brotli/encode.go

package brotli

import "io"

/* Copyright 2016 Google Inc. All Rights Reserved.

   Distributed under MIT license.
   See file LICENSE for detail or copy at https://opensource.org/licenses/MIT
*/

/* Constants and formulas that affect speed-ratio trade-offs and thus define
   quality levels. */
/* Copyright 2013 Google Inc. All Rights Reserved.

   Distributed under MIT license.
   See file LICENSE for detail or copy at https://opensource.org/licenses/MIT
*/

/**
 * @file
 * API for Brotli compression.
 */

/** Minimal value for ::BROTLI_PARAM_LGWIN parameter. */
const BROTLI_MIN_WINDOW_BITS = 10

/**
 * Maximal value for ::BROTLI_PARAM_LGWIN parameter.
 *
 * @note equal to @c BROTLI_MAX_DISTANCE_BITS constant.
 */
const BROTLI_MAX_WINDOW_BITS = 24

/**
 * Maximal value for ::BROTLI_PARAM_LGWIN parameter
 * in "Large Window Brotli" (32-bit).
 */
const BROTLI_LARGE_MAX_WINDOW_BITS = 30

/** Minimal value for ::BROTLI_PARAM_LGBLOCK parameter. */
const BROTLI_MIN_INPUT_BLOCK_BITS = 16

/** Maximal value for ::BROTLI_PARAM_LGBLOCK parameter. */
const BROTLI_MAX_INPUT_BLOCK_BITS = 24

/** Minimal value for ::BROTLI_PARAM_QUALITY parameter. */
const BROTLI_MIN_QUALITY = 0

/** Maximal value for ::BROTLI_PARAM_QUALITY parameter. */
const BROTLI_MAX_QUALITY = 11

/** Options for ::BROTLI_PARAM_MODE parameter. */
const (
	BROTLI_MODE_GENERIC = 0
	BROTLI_MODE_TEXT    = 1
	BROTLI_MODE_FONT    = 2
)

/** Default value for ::BROTLI_PARAM_QUALITY parameter. */
const BROTLI_DEFAULT_QUALITY = 11

/** Default value for ::BROTLI_PARAM_LGWIN parameter. */
const BROTLI_DEFAULT_WINDOW = 22

/** Default value for ::BROTLI_PARAM_MODE parameter. */
const BROTLI_DEFAULT_MODE = BROTLI_MODE_GENERIC

/** Operations that can be performed by streaming encoder. */
const (
	BROTLI_OPERATION_PROCESS       = 0
	BROTLI_OPERATION_FLUSH         = 1
	BROTLI_OPERATION_FINISH        = 2
	BROTLI_OPERATION_EMIT_METADATA = 3
)

/** Options to be used with ::BrotliEncoderSetParameter. */
const (
	BROTLI_PARAM_MODE                             = 0
	BROTLI_PARAM_QUALITY                          = 1
	BROTLI_PARAM_LGWIN                            = 2
	BROTLI_PARAM_LGBLOCK                          = 3
	BROTLI_PARAM_DISABLE_LITERAL_CONTEXT_MODELING = 4
	BROTLI_PARAM_SIZE_HINT                        = 5
	BROTLI_PARAM_LARGE_WINDOW                     = 6
	BROTLI_PARAM_NPOSTFIX                         = 7
	BROTLI_PARAM_NDIRECT                          = 8
)

const (
	BROTLI_STREAM_PROCESSING      = 0
	BROTLI_STREAM_FLUSH_REQUESTED = 1
	BROTLI_STREAM_FINISHED        = 2
	BROTLI_STREAM_METADATA_HEAD   = 3
	BROTLI_STREAM_METADATA_BODY   = 4
)

type Writer struct {
	dst io.Writer

	params              BrotliEncoderParams
	hasher_             HasherHandle
	input_pos_          uint64
	ringbuffer_         RingBuffer
	cmd_alloc_size_     uint
	commands_           []Command
	num_commands_       uint
	num_literals_       uint
	last_insert_len_    uint
	last_flush_pos_     uint64
	last_processed_pos_ uint64
	dist_cache_         [BROTLI_NUM_DISTANCE_SHORT_CODES]int
	saved_dist_cache_   [4]int
	last_bytes_         uint16
	last_bytes_bits_    byte
	prev_byte_          byte
	prev_byte2_         byte
	storage_size_       uint
	storage_            []byte
	small_table_        [1 << 10]int
	large_table_        []int
	large_table_size_   uint
	cmd_depths_         [128]byte
	cmd_bits_           [128]uint16
	cmd_code_           [512]byte
	cmd_code_numbits_   uint
	command_buf_        []uint32
	literal_buf_        []byte
	next_out_           []byte
	available_out_      uint
	total_out_          uint
	tiny_buf_           struct {
		u64 [2]uint64
		u8  [16]byte
	}
	remaining_metadata_bytes_ uint32
	stream_state_             int
	is_last_block_emitted_    bool
	is_initialized_           bool
}

func InputBlockSize(s *Writer) uint {
	return uint(1) << uint(s.params.lgblock)
}

func UnprocessedInputSize(s *Writer) uint64 {
	return s.input_pos_ - s.last_processed_pos_
}

func RemainingInputBlockSize(s *Writer) uint {
	var delta uint64 = UnprocessedInputSize(s)
	var block_size uint = InputBlockSize(s)
	if delta >= uint64(block_size) {
		return 0
	}
	return block_size - uint(delta)
}

func BrotliEncoderSetParameter(state *Writer, p int, value uint32) bool {
	/* Changing parameters on the fly is not implemented yet. */
	if state.is_initialized_ {
		return false
	}

	/* TODO: Validate/clamp parameters here. */
	switch p {
	case BROTLI_PARAM_MODE:
		state.params.mode = int(value)
		return true

	case BROTLI_PARAM_QUALITY:
		state.params.quality = int(value)
		return true

	case BROTLI_PARAM_LGWIN:
		state.params.lgwin = uint(int(value))
		return true

	case BROTLI_PARAM_LGBLOCK:
		state.params.lgblock = int(value)
		return true

	case BROTLI_PARAM_DISABLE_LITERAL_CONTEXT_MODELING:
		if (value != 0) && (value != 1) {
			return false
		}
		state.params.disable_literal_context_modeling = (!(value == 0))
		return true

	case BROTLI_PARAM_SIZE_HINT:
		state.params.size_hint = uint(value)
		return true

	case BROTLI_PARAM_LARGE_WINDOW:
		state.params.large_window = (!(value == 0))
		return true

	case BROTLI_PARAM_NPOSTFIX:
		state.params.dist.distance_postfix_bits = value
		return true

	case BROTLI_PARAM_NDIRECT:
		state.params.dist.num_direct_distance_codes = value
		return true

	default:
		return false
	}
}

/* Wraps 64-bit input position to 32-bit ring-buffer position preserving
   "not-a-first-lap" feature. */
func WrapPosition(position uint64) uint32 {
	var result uint32 = uint32(position)
	var gb uint64 = position >> 30
	if gb > 2 {
		/* Wrap every 2GiB; The first 3GB are continuous. */
		result = result&((1<<30)-1) | (uint32((gb-1)&1)+1)<<30
	}

	return result
}

func GetBrotliStorage(s *Writer, size uint) []byte {
	if s.storage_size_ < size {
		s.storage_ = nil
		s.storage_ = make([]byte, size)
		s.storage_size_ = size
	}

	return s.storage_
}

func HashTableSize(max_table_size uint, input_size uint) uint {
	var htsize uint = 256
	for htsize < max_table_size && htsize < input_size {
		htsize <<= 1
	}

	return htsize
}

func GetHashTable(s *Writer, quality int, input_size uint, table_size *uint) []int {
	var max_table_size uint = MaxHashTableSize(quality)
	var htsize uint = HashTableSize(max_table_size, input_size)
	/* Use smaller hash table when input.size() is smaller, since we
	   fill the table, incurring O(hash table size) overhead for
	   compression, and if the input is short, we won't need that
	   many hash table entries anyway. */

	var table []int
	assert(max_table_size >= 256)
	if quality == FAST_ONE_PASS_COMPRESSION_QUALITY {
		/* Only odd shifts are supported by fast-one-pass. */
		if htsize&0xAAAAA == 0 {
			htsize <<= 1
		}
	}

	if htsize <= uint(len(s.small_table_)) {
		table = s.small_table_[:]
	} else {
		if htsize > s.large_table_size_ {
			s.large_table_size_ = htsize
			s.large_table_ = nil
			s.large_table_ = make([]int, htsize)
		}

		table = s.large_table_
	}

	*table_size = htsize
	for i := 0; i < int(htsize); i++ {
		table[i] = 0
	}
	return table
}

func EncodeWindowBits(lgwin int, large_window bool, last_bytes *uint16, last_bytes_bits *byte) {
	if large_window {
		*last_bytes = uint16((lgwin&0x3F)<<8 | 0x11)
		*last_bytes_bits = 14
	} else {
		if lgwin == 16 {
			*last_bytes = 0
			*last_bytes_bits = 1
		} else if lgwin == 17 {
			*last_bytes = 1
			*last_bytes_bits = 7
		} else if lgwin > 17 {
			*last_bytes = uint16((lgwin-17)<<1 | 0x01)
			*last_bytes_bits = 4
		} else {
			*last_bytes = uint16((lgwin-8)<<4 | 0x01)
			*last_bytes_bits = 7
		}
	}
}

