Project 2 (Scanning)
Goal
1. Support multiline comments.
2. Support additional tokens (reserved words and operators).
3. Support long and double literals.
Grammars
The lexical and syntactic grammars for j– and Java can be found at https://www.cs.umb.edu/j–/grammar.pdf W.
Download the Project Tests
Download and unzip the tests W for this project under $j/j–.
In this project, you will only be updating the hand-crafted scanner, which means that the only program files you will be
modifying under $j/j–/src/jminusminus are TokenInfo.java and Scanner.java.
Run the following command inside the $j/j– directory to compile the j– compiler with your changes.
& ~/workspace/j–
$ ant
Run the following command to compile (just scan for now) a test program, say project2/XYZ.java using the j– compiler.
& ~/workspace/j–
$ bash ./bin/j– -t project2/XYZ.java
which only scans XYZ.java and prints the tokens in the program along with the line number where each token appears. The
file project2/XYZ.tokens provides the reference (ie, expected) output.
Problem 1. (Multiline Comment) Add support for multiline comment, where all the text from the ASCII characters /* to
the ASCII characters */ is ignored.
Directions:
• Draw a state transition diagram for recognizing multiline comments
• Use the diagram to implement code in Scanner.java for scanning multiline comments
Problem 2. (Operators) Add support for the following operators. Note that scanning support for some of the operators
was added to j– in Project 1.
? : ~ != / /= -= *= % %=
>> >>= >>> >>>= >= << <<= < ^ ^= | |= || & &= Directions: • Define tokens for the operators in TokenInfo.java • Modify Scanner.java to recognize the operators Problem 3. (Reserved Words) Add support for the following reserved words. 1 / 2 https://www.cs.umb.edu/j--/grammar.pdf http://www.swamiiyer.net/cs451/project2.zip Project 2 (Scanning) break case catch continue default do double finally for implements interface long switch throw throws try Directions: • Define tokens for the reserved words in TokenInfo.java • Modify Scanner.java to recognize the reserved words Problem 4. (Literals) Add support for long and double literals (just decimal). Your are not allowed to use regular expressions to scan literals. Directions: • Define tokens for long and double literals in TokenInfo.java • Extend the state transition diagram for recognizing integer literals to also recognize long and double literals • Use the diagram to implement code in Scanner.java for scanning long and double literals Before you submit your files, make sure: • Your code is adequately commented and follows good programming principles. • You use the template file report.txt for your report. • Your report meets the prescribed guidelines. Files to submit: 1. TokenInfo.java 2. Scanner.java 3. Parser.java 4. JBinaryExpression.java 5. JUnaryExpression.java 6. report.txt 2 / 2