API tools faq
paste
Guest User

Untitled

a guest
May 23rd, 2017
635
0
Never
Add comment
Not a member of Pastebin yet? Sign Up, it unlocks many cool features!
C 9.96 KB | None | 0 0
raw download clone embed print report
  1. /* K&R version, 1388 tokens excluding whitespace */
  2. /****************************************************************************/
  3. /*                                                                          */
  4. /*       Program: tokount.c                                                 */
  5. /*                                                                          */
  6. /*       Author:  Barry Gorman ([email protected])                    */
  7. /*                                                                          */
  8. /*       Date:    April 15th 1993.                                          */
  9. /*                                                                          */
  10. /*       Purpose: Counts tokens in C programs, based on ANSI C Standard     */
  11. /*                                                                          */
  12. /*       Note:    This program is to be used only on correct program texts  */
  13. /*                                                                          */
  14. /****************************************************************************/
  15. #include <ctype.h>                      /* isalnum isdigit isspace          */
  16. /* #include <errno.h>                   /* errno                            */
  17. extern int errno;
  18. #include <stdio.h>                      /* EOF ferror FILE fopen fputs getc */
  19.                                         /* perror printf stderr ungetc      */
  20. /* #include <stdlib.h>                  /* EXIT_FAILURE EXIT_SUCCESS        */
  21. #include <string.h>                     /* strchr strcmp                    */
  22. /****************************************************************************/
  23. #define TOKEN_MAX  512
  24. typedef int E_CHAR;                                       /* all chars +EOF */
  25. typedef enum {FALSE=0, TRUE=1} bool;                      /* suits my debug */
  26. typedef enum
  27. {HEADER, IDENTIFIER, NUMBER, CHARACTER, STRING, OPERATOR, PUNCTUATOR,
  28.     OTHER, UNKNOWN, END_OF_FILE, WHITE_VERTICAL, WHITE_HORIZONTAL}
  29. TOKEN_CLASS;
  30. E_CHAR current_char, look_ahead='\n';                     /* start with new */
  31. FILE *program_file;
  32. char token_chars[TOKEN_MAX+1];
  33. int token_count, token_length;
  34. bool header_flag, echo_tokens;                            /* handle <....h> */
  35. TOKEN_CLASS current_token;
  36. /****************************************************************************/
  37. E_CHAR next_char(/*void*/)  /* a function for ANSI C Translation Phases 1+2 */
  38. /****************************************************************************/
  39. {static E_CHAR next_one='\n';                /* internal character cache */
  40.     current_char=look_ahead;                    /* copy external lookaheads */
  41.     if(token_length<TOKEN_MAX)
  42.       {token_chars[token_length++]=current_char;           /* assemble val */
  43.         token_chars[token_length]='\0';}
  44.     do {look_ahead=next_one; next_one=getc(program_file);   /* get next raw */
  45.         if(look_ahead=='?' && next_one=='?')    /* ANSI trigraph 2.1.1.2(1) */
  46.       {int next_two=getc(program_file);    /* look ahead one more char */
  47.             char *translated=strchr("#[\\]^{|}~=(/)'<!>-"+9, next_two);
  48.             if(translated)                      /* test trigraph recognised */
  49.           {look_ahead=*(translated-9);     /* copy char to replace ??x */
  50.                 next_one=getc(program_file);}   /* read following character */
  51.             else ungetc(next_two, program_file);}  /* otherwise put ch back */
  52.         if(look_ahead!='\\' || next_one!='\n') return(current_char);  /* OK */
  53.         next_one=getc(program_file);}           /* ANSI splicing 2.1.1.2(2) */
  54.     while(1);}  /* I don't like infinite loops; this one makes code clearer */
  55. /****************************************************************************/
  56. TOKEN_CLASS next_token(/*void*/) /* function for ANSI C Translation Phase 3 */
  57. /****************************************************************************/
  58. #define Digits     '0': case '1': case '2': case '3': case '4': case '5':\
  59.    case '6': case '7': case '8': case '9'
  60. #define Double     '#': case '+': case '&': case '|'
  61. #define Letter     'A': case 'B': case 'C': case 'D': case 'E': case 'F':\
  62.  case 'G': case 'H': case 'I': case 'J': case 'K': case 'M': case 'N':\
  63.  case 'O': case 'P': case 'Q': case 'R': case 'S': case 'T': case 'U':\
  64.  case 'V': case 'W': case 'X': case 'Y': case 'Z': case 'a': case 'b':\
  65.  case 'c': case 'd': case 'e': case 'f': case 'g': case 'h': case 'i':\
  66.  case 'j': case 'k': case 'l': case 'm': case 'n': case 'o': case 'p':\
  67.  case 'q': case 'r': case 's': case 't': case 'u': case 'v': case 'w':\
  68.  case 'x': case 'y': case 'z': case '_'
  69. #define Others     '$': case '@': case '`': case '\\'
  70. #define Puncts     '(': case ')': case ',': case ';': case '[': case ']':\
  71.  case '{': case '}'
  72. #define Quotes     '"': case'\''
  73. #define Simple     ':': case '?': case '~'
  74. #define Single     '!': case '%': case '*': case'=' : case'^'
  75. #define WhiteH     ' ': case'\t'
  76. #define WhiteV    '\n': case'\v': case'\f'
  77. {token_length=0; switch(next_char())
  78.  case '.':   {if(!isdigit(look_ahead))                    /* period/ellipsis */
  79.                 {while(look_ahead=='.') next_char();     /* grab any number */
