how to compress file in c code example

Example 1: how to compress a file in c

#include <stdint.h>

uint32_t lz77_compress (uint8_t *uncompressed_text, uint32_t uncompressed_size, uint8_t *compressed_text)
{
    uint8_t pointer_length, temp_pointer_length;
    uint16_t pointer_pos, temp_pointer_pos, output_pointer;
    uint32_t compressed_pointer, output_size, coding_pos, output_lookahead_ref, look_behind, look_ahead;
    
    *((uint32_t *) compressed_text) = uncompressed_size;
    compressed_pointer = output_size = 4;
    
    for(coding_pos = 0; coding_pos < uncompressed_size; ++coding_pos)
    {
        pointer_pos = 0;
        pointer_length = 0;
        for(temp_pointer_pos = 1; (temp_pointer_pos < 4096) && (temp_pointer_pos <= coding_pos); ++temp_pointer_pos)
        {
            look_behind = coding_pos - temp_pointer_pos;
            look_ahead = coding_pos;
            for(temp_pointer_length = 0; uncompressed_text[look_ahead++] == uncompressed_text[look_behind++]; ++temp_pointer_length)
                if(temp_pointer_length == 15)
                    break;
            if(temp_pointer_length > pointer_length)
            {
                pointer_pos = temp_pointer_pos;
                pointer_length = temp_pointer_length;
                if(pointer_length == 15)
                    break;
            }
        }
        coding_pos += pointer_length;
        if(pointer_length && (coding_pos == uncompressed_size))
        {
            output_pointer = (pointer_pos << 4) | (pointer_length - 1);
            output_lookahead_ref = coding_pos - 1;
        }
        else
        {
            output_pointer = (pointer_pos << 4) | pointer_length;
            output_lookahead_ref = coding_pos;
        }
        *((uint32_t *) (compressed_text + compressed_pointer)) = output_pointer;
        compressed_pointer += 2;
        *(compressed_text + compressed_pointer++) = *(uncompressed_text + output_lookahead_ref);
        output_size += 3;
    }
    
    return output_size;
}

uint32_t lz77_decompress (uint8_t *compressed_text, uint8_t *uncompressed_text)
{
    uint8_t pointer_length;
    uint16_t input_pointer, pointer_pos;
    uint32_t compressed_pointer, coding_pos, pointer_offset, uncompressed_size;
    
    uncompressed_size = *((uint32_t *) compressed_text);
    compressed_pointer = 4;
    
    for(coding_pos = 0; coding_pos < uncompressed_size; ++coding_pos)
    {
        input_pointer = *((uint32_t *) (compressed_text + compressed_pointer));
        compressed_pointer += 2;
        pointer_pos = input_pointer >> 4;
        pointer_length = input_pointer & 15;
        if(pointer_pos)
            for(pointer_offset = coding_pos - pointer_pos; pointer_length > 0; --pointer_length)
                uncompressed_text[coding_pos++] = uncompressed_text[pointer_offset++];
        *(uncompressed_text + coding_pos) = *(compressed_text + compressed_pointer++);
    }
    
    return coding_pos;
}

Example 2: how to compress a file in c

#include <stdio.h>
#include <string.h>

//comp()
void Compress(unsigned char *szOut, const char *szMessage) {
	unsigned long long nBuffer = 0; //This is enough to store 8 uncompressed characters or 9 compressed. We will only use 8 tho.
	char nBitsInBuffer = 0;
	while (*szMessage != 0) {
		nBuffer |= (unsigned long long)(*szMessage++ & 0x7F) << nBitsInBuffer;
		nBitsInBuffer += 7;
		if (nBitsInBuffer == 7 * 8) { //We have 8 chars in the buffer, dump them
			while (nBitsInBuffer > 0) {
				*szOut++ = nBuffer & 0xFF;
				nBuffer >>= 8;
				nBitsInBuffer -= 8;
			}
			//The following should be redundant, but just to be safe
			nBitsInBuffer = 0;
			nBuffer = 0;
		}
	}
	//Write out whatever is left in the buffer
	while (nBitsInBuffer > 0) {
		*szOut++ = nBuffer & 0xFF;
		nBuffer >>= 8;
		nBitsInBuffer -= 8;
	}
}

//uncomp()
//nCompressedLen is the number of bytes in the compressed buffer.
//NOTE: the compressed buffer does not have a NULL terminating character
void Uncompress(char *szOut, const unsigned char *szCompressed, unsigned nCompressedLen) {
	unsigned long long nBuffer = 0; //This is enough to store 8 uncompressed characters or 9 compressed. We will only use 8 tho.
	char nBitsInBuffer = 0;
	while (nCompressedLen) {
		while (nCompressedLen && nBitsInBuffer < 7 * 8) {
			nBuffer |= (unsigned long long)*szCompressed++ << nBitsInBuffer;
			nBitsInBuffer += 8;
			--nCompressedLen;
		}
		while (nBitsInBuffer > 0) {
			*szOut++ = nBuffer & 0x7F;
			nBuffer >>= 7;
			nBitsInBuffer -= 7;
		}
		//The following should be redundant, but just to be safe
		nBitsInBuffer = 0;
		nBuffer = 0;
	}
}

int main() {
	//char szMessage[] = "\x53\x7F\x63\x4B";
	char szMessage[] = "hello world. this is a compressed long string";
	static const unsigned nCompressedSize = sizeof(szMessage) * 7 / 8; //This does not include the NULL terminating character from the string
	unsigned char pCompressedBytes[nCompressedSize];
	char szUncompressed[sizeof(szMessage)];
	printf("     Message: %s\n", szMessage);
	Compress(pCompressedBytes, szMessage);
	printf("  Compressed: ");
	for (int nByte = 0; nByte < nCompressedSize; ++nByte) {
		printf("%02X ", pCompressedBytes[nByte]);
	}
	printf("\n");
	Uncompress(szUncompressed, pCompressedBytes, nCompressedSize);
	szUncompressed[sizeof(szMessage) - 1] = 0; //We need to terminate the string. The NULL terminator is not stored in the compressed bytes
	printf("Uncompressed: %s\n", szUncompressed);
	//Now just verify that we ended up with the same message we started with
	if (strcmp(szMessage, szUncompressed) == 0) {
		printf("Compression works\n");
	} else {
		printf("Compression failed\n");
	}
	return 0;
}

Tags:

C Example