pdftops/PSTokenizer.cxx

   1 //========================================================================
   2 //
   3 // PSTokenizer.cc
   4 //
   5 // Copyright 2002-2003 Glyph & Cog, LLC
   6 //
   7 //========================================================================
   8
   9 #include <config.h>
  10
  11 #ifdef USE_GCC_PRAGMAS
  12 #pragma implementation
  13 #endif
  14
  15 #include <stdio.h>
  16 #include <stdlib.h>
  17 #include "PSTokenizer.h"
  18
  19 //------------------------------------------------------------------------
  20
  21 // A '1' in this array means the character is white space.  A '1' or
  22 // '2' means the character ends a name or command.
  23 static char specialChars[256] = {
  24   1, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 0, 1, 1, 0, 0,   // 0x
  25   0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,   // 1x
  26   1, 0, 0, 0, 0, 2, 0, 0, 2, 2, 0, 0, 0, 0, 0, 2,   // 2x
  27   0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 2, 0, 2, 0,   // 3x
  28   0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,   // 4x
  29   0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 2, 0, 2, 0, 0,   // 5x
  30   0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,   // 6x
  31   0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 2, 0, 2, 0, 0,   // 7x
  32   0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,   // 8x
  33   0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,   // 9x
  34   0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,   // ax
  35   0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,   // bx
  36   0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,   // cx
  37   0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,   // dx
  38   0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,   // ex
  39   0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0    // fx
  40 };
  41
  42 //------------------------------------------------------------------------
  43
  44 PSTokenizer::PSTokenizer(int (*getCharFuncA)(void *), void *dataA) {
  45   getCharFunc = getCharFuncA;
  46   data = dataA;
  47   charBuf = -1;
  48 }
  49
  50 PSTokenizer::~PSTokenizer() {
  51 }
  52
  53 GBool PSTokenizer::getToken(char *buf, int size, int *length) {
  54   GBool comment, backslash;
  55   int c;
  56   int i;
  57
  58   // skip whitespace and comments
  59   comment = gFalse;
  60   while (1) {
  61     if ((c = getChar()) == EOF) {
  62       buf[0] = '\0';
  63       *length = 0;
  64       return gFalse;
  65     }
  66     if (comment) {
  67       if (c == '\x0a' || c == '\x0d') {
  68         comment = gFalse;
  69       }
  70     } else if (c == '%') {
  71       comment = gTrue;
  72     } else if (specialChars[c] != 1) {
  73       break;
  74     }
  75   }
  76
  77   // read a token
  78   i = 0;
  79   buf[i++] = c;
  80   if (c == '(') {
  81     backslash = gFalse;
  82     while ((c = lookChar()) != EOF) {
  83       if (i < size - 1) {
  84         buf[i++] = c;
  85       }
  86       getChar();
  87       if (c == '\\') {
  88         backslash = gTrue;
  89       } else if (!backslash && c == ')') {
  90         break;
  91       } else {
  92         backslash = gFalse;
  93       }
  94     }
  95   } else if (c == '<') {
  96     while ((c = lookChar()) != EOF) {
  97       getChar();
  98       if (i < size - 1) {
  99         buf[i++] = c;
 100       }
 101       if (c == '>') {
 102         break;
 103       }
 104     }
 105   } else if (c != '[' && c != ']') {
 106     while ((c = lookChar()) != EOF && !specialChars[c]) {
 107       getChar();
 108       if (i < size - 1) {
 109         buf[i++] = c;
 110       }
 111     }
 112   }
 113   buf[i] = '\0';
 114   *length = i;
 115
 116   return gTrue;
 117 }
 118
 119 int PSTokenizer::lookChar() {
 120   if (charBuf < 0) {
 121     charBuf = (*getCharFunc)(data);
 122   }
 123   return charBuf;
 124 }
 125
 126 int PSTokenizer::getChar() {
 127   int c;
 128
 129   if (charBuf < 0) {
 130     charBuf = (*getCharFunc)(data);
 131   }
 132   c = charBuf;
 133   charBuf = -1;
 134   return c;
 135 }