CS143 -PA3 - Solved

Introduction
In this assignment you will write a parser for Cool. The assignment makes use of two tools: the parser generator (the C++ tool is called bison; the Java tool is called CUP) and a package for manipulating trees. The output of your parser will be an abstract syntax tree (AST). You will construct this AST using semantic actions of the parser generator.

You certainly will need to refer to the syntactic structure of Cool, found in Figure 1 of The Cool Reference Manual (as well as other parts). The documentation for bison and CUP is available online. The C++ version of the tree package is described in the Tour of Cool Support Code handout. The documentation for the Java version is available online. You will need the tree package information for this and future assignments.

There is a lot of information in this handout, and you need to know most of it to write a working parser. Please read the handout thoroughly.



Once you have a working copy of the programming assignment source tree, change into the directory for the current assignment. For the C++ version of the assignment, navigate to

[cool root]/assignments/PA3/

For Java, navigate to

[cool root]/assignments/PA3J/

(notice the “J” in the path name). Typing make in this directory will set up the workspace and copy a number of files to your directory. Some of the files in the directory will be read-only (using symbolic links). You should not edit these files. In fact, if you make and modify private copies of these files, you may find it impossible to complete the assignment. See the instructions in the README file. The files that you will need to modify are:

•    cool.y (in the C++ version) / cool.cup (in the Java version)

This file contains a start towards a parser description for Cool. The declaration section is mostly complete, but you will need to add additional type declarations for new nonterminals you introduce. We have given you names and type declarations for the terminals. You might also need to add precedence declarations. The rule section, however, is rather incomplete. We have provided some parts of some rules. You should not need to modify this code to get a working solution, but you are welcome to if you like. However, do not assume that any particular rule is complete.

•    good.cl and bad.cl

These files test a few features of the grammar. You should add tests to ensure that good.cl exercises every legal construction of the grammar and that bad.cl exercises as many types of parsing errors as possible in a single file. Explain your tests in these files and put any overall comments in the README file.

•    README

As usual, this file will contain the write-up for your assignment. Explain your design decisions, your test cases, and why you believe your program is correct and robust. It is part of the assignment to explain things in text, as well as to comment your code.

3           Testing the Parser
You will need a working scanner to test the parser. You may use either your own scanner or the coolc scanner. By default, the coolc scanner is used; to change this behavior, replace the lexer executable (which is a symbolic link in your project directory) with your own scanner. Don’t automatically assume that the scanner (whichever one you use!) is bug free—latent bugs in the scanner may cause mysterious problems in the parser.

You will run your parser using myparser, a shell script that “glues” together the parser with the scanner. Note that myparser takes a -p flag for debugging the parser; using this flag causes lots of information about what the parser is doing to be printed on stdout. Both bison and CUP produce a human-readable dump of the LALR(1) parsing tables in the cool.output file. Examining this dump may sometimes be useful for debugging the parser definition.

You should test this compiler on both good and bad inputs to see if everything is working. Remember, bugs in your parser may manifest themselves anywhere.

4           Parser Output
Your semantic actions should build an AST. The root (and only the root) of the AST should be of type program. For programs that parse successfully, the output of parser is a listing of the AST.

For programs that have errors, the output is the error messages of the parser. We have supplied you with an error reporting routine that prints error messages in a standard format; please do not modify it. You should not invoke this routine directly in the semantic actions; bison/CUP automatically invokes it when a problem is detected.

For constructs that span multiple lines, you are free to set the line number to any line that is part of the construct. Do not worry if the lines reported by your parser do not exactly match the reference compiler. Also, your parser need only work for programs contained in a single file—don’t worry about compiling multiple files.

5           Error Handling
You should use the error pseudo-nonterminal to add error handling capabilities in the parser. The purpose of error is to permit the parser to continue after some anticipated error. It is not a panacea and the parser may become completely confused. See the bison/CUP documentation for how best to use error. In your README, describe which errors you attempt to catch. Your test file bad.cl should have some instances that illustrate the errors from which your parser can recover. Your parser should recover in at least the following situations:

•    If there is an error in a class definition but the class is terminated properly and the next class is syntactically correct, the parser should be able to restart at the next class definition.

•    Similarly, the parser should recover from errors in features (going on to the next feature), a let binding (going on to the next variable), and an expression inside a {...} block.

Do not be overly concerned about the line numbers that appear in the error messages your parser generates. If your parser is working correctly, the line number will generally be the line where the error occurred. For erroneous constructs broken across multiple lines, the line number will probably be the last line of the construct.

6           The Tree Package
There is an extensive discussion of the C++ version of the tree package for Cool abstract syntax trees in the Tour section of the Cool documentation. The documentation for the Java version is available on the course web page. You will need most of that information to write a working parser.

7           Remarks
You may use precedence declarations, but only for expressions. Do not use precedence declarations blindly (i.e., do not respond to a shift-reduce conflict in your grammar by adding precedence rules until it goes away).

The Cool let construct introduces an ambiguity into the language (try to construct an example if you are not convinced). The manual resolves the ambiguity by saying that a let expression extends as far to the right as possible. Depending on the way your grammar is written, this ambiguity may show up in your parser as a shift-reduce conflict involving the productions for let. If you run into such a conflict, you might want to consider solving the problem by using a bison/CUP feature that allows precedence to be associated with productions (not just operators). See the bison/CUP documentation for information on how to use this feature.

Since the mycoolc compiler uses pipes to communicate from one stage to the next, any extraneous characters produced by the parser can cause errors; in particular, the semantic analyzer may not be able to parse the AST your parser produces.

8           Notes for the C++ version of the assignment
If you are working on the Java version, skip to the following section.

• You must declare bison “types” for your non-terminals and terminals that have attributes. For example, in the skeleton cool.y is the declaration:

%type <program> program

This declaration says that the non-terminal program has type <program>. The use of the word “type” is misleading here; what it really means is that the attribute for the non-terminal program is stored in the program member of the union declaration in cool.y, which has type Program. By specifying the type

%type <member_name> X Y Z ...

you instruct bison that the attributes of non-terminals (or terminals) X, Y, and Z have a type appropriate for the member member name of the union.

All the union members and their types have similar names by design. It is a coincidence in the example above that the non-terminal program has the same name as a union member.

It is critical that you declare the correct types for the attributes of grammar symbols; failure to do so virtually guarantees that your parser won’t work. You do not need to declare types for symbols of your grammar that do not have attributes.

•    Running bison on the initial skeleton file will produce some warnings about “useless noterminals” and “useless rules”. This is because some of the nonterminals and rules will never be used, but these should go away when your parser is complete.

•    The g++ type checker complains if you use the tree constructors with the wrong type parameters. If you ignore the warnings, your program may crash when the constructor notices that it is being used incorrectly. Moreover, bison may complain if you make type errors. Heed any warnings. Don’t be surprised if your program crashes when bison or g++ give warning messages.