File Compression and multi-dimensional arrays

Question

Compression Algorithm:

One method of compressing a file is to recognize repeated digits and replace the runs of

repetition with a number representing the repetition length. For instance for a line in you file

such as ;

0110000000000110

Can be represented with the line:

1 2 10 2 1

Because there is 1 zero, 2 ones, 10 zeros, 2 ones and 1 zero. It is already understood that only

zeros and ones appear and that zeros apper first in a given line. Thus storing the new line of

data is much shorter (5 numbers) than the original line (16 numbers)

To decompress the encoded file you can follow the reverse implementation of the method

stated above, for example for the encoded data:

4 2 4 2 4

The decompressed data is:

0000110000110000

Answer 1

using namespace std;

#include <iostream>

#include <string.h>

#define WORD_LEN (16)

#define MAX_LINES (1024)

typedef struct _TWord

{

char bitstr[WORD_LEN+1];

int intstr[WORD_LEN];

int iCount;

int startingBit;

}*TWord;

#define WORD_SIZE (sizeof(struct _TWord))

#define TWORD_SIZE (sizeof(TWord))

int bits2Ints(TWord word)

{

int bitStreak=1;

word->startingBit = (word->bitstr[0]=='1') ? 1 : 0;

word->iCount=0;

for (int iLoop=0; iLoop<WORD_LEN-1; iLoop++)

{

if ( word->bitstr[iLoop]!=word->bitstr[iLoop+1])

{

word->intstr[word->iCount++] = bitStreak;

bitStreak=1;

}

else

{

bitStreak++;

}

}

word->intstr[word->iCount++] = bitStreak;

}

ints2Bits(TWord word)

{

char curBit = (word->startingBit==1) ? '1' : '0';

int iIndexPos=-1;

for (int iLoop=0; iLoop<word->iCount; iLoop++)

{

for (int jLoop=0; jLoop<word->intstr[iLoop]; jLoop++)

{

word->bitstr[++iIndexPos]=curBit;

}

if (curBit=='1')

{

curBit='0';

}

else

{

curBit='1';

}

}

word->bitstr[WORD_LEN]=0;

}

DumpWord(TWord word)

{

cout << word->bitstr << endl;

for (int iLoop=0; iLoop<word->iCount; iLoop++)

{

cout << word->intstr[iLoop] << endl;

}

cout << " starting bit " << word->startingBit << endl;

}

int Compress ( char * input_filename, char * output_filename )

{

char A[MAX_LINES][WORD_LEN];

struct _TWord word[MAX_LINES];

int iLineCount=0;

int iLoop;

int wordCount =0;

FILE * infile = fopen(input_filename,"r");

FILE * outfile = fopen(output_filename,"w");

for (iLoop=0; iLoop<MAX_LINES; iLoop++)

{

if (fscanf(infile,"%s",A[iLoop])== EOF)

{

cout << A[iLoop] << endl;

break;

}

cout << A[iLoop] << endl;

}

iLineCount = iLoop;

cout << iLineCount << " lines compressed " << endl;

for (iLoop=0; iLoop<iLineCount; iLoop++)

{

strcpy(word[iLoop].bitstr,A[iLoop]);

bits2Ints(&word[iLoop]); //compression takes place here

DumpWord(&word[iLoop]);

fprintf(outfile,"%d %d ",word[iLoop].startingBit,word[iLoop].iCount);

for (int jLoop=0; jLoop<word[iLoop].iCount; jLoop++)

{

fprintf(outfile," %d ",word[iLoop].intstr[jLoop]);

}

fprintf(outfile,"\n");

}

fclose(infile);

fclose(outfile);

return(iLoop);

}

int Decompress( char * input_filename, char * output_filename)

{

struct _TWord word[MAX_LINES];

int iLineCount=0;

int iLoop;

int wordCount =0;

FILE * infile = fopen(input_filename,"r");

FILE * outfile = fopen(output_filename,"w");

for (iLoop=0; iLoop<MAX_LINES; iLoop++)

{

if (fscanf(infile,"%d %d",&word[iLoop].startingBit, &word[iLoop].iCount)!=EOF)

{

for (int jLoop=0; jLoop<word[iLoop].iCount; jLoop++)

{

fscanf(infile,"%d",&word[iLoop].intstr[jLoop]);

}

ints2Bits(&word[iLoop]);

fprintf(outfile,"%s\n",word[iLoop].bitstr);

}

else

{

break;

}

}

fclose(infile);

fclose(outfile);

return(iLoop);

}

int main()

{

Compress( (char*)"data.txt",(char*)"compressed.txt");

Decompress ((char*) "compressed.txt",(char *) "copydata.txt ");

}