CS 2213 Advanced Programming
Multiple Precision Arithmetic

Checking for valid values

/* Return as follows:
     0:        OK
     -1:       bad sign
     INTSIZE:  no digits, just blanks
     n         position of first invalid digit
*/
int checkvalid(char a[]) {
   int i;

   if ( (a[0] != '+') && (a[0] != '-') ) return -1;

   for (i=1;i<INTSIZE;i++)
     if (a[i] != ' ') break;
   if (i == INTSIZE) return INTSIZE;
   if ( (i == INTSIZE-1) && (a[i]=='0') && (a[0]=='-') ) return -1;
   if ( (i < INTSIZE-1) && (a[i]=='0') ) return i;


   for ( ; i<INTSIZE;i++)
     if ( (a[i] < '0') || (a[i] > '9') ) return i;

   return 0;
}

Functions to write

#define INTSIZE 50

int checkvalid(char a[]);
int readint(char a[]);
void printint(char a[]);
int addint(char a[], char b[], char c[]);
int subint(char a[], char b[], char c[]);
int multint(char a[], char b[], char c[]);

readint will read in a line from standard input and store the result in the array a in the appropriate format. It will return 0 if successful and 1 if an error occurs. The leading sign will be optional and it will ignore leading blanks on the input.

What are the possible error conditions?

• too many characters to fit in the array
• no digits entered
• invalid character entered

int readint(char a[]) {
   int count;
   int i;
   int j;
   int c;

   count = 0;
   while ((c = getchar()) == ' ');       /* skip leading blanks */
   if ( (c == '\n') || (c == EOF) )
      return 1;                          /* premature end of file */
   if (c == '-') a[0] = '-';             /* put in sign */
   else {
      a[0] = '+';
      if (c != '+')                      /* or first digit */
          a[++count] = (char)c;
   }
   while ( ((c=getchar()) !='\n') && (c!=EOF) ) { /* go until EOL or EOF */
      if (count >= INTSIZE-1)                     /* not enough room */
         return 1;
      a[++count] = (char)c;                       /* put it in, adjust count */
   }
   if (count == 0) return 1;                      /* nothing was put in */
   for (i=1;i<=count;i++)                        /* check that all are digits */
      if ( (a[i] < '0') || (a[i] > '9') )
         return 1;
   for (i=count,j=INTSIZE-1;i>0;i--,j--)      /* move them to the right place */
      a[j] = a[i];
   while (j > 0)                              /* fill beginning with blanks */
      a[j--] = ' ';
   return 0;
}

printint prints out the integer to standard output with the leading sign but not the leading blanks. It assumes that the array is valid.

addint puts the result of a + b in c if the result will fit and returns 0 in this case. It returns 1 if there is an overflow. It assumes that a and b are valid and that c is an array of the appropriate size.

subint and multint are similar.

Some useful functions

int addposint(char a[], char b[], char c[]);
will add a and b and put the result in c ignoring the signs of a and b assuming they are positive. The return value will be like addint.

void subposint(char a[], char b[], char c[]);
will be similar but will subtract b from a assuming that a is at least as big as b. This will never produce an error.

int greater(char a[], char b[]);
will return true if a is at least as big as b ignoring the signs of a and b.

From these it will be simple to write addint and subint by looking at the various cases.

Here are some issues to keep in mind when writing addposint

• Start adding from the end of the array.
• Elements of the array are characters so adding will look something like this:
  c[i] = a[i] + b[i] - '0';
• This might need to be adjusted based on a carry.
• You need to handle the case in which one of a[i] or b[i] is a blank.
• If they are both blank, you are done (except for the carry generated by the previsou step).
• Watch out for the carry overflowing into the sign position.
• Fill the rest of c with blanks.
• Fill in a plus sign in c.

This is what we got done in class on Tuesday, January 23:

int addint(char a[], char b[], char c[]) {
   if ( (a[0]=='+') && (b[0]=='+') )
      return addposint(a,b,c);
   if ( (a[0]=='-') && (b[0]=='-') ) {
      if (addposint(a,b,c) != 0) return 1;
      c[0] = '-';
      return 0;
   }
   if (greater(a,b)) {
   ...
   }
}