CSE P 501 Project III
Static Semantics, Type Checking & Symbol Tables

Due: Thursday, Nov. 17, by 11:00 pm. Turn in your project using the assignment drop box (links on the project page).

Overview

Add static semantics checking to your compiler. In particular, you should do the following:

• Add global symbol table(s) storing information about classes and their members, and local symbol tables for each method to store information about parameters and local variables.
• Calculate and store type information for classes, class members, parameters, and variables.
• Calculate type information for expressions and other appropriate parts of the abstract syntax.
• Do some basic error checking to verify at least the following properties:
  1. Every name is properly declared.
  2. Components of expressions have appropriate types (e.g., + is only applied to values of type int, && is only applied to values of type boolean, the expression in parentheses in an if or while statement is boolean, etc.).
  3. If a method is selected from a value with a reference type, then that name is defined as a member of that type.
  4. Methods are called with the correct number of arguments.
  5. In assignment statements and method call parameter lists, the values being assigned have appropriate types (i.e., the value either has the same type as the variable or is a subclass of the variable's class).
  6. If a method in a subclass overrides one in any of its superclasses, the overriding method has the same parameter list as the original method, and the result type of the method is the same as the result type of the original one, or a subclass if the original result type is a reference type.
  Feel free to add additional checks, but you should try to get this much done. It may be possible to do some of these checks on the fly while building symbol tables and type information, or it may require an additional visitor pass.
• If your parser grammar accepts input that is not part of MiniJava (i.e., uses a covering grammar or otherwise parses constructs that are syntactically legal but have semantics errors), check that the program is actually legal. If you implemented extensions like casts, you should add appropriate error and type checking.
• Print suitable messages for errors detected. It's fine to supress useless error messages - for instance, feel free to complain only once about an undeclared variable instead of repeating the message each time the variable is used in the code.
• Provide a mechanism to print the contents of the symbol tables. The output does not need to be verbose, but it should be sufficient to show each of the identifiers declared in the program, their types, and any other important information. The output should label each scope so it is clear which identifier(s) are declared where.
• You should include a TestSemantics main program for this phase of the project in a file TestSemantics.java in the src directory of the MiniJava project tree. When executed without arguments, this program should read a MiniJava program from System.in, parse and check it, then print the symbol tables and the information in them. If any errors are detected in the semantics phase or earlier, the program should exit with a non-zero return code, otherwise it should exit normally. If you are implementing your compiler using a language different from Java you should do the closest equivalent in that language.

Details and Suggestions

• It's probably easiest to collect the type information in a pass or two over the AST. Pass I should collect information about classes and fields (both data and methods), and build the global symbol tables. Pass II would then analyze method bodies, build the local symbol tables, and perform type and other error checking. You might find it more convenient to break this down into more passes each of which does fewer things, particularly for Pass I, where it can be easier to build a global symbol table of class names before processing individual classes to build class symbol tables with information about variables, methods, and their types.
• You should add appropriate fields in some or all AST nodes to store type and other information as necessary.
• Use the visitor pattern! This is where it pays off to have gone to the trouble to set up the visitor machinery. Provide new implementations of the Visitor interface for the various semantics checks.
• Take advantage of the standard library container classes and data structures in Java (or whatever implementation language you're using) to simplify your implementation. Class HashMap should be particularly useful for symbol tables. Use the List classes (ArrayList or LinkedList) for things like argument lists and field lists in classes. Don't reinvent any more wheels than necessary.
• It can be useful to include a few auxiliary methods that perform common operations on types. Possibilities include a method that returns true if two types are the same, and a method that returns true if a value of one type is assignable to another. Also possibly useful: a method that tries to add an entry to a symbol table and reports an error if the name is already declared, and another that looks up an identifier and reports an error if it is not found (and maybe adds it to the symbol table with an "undefined" type, which can be used to suppress additional, extraneous error messages about the same identifier).
• A useful tool is to augment the AST print visitor from the last assignment so it also prints type information where appropriate.  This isn't required, but it might save some debugging time. In any case you will want to provide toString() methods or other mechanisms to print type information, at least when printing symbol tables.

Testing

You should test your compiler on both correct MiniJava programs and on programs that contain various sorts of static semantics problems, including programs that are syntactically legal (i.e., accepted by the parser), but have static semantics errors.

What to Turn In

Turn in the following:

• Java (or other) source code for your the compiler, including code and data structures used to represent types and symbol tables.
• An INFO file explaining how to build and run your compiler, including a description of any required software or libraries if you're using your own project setup, along with any notes about extensions, additional checks performed, or other unusual or interesting things in this phase of your compiler. You should be sure to describe how much of the requested checking was actually implemented, and any major surprises (either good or bad) you encountered along the way.
• Files showing the contents of symbol tables for one or two sample programs. This must include the Factorial.java sample program from the MiniJava web site.
• A few files demonstrating the output produced by your compiler (if any) for both error-free and erroneous source programs.

As before, your code should run on attu when built with ant. You should do an ant clean, then bundle up your compiler directory in a tar file and turn that in. That will ensure that we have all the pieces of your compiler if we need to check something. If you are using a different implementation language, please do the equivalent in that language and include instructions describing how to build and test your parser/AST code in your INFO file.

If you are working with others, you should turn in only one copy per group with your names listed in the same order as usual in the INFO file.