pdf2swf/xpdf/PSTokenizer.cc

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