/* Initializes the command and distance prefix codes for the first block. */

var InitCommandPrefixCodes_kDefaultCommandDepths = [128]byte{
	0,
	4,
	4,
	5,
	6,
	6,
	7,
	7,
	7,
	7,
	7,
	8,
	8,
	8,
	8,
	8,
	0,
	0,
	0,
	4,
	4,
	4,
	4,
	4,
	5,
	5,
	6,
	6,
	6,
	6,
	7,
	7,
	7,
	7,
	10,
	10,
	10,
	10,
	10,
	10,
	0,
	4,
	4,
	5,
	5,
	5,
	6,
	6,
	7,
	8,
	8,
	9,
	10,
	10,
	10,
	10,
	10,
	10,
	10,
	10,
	10,
	10,
	10,
	10,
	5,
	0,
	0,
	0,
	0,
	0,
	0,
	0,
	0,
	0,
	0,
	0,
	0,
	0,
	0,
	0,
	6,
	6,
	6,
	6,
	6,
	6,
	5,
	5,
	5,
	5,
	5,
	5,
	4,
	4,
	4,
	4,
	4,
	4,
	4,
	5,
	5,
	5,
	5,
	5,
	5,
	6,
	6,
	7,
	7,
	7,
	8,
	10,
	12,
	12,
	12,
	12,
	12,
	12,
	12,
	12,
	12,
	12,
	12,
	12,
}
var InitCommandPrefixCodes_kDefaultCommandBits = [128]uint16{
	0,
	0,
	8,
	9,
	3,
	35,
	7,
	71,
	39,
	103,
	23,
	47,
	175,
	111,
	239,
	31,
	0,
	0,
	0,
	4,
	12,
	2,
	10,
	6,
	13,
	29,
	11,
	43,
	27,
	59,
	87,
	55,
	15,
	79,
	319,
	831,
	191,
	703,
	447,
	959,
	0,
	14,
	1,
	25,
	5,
	21,
	19,
	51,
	119,
	159,
	95,
	223,
	479,
	991,
	63,
	575,
	127,
	639,
	383,
	895,
	255,
	767,
	511,
	1023,
	14,
	0,
	0,
	0,
	0,
	0,
	0,
	0,
	0,
	0,
	0,
	0,
	0,
	0,
	0,
	0,
	27,
	59,
	7,
	39,
	23,
	55,
	30,
	1,
	17,
	9,
	25,
	5,
	0,
	8,
	4,
	12,
	2,
	10,
	6,
	21,
	13,
	29,
	3,
	19,
	11,
	15,
	47,
	31,
	95,
	63,
	127,
	255,
	767,
	2815,
	1791,
	3839,
	511,
	2559,
	1535,
	3583,
	1023,
	3071,
	2047,
	4095,
}
var InitCommandPrefixCodes_kDefaultCommandCode = []byte{
	0xff,
	0x77,
	0xd5,
	0xbf,
	0xe7,
	0xde,
	0xea,
	0x9e,
	0x51,
	0x5d,
	0xde,
	0xc6,
	0x70,
	0x57,
	0xbc,
	0x58,
	0x58,
	0x58,
	0xd8,
	0xd8,
	0x58,
	0xd5,
	0xcb,
	0x8c,
	0xea,
	0xe0,
	0xc3,
	0x87,
	0x1f,
	0x83,
	0xc1,
	0x60,
	0x1c,
	0x67,
	0xb2,
	0xaa,
	0x06,
	0x83,
	0xc1,
	0x60,
	0x30,
	0x18,
	0xcc,
	0xa1,
	0xce,
	0x88,
	0x54,
	0x94,
	0x46,
	0xe1,
	0xb0,
	0xd0,
	0x4e,
	0xb2,
	0xf7,
	0x04,
	0x00,
}
var InitCommandPrefixCodes_kDefaultCommandCodeNumBits uint = 448

func InitCommandPrefixCodes(cmd_depths []byte, cmd_bits []uint16, cmd_code []byte, cmd_code_numbits *uint) {
	copy(cmd_depths, InitCommandPrefixCodes_kDefaultCommandDepths[:])
	copy(cmd_bits, InitCommandPrefixCodes_kDefaultCommandBits[:])

	/* Initialize the pre-compressed form of the command and distance prefix
	   codes. */
	copy(cmd_code, InitCommandPrefixCodes_kDefaultCommandCode)

	*cmd_code_numbits = InitCommandPrefixCodes_kDefaultCommandCodeNumBits
}

/* Decide about the context map based on the ability of the prediction
   ability of the previous byte UTF8-prefix on the next byte. The
   prediction ability is calculated as Shannon entropy. Here we need
   Shannon entropy instead of 'BitsEntropy' since the prefix will be
   encoded with the remaining 6 bits of the following byte, and
   BitsEntropy will assume that symbol to be stored alone using Huffman
   coding. */

var ChooseContextMap_kStaticContextMapContinuation = [64]uint32{
	1,
	1,
	2,
	2,
	0,
	0,
	0,
	0,
	0,
	0,
	0,
	0,
	0,
	0,
	0,
	0,
	0,
	0,
	0,
	0,
	0,
	0,
	0,
	0,
	0,
	0,
	0,
	0,
	0,
	0,
	0,
	0,
	0,
	0,
	0,
	0,
	0,
	0,
	0,
	0,
	0,
	0,
	0,
	0,
	0,
	0,
	0,
	0,
	0,
	0,
	0,
	0,
	0,
	0,
	0,
	0,
	0,
	0,
	0,
	0,
	0,
	0,
	0,
	0,
}
var ChooseContextMap_kStaticContextMapSimpleUTF8 = [64]uint32{
	0,
	0,
	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,
}

func ChooseContextMap(quality int, bigram_histo []uint32, num_literal_contexts *uint, literal_context_map *[]uint32) {
	var monogram_histo = [3]uint32{0}
	var two_prefix_histo = [6]uint32{0}
	var total uint
	var i uint
	var dummy uint
	var entropy [4]float64
	for i = 0; i < 9; i++ {
		monogram_histo[i%3] += bigram_histo[i]
		two_prefix_histo[i%6] += bigram_histo[i]
	}

	entropy[1] = ShannonEntropy(monogram_histo[:], 3, &dummy)
	entropy[2] = (ShannonEntropy(two_prefix_histo[:], 3, &dummy) + ShannonEntropy(two_prefix_histo[3:], 3, &dummy))
	entropy[3] = 0
	for i = 0; i < 3; i++ {
		entropy[3] += ShannonEntropy(bigram_histo[3*i:], 3, &dummy)
	}

	total = uint(monogram_histo[0] + monogram_histo[1] + monogram_histo[2])
	assert(total != 0)
	entropy[0] = 1.0 / float64(total)
	entropy[1] *= entropy[0]
	entropy[2] *= entropy[0]
	entropy[3] *= entropy[0]

	if quality < MIN_QUALITY_FOR_HQ_CONTEXT_MODELING {
		/* 3 context models is a bit slower, don't use it at lower qualities. */
		entropy[3] = entropy[1] * 10
	}

	/* If expected savings by symbol are less than 0.2 bits, skip the
	   context modeling -- in exchange for faster decoding speed. */
	if entropy[1]-entropy[2] < 0.2 && entropy[1]-entropy[3] < 0.2 {
		*num_literal_contexts = 1
	} else if entropy[2]-entropy[3] < 0.02 {
		*num_literal_contexts = 2
		*literal_context_map = ChooseContextMap_kStaticContextMapSimpleUTF8[:]
	} else {
		*num_literal_contexts = 3
		*literal_context_map = ChooseContextMap_kStaticContextMapContinuation[:]
	}
}

/* Decide if we want to use a more complex static context map containing 13
   context values, based on the entropy reduction of histograms over the
   first 5 bits of literals. */

var ShouldUseComplexStaticContextMap_kStaticContextMapComplexUTF8 = [64]uint32{
	11,
	11,
	12,
	12,
	0,
	0,
	0,
	0,
	1,
	1,
	9,
	9,
	2,
	2,
	2,
	2,
	1,
	1,
	1,
	1,
	8,
	3,
	3,
	3,
	1,
	1,
	1,
	1,
	2,
	2,
	2,
	2,
	8,
	4,
	4,
	4,
	8,
	7,
	4,
	4,
	8,
	0,
	0,
	0,
	3,
	3,
	3,
	3,
	5,
	5,
	10,
	5,
	5,
	5,
	10,
	5,
	6,
	6,
	6,
	6,
	6,
	6,
	6,
	6,
}

func ShouldUseComplexStaticContextMap(input []byte, start_pos uint, length uint, mask uint, quality int, size_hint uint, num_literal_contexts *uint, literal_context_map *[]uint32) bool {
	/* Try the more complex static context map only for long data. */
	if size_hint < 1<<20 {
		return false
	} else {
		var end_pos uint = start_pos + length
		var combined_histo = [32]uint32{0}
		var context_histo = [13][32]uint32{[32]uint32{0}}
		var total uint32 = 0
		var entropy [3]float64
		var dummy uint
		var i uint
		var utf8_lut ContextLut = BROTLI_CONTEXT_LUT(CONTEXT_UTF8)
		/* To make entropy calculations faster and to fit on the stack, we collect
		   histograms over the 5 most significant bits of literals. One histogram
		   without context and 13 additional histograms for each context value. */
		for ; start_pos+64 <= end_pos; start_pos += 4096 {
			var stride_end_pos uint = start_pos + 64
			var prev2 byte = input[start_pos&mask]
			var prev1 byte = input[(start_pos+1)&mask]
			var pos uint

