forked from Shiloh/githaven
792b4dba2c
* update github.com/blevesearch/bleve v2.0.2 -> v2.0.3 * github.com/denisenkom/go-mssqldb v0.9.0 -> v0.10.0 * github.com/editorconfig/editorconfig-core-go v2.4.1 -> v2.4.2 * github.com/go-chi/cors v1.1.1 -> v1.2.0 * github.com/go-git/go-billy v5.0.0 -> v5.1.0 * github.com/go-git/go-git v5.2.0 -> v5.3.0 * github.com/go-ldap/ldap v3.2.4 -> v3.3.0 * github.com/go-redis/redis v8.6.0 -> v8.8.2 * github.com/go-sql-driver/mysql v1.5.0 -> v1.6.0 * github.com/go-swagger/go-swagger v0.26.1 -> v0.27.0 * github.com/lib/pq v1.9.0 -> v1.10.1 * github.com/mattn/go-sqlite3 v1.14.6 -> v1.14.7 * github.com/go-testfixtures/testfixtures v3.5.0 -> v3.6.0 * github.com/issue9/identicon v1.0.1 -> v1.2.0 * github.com/klauspost/compress v1.11.8 -> v1.12.1 * github.com/mgechev/revive v1.0.3 -> v1.0.6 * github.com/microcosm-cc/bluemonday v1.0.7 -> v1.0.8 * github.com/niklasfasching/go-org v1.4.0 -> v1.5.0 * github.com/olivere/elastic v7.0.22 -> v7.0.24 * github.com/pelletier/go-toml v1.8.1 -> v1.9.0 * github.com/prometheus/client_golang v1.9.0 -> v1.10.0 * github.com/xanzy/go-gitlab v0.44.0 -> v0.48.0 * github.com/yuin/goldmark v1.3.3 -> v1.3.5 * github.com/6543/go-version v1.2.4 -> v1.3.1 * do github.com/lib/pq v1.10.0 -> v1.10.1 again ...
179 lines
4.3 KiB
Go
Vendored
179 lines
4.3 KiB
Go
Vendored
package zstd
|
|
|
|
import (
|
|
"fmt"
|
|
"math/bits"
|
|
|
|
"github.com/klauspost/compress/zstd/internal/xxhash"
|
|
)
|
|
|
|
const (
|
|
dictShardBits = 6
|
|
)
|
|
|
|
type fastBase struct {
|
|
// cur is the offset at the start of hist
|
|
cur int32
|
|
// maximum offset. Should be at least 2x block size.
|
|
maxMatchOff int32
|
|
hist []byte
|
|
crc *xxhash.Digest
|
|
tmp [8]byte
|
|
blk *blockEnc
|
|
lastDictID uint32
|
|
lowMem bool
|
|
}
|
|
|
|
// CRC returns the underlying CRC writer.
|
|
func (e *fastBase) CRC() *xxhash.Digest {
|
|
return e.crc
|
|
}
|
|
|
|
// AppendCRC will append the CRC to the destination slice and return it.
|
|
func (e *fastBase) AppendCRC(dst []byte) []byte {
|
|
crc := e.crc.Sum(e.tmp[:0])
|
|
dst = append(dst, crc[7], crc[6], crc[5], crc[4])
|
|
return dst
|
|
}
|
|
|
|
// WindowSize returns the window size of the encoder,
|
|
// or a window size small enough to contain the input size, if > 0.
|
|
func (e *fastBase) WindowSize(size int) int32 {
|
|
if size > 0 && size < int(e.maxMatchOff) {
|
|
b := int32(1) << uint(bits.Len(uint(size)))
|
|
// Keep minimum window.
|
|
if b < 1024 {
|
|
b = 1024
|
|
}
|
|
return b
|
|
}
|
|
return e.maxMatchOff
|
|
}
|
|
|
|
// Block returns the current block.
|
|
func (e *fastBase) Block() *blockEnc {
|
|
return e.blk
|
|
}
|
|
|
|
func (e *fastBase) addBlock(src []byte) int32 {
|
|
if debugAsserts && e.cur > bufferReset {
|
|
panic(fmt.Sprintf("ecur (%d) > buffer reset (%d)", e.cur, bufferReset))
|
|
}
|
|
// check if we have space already
|
|
if len(e.hist)+len(src) > cap(e.hist) {
|
|
if cap(e.hist) == 0 {
|
|
e.ensureHist(len(src))
|
|
} else {
|
|
if cap(e.hist) < int(e.maxMatchOff+maxCompressedBlockSize) {
|
|
panic(fmt.Errorf("unexpected buffer cap %d, want at least %d with window %d", cap(e.hist), e.maxMatchOff+maxCompressedBlockSize, e.maxMatchOff))
|
|
}
|
|
// Move down
|
|
offset := int32(len(e.hist)) - e.maxMatchOff
|
|
copy(e.hist[0:e.maxMatchOff], e.hist[offset:])
|
|
e.cur += offset
|
|
e.hist = e.hist[:e.maxMatchOff]
|
|
}
|
|
}
|
|
s := int32(len(e.hist))
|
|
e.hist = append(e.hist, src...)
