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
#![feature(option_result_unwrap_unchecked)] //! This crate provides functions to read utf-8 text from any type implementing [`io::BufRead`] //! through a trait, [`BufRead`], without waiting for newline delimiters. These functions take //! advantage of buffering and either return `&`[`str`] or [`char`]s. Each has an associated //! iterator, some have an equivalent to a [`Map`] iterator that avoids allocation and cloning as //! well. //! //! # Quick Start //! //! The simplest way to read a file using this crate may be something along the following: //! //! ``` //! use utf8_bufread::BufRead; //! use std::io::{Cursor, ErrorKind}; //! use std::borrow::Cow; //! //! // Reader may be any type implementing io::BufRead //! // We'll just use a cursor wrapping a slice for this example //! let mut reader = Cursor::new("Löwe 老虎 Léopard"); //! loop { // Loop until EOF //! match reader.read_str() { //! Ok(s) => { //! if s.is_empty() { //! break; // EOF //! } //! // Do something with `s` ... //! print!("{}", s); //! } //! Err(e) => { //! // We should try again if we get interrupted //! if e.kind() != ErrorKind::Interrupted { //! break; //! } //! } //! } //! } //! ``` //! //! # Reading arbitrary-length string slices //! //! The [`read_str`] function returns a `&`[`str`] of arbitrary length (up to the reader's buffer //! capacity) read from the inner reader, without cloning data, unless a valid codepoint ends up //! cut at the end of the reader's buffer. Its associated iterator can be obtained by calling //! [`str_iter`], and since it involves cloning the data at each iteration, [`str_map`] is also //! provided. //! //! # Reading codepoints //! //! The [`read_char`] function returns a [`char`] read from the inner reader. Its associated //! iterator can be obtained by calling [`char_iter`]. //! //! # Iterator types //! //! This crate provides several structs for several ways of iterating over the inner reader's data: //! - [`StrIter`] and [`CodepointIter`] clone the data on each iteration, but use an [`Rc`] to //! check if the returned [`String`] buffer is still used. If not, it is re-used to avoid //! re-allocating. //! ``` //! use utf8_bufread::BufRead; //! use std::io::Cursor; //! //! let mut reader = Cursor::new("Löwe 老虎 Léopard"); //! for s in reader.str_iter().filter_map(|r| r.ok()) { //! // Do something with s ... //! print!("{}", s); //! } //! ``` //! - [`StrMap`] and [`CodepointMap`] allow having access to read data without cloning, but then it //! cannot be passed to further iterator adapters. //! ``` //! use utf8_bufread::BufRead; //! use std::io::Cursor; //! //! let s = "Löwe 老虎 Léopard"; //! let mut reader = Cursor::new(s); //! let count: usize = reader.str_map(|s| s.len()).filter_map(Result::ok).sum(); //! println!("There is {} valid utf-8 bytes in {}", count, s); //! ``` //! - [`CharIter`] is similar to [`StrIter`] and others, except it relies on [`char`]s implementing //! [`Copy`] and thus doesn't need a buffer nor the "`Rc` trick". //! ``` //! use utf8_bufread::BufRead; //! use std::io::Cursor; //! //! let s = "Löwe 老虎 Léopard"; //! let mut reader = Cursor::new(s); //! let count = reader.char_iter().filter_map(Result::ok).filter(|c| c.is_lowercase()).count(); //! assert_eq!(count, 9); //! ``` //! //! All these iterators may read data until EOF or an invalid codepoint is found. If valid //! codepoints are read from the inner reader, they *will* be returned before reporting an error. //! After encountering an error or EOF, they always return `None`. They always ignore any //! [`Interrupted`] error. //! //! [`read_str`]: self::BufRead::read_str //! [`str_iter`]: self::BufRead::str_iter //! [`str_map`]: self::BufRead::str_map //! [`read_char`]: self::BufRead::read_char //! [`char_iter`]: self::BufRead::char_iter //! [`Map`]: std::iter::Map //! [`Interrupted`]: std::io::ErrorKind::Interrupted #[deny(missing_crate_level_docs, missing_docs, missing_doc_code_examples)] mod error; use error::Result; use std::borrow::Cow; use std::io::{self, ErrorKind}; use std::rc::Rc; use std::slice::from_raw_parts; use std::str::{from_utf8, from_utf8_unchecked, FromStr}; pub use error::Error; /// A trait implemented for all types implementing [`io::BufRead`], providing functions to /// read utf-8 text streams without waiting for newline delimiters. /// /// [`io::BufRead`]: std::io::BufRead /// /// # Examples /// /// ``` /// use std::io::Cursor; /// use utf8_bufread::BufRead; /// /// // Prints "I luv you too !" /// if Cursor::new("💖").read_str().map_or(false, |s| s == "💖") { /// println!