  80.            case Puncts:     return(PUNCTUATOR);}                    /* arbitrary class */
  81.         case Digits: while(isalnum(look_ahead) || look_ahead=='.' || look_ahead=='_')
  82.                  if((next_char()=='e' || current_char=='E') /* opt exponent */
  83.                  && ( look_ahead=='+' || look_ahead=='-')) next_char();
  84.              return(NUMBER);                             /* unusual syntax! */
  85.         case 'L':    if(look_ahead=='"' || look_ahead=='\'') next_char();  /* L"... */
  86.              else                                        /* reserved words, */
  87.          case Letter:    {while(isalnum(look_ahead) || look_ahead=='_') next_char();
  88.                  return(IDENTIFIER);                     /* all identifiers */
  89.                 case '<':        if(header_flag) current_char='>';       /* flag set in Ph4 */
  90.                  else
  91.          case '>':           {if(look_ahead==current_char) next_char(); /* >> and << */
  92.                      if(look_ahead=='=') next_char();    /*  >= >>= <= <<=  */
  93.                      return(OPERATOR);}}                 /* compare & shift */
  94.         case Quotes:{E_CHAR quote_char=current_char;             /* "_" '_' and <_> */
  95.              do if(look_ahead=='\\') {next_char(); next_char();}  /* \n etc */
  96.              while(next_char()!=quote_char && current_char!=EOF); /* match? */
  97.              return(header_flag? HEADER: quote_char=='"'? STRING: CHARACTER);}
  98.         case '/':    if(look_ahead=='*')                         /* start a comment */
  99.           {next_char();                            /* kill current /, */
  100.                  while(next_char()!='*' || look_ahead!='/')  /* and first * */
  101.                      if(current_char==EOF) break;        /* don't lock loop */
  102.                  next_char();    token_chars[0]=' ';     /* replace comment */
  103.                  token_length=1; token_chars[1]='\0';    /* by single space */
  104.          case WhiteH:     return(WHITE_HORIZONTAL);               /* may consolidate */
  105.          case WhiteV:     return(WHITE_VERTICAL);                 /* in later phases */
  106.          case '-':        if(look_ahead=='>') {next_char(); return(OPERATOR);} /* -> */
  107.          case Double:     if(look_ahead==current_char){next_char(); return(OPERATOR);}}
  108.         case Single: if(look_ahead=='=') next_char();            /* finds op= forms */
  109.         case Simple: return(OPERATOR);                           /* deal with ++ += */
  110.         case Others: return(OTHER);                              /* few unused char */
  111.         case EOF:    return(END_OF_FILE);                        /* want EOF repeat */
  112.         default:     return(UNKNOWN);}}                          /* is non printing */
  113. /****************************************************************************/
  114. TOKEN_CLASS print_token(/*void*/)          /* application specific function */
  115. /****************************************************************************/
  116. {current_token=next_token();    /* alter function to print selected info */
  117.     if(current_token<END_OF_FILE)
  118.       {switch (current_token)
  119.      {
  120.      case HEADER:
  121.             token_count-=2;
  122.         if (echo_tokens)
  123.               printf("    #include %s shouldn't have counted; subtracting\n",
  124.                    token_chars);
  125.             break;
  126.       case STRING:  
  127.             if (echo_tokens)
  128.                 printf("%5d-%d: %d [%s]\n", token_count+1, token_count+token_length,
  129.                     current_token, token_chars);
  130.             token_count += token_length;
  131.             break;
  132.       default:
  133.             token_count++;
  134.             if (echo_tokens)
  135.                 printf("%5d: %d [%s]\n", token_count, current_token, token_chars);
  136.       }
  137.      }
  138.     return(current_token);}
  139. /****************************************************************************/
  140. /* int main(int argc, char *argv[])/* program expects file-name as argument */
  141. main(argc, argv) int argc; char *argv[];
  142. /****************************************************************************/
  143. {
  144.     if((argc<2)||(argc>3)||((argc==3)&&(strcmp(argv[1],"-p"))))
  145.    {fputs("Usage: tokount [-p] file.c\n",stderr); return(/*EXIT_FAILURE*/1);}
  146.     echo_tokens=(argc==3);                     /* Print tokens if -p option */
  147.     if(!(program_file=fopen(argv[argc-1], "r"))) {perror(argv[argc-1]); return(errno);}
  148.         /*** do not alter this loop, put changes in print_token ***/
  149.     do {while(print_token()==WHITE_VERTICAL||current_token==WHITE_HORIZONTAL);
  150.         if(strcmp(token_chars, "#"))       continue;
  151.         while(print_token()==WHITE_HORIZONTAL);
  152.         if(strcmp(token_chars, "include")) continue;   /* double negative!! */
  153.         header_flag=TRUE;
  154.         while(print_token()==WHITE_HORIZONTAL);
  155.         header_flag=FALSE;}
  156.     while(current_token!=END_OF_FILE);
  157.     if(ferror(program_file)) {perror(argv[argc-1]); return(errno);}
  158.     printf("%d\n", token_count);
  159.     return(/*EXIT_SUCCESS*/0);}
  160. /********************************* the end **********************************/
Advertisement
Add Comment
Please, Sign In to add comment
Public Pastes
We use cookies for various purposes including analytics. By continuing to use Pastebin, you agree to our use of cookies as described in the Cookies Policy.  OK, I Understand
Not a member of Pastebin yet?
Sign Up, it unlocks many cool features!