mirror of
https://github.com/ZDoom/qzdoom.git
synced 2024-11-10 23:02:08 +00:00
2e56925c36
installation, not the bundled copies which might not match what is installed. - Upgraded bundled FLAC from version 1.1.2 to version 1.2.1. SVN r575 (trunk)
1376 lines
41 KiB
C
1376 lines
41 KiB
C
/* libFLAC - Free Lossless Audio Codec library
|
|
* Copyright (C) 2000,2001,2002,2003,2004,2005,2006,2007 Josh Coalson
|
|
*
|
|
* Redistribution and use in source and binary forms, with or without
|
|
* modification, are permitted provided that the following conditions
|
|
* are met:
|
|
*
|
|
* - Redistributions of source code must retain the above copyright
|
|
* notice, this list of conditions and the following disclaimer.
|
|
*
|
|
* - Redistributions in binary form must reproduce the above copyright
|
|
* notice, this list of conditions and the following disclaimer in the
|
|
* documentation and/or other materials provided with the distribution.
|
|
*
|
|
* - Neither the name of the Xiph.org Foundation nor the names of its
|
|
* contributors may be used to endorse or promote products derived from
|
|
* this software without specific prior written permission.
|
|
*
|
|
* THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
|
|
* ``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
|
|
* LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
|
|
* A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE FOUNDATION OR
|
|
* CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL,
|
|
* EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
|
|
* PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR
|
|
* PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF
|
|
* LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING
|
|
* NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
|
|
* SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
|
|
*/
|
|
|
|
#if HAVE_CONFIG_H
|
|
# include <config.h>
|
|
#endif
|
|
|
|
#include <stdlib.h> /* for malloc() */
|
|
#include <string.h> /* for memcpy(), memset() */
|
|
#ifdef _MSC_VER
|
|
#include <winsock.h> /* for ntohl() */
|
|
#elif defined FLAC__SYS_DARWIN
|
|
#include <machine/endian.h> /* for ntohl() */
|
|
#elif defined __MINGW32__
|
|
#include <winsock.h> /* for ntohl() */
|
|
#else
|
|
#include <netinet/in.h> /* for ntohl() */
|
|
#endif
|
|
#include "private/bitmath.h"
|
|
#include "private/bitreader.h"
|
|
#include "private/crc.h"
|
|
#include "FLAC/assert.h"
|
|
|
|
/* Things should be fastest when this matches the machine word size */
|
|
/* WATCHOUT: if you change this you must also change the following #defines down to COUNT_ZERO_MSBS below to match */
|
|
/* WATCHOUT: there are a few places where the code will not work unless brword is >= 32 bits wide */
|
|
/* also, some sections currently only have fast versions for 4 or 8 bytes per word */
|
|
typedef FLAC__uint32 brword;
|
|
#define FLAC__BYTES_PER_WORD 4
|
|
#define FLAC__BITS_PER_WORD 32
|
|
#define FLAC__WORD_ALL_ONES ((FLAC__uint32)0xffffffff)
|
|
/* SWAP_BE_WORD_TO_HOST swaps bytes in a brword (which is always big-endian) if necessary to match host byte order */
|
|
#if WORDS_BIGENDIAN
|
|
#define SWAP_BE_WORD_TO_HOST(x) (x)
|
|
#else
|
|
#ifdef _MSC_VER
|
|
#define SWAP_BE_WORD_TO_HOST(x) local_swap32_(x)
|
|
#else
|
|
#define SWAP_BE_WORD_TO_HOST(x) ntohl(x)
|
|
#endif
|
|
#endif
|
|
/* counts the # of zero MSBs in a word */
|
|
#define COUNT_ZERO_MSBS(word) ( \
|
|
(word) <= 0xffff ? \
|
|
( (word) <= 0xff? byte_to_unary_table[word] + 24 : byte_to_unary_table[(word) >> 8] + 16 ) : \
|
|
( (word) <= 0xffffff? byte_to_unary_table[word >> 16] + 8 : byte_to_unary_table[(word) >> 24] ) \
|
|
)
|
|
/* this alternate might be slightly faster on some systems/compilers: */
|
|
#define COUNT_ZERO_MSBS2(word) ( (word) <= 0xff ? byte_to_unary_table[word] + 24 : ((word) <= 0xffff ? byte_to_unary_table[(word) >> 8] + 16 : ((word) <= 0xffffff ? byte_to_unary_table[(word) >> 16] + 8 : byte_to_unary_table[(word) >> 24])) )
|
|
|
|
|
|
/*
|
|
* This should be at least twice as large as the largest number of words
|
|
* required to represent any 'number' (in any encoding) you are going to
|
|
* read. With FLAC this is on the order of maybe a few hundred bits.
|
|
* If the buffer is smaller than that, the decoder won't be able to read
|
|
* in a whole number that is in a variable length encoding (e.g. Rice).
|
|
* But to be practical it should be at least 1K bytes.
|
|
*
|
|
* Increase this number to decrease the number of read callbacks, at the
|
|
* expense of using more memory. Or decrease for the reverse effect,
|
|
* keeping in mind the limit from the first paragraph. The optimal size
|
|
* also depends on the CPU cache size and other factors; some twiddling
|
|
* may be necessary to squeeze out the best performance.
|
|
*/
|
|
static const unsigned FLAC__BITREADER_DEFAULT_CAPACITY = 65536u / FLAC__BITS_PER_WORD; /* in words */
|
|
|
|
static const unsigned char byte_to_unary_table[] = {
|
|
8, 7, 6, 6, 5, 5, 5, 5, 4, 4, 4, 4, 4, 4, 4, 4,
|
|
3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3,
|
|
2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
|
|
2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
|
|
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, 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
|
|
};
|
|
|
|
#ifdef min
|
|
#undef min
|
|
#endif
|
|
#define min(x,y) ((x)<(y)?(x):(y))
|
|
#ifdef max
|
|
#undef max
|
|
#endif
|
|
#define max(x,y) ((x)>(y)?(x):(y))
|
|
|
|
/* adjust for compilers that can't understand using LLU suffix for uint64_t literals */
|
|
#ifdef _MSC_VER
|
|
#define FLAC__U64L(x) x
|
|
#else
|
|
#define FLAC__U64L(x) x##LLU
|
|
#endif
|
|
|
|
#ifndef FLaC__INLINE
|
|
#define FLaC__INLINE
|
|
#endif
|
|
|
|