("I luv you too !"); /// } /// ``` pub trait BufRead: io::BufRead { /// Reads some bytes from the inner reader and returns a [`Cow`]`<&`[`str`]`>` of it referring /// to all valid codepoints read, wrapped in an [`io::Result`]. /// /// This function will read all bytes from the underlying stream until its buffer is full, an /// invalid or incomplete codepoint is found, or EOF is found. Once found, all codepoints up /// to, including the EOF (if found), but not including the invalid or incomplete codepoint /// (if found), will be returned. This function may read an arbitrary number of byte, between 1 /// and this reader's buffer capacity (unless the buffer is not big enough to fit a unicode /// codepoint). /// /// The returned reference points to this reader's actual buffer, meaning it borrows the /// reader. /// /// A [`Cow`] is used to gracefully handle cases where a valid codepoint is cut by the end of /// the buffer of this reader, and more bytes may need to be read from the inner reader to form /// a hopefully valid codepoint. This function only allocates a new [`String`] and clones data /// in that scenario. In worst case it happens once every two calls, allocating and cloning /// 4 bytes every `c` bytes read, where `c` is this reader's buffer capacity. /// /// If this function returns [`Ok`]`("")`, the stream has reached EOF. /// /// # Errors /// /// If the internal call to [`fill_buf`] returns an [`io::Error`] or this function returns /// immediately an [`Error`] wrapping the original error. /// /// If the first codepoint read from the inner reader is invalid, an [`Error`] wrapping the /// original [`Utf8Error`] or [`FromUtf8Error`] is returned. /// /// If the codepoint is complete but invalid, the returned error will have a [`kind`] of /// [`ErrorKind`]`::`[`InvalidData`]. If EOF was encountered before the end of a codepoint, /// the error will have a [`kind`] of [`ErrorKind`]`::`[`UnexpectedEof`]. /// /// The returned [`Error`] may contain a non-zero amount of "leftover" bytes (see /// [`Error::leftovers`] for more info). When it is the case, it is guaranteed that the /// following read operation will not return any of those bytes, nor "skip" bytes from this /// reader. /// /// Note that if the buffer of this reader is less than 4 bytes long it may fail to read /// complete codepoints and "spuriously" return the same error as when it unexpectedly /// encounters EOF, since we're unable to load enough bytes to form a valid codepoint. We /// cannot check the capacity of the buffer using the [`io::BufRead`] API only, it is then up /// to the user to ensure this won't happen. *It should not happen unless you explicitly set /// the capacity yourself*. /// /// # Examples /// /// This example simply reads from a stream and prints it to standard output. /// /// ``` /// use std::io::{Cursor, Error, ErrorKind}; /// use utf8_bufread::BufRead; /// use std::borrow::Cow; /// /// // We could use any type implementing io::BufRead, we'll just use a cursor here /// let mut reader = Cursor::new("Löwe 老虎 Léopard"); /// /// loop { /// match reader.read_str() { /// Ok(s) => { /// if s.is_empty() { /// break; // EOF /// } /// print!("{}", s) /// } /// Err(e) => { /// if ErrorKind::Interrupted != e.kind() { /// // Ignore interrupted errors /// eprintln!("{}", e); /// } /// } /// } /// } /// ``` /// /// [`kind`]: self::Error::kind /// [`fill_buf`]: std::io::BufRead::fill_buf /// [`Interrupted`]: std::io::ErrorKind::Interrupted /// [`InvalidData`]: std::io::ErrorKind::InvalidData /// [`UnexpectedEof`]: std::io::ErrorKind::UnexpectedEof /// [`Utf8Error`]: std::str::Utf8Error /// [`FromUtf8Error`]: std::string::FromUtf8Error fn read_str(&mut self) -> Result<Cow<str>> { // Fill the buffer from inner reader's data and get its content let read_bytes = self.fill_buf()?; let read_len = read_bytes.len(); if read_len == 0 { return Ok(Cow::from("")); } let ptr = read_bytes.as_ptr(); // We attempt converting read bytes to utf8 match from_utf8(read_bytes) { Ok(_) => { self.consume(read_len); // The