			/* To make the analysis of the data faster we only examine 64 byte long
			   strides at every 4kB intervals. */
			for pos = start_pos + 2; pos < stride_end_pos; pos++ {
				var literal byte = input[pos&mask]
				var context byte = byte(ShouldUseComplexStaticContextMap_kStaticContextMapComplexUTF8[BROTLI_CONTEXT(prev1, prev2, utf8_lut)])
				total++
				combined_histo[literal>>3]++
				context_histo[context][literal>>3]++
				prev2 = prev1
				prev1 = literal
			}
		}

		entropy[1] = ShannonEntropy(combined_histo[:], 32, &dummy)
		entropy[2] = 0
		for i = 0; i < 13; i++ {
			entropy[2] += ShannonEntropy(context_histo[i][0:], 32, &dummy)
		}

		entropy[0] = 1.0 / float64(total)
		entropy[1] *= entropy[0]
		entropy[2] *= entropy[0]

		/* The triggering heuristics below were tuned by compressing the individual
		   files of the silesia corpus. If we skip this kind of context modeling
		   for not very well compressible input (i.e. entropy using context modeling
		   is 60% of maximal entropy) or if expected savings by symbol are less
		   than 0.2 bits, then in every case when it triggers, the final compression
		   ratio is improved. Note however that this heuristics might be too strict
		   for some cases and could be tuned further. */
		if entropy[2] > 3.0 || entropy[1]-entropy[2] < 0.2 {
			return false
		} else {
			*num_literal_contexts = 13
			*literal_context_map = ShouldUseComplexStaticContextMap_kStaticContextMapComplexUTF8[:]
			return true
		}
	}
}

func DecideOverLiteralContextModeling(input []byte, start_pos uint, length uint, mask uint, quality int, size_hint uint, num_literal_contexts *uint, literal_context_map *[]uint32) {
	if quality < MIN_QUALITY_FOR_CONTEXT_MODELING || length < 64 {
		return
	} else if ShouldUseComplexStaticContextMap(input, start_pos, length, mask, quality, size_hint, num_literal_contexts, literal_context_map) {
	} else /* Context map was already set, nothing else to do. */
	{
		var end_pos uint = start_pos + length
		/* Gather bi-gram data of the UTF8 byte prefixes. To make the analysis of
		   UTF8 data faster we only examine 64 byte long strides at every 4kB
		   intervals. */

		var bigram_prefix_histo = [9]uint32{0}
		for ; start_pos+64 <= end_pos; start_pos += 4096 {
			var lut = [4]int{0, 0, 1, 2}
			var stride_end_pos uint = start_pos + 64
			var prev int = lut[input[start_pos&mask]>>6] * 3
			var pos uint
			for pos = start_pos + 1; pos < stride_end_pos; pos++ {
				var literal byte = input[pos&mask]
				bigram_prefix_histo[prev+lut[literal>>6]]++
				prev = lut[literal>>6] * 3
			}
		}

		ChooseContextMap(quality, bigram_prefix_histo[0:], num_literal_contexts, literal_context_map)
	}
}

func ShouldCompress_encode(data []byte, mask uint, last_flush_pos uint64, bytes uint, num_literals uint, num_commands uint) bool {
	/* TODO: find more precise minimal block overhead. */
	if bytes <= 2 {
		return false
	}
	if num_commands < (bytes>>8)+2 {
		if float64(num_literals) > 0.99*float64(bytes) {
			var literal_histo = [256]uint32{0}
			var kSampleRate uint32 = 13
			var kMinEntropy float64 = 7.92
			var bit_cost_threshold float64 = float64(bytes) * kMinEntropy / float64(kSampleRate)
			var t uint = uint((uint32(bytes) + kSampleRate - 1) / kSampleRate)
			var pos uint32 = uint32(last_flush_pos)
			var i uint
			for i = 0; i < t; i++ {
				literal_histo[data[pos&uint32(mask)]]++
				pos += kSampleRate
			}

			if BitsEntropy(literal_histo[:], 256) > bit_cost_threshold {
				return false
			}
		}
	}

	return true
}

/* Chooses the literal context mode for a metablock */
func ChooseContextMode(params *BrotliEncoderParams, data []byte, pos uint, mask uint, length uint) int {
	/* We only do the computation for the option of something else than
	   CONTEXT_UTF8 for the highest qualities */
	if params.quality >= MIN_QUALITY_FOR_HQ_BLOCK_SPLITTING && !BrotliIsMostlyUTF8(data, pos, mask, length, kMinUTF8Ratio) {
		return CONTEXT_SIGNED
	}

	return CONTEXT_UTF8
}

func WriteMetaBlockInternal(data []byte, mask uint, last_flush_pos uint64, bytes uint, is_last bool, literal_context_mode int, params *BrotliEncoderParams, prev_byte byte, prev_byte2 byte, num_literals uint, num_commands uint, commands []Command, saved_dist_cache []int, dist_cache []int, storage_ix *uint, storage []byte) {
	var wrapped_last_flush_pos uint32 = WrapPosition(last_flush_pos)
	var last_bytes uint16
	var last_bytes_bits byte
	var literal_context_lut ContextLut = BROTLI_CONTEXT_LUT(literal_context_mode)
	var block_params BrotliEncoderParams = *params

	if bytes == 0 {
		/* Write the ISLAST and ISEMPTY bits. */
		BrotliWriteBits(2, 3, storage_ix, storage)

		*storage_ix = (*storage_ix + 7) &^ 7
		return
	}

	if !ShouldCompress_encode(data, mask, last_flush_pos, bytes, num_literals, num_commands) {
		/* Restore the distance cache, as its last update by
		   CreateBackwardReferences is now unused. */
		copy(dist_cache, saved_dist_cache[:4])

		BrotliStoreUncompressedMetaBlock(is_last, data, uint(wrapped_last_flush_pos), mask, bytes, storage_ix, storage)
		return
	}

	assert(*storage_ix <= 14)
	last_bytes = uint16(storage[1]<<8 | storage[0])
	last_bytes_bits = byte(*storage_ix)
	if params.quality <= MAX_QUALITY_FOR_STATIC_ENTROPY_CODES {
		BrotliStoreMetaBlockFast(data, uint(wrapped_last_flush_pos), bytes, mask, is_last, params, commands, num_commands, storage_ix, storage)
	} else if params.quality < MIN_QUALITY_FOR_BLOCK_SPLIT {
		BrotliStoreMetaBlockTrivial(data, uint(wrapped_last_flush_pos), bytes, mask, is_last, params, commands, num_commands, storage_ix, storage)
	} else {
		var mb MetaBlockSplit
		InitMetaBlockSplit(&mb)
		if params.quality < MIN_QUALITY_FOR_HQ_BLOCK_SPLITTING {
			var num_literal_contexts uint = 1
			var literal_context_map []uint32 = nil
			if !params.disable_literal_context_modeling {
				DecideOverLiteralContextModeling(data, uint(wrapped_last_flush_pos), bytes, mask, params.quality, params.size_hint, &num_literal_contexts, &literal_context_map)
			}

			BrotliBuildMetaBlockGreedy(data, uint(wrapped_last_flush_pos), mask, prev_byte, prev_byte2, literal_context_lut, num_literal_contexts, literal_context_map, commands, num_commands, &mb)
		} else {
			BrotliBuildMetaBlock(data, uint(wrapped_last_flush_pos), mask, &block_params, prev_byte, prev_byte2, commands, num_commands, literal_context_mode, &mb)
		}

		if params.quality >= MIN_QUALITY_FOR_OPTIMIZE_HISTOGRAMS {
			/* The number of distance symbols effectively used for distance
			   histograms. It might be less than distance alphabet size
			   for "Large Window Brotli" (32-bit). */
			var num_effective_dist_codes uint32 = block_params.dist.alphabet_size
			if num_effective_dist_codes > BROTLI_NUM_HISTOGRAM_DISTANCE_SYMBOLS {
				num_effective_dist_codes = BROTLI_NUM_HISTOGRAM_DISTANCE_SYMBOLS
			}

