source/libs/poppler/poppler-src/poppler/PSTokenizer.cc

   1 //========================================================================
   2 //
   3 // PSTokenizer.cc
   4 //
   5 // Copyright 2002-2003 Glyph & Cog, LLC
   6 //
   7 //========================================================================
   8
   9 //========================================================================
  10 //
  11 // Modified under the Poppler project - http://poppler.freedesktop.org
  12 //
  13 // All changes made under the Poppler project to this file are licensed
  14 // under GPL version 2 or later
  15 //
  16 // Copyright (C) 2006 Scott Turner <scotty1024@mac.com>
  17 // Copyright (C) 2008 Albert Astals Cid <aacid@kde.org>
  18 //
  19 // To see a description of the changes please see the Changelog file that
  20 // came with your tarball or type make ChangeLog if you are building from git
  21 //
  22 //========================================================================
  23
  24 #include <config.h>
  25
  26 #ifdef USE_GCC_PRAGMAS
  27 #pragma implementation
  28 #endif
  29
  30 #include <stdio.h>
  31 #include <stdlib.h>
  32 #include "PSTokenizer.h"
  33
  34 //------------------------------------------------------------------------
  35
  36 // A '1' in this array means the character is white space.  A '1' or
  37 // '2' means the character ends a name or command.
  38 static const char specialChars[256] = {
  39   1, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 0, 1, 1, 0, 0,   // 0x
  40   0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,   // 1x
  41   1, 0, 0, 0, 0, 2, 0, 0, 2, 2, 0, 0, 0, 0, 0, 2,   // 2x
  42   0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 2, 0, 2, 0,   // 3x
  43   0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,   // 4x
  44   0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 2, 0, 2, 0, 0,   // 5x
  45   0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,   // 6x
  46   0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 2, 0, 2, 0, 0,   // 7x
  47   0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,   // 8x
  48   0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,   // 9x
  49   0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,   // ax
  50   0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,   // bx
  51   0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,   // cx
  52   0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,   // dx
  53   0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,   // ex
  54   0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0    // fx
  55 };
  56
  57 //------------------------------------------------------------------------
  58
  59 PSTokenizer::PSTokenizer(int (*getCharFuncA)(void *), void *dataA) {
  60   getCharFunc = getCharFuncA;
  61   data = dataA;
  62   charBuf = -1;
  63 }
  64
  65 PSTokenizer::~PSTokenizer() {
  66 }
  67
  68 GBool PSTokenizer::getToken(char *buf, int size, int *length) {
  69   GBool comment, backslash;
  70   int c;
  71   int i;
  72
  73   // skip leading whitespace and comments
  74   comment = gFalse;
  75   while (1) {
  76     if ((c = getChar()) == EOF) {
  77       buf[0] = '\0';
  78       *length = 0;
  79       return gFalse;
  80     }
  81     if (comment) {
  82       if (c == '\x0a' || c == '\x0d') {
  83         comment = gFalse;
  84       }
  85     } else if (c == '%') {
  86       comment = gTrue;
  87     } else if (specialChars[c] != 1) {
  88       break;
  89     }
  90   }
  91
  92   // Reserve room for terminating '\0'
  93   size--;
  94
  95   // read a token
  96   i = 0;
  97   buf[i++] = c;
  98   if (c == '(') {
  99     backslash = gFalse;
 100     while ((c = lookChar()) != EOF) {
 101       consumeChar();
 102       if (i < size) {
 103         buf[i++] = c;
 104       }
 105       if (c == '\\') {
 106         backslash = gTrue;
 107       } else if (!backslash && c == ')') {
 108         break;
 109       } else {
 110         backslash = gFalse;
 111       }
 112     }
 113   } else if (c == '<') {
 114     while ((c = lookChar()) != EOF) {
 115       consumeChar();
 116       if (i < size && specialChars[c] != 1) {
 117         buf[i++] = c;
 118       }
 119       if (c == '>') {
 120         break;
 121       }
 122     }
 123   } else if (c != '[' && c != ']') {
 124     while ((c = lookChar()) != EOF && !specialChars[c]) {
 125       consumeChar();
 126       if (i < size) {
 127         buf[i++] = c;
 128       }
 129     }
 130   }
 131   // Zero terminate token string
 132   buf[i] = '\0';
 133   // Return length of token
 134   *length = i;
 135
 136   return gTrue;
 137 }
 138
 139 int PSTokenizer::lookChar() {
 140   if (charBuf < 0) {
 141     charBuf = (*getCharFunc)(data);
 142   }
 143   return charBuf;
 144 }
 145
 146 void PSTokenizer::consumeChar() {
 147   charBuf = -1;
 148 }
 149
 150 int PSTokenizer::getChar() {
 151   int c = charBuf;
 152
 153   if (c < 0) {
 154     c = (*getCharFunc)(data);
 155   } else {
 156     charBuf = -1;
 157   }
 158   return c;
 159 }