Assignment 3 Extra Credit

• Robert Nguyen

• Sean Javiya

• Allison Atienza

• Eric Britten

Follow the instructions below on building and running tests:

Build

sh ./build.sh

test with the Bottom up LR parser

./tokenizer tests/lr-table.txt -lr

The output on the screen will show what is happening with the stack and table, for instance:

--------------------------------------------------
stack: $0T2*8c5
top of stack: 5 and current token: $
T[5, $] = R8; Action -> Reduce(8): pop(5);pop(c); Production: F -> i;  T[8, F]; push(13)
--------------------------------------------------

And after that a parse tree that shows what rule was used to reduce expressions with operators:

<Program>
  * : T -> T*F
    + : E -> E+T
      a
      b
    c

Additionally, the tests/ folder will contain newly generated files that show the parse tree in several formats (code, tree, productions).

test with the Bottom up LR parser to show syntax error

./tokenizer tests/lr-error.txt -lr

#Previous Documentation from Assignment 2

Run with a test file as the first argument

./tokenizer tests/declarations.txt

The output to the screen will be the tokens and productions as they are being processed followed by a represenation of the parse tree that shows a token, followed by the productions that were used.

-------Parse Tree Productions -----------
KEYWORD     = float
     <Program>
     <StatementList> -> <Statement> <MoreStatements>
     <Statement> -> <Assign> | <Declaration>
     <Declaration> -> <Type><ID>
     <Type> -> float | int | bool
IDENTIFIER  = fahr
     <ID> -> identifier
SEPARATOR   = ;
     <MoreStatements> -> ; <Statement> <MoreStatements> | epsilon

Several files are generated in the same folder as the test file that will be named: <testfile>-<type>.txt where type is one of the following.

1. states - All of the state transitions from the lexical scanner
2. symbols - A list of all the defined symbols (identifiers in the code) with type and name with the location in the file where they were defined.
3. code - A view of the parse tree that only shows the terminal nodes (leaves) that will look like the source code with comments removed and different spacing.
4. productions - The view of the parse tree that lists tokens and then the productions used to verify the syntax
5. parse tree - A view of the parse tree that will show indented nodes for children. This shows terminal and non-terminal nodes.

Demonstration of Features

Assignments and Declaratives

This shows that the syntax analyzer can handle statements that involve declarations, assignment and arthmetic expressions including identifers and numbers.

./tokenizer tests/declarations.txt

or a comprehensive test can be run with:

sh tests2.sh

This command will run several tests and only report the errors to the screen. Some source files should have errors to demonstrate error handling.

While Loop

This example has a while loop inside of a while loop.

./tokenizer tests/while.txt

Error handling:

The syntax analyzer stops at the first error.

Invalid Lexeme:

./tokenizer tests/lexicalerror.txt

Output includes:

Lexical: tests/lexicalerror.txt:2:5 - _ is only allowed inside of an identifier: _invalid_name

Undefined Symbols:

This will show what happens when an indentifier is used that was not previously defined:

./tokenizer tests/undefined.txt 

The output includes:

Syntactic:  tests/undefined.txt:5:4 - sum was not previously defined

Syntax Error: In this example a while loop is missing the do keyword:

./tokenizer tests/syntaxerror.txt

In this case the error refers to while and a production is not satisfied, though it doesn't give the exact cause.

Syntactic:  tests/syntaxerror.txt:4:6 - this rule <TermPrime> -> *<Factor><TermPrime> | /<Factor><TermPrime> | epsilon could not be met with 'while'