mirror of
https://github.com/deater/dos33fsprogs.git
synced 2024-08-20 06:29:39 +00:00
385 lines
11 KiB
C
385 lines
11 KiB
C
/**************************************************************
|
|
LZSS.C -- A Data Compression Program
|
|
(tab = 4 spaces)
|
|
***************************************************************
|
|
4/6/1989 Haruhiko Okumura
|
|
Use, distribute, and modify this program freely.
|
|
Please send me your improved versions.
|
|
PC-VAN SCIENCE
|
|
NIFTY-Serve PAF01022
|
|
CompuServe 74050,1022
|
|
**************************************************************
|
|
WARNING: order of match_position and match_lenght changed!
|
|
see lines 178 to 182
|
|
Mofication by <stephan.walter@gmx.ch>
|
|
|
|
Also modified to have N,F,etc, etc to be parameters, not
|
|
hard-coded -- vmw
|
|
*/
|
|
|
|
#include <stdio.h>
|
|
#include <stdlib.h>
|
|
#include <string.h>
|
|
#include <ctype.h>
|
|
|
|
|
|
/* output types */
|
|
#define NORMAL_ASM 0
|
|
|
|
#define num_trees 256
|
|
|
|
/* initialize trees */
|
|
static void newInitTree(int ring_buffer_size,int binary_search_index,
|
|
int *rson, int *dad) {
|
|
|
|
int i;
|
|
|
|
/* For i = 0 to N - 1, rson[i] and lson[i] will be the right and
|
|
left children of node i. These nodes need not be initialized.
|
|
Also, dad[i] is the parent of node i. These are initialized to
|
|
NIL (= N), which stands for 'not used.'
|
|
For i = 0 to 255, rson[N + i + 1] is the root of the tree
|
|
for strings that begin with character i. These are initialized
|
|
to NIL. Note there are 256 trees. */
|
|
|
|
for (i=ring_buffer_size+1; i<=ring_buffer_size+num_trees; i++) {
|
|
rson[i] = binary_search_index;
|
|
}
|
|
for (i=0; i<ring_buffer_size; i++) {
|
|
dad[i] = binary_search_index;
|
|
}
|
|
}
|
|
|
|
static void newInsertNode(
|
|
int r, int ring_buffer_size, int binary_search_index,
|
|
int match_length_limit,
|
|
unsigned char *text_buf, int *rson,int *lson, int *dad,
|
|
int *match_length, int *match_position) {
|
|
/* Inserts string of length F, text_buf[r..r+F-1], into one of the
|
|
trees (text_buf[r]'th tree) and returns the longest-match position
|
|
and length via the global variables match_position and match_length.
|
|
If match_length = F, then removes the old node in favor of the new
|
|
one, because the old one will be deleted sooner.
|
|
Note r plays double role, as tree node and position in buffer. */
|
|
|
|
int i, p, cmp;
|
|
unsigned char *key;
|
|
|
|
cmp = 1;
|
|
key = text_buf+r;
|
|
p = ring_buffer_size + 1 + key[0];
|
|
rson[r] = lson[r] = binary_search_index;
|
|
*match_length = 0;
|
|
|
|
for( ; ; ) {
|
|
|
|
if (cmp >= 0) {
|
|
if (rson[p] != binary_search_index) p = rson[p];
|
|
else {
|
|
rson[p] = r;
|
|
dad[r] = p;
|
|
return;
|
|
}
|
|
}
|
|
else {
|
|
if (lson[p] != binary_search_index) p = lson[p];
|
|
else {
|
|
lson[p] = r;
|
|
dad[r] = p;
|
|
return;
|
|
}
|
|
}
|
|
|
|
for(i = 1; i < match_length_limit; i++)
|
|
if ((cmp = key[i] - text_buf[p + i]) != 0) break;
|
|
if (i > *match_length) {