/* WATCHOUT: assembly routines rely on the order in which these fields are declared */
|
|
struct FLAC__BitReader {
|
|
/* any partially-consumed word at the head will stay right-justified as bits are consumed from the left */
|
|
/* any incomplete word at the tail will be left-justified, and bytes from the read callback are added on the right */
|
|
brword *buffer;
|
|
unsigned capacity; /* in words */
|
|
unsigned words; /* # of completed words in buffer */
|
|
unsigned bytes; /* # of bytes in incomplete word at buffer[words] */
|
|
unsigned consumed_words; /* #words ... */
|
|
unsigned consumed_bits; /* ... + (#bits of head word) already consumed from the front of buffer */
|
|
unsigned read_crc16; /* the running frame CRC */
|
|
unsigned crc16_align; /* the number of bits in the current consumed word that should not be CRC'd */
|
|
FLAC__BitReaderReadCallback read_callback;
|
|
void *client_data;
|
|
FLAC__CPUInfo cpu_info;
|
|
};
|
|
|
|
#ifdef _MSC_VER
|
|
/* OPT: an MSVC built-in would be better */
|
|
static _inline FLAC__uint32 local_swap32_(FLAC__uint32 x)
|
|
{
|
|
x = ((x<<8)&0xFF00FF00) | ((x>>8)&0x00FF00FF);
|
|
return (x>>16) | (x<<16);
|
|
}
|
|
static void local_swap32_block_(FLAC__uint32 *start, FLAC__uint32 len)
|
|
{
|
|
__asm {
|
|
mov edx, start
|
|
mov ecx, len
|
|
test ecx, ecx
|
|
loop1:
|
|
jz done1
|
|
mov eax, [edx]
|
|
bswap eax
|
|
mov [edx], eax
|
|
add edx, 4
|
|
dec ecx
|
|
jmp short loop1
|
|
done1:
|
|
}
|
|
}
|
|
#endif
|
|
|
|
static FLaC__INLINE void crc16_update_word_(FLAC__BitReader *br, brword word)
|
|
{
|
|
register unsigned crc = br->read_crc16;
|
|
#if FLAC__BYTES_PER_WORD == 4
|
|
switch(br->crc16_align) {
|
|
case 0: crc = FLAC__CRC16_UPDATE((unsigned)(word >> 24), crc);
|
|
case 8: crc = FLAC__CRC16_UPDATE((unsigned)((word >> 16) & 0xff), crc);
|
|
case 16: crc = FLAC__CRC16_UPDATE((unsigned)((word >> 8) & 0xff), crc);
|
|
case 24: br->read_crc16 = FLAC__CRC16_UPDATE((unsigned)(word & 0xff), crc);
|
|
}
|
|
#elif FLAC__BYTES_PER_WORD == 8
|
|
switch(br->crc16_align) {
|
|
case 0: crc = FLAC__CRC16_UPDATE((unsigned)(word >> 56), crc);
|
|
case 8: crc = FLAC__CRC16_UPDATE((unsigned)((word >> 48) & 0xff), crc);
|
|
case 16: crc = FLAC__CRC16_UPDATE((unsigned)((word >> 40) & 0xff), crc);
|
|
case 24: crc = FLAC__CRC16_UPDATE((unsigned)((word >> 32) & 0xff), crc);
|
|
case 32: crc = FLAC__CRC16_UPDATE((unsigned)((word >> 24) & 0xff), crc);
|
|
case 40: crc = FLAC__CRC16_UPDATE((unsigned)((word >> 16) & 0xff), crc);
|
|
case 48: crc = FLAC__CRC16_UPDATE((unsigned)((word >> 8) & 0xff), crc);
|
|
case 56: br->read_crc16 = FLAC__CRC16_UPDATE((unsigned)(word & 0xff), crc);
|
|
}
|
|
#else
|
|
for( ; br->crc16_align < FLAC__BITS_PER_WORD; br->crc16_align += 8)
|
|
crc = FLAC__CRC16_UPDATE((unsigned)((word >> (FLAC__BITS_PER_WORD-8-br->crc16_align)) & 0xff), crc);
|
|
br->read_crc16 = crc;
|
|
#endif
|
|
br->crc16_align = 0;
|
|
}
|
|
|
|
/* would be static except it needs to be called by asm routines */
|
|
FLAC__bool bitreader_read_from_client_(FLAC__BitReader *br)
|
|
{
|
|
unsigned start, end;
|
|
size_t bytes;
|
|
FLAC__byte *target;
|
|
|
|
/* first shift the unconsumed buffer data toward the front as much as possible */
|
|
if(br->consumed_words > 0) {
|
|
start = br->consumed_words;
|
|
end = br->words + (br->bytes? 1:0);
|
|
memmove(br->buffer, br->buffer+start, FLAC__BYTES_PER_WORD * (end - start));
|
|
|
|
br->words -= start;
|
|
br->consumed_words = 0;
|
|
}
|
|
|
|
/*
|
|
* set the target for reading, taking into account word alignment and endianness
|
|
*/
|
|
bytes = (br->capacity - br->words) * FLAC__BYTES_PER_WORD - br->bytes;
|
|
if(bytes == 0)
|
|
return false; /* no space left, buffer is too small; see note for FLAC__BITREADER_DEFAULT_CAPACITY */
|
|
target = ((FLAC__byte*)(br->buffer+br->words)) + br->bytes;
|
|
|
|
/* before reading, if the existing reader looks like this (say brword is 32 bits wide)
|
|
* bitstream : 11 22 33 44 55 br->words=1 br->bytes=1 (partial tail word is left-justified)
|
|
* buffer[BE]: 11 22 33 44 55 ?? ?? ?? (shown layed out as bytes sequentially in memory)
|
|
* buffer[LE]: 44 33 22 11 ?? ?? ?? 55 (?? being don't-care)
|
|
* ^^-------target, bytes=3
|
|
* on LE machines, have to byteswap the odd tail word so nothing is
|
|
* overwritten:
|
|
*/
|
|
#if WORDS_BIGENDIAN
|
|
#else
|
|
if(br->bytes)
|
|
br->buffer[br->words] = SWAP_BE_WORD_TO_HOST(br->buffer[br->words]);
|
|
#endif
|
|
|
|
/* now it looks like:
|
|
* bitstream : 11 22 33 44 55 br->words=1 br->bytes=1
|
|
* buffer[BE]: 11 22 33 44 55 ?? ?? ??
|
|
* buffer[LE]: 44 33 22 11 55 ?? ?? ??
|
|
* ^^-------target, bytes=3
|
|
*/
|
|
|
|
/* read in the data; note that the callback may return a smaller number of bytes */
|
|
if(!br->read_callback(target, &bytes, br->client_data))
|
|
return false;
|
|
|
|
/* after reading bytes 66 77 88 99 AA BB CC DD EE FF from the client:
|
|
* bitstream : 11 22 33 44 55 66 77 88 99 AA BB CC DD EE FF
|
|
* buffer[BE]: 11 22 33 44 55 66 77 88 99 AA BB CC DD EE FF ??
|
|
* buffer[LE]: 44 33 22 11 55 66 77 88 99 AA BB CC DD EE FF ??
|
|
* now have to byteswap on LE machines:
|
|
*/
|
|
#if WORDS_BIGENDIAN
|
|
#else
|
|
end = (br->words*FLAC__BYTES_PER_WORD + br->bytes + bytes + (FLAC__BYTES_PER_WORD-1)) / FLAC__BYTES_PER_WORD;
|
|
# if defined(_MSC_VER) && (FLAC__BYTES_PER_WORD == 4)
|
|
if(br->cpu_info.type == FLAC__CPUINFO_TYPE_IA32 && br->cpu_info.data.ia32.bswap) {
|
|
start = br->words;
|
|
local_swap32_block_(br->buffer + start, end - start);
|
|
}
|
|
else
|
|
# endif
|
|
for(start = br->words; start < end; start++)
|
|
br->buffer[start] = SWAP_BE_WORD_TO_HOST(br->buffer[start]);
|
|
#endif
|
|
|
|
/* now it looks like:
|
|
* bitstream : 11 22 33 44 55 66 77 88 99 AA BB CC DD EE FF
|
|
* buffer[BE]: 11 22 33 44 55 66 77 88 99 AA BB CC DD EE FF ??
