Need help regarding lzss compression in C - c

Question

I had used this code from https://oku.edu.mie-u.ac.jp/~okumura/compression/lzss.c
This code was for file compression. I had modified it for a given string. For example :
d(2306):AuthorisationScheme:RADIUSserveratfd04:bd3:80e8:1::1usingPAPD/6LoWPANd(2306):WritingModule:SecurityConfigD/6LoWPANd(2306):WritingModule:RunCoordinatorD/6LoWPANd(2306):RequestingmoduleaddressD/6LoWPANd(2306):WritingModule:GetAddressD/smartcard-jni(2781):SmartCard_state_update_callback:status=6D/SmartCardNative(2781):status=6D/smartcard(2781):PN532Smartcard_loop_threadexitD/smartcard-jni(2781):SmartCard_loop_thread:SMARTCARD_STATUS_ABORTD/smartcard(2781):Smartcard_loop_uninitD/smartcard(2781):2D/serialcomm_pn532(2781):PN532:Readingfrom/dev/ttyUSB0-->D/smartcard(2781):Received(0x3)fromPN532:(dataBuf[0]:0x1)(0x1f5988)D/smartcard(2781):ReceivedStatusfromPN532:OK(cmd:0x2)D/smartcard-jni(2781):SmartCard_listener_update_callbackD/smartcard(2781):Received(0x1c2)fromPN532:(dataBuf[0]:0x32)(0x1f5988)D/smartcard(2781):vd(2306):AuthorisationScheme:RADIUSserveratfd04:bd3:80e8:1::1usingPAPD/6LoWPANd(2306):
The problem I am facing is, if I want to compress a file containing this , the code is able to do that. But the modified code that I had done having little bit of problem. The last 2 bits of compressed files are different. Though the content of file and the buffer both are exactly same.
The original code was reading from file. Here, I provided the string of same content.
The code goes here :
/* LZSS encoder-decoder (c) Haruhiko Okumura */
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#define EI 11 /* typically 10..13 */
#define EJ 4 /* typically 4..5 */
#define P 1 /* If match length <= P then output one character */
#define N (1 << EI) /* buffer size */
#define F ((1 << EJ) + P) /* lookahead buffer size */
int bit_buffer = 0, bit_mask = 128;
unsigned long codecount = 0, textcount = 0;
unsigned char buffer[N * 2];
FILE *infile, *outfile, *outfile2, *outfile3;
/*----*/
unsigned long payload_i = 0;
int tt = 0;
unsigned int buf_load[16000];
unsigned char *string_buf= "d(2306):AuthorisationScheme:RADIUSserveratfd04:bd3:80e8:1::1usingPAPD/6LoWPANd(2306):WritingModule:SecurityConfigD/6LoWPANd(2306):WritingModule:RunCoordinatorD/6LoWPANd(2306):RequestingmoduleaddressD/6LoWPANd(2306):WritingModule:GetAddressD/smartcard-jni(2781):SmartCard_state_update_callback:status=6D/SmartCardNative(2781):status=6D/smartcard(2781):PN532Smartcard_loop_threadexitD/smartcard-jni(2781):SmartCard_loop_thread:SMARTCARD_STATUS_ABORTD/smartcard(2781):Smartcard_loop_uninitD/smartcard(2781):2D/serialcomm_pn532(2781):PN532:Readingfrom/dev/ttyUSB0-->D/smartcard(2781):Received(0x3)fromPN532:(dataBuf[0]:0x1)(0x1f5988)D/smartcard(2781):ReceivedStatusfromPN532:OK(cmd:0x2)D/smartcard-jni(2781):SmartCard_listener_update_callbackD/smartcard(2781):Received(0x1c2)fromPN532:(dataBuf[0]:0x32)(0x1f5988)D/smartcard(2781):vd(2306):AuthorisationScheme:RADIUSserveratfd04:bd3:80e8:1::1usingPAPD/6LoWPANd(2306):";
/*----*/
void error(void)
{
printf("Output error\n"); exit(1);
}
void putbit1(void)
{
outfile2 = fopen("file2.lzss", "a");
bit_buffer |= bit_mask;
if ((bit_mask >>= 1) == 0) {
/*----*/
buf_load[payload_i] = bit_buffer;
if (fputc(buf_load[payload_i], outfile3) == EOF) error();
payload_i++;
/*----*/
if (fputc(bit_buffer, outfile2) == EOF) error();
if (fputc(bit_buffer, outfile) == EOF) error();
bit_buffer = 0; bit_mask = 128; codecount++;
}
fclose(outfile2);
}
void putbit0(void)
{
outfile2 = fopen("file2.lzss", "a");
if ((bit_mask >>= 1) == 0) {
/*----*/
buf_load[payload_i] = bit_buffer;
if (fputc(buf_load[payload_i], outfile3) == EOF) error();
payload_i++;
/*----*/
if (fputc(bit_buffer, outfile2) == EOF) error();
if (fputc(bit_buffer, outfile) == EOF) error();
bit_buffer = 0; bit_mask = 128; codecount++;
}
fclose(outfile2);
}
