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
// Copyright 2017 Brian Langenberger // // 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. //! Traits and implementations for reading bits from a stream. //! //! ## Example //! ``` //! use std::io::{Cursor, Read}; //! use bitstream_io::{BitRead, BitReaderBE}; //! //! let flac: Vec<u8> = vec![0x66,0x4C,0x61,0x43,0x00,0x00,0x00,0x22, //! 0x10,0x00,0x10,0x00,0x00,0x06,0x06,0x00, //! 0x21,0x62,0x0A,0xC4,0x42,0xF0,0x00,0x04, //! 0xA6,0xCC,0xFA,0xF2,0x69,0x2F,0xFD,0xEC, //! 0x2D,0x5B,0x30,0x01,0x76,0xB4,0x62,0x88, //! 0x7D,0x92]; //! //! let mut cursor = Cursor::new(&flac); //! { //! let mut reader = BitReaderBE::new(&mut cursor); //! let mut file_header: [u8; 4] = [0, 0, 0, 0]; //! reader.read_bytes(&mut file_header).unwrap(); //! assert_eq!(&file_header, b"fLaC"); //! //! let last_block: bool = reader.read_bit().unwrap(); //! let block_type: u8 = reader.read(7).unwrap(); //! let block_size: u32 = reader.read(24).unwrap(); //! assert_eq!(last_block, false); //! assert_eq!(block_type, 0); //! assert_eq!(block_size, 34); //! //! let minimum_block_size: u16 = reader.read(16).unwrap(); //! let maximum_block_size: u16 = reader.read(16).unwrap(); //! let minimum_frame_size: u32 = reader.read(24).unwrap(); //! let maximum_frame_size: u32 = reader.read(24).unwrap(); //! let sample_rate: u32 = reader.read(20).unwrap(); //! let channels = reader.read::<u8>(3).unwrap() + 1; //! let bits_per_sample = reader.read::<u8>(5).unwrap() + 1; //! let total_samples: u64 = reader.read(36).unwrap(); //! assert_eq!(minimum_block_size, 4096); //! assert_eq!(maximum_block_size, 4096); //! assert_eq!(minimum_frame_size, 1542); //! assert_eq!(maximum_frame_size, 8546); //! assert_eq!(sample_rate, 44100); //! assert_eq!(channels, 2); //! assert_eq!(bits_per_sample, 16); //! assert_eq!(total_samples, 304844); //! } //! //! // the wrapped reader can be used once bitstream reading is finished //! // at exactly the position one would expect //! let mut md5: [u8; 16] = [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0]; //! cursor.read_exact(&mut md5).unwrap(); //! assert_eq!(&md5, //! b"\xFA\xF2\x69\x2F\xFD\xEC\x2D\x5B\x30\x01\x76\xB4\x62\x88\x7D\x92"); //! ``` use std::io; use super::{Numeric, SignedNumeric, BitQueueBE, BitQueueLE, BitQueue}; use huffman::ReadHuffmanTree; /// For reading bit values from an underlying stream in a given endianness. pub trait BitRead { /// Reads a single bit from the stream. /// `true` indicates 1, `false` indicates 0 /// /// # Examples /// /// ``` /// use std::io::{Read, Cursor}; /// use bitstream_io::{BitRead, BitReaderBE}; /// let data = [0b10110111]; /// let mut cursor = Cursor::new(&data); /// let mut reader = BitReaderBE::new(&mut cursor); /// assert_eq!(reader.read_bit().unwrap(), true); /// assert_eq!(reader.read_bit().unwrap(), false); /// assert_eq!(reader.read_bit().unwrap(), true); /// assert_eq!(reader.read_bit().unwrap(), true); /// assert_eq!(reader.read_bit().unwrap(), false); /// assert_eq!(reader.read_bit().unwrap(), true); /// assert_eq!(reader.read_bit().unwrap(), true); /// assert_eq!(reader.read_bit().unwrap(), true); /// ``` /// /// ``` /// use std::io::{Read, Cursor}; /// use bitstream_io::{BitRead, BitReaderLE}; /// let data = [0b10110111]; /// let mut cursor = Cursor::new(&data); /// let mut reader = BitReaderLE::new(&mut cursor); /// assert_eq!(reader.read_bit().unwrap(), true); /// assert_eq!(reader.read_bit().unwrap(), true); /// assert_eq!