call to `from_raw_parts` is safe, as: // a. It is within the memory region of the reader's now filled buffer. // b. Implicit lifetimes imply the reader is mutably borrowed for the lifetime of the // returned str reference // TODO: ask for review of point b. above // The call to `from_utf8_unchecked` is safe as we just ran the validation on the same // memory region above Ok(Cow::from(unsafe { from_utf8_unchecked(from_raw_parts(ptr, read_len)) })) } Err(e) => { // If we have an error, we will first attempt to return all valid read bytes, // putting the invalid or incomplete codepoint at the beginning of the buffer. // This allows us to recover from reading up to a byte that isn't on a char // boundary by reading the complete codepoint on the next call let len = e.valid_up_to(); if len != 0 { self.consume(len); // This is safe, see `Utf8Error::valid_up_to(&self)` doc Ok(Cow::from(unsafe { from_utf8_unchecked(from_raw_parts(ptr, len)) })) } else if read_len >= codepoint_length(read_bytes[0]) { // If we cannot decode any valid utf8 byte from the buffer, it either means // - We reached EOF with an incomplete codepoint, we should return an // UnexpectedEof Error // - There was a parse error earlier, and we read everything up to this // point in a previous read call, there is two possible situations again: // - There is more than 2 bytes following the first byte of the invalid // slice, this means there truly is an invalid codepoint, we should // return an Utf8Error // - There is less than 4 bytes left in the buffer, meaning we may have // an incomplete codepoint and need to read up to 3 bytes further. // We know read_bytes is not empty // We couldn't get a valid codepoint despite reading enough bytes Err(Error::from(e)) } else { // Not enough bytes read, we will try to read more bytes // Consume the last bytes, so that the next call to `fill_buff` will read // more bytes from the underlying stream self.consume(read_len); read_across_boundary(self, Vec::from(unsafe { from_raw_parts(ptr, read_len) })) } } } } /// Reads 1 to 4 bytes from the inner reader and returns a [`Cow`]`<&`[`str`]`>` of it /// referring to the valid codepoints read, wrapped in an [`io::Result`]. /// /// This function will read bytes from the underlying stream until one codepoint is read, an /// invalid or incomplete codepoint is found, or EOF is found. /// /// The returned reference points to this reader's actual buffer, meaning it borrows the /// reader. /// /// A [`Cow`] is used to gracefully handle cases where a valid codepoint is cut by the end of /// the buffer of this reader, and more bytes may need to be read from the inner reader to form /// a hopefully valid codepoint. This function only allocates a new [`String`] and clones data /// in that scenario. In worst case it allocates and clones 4 bytes every `c` bytes read, /// where `c` is this reader's buffer capacity. /// /// If this function returns [`Ok`]`("")`, the stream has reached EOF. /// /// # Errors /// /// If the internal call to [`fill_buf`] returns an [`io::Error`] or this function returns /// immediately an [`Error`] wrapping the original error. /// /// If the first codepoint read from the inner reader is invalid or incomplete, an [`Error`] /// wrapping the original [`Utf8Error`] or [`FromUtf8Error`] is returned. /// /// If the codepoint is complete but invalid, the returned error will have a [`kind`] of /// [`ErrorKind`]`::`[`InvalidData`]. If EOF was encountered before the end of a codepoint, /// the error will have a [`kind`] of [`ErrorKind`]`::`[`UnexpectedEof`]. /// /// The returned [`Error`] may contain a non-zero amount of "leftover" bytes (see /// [`Error::leftovers`] for more info). When it is the case, it is guaranteed that the /// following read operation will not return any of those bytes, nor "skip" bytes from this /// reader. /// /// Note that if the buffer of this reader is less than 4 bytes long it may fail to read /// complete codepoints and "spuriously" return the same error as when it unexpectedly /// encounters EOF, since we're unable to load enough bytes to form a valid codepoint. We /// cannot check the capacity of the buffer using the [`io::BufRead`] API only, it is then up /// to the user to ensure this won't happen. *It should not happen unless you explicitly