| / This encoding algorithm uses the following procedures: |
| input( buffer + offset, stream, length ) : Data inputting from the character stream; |
| output( length, index ) : Data outputting as a length/index code-word; |
| output_Char( character ) : Data outputting as a single character code-word. |
| / |
| function encodeLZ_MLB( stream ) { / stream is the input stream / |
| buf = array( 0..M-1 ); / buf is the data stream buffer, and Eq (5) is always working / |
| pos = N; / pos traces the current position of the data window (M=2N) / |
| define: s = buf + pos; / s[0..N-1] is the data window, and Eq (6) is always working / |
| link = array( 0..M-1 ); / link stores the matching links for buf / |
| bucket = array( 0..2563-1); / bucket stores the link headers of Eq (3) / |
| link[ 0..M-1 ] = null; bucket[ 0..2563-1 ] = null; / initialize the matching links and link headers / |
| repeat do { / Hint: positions of buf form a congruent group / |
| remain = input( buf + N - pos, stream, N); / remain gets the length of current input data / |
| if ( remain =0 ) do { return; } / this algorithm ends / |
| for n = 0..remain-1 do { / Phase (a): build matching links / |
| i = N – pos + n; j = buf[ i..i+2 ]; p = bucket[ j ]; |
| if not ( p = null ) and not ( buf[ p..p+2 ] = j ) do { bucket[ j ] = null; } / clear the expired header value / |
| if ( bucket[ j ] = null ) do { link[ i ] = null; } |
| else do { link[ i ] = i - bucket[ j ]; } / link stores the distance (i.e. relative position) of Eq (4) / |
| bucket[ j ] = i; / bucket stores the plain position of Eq (3) / |
| } |
| while ( remain >0 ) do { / Phase (b): LZ77 string match and encode / |
| if ( remain <3 ) do { |
| for i = 1..remain do { output_Char( s[N] ); pos = pos+1; } / output the remaining characters / |
| break; / jump out of the “while” cycle / |
| } |
| n = pos + N; dist =0; / dist traces the distance between s[N] and current matching point / |
| len =1; index =0; / len and index trace the current maximum matching length and its relative position / |
| for i = 1..sight do { / sight controls the matching times / |
| d = link[ n - index ]; / d is the distance from current matching point / |
| if ( d = null ) or ( dist + d >= N ) do { break; } / jump out of the “for” cycle / |
| dist = dist + d; x = N; y = N - dist; |
| for ( l = 0..remain-1 ) and ( s[x] = s [y] ) do { x = x+1; y = y+1; } / get the matching length l / |
| if ( l > len ) do { len = l; index = dist; } / find a better matching point / |
| } |
| If ( len >2 ) do { output( len, index ); } / output a length/index code-word / |
| else do { output_Char( s[N] ); len =1; } / output a single character /l |
| pos = pos + len; remain = remain - len; |
| } |
| } |
| } |