|
|
*match_position = p;
|
|
if ((*match_length = i) >= match_length_limit) break;
|
|
}
|
|
}
|
|
|
|
dad[r] = dad[p];
|
|
lson[r] = lson[p];
|
|
rson[r] = rson[p];
|
|
dad[lson[p]] = r;
|
|
dad[rson[p]] = r;
|
|
|
|
if (rson[dad[p]] == p) rson[dad[p]] = r;
|
|
else lson[dad[p]] = r;
|
|
dad[p] = binary_search_index; /* remove p */
|
|
|
|
}
|
|
|
|
/* deletes node p from tree */
|
|
static void newDeleteNode(int p, int binary_search_index,
|
|
int *dad, int *rson, int *lson) {
|
|
|
|
int q;
|
|
|
|
if (dad[p] == binary_search_index) return; /* not in tree */
|
|
if (rson[p] == binary_search_index) q = lson[p];
|
|
else if (lson[p] == binary_search_index) q = rson[p];
|
|
else {
|
|
q = lson[p];
|
|
if (rson[q] != binary_search_index) {
|
|
do { q = rson[q]; } while (rson[q] != binary_search_index);
|
|
rson[dad[q]] = lson[q];
|
|
dad[lson[q]] = dad[q];
|
|
lson[q] = lson[p];
|
|
dad[lson[p]] = q;
|
|
}
|
|
rson[q] = rson[p]; dad[rson[p]] = q;
|
|
}
|
|
dad[q] = dad[p];
|
|
if (rson[dad[p]] == p) rson[dad[p]] = q;
|
|
else lson[dad[p]] = q;
|
|
dad[p] = binary_search_index;
|
|
}
|
|
|
|
static int lzss_encode_better(
|
|
FILE *infile,
|
|
unsigned char frequent_char,
|
|
int ring_buffer_size, int position_length_threshold,
|
|
int output_type) {
|
|
|
|
// unsigned char frequent_char='#';
|
|
// int ring_buffer_size=1024; /* N */
|
|
int match_length_limit; //=64; /* F */
|
|
/*int position_length_threshold=2; THRESHOLD */
|
|
int binary_search_index=ring_buffer_size; /* NIL */
|
|
int position_bits; //=10;
|
|
// int length_bits=16-position_bits;
|
|
|
|
unsigned long int codesize = 0; /* code size counter */
|
|
|
|
int i, c, len, r, s, last_match_length, code_buf_ptr;
|
|
|
|
unsigned char code_buf[8*2+1], mask;
|
|
|
|
unsigned char *text_buf;
|
|
|
|
int match_position, match_length; /* of longest match. These are
|
|
set by the InsertNode() procedure. */
|
|
int *lson, *rson, *dad; /* left & right children &
|
|
parents -- These constitute
|
|
binary search trees. */
|
|
|
|
/* determine stuff from ring_buffer_size */
|
|
/* fake log2 algorithm */
|
|
|
|
i=1;
|
|
while (((ring_buffer_size-1)>>i) >=1) {
|
|
i++;
|
|
};
|
|
|
|
position_bits=i;
|
|
match_length_limit=1<<(16-position_bits);
|
|
// printf("%i, %i %i %i '%c'\n",ring_buffer_size,position_bits,match_length_limit
|
|
// ,position_length_threshold,frequent_char);
|
|
|
|
/* ring buffer of size N, with extra F-1
|
|
bytes to facilitate string comparison */
|
|
text_buf=calloc(ring_buffer_size+match_length_limit-1,sizeof(unsigned char));
|
|
lson=calloc(ring_buffer_size+1,sizeof(int));
|
|
rson=calloc(ring_buffer_size+num_trees+1,sizeof(int));
|
|
dad=calloc(ring_buffer_size+1,sizeof(int));
|
|
|
|
/* initialize trees */
|
|
newInitTree(ring_buffer_size,binary_search_index,rson,dad);
|
|
|
|
|
|
|
|
/* code_buf[1..16] saves eight units of code, and
|
|
code_buf[0] works as eight flags, "1" representing
|
|
that the unit is an unencoded letter (1 byte),
|
|
"0" a position-and-length pair (2 bytes).