set /// the capacity yourself*. /// /// # Examples /// /// This example simply reads from a stream and counts the number of `🏳` character. /// /// ``` /// use std::io::{Cursor, Error, ErrorKind}; /// use utf8_bufread::BufRead; /// use std::borrow::Cow; /// /// // We could use any type implementing io::BufRead, we'll just use a cursor here /// let mut reader = Cursor::new("Löwe 老虎 🏳Léopard"); /// let mut count = 0; /// /// loop { /// match reader.read_codepoint() { /// Ok(s) => { /// if s.is_empty() { /// break; // EOF /// } /// if s == "🏳" { /// count += 1; /// } /// } /// Err(e) => { /// if ErrorKind::Interrupted != e.kind() { /// // Ignore interrupted errors /// eprintln!("{}", e); /// } /// } /// } /// } /// assert_eq!(count, 1); /// ``` #[doc(hidden)] fn read_codepoint(&mut self) -> Result<Cow<str>> { // Fill the buffer from inner reader's data and get its content let read_bytes = self.fill_buf()?; let read_len = read_bytes.len(); if read_len == 0 { return Ok(Cow::from("")); } let ptr = read_bytes.as_ptr(); let len = codepoint_length(read_bytes[0]); if read_len < len { // Not enough bytes read, we will try to read more bytes // Consume the last bytes, so that the next call to `fill_buff` will read // more bytes from the underlying stream self.consume(read_len); read_across_boundary(self, Vec::from(unsafe { from_raw_parts(ptr, read_len) })) } else { match from_utf8(&read_bytes[..len]) { Ok(_) => { self.consume(len); // The call to `from_raw_parts` is safe, as: // a. It is within the memory region of the reader's now filled buffer. // b. Implicit lifetimes imply the reader is mutably borrowed for the lifetime of the // returned str reference // TODO: ask for review of point b. above // The call to `from_utf8_unchecked` is safe as we just ran the validation on the same // memory region above Ok(Cow::from(unsafe { from_utf8_unchecked(from_raw_parts(ptr, len)) })) } Err(e) => Err(Error::from(e)), } } } /// Reads 1 to 4 bytes from the inner reader and returns the [`char`] read, wrapped in an /// [`io::Result`]. /// /// This function will read bytes from the underlying stream until one codepoint is read, an /// invalid or incomplete codepoint is found, or EOF is found. /// /// If this function returns [`Ok`]`('\0')`, the stream has reached EOF. /// /// # Errors /// /// If the internal call to [`fill_buf`] returns an [`io::Error`] or this function returns /// immediately an [`Error`] wrapping the original error. /// /// If the first codepoint read from the inner reader is invalid or incomplete, an [`Error`] /// wrapping the original [`Utf8Error`] or [`FromUtf8Error`] is returned. /// /// If the codepoint is complete but invalid, the returned error will have a [`kind`] of /// [`ErrorKind`]`::`[`InvalidData`]. If EOF was encountered before the end of a codepoint, /// the error will have a [`kind`] of [`ErrorKind`]`::`[`UnexpectedEof`]. /// /// The returned [`Error`] may contain a non-zero amount of "leftover" bytes (see /// [`Error::leftovers`] for more info). When it is the case, it is guaranteed that the /// following read operation will not return any of those bytes, nor "skip" bytes from this /// reader. /// /// Note that if the buffer of this reader is less than 4 bytes long it may fail to read /// complete codepoints and "spuriously" return the same error as when it unexpectedly /// encounters EOF, since we're unable to load enough bytes to form a valid codepoint. We /// cannot check the capacity of the buffer using the [`io::BufRead`] API only, it is then up /// to the user to ensure this won't happen. *It should not happen unless you explicitly set /// the capacity yourself*. /// /// # Examples /// /// This example simply reads from a stream and counts the number of lowercase characters /// /// ``` /// use std::io::{Cursor, Error, ErrorKind}; /// use utf8_bufread::BufRead; /// use std::borrow::Cow; /// /// // We could use any type implementing io::BufRead, we'll just use a cursor here /// let mut reader = Cursor::new("Löwe 老虎 Léopard"); /// let mut count = 0; /// /// loop { /// match reader.read_char() { /// Ok('\0') => break, // EOF /// Ok(c) => { /// if c.is_lowercase() { /// count += 1; /// } /// } /// Err(e) => { /// if ErrorKind::Interrupted != e.kind() { /// // Ignore interrupted errors /// eprintln!