			BrotliOptimizeHistograms(num_effective_dist_codes, &mb)
		}

		BrotliStoreMetaBlock(data, uint(wrapped_last_flush_pos), bytes, mask, prev_byte, prev_byte2, is_last, &block_params, literal_context_mode, commands, num_commands, &mb, storage_ix, storage)
		DestroyMetaBlockSplit(&mb)
	}

	if bytes+4 < *storage_ix>>3 {
		/* Restore the distance cache and last byte. */
		copy(dist_cache, saved_dist_cache[:4])

		storage[0] = byte(last_bytes)
		storage[1] = byte(last_bytes >> 8)
		*storage_ix = uint(last_bytes_bits)
		BrotliStoreUncompressedMetaBlock(is_last, data, uint(wrapped_last_flush_pos), mask, bytes, storage_ix, storage)
	}
}

func ChooseDistanceParams(params *BrotliEncoderParams) {
	var distance_postfix_bits uint32 = 0
	var num_direct_distance_codes uint32 = 0

	if params.quality >= MIN_QUALITY_FOR_NONZERO_DISTANCE_PARAMS {
		var ndirect_msb uint32
		if params.mode == BROTLI_MODE_FONT {
			distance_postfix_bits = 1
			num_direct_distance_codes = 12
		} else {
			distance_postfix_bits = params.dist.distance_postfix_bits
			num_direct_distance_codes = params.dist.num_direct_distance_codes
		}

		ndirect_msb = (num_direct_distance_codes >> distance_postfix_bits) & 0x0F
		if distance_postfix_bits > BROTLI_MAX_NPOSTFIX || num_direct_distance_codes > BROTLI_MAX_NDIRECT || ndirect_msb<<distance_postfix_bits != num_direct_distance_codes {
			distance_postfix_bits = 0
			num_direct_distance_codes = 0
		}
	}

	BrotliInitDistanceParams(params, distance_postfix_bits, num_direct_distance_codes)
}

func EnsureInitialized(s *Writer) bool {
	if s.is_initialized_ {
		return true
	}

	s.last_bytes_bits_ = 0
	s.last_bytes_ = 0
	s.remaining_metadata_bytes_ = BROTLI_UINT32_MAX

	SanitizeParams(&s.params)
	s.params.lgblock = ComputeLgBlock(&s.params)
	ChooseDistanceParams(&s.params)

	RingBufferSetup(&s.params, &s.ringbuffer_)

	/* Initialize last byte with stream header. */
	{
		var lgwin int = int(s.params.lgwin)
		if s.params.quality == FAST_ONE_PASS_COMPRESSION_QUALITY || s.params.quality == FAST_TWO_PASS_COMPRESSION_QUALITY {
			lgwin = brotli_max_int(lgwin, 18)
		}

		EncodeWindowBits(lgwin, s.params.large_window, &s.last_bytes_, &s.last_bytes_bits_)
	}

	if s.params.quality == FAST_ONE_PASS_COMPRESSION_QUALITY {
		InitCommandPrefixCodes(s.cmd_depths_[:], s.cmd_bits_[:], s.cmd_code_[:], &s.cmd_code_numbits_)
	}

	s.is_initialized_ = true
	return true
}

func BrotliEncoderInitParams(params *BrotliEncoderParams) {
	params.mode = BROTLI_DEFAULT_MODE
	params.large_window = false
	params.quality = BROTLI_DEFAULT_QUALITY
	params.lgwin = BROTLI_DEFAULT_WINDOW
	params.lgblock = 0
	params.size_hint = 0
	params.disable_literal_context_modeling = false
	BrotliInitEncoderDictionary(&params.dictionary)
	params.dist.distance_postfix_bits = 0
	params.dist.num_direct_distance_codes = 0
	params.dist.alphabet_size = uint32(BROTLI_DISTANCE_ALPHABET_SIZE(0, 0, BROTLI_MAX_DISTANCE_BITS))
	params.dist.max_distance = BROTLI_MAX_DISTANCE
}

func BrotliEncoderInitState(s *Writer) {
	BrotliEncoderInitParams(&s.params)
	s.input_pos_ = 0
	s.num_commands_ = 0
	s.num_literals_ = 0
	s.last_insert_len_ = 0
	s.last_flush_pos_ = 0
	s.last_processed_pos_ = 0
	s.prev_byte_ = 0
	s.prev_byte2_ = 0
	s.storage_size_ = 0
	s.storage_ = nil
	s.hasher_ = nil
	s.large_table_ = nil
	s.large_table_size_ = 0
	s.cmd_code_numbits_ = 0
	s.command_buf_ = nil
	s.literal_buf_ = nil
	s.next_out_ = nil
	s.available_out_ = 0
	s.total_out_ = 0
	s.stream_state_ = BROTLI_STREAM_PROCESSING
	s.is_last_block_emitted_ = false
	s.is_initialized_ = false

	RingBufferInit(&s.ringbuffer_)

	s.commands_ = nil
	s.cmd_alloc_size_ = 0

	/* Initialize distance cache. */
	s.dist_cache_[0] = 4

	s.dist_cache_[1] = 11
	s.dist_cache_[2] = 15
	s.dist_cache_[3] = 16

	/* Save the state of the distance cache in case we need to restore it for
	   emitting an uncompressed block. */
	copy(s.saved_dist_cache_[:], s.dist_cache_[:])
}

func BrotliEncoderCreateInstance() *Writer {
	var state *Writer = nil
	state = new(Writer)
	if state == nil {
		/* BROTLI_DUMP(); */
		return nil
	}

	BrotliEncoderInitState(state)
	return state
}

func BrotliEncoderCleanupState(s *Writer) {
	s.storage_ = nil
	s.commands_ = nil
	RingBufferFree(&s.ringbuffer_)
	DestroyHasher(&s.hasher_)
	s.large_table_ = nil
	s.command_buf_ = nil
	s.literal_buf_ = nil
}

/* Deinitializes and frees BrotliEncoderState instance. */
func BrotliEncoderDestroyInstance(state *Writer) {
	if state == nil {
		return
	} else {
		BrotliEncoderCleanupState(state)
	}
}

/*
   Copies the given input data to the internal ring buffer of the compressor.
   No processing of the data occurs at this time and this function can be
   called multiple times before calling WriteBrotliData() to process the
   accumulated input. At most input_block_size() bytes of input data can be
   copied to the ring buffer, otherwise the next WriteBrotliData() will fail.
*/
func CopyInputToRingBuffer(s *Writer, input_size uint, input_buffer []byte) {
	var ringbuffer_ *RingBuffer = &s.ringbuffer_
	RingBufferWrite(input_buffer, input_size, ringbuffer_)
	s.input_pos_ += uint64(input_size)

	/* TL;DR: If needed, initialize 7 more bytes in the ring buffer to make the
	   hashing not depend on uninitialized data. This makes compression
	   deterministic and it prevents uninitialized memory warnings in Valgrind.
	   Even without erasing, the output would be valid (but nondeterministic).

	   Background information: The compressor stores short (at most 8 bytes)
	   substrings of the input already read in a hash table, and detects
	   repetitions by looking up such substrings in the hash table. If it
	   can find a substring, it checks whether the substring is really there
	   in the ring buffer (or it's just a hash collision). Should the hash
	   table become corrupt, this check makes sure that the output is
	   still valid, albeit the compression ratio would be bad.

	   The compressor populates the hash table from the ring buffer as it's
	   reading new bytes from the input. However, at the last few indexes of
	   the ring buffer, there are not enough bytes to build full-length
	   substrings from. Since the hash table always contains full-length
	   substrings, we erase with dummy zeros here to make sure that those
	   substrings will contain zeros at the end instead of uninitialized
	   data.

	   Please note that erasing is not necessary (because the
	   memory region is already initialized since he ring buffer
	   has a `tail' that holds a copy of the beginning,) so we
	   skip erasing if we have already gone around at least once in
	   the ring buffer.

	   Only clear during the first round of ring-buffer writes. On
	   subsequent rounds data in the ring-buffer would be affected. */
	if ringbuffer_.pos_ <= ringbuffer_.mask_ {
		/* This is the first time when the ring buffer is being written.
		   We clear 7 bytes just after the bytes that have been copied from
		   the input buffer.

		   The ring-buffer has a "tail" that holds a copy of the beginning,
		   but only once the ring buffer has been fully written once, i.e.,
		   pos <= mask. For the first time, we need to write values
		   in this tail (where index may be larger than mask), so that
		   we have exactly defined behavior and don't read uninitialized
		   memory. Due to performance reasons, hashing reads data using a
		   LOAD64, which can go 7 bytes beyond the bytes written in the
		   ring-buffer. */
		for i := 0; i < int(7); i++ {
			ringbuffer_.buffer_[ringbuffer_.pos_:][i] = 0
		}
	}
}

/* Marks all input as processed.
   Returns true if position wrapping occurs. */
func UpdateLastProcessedPos(s *Writer) bool {
	var wrapped_last_processed_pos uint32 = WrapPosition(s.last_processed_pos_)
	var wrapped_input_pos uint32 = WrapPosition(s.input_pos_)
	s.last_processed_pos_ = s.input_pos_
	return wrapped_input_pos < wrapped_last_processed_pos
}

func ExtendLastCommand(s *Writer, bytes *uint32, wrapped_last_processed_pos *uint32) {
	var last_command *Command = &s.commands_[s.num_commands_-1]
	var data []byte = s.ringbuffer_.buffer_
	var mask uint32 = s.ringbuffer_.mask_
	var max_backward_distance uint64 = ((uint64(1)) << s.params.lgwin) - BROTLI_WINDOW_GAP
	var last_copy_len uint64 = uint64(last_command.copy_len_) & 0x1FFFFFF
	var last_processed_pos uint64 = s.last_processed_pos_ - last_copy_len
	var max_distance uint64
	if last_processed_pos < max_backward_distance {
		max_distance = last_processed_pos
	} else {
		max_distance = max_backward_distance
	}
	var cmd_dist uint64 = uint64(s.dist_cache_[0])
	var distance_code uint32 = CommandRestoreDistanceCode(last_command, &s.params.dist)
	if distance_code < BROTLI_NUM_DISTANCE_SHORT_CODES || uint64(distance_code-(BROTLI_NUM_DISTANCE_SHORT_CODES-1)) == cmd_dist {
		if cmd_dist <= max_distance {
			for *bytes != 0 && data[*wrapped_last_processed_pos&mask] == data[(uint64(*wrapped_last_processed_pos)-cmd_dist)&uint64(mask)] {
				last_command.copy_len_++
				(*bytes)--
				(*wrapped_last_processed_pos)++
			}
		}