|
|
Thus, eight units require at most 16 bytes of code. */
|
|
|
|
code_buf[0] = 0;
|
|
|
|
code_buf_ptr = mask = 1;
|
|
s = 0;
|
|
r = ring_buffer_size - match_length_limit;
|
|
|
|
if (frequent_char<32) {
|
|
printf(";FREQUENT_CHAR EQU %d\n",frequent_char);
|
|
}
|
|
else {
|
|
printf(";FREQUENT_CHAR EQU '%c'\n",frequent_char);
|
|
}
|
|
printf(";N EQU %i\n",ring_buffer_size);
|
|
printf(";F EQU %i\n",match_length_limit);
|
|
printf(";THRESHOLD EQU %i\n",position_length_threshold);
|
|
printf(";P_BITS EQU %i\n",position_bits);
|
|
printf(";POSITION_MASK EQU %i\n",(0xff>>(8-(position_bits-8))));
|
|
|
|
printf("logo:\n");
|
|
|
|
/* Clear the buffer with any character that will appear often. */
|
|
for(i=0; i<(ring_buffer_size-match_length_limit); i++) {
|
|
text_buf[i]=frequent_char;
|
|
}
|
|
|
|
// printf("%i to %i = %i\n",0,ring_buffer_size-match_length_limit,frequent_char);
|
|
|
|
// printf("%i to %i = ",r,r+match_length_limit);
|
|
for(len=0; len<match_length_limit && (c=getc(infile))!=EOF; len++) {
|
|
/* Read F bytes into the last F bytes of the buffer */
|
|
text_buf[r+len]=c;
|
|
// printf("%i ",text_buf[r+len]);
|
|
}
|
|
// printf("\n");
|
|
if (len== 0) return 0; /* trying to compress empty file */
|
|
|
|
for(i = 1; i <= match_length_limit; i++) {
|
|
newInsertNode(r-i,ring_buffer_size,binary_search_index,
|
|
match_length_limit,text_buf,rson,lson,dad,
|
|
&match_length,&match_position);
|
|
}
|
|
|
|
/* Insert the F strings,
|
|
each of which begins with one or more 'space' characters. Note
|
|
the order in which these strings are inserted. This way,
|
|
degenerate trees will be less likely to occur. */
|
|
newInsertNode(r,ring_buffer_size,binary_search_index,
|
|
match_length_limit,text_buf,rson,lson,dad,
|
|
&match_length,&match_position);
|
|
/* Finally, insert the whole string just read. The
|
|
global variables match_length and match_position are set. */
|
|
do {
|
|
/* match_length may be spuriously long near the end of text. */
|
|
if (match_length > len) match_length = len;
|
|
|
|
if (match_length <= position_length_threshold) {
|
|
match_length=1; /* Not long enough match. Send one byte. */
|
|
code_buf[0] |= mask; /* 'send one byte' flag */
|
|
code_buf[code_buf_ptr++] = text_buf[r]; /* Send uncoded. */
|
|
// printf("single: %i @ %i\n",text_buf[r],r);
|
|
} else {
|
|
// printf("pos : %i\tlen : %i\n",match_position,match_length);
|
|
|
|
code_buf[code_buf_ptr++] = (unsigned char) match_position;
|
|
|
|
code_buf[code_buf_ptr++] = (unsigned char)
|
|
( ((match_position>>8) & (0xff >> (8-(position_bits-8)))) |
|
|
((match_length-(position_length_threshold+1))<<(position_bits-8)) );
|
|
|
|
// code_buf[code_buf_ptr++] = (unsigned char)
|
|
// (((match_position >> 8) & 7) |
|
|
// (match_length - (position_length_threshold+1))<<3);
|
|
}
|
|
if ((mask <<= 1) == 0) { /* Shift mask left one bit. */