|
|
* buffer[LE]: 44 33 22 11 88 77 66 55 CC BB AA 99 ?? FF EE DD
|
|
* finally we'll update the reader values:
|
|
*/
|
|
end = br->words*FLAC__BYTES_PER_WORD + br->bytes + bytes;
|
|
br->words = end / FLAC__BYTES_PER_WORD;
|
|
br->bytes = end % FLAC__BYTES_PER_WORD;
|
|
|
|
return true;
|
|
}
|
|
|
|
/***********************************************************************
|
|
*
|
|
* Class constructor/destructor
|
|
*
|
|
***********************************************************************/
|
|
|
|
FLAC__BitReader *FLAC__bitreader_new(void)
|
|
{
|
|
FLAC__BitReader *br = (FLAC__BitReader*)calloc(1, sizeof(FLAC__BitReader));
|
|
|
|
/* calloc() implies:
|
|
memset(br, 0, sizeof(FLAC__BitReader));
|
|
br->buffer = 0;
|
|
br->capacity = 0;
|
|
br->words = br->bytes = 0;
|
|
br->consumed_words = br->consumed_bits = 0;
|
|
br->read_callback = 0;
|
|
br->client_data = 0;
|
|
*/
|
|
return br;
|
|
}
|
|
|
|
void FLAC__bitreader_delete(FLAC__BitReader *br)
|
|
{
|
|
FLAC__ASSERT(0 != br);
|
|
|
|
FLAC__bitreader_free(br);
|
|
free(br);
|
|
}
|
|
|
|
/***********************************************************************
|
|
*
|
|
* Public class methods
|
|
*
|
|
***********************************************************************/
|
|
|
|
FLAC__bool FLAC__bitreader_init(FLAC__BitReader *br, FLAC__CPUInfo cpu, FLAC__BitReaderReadCallback rcb, void *cd)
|
|
{
|
|
FLAC__ASSERT(0 != br);
|
|
|
|
br->words = br->bytes = 0;
|
|
br->consumed_words = br->consumed_bits = 0;
|
|
br->capacity = FLAC__BITREADER_DEFAULT_CAPACITY;
|
|
br->buffer = (brword*)malloc(sizeof(brword) * br->capacity);
|
|
if(br->buffer == 0)
|
|
return false;
|
|
br->read_callback = rcb;
|
|
br->client_data = cd;
|
|
br->cpu_info = cpu;
|
|
|
|
return true;
|
|
}
|
|
|
|
void FLAC__bitreader_free(FLAC__BitReader *br)
|
|
{
|
|
FLAC__ASSERT(0 != br);
|
|
|
|
if(0 != br->buffer)
|
|
free(br->buffer);
|
|
br->buffer = 0;
|
|
br->capacity = 0;
|
|
br->words = br->bytes = 0;
|
|
br->consumed_words = br->consumed_bits = 0;
|
|
br->read_callback = 0;
|
|
br->client_data = 0;
|
|
}
|
|
|
|
FLAC__bool FLAC__bitreader_clear(FLAC__BitReader *br)
|
|
{
|
|
br->words = br->bytes = 0;
|
|
br->consumed_words = br->consumed_bits = 0;
|
|
return true;
|
|
}
|
|
|
|
void FLAC__bitreader_dump(const FLAC__BitReader *br, FILE *out)
|
|
{
|
|
unsigned i, j;
|
|
if(br == 0) {
|
|
fprintf(out, "bitreader is NULL\n");
|
|
}
|
|
else {
|
|
fprintf(out, "bitreader: capacity=%u words=%u bytes=%u consumed: words=%u, bits=%u\n", br->capacity, br->words, br->bytes, br->consumed_words, br->consumed_bits);
|
|
|
|
for(i = 0; i < br->words; i++) {
|
|
fprintf(out, "%08X: ", i);
|
|
for(j = 0; j < FLAC__BITS_PER_WORD; j++)
|
|
if(i < br->consumed_words || (i == br->consumed_words && j < br->consumed_bits))
|
|
fprintf(out, ".");
|
|
else
|
|
fprintf(out, "%01u", br->buffer[i] & (1 << (FLAC__BITS_PER_WORD-j-1)) ? 1:0);
|
|
fprintf(out, "\n");
|
|
}
|
|
if(br->bytes > 0) {
|
|
fprintf(out, "%08X: ", i);
|
|
for(j = 0; j < br->bytes*8; j++)
|
|
if(i < br->consumed_words || (i == br->consumed_words && j < br->consumed_bits))
|
|
fprintf(out, ".");
|
|
else
|
|
fprintf(out, "%01u", br->buffer[i] & (1 << (br->bytes*8-j-1)) ? 1:0);
|
|
fprintf(out, "\n");
|
|
}
|
|
}
|
|
}
|
|
|
|
void FLAC__bitreader_reset_read_crc16(FLAC__BitReader *br, FLAC__uint16 seed)
|
|
{
|
|
FLAC__ASSERT(0 != br);
|
|
FLAC__ASSERT(0 != br->buffer);
|
|
FLAC__ASSERT((br->consumed_bits & 7) == 0);
|
|
|
|
br->read_crc16 = (unsigned)seed;
|
|
br->crc16_align = br->consumed_bits;
|
|
}
|
|
|
|
FLAC__uint16 FLAC__bitreader_get_read_crc16(FLAC__BitReader *br)
|
|
{
|
|
FLAC__ASSERT(0 != br);
|
|
FLAC__ASSERT(0 != br->buffer);
|
|
FLAC__ASSERT((br->consumed_bits & 7) == 0);
|
|
FLAC__ASSERT(br->crc16_align <= br->consumed_bits);
|
|
|
|
/* CRC any tail bytes in a partially-consumed word */
|
|
if(br->consumed_bits) {
|
|
const brword tail = br->buffer[br->consumed_words];
|
|
for( ; br->crc16_align < br->consumed_bits; br->crc16_align += 8)
|
|
br->read_crc16 = FLAC__CRC16_UPDATE((unsigned)((tail >> (FLAC__BITS_PER_WORD-8-br->crc16_align)) & 0xff), br->read_crc16);
|
|
}
|
|
return br->read_crc16;
|
|
}
|
|
|
|
FLaC__INLINE FLAC__bool FLAC__bitreader_is_consumed_byte_aligned(const FLAC__BitReader *br)
|
|
{
|
|
return ((br->consumed_bits & 7) == 0);
|
|
}
|
|
|
|
FLaC__INLINE unsigned FLAC__bitreader_bits_left_for_byte_alignment(const FLAC__BitReader *br)
|
|
{
|
|
return 8 - (br->consumed_bits & 7);
|
|
}
|
|
|
|
FLaC__INLINE unsigned FLAC__bitreader_get_input_bits_unconsumed(const FLAC__BitReader *br)
|
|
{
|
|
return (br->words-br->consumed_words)*FLAC__BITS_PER_WORD + br->bytes*8 - br->consumed_bits;
|
|
}
|
|
|
|
FLaC__INLINE FLAC__bool FLAC__bitreader_read_raw_uint32(FLAC__BitReader *br, FLAC__uint32 *val, unsigned bits)
|
|
{
|
|
FLAC__ASSERT(0 != br);
|
|
FLAC__ASSERT(0 != br->buffer);
|
|
|
|
FLAC__ASSERT(bits <= 32);
|
|
FLAC__ASSERT((br->capacity*FLAC__BITS_PER_WORD) * 2 >= bits);
|
|
FLAC__ASSERT(br->consumed_words <= br->words);
|
|
|
|
/* WATCHOUT: code does not work with <32bit words; we can make things much faster with this assertion */
|
|
FLAC__ASSERT(FLAC__BITS_PER_WORD >= 32);
|
|
|
|
if(bits == 0) { /* OPT: investigate if this can ever happen, maybe change to assertion */
|
|
*val = 0;
|
|
return true;
|
|
}
|
|
|
|
while((br->words-br->consumed_words)*FLAC__BITS_PER_WORD + br->bytes*8 - br->consumed_bits < bits) {
|
|
if(!bitreader_read_from_client_(br))
|
|
return false;
|
|
}
|
|
if(br->consumed_words < br->words) { /* if we've not consumed up to a partial tail word... */
|
|
/* OPT: taking out the consumed_bits==0 "else" case below might make things faster if less code allows the compiler to inline this function */
|
|
if(br->consumed_bits) {
|
|
/* this also works when consumed_bits==0, it's just a little slower than necessary for that case */
|
|
const unsigned n = FLAC__BITS_PER_WORD - br->consumed_bits;
|
|
const brword word = br->buffer[br->consumed_words];
|
|
if(bits < n) {
|
|
*val = (word & (FLAC__WORD_ALL_ONES >> br->consumed_bits)) >> (n-bits);
|
|
br->consumed_bits += bits;
|
|
return true;
|
|
}
|
|
*val = word & (FLAC__WORD_ALL_ONES >> br->consumed_bits);
|
|
bits -= n;
|
|
crc16_update_word_(br, word);
|
|
br->consumed_words++;
|
|
br->consumed_bits = 0;
|
|
if(bits) { /* if there are still bits left to read, there have to be less than 32 so they will all be in the next word */
|
|
*val <<= bits;
|
|
*val |= (br->buffer[br->consumed_words] >> (FLAC__BITS_PER_WORD-bits));
|
|
br->consumed_bits = bits;
|
|
}
|
|
return true;
|
|
}
|
|
else {
|
|
const brword word = br->buffer[br->consumed_words];
|
|
if(bits < FLAC__BITS_PER_WORD) {
|
|
*val = word >> (FLAC__BITS_PER_WORD-bits);
|
|
br->consumed_bits = bits;
|
|
return true;
|
|
}
|
|
/* at this point 'bits' must be == FLAC__BITS_PER_WORD; because of previous assertions, it can't be larger */
|
|
*val = word;
|
|
crc16_update_word_(br, word);
|
|
br->consumed_words++;
|
|
return true;
|
|
}
|
|
}
|
|
else {
|
|
/* in this case we're starting our read at a partial tail word;
|
|
* the reader has guaranteed that we have at least 'bits' bits
|
|
* available to read, which makes this case simpler.