(reader.read_bit().unwrap(), true); /// assert_eq!(reader.read_bit().unwrap(), false); /// assert_eq!(reader.read_bit().unwrap(), true); /// assert_eq!(reader.read_bit().unwrap(), true); /// assert_eq!(reader.read_bit().unwrap(), false); /// assert_eq!(reader.read_bit().unwrap(), true); /// ``` fn read_bit(&mut self) -> Result<bool, io::Error>; /// Reads an unsigned value from the stream with /// the given number of bits. This method assumes /// that the programmer is using an output type /// sufficiently large to hold those bits. /// /// # Examples /// ``` /// use std::io::{Read, Cursor}; /// use bitstream_io::{BitRead, BitReaderBE}; /// let data = [0b10110111]; /// let mut cursor = Cursor::new(&data); /// let mut reader = BitReaderBE::new(&mut cursor); /// assert_eq!(reader.read::<u8>(1).unwrap(), 0b1); /// assert_eq!(reader.read::<u8>(2).unwrap(), 0b01); /// assert_eq!(reader.read::<u8>(5).unwrap(), 0b10111); /// ``` /// /// ``` /// use std::io::{Read, Cursor}; /// use bitstream_io::{BitRead, BitReaderLE}; /// let data = [0b10110111]; /// let mut cursor = Cursor::new(&data); /// let mut reader = BitReaderLE::new(&mut cursor); /// assert_eq!(reader.read::<u8>(1).unwrap(), 0b1); /// assert_eq!(reader.read::<u8>(2).unwrap(), 0b11); /// assert_eq!(reader.read::<u8>(5).unwrap(), 0b10110); /// ``` fn read<U>(&mut self, bits: u32) -> Result<U, io::Error> where U: Numeric; /// Reads a twos-complement signed value from the stream with /// the given number of bits. This method assumes /// that the programmer is using an output type /// sufficiently large to hold those bits. /// /// # Examples /// ``` /// use std::io::{Read, Cursor}; /// use bitstream_io::{BitRead, BitReaderBE}; /// let data = [0b10110111]; /// let mut cursor = Cursor::new(&data); /// let mut reader = BitReaderBE::new(&mut cursor); /// assert_eq!(reader.read_signed::<i8>(4).unwrap(), -5); /// assert_eq!(reader.read_signed::<i8>(4).unwrap(), 7); /// ``` /// /// ``` /// use std::io::{Read, Cursor}; /// use bitstream_io::{BitRead, BitReaderLE}; /// let data = [0b10110111]; /// let mut cursor = Cursor::new(&data); /// let mut reader = BitReaderLE::new(&mut cursor); /// assert_eq!(reader.read_signed::<i8>(4).unwrap(), 7); /// assert_eq!(reader.read_signed::<i8>(4).unwrap(), -5); /// ``` fn read_signed<S>(&mut self, bits: u32) -> Result<S, io::Error> where S: SignedNumeric; /// Skips the given number of bits in the stream. /// Since this method does not need an accumulator, /// it may be slightly faster than reading to an empty variable. /// In addition, since there is no accumulator, /// there is no upper limit on the number of bits /// which may be skipped. /// These bits are still read from the stream, however, /// and are never skipped via a `seek` method. /// /// # Examples /// ``` /// use std::io::{Read, Cursor}; /// use bitstream_io::{BitRead, BitReaderBE}; /// let data = [0b10110111]; /// let mut cursor = Cursor::new(&data); /// let mut reader = BitReaderBE::new(&mut cursor); /// assert!(reader.skip(3).is_ok()); /// assert_eq!(reader.read::<u8>(5).unwrap(), 0b10111); /// ``` /// /// ``` /// use std::io::{Read, Cursor}; /// use bitstream_io::{BitRead, BitReaderLE}; /// let data = [0b10110111]; /// let mut cursor = Cursor::new(&data); /// let mut reader = BitReaderLE::new(&mut cursor); /// assert!(reader.skip(3).is_ok()); /// assert_eq!