("{}", e); /// } /// } /// } /// } /// assert_eq!(count, 9); /// ``` /// /// [`fill_buf`]: std::io::BufRead::fill_buf /// [`Interrupted`]: std::io::ErrorKind::Interrupted /// [`InvalidData`]: std::io::ErrorKind::InvalidData /// [`UnexpectedEof`]: std::io::ErrorKind::UnexpectedEof /// [`Utf8Error`]: std::str::Utf8Error /// [`FromUtf8Error`]: std::string::FromUtf8Error /// [`kind`]: crate::error::Error::kind fn read_char(&mut self) -> Result<char> { // We guarantee that self.read_codepoint returns: // - An empty string or // - Exactly one valid codepoint let c = self.read_codepoint()?; if c.is_empty() { return Ok('\0'); } Ok(unsafe { char::from_str(c.as_ref()).unwrap_unchecked() }) } /// Returns an iterator over string slices of this reader. /// /// It is equivalent to calling [`read_str`] in a loop, ignoring /// [`ErrorKind`]`::`[`Interrupted`] errors, until EOF or the first error encountered. /// /// The iterator returned by this function will yield instances of /// [`io::Result`]`<`[`Rc`]`<`[`String`]`>>`. We use the [`Rc`] to check while iterating if the /// iterator is the only one holding a reference to it, avoiding allocating a new buffer if /// that's the case. /// /// The iterator returned will yield at most one [`io::Error`]. Once an error is yielded, it /// will only yield [`None`]. /// /// # Examples /// /// This example simply reads from a string and prints it to standard output: /// /// ``` /// use std::io::Cursor; /// use utf8_bufread::BufRead; /// /// // We could use any type implementing io::BufRead, we'll just use a cursor here /// let mut reader = Cursor::new("Löwe 老虎 Léopard"); /// // We ignore any error, we know once we encounter one we can't read any further anyway /// reader.str_iter().filter_map(Result::ok).for_each(|s| print!("{}", s)); /// ``` /// /// [`read_str`]: self::BufRead::read_str /// [`Interrupted`]: std::io::ErrorKind::Interrupted fn str_iter(&mut self) -> StrIter<'_, Self> { let default_cap = 8 * 1024; StrIter { reader: self, buf: Rc::new(String::with_capacity(default_cap)), default_cap, ended: false, } } /// Returns an iterator over codepoints of this reader. /// /// It is equivalent to calling [`read_codepoint`] in a loop, ignoring /// [`ErrorKind`]`::`[`Interrupted`] errors, until EOF or the first error encountered. /// /// The iterator returned by this function will yield instances of /// [`io::Result`]`<`[`Rc`]`<`[`String`]`>>`. We use the [`Rc`] to check while iterating if the /// iterator is the only one holding a reference to it, avoiding allocating a new buffer if /// that's the case. /// /// The iterator returned will yield at most one [`io::Error`]. Once an error is yielded, it /// will only yield [`None`]. /// /// # Examples /// /// This example simply reads from a stream and counts the number of `🏳` character. /// /// ``` /// use std::io::Cursor; /// use utf8_bufread::BufRead; /// /// // We could use any type implementing io::BufRead, we'll just use a cursor here /// let mut reader = Cursor::new("Löwe 老虎 🏳Léopard"); /// let count = reader.codepoint_iter() /// .filter_map(Result::ok) /// .filter(|s| s.as_ref() == "🏳") /// .count(); /// assert_eq!(count, 1); /// ``` #[doc(hidden)] fn codepoint_iter(&mut self) -> CodepointIter<'_, Self> { let default_cap = 4; CodepointIter { reader: self, buf: Rc::new(String::with_capacity(default_cap)), default_cap, ended: false, } } /// Returns an iterator over chars of this reader. /// /// It is equivalent to calling [`read_char`] in a loop, ignoring /// [`ErrorKind`]`::`[`Interrupted`] errors, until EOF or the first error encountered. /// /// The iterator returned by this function will yield instances of /// [`io::Result`]`<`[`char`]`>`. /// /// The iterator returned will yield at most one [`io::Error`]. Once an error is yielded, it /// will only yield [`None`]. /// /// # Examples /// /// This example simply reads from a stream, filtering out any whitespace: /// /// ``` /// use std::io::Cursor; /// use utf8_bufread::BufRead; /// /// // We could use any type implementing io::BufRead, we'll just use a cursor here /// let mut reader = Cursor::new("Löwe 老虎 Léopard"); /// let result: String = reader.char_iter() /// .filter_map(Result::ok) /// .filter(|c| !c.is_whitespace()) /// .collect(); /// assert_eq!