Patrick Mealey (plm051000@utdallas.edu) and Joseph Gebhard (jag054100@utdallas.edu)

To compile the programs, run make.


To use any of the 3 parts:
    - Run the program
        Then input the opcodes one per line.
    - Run program < scriptfile
        scriptfile must contain opcodes for the machine to run. 


jc reads opcodes and immediately executes them if possible.
On errors, the machine will attempt to continue, but warn the user of the problem.

jvm will parse the entire input before beginning execution. 
As it reads in the program, when a line containing a label is seen, an entry will be added to a map which stores the "address" of that instruction.
For each line read, an Opcode struct is created that indicates the type and stores the arguments (if any).
A Program Counter is initialized to 0, and the program executes in a loop while the PC is pointing within the program.

ejvm adds support for methods to jvm.
Additional parsing is added for the new commands and the method declarations.
The symbol table now maps method/label names to a struct containing the "address" and the variable limit for methods (labels are given a limit of -1).
The machine initialization is modified to set PC to the "address" of the first instruction of the main method.

In all 3 parts, input parsing is done using stringstreams. This allows for ignoring of whitespace within a line easily. The implementation does, however require that lines containing labels have some form of whitespace between the label: and the opcode.

Since labels and methods are stored in the same symbol table, there cannot be both a method and a label with the same name. 

The stack size for all parts is set to 50000.  This allows for significant recursion in the ejvm. The fib.jvm program included here successfully calculated the 40th fibonacci number.