|
|
*/
|
|
/* OPT: taking out the consumed_bits==0 "else" case below might make things faster if less code allows the compiler to inline this function */
|
|
if(br->consumed_bits) {
|
|
/* this also works when consumed_bits==0, it's just a little slower than necessary for that case */
|
|
FLAC__ASSERT(br->consumed_bits + bits <= br->bytes*8);
|
|
*val = (br->buffer[br->consumed_words] & (FLAC__WORD_ALL_ONES >> br->consumed_bits)) >> (FLAC__BITS_PER_WORD-br->consumed_bits-bits);
|
|
br->consumed_bits += bits;
|
|
return true;
|
|
}
|
|
else {
|
|
*val = br->buffer[br->consumed_words] >> (FLAC__BITS_PER_WORD-bits);
|
|
br->consumed_bits += bits;
|
|
return true;
|
|
}
|
|
}
|
|
}
|
|
|
|
FLAC__bool FLAC__bitreader_read_raw_int32(FLAC__BitReader *br, FLAC__int32 *val, unsigned bits)
|
|
{
|
|
/* OPT: inline raw uint32 code here, or make into a macro if possible in the .h file */
|
|
if(!FLAC__bitreader_read_raw_uint32(br, (FLAC__uint32*)val, bits))
|
|
return false;
|
|
/* sign-extend: */
|
|
*val <<= (32-bits);
|
|
*val >>= (32-bits);
|
|
return true;
|
|
}
|
|
|
|
FLAC__bool FLAC__bitreader_read_raw_uint64(FLAC__BitReader *br, FLAC__uint64 *val, unsigned bits)
|
|
{
|
|
FLAC__uint32 hi, lo;
|
|
|
|
if(bits > 32) {
|
|
if(!FLAC__bitreader_read_raw_uint32(br, &hi, bits-32))
|
|
return false;
|
|
if(!FLAC__bitreader_read_raw_uint32(br, &lo, 32))
|
|
return false;
|
|
*val = hi;
|
|
*val <<= 32;
|
|
*val |= lo;
|
|
}
|
|
else {
|
|
if(!FLAC__bitreader_read_raw_uint32(br, &lo, bits))
|
|
return false;
|
|
*val = lo;
|
|
}
|
|
return true;
|
|
}
|
|
|
|
FLaC__INLINE FLAC__bool FLAC__bitreader_read_uint32_little_endian(FLAC__BitReader *br, FLAC__uint32 *val)
|
|
{
|
|
FLAC__uint32 x8, x32 = 0;
|
|
|
|
/* this doesn't need to be that fast as currently it is only used for vorbis comments */
|
|
|
|
if(!FLAC__bitreader_read_raw_uint32(br, &x32, 8))
|
|
return false;
|
|
|
|
if(!FLAC__bitreader_read_raw_uint32(br, &x8, 8))
|
|
return false;
|
|
x32 |= (x8 << 8);
|
|
|
|
if(!FLAC__bitreader_read_raw_uint32(br, &x8, 8))
|
|
return false;
|
|
x32 |= (x8 << 16);
|
|
|
|
if(!FLAC__bitreader_read_raw_uint32(br, &x8, 8))
|
|
return false;
|
|
x32 |= (x8 << 24);
|
|
|
|
*val = x32;
|
|
return true;
|
|
}
|
|
|
|
FLAC__bool FLAC__bitreader_skip_bits_no_crc(FLAC__BitReader *br, unsigned bits)
|
|
{
|
|
/*
|
|
* OPT: a faster implementation is possible but probably not that useful
|
|
* since this is only called a couple of times in the metadata readers.
|
|
*/
|
|
FLAC__ASSERT(0 != br);
|
|
FLAC__ASSERT(0 != br->buffer);
|
|
|
|
if(bits > 0) {
|
|
const unsigned n = br->consumed_bits & 7;
|
|
unsigned m;
|
|
FLAC__uint32 x;
|
|
|
|
if(n != 0) {
|
|
m = min(8-n, bits);
|
|
if(!FLAC__bitreader_read_raw_uint32(br, &x, m))
|
|
return false;
|
|
bits -= m;
|
|
}
|
|
m = bits / 8;
|
|
if(m > 0) {
|
|
if(!FLAC__bitreader_skip_byte_block_aligned_no_crc(br, m))
|
|
return false;
|
|
bits %= 8;
|
|
}
|
|
if(bits > 0) {
|
|
if(!FLAC__bitreader_read_raw_uint32(br, &x, bits))
|
|
return false;
|
|
}
|
|
}
|
|
|
|
return true;
|
|
}
|
|
|
|
FLAC__bool FLAC__bitreader_skip_byte_block_aligned_no_crc(FLAC__BitReader *br, unsigned nvals)
|
|
{
|
|
FLAC__uint32 x;
|
|
|
|
FLAC__ASSERT(0 != br);
|
|
FLAC__ASSERT(0 != br->buffer);
|
|
FLAC__ASSERT(FLAC__bitreader_is_consumed_byte_aligned(br));
|
|
|
|
/* step 1: skip over partial head word to get word aligned */
|
|
while(nvals && br->consumed_bits) { /* i.e. run until we read 'nvals' bytes or we hit the end of the head word */
|
|
if(!FLAC__bitreader_read_raw_uint32(br, &x, 8))
|
|
return false;
|
|
nvals--;
|
|
}
|
|
if(0 == nvals)
|
|
return true;
|
|
/* step 2: skip whole words in chunks */
|
|
while(nvals >= FLAC__BYTES_PER_WORD) {
|
|
if(br->consumed_words < br->words) {
|
|
br->consumed_words++;
|
|
nvals -= FLAC__BYTES_PER_WORD;
|
|
}
|
|
else if(!bitreader_read_from_client_(br))
|
|
return false;
|
|
}
|
|
/* step 3: skip any remainder from partial tail bytes */
|
|
while(nvals) {
|
|
if(!FLAC__bitreader_read_raw_uint32(br, &x, 8))
|
|
return false;
|
|
nvals--;
|
|
}
|
|
|
|
return true;
|
|
}
|
|
|
|
FLAC__bool FLAC__bitreader_read_byte_block_aligned_no_crc(FLAC__BitReader *br, FLAC__byte *val, unsigned nvals)
|
|
{
|
|
FLAC__uint32 x;
|
|
|
|
FLAC__ASSERT(0 != br);
|
|
FLAC__ASSERT(0 != br->buffer);
|
|
FLAC__ASSERT(FLAC__bitreader_is_consumed_byte_aligned(br));
|
|
|
|
/* step 1: read from partial head word to get word aligned */
|
|
while(nvals && br->consumed_bits) { /* i.e. run until we read 'nvals' bytes or we hit the end of the head word */
|
|
if(!FLAC__bitreader_read_raw_uint32(br, &x, 8))
|
|
return false;
|
|
*val++ = (FLAC__byte)x;
|
|
nvals--;
|
|
}
|
|
if(0 == nvals)
|
|
return true;
|
|
/* step 2: read whole words in chunks */
|
|
while(nvals >= FLAC__BYTES_PER_WORD) {
|
|
if(br->consumed_words < br->words) {
|
|
const brword word = br->buffer[br->consumed_words++];
|
|
#if FLAC__BYTES_PER_WORD == 4
|
|
val[0] = (FLAC__byte)(word >> 24);
|
|
val[1] = (FLAC__byte)(word >> 16);
|
|
val[2] = (FLAC__byte)(word >> 8);
|
|
val[3] = (FLAC__byte)word;
|
|
#elif FLAC__BYTES_PER_WORD == 8
|
|
val[0] = (FLAC__byte)(word >> 56);
|
|
val[1] = (FLAC__byte)(word >> 48);
|
|
val[2] = (FLAC__byte)(word >> 40);
|
|
val[3] = (FLAC__byte)(word >> 32);
|
|
val[4] = (FLAC__byte)(word >> 24);
|
|
val[5] = (FLAC__byte)(word >> 16);
|
|
val[6] = (FLAC__byte)(word >> 8);
|
|
val[7] = (FLAC__byte)word;
|
|
#else
|
|
for(x = 0; x < FLAC__BYTES_PER_WORD; x++)
|
|
val[x] = (FLAC__byte)(word >> (8*(FLAC__BYTES_PER_WORD-x-1)));
|
|
#endif
|
|
val += FLAC__BYTES_PER_WORD;
|
|
nvals -= FLAC__BYTES_PER_WORD;
|
|
}
|
|
