A8P3 - 30 marks (LR.java or lr.ss) Write a Java or Scheme program that reads an LR1 file representing a context-free grammar, an LR(1) machine, and a sequence to be recognized. If the sequence is in the language, output the unindented reversed rightmost derivation, as defined in the cfg-r file format. If the input is not recognized, output "ERROR at k" to standard error, where k is one greater than the number of tokens in the longest correct prefix. For the sample lr1 file as input, the correct output is:
term id expr term term id expr term term ( expr ) expr expr - term term id expr expr - term S BOF expr EOF
If you replace the last line of this file by BOF id - id ) - id EOF the correct output (to standard error) is ERROR at 5
You may also test your program with the WL grammar and parse tables, in .lr1 format after adding a sequence to be recognized. Note that the parsing handout contains LR pseudocode.
A8P4 - 16 marks (WLParse.java or wlparse.ss) Write a WL parser encapsulated as a Java or Scheme program. The input to your parser is a file representing the sequence of tokens in a WL program, as output by the scanner specified in A6P4. That is, each token is represented by a line containing the name of the token, followed by a space, followed by the string recognized as the token in the WL source program. Note: Unlike the input of A8P3, the input to your parser in this problem will not have lines containing BOF and EOF at the beginning and end.
If the input represents a syntactically valid WL program, the output from your program should be a .wli file. The format of a .wli represents the derivation and also the recognized token strings.
If the input does not represent a syntactically valid WL program, output "ERROR at k" to System.err where k is one greater the number of tokens in the longest correct prefix of the input (that is, not counting BOF).
The student.cs undergraduate enviroment includes two commands to help you test your parser:
java cs241.WLScan reads a WL program and outputs a token list that may be used as input to your parser; java cs241.WLICheck displays and verifies the format of a .wli file.