(result.as_str(), "Löwe老虎Léopard"); /// ``` /// /// [`read_char`]: self::BufRead::read_char /// [`Interrupted`]: std::io::ErrorKind::Interrupted fn char_iter(&mut self) -> CharIter<'_, Self> { CharIter { reader: self, ended: false, } } /// Returns an mapping iterator over string slices of this reader. /// /// It is equivalent to calling [`read_str`] in a loop, ignoring /// [`ErrorKind`]`::`[`Interrupted`] errors, until EOF or the first error encountered. /// /// The iterator returned by this function will call `f` with instances of [`Cow`]`<`[`str`]`>` /// as returned by [`read_str`], and yield instances of [`io::Result`]`<T>`. This may help /// avoids the allocations and clonings [`str_iter`] does. /// /// The iterator returned will yield at most one [`io::Error`], and if one is yielded it will /// always be the last item. /// /// # Examples /// /// This example simply reads from a stream and counts the number of bytes read: /// /// ``` /// use std::io::Cursor; /// use utf8_bufread::BufRead; /// /// // We could use any type implementing io::BufRead, we'll just use a cursor here /// let mut reader = Cursor::new("Löwe 老虎 Léopard"); /// let count: usize = reader.str_map(|s| s.len()).filter_map(Result::ok).sum(); /// assert_eq!(count, 21); /// ``` /// /// [`read_str`]: self::BufRead::read_str /// [`str_iter`]: self::BufRead::str_iter /// [`Interrupted`]: std::io::ErrorKind::Interrupted fn str_map<F, T>(&mut self, f: F) -> StrMap<'_, Self, F> where F: FnMut(Cow<str>) -> T, { StrMap { reader: self, map: Rc::new(f), ended: false, } } /// Returns an mapping iterator over codepoints of this reader. /// /// It is equivalent to calling [`read_codepoint`] in a loop, ignoring /// [`ErrorKind`]`::`[`Interrupted`] errors, until EOF or the first error encountered. /// /// The iterator returned by this function will call `f` with instances of [`Cow`]`<`[`str`]`>` /// as returned by [`read_str`], and yield instances of [`io::Result`]`<T>`. This may help /// avoids the allocations and clonings [`str_iter`] does. /// /// The iterator returned will yield at most one [`io::Error`], and if one is yielded it will /// always be the last item. /// /// # Examples /// /// This example simply reads maps each codepoints to their length in bytes: /// /// ``` /// use std::io::Cursor; /// use utf8_bufread::BufRead; /// /// // We could use any type implementing io::BufRead, we'll just use a cursor here /// let mut reader = Cursor::new("Löwe 老虎 Léopard"); /// let lengths: Vec<_> = reader.codepoint_map(|s| s.len()).filter_map(Result::ok).collect(); /// assert_eq!(lengths.as_ref(), [1, 2, 1, 1, 1, 3, 3, 1, 1, 2, 1, 1, 1, 1, 1]); /// ``` #[doc(hidden)] fn codepoint_map<F, T>(&mut self, f: F) -> CodepointMap<'_, Self, F> where F: FnMut(Cow<str>) -> T, { CodepointMap { reader: self, map: Rc::new(f), ended: false, } } } impl<R: io::BufRead> BufRead for R {} /// An iterator over string slices of an instance of [`io::BufRead`], created by [`str_iter`], see /// its documentation for more details. /// /// [`str_iter`]: self::BufRead::str_iter pub struct StrIter<'r, R> where R: ?Sized, { reader: &'r mut R, buf: Rc<String>, default_cap: usize, ended: bool, } impl<R> Iterator for StrIter<'_, R> where R: io::BufRead, { type Item = Result<Rc<String>>; //noinspection DuplicatedCode fn next(&mut self) -> Option<Self::Item> { if self.ended { return None; } let buf = match Rc::get_mut(&mut self.buf) { None => { self.buf = Rc::new(String::with_capacity(self.default_cap)); Rc::make_mut(&mut self.buf) } Some(buf) => { buf.clear(); buf } }; loop { match self.reader.read_str() { Err(e) => { if let ErrorKind::Interrupted = e.kind() { continue; } self.ended = true; break Some(Err(e)); } Ok(s) => { if s.is_empty() { self.ended = true; break None; } else { buf.push_str(s.as_ref()); break Some(Ok(Rc::clone(&self.buf))); } } } } } } /// An iterator over string slices of an instance of [`io::BufRead`], created by /// [`codepoints_iter`], see its documentation for more details. /// /// [`codepoints_iter`]: self::BufRead::codepoints_iter #[doc(hidden)] pub struct CodepointIter<'r, R> where R: ?Sized, { reader: &'r mut R, buf: Rc<String>, default_cap: usize, ended: bool, } impl<R> Iterator for CodepointIter<'_, R> where R: io::BufRead, { type Item = Result<Rc<String>>; //noinspection DuplicatedCode fn next(&mut self) -> Option<Self::Item> { if self.ended { return None; } let