else if(!bitreader_read_from_client_(br))
|
|
return false;
|
|
}
|
|
/* step 3: read any remainder from partial tail bytes */
|
|
while(nvals) {
|
|
if(!FLAC__bitreader_read_raw_uint32(br, &x, 8))
|
|
return false;
|
|
*val++ = (FLAC__byte)x;
|
|
nvals--;
|
|
}
|
|
|
|
return true;
|
|
}
|
|
|
|
FLaC__INLINE FLAC__bool FLAC__bitreader_read_unary_unsigned(FLAC__BitReader *br, unsigned *val)
|
|
#if 0 /* slow but readable version */
|
|
{
|
|
unsigned bit;
|
|
|
|
FLAC__ASSERT(0 != br);
|
|
FLAC__ASSERT(0 != br->buffer);
|
|
|
|
*val = 0;
|
|
while(1) {
|
|
if(!FLAC__bitreader_read_bit(br, &bit))
|
|
return false;
|
|
if(bit)
|
|
break;
|
|
else
|
|
*val++;
|
|
}
|
|
return true;
|
|
}
|
|
#else
|
|
{
|
|
unsigned i;
|
|
|
|
FLAC__ASSERT(0 != br);
|
|
FLAC__ASSERT(0 != br->buffer);
|
|
|
|
*val = 0;
|
|
while(1) {
|
|
while(br->consumed_words < br->words) { /* if we've not consumed up to a partial tail word... */
|
|
brword b = br->buffer[br->consumed_words] << br->consumed_bits;
|
|
if(b) {
|
|
i = COUNT_ZERO_MSBS(b);
|
|
*val += i;
|
|
i++;
|
|
br->consumed_bits += i;
|
|
if(br->consumed_bits >= FLAC__BITS_PER_WORD) { /* faster way of testing if(br->consumed_bits == FLAC__BITS_PER_WORD) */
|
|
crc16_update_word_(br, br->buffer[br->consumed_words]);
|
|
br->consumed_words++;
|
|
br->consumed_bits = 0;
|
|
}
|
|
return true;
|
|
}
|
|
else {
|
|
*val += FLAC__BITS_PER_WORD - br->consumed_bits;
|
|
crc16_update_word_(br, br->buffer[br->consumed_words]);
|
|
br->consumed_words++;
|
|
br->consumed_bits = 0;
|
|
/* didn't find stop bit yet, have to keep going... */
|
|
}
|
|
}
|
|
/* at this point we've eaten up all the whole words; have to try
|
|
* reading through any tail bytes before calling the read callback.
|
|
* this is a repeat of the above logic adjusted for the fact we
|
|
* don't have a whole word. note though if the client is feeding
|
|
* us data a byte at a time (unlikely), br->consumed_bits may not
|
|
* be zero.
|
|
*/
|
|
if(br->bytes) {
|
|
const unsigned end = br->bytes * 8;
|
|
brword b = (br->buffer[br->consumed_words] & (FLAC__WORD_ALL_ONES << (FLAC__BITS_PER_WORD-end))) << br->consumed_bits;
|
|
if(b) {
|
|
i = COUNT_ZERO_MSBS(b);
|
|
*val += i;
|
|
i++;
|
|
br->consumed_bits += i;
|
|
FLAC__ASSERT(br->consumed_bits < FLAC__BITS_PER_WORD);
|
|
return true;
|
|
}
|
|
else {
|
|
*val += end - br->consumed_bits;
|
|
br->consumed_bits += end;
|
|
FLAC__ASSERT(br->consumed_bits < FLAC__BITS_PER_WORD);
|
|
/* didn't find stop bit yet, have to keep going... */
|
|
}
|
|
}
|
|
if(!bitreader_read_from_client_(br))
|
|
return false;
|
|
}
|
|
}
|
|
#endif
|
|
|
|
FLAC__bool FLAC__bitreader_read_rice_signed(FLAC__BitReader *br, int *val, unsigned parameter)
|
|
{
|
|
FLAC__uint32 lsbs = 0, msbs = 0;
|
|
unsigned uval;
|
|
|
|
FLAC__ASSERT(0 != br);
|
|
FLAC__ASSERT(0 != br->buffer);
|
|
FLAC__ASSERT(parameter <= 31);
|
|
|
|
/* read the unary MSBs and end bit */
|
|
if(!FLAC__bitreader_read_unary_unsigned(br, &msbs))
|
|
return false;
|
|
|
|
/* read the binary LSBs */
|
|
if(!FLAC__bitreader_read_raw_uint32(br, &lsbs, parameter))
|
|
return false;
|
|
|
|
/* compose the value */
|
|
uval = (msbs << parameter) | lsbs;
|
|
if(uval & 1)
|
|
*val = -((int)(uval >> 1)) - 1;
|
|
else
|
|
*val = (int)(uval >> 1);
|
|
|
|
return true;
|
|
}
|
|
|
|
/* this is by far the most heavily used reader call. it ain't pretty but it's fast */
|
|
/* a lot of the logic is copied, then adapted, from FLAC__bitreader_read_unary_unsigned() and FLAC__bitreader_read_raw_uint32() */
|
|
FLAC__bool FLAC__bitreader_read_rice_signed_block(FLAC__BitReader *br, int vals[], unsigned nvals, unsigned parameter)
|
|
/* OPT: possibly faster version for use with MSVC */
|
|
#ifdef _MSC_VER
|
|
{
|
|
unsigned i;
|
|
unsigned uval = 0;
|
|
unsigned bits; /* the # of binary LSBs left to read to finish a rice codeword */
|
|
|
|
/* try and get br->consumed_words and br->consumed_bits into register;
|
|
* must remember to flush them back to *br before calling other
|
|
* bitwriter functions that use them, and before returning */
|
|
register unsigned cwords;
|
|
register unsigned cbits;
|
|
|
|
FLAC__ASSERT(0 != br);
|
|
FLAC__ASSERT(0 != br->buffer);
|
|
/* WATCHOUT: code does not work with <32bit words; we can make things much faster with this assertion */
|
|
FLAC__ASSERT(FLAC__BITS_PER_WORD >= 32);
|
|
FLAC__ASSERT(parameter < 32);
|
|
/* the above two asserts also guarantee that the binary part never straddles more that 2 words, so we don't have to loop to read it */
|
|
|
|
if(nvals == 0)
|
|
return true;
|
|
|
|
cbits = br->consumed_bits;
|
|
cwords = br->consumed_words;
|
|
|
|
while(1) {
|
|
|
|
/* read unary part */
|
|
while(1) {
|
|
while(cwords < br->words) { /* if we've not consumed up to a partial tail word... */
|
|
brword b = br->buffer[cwords] << cbits;
|
|
if(b) {
|
|
#if 0 /* slower, probably due to bad register allocation... */ && defined FLAC__CPU_IA32 && !defined FLAC__NO_ASM && FLAC__BITS_PER_WORD == 32
|
|
__asm {
|
|
bsr eax, b
|
|
not eax
|
|
and eax, 31
|
|
mov i, eax
|
|
}
|
|
#else
|
|
i = COUNT_ZERO_MSBS(b);
|
|
#endif
|
|
uval += i;
|
|
bits = parameter;
|
|
i++;
|
|
cbits += i;
|
|
if(cbits == FLAC__BITS_PER_WORD) {
|
|
crc16_update_word_(br, br->buffer[cwords]);
|
|
cwords++;
|
|
cbits = 0;
|
|
}
|
|
goto break1;
|
|
}
|
|
else {
|
|
uval += FLAC__BITS_PER_WORD - cbits;
|
|
crc16_update_word_(br, br->buffer[cwords]);
|
|
cwords++;
|
|
cbits = 0;
|
|
/* didn't find stop bit yet, have to keep going... */
|
|
}
|
|
}
|
|
/* at this point we've eaten up all the whole words; have to try
|
|
* reading through any tail bytes before calling the read callback.
