This is a simple C parser script I wrote for a university project (Principles of Compiler Design course).

scanner.pl and parser.pl are separate and do not depend on each other. A simplified version of scanner.pl is used in parser.pl.

Scanner does the lexical analysis(using regular expressions) and Parser does the syntax analysis.

The simple C grammar along with FIRST and FOLLOW sets and parse table are in the documents directory. The LL(1) C grammar is the LL(1) version of Simple C grammar you get after left factoring and left recursion removal.

Text::SimpleTable module is required for drawing symbol table and output of parser. You can install it with cpanminus:

sudo cpanm Text::SimpleTable

Perl scripts should be marked as executable. You can achieve this with the following command:

chmod +x scanner.pl parser.pl

Scanning C source code:

./scanner.pl [file_name]

Parsing C source code:

./parser.pl [-s] [file_name]

-s option tells the script to print the symbol table.

If you want to type the C code directly, don't enter the file name. When you finished entering C code, hit Ctrl-D.

If you can't see the whole output in the terminal, pipe it to less:

./parser.pl [-s] [file_name] | less

Here is some dummy code to show you what can be parsed and scanned:

For scanner:

/* Multi-line
comment */
int main() {
 int i = 2.56e+2;
 // Comment
 char ch = 'a';
 i++;
 i--;
 i*=5;
 i/=2;
 return 0;
}

For parser:

main() {
 int a[10];
 int i;
 int j;
 i = 4;
 j=i*6;
 if(j<10) {
 j = j - 1;
 }
 return 0;
}