Assignment 1


Write 2 functions with the calling sequences

   int dumadd(int *a, int *b, int *c, int len),

   int dumsub(int *a, int *b, int *c, int len),

performing addition and subtraction of "long" real numbers,
represented by integer arrays of length len+2. In each of the above
functions, arrays *a, *b contain the input operands, whose contents
may not be altered; on exit, the array *c will contain the result of
the operation. The structure of the representation is illustrated by
the following two examples.



Example 1.

Integer array of length 7 (here, len=5)


1  3  2  7  4  3  5 

represents the real number

                  3
1  * 0.27435 * 10   .



Example 2.

Integer array of length 10 (here, len=8)


-1  -2  7  4  1  2  8  2  1 

represents the real number

                    -2
-1  * 0.7412821 * 10   .


In other words, the first element (in C, this is element number 0) of
the array represents the sign of the number (-1 means negative, 1
means positive), the second element is the decimal exponent, and the
remaining len elements represent the decimal mantissa. You may assume
that in all cases, len < 1001.

The maximal permitted absolute value of the decimal exponent is 1000
000; if the result of any operation has the exponent whose absolute
value exceeds 1000 000, an error code is returned (see below); in such
cases, the function does not alter the contents of *c.

The error code returned by each function is its own integer value.
The possible values for the functions dumadd, dumsub are 0 (successful
conclusion) and 1 (overflow, the absolute value of the exponent is
bigger than 1000 000).


This assignment is due on Wednesday, September 27.