(reader.read::<u8>(5).unwrap(), 0b10110); /// ``` fn skip(&mut self, bits: u32) -> Result<(), io::Error>; /// Completely fills the given buffer with whole bytes. /// If the stream is already byte-aligned, it will typically map /// to a faster `read_exact` call. Otherwise it will read /// bytes individually in 8-bit increments. /// /// # Example /// ``` /// use std::io::{Read, Cursor}; /// use bitstream_io::{BitRead, BitReaderBE}; /// let data = b"foobar"; /// let mut cursor = Cursor::new(&data); /// let mut reader = BitReaderBE::new(&mut cursor); /// assert!(reader.skip(24).is_ok()); /// let mut buf = [0;3]; /// assert!(reader.read_bytes(&mut buf).is_ok()); /// assert_eq!(&buf, b"bar"); /// ``` fn read_bytes(&mut self, buf: &mut [u8]) -> Result<(), io::Error>; /// Counts the number of 1 bits in the stream until the next /// 0 bit and returns the amount read. /// Because this field is variably-sized and may be large, /// its output is always a `u32` type. /// /// # Examples /// ``` /// use std::io::{Read, Cursor}; /// use bitstream_io::{BitRead, BitReaderBE}; /// let data = [0b01110111, 0b11111110]; /// let mut cursor = Cursor::new(&data); /// let mut reader = BitReaderBE::new(&mut cursor); /// assert_eq!(reader.read_unary0().unwrap(), 0); /// assert_eq!(reader.read_unary0().unwrap(), 3); /// assert_eq!(reader.read_unary0().unwrap(), 10); /// ``` /// /// ``` /// use std::io::{Read, Cursor}; /// use bitstream_io::{BitRead, BitReaderLE}; /// let data = [0b11101110, 0b01111111]; /// let mut cursor = Cursor::new(&data); /// let mut reader = BitReaderLE::new(&mut cursor); /// assert_eq!(reader.read_unary0().unwrap(), 0); /// assert_eq!(reader.read_unary0().unwrap(), 3); /// assert_eq!(reader.read_unary0().unwrap(), 10); /// ``` fn read_unary0(&mut self) -> Result<u32, io::Error>; /// Counts the number of 0 bits in the stream until the next /// 1 bit and returns the amount read. /// Because this field is variably-sized and may be large, /// its output is always a `u32` type. /// /// # Examples /// ``` /// use std::io::{Read, Cursor}; /// use bitstream_io::{BitRead, BitReaderBE}; /// let data = [0b10001000, 0b00000001]; /// let mut cursor = Cursor::new(&data); /// let mut reader = BitReaderBE::new(&mut cursor); /// assert_eq!(reader.read_unary1().unwrap(), 0); /// assert_eq!(reader.read_unary1().unwrap(), 3); /// assert_eq!(reader.read_unary1().unwrap(), 10); /// ``` /// /// ``` /// use std::io::{Read, Cursor}; /// use bitstream_io::{BitRead, BitReaderLE}; /// let data = [0b00010001, 0b10000000]; /// let mut cursor = Cursor::new(&data); /// let mut reader = BitReaderLE::new(&mut cursor); /// assert_eq!(reader.read_unary1().unwrap(), 0); /// assert_eq!(reader.read_unary1().unwrap(), 3); /// assert_eq!(reader.read_unary1().unwrap(), 10); /// ``` fn read_unary1(&mut self) -> Result<u32, io::Error>; /// Returns true if the stream is aligned at a whole byte. /// /// # Example /// ``` /// use std::io::{Read, Cursor}; /// use bitstream_io::{BitRead, BitReaderBE}; /// let data = [0]; /// let mut cursor = Cursor::new(&data); /// let mut reader = BitReaderBE::new(&mut cursor); /// assert_eq!(reader.byte_aligned(), true); /// assert!(reader.skip(1).is_ok()); /// assert_eq!(reader.byte_aligned(), false); /// assert!(reader.skip(7).is_ok()); /// assert_eq!(reader.byte_aligned(), true); /// ``` fn byte_aligned(&self) -> bool; /// Throws away all unread bit values until the next whole byte. /// /// # Example /// ``` /// use std::io::{Read, Cursor}; /// use bitstream_io::{BitRead, BitReaderBE}; /// let data = [0x00, 0xFF]; /// let mut cursor = Cursor::new(&data); /// let mut reader = BitReaderBE::new(&mut cursor); /// assert_eq!