...

Source file src/github.com/klauspost/compress/flate/level3.go

Documentation: github.com/klauspost/compress/flate

     1  package flate
     2  
     3  import "fmt"
     4  
     5  // fastEncL3
     6  type fastEncL3 struct {
     7  	fastGen
     8  	table [1 << 16]tableEntryPrev
     9  }
    10  
    11  // Encode uses a similar algorithm to level 2, will check up to two candidates.
    12  func (e *fastEncL3) Encode(dst *tokens, src []byte) {
    13  	const (
    14  		inputMargin            = 12 - 1
    15  		minNonLiteralBlockSize = 1 + 1 + inputMargin
    16  		tableBits              = 16
    17  		tableSize              = 1 << tableBits
    18  		hashBytes              = 5
    19  	)
    20  
    21  	if debugDeflate && e.cur < 0 {
    22  		panic(fmt.Sprint("e.cur < 0: ", e.cur))
    23  	}
    24  
    25  	// Protect against e.cur wraparound.
    26  	for e.cur >= bufferReset {
    27  		if len(e.hist) == 0 {
    28  			for i := range e.table[:] {
    29  				e.table[i] = tableEntryPrev{}
    30  			}
    31  			e.cur = maxMatchOffset
    32  			break
    33  		}
    34  		// Shift down everything in the table that isn't already too far away.
    35  		minOff := e.cur + int32(len(e.hist)) - maxMatchOffset
    36  		for i := range e.table[:] {
    37  			v := e.table[i]
    38  			if v.Cur.offset <= minOff {
    39  				v.Cur.offset = 0
    40  			} else {
    41  				v.Cur.offset = v.Cur.offset - e.cur + maxMatchOffset
    42  			}
    43  			if v.Prev.offset <= minOff {
    44  				v.Prev.offset = 0
    45  			} else {
    46  				v.Prev.offset = v.Prev.offset - e.cur + maxMatchOffset
    47  			}
    48  			e.table[i] = v
    49  		}
    50  		e.cur = maxMatchOffset
    51  	}
    52  
    53  	s := e.addBlock(src)
    54  
    55  	// Skip if too small.
    56  	if len(src) < minNonLiteralBlockSize {
    57  		// We do not fill the token table.
    58  		// This will be picked up by caller.
    59  		dst.n = uint16(len(src))
    60  		return
    61  	}
    62  
    63  	// Override src
    64  	src = e.hist
    65  	nextEmit := s
    66  
    67  	// sLimit is when to stop looking for offset/length copies. The inputMargin
    68  	// lets us use a fast path for emitLiteral in the main loop, while we are
    69  	// looking for copies.
    70  	sLimit := int32(len(src) - inputMargin)
    71  
    72  	// nextEmit is where in src the next emitLiteral should start from.
    73  	cv := load6432(src, s)
    74  	for {
    75  		const skipLog = 7
    76  		nextS := s
    77  		var candidate tableEntry
    78  		for {
    79  			nextHash := hashLen(cv, tableBits, hashBytes)
    80  			s = nextS
    81  			nextS = s + 1 + (s-nextEmit)>>skipLog
    82  			if nextS > sLimit {
    83  				goto emitRemainder
    84  			}
    85  			candidates := e.table[nextHash]
    86  			now := load6432(src, nextS)
    87  
    88  			// Safe offset distance until s + 4...
    89  			minOffset := e.cur + s - (maxMatchOffset - 4)
    90  			e.table[nextHash] = tableEntryPrev{Prev: candidates.Cur, Cur: tableEntry{offset: s + e.cur}}
    91  
    92  			// Check both candidates
    93  			candidate = candidates.Cur
    94  			if candidate.offset < minOffset {
    95  				cv = now
    96  				// Previous will also be invalid, we have nothing.
    97  				continue
    98  			}
    99  
   100  			if uint32(cv) == load3232(src, candidate.offset-e.cur) {
   101  				if candidates.Prev.offset < minOffset || uint32(cv) != load3232(src, candidates.Prev.offset-e.cur) {
   102  					break
   103  				}
   104  				// Both match and are valid, pick longest.
   105  				offset := s - (candidate.offset - e.cur)
   106  				o2 := s - (candidates.Prev.offset - e.cur)
   107  				l1, l2 := matchLen(src[s+4:], src[s-offset+4:]), matchLen(src[s+4:], src[s-o2+4:])
   108  				if l2 > l1 {
   109  					candidate = candidates.Prev
   110  				}
   111  				break
   112  			} else {
   113  				// We only check if value mismatches.
   114  				// Offset will always be invalid in other cases.
   115  				candidate = candidates.Prev
   116  				if candidate.offset > minOffset && uint32(cv) == load3232(src, candidate.offset-e.cur) {
   117  					break
