Thread: Parsing an escaped quote inside a string

Sorry if this is a double post, I didn't see my post on the board so I am reposting.

I've written a rudimentary syntax checking program for c code, which just checks to make sure every {, (, or [ is closed in the appropriate order. It works fine right now, except for when I try and break it by adding (,[,{ characters inside strings.

So if it checks this file:

Code:

#include <stdio.h>
int main(int argc, char **argv){
     printf("hello world");
     return(1);
}

It works fine, but if I do this:

Code:

printf("there is escaped quotes \" in this string and now ( ( [ ) } will break the syntax check");

I'm trying to figure out how to parse the character " differently than \" if that makes any sense. Here is my code, and I'll bold out the relevant section.

Code:

#include <stdio.h>
#include <stdlib.h>
#include "stack.h"

#define STRLEN 256
#define IN_LINE 1
#define IN_MULTILINE 2
#define IN_SINGLE_STRING 3
#define IN_DOUBLE_STRING 4
#define OUT 0
#define TRUE 1
#define FALSE 0
#define BRACKETS 1
#define PARENTHESES 2

int check_syntax(FILE *fp);

int main(int argc, char *argv[]){
    FILE *fp;
    int bad_line;

	if (argc < 2){
	    printf("Usage is syntax [FILE]");
	    exit(EXIT_FAILURE);
	}

	if( (fp=fopen(argv[1], "r")) == NULL)
        exit(EXIT_FAILURE);

    bad_line = check_syntax(fp);

    fclose(fp);

    if(bad_line)
        printf("Error on line %d\n", bad_line);
    else
        printf("syntax is correct\n");

	exit(EXIT_SUCCESS);
}//}99[}


int check_syntax(FILE *fp){
	STACK *stack;

	char cp, nextc;
	int state, ln;
	int error;

    stack = (STACK *)malloc(sizeof stack);

	state = OUT;
	ln  = 1;
	error = FALSE;

	while( (cp = fgetc(fp)) != EOF){
	    nextc = fgetc(fp);
	    ungetc(nextc, fp);

		//check for comments first
		if(state == OUT && cp == '/'){
			if( nextc == '/'){ //we're in a single line comment
				state = IN_LINE;
				continue;
			}
			if(nextc == '*'){ //we're in a multi line comment
				state = IN_MULTILINE;
				continue;
			}
		}

		//check for end of comments next
		if(state == IN_LINE && cp == '\n'){
			state = OUT;
			continue;
		}
		if(state == IN_MULTILINE && cp == '*' && nextc == '/'){
			state = OUT;
			continue;
		}

		//check for in string
        if(state == OUT && cp == '\"')
            state = IN_DOUBLE_STRING;
        if(state == OUT && cp == '\'')
            state = IN_SINGLE_STRING;
        if(state == IN_DOUBLE_STRING && cp == '\\' && nextc == '\"'){
            cp = fgetc(fp);
            cp = fgetc(fp);
        }
        if(state == IN_SINGLE_STRING && cp == '\\' && nextc == '\''){
            cp = fgetc(fp);
            cp = fgetc(fp);
        }
        if(state == IN_DOUBLE_STRING && cp == '\"')
            state = OUT;
        if(state == IN_SINGLE_STRING && cp == '\'')
            state = OUT;

		//we are out of a comment, time to check syntax
		if(state == OUT){
            switch(cp){
                case '\n':
                    ln++;
                    break;
                case '{':
                case '}':
                    if(stack->state != OUT)
                        error = TRUE;
                    break;
                case '(':
                    stack = add_to_stack(stack, PARENTHESES);
                    break;
                case ')':
                    if(stack->state != PARENTHESES)
                        error = TRUE;
                    stack = pop_stack(stack);
                    break;
                case '[':
                    stack = add_to_stack(stack, BRACKETS);
                    break;
                case ']':
                    if(stack->state != BRACKETS)
                        error = TRUE;
                    stack = pop_stack(stack);
                    break;
            }
            if(error == TRUE)
                return(ln);
		}
	}
	return(0);
}

I hope this makes sense, thanks for looking

So how should I know when to skip over the escaped quotes in this line.

Code:

printf("this is a \"quoted string");

My code is checking each character one by one, and if it hits a " character, it sets the state to DOUBLE_QUOTE, and will ignore the every character until it gets to another " character setting the state back to OUT. When it checks the above line it will thing the quote has finished at the character before the q. What can I do??

Thanks for the fast reply by the way.

Thanks Quzah, ended up optimizing and cleaning up my code and it works great, I'm still not 100% sure why it wasn't working before but it looks better now