Counting the mose frequent char in a file - c

Question

For my CS class I need to write a program that reads an entire file. I've researched a whole bunch of different ways to do this with a string (the two for loops inside the while loops) and I've combined it with the way I was taught to read through a whole file. The problem is you can't index the frequency list with a char variable type (line). Is there an easier way to read through the file and do this?
# define MAX 200
void replace_most_freq(const char *filename, char c, FILE *destination) {
// your code here
FILE *in_file = NULL;
in_file = fopen(filename, "r");
if (!in_file) {
fprintf(destination,
"Error(replace_most_freq): Could not open file %s\n", filename);
fclose(in_file);
return;
}
int i, max = -1, len;
int freq[256] = { 0 };
char line[MAX], result;
while (fgets(line, sizeof(line), in_file)) {
len = strlen(line);
for (i = 0; i < len; i++) {
freq[line[i]]++;
}
}
while (fgets(line, sizeof(line), in_file)) {
len = strlen(line);
for (i = 0; i < len; i++) {
if (max < freq[line[i]]) {
max = freq[line[i]];
result = line[i];
}
}
}
printf("Most frequent char = %c\n", result);
return;
}

Answer 1

Your initial loop is almost correct: you should convert the char to an unsigned char to avoid undefined behavior on negative char values on platforms where char is signed.
The second loop is incorrect: there is no need to read from the file, just iterate over the freq array to find the largest count.
Here is a modified version:
#include <limits.h>
#include <stdio.h>
void replace_most_freq(const char *filename, char newc, FILE *destination) {
FILE *in_file = fopen(filename, "r");
if (!in_file) {
fprintf(stderr,
"Error(replace_most_freq): Could not open file %s\n", filename);
return;
}
int c, max, maxc;
int freq[UCHAR_MAX] = { 0 };
while ((c = getc(in_file)) != EOF) {
freq[c]++;
}
max = freq[maxc = 0];
for (c = 1; c < UCHAR_MAX; c++) {
if (max < freq[c])
max = freq[maxc = c];
}
printf("Most frequent char = %c (%d)\n", max, max);
rewind(in_file);
while ((c = getc(in_file)) != EOF) {
if (c == maxc)
c = newc;
putc(c, destination);
}
}

Answer 2

You can read file in much larger chunks:
#define BUFFSIZE (4*1024*1024)
int findMax(const size_t *, size_t);
int replace_most_freq(const char *filename, char c, FILE *destination) {
int result = 1;
FILE *fi ;
size_t freq[256] = { 0 };
size_t dataChunkLength;
long fileLength;
unsigned char *databuff = malloc(BUFFSIZE);
if(!databuff)
{
result = -2;
goto function_exit;
}
fi = fopen(filename, "r");
if (!fi)
{
result = -1;
goto function_exit;
}
if (fseek(fi, 0, SEEK_END) == -1)
{
result = -3;
goto function_exit;
}
fileLength = ftell(fi);
if (fileLength == -1)
{
result = -4;
goto function_exit;
}
if (fseek(fi, 0, SEEK_SET) == -1)
{
result = -3;
goto function_exit;
}
while(fileLength)
{
if(fileLength <= BUFFSIZE) dataChunkLength = fileLength;
else dataChunkLength = BUFFSIZE;
size_t bytesRead = fread(databuff, 1, dataChunkLength, fi);
if(bytesRead != dataChunkLength)
{
if(feof(fi) || ferror(fi))
{
result = -4;
goto function_exit;
}
}
for(size_t index = 0; index < bytesRead; index++)
{
freq[databuff[index]]++;
}
fileLength -= bytesRead;
}
int mostFrequent;
printf("The most freq char is 0x%02x\n", mostFrequent = findMax(freq, 256));
function_exit:
free(databuff);
if (fi) fclose(fi);
return result;
}