|
|
* this is a repeat of the above logic adjusted for the fact we
|
|
* don't have a whole word. note though if the client is feeding
|
|
* us data a byte at a time (unlikely), br->consumed_bits may not
|
|
* be zero.
|
|
*/
|
|
if(br->bytes) {
|
|
const unsigned end = br->bytes * 8;
|
|
brword b = (br->buffer[cwords] & (FLAC__WORD_ALL_ONES << (FLAC__BITS_PER_WORD-end))) << cbits;
|
|
if(b) {
|
|
i = COUNT_ZERO_MSBS(b);
|
|
uval += i;
|
|
bits = parameter;
|
|
i++;
|
|
cbits += i;
|
|
FLAC__ASSERT(cbits < FLAC__BITS_PER_WORD);
|
|
goto break1;
|
|
}
|
|
else {
|
|
uval += end - cbits;
|
|
cbits += end;
|
|
FLAC__ASSERT(cbits < FLAC__BITS_PER_WORD);
|
|
/* didn't find stop bit yet, have to keep going... */
|
|
}
|
|
}
|
|
/* flush registers and read; bitreader_read_from_client_() does
|
|
* not touch br->consumed_bits at all but we still need to set
|
|
* it in case it fails and we have to return false.
|
|
*/
|
|
br->consumed_bits = cbits;
|
|
br->consumed_words = cwords;
|
|
if(!bitreader_read_from_client_(br))
|
|
return false;
|
|
cwords = br->consumed_words;
|
|
}
|
|
break1:
|
|
/* read binary part */
|
|
FLAC__ASSERT(cwords <= br->words);
|
|
|
|
if(bits) {
|
|
while((br->words-cwords)*FLAC__BITS_PER_WORD + br->bytes*8 - cbits < bits) {
|
|
/* flush registers and read; bitreader_read_from_client_() does
|
|
* not touch br->consumed_bits at all but we still need to set
|
|
* it in case it fails and we have to return false.
|
|
*/
|
|
br->consumed_bits = cbits;
|
|
br->consumed_words = cwords;
|
|
if(!bitreader_read_from_client_(br))
|
|
return false;
|
|
cwords = br->consumed_words;
|
|
}
|
|
if(cwords < br->words) { /* if we've not consumed up to a partial tail word... */
|
|
if(cbits) {
|
|
/* this also works when consumed_bits==0, it's just a little slower than necessary for that case */
|
|
const unsigned n = FLAC__BITS_PER_WORD - cbits;
|
|
const brword word = br->buffer[cwords];
|
|
if(bits < n) {
|
|
uval <<= bits;
|
|
uval |= (word & (FLAC__WORD_ALL_ONES >> cbits)) >> (n-bits);
|
|
cbits += bits;
|
|
goto break2;
|
|
}
|
|
uval <<= n;
|
|
uval |= word & (FLAC__WORD_ALL_ONES >> cbits);
|
|
bits -= n;
|
|
crc16_update_word_(br, word);
|
|
cwords++;
|
|
cbits = 0;
|
|
if(bits) { /* if there are still bits left to read, there have to be less than 32 so they will all be in the next word */
|
|
uval <<= bits;
|
|
uval |= (br->buffer[cwords] >> (FLAC__BITS_PER_WORD-bits));
|
|
cbits = bits;
|
|
}
|
|
goto break2;
|
|
}
|
|
else {
|
|
FLAC__ASSERT(bits < FLAC__BITS_PER_WORD);
|
|
uval <<= bits;
|
|
uval |= br->buffer[cwords] >> (FLAC__BITS_PER_WORD-bits);
|
|
cbits = bits;
|
|
goto break2;
|
|
}
|
|
}
|
|
else {
|
|
/* in this case we're starting our read at a partial tail word;
|
|
* the reader has guaranteed that we have at least 'bits' bits
|
|
* available to read, which makes this case simpler.