buf = match Rc::get_mut(&mut self.buf) { None => { self.buf = Rc::new(String::with_capacity(self.default_cap)); Rc::make_mut(&mut self.buf) } Some(buf) => { buf.clear(); buf } }; loop { match self.reader.read_codepoint() { Err(e) => { if let ErrorKind::Interrupted = e.kind() { continue; } self.ended = true; break Some(Err(e)); } Ok(s) => { if s.is_empty() { self.ended = true; break None; } else { buf.push_str(s.as_ref()); break Some(Ok(Rc::clone(&self.buf))); } } } } } } /// A mapping iterator over string slices of an instance of [`io::BufRead`], created by /// [`str_map`], see its documentation for more details. /// /// [`str_map`]: self::BufRead::str_map pub struct StrMap<'r, R, F> where R: ?Sized, { reader: &'r mut R, map: Rc<F>, ended: bool, } impl<R, F, T> Iterator for StrMap<'_, R, F> where R: io::BufRead, F: FnMut(Cow<str>) -> T, { type Item = Result<T>; //noinspection DuplicatedCode fn next(&mut self) -> Option<Self::Item> { if self.ended { return None; } loop { match self.reader.read_str() { Ok(s) => { if s.is_empty() { self.ended = true; break None; } else { break Some(Ok((Rc::get_mut(&mut self.map) .expect("MappingIter's mapping function cannot be shared !"))( s ))); } } Err(e) => { if let ErrorKind::Interrupted = e.kind() { continue; } self.ended = true; break Some(Err(e)); } } } } } /// A mapping iterator over codepoints of an instance of [`io::BufRead`], created by [`str_map`], /// see its documentation for more details. #[doc(hidden)] pub struct CodepointMap<'r, R, F> where R: ?Sized, { reader: &'r mut R, map: Rc<F>, ended: bool, } impl<R, F, T> Iterator for CodepointMap<'_, R, F> where R: io::BufRead, F: FnMut(Cow<str>) -> T, { type Item = Result<T>; //noinspection DuplicatedCode fn next(&mut self) -> Option<Self::Item> { if self.ended { return None; } loop { match self.reader.read_codepoint() { Ok(s) => { if s.is_empty() { self.ended = true; break None; } else { break Some(Ok((Rc::get_mut(&mut self.map) .expect("MappingIter's mapping function cannot be shared !"))( s ))); } } Err(e) => { if let ErrorKind::Interrupted = e.kind() { continue; } self.ended = true; break Some(Err(e)); } } } } } /// An iterator over chars of an instance of [`io::BufRead`], created by [`char_iter`], see its /// documentation for more details. /// /// [`char_iter`]: self::BufRead::char_iter pub struct CharIter<'r, R> where R: ?Sized, { reader: &'r mut R, ended: bool, } impl<R> Iterator for CharIter<'_, R> where R: io::BufRead, { type Item = Result<char>; fn next(&mut self) -> Option<Self::Item> { if self.ended { return None; } match self.reader.read_char() { Ok(c) => { if c == '\0' { self.ended = true; None } else { Some(Ok(c)) } } Err(e) => { self.ended = true; Some(Err(e)) } } } } fn read_across_boundary<R>(reader: &mut R, mut leftovers: Vec<u8>) -> Result<Cow<str>> where R: io::BufRead + ?Sized, { debug_assert!(!leftovers.is_empty()); // We know leftovers is not empty let len = codepoint_length(leftovers[0]); let first_read_len = leftovers.len(); debug_assert!(len > first_read_len); let additional_len = (len - first_read_len) as usize; // Let's try reading more bytes let additional_bytes = &reader.fill_buf()?; if additional_bytes.len() < additional_len { // Not enough additional bytes, we reached EOF on an incomplete codepoint return Err(Error::from(ErrorKind::UnexpectedEof).with_leftovers(leftovers)); } // we know we have enough data leftovers.extend_from_slice(&additional_bytes[..additional_len]); reader.consume(additional_len); match String::from_utf8(leftovers) { Ok(s) => Ok(Cow::from(s)), // We read enough bytes, they simply were not valid Err(e) => Err(Error::from(e)), } } #[inline] fn codepoint_length(x: u8) -> usize { if x < 0x80 { 1 } else if x < 0xE0 { 2 } else if x < 0xF0 { 3 } else { 4 } } #[cfg(test)] mod read_str_tests { use crate::BufRead; use std::io::{BufReader, Cursor, ErrorKind}; use std::str::Utf8Error; use std::string::FromUtf8Error; #[test] fn empty_read() { let mut r = Cursor::new(""); let s = r.read_str(); assert!(s.is_ok()); let s = s.unwrap(); assert!(s.is_empty()); } #[test] fn invalid_in_buffer() { let mut r = Cursor::new([0x9fu8, 0x92, 0x96, 0x0]); let e = r.read_str(); assert!(e.is_err()); let e = e.unwrap_err(); assert_eq!(e.kind(), ErrorKind::InvalidData); let e = e.into_inner_checked(); assert!