|
|
return s
|
|
}
|
|
|
|
// ensureHist will ensure that history can keep at least this many bytes.
|
|
func (e *fastBase) ensureHist(n int) {
|
|
if cap(e.hist) >= n {
|
|
return
|
|
}
|
|
l := e.maxMatchOff
|
|
if (e.lowMem && e.maxMatchOff > maxCompressedBlockSize) || e.maxMatchOff <= maxCompressedBlockSize {
|
|
l += maxCompressedBlockSize
|
|
} else {
|
|
l += e.maxMatchOff
|
|
}
|
|
// Make it at least 1MB.
|
|
if l < 1<<20 && !e.lowMem {
|
|
l = 1 << 20
|
|
}
|
|
// Make it at least the requested size.
|
|
if l < int32(n) {
|
|
l = int32(n)
|
|
}
|
|
e.hist = make([]byte, 0, l)
|
|
}
|
|
|
|
// useBlock will replace the block with the provided one,
|
|
// but transfer recent offsets from the previous.
|
|
func (e *fastBase) UseBlock(enc *blockEnc) {
|
|
enc.reset(e.blk)
|
|
e.blk = enc
|
|
}
|
|
|
|
func (e *fastBase) matchlenNoHist(s, t int32, src []byte) int32 {
|
|
// Extend the match to be as long as possible.
|
|
return int32(matchLen(src[s:], src[t:]))
|
|
}
|
|
|
|
func (e *fastBase) matchlen(s, t int32, src []byte) int32 {
|
|
if debugAsserts {
|
|
if s < 0 {
|
|
err := fmt.Sprintf("s (%d) < 0", s)
|
|
panic(err)
|
|
}
|
|
if t < 0 {
|
|
err := fmt.Sprintf("s (%d) < 0", s)
|
|
panic(err)
|
|
}
|
|
if s-t > e.maxMatchOff {
|
|
err := fmt.Sprintf("s (%d) - t (%d) > maxMatchOff (%d)", s, t, e.maxMatchOff)
|
|
panic(err)
|
|
}
|
|
if len(src)-int(s) > maxCompressedBlockSize {
|
|
panic(fmt.Sprintf("len(src)-s (%d) > maxCompressedBlockSize (%d)", len(src)-int(s), maxCompressedBlockSize))
|
|
}
|
|
}
|
|
|
|
// Extend the match to be as long as possible.
|
|
return int32(matchLen(src[s:], src[t:]))
|
|
}
|
|
|
|
// Reset the encoding table.
|
|
func (e *fastBase) resetBase(d *dict, singleBlock bool) {
|
|
if e.blk == nil {
|
|
e.blk = &blockEnc{lowMem: e.lowMem}
|
|
e.blk.init()
|
|
} else {
|
|
e.blk.reset(nil)
|
|
}
|
|
e.blk.initNewEncode()
|
|
if e.crc == nil {
|
|
e.crc = xxhash.New()
|
|
} else {
|
|
e.crc.Reset()
|
|
}
|
|
if d != nil {
|
|
low := e.lowMem
|
|
if singleBlock {
|
|
e.lowMem = true
|
|
}
|
|
e.ensureHist(d.DictContentSize() + maxCompressedBlockSize)
|
|
e.lowMem = low
|
|
}
|
|
|
|
// We offset current position so everything will be out of reach.
|
|
// If above reset line, history will be purged.
|
|
if e.cur < bufferReset {
|
|
e.cur += e.maxMatchOff + int32(len(e.hist))
|
|
}
|
|
e.hist = e.hist[:0]
|
|
if d != nil {
|
|
// Set offsets (currently not used)
|
|
for i, off := range d.offsets {
|
|
e.blk.recentOffsets[i] = uint32(off)
|
|
e.blk.prevRecentOffsets[i] = e.blk.recentOffsets[i]
|
|
}
|
|
// Transfer litenc.
|
|
e.blk.dictLitEnc = d.litEnc
|
|
e.hist = append(e.hist, d.content...)
|
|
}
|
|
}
|