(reader.read::<u8>(4).unwrap(), 0); /// reader.byte_align(); /// assert_eq!(reader.read::<u8>(8).unwrap(), 0xFF); /// ``` fn byte_align(&mut self); /// Given a compiled Huffman tree, reads bits from the stream /// until the next symbol is encountered. /// /// # Example /// ``` /// use std::io::{Read, Cursor}; /// use bitstream_io::{BitRead, BitReaderBE}; /// use bitstream_io::huffman::ReadHuffmanTree; /// let tree = ReadHuffmanTree::new( /// vec![('a', vec![0]), /// ('b', vec![1, 0]), /// ('c', vec![1, 1, 0]), /// ('d', vec![1, 1, 1])]).unwrap(); /// let data = [0b10110111]; /// let mut cursor = Cursor::new(&data); /// let mut reader = BitReaderBE::new(&mut cursor); /// assert_eq!(reader.read_huffman(&tree).unwrap(), 'b'); /// assert_eq!(reader.read_huffman(&tree).unwrap(), 'c'); /// assert_eq!(reader.read_huffman(&tree).unwrap(), 'd'); /// ``` fn read_huffman<T>(&mut self, mut tree: &ReadHuffmanTree<T>) -> Result<T,io::Error> where T: Clone { loop { match tree { &ReadHuffmanTree::Leaf(ref v) => {return Ok(v.clone());} &ReadHuffmanTree::Tree(ref zero, ref one) => { tree = match self.read_bit() { Ok(false) => {zero} Ok(true) => {one} Err(err) => {return Err(err);} }; } } } } } macro_rules! define_read_bit { () => { #[inline(always)] fn read_bit(&mut self) -> Result<bool, io::Error> { if self.bitqueue.is_empty() { self.bitqueue.set(read_byte(self.reader)?, 8); } Ok(self.bitqueue.pop(1) == 1) } } } macro_rules! define_read { ($bitqueue:ident) => { fn read<U>(&mut self, mut bits: u32) -> Result<U, io::Error> where U: Numeric { use std::cmp::min; let mut acc: $bitqueue<U> = $bitqueue::new(); /*transfer un-processed bits from queue to accumulator*/ let queue_len = self.bitqueue.len(); if queue_len > 0 { let to_transfer = min(queue_len, bits); acc.push(to_transfer, U::from_u8(self.bitqueue.pop(to_transfer))); bits -= to_transfer; } read_aligned(&mut self.reader, bits / 8, &mut acc) .and_then(|()| read_unaligned(&mut self.reader, bits % 8, &mut acc, &mut self.bitqueue)) .map(|()| acc.value()) } } } macro_rules! define_skip { () => { fn skip(&mut self, mut bits: u32) -> Result<(), io::Error> { use std::cmp::min; let queue_len = self.bitqueue.len(); if queue_len > 0 { let to_drop = min(queue_len, bits); self.bitqueue.drop(to_drop); bits -= to_drop; } skip_aligned(&mut self.reader, bits / 8) .and_then(|()| skip_unaligned(&mut self.reader, bits % 8, &mut self.bitqueue)) } } } macro_rules! define_read_bytes { () => { fn read_bytes(&mut self, buf: &mut [u8]) -> Result<(), io::Error> { if self.byte_aligned() { self.reader.read_exact(buf) } else { for b in buf.iter_mut() { *b = self.read::<u8>(8)?; } Ok(()) } } } } macro_rules! define_read_unary { ($method_name:ident, $aligned_cont_val: expr, $bitqueue_check: ident, $bitqueue_pop: ident) => { fn $method_name(&mut self) -> Result<u32, io::Error> { if self.bitqueue.is_empty() { read_aligned_unary(&mut self.reader, $aligned_cont_val, &mut self.bitqueue).map( |u| u + self.bitqueue.$bitqueue_pop()) } else if self.bitqueue.$bitqueue_check() { let base = self.bitqueue.len(); self.bitqueue.clear(); read_aligned_unary(&mut self.reader, $aligned_cont_val, &mut self.bitqueue).map( |u| base + u + self.bitqueue.$bitqueue_pop()) } else { Ok(self.bitqueue.