		/* The copy length is at most the metablock size, and thus expressible. */
		GetLengthCode(uint(last_command.insert_len_), uint(int(last_command.copy_len_&0x1FFFFFF)+int(last_command.copy_len_>>25)), (last_command.dist_prefix_&0x3FF == 0), &last_command.cmd_prefix_)
	}
}

/*
   Processes the accumulated input data and sets |*out_size| to the length of
   the new output meta-block, or to zero if no new output meta-block has been
   created (in this case the processed input data is buffered internally).
   If |*out_size| is positive, |*output| points to the start of the output
   data. If |is_last| or |force_flush| is true, an output meta-block is
   always created. However, until |is_last| is true encoder may retain up
   to 7 bits of the last byte of output. To force encoder to dump the remaining
   bits use WriteMetadata() to append an empty meta-data block.
   Returns false if the size of the input data is larger than
   input_block_size().
*/
func EncodeData(s *Writer, is_last bool, force_flush bool, out_size *uint, output *[]byte) bool {
	var delta uint64 = UnprocessedInputSize(s)
	var bytes uint32 = uint32(delta)
	var wrapped_last_processed_pos uint32 = WrapPosition(s.last_processed_pos_)
	var data []byte
	var mask uint32
	var literal_context_mode int

	data = s.ringbuffer_.buffer_
	mask = s.ringbuffer_.mask_

	/* Adding more blocks after "last" block is forbidden. */
	if s.is_last_block_emitted_ {
		return false
	}
	if is_last {
		s.is_last_block_emitted_ = true
	}

	if delta > uint64(InputBlockSize(s)) {
		return false
	}

	if s.params.quality == FAST_TWO_PASS_COMPRESSION_QUALITY && s.command_buf_ == nil {
		s.command_buf_ = make([]uint32, kCompressFragmentTwoPassBlockSize)
		s.literal_buf_ = make([]byte, kCompressFragmentTwoPassBlockSize)
	}

	if s.params.quality == FAST_ONE_PASS_COMPRESSION_QUALITY || s.params.quality == FAST_TWO_PASS_COMPRESSION_QUALITY {
		var storage []byte
		var storage_ix uint = uint(s.last_bytes_bits_)
		var table_size uint
		var table []int

		if delta == 0 && !is_last {
			/* We have no new input data and we don't have to finish the stream, so
			   nothing to do. */
			*out_size = 0

			return true
		}

		storage = GetBrotliStorage(s, uint(2*bytes+503))
		storage[0] = byte(s.last_bytes_)
		storage[1] = byte(s.last_bytes_ >> 8)
		table = GetHashTable(s, s.params.quality, uint(bytes), &table_size)
		if s.params.quality == FAST_ONE_PASS_COMPRESSION_QUALITY {
			BrotliCompressFragmentFast(data[wrapped_last_processed_pos&mask:], uint(bytes), is_last, table, table_size, s.cmd_depths_[:], s.cmd_bits_[:], &s.cmd_code_numbits_, s.cmd_code_[:], &storage_ix, storage)
		} else {
			BrotliCompressFragmentTwoPass(data[wrapped_last_processed_pos&mask:], uint(bytes), is_last, s.command_buf_, s.literal_buf_, table, table_size, &storage_ix, storage)
		}

		s.last_bytes_ = uint16(storage[storage_ix>>3])
		s.last_bytes_bits_ = byte(storage_ix & 7)
		UpdateLastProcessedPos(s)
		*output = storage[0:]
		*out_size = storage_ix >> 3
		return true
	}
	{
		/* Theoretical max number of commands is 1 per 2 bytes. */
		var newsize uint = uint(uint32(s.num_commands_) + bytes/2 + 1)
		if newsize > s.cmd_alloc_size_ {
			var new_commands []Command

			/* Reserve a bit more memory to allow merging with a next block
			   without reallocation: that would impact speed. */
			newsize += uint((bytes / 4) + 16)

			s.cmd_alloc_size_ = newsize
			new_commands = make([]Command, newsize)
			if s.commands_ != nil {
				copy(new_commands, s.commands_[:s.num_commands_])
				s.commands_ = nil
			}

			s.commands_ = new_commands
		}
	}

	InitOrStitchToPreviousBlock(&s.hasher_, data, uint(mask), &s.params, uint(wrapped_last_processed_pos), uint(bytes), is_last)

	literal_context_mode = ChooseContextMode(&s.params, data, uint(WrapPosition(s.last_flush_pos_)), uint(mask), uint(s.input_pos_-s.last_flush_pos_))

	if s.num_commands_ != 0 && s.last_insert_len_ == 0 {
		ExtendLastCommand(s, &bytes, &wrapped_last_processed_pos)
	}

	if s.params.quality == ZOPFLIFICATION_QUALITY {
		assert(s.params.hasher.type_ == 10)
		BrotliCreateZopfliBackwardReferences(uint(bytes), uint(wrapped_last_processed_pos), data, uint(mask), &s.params, s.hasher_.(*H10), s.dist_cache_[:], &s.last_insert_len_, s.commands_[s.num_commands_:], &s.num_commands_, &s.num_literals_)
	} else if s.params.quality == HQ_ZOPFLIFICATION_QUALITY {
		assert(s.params.hasher.type_ == 10)
		BrotliCreateHqZopfliBackwardReferences(uint(bytes), uint(wrapped_last_processed_pos), data, uint(mask), &s.params, s.hasher_, s.dist_cache_[:], &s.last_insert_len_, s.commands_[s.num_commands_:], &s.num_commands_, &s.num_literals_)
	} else {
		BrotliCreateBackwardReferences(uint(bytes), uint(wrapped_last_processed_pos), data, uint(mask), &s.params, s.hasher_, s.dist_cache_[:], &s.last_insert_len_, s.commands_[s.num_commands_:], &s.num_commands_, &s.num_literals_)
	}
	{
		var max_length uint = MaxMetablockSize(&s.params)
		var max_literals uint = max_length / 8
		var max_commands uint = max_length / 8
		var processed_bytes uint = uint(s.input_pos_ - s.last_flush_pos_)
		var next_input_fits_metablock bool = (processed_bytes+InputBlockSize(s) <= max_length)
		var should_flush bool = (s.params.quality < MIN_QUALITY_FOR_BLOCK_SPLIT && s.num_literals_+s.num_commands_ >= MAX_NUM_DELAYED_SYMBOLS)
		/* If maximal possible additional block doesn't fit metablock, flush now. */
		/* TODO: Postpone decision until next block arrives? */

		/* If block splitting is not used, then flush as soon as there is some
		   amount of commands / literals produced. */
		if !is_last && !force_flush && !should_flush && next_input_fits_metablock && s.num_literals_ < max_literals && s.num_commands_ < max_commands {
			/* Merge with next input block. Everything will happen later. */
			if UpdateLastProcessedPos(s) {
				HasherReset(s.hasher_)
			}

			*out_size = 0
			return true
		}
	}

	/* Create the last insert-only command. */
	if s.last_insert_len_ > 0 {
		InitInsertCommand(&s.commands_[s.num_commands_], s.last_insert_len_)
		s.num_commands_++
		s.num_literals_ += s.last_insert_len_
		s.last_insert_len_ = 0
	}

	if !is_last && s.input_pos_ == s.last_flush_pos_ {
		/* We have no new input data and we don't have to finish the stream, so
		   nothing to do. */
		*out_size = 0

		return true
	}

	assert(s.input_pos_ >= s.last_flush_pos_)
	assert(s.input_pos_ > s.last_flush_pos_ || is_last)
	assert(s.input_pos_-s.last_flush_pos_ <= 1<<24)
	{
		var metablock_size uint32 = uint32(s.input_pos_ - s.last_flush_pos_)
		var storage []byte = GetBrotliStorage(s, uint(2*metablock_size+503))
		var storage_ix uint = uint(s.last_bytes_bits_)
		storage[0] = byte(s.last_bytes_)
		storage[1] = byte(s.last_bytes_ >> 8)
		WriteMetaBlockInternal(data, uint(mask), s.last_flush_pos_, uint(metablock_size), is_last, literal_context_mode, &s.params, s.prev_byte_, s.prev_byte2_, s.num_literals_, s.num_commands_, s.commands_, s.saved_dist_cache_[:], s.dist_cache_[:], &storage_ix, storage)
		s.last_bytes_ = uint16(storage[storage_ix>>3])
		s.last_bytes_bits_ = byte(storage_ix & 7)
		s.last_flush_pos_ = s.input_pos_
		if UpdateLastProcessedPos(s) {
			HasherReset(s.hasher_)
		}

		if s.last_flush_pos_ > 0 {
			s.prev_byte_ = data[(uint32(s.last_flush_pos_)-1)&mask]
		}

		if s.last_flush_pos_ > 1 {
			s.prev_byte2_ = data[uint32(s.last_flush_pos_-2)&mask]
		}

		s.num_commands_ = 0
		s.num_literals_ = 0

		/* Save the state of the distance cache in case we need to restore it for
		   emitting an uncompressed block. */
		copy(s.saved_dist_cache_[:], s.dist_cache_[:])