|
|
*/
|
|
uval <<= bits;
|
|
if(cbits) {
|
|
/* this also works when consumed_bits==0, it's just a little slower than necessary for that case */
|
|
FLAC__ASSERT(cbits + bits <= br->bytes*8);
|
|
uval |= (br->buffer[cwords] & (FLAC__WORD_ALL_ONES >> cbits)) >> (FLAC__BITS_PER_WORD-cbits-bits);
|
|
cbits += bits;
|
|
goto break2;
|
|
}
|
|
else {
|
|
uval |= br->buffer[cwords] >> (FLAC__BITS_PER_WORD-bits);
|
|
cbits += bits;
|
|
goto break2;
|
|
}
|
|
}
|
|
}
|
|
break2:
|
|
/* compose the value */
|
|
*vals = (int)(uval >> 1 ^ -(int)(uval & 1));
|
|
|
|
/* are we done? */
|
|
--nvals;
|
|
if(nvals == 0) {
|
|
br->consumed_bits = cbits;
|
|
br->consumed_words = cwords;
|
|
return true;
|
|
}
|
|
|
|
uval = 0;
|
|
++vals;
|
|
|
|
}
|
|
}
|
|
#else
|
|
{
|
|
unsigned i;
|
|
unsigned uval = 0;
|
|
|
|
/* try and get br->consumed_words and br->consumed_bits into register;
|
|
* must remember to flush them back to *br before calling other
|
|
* bitwriter functions that use them, and before returning */
|
|
register unsigned cwords;
|
|
register unsigned cbits;
|
|
unsigned ucbits; /* keep track of the number of unconsumed bits in the buffer */
|
|
|
|
FLAC__ASSERT(0 != br);
|
|
FLAC__ASSERT(0 != br->buffer);
|
|
/* WATCHOUT: code does not work with <32bit words; we can make things much faster with this assertion */
|
|
FLAC__ASSERT(FLAC__BITS_PER_WORD >= 32);
|
|
FLAC__ASSERT(parameter < 32);
|
|
/* the above two asserts also guarantee that the binary part never straddles more than 2 words, so we don't have to loop to read it */
|
|
|
|
if(nvals == 0)
|
|
return true;
|
|
|
|
cbits = br->consumed_bits;
|
|
cwords = br->consumed_words;
|
|
ucbits = (br->words-cwords)*FLAC__BITS_PER_WORD + br->bytes*8 - cbits;
|
|
|
|
while(1) {
|
|
|
|
/* read unary part */
|
|
while(1) {
|
|
while(cwords < br->words) { /* if we've not consumed up to a partial tail word... */
|
|
brword b = br->buffer[cwords] << cbits;
|
|
if(b) {
|
|
#if 0 /* is not discernably faster... */ && defined FLAC__CPU_IA32 && !defined FLAC__NO_ASM && FLAC__BITS_PER_WORD == 32 && defined __GNUC__
|
|
asm volatile (
|
|
"bsrl %1, %0;"
|
|
"notl %0;"
|
|
"andl $31, %0;"
|
|
: "=r"(i)
|
|
: "r"(b)
|
|
);
|
|
#else
|
|
i = COUNT_ZERO_MSBS(b);
|
|
#endif
|
|
uval += i;
|
|
cbits += i;
|
|
cbits++; /* skip over stop bit */
|
|
if(cbits >= FLAC__BITS_PER_WORD) { /* faster way of testing if(cbits == FLAC__BITS_PER_WORD) */
|
|
crc16_update_word_(br, br->buffer[cwords]);
|
|
cwords++;
|
|
cbits = 0;
|
|
}
|
|
goto break1;
|
|
}
|
|
else {
|
|
uval += FLAC__BITS_PER_WORD - cbits;
|
|
crc16_update_word_(br, br->buffer[cwords]);
|
|
cwords++;
|
|
cbits = 0;
|
|
/* didn't find stop bit yet, have to keep going... */
|
|
}
|
|
}
|
|
/* at this point we've eaten up all the whole words; have to try
|
|
* reading through any tail bytes before calling the read callback.
|
|
* this is a repeat of the above logic adjusted for the fact we
|
|
* don't have a whole word. note though if the client is feeding
|
|
* us data a byte at a time (unlikely), br->consumed_bits may not
|
|
* be zero.
|
|
*/
|
|
if(br->bytes) {
|
|
const unsigned end = br->bytes * 8;
|
|
brword b = (br->buffer[cwords] & ~(FLAC__WORD_ALL_ONES >> end)) << cbits;
|
|
if(b) {
|
|
i = COUNT_ZERO_MSBS(b);
|
|
uval += i;
|
|
cbits += i;
|
|
cbits++; /* skip over stop bit */
|
|
FLAC__ASSERT(cbits < FLAC__BITS_PER_WORD);
|
|
goto break1;
|
|
}
|
|
else {
|
|
uval += end - cbits;
|
|
cbits += end;
|
|
FLAC__ASSERT(cbits < FLAC__BITS_PER_WORD);
|
|
/* didn't find stop bit yet, have to keep going... */
|
|
}
|
|
}
|
|
/* flush registers and read; bitreader_read_from_client_() does
|
|
* not touch br->consumed_bits at all but we still need to set
|
|
* it in case it fails and we have to return false.
|
|
*/
|
|
br->consumed_bits = cbits;
|
|
br->consumed_words = cwords;
|
|
if(!bitreader_read_from_client_(br))
|
|
return false;
|
|
cwords = br->consumed_words;
|
|
ucbits = (br->words-cwords)*FLAC__BITS_PER_WORD + br->bytes*8 - cbits + uval;
|
|
/* + uval to offset our count by the # of unary bits already
|
|
* consumed before the read, because we will add these back
|
|
* in all at once at break1
|
|
*/
|
|
}
|
|
break1:
|
|
ucbits -= uval;
|
|
ucbits--; /* account for stop bit */
|
|
|
|
/* read binary part */
|
|
FLAC__ASSERT(cwords <= br->words);
|
|
|
|
if(parameter) {
|
|
while(ucbits < parameter) {
|
|
/* flush registers and read; bitreader_read_from_client_() does
|
|
* not touch br->consumed_bits at all but we still need to set
|
|
* it in case it fails and we have to return false.
|
|
*/
|
|
br->consumed_bits = cbits;
|
|
br->consumed_words = cwords;
|
|
if(!bitreader_read_from_client_(br))
|
|
return false;
|
|
cwords = br->consumed_words;
|
|
ucbits = (br->words-cwords)*FLAC__BITS_PER_WORD + br->bytes*8 - cbits;
|
|
}
|
|
if(cwords < br->words) { /* if we've not consumed up to a partial tail word... */
|
|
if(cbits) {
|
|
/* this also works when consumed_bits==0, it's just slower than necessary for that case */
|
|
const unsigned n = FLAC__BITS_PER_WORD - cbits;
|
|
const brword word = br->buffer[cwords];
|
|
if(parameter < n) {
|
|
uval <<= parameter;
|
|
uval |= (word & (FLAC__WORD_ALL_ONES >> cbits)) >> (n-parameter);
|
|
cbits += parameter;
|
|
}
|
|
else {
|
|
uval <<= n;
|
|
uval |= word & (FLAC__WORD_ALL_ONES >> cbits);
|
|
crc16_update_word_(br, word);
|
|
cwords++;
|
|
cbits = parameter - n;
|
|
if(cbits) { /* parameter > n, i.e. if there are still bits left to read, there have to be less than 32 so they will all be in the next word */
|
|
uval <<= cbits;
|
|
uval |= (br->buffer[cwords] >> (FLAC__BITS_PER_WORD-cbits));
|
|
}
|
|
}
|
|
}
|
|
else {
|
|
cbits = parameter;
|
|
uval <<= parameter;
|
|
uval |= br->buffer[cwords] >> (FLAC__BITS_PER_WORD-cbits);
|
|
}
|
|
}
|
|
else {
|
|
/* in this case we're starting our read at a partial tail word;
|
|
* the reader has guaranteed that we have at least 'parameter'
|
|
* bits available to read, which makes this case simpler.
