Removing comments with a sliding window without nested while loops

Question

I\'m trying to remove comments and strings from a c file with c code. I\'ll just stick to comments for the examples. I have a sliding window so I only have character n

Answer 1

Since you only wish to use two characters for the buffer and only one while loop, I would suggest a third char to track your state (whether skipping text or not). I've put together a test program for you with inline comments explaining the logic:

// Program to strip comments and strings from a C file
//
//  Build:
//     gcc -o strip-comments strip-comments.c
//
//  Test:
//     ./strip-comments strip-comments.c

#include 
#include 
#include 
#include 
#include 
#include 

/* The following is a block of strings, and comments for testing
 * the code.
 */
/* test if three comments *//* chained together */// will be removed.
static int value = 128 /* test comment within valid code *// 2;
const char * test1 = "This is a test of \" processing"; /* testing inline comment */
const char * test2 = "this is a test of \n within strings."; // testing inline comment
// this is a the last test


int strip_c_code(FILE * in, FILE * out)
{
   char      buff[2];
   char      skipping;

   skipping = '\0';
   buff[0]  = '\0';
   buff[1]  = '\0';

   // loop through the file
   while((buff[0] =  fgetc(in)) != EOF)
   {
      // checking for start of comment or string block
      if (!(skipping))
      {
         // start skipping in "//"  comments
         if ((buff[1] == '/') && (buff[0] == '/'))
            skipping = '/';

         // start skipping in "/*"  comments
         else if ((buff[1] == '/') && (buff[0] == '*'))
            skipping = '*';

         // start skipping at start of strings, but not character assignments
         else if ( ((buff[1] != '\'') && (buff[0] == '"')) &&
                   ((buff[1] != '\\') && (buff[0] == '"')) )
         {
            fputc(buff[1], out);
            skipping = '"';
         };

         // clear buffer so that processed characters are not interpreted as
         // end of skip characters.
         if ((skipping))
         {
            buff[0] = '\0';
            buff[1] = '\0';
         };
      };

      // check for characters which terminate skip block
      switch(skipping)
      {
         // if skipping "//" comments, look for new line
         case '/':
         if (buff[1] == '\n')
            skipping = '\0';
         break;

         // if skipping "/*" comments, look for "*/" terminating string
         case '*':
         if ((buff[1] == '*') && (buff[0] == '/'))
         {
            buff[0]  = '\0';
            buff[1]  = '\0';
            skipping = '\0';
         };
         break;

         // if skipping strings, look for terminating '"' character
         case '"':
         if ((buff[1] != '\\') && (buff[0] == '"'))
         {
            skipping = '\0';
            buff[0]  = '\0';
            buff[1]  = '\0';
            fprintf(out, "NULL"); // replace string with NULL
         };
         break;

         default:
         break;
      };

      // if not skipping, write character out
      if ( (!(skipping)) && ((buff[1])) )
         fputc(buff[1], out);

      // shift new character to old character position
      buff[1] = buff[0];
   };

   // verify that the comment or string was terminated properly
   if ((skipping))
   {
      fprintf(stderr, "Unterminated comment or string\n");
      return(-1);
   };

   // write last character
   fputc(buff[1], out);

   return(0);
}


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

   if (argc != 2)
   {
      fprintf(stderr, "Usage: %s \n", argv[0]);
      return(1);
   };

   if ((fs = fopen(argv[1], "r")) == NULL)
   {
      perror("fopen()");
      return(1);
   };

   strip_c_code(fs, stdout);

   fclose(fs);

   return(0);
}

/* end of source file */

I've also posted this code on Github to make it easier to download and compile:

https://gist.github.com/syzdek/5417109