MOP: Matrix operations program

Abstract

MOP performs matrix algebraic operations. All MOP instructions are input in a free-field fashion. Input consists of a regular form of algebraic syntax for matrix addition, subtraction, and multiplication. Various mathematical and utility procedures are available for operations such as inversion, eigen analysis, reading, writing, and printing. Matrices are referenced by alphabetic names with no embedded blanks. No distinction is made between matrices and vectors. Column or row vectors are easily accommodated as matrices with only one column or row, respectively. In addition, the user may set row, column, or dimension specifications to nonnegative integers. These variables are referenced by alphanumeric names that are separate from matrix names. Integer variables may replace size specifications in most MOP procedure calls. There are two input and two output files. All MOP commands are read from INPUT and all MOP output is placed on OUTPUT. Matrices may be read from INPUT or ALTINP using the READ procedure; matrices may be written to OUTPUT or ALTOUT using the WRITE procedure. MOP instruction lines may be no more than 80 characters in length. No continuation lines are allowed. Algebraic Expressions. Matrix addition, subtraction, and multiplication follow customary algebraic syntax. The only restriction to forming such expressions is that the result must be on the left. Conformity of operands is checked. It is a fatal error when the operands in an algebraic statement do not conform for the intended matrix operation. The possible operations are A =B + C, A = B C, A = B * C, and A = B, where A, B, and C are matrices and where they conform for the indicated operation. In each of these, the operands must exist prior to the operation. The result of the operation is created if it does not exist. If the result exists, it is replaced (overwritten) if it is the appropriate size;