   118  				}
   119  			}
   120  			cv = now
   121  		}
   122  
   123  		// Call emitCopy, and then see if another emitCopy could be our next
   124  		// move. Repeat until we find no match for the input immediately after
   125  		// what was consumed by the last emitCopy call.
   126  		//
   127  		// If we exit this loop normally then we need to call emitLiteral next,
   128  		// though we don't yet know how big the literal will be. We handle that
   129  		// by proceeding to the next iteration of the main loop. We also can
   130  		// exit this loop via goto if we get close to exhausting the input.
   131  		for {
   132  			// Invariant: we have a 4-byte match at s, and no need to emit any
   133  			// literal bytes prior to s.
   134  
   135  			// Extend the 4-byte match as long as possible.
   136  			//
   137  			t := candidate.offset - e.cur
   138  			l := e.matchlenLong(s+4, t+4, src) + 4
   139  
   140  			// Extend backwards
   141  			for t > 0 && s > nextEmit && src[t-1] == src[s-1] {
   142  				s--
   143  				t--
   144  				l++
   145  			}
   146  			if nextEmit < s {
   147  				if false {
   148  					emitLiteral(dst, src[nextEmit:s])
   149  				} else {
   150  					for _, v := range src[nextEmit:s] {
   151  						dst.tokens[dst.n] = token(v)
   152  						dst.litHist[v]++
   153  						dst.n++
   154  					}
   155  				}
   156  			}
   157  
   158  			dst.AddMatchLong(l, uint32(s-t-baseMatchOffset))
   159  			s += l
   160  			nextEmit = s
   161  			if nextS >= s {
   162  				s = nextS + 1
   163  			}
   164  
   165  			if s >= sLimit {
   166  				t += l
   167  				// Index first pair after match end.
   168  				if int(t+8) < len(src) && t > 0 {
   169  					cv = load6432(src, t)
   170  					nextHash := hashLen(cv, tableBits, hashBytes)
   171  					e.table[nextHash] = tableEntryPrev{
   172  						Prev: e.table[nextHash].Cur,
   173  						Cur:  tableEntry{offset: e.cur + t},
   174  					}
   175  				}
   176  				goto emitRemainder
   177  			}
   178  
   179  			// Store every 5th hash in-between.
   180  			for i := s - l + 2; i < s-5; i += 6 {
   181  				nextHash := hashLen(load6432(src, i), tableBits, hashBytes)
   182  				e.table[nextHash] = tableEntryPrev{
   183  					Prev: e.table[nextHash].Cur,
   184  					Cur:  tableEntry{offset: e.cur + i}}
   185  			}
   186  			// We could immediately start working at s now, but to improve
   187  			// compression we first update the hash table at s-2 to s.
   188  			x := load6432(src, s-2)
   189  			prevHash := hashLen(x, tableBits, hashBytes)
   190  
   191  			e.table[prevHash] = tableEntryPrev{
   192  				Prev: e.table[prevHash].Cur,
   193  				Cur:  tableEntry{offset: e.cur + s - 2},
   194  			}
   195  			x >>= 8
   196  			prevHash = hashLen(x, tableBits, hashBytes)
   197  
   198  			e.table[prevHash] = tableEntryPrev{
   199  				Prev: e.table[prevHash].Cur,
   200  				Cur:  tableEntry{offset: e.cur + s - 1},
   201  			}
   202  			x >>= 8
   203  			currHash := hashLen(x, tableBits, hashBytes)
   204  			candidates := e.table[currHash]
   205  			cv = x
   206  			e.table[currHash] = tableEntryPrev{
   207  				Prev: candidates.Cur,
   208  				Cur:  tableEntry{offset: s + e.cur},
   209  			}
   210  
   211  			// Check both candidates
   212  			candidate = candidates.Cur
   213  			minOffset := e.cur + s - (maxMatchOffset - 4)
   214  
   215  			if candidate.offset > minOffset {
   216  				if uint32(cv) == load3232(src, candidate.offset-e.cur) {
   217  					// Found a match...
   218  					continue
   219  				}
   220  				candidate = candidates.Prev
   221  				if candidate.offset > minOffset && uint32(cv) == load3232(src, candidate.offset-e.cur) {
   222  					// Match at prev...
   223  					continue
   224  				}
   225  			}
   226  			cv = x >> 8
   227  			s++
   228  			break
   229  		}
   230  	}
   231  
   232  emitRemainder:
   233  	if int(nextEmit) < len(src) {
   234  		// If nothing was added, don't encode literals.
   235  		if dst.n == 0 {
   236  			return
   237  		}
   238  
   239  		emitLiteral(dst, src[nextEmit:])
   240  	}
   241  }
   242  

View as plain text