types - calculator error in C -

i'm trying make calculator program in c. i've done of think...just output incorrect. output supposed come out doesn't. below code calculator program!.

#include <stdio.h> #include <string.h> #include <stdlib.h>  void add(double num1,double num2); void del(double num1,double num2); void mul(double num1,double num2); void divide(double num1,double num2); main(){     file *fp;     char sym;     float num1, num2;     int ret;     fp = fopen("input.txt","r");     if(fp==null){         exit(1);     }     while(fscanf(fp,"%f%c%f", &num1,&sym,&num2)!=eof){         switch(sym){             case '+': add(num1,num2);                     break;             case '-': del(num1,num2);                     break;             case '*': mul(num1,num2);                     break;             case '/': divide(num1,num2);                     break;             default: printf("%f%c%f", num1,sym,num2);         }     }     fclose(fp); }  void add(double num1,double num2){     double result;     result = num1+num2;     printf("%f\n", result); }  void del(double num1,double num2){     double result;     result = num1-num2;     printf("%f\n", result); }  void mul(double num1,double num2){     double result;     result = num1*num2;     printf("%f\n", result); }  void divide(double num1,double num2){     double result;     result = num1/num2;     printf("%f\n", result); } 

when have inputs of:

123456789+987654321 123456789-987654321 12345*54321 1349.238912+12384.12871 3918.381631-1287.38272 

the results i'm supposed following:

1111111110.000000 -864197532.000000 670592745.000000 13733.367622 2090.998911 

but these results get:

1111111128.000000 -864197544.000000 670592745.000000 13733.367798 2090.998901 

and i'm not sure why middle operation works when rest doesn't i've tried using float data type num1 , num2, results far more off using double.

in response question in comments said regarding streamlining program, consider following code:

note included revisions code reflecting original problem. namely, using float data type being implicitly cast double followed printf() call float specifiers.

#include <stdio.h> #include <stdlib.h>  main(){     file *fp;     char sym;     double num1, num2;      fp = fopen("input.txt","r");     if(fp == null)         exit(1);      while( fscanf(fp,"%f%c%f", &num1,&sym,&num2) != eof )         switch(sym){             case '+':                     printf("%f\n", num1+num2);                     break;             case '-':                      printf("%f\n", num1-num2);                     break;             case '*':                     printf("%f\n", num1*num2);                     break;             case '/':                      printf("%f\n", num1/num2);                     break;             default:                      printf("%f%c%f", num1,sym,num2);         }    fclose(fp); } 

notice reduced brackets well. required include brackets statements have multiple statements depend on it.

as in:

if(myvar == 1){     printsomething();     dosomething();     dosomethingelse(); } 

however, if require 1 statement can drop brackets.

if(myvar == 1)     printsomething(); 

note if have single block of code follows still can drop brackets (as same code above - while loop has single switch-case).