|
|
printf("\t.byte\t");
|
|
for(i=0; i<code_buf_ptr; i++) { /* Send at most 8 units of */
|
|
printf("$%02X%c",code_buf[i],(i==code_buf_ptr-1)?'\n':',');
|
|
}
|
|
codesize += code_buf_ptr;
|
|
code_buf[0] = 0; code_buf_ptr = mask = 1;
|
|
}
|
|
|
|
last_match_length = match_length;
|
|
for (i = 0; i < last_match_length && (c = getc(infile)) != EOF; i++) {
|
|
/* Delete old strings and */
|
|
newDeleteNode(s,binary_search_index,dad,rson,lson);
|
|
/* read new bytes */
|
|
text_buf[s] = c;
|
|
|
|
/* If the position is near the end of buffer,
|
|
extend the buffer to make string comparison easier. */
|
|
if (s < match_length_limit - 1) {
|
|
text_buf[s + ring_buffer_size] = c;
|
|
}
|
|
|
|
s = (s + 1) & (ring_buffer_size - 1);
|
|
r = (r + 1) & (ring_buffer_size - 1);
|
|
/* Since this is a ring buffer, increment the position
|
|
modulo N. */
|
|
newInsertNode(r,ring_buffer_size,binary_search_index,
|
|
match_length_limit,text_buf,rson,lson,dad,
|
|
&match_length,&match_position);
|
|
/* Register the string in text_buf[r..r+F-1] */
|
|
}
|
|
|
|
/* After the end of text, */
|
|
while (i++ < last_match_length) {
|
|
|
|
/* no need to read, but */
|
|
newDeleteNode(s,binary_search_index,dad,rson,lson);
|
|
|
|
s = (s + 1) & (ring_buffer_size - 1);
|
|
r = (r + 1) & (ring_buffer_size - 1);
|
|
if (--len) {
|
|
newInsertNode(r,ring_buffer_size,binary_search_index,
|
|
match_length_limit,text_buf,rson,lson,dad,
|
|
&match_length,&match_position);
|
|
}
|
|
/* buffer may not be empty. */
|
|
}
|
|
/* until length of string to be processed is zero */
|
|
} while (len > 0);
|
|
|
|
|
|
/* Send remaining code. */
|
|
if (code_buf_ptr > 1) {
|
|
|
|
|
|
printf("\t.byte\t");
|
|
|
|
for(i=0; i<code_buf_ptr; i++) {
|
|
printf("$%02x",code_buf[i]);
|
|
|
|
if (i==code_buf_ptr-1) {
|
|
printf("\n");
|
|
} else {
|
|
printf(",");
|
|
}
|
|
}
|
|
codesize += code_buf_ptr;
|
|
}
|
|
|
|
printf("logo_end:\n");
|
|
|
|
free(text_buf);
|
|
free(lson);
|
|
free(rson);
|
|
free(dad);
|
|
return codesize;
|
|
}
|
|
|
|
int main(int argc, char **argv) {
|
|
|
|
FILE *input;
|
|
int ch,byte_count=0,orig_byte_count=0;
|
|
|
|
/* Check command line arguments */
|
|
if (argc<2) {
|
|
printf("\nUsage:\n\t%s file_to_compress\n\n",argv[0]);
|
|
return 3;
|
|
}
|
|
|
|
/* Open input file */
|
|
input=fopen(argv[1],"r");
|
|
if (input==NULL) {
|
|
printf("Could not open \"%s\"\n",argv[1]);
|
|
return 2;
|
|
}
|
|
|
|
byte_count=lzss_encode_better(input,'\0',1024,2,NORMAL_ASM);
|
|
|
|
rewind(input);
|
|
|
|
/* calculate size of original */
|
|
/* Why aren't we using stat()? */
|
|
while((ch=fgetc(input))!=EOF) {
|
|
orig_byte_count++;
|
|
}
|
|
|
|
fclose(input);
|
|
|
|
printf("; Size of original version: %i\n",orig_byte_count);
|
|
printf("; Size of compressed version: %i\n",byte_count);
|
|
|
|
return 0;
|
|
|
|
}
|