void flush_bit_buffer(void)
{
outfile2 = fopen("file2.lzss", "a");
if (bit_mask != 128) {
if (fputc(buf_load[payload_i], outfile3) == EOF) error();
if (fputc(bit_buffer, outfile2) == EOF) error();
if (fputc(bit_buffer, outfile) == EOF) error();
codecount++;
}
fclose(outfile2);
}
void output1(int c)
{
int mask;
putbit1();
mask = 256;
while (mask >>= 1) {
if (c & mask) putbit1();
else putbit0();
}
}
void output2(int x, int y)
{
int mask;
putbit0();
mask = N;
while (mask >>= 1) {
if (x & mask) putbit1();
else putbit0();
}
mask = (1 << EJ);
while (mask >>= 1) {
if (y & mask) putbit1();
else putbit0();
}
}
void encode(void)
{
int i, j, f1, x, y, r, s, bufferend, c;
for (i = 0; i < N - F; i++) buffer[i] = ' ';
for (i = N - F; i < N * 2; i++) {
if ((c = fgetc(infile)) == EOF) break;
if((c = string_buf[tt++]) == '\0') break;
buffer[i] = c; textcount++; //tt++;
}
bufferend = i; r = N - F; s = 0;
while (r < bufferend) {
f1 = (F <= bufferend - r) ? F : bufferend - r;
x = 0; y = 1; c = buffer[r];
for (i = r - 1; i >= s; i--)
if (buffer[i] == c) {
for (j = 1; j < f1; j++)
if (buffer[i + j] != buffer[r + j]) break;
if (j > y) {
x = i; y = j;
}
}
if (y <= P) output1(c);
else output2(x & (N - 1), y - 2);
r += y; s += y;
if (r >= N * 2 - F) {
for (i = 0; i < N; i++) buffer[i] = buffer[i + N];
bufferend -= N; r -= N; s -= N;
while (bufferend < N * 2) {
if ((c = fgetc(infile)) == EOF) break;
if((c = string_buf[tt++]) == '\0') break;
//tt++;
buffer[bufferend++] = c; textcount++;
}
}
}
flush_bit_buffer();
printf("text: %ld bytes\n", textcount);
printf("code: %ld bytes (%ld%%)\n",
codecount, (codecount * 100) / textcount);
}
int getbit(int n) /* get n bits */
{
int i, x;
static int buf, mask = 0;
x = 0;
for (i = 0; i < n; i++) {
if (mask == 0) {
if ((buf = fgetc(infile)) == EOF) return EOF;
mask = 128;
}
x <<= 1;
if (buf & mask) x++;
mask >>= 1;
}
return x;
}
void decode(void)
{
int i, j, k, r, c;
for (i = 0; i < N - F; i++) buffer[i] = ' ';
r = N - F;
while ((c = getbit(1)) != EOF) {
if (c) {
if ((c = getbit(8)) == EOF) break;
fputc(c, outfile);
buffer[r++] = c; r &= (N - 1);
} else {
if ((i = getbit(EI)) == EOF) break;
if ((j = getbit(EJ)) == EOF) break;
for (k = 0; k <= j + 1; k++) {
c = buffer[(i + k) & (N - 1)];
fputc(c, outfile);
buffer[r++] = c; r &= (N - 1);
}
}
}
}
int main(int argc, char *argv[])
{
int enc;
char *s;
memset(buf_load, '\0', sizeof(buf_load));
outfile3 = fopen ("file1.lzss", "wb");
if (argc != 4) {
printf("Usage: lzss e/d infile outfile\n\te = encode\td = decode\n");
return 1;
}
s = argv[1];
if (s[1] == 0 && (*s == 'd' || *s == 'D' || *s == 'e' || *s == 'E'))
enc = (*s == 'e' || *s == 'E');
else {
printf("? %s\n", s); return 1;
}
if ((infile = fopen(argv[2], "rb")) == NULL) {
printf("? %s\n", argv[2]); return 1;
}
if ((outfile = fopen(argv[3], "wb")) == NULL) {
printf("? %s\n", argv[3]); return 1;
}
if (enc) encode(); else decode();
fclose(infile); fclose(outfile);
fclose(outfile3);
return 0;
}
Looks like the issue lies in buffer reading-writing vs file read write. In file read-write the pointer is increased to next mem loc and reading and writing the same way. In string, each array is being read by incrementing index and then the compressed value is written in file in file write. In file, it is read as binary (wb), In string it is being read as array element. Can there be a problem for that ? Need expert's advice on that.

Answer 1

void flush_bit_buffer(void)
{
outfile2 = fopen("file2.lzss", "a");
if (bit_mask != 128) {
flush_bit[payload_i] = bit_buffer; /*This last bit should be put which I missed. It solved my issue*/
if (fputc(buf_load[payload_i], outfile3) == EOF) error();
if (fputc(bit_buffer, outfile2) == EOF) error();
if (fputc(bit_buffer, outfile) == EOF) error();
codecount++;
}
fclose(outfile2);
}
I solved it by adding the line in the code snippet. I needed to add last bit of bit buffer which solved the problem.