Reading a little endian float on a big endian machine

Hi,

I have a problem reading floats that have been written to a binary file on a little endian machine (x86). When I try to read the file on a big endian machine (PPC), the floats get read in as 0.0000.

Integers get read in correctly and I just need to switch the endian to get the correct value. Here is the code that I'm using

      float v[5];
      int w[5];
      
      FILE *fp = fopen("test.dat", "rb");
      
      if (fp == NULL)
      {
            fp = fopen("test.dat", "wb");
            
            v[0] = 1.2f;
            v[1] = 2.3f;
            v[2] = 3.4f;
            v[3] = 4.5f;
            v[4] = 5.6f;
            
            w[0] = 10;
            w[1] = 20;
            w[2] = 30;
            w[3] = 40;
            w[4] = 50;
            
            fwrite(v, sizeof(float), 5, fp);
            fwrite(w, sizeof(int), 5, fp);
            
            fclose(fp);
      }
      
      else
      {
            fp = fopen("test.dat", "rb");
            
            fread(v, sizeof(float), 5, fp);
            fread(w, sizeof(int), 5, fp);
            
            printf("Got %f %f %f %f %f\n", v[0], v[1], v[2], v[3], v[4]);
            printf("Got %d %d %d %d %d\n", w[0], w[1], w[2], w[3], w[4]);
      }
      
      exit(0);

The first if statement just writes the file if it doesn't exist. The output of the program when run on 68k or PPC is

Got -0.00000 0.00000 -0.00000 0.00000 0.00000
Got 167772160 335544320 503316480 671088640 838860800

Does anyone know what I can do to get the floats to read in correctly so that I can then swap the endian?

Thanks

RichardStart Free Trial