		*output = storage[0:]
		*out_size = storage_ix >> 3
		return true
	}
}

/* Dumps remaining output bits and metadata header to |header|.
   Returns number of produced bytes.
   REQUIRED: |header| should be 8-byte aligned and at least 16 bytes long.
   REQUIRED: |block_size| <= (1 << 24). */
func WriteMetadataHeader(s *Writer, block_size uint, header []byte) uint {
	var storage_ix uint
	storage_ix = uint(s.last_bytes_bits_)
	header[0] = byte(s.last_bytes_)
	header[1] = byte(s.last_bytes_ >> 8)
	s.last_bytes_ = 0
	s.last_bytes_bits_ = 0

	BrotliWriteBits(1, 0, &storage_ix, header)
	BrotliWriteBits(2, 3, &storage_ix, header)
	BrotliWriteBits(1, 0, &storage_ix, header)
	if block_size == 0 {
		BrotliWriteBits(2, 0, &storage_ix, header)
	} else {
		var nbits uint32
		if block_size == 1 {
			nbits = 0
		} else {
			nbits = Log2FloorNonZero(uint(uint32(block_size)-1)) + 1
		}
		var nbytes uint32 = (nbits + 7) / 8
		BrotliWriteBits(2, uint64(nbytes), &storage_ix, header)
		BrotliWriteBits(uint(8*nbytes), uint64(block_size)-1, &storage_ix, header)
	}

	return (storage_ix + 7) >> 3
}

func BrotliCompressBufferQuality10(lgwin int, input_size uint, input_buffer []byte, encoded_size *uint, encoded_buffer []byte) bool {
	var mask uint = BROTLI_SIZE_MAX >> 1
	var dist_cache = [4]int{4, 11, 15, 16}
	var saved_dist_cache = [4]int{4, 11, 15, 16}
	var ok bool = true
	var max_out_size uint = *encoded_size
	var total_out_size uint = 0
	var last_bytes uint16
	var last_bytes_bits byte
	var hasher HasherHandle = nil
	var hasher_eff_size uint = brotli_min_size_t(input_size, BROTLI_MAX_BACKWARD_LIMIT(uint(lgwin))+BROTLI_WINDOW_GAP)
	var params BrotliEncoderParams
	var lgmetablock int = brotli_min_int(24, lgwin+1)
	var max_block_size uint
	var max_metablock_size uint = uint(1) << uint(lgmetablock)
	var max_literals_per_metablock uint = max_metablock_size / 8
	var max_commands_per_metablock uint = max_metablock_size / 8
	var metablock_start uint = 0
	var prev_byte byte = 0
	var prev_byte2 byte = 0

	BrotliEncoderInitParams(&params)
	params.quality = 10
	params.lgwin = uint(lgwin)
	if lgwin > BROTLI_MAX_WINDOW_BITS {
		params.large_window = true
	}

	SanitizeParams(&params)
	params.lgblock = ComputeLgBlock(&params)
	ChooseDistanceParams(&params)
	max_block_size = uint(1) << uint(params.lgblock)

	assert(input_size <= mask+1)
	EncodeWindowBits(lgwin, params.large_window, &last_bytes, &last_bytes_bits)
	InitOrStitchToPreviousBlock(&hasher, input_buffer, mask, &params, 0, hasher_eff_size, true)

	for ok && metablock_start < input_size {
		var metablock_end uint = brotli_min_size_t(input_size, metablock_start+max_metablock_size)
		var expected_num_commands uint = (metablock_end-metablock_start)/12 + 16
		var commands []Command = nil
		var num_commands uint = 0
		var last_insert_len uint = 0
		var num_literals uint = 0
		var metablock_size uint = 0
		var cmd_alloc_size uint = 0
		var is_last bool
		var storage []byte
		var storage_ix uint
		var literal_context_mode int = ChooseContextMode(&params, input_buffer, metablock_start, mask, metablock_end-metablock_start)
		var block_start uint
		for block_start = metablock_start; block_start < metablock_end; {
			var block_size uint = brotli_min_size_t(metablock_end-block_start, max_block_size)
			var nodes []ZopfliNode = make([]ZopfliNode, (block_size + 1))
			var path_size uint
			var new_cmd_alloc_size uint
			BrotliInitZopfliNodes(nodes, block_size+1)
			hasher.StitchToPreviousBlock(block_size, block_start, input_buffer, mask)
			path_size = BrotliZopfliComputeShortestPath(block_size, block_start, input_buffer, mask, &params, dist_cache[:], hasher.(*H10), nodes)

			/* We allocate a command buffer in the first iteration of this loop that
			   will be likely big enough for the whole metablock, so that for most
			   inputs we will not have to reallocate in later iterations. We do the
			   allocation here and not before the loop, because if the input is small,
			   this will be allocated after the Zopfli cost model is freed, so this
			   will not increase peak memory usage.
			   TODO: If the first allocation is too small, increase command
			   buffer size exponentially. */
			new_cmd_alloc_size = brotli_max_size_t(expected_num_commands, num_commands+path_size+1)

			if cmd_alloc_size != new_cmd_alloc_size {
				var new_commands []Command = make([]Command, new_cmd_alloc_size)
				cmd_alloc_size = new_cmd_alloc_size
				if commands != nil {
					copy(new_commands, commands[:num_commands])
					commands = nil
				}

				commands = new_commands
			}

			BrotliZopfliCreateCommands(block_size, block_start, nodes[0:], dist_cache[:], &last_insert_len, &params, commands[num_commands:], &num_literals)
			num_commands += path_size
			block_start += block_size
			metablock_size += block_size
			nodes = nil
			if num_literals > max_literals_per_metablock || num_commands > max_commands_per_metablock {
				break
			}
		}

		if last_insert_len > 0 {
			InitInsertCommand(&commands[num_commands], last_insert_len)
			num_commands++
			num_literals += last_insert_len
		}

		is_last = (metablock_start+metablock_size == input_size)
		storage = nil
		storage_ix = uint(last_bytes_bits)

		if metablock_size == 0 {
			/* Write the ISLAST and ISEMPTY bits. */
			storage = make([]byte, 16)

			storage[0] = byte(last_bytes)
			storage[1] = byte(last_bytes >> 8)
			BrotliWriteBits(2, 3, &storage_ix, storage)
			storage_ix = (storage_ix + 7) &^ 7
		} else if !ShouldCompress_encode(input_buffer, mask, uint64(metablock_start), metablock_size, num_literals, num_commands) {
			/* Restore the distance cache, as its last update by
			   CreateBackwardReferences is now unused. */
			copy(dist_cache[:], saved_dist_cache[:4])

			storage = make([]byte, (metablock_size + 16))
			storage[0] = byte(last_bytes)
			storage[1] = byte(last_bytes >> 8)
			BrotliStoreUncompressedMetaBlock(is_last, input_buffer, metablock_start, mask, metablock_size, &storage_ix, storage)
		} else {
			var mb MetaBlockSplit
			var block_params BrotliEncoderParams = params
			InitMetaBlockSplit(&mb)
			BrotliBuildMetaBlock(input_buffer, metablock_start, mask, &block_params, prev_byte, prev_byte2, commands, num_commands, literal_context_mode, &mb)
			{
				/* The number of distance symbols effectively used for distance
				   histograms. It might be less than distance alphabet size
				   for "Large Window Brotli" (32-bit). */
				var num_effective_dist_codes uint32 = block_params.dist.alphabet_size
				if num_effective_dist_codes > BROTLI_NUM_HISTOGRAM_DISTANCE_SYMBOLS {
					num_effective_dist_codes = BROTLI_NUM_HISTOGRAM_DISTANCE_SYMBOLS
				}

				BrotliOptimizeHistograms(num_effective_dist_codes, &mb)
			}

			storage = make([]byte, (2*metablock_size + 503))
			storage[0] = byte(last_bytes)
			storage[1] = byte(last_bytes >> 8)
			BrotliStoreMetaBlock(input_buffer, metablock_start, metablock_size, mask, prev_byte, prev_byte2, is_last, &block_params, literal_context_mode, commands, num_commands, &mb, &storage_ix, storage)
			if metablock_size+4 < storage_ix>>3 {
				/* Restore the distance cache and last byte. */
				copy(dist_cache[:], saved_dist_cache[:4])

				storage[0] = byte(last_bytes)
				storage[1] = byte(last_bytes >> 8)
				storage_ix = uint(last_bytes_bits)
				BrotliStoreUncompressedMetaBlock(is_last, input_buffer, metablock_start, mask, metablock_size, &storage_ix, storage)
			}

			DestroyMetaBlockSplit(&mb)
		}

		last_bytes = uint16(storage[storage_ix>>3])
		last_bytes_bits = byte(storage_ix & 7)
		metablock_start += metablock_size
		if metablock_start < input_size {
			prev_byte = input_buffer[metablock_start-1]
			prev_byte2 = input_buffer[metablock_start-2]
		}