$bitqueue_pop()) } } } } /// A wrapper for reading values from a big-endian stream. pub struct BitReaderBE<'a> { reader: &'a mut io::Read, bitqueue: BitQueueBE<u8> } impl<'a> BitReaderBE<'a> { /// Wraps a big-endian reader around a `Read` reference. pub fn new(reader: &mut io::Read) -> BitReaderBE { BitReaderBE{reader: reader, bitqueue: BitQueueBE::new()} } } impl<'a> BitRead for BitReaderBE<'a> { define_read_bit!(); define_read!(BitQueueBE); define_skip!(); define_read_bytes!(); define_read_unary!(read_unary0, 0xFF, all_1, pop_1); define_read_unary!(read_unary1, 0x00, all_0, pop_0); fn read_signed<S>(&mut self, bits: u32) -> Result<S, io::Error> where S: SignedNumeric { debug_assert!(bits >= 1); let is_negative = self.read_bit()?; let unsigned = self.read::<S>(bits - 1)?; Ok(if is_negative {unsigned.as_negative(bits)} else {unsigned}) } #[inline] fn byte_aligned(&self) -> bool {self.bitqueue.is_empty()} #[inline] fn byte_align(&mut self) {self.bitqueue.clear()} } /// A wrapper for reading values from a little-endian stream. pub struct BitReaderLE<'a> { reader: &'a mut io::Read, bitqueue: BitQueueLE<u8> } impl<'a> BitReaderLE<'a> { /// Wraps a little-endian reader around a `Read` reference. pub fn new(reader: &mut io::Read) -> BitReaderLE { BitReaderLE{reader: reader, bitqueue: BitQueueLE::new()} } } impl<'a> BitRead for BitReaderLE<'a> { define_read_bit!(); define_read!(BitQueueLE); define_skip!(); define_read_bytes!(); define_read_unary!(read_unary0, 0xFF, all_1, pop_1); define_read_unary!(read_unary1, 0x00, all_0, pop_0); fn read_signed<S>(&mut self, bits: u32) -> Result<S, io::Error> where S: SignedNumeric { debug_assert!(bits >= 1); let unsigned = self.read::<S>(bits - 1)?; let is_negative = self.read_bit()?; Ok(if is_negative {unsigned.as_negative(bits)} else {unsigned}) } #[inline] fn byte_aligned(&self) -> bool {self.bitqueue.is_empty()} #[inline] fn byte_align(&mut self) {self.bitqueue.clear()} } #[inline] fn read_byte(reader: &mut io::Read) -> Result<u8,io::Error> { let mut buf = [0; 1]; reader.read_exact(&mut buf).map(|()| buf[0]) } fn read_aligned<N>(reader: &mut io::Read, mut bytes: u32, acc: &mut BitQueue<N>) -> Result<(), io::Error> where N: Numeric { use std::cmp::min; // for native types, it's difficult to imagine a situation // in which this would require more than a single pass let mut buf = [0; 8]; while bytes > 0 { let to_read = min(8, bytes); reader.read_exact(&mut buf[0..to_read as usize])?; for b in buf.iter().take(to_read as usize) { acc.push(8, N::from_u8(*b)); } bytes -= to_read; } Ok(()) } fn skip_aligned(reader: &mut io::Read, mut bytes: u32) -> Result<(), io::Error> { use std::cmp::min; // for native types, it's difficult to imagine a situation // in which this would require more than a single pass let mut buf = [0; 8]; while bytes > 0 { let to_read = min(8, bytes); reader.read_exact(&mut buf[0..to_read as usize])?; bytes -= to_read; } Ok(()) } #[inline] fn read_unaligned<N>(reader: &mut io::Read, bits: u32, acc: &mut BitQueue<N>, rem: &mut BitQueue<u8>) -> Result<(), io::Error> where N: Numeric { debug_assert!(bits <= 8); if bits > 0 { rem.set(read_byte(reader)?, 8); acc.push(bits, N::from_u8(rem.pop(bits))); } Ok(()) } #[inline] fn skip_unaligned(reader: &mut io::Read, bits: u32, rem: &mut BitQueue<u8>) -> Result<(), io::Error> { debug_assert!(bits <= 8); if bits > 0 { rem.set(read_byte(reader)?, 8); rem.pop(bits); } Ok(()) } #[inline] fn read_aligned_unary(reader: &mut io::Read, continue_val: u8, rem: &mut BitQueue<u8>) -> Result<u32,io::Error> { let mut acc = 0; let mut byte = read_byte(reader)?; while byte == continue_val { acc += 8; byte = read_byte(reader)?; } rem.set(byte, 8); Ok(acc) }