(e.is_ok()); let e = e.unwrap(); assert!(e.is_some()); let e = e.unwrap(); assert!(e.is::<Utf8Error>()); } #[test] fn incomplete_in_buffer() { let mut r = Cursor::new(&"💖".as_bytes()[..3]); let e = r.read_str(); assert!(e.is_err()); let e = e.unwrap_err(); assert_eq!(e.kind(), ErrorKind::UnexpectedEof); assert!(!e.leftovers().is_empty()); let e = e.into_inner_lossy(); assert!(e.is_none()); } #[test] fn invalid_across_boundary() { let mut r = BufReader::<&[u8]>::with_capacity(2, [0xffu8, 0x92, 0x96, 0x0].as_ref()); let e = r.read_str(); assert!(e.is_err()); let e = e.unwrap_err(); assert_eq!(e.kind(), ErrorKind::InvalidData); assert!(!e.leftovers().is_empty()); let e = e.into_inner_lossy(); assert!(e.is_some()); let e = e.unwrap(); assert!(e.is::<FromUtf8Error>()); } #[test] fn incomplete_across_boundary() { let mut r = BufReader::<&[u8]>::with_capacity(2, &"💖".as_bytes()[..3]); let e = r.read_str(); assert!(e.is_err()); let e = e.unwrap_err(); assert_eq!(e.kind(), ErrorKind::UnexpectedEof); let e = e.into_inner_lossy(); assert!(e.is_none()); } #[test] fn complete_successful_read() { let mut r = Cursor::new("💖"); let s = r.read_str(); assert!(s.is_ok()); let s = s.unwrap(); assert_eq!(s, "💖"); } #[test] fn incomplete_successful_read() { let mut r = Cursor::new([0x6fu8, 0xa, 0x9f, 0x92, 0x96, 0x0]); let s = r.read_str(); assert!(s.is_ok()); let s = s.unwrap(); assert_eq!(s, "o\n"); } #[test] fn read_across_boundary() { let mut r = BufReader::<&[u8]>::with_capacity(2, "💖".as_ref()); let s = r.read_str(); assert!(s.is_ok()); let s = s.unwrap(); assert_eq!(s, "💖"); } #[test] fn multi_codepoints_read() { let mut r = Cursor::new("foo💖bär€"); let s = r.read_str(); assert!(s.is_ok()); let s = s.unwrap(); assert_eq!(s, "foo💖bär€"); let s = r.read_str(); assert!(s.is_ok()); let s = s.unwrap(); assert_eq!(s, ""); } } #[cfg(test)] mod buf_too_small_tests { macro_rules! buf_too_small_test { ($name:ident $cap:literal $input:literal: success) => { #[test] fn $name() { let mut r = BufReader::<&[u8]>::with_capacity($cap, $input.as_bytes()); let mut call_count = 0; // Reading until EOF loop { let s = r.read_str(); assert!(s.is_ok()); let s = s.unwrap(); if s.is_empty() { break; } else { call_count += 1; } } // Asserting we did not encounter EOF on the first call assert_ne!(call_count, 0); } }; ($name:ident $cap:literal $input:literal: failure) => { #[test] fn $name() { let mut r = BufReader::<&[u8]>::with_capacity($cap, $input.as_bytes()); // Reading until we fail loop { let e = r.read_str(); match e { Ok(s) => { // We shouldn't reach EOF without failing a read assert!(!s.is_empty()); } Err(e) => { assert_eq!(e.kind(), ErrorKind::UnexpectedEof); assert!(!e.leftovers().is_empty()); let e = e.into_inner_lossy(); assert!(e.is_none()); break; } } } } }; } mod buf_capacity_1 { use crate::BufRead; use std::io::{BufReader, ErrorKind}; buf_too_small_test!(codepoint_length_1_offset_0 1 "f": success); buf_too_small_test!(codepoint_length_2_offset_0 1 "ä": success); buf_too_small_test!(codepoint_length_3_offset_0 1 "€": failure); buf_too_small_test!(codepoint_length_4_offset_0 1 "💖": failure); } mod buf_capacity_2 { use crate::BufRead; use std::io::{BufReader, ErrorKind}; buf_too_small_test!(codepoint_length_1_offset_0 2 "f": success); buf_too_small_test!(codepoint_length_2_offset_0 2 "ä": success); buf_too_small_test!(codepoint_length_2_offset_1 2 "xä": success); buf_too_small_test!(codepoint_length_3_offset_0 2 "€": success); buf_too_small_test!(codepoint_length_3_offset_1 2 "x€": success); buf_too_small_test!(codepoint_length_4_offset_0 2 "💖": success); buf_too_small_test!(codepoint_length_4_offset_1 2 "x💖": failure); } mod buf_capacity_3 { use crate::BufRead; use std::io::BufReader; buf_too_small_test!(codepoint_length_1_offset_0 3 "f": success); buf_too_small_test!(codepoint_length_2_offset_0 3 "ä": success); buf_too_small_test!(codepoint_length_2_offset_1 3 "xä": success); buf_too_small_test!(codepoint_length_3_offset_0 3 "€": success); buf_too_small_test!(codepoint_length_3_offset_1 3 "x€": success); buf_too_small_test!(codepoint_length_3_offset_2 3 "xx€": success); buf_too_small_test!(codepoint_length_4_offset_0 3 "💖": success); buf_too_small_test!(codepoint_length_4_offset_1 3 "x💖": success); buf_too_small_test!(codepoint_length_4_offset_2 3 "xx💖": success); } }