.1",
        Ok(Status::Partial),
        |_res| {}
    }

    res! {
        test_response_code_missing_space,
        b"HTTP/1.1 200",
        Ok(Status::Partial),
        |_res| {}
    }

    res! {
        test_response_empty_lines_prefix_lf_only,
        b"\n\nHTTP/1.1 200 OK\n\n",
        |_res| {}
    }

    #[test]
    fn test_chunk_size() {
        assert_eq!(parse_chunk_size(b"0\r\n"), Ok(Status::Complete((3, 0))));
        assert_eq!(parse_chunk_size(b"12\r\nchunk"), Ok(Status::Complete((4, 18))));
        assert_eq!(parse_chunk_size(b"3086d\r\n"), Ok(Status::Complete((7, 198765))));
        assert_eq!(parse_chunk_size(b"3735AB1;foo bar*\r\n"), Ok(Status::Complete((18, 57891505))));
        assert_eq!(parse_chunk_size(b"3735ab1 ; baz \r\n"), Ok(Status::Complete((16, 57891505))));
        assert_eq!(parse_chunk_size(b"77a65\r"), Ok(Status::Partial));
        assert_eq!(parse_chunk_size(b"ab"), Ok(Status::Partial));
        assert_eq!(parse_chunk_size(b"567f8a\rfoo"), Err(::InvalidChunkSize));
        assert_eq!(parse_chunk_size(b"567f8a\rfoo"), Err(::InvalidChunkSize));
        assert_eq!(parse_chunk_size(b"567xf8a\r\n"), Err(::InvalidChunkSize));
        assert_eq!(parse_chunk_size(b"ffffffffffffffff\r\n"), Ok(Status::Complete((18, ::core::u64::MAX))));
        assert_eq!(parse_chunk_size(b"1ffffffffffffffff\r\n"), Err(::InvalidChunkSize));
        assert_eq!(parse_chunk_size(b"Affffffffffffffff\r\n"), Err(::InvalidChunkSize));
        assert_eq!(parse_chunk_size(b"fffffffffffffffff\r\n"), Err(::InvalidChunkSize));
    }

    #[cfg(feature = "std")]
    #[test]
    fn test_std_error() {
        use super::Error;
        use std::error::Error as StdError;
        let err = Error::HeaderName;
        assert_eq!(err.to_string(), err.description());
    }
}