/* entropy.c - Entropy calculator - Don Yang (uguu.org)

   02/17/03
                                                                           */

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

#define BUFSIZE      4096


int main(int argc, char **argv)
{
   static unsigned char buffer[BUFSIZE];
   static unsigned int freq[256];
   double p, entropy, log2;
   FILE *infile;
   size_t i, size, filesize;
   int a;

   if( argc < 2 )
      return printf("%s <filename> [...]\n", *argv);

   log2 = log(2);
   for(a = 1; a < argc; a++)
   {
      printf("%s: ", argv[a]);
      if( (infile = fopen(argv[a], "rb")) == NULL )
      {
         (void)puts("fopen() failed");
         continue;
      }

      memset(freq, 0, 256 * sizeof(int));
      filesize = 0;
      while( feof(infile) == 0 )
      {
         if( (size = fread(buffer, 1, BUFSIZE, infile)) == 0 )
            break;
         if( size > BUFSIZE )
            break;
         for(i = 0; i < size; freq[(unsigned int)(buffer[i++])]++);
         filesize += size;
      }

      (void)fclose(infile);
      entropy = 0.0;
      for(i = 0; i < 256; i++)
      {
         if( freq[i] > 0 )
         {
            p = (double)(freq[i]) / (double)filesize;
            entropy -= p * log(p) / log2;
         }
      }
      printf("%g bits\n", entropy);
   }

   return 0;
}