		/* Save the state of the distance cache in case we need to restore it for
		   emitting an uncompressed block. */
		copy(saved_dist_cache[:], dist_cache[:4])
		{
			var out_size uint = storage_ix >> 3
			total_out_size += out_size
			if total_out_size <= max_out_size {
				copy(encoded_buffer, storage[:out_size])
				encoded_buffer = encoded_buffer[out_size:]
			} else {
				ok = false
			}
		}

		storage = nil
		commands = nil
	}

	*encoded_size = total_out_size
	DestroyHasher(&hasher)
	return ok
}

func BrotliEncoderMaxCompressedSize(input_size uint) uint {
	var num_large_blocks uint = input_size >> 14
	var overhead uint = 2 + (4 * num_large_blocks) + 3 + 1
	/* [window bits / empty metadata] + N * [uncompressed] + [last empty] */

	var result uint = input_size + overhead
	if input_size == 0 {
		return 2
	}
	if result < input_size {
		return 0
	} else {
		return result
	}
}

/* Wraps data to uncompressed brotli stream with minimal window size.
   |output| should point at region with at least BrotliEncoderMaxCompressedSize
   addressable bytes.
   Returns the length of stream. */
func MakeUncompressedStream(input []byte, input_size uint, output []byte) uint {
	var size uint = input_size
	var result uint = 0
	var offset uint = 0
	if input_size == 0 {
		output[0] = 6
		return 1
	}

	output[result] = 0x21
	result++ /* window bits = 10, is_last = false */
	output[result] = 0x03
	result++ /* empty metadata, padding */
	for size > 0 {
		var nibbles uint32 = 0
		var chunk_size uint32
		var bits uint32
		if size > 1<<24 {
			chunk_size = 1 << 24
		} else {
			chunk_size = uint32(size)
		}
		if chunk_size > 1<<16 {
			if chunk_size > 1<<20 {
				nibbles = 2
			} else {
				nibbles = 1
			}
		}
		bits = nibbles<<1 | (chunk_size-1)<<3 | 1<<(19+4*nibbles)
		output[result] = byte(bits)
		result++
		output[result] = byte(bits >> 8)
		result++
		output[result] = byte(bits >> 16)
		result++
		if nibbles == 2 {
			output[result] = byte(bits >> 24)
			result++
		}
		copy(output[result:], input[offset:][:chunk_size])
		result += uint(chunk_size)
		offset += uint(chunk_size)
		size -= uint(chunk_size)
	}

	output[result] = 3
	result++
	return result
}

func BrotliEncoderCompress(quality int, lgwin int, mode int, input_size uint, input_buffer []byte, encoded_size *uint, encoded_buffer []byte) bool {
	var s *Writer
	var out_size uint = *encoded_size
	var input_start []byte = input_buffer
	var output_start []byte = encoded_buffer
	var max_out_size uint = BrotliEncoderMaxCompressedSize(input_size)
	if out_size == 0 {
		/* Output buffer needs at least one byte. */
		return false
	}

	if input_size == 0 {
		/* Handle the special case of empty input. */
		*encoded_size = 1

		encoded_buffer[0] = 6
		return true
	}

	if quality == 10 {
		var lg_win int = brotli_min_int(BROTLI_LARGE_MAX_WINDOW_BITS, brotli_max_int(16, lgwin))
		/* TODO: Implement this direct path for all quality levels. */

		var ok bool = BrotliCompressBufferQuality10(lg_win, input_size, input_buffer, encoded_size, encoded_buffer)
		if !ok || (max_out_size != 0 && *encoded_size > max_out_size) {
			goto fallback
		}

		return true
	}

	s = BrotliEncoderCreateInstance()
	if s == nil {
		return false
	} else {
		var available_in uint = input_size
		var next_in []byte = input_buffer
		var available_out uint = *encoded_size
		var next_out []byte = encoded_buffer
		var total_out uint = 0
		var result bool = false
		BrotliEncoderSetParameter(s, BROTLI_PARAM_QUALITY, uint32(quality))
		BrotliEncoderSetParameter(s, BROTLI_PARAM_LGWIN, uint32(lgwin))
		BrotliEncoderSetParameter(s, BROTLI_PARAM_MODE, uint32(mode))
		BrotliEncoderSetParameter(s, BROTLI_PARAM_SIZE_HINT, uint32(input_size))
		if lgwin > BROTLI_MAX_WINDOW_BITS {
			BrotliEncoderSetParameter(s, BROTLI_PARAM_LARGE_WINDOW, 1)
		}

		result = BrotliEncoderCompressStream(s, int(BROTLI_OPERATION_FINISH), &available_in, &next_in, &available_out, &next_out, &total_out)
		if !BrotliEncoderIsFinished(s) {
			result = false
		}
		*encoded_size = total_out
		BrotliEncoderDestroyInstance(s)
		if !result || (max_out_size != 0 && *encoded_size > max_out_size) {
			goto fallback
		}

		return true
	}

fallback:
	*encoded_size = 0
	if max_out_size == 0 {
		return false
	}
	if out_size >= max_out_size {
		*encoded_size = MakeUncompressedStream(input_start, input_size, output_start)
		return true
	}

	return false
}

func InjectBytePaddingBlock(s *Writer) {
	var seal uint32 = uint32(s.last_bytes_)
	var seal_bits uint = uint(s.last_bytes_bits_)
	var destination []byte
	s.last_bytes_ = 0
	s.last_bytes_bits_ = 0

	/* is_last = 0, data_nibbles = 11, reserved = 0, meta_nibbles = 00 */
	seal |= 0x6 << seal_bits

	seal_bits += 6

	/* If we have already created storage, then append to it.
	   Storage is valid until next block is being compressed. */
	if s.next_out_ != nil {
		destination = s.next_out_[s.available_out_:]
	} else {
		destination = s.tiny_buf_.u8[:]
		s.next_out_ = destination
	}

	destination[0] = byte(seal)
	if seal_bits > 8 {
		destination[1] = byte(seal >> 8)
	}
	if seal_bits > 16 {
		destination[2] = byte(seal >> 16)
	}
	s.available_out_ += (seal_bits + 7) >> 3
}

/* Injects padding bits or pushes compressed data to output.
   Returns false if nothing is done. */
func InjectFlushOrPushOutput(s *Writer, available_out *uint, next_out *[]byte, total_out *uint) bool {
	if s.stream_state_ == BROTLI_STREAM_FLUSH_REQUESTED && s.last_bytes_bits_ != 0 {
		InjectBytePaddingBlock(s)
		return true
	}

	if s.available_out_ != 0 && *available_out != 0 {
		var copy_output_size uint = brotli_min_size_t(s.available_out_, *available_out)
		copy(*next_out, s.next_out_[:copy_output_size])
		*next_out = (*next_out)[copy_output_size:]
		*available_out -= copy_output_size
		s.next_out_ = s.next_out_[copy_output_size:]
		s.available_out_ -= copy_output_size
		s.total_out_ += copy_output_size
		if total_out != nil {
			*total_out = s.total_out_
		}
		return true
	}

	return false
}

func CheckFlushComplete(s *Writer) {
	if s.stream_state_ == BROTLI_STREAM_FLUSH_REQUESTED && s.available_out_ == 0 {
		s.stream_state_ = BROTLI_STREAM_PROCESSING
		s.next_out_ = nil
	}
}

func BrotliEncoderCompressStreamFast(s *Writer, op int, available_in *uint, next_in *[]byte, available_out *uint, next_out *[]byte, total_out *uint) bool {
	var block_size_limit uint = uint(1) << s.params.lgwin
	var buf_size uint = brotli_min_size_t(kCompressFragmentTwoPassBlockSize, brotli_min_size_t(*available_in, block_size_limit))
	var tmp_command_buf []uint32 = nil
	var command_buf []uint32 = nil
	var tmp_literal_buf []byte = nil
	var literal_buf []byte = nil
	if s.params.quality != FAST_ONE_PASS_COMPRESSION_QUALITY && s.params.quality != FAST_TWO_PASS_COMPRESSION_QUALITY {
		return false
	}

	if s.params.quality == FAST_TWO_PASS_COMPRESSION_QUALITY {
		if s.command_buf_ == nil && buf_size == kCompressFragmentTwoPassBlockSize {
			s.command_buf_ = make([]uint32, kCompressFragmentTwoPassBlockSize)
			s.literal_buf_ = make([]byte, kCompressFragmentTwoPassBlockSize)
		}

		if s.command_buf_ != nil {
			command_buf = s.command_buf_
			literal_buf = s.literal_buf_
		} else {
			tmp_command_buf = make([]uint32, buf_size)
			tmp_literal_buf = make([]byte, buf_size)
			command_buf = tmp_command_buf
			literal_buf = tmp_literal_buf
		}
	}

	for {
		if InjectFlushOrPushOutput(s, available_out, next_out, total_out) {
			continue
		}