|
|
*/
|
|
uval <<= parameter;
|
|
if(cbits) {
|
|
/* this also works when consumed_bits==0, it's just a little slower than necessary for that case */
|
|
FLAC__ASSERT(cbits + parameter <= br->bytes*8);
|
|
uval |= (br->buffer[cwords] & (FLAC__WORD_ALL_ONES >> cbits)) >> (FLAC__BITS_PER_WORD-cbits-parameter);
|
|
cbits += parameter;
|
|
}
|
|
else {
|
|
cbits = parameter;
|
|
uval |= br->buffer[cwords] >> (FLAC__BITS_PER_WORD-cbits);
|
|
}
|
|
}
|
|
}
|
|
|
|
ucbits -= parameter;
|
|
|
|
/* compose the value */
|
|
*vals = (int)(uval >> 1 ^ -(int)(uval & 1));
|
|
|
|
/* are we done? */
|
|
--nvals;
|
|
if(nvals == 0) {
|
|
br->consumed_bits = cbits;
|
|
br->consumed_words = cwords;
|
|
return true;
|
|
}
|
|
|
|
uval = 0;
|
|
++vals;
|
|
|
|
}
|
|
}
|
|
#endif
|
|
|
|
#if 0 /* UNUSED */
|
|
FLAC__bool FLAC__bitreader_read_golomb_signed(FLAC__BitReader *br, int *val, unsigned parameter)
|
|
{
|
|
FLAC__uint32 lsbs = 0, msbs = 0;
|
|
unsigned bit, uval, k;
|
|
|
|
FLAC__ASSERT(0 != br);
|
|
FLAC__ASSERT(0 != br->buffer);
|
|
|
|
k = FLAC__bitmath_ilog2(parameter);
|
|
|
|
/* read the unary MSBs and end bit */
|
|
if(!FLAC__bitreader_read_unary_unsigned(br, &msbs))
|
|
return false;
|
|
|
|
/* read the binary LSBs */
|
|
if(!FLAC__bitreader_read_raw_uint32(br, &lsbs, k))
|
|
return false;
|
|
|
|
if(parameter == 1u<<k) {
|
|
/* compose the value */
|
|
uval = (msbs << k) | lsbs;
|
|
}
|
|
else {
|
|
unsigned d = (1 << (k+1)) - parameter;
|
|
if(lsbs >= d) {
|
|
if(!FLAC__bitreader_read_bit(br, &bit))
|
|
return false;
|
|
lsbs <<= 1;
|
|
lsbs |= bit;
|
|
lsbs -= d;
|
|
}
|
|
/* compose the value */
|
|
uval = msbs * parameter + lsbs;
|
|
}
|
|
|
|
/* unfold unsigned to signed */
|
|
if(uval & 1)
|
|
*val = -((int)(uval >> 1)) - 1;
|
|
else
|
|
*val = (int)(uval >> 1);
|
|
|
|
return true;
|
|
}
|
|
|
|
FLAC__bool FLAC__bitreader_read_golomb_unsigned(FLAC__BitReader *br, unsigned *val, unsigned parameter)
|
|
{
|
|
FLAC__uint32 lsbs, msbs = 0;
|
|
unsigned bit, k;
|
|
|
|
FLAC__ASSERT(0 != br);
|
|
FLAC__ASSERT(0 != br->buffer);
|
|
|
|
k = FLAC__bitmath_ilog2(parameter);
|
|
|
|
/* read the unary MSBs and end bit */
|
|
if(!FLAC__bitreader_read_unary_unsigned(br, &msbs))
|
|
return false;
|
|
|
|
/* read the binary LSBs */
|
|
if(!FLAC__bitreader_read_raw_uint32(br, &lsbs, k))
|
|
return false;
|
|
|
|
if(parameter == 1u<<k) {
|
|
/* compose the value */
|
|
*val = (msbs << k) | lsbs;
|
|
}
|
|
else {
|
|
unsigned d = (1 << (k+1)) - parameter;
|
|
if(lsbs >= d) {
|
|
if(!FLAC__bitreader_read_bit(br, &bit))
|
|
return false;
|
|
lsbs <<= 1;
|
|
lsbs |= bit;
|
|
lsbs -= d;
|
|
}
|
|
/* compose the value */
|
|
*val = msbs * parameter + lsbs;
|
|
}
|
|
|
|
return true;
|
|
}
|
|
#endif /* UNUSED */
|
|
|
|
/* on return, if *val == 0xffffffff then the utf-8 sequence was invalid, but the return value will be true */
|
|
FLAC__bool FLAC__bitreader_read_utf8_uint32(FLAC__BitReader *br, FLAC__uint32 *val, FLAC__byte *raw, unsigned *rawlen)
|
|
{
|
|
FLAC__uint32 v = 0;
|
|
FLAC__uint32 x;
|
|
unsigned i;
|
|
|
|
if(!FLAC__bitreader_read_raw_uint32(br, &x, 8))
|
|
return false;
|
|
if(raw)
|
|
raw[(*rawlen)++] = (FLAC__byte)x;
|
|
if(!(x & 0x80)) { /* 0xxxxxxx */
|
|
v = x;
|
|
i = 0;
|
|
}
|
|
else if(x & 0xC0 && !(x & 0x20)) { /* 110xxxxx */
|
|
v = x & 0x1F;
|
|
i = 1;
|
|
}
|
|
else if(x & 0xE0 && !(x & 0x10)) { /* 1110xxxx */
|
|
v = x & 0x0F;
|
|
i = 2;
|
|
}
|
|
else if(x & 0xF0 && !(x & 0x08)) { /* 11110xxx */
|
|
v = x & 0x07;
|
|
i = 3;
|
|
}
|
|
else if(x & 0xF8 && !(x & 0x04)) { /* 111110xx */
|
|
v = x & 0x03;
|
|
i = 4;
|
|
}
|
|
else if(x & 0xFC && !(x & 0x02)) { /* 1111110x */
|
|
v = x & 0x01;
|
|
i = 5;
|
|
}
|
|
else {
|
|
*val = 0xffffffff;
|
|
return true;
|
|
}
|
|
for( ; i; i--) {
|
|
if(!FLAC__bitreader_read_raw_uint32(br, &x, 8))
|
|
return false;
|
|
if(raw)
|
|
raw[(*rawlen)++] = (FLAC__byte)x;
|
|
if(!(x & 0x80) || (x & 0x40)) { /* 10xxxxxx */
|
|
*val = 0xffffffff;
|
|
return true;
|
|
}
|
|
v <<= 6;
|
|
v |= (x & 0x3F);
|
|
}
|
|
*val = v;
|
|
return true;
|
|
}
|
|
|
|
/* on return, if *val == 0xffffffffffffffff then the utf-8 sequence was invalid, but the return value will be true */
|
|
FLAC__bool FLAC__bitreader_read_utf8_uint64(FLAC__BitReader *br, FLAC__uint64 *val, FLAC__byte *raw, unsigned *rawlen)
|
|
{
|
|
FLAC__uint64 v = 0;
|
|
FLAC__uint32 x;
|
|
unsigned i;
|
|
|
|
if(!FLAC__bitreader_read_raw_uint32(br, &x, 8))
|
|
return false;
|
|
if(raw)
|
|
raw[(*rawlen)++] = (FLAC__byte)x;
|
|
if(!(x & 0x80)) { /* 0xxxxxxx */
|
|
v = x;
|
|
i = 0;
|
|
}
|
|
else if(x & 0xC0 && !(x & 0x20)) { /* 110xxxxx */
|
|
v = x & 0x1F;
|
|
i = 1;
|
|
}
|
|
else if(x & 0xE0 && !(x & 0x10)) { /* 1110xxxx */
|
|
v = x & 0x0F;
|
|
i = 2;
|
|
}
|
|
else if(x & 0xF0 && !(x & 0x08)) { /* 11110xxx */
|
|
v = x & 0x07;
|
|
i = 3;
|
|
}
|
|
else if(x & 0xF8 && !(x & 0x04)) { /* 111110xx */
|
|
v = x & 0x03;
|
|
i = 4;
|
|
}
|
|
else if(x & 0xFC && !(x & 0x02)) { /* 1111110x */
|
|
v = x & 0x01;
|
|
i = 5;
|
|
}
|
|
else if(x & 0xFE && !(x & 0x01)) { /* 11111110 */
|
|
v = 0;
|
|
i = 6;
|
|
}
|
|
else {
|
|
*val = FLAC__U64L(0xffffffffffffffff);
|
|
return true;
|
|
}
|
|
for( ; i; i--) {
|
|
if(!FLAC__bitreader_read_raw_uint32(br, &x, 8))
|
|
return false;
|
|
if(raw)
|
|
raw[(*rawlen)++] = (FLAC__byte)x;
|
|
if(!(x & 0x80) || (x & 0x40)) { /* 10xxxxxx */
|
|
*val = FLAC__U64L(0xffffffffffffffff);
|
|
return true;
|
|
}
|
|
v <<= 6;
|
|
v |= (x & 0x3F);
|
|
}
|
|
*val = v;
|
|
return true;
|
|
}
|