		/* Compress block only when internal output buffer is empty, stream is not
		   finished, there is no pending flush request, and there is either
		   additional input or pending operation. */
		if s.available_out_ == 0 && s.stream_state_ == BROTLI_STREAM_PROCESSING && (*available_in != 0 || op != int(BROTLI_OPERATION_PROCESS)) {
			var block_size uint = brotli_min_size_t(block_size_limit, *available_in)
			var is_last bool = (*available_in == block_size) && (op == int(BROTLI_OPERATION_FINISH))
			var force_flush bool = (*available_in == block_size) && (op == int(BROTLI_OPERATION_FLUSH))
			var max_out_size uint = 2*block_size + 503
			var inplace bool = true
			var storage []byte = nil
			var storage_ix uint = uint(s.last_bytes_bits_)
			var table_size uint
			var table []int

			if force_flush && block_size == 0 {
				s.stream_state_ = BROTLI_STREAM_FLUSH_REQUESTED
				continue
			}

			if max_out_size <= *available_out {
				storage = *next_out
			} else {
				inplace = false
				storage = GetBrotliStorage(s, max_out_size)
			}

			storage[0] = byte(s.last_bytes_)
			storage[1] = byte(s.last_bytes_ >> 8)
			table = GetHashTable(s, s.params.quality, block_size, &table_size)

			if s.params.quality == FAST_ONE_PASS_COMPRESSION_QUALITY {
				BrotliCompressFragmentFast(*next_in, block_size, is_last, table, table_size, s.cmd_depths_[:], s.cmd_bits_[:], &s.cmd_code_numbits_, s.cmd_code_[:], &storage_ix, storage)
			} else {
				BrotliCompressFragmentTwoPass(*next_in, block_size, is_last, command_buf, literal_buf, table, table_size, &storage_ix, storage)
			}

			*next_in = (*next_in)[block_size:]
			*available_in -= block_size
			if inplace {
				var out_bytes uint = storage_ix >> 3
				assert(out_bytes <= *available_out)
				assert(storage_ix&7 == 0 || out_bytes < *available_out)
				*next_out = (*next_out)[out_bytes:]
				*available_out -= out_bytes
				s.total_out_ += out_bytes
				if total_out != nil {
					*total_out = s.total_out_
				}
			} else {
				var out_bytes uint = storage_ix >> 3
				s.next_out_ = storage
				s.available_out_ = out_bytes
			}

			s.last_bytes_ = uint16(storage[storage_ix>>3])
			s.last_bytes_bits_ = byte(storage_ix & 7)

			if force_flush {
				s.stream_state_ = BROTLI_STREAM_FLUSH_REQUESTED
			}
			if is_last {
				s.stream_state_ = BROTLI_STREAM_FINISHED
			}
			continue
		}

		break
	}

	tmp_command_buf = nil
	tmp_literal_buf = nil
	CheckFlushComplete(s)
	return true
}

func ProcessMetadata(s *Writer, available_in *uint, next_in *[]byte, available_out *uint, next_out *[]byte, total_out *uint) bool {
	if *available_in > 1<<24 {
		return false
	}

	/* Switch to metadata block workflow, if required. */
	if s.stream_state_ == BROTLI_STREAM_PROCESSING {
		s.remaining_metadata_bytes_ = uint32(*available_in)
		s.stream_state_ = BROTLI_STREAM_METADATA_HEAD
	}

	if s.stream_state_ != BROTLI_STREAM_METADATA_HEAD && s.stream_state_ != BROTLI_STREAM_METADATA_BODY {
		return false
	}

	for {
		if InjectFlushOrPushOutput(s, available_out, next_out, total_out) {
			continue
		}

		if s.available_out_ != 0 {
			break
		}

		if s.input_pos_ != s.last_flush_pos_ {
			var result bool = EncodeData(s, false, true, &s.available_out_, &s.next_out_)
			if !result {
				return false
			}
			continue
		}

		if s.stream_state_ == BROTLI_STREAM_METADATA_HEAD {
			s.next_out_ = s.tiny_buf_.u8[:]
			s.available_out_ = WriteMetadataHeader(s, uint(s.remaining_metadata_bytes_), s.next_out_)
			s.stream_state_ = BROTLI_STREAM_METADATA_BODY
			continue
		} else {
			/* Exit workflow only when there is no more input and no more output.
			   Otherwise client may continue producing empty metadata blocks. */
			if s.remaining_metadata_bytes_ == 0 {
				s.remaining_metadata_bytes_ = BROTLI_UINT32_MAX
				s.stream_state_ = BROTLI_STREAM_PROCESSING
				break
			}

			if *available_out != 0 {
				/* Directly copy input to output. */
				var c uint32 = uint32(brotli_min_size_t(uint(s.remaining_metadata_bytes_), *available_out))
				copy(*next_out, (*next_in)[:c])
				*next_in = (*next_in)[c:]
				*available_in -= uint(c)
				s.remaining_metadata_bytes_ -= c
				*next_out = (*next_out)[c:]
				*available_out -= uint(c)
			} else {
				/* This guarantees progress in "TakeOutput" workflow. */
				var c uint32 = brotli_min_uint32_t(s.remaining_metadata_bytes_, 16)
				s.next_out_ = s.tiny_buf_.u8[:]
				copy(s.next_out_, (*next_in)[:c])
				*next_in = (*next_in)[c:]
				*available_in -= uint(c)
				s.remaining_metadata_bytes_ -= c
				s.available_out_ = uint(c)
			}

			continue
		}
	}

	return true
}

func UpdateSizeHint(s *Writer, available_in uint) {
	if s.params.size_hint == 0 {
		var delta uint64 = UnprocessedInputSize(s)
		var tail uint64 = uint64(available_in)
		var limit uint32 = 1 << 30
		var total uint32
		if (delta >= uint64(limit)) || (tail >= uint64(limit)) || ((delta + tail) >= uint64(limit)) {
			total = limit
		} else {
			total = uint32(delta + tail)
		}

		s.params.size_hint = uint(total)
	}
}

func BrotliEncoderCompressStream(s *Writer, op int, available_in *uint, next_in *[]byte, available_out *uint, next_out *[]byte, total_out *uint) bool {
	if !EnsureInitialized(s) {
		return false
	}

	/* Unfinished metadata block; check requirements. */
	if s.remaining_metadata_bytes_ != BROTLI_UINT32_MAX {
		if uint32(*available_in) != s.remaining_metadata_bytes_ {
			return false
		}
		if op != int(BROTLI_OPERATION_EMIT_METADATA) {
			return false
		}
	}

	if op == int(BROTLI_OPERATION_EMIT_METADATA) {
		UpdateSizeHint(s, 0) /* First data metablock might be emitted here. */
		return ProcessMetadata(s, available_in, next_in, available_out, next_out, total_out)
	}

	if s.stream_state_ == BROTLI_STREAM_METADATA_HEAD || s.stream_state_ == BROTLI_STREAM_METADATA_BODY {
		return false
	}

	if s.stream_state_ != BROTLI_STREAM_PROCESSING && *available_in != 0 {
		return false
	}

	if s.params.quality == FAST_ONE_PASS_COMPRESSION_QUALITY || s.params.quality == FAST_TWO_PASS_COMPRESSION_QUALITY {
		return BrotliEncoderCompressStreamFast(s, op, available_in, next_in, available_out, next_out, total_out)
	}

	for {
		var remaining_block_size uint = RemainingInputBlockSize(s)

		if remaining_block_size != 0 && *available_in != 0 {
			var copy_input_size uint = brotli_min_size_t(remaining_block_size, *available_in)
			CopyInputToRingBuffer(s, copy_input_size, *next_in)
			*next_in = (*next_in)[copy_input_size:]
			*available_in -= copy_input_size
			continue
		}

		if InjectFlushOrPushOutput(s, available_out, next_out, total_out) {
			continue
		}

		/* Compress data only when internal output buffer is empty, stream is not
		   finished and there is no pending flush request. */
		if s.available_out_ == 0 && s.stream_state_ == BROTLI_STREAM_PROCESSING {
			if remaining_block_size == 0 || op != int(BROTLI_OPERATION_PROCESS) {
				var is_last bool = ((*available_in == 0) && op == int(BROTLI_OPERATION_FINISH))
				var force_flush bool = ((*available_in == 0) && op == int(BROTLI_OPERATION_FLUSH))
				var result bool
				UpdateSizeHint(s, *available_in)
				result = EncodeData(s, is_last, force_flush, &s.available_out_, &s.next_out_)
				if !result {
					return false
				}
				if force_flush {
					s.stream_state_ = BROTLI_STREAM_FLUSH_REQUESTED
				}
				if is_last {
					s.stream_state_ = BROTLI_STREAM_FINISHED
				}
				continue
			}
		}

		break
	}

	CheckFlushComplete(s)
	return true
}

func BrotliEncoderIsFinished(s *Writer) bool {
	return s.stream_state_ == BROTLI_STREAM_FINISHED && !BrotliEncoderHasMoreOutput(s)
}

func BrotliEncoderHasMoreOutput(s *Writer) bool {
	return s.available_out_ != 0
}

func BrotliEncoderTakeOutput(s *Writer, size *uint) []byte {
	var consumed_size uint = s.available_out_
	var result []byte = s.next_out_
	if *size != 0 {
		consumed_size = brotli_min_size_t(*size, s.available_out_)
	}

	if consumed_size != 0 {
		s.next_out_ = s.next_out_[consumed_size:]
		s.available_out_ -= consumed_size
		s.total_out_ += consumed_size
		CheckFlushComplete(s)
		*size = consumed_size
	} else {
		*size = 0
		result = nil
	}

	return result
}

func BrotliEncoderVersion() uint32 {
	return BROTLI_VERSION
}