thirdparty/libantlr3c-3.2/src/antlr3ucs2inputstream.c

   1 /// \file
   2 /// Base functions to initialize and manipulate a UCS2 input stream
   3 ///
   4 #include    <antlr3input.h>
   5
   6 // [The "BSD licence"]
   7 // Copyright (c) 2005-2009 Jim Idle, Temporal Wave LLC
   8 // http://www.temporal-wave.com
   9 // http://www.linkedin.com/in/jimidle
  10 //
  11 // All rights reserved.
  12 //
  13 // Redistribution and use in source and binary forms, with or without
  14 // modification, are permitted provided that the following conditions
  15 // are met:
  16 // 1. Redistributions of source code must retain the above copyright
  17 //    notice, this list of conditions and the following disclaimer.
  18 // 2. Redistributions in binary form must reproduce the above copyright
  19 //    notice, this list of conditions and the following disclaimer in the
  20 //    documentation and/or other materials provided with the distribution.
  21 // 3. The name of the author may not be used to endorse or promote products
  22 //    derived from this software without specific prior written permission.
  23 //
  24 // THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR
  25 // IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
  26 // OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.
  27 // IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT,
  28 // INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
  29 // NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
  30 // DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
  31 // THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
  32 // (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
  33 // THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
  34
  35 // INT Stream API
  36 //
  37 static      void                        antlr3UCS2Consume       (pANTLR3_INT_STREAM is);
  38 static      ANTLR3_UCHAR    antlr3UCS2LA                (pANTLR3_INT_STREAM is, ANTLR3_INT32 la);
  39 static      ANTLR3_MARKER   antlr3UCS2Index             (pANTLR3_INT_STREAM is);
  40 static      void                        antlr3UCS2Seek          (pANTLR3_INT_STREAM is, ANTLR3_MARKER seekPoint);
  41
  42 // ucs2 Charstream API functions
  43 //
  44 static      pANTLR3_STRING      antlr3UCS2Substr        (pANTLR3_INPUT_STREAM input, ANTLR3_MARKER start, ANTLR3_MARKER stop);
  45
  46 /// \brief Common function to setup function interface for a 16 bit "UCS2" input stream.
  47 ///
  48 /// \param input Input stream context pointer
  49 ///
  50 /// \remark
  51 ///  - Strictly speaking, there is no such thing as a UCS2 input stream as the term
  52 ///    tends to confuse the notions of character encoding, unicode and so on. However
  53 ///    because there will possibly be a need for a UTF-16 stream, I needed to identify 16 bit
  54 ///    streams that do not support surrogate encodings and UCS2 is how it is mostly referred to.
  55 ///    For instance Java, Oracle and others use a 16 bit encoding of characters and so this type
  56 ///    of stream is very common.
  57 ///    Take it to mean, therefore, a straight 16 bit uncomplicated encoding of Unicode code points.
  58 ///
  59 void
  60 antlr3UCS2SetupStream   (pANTLR3_INPUT_STREAM input, ANTLR3_UINT32 type)
  61 {
  62     // Build a string factory for this stream. This is a 16 bit string "UCS2" factory which is a standard
  63     // part of the ANTLR3 string. The string factory is then passed through the whole chain of lexer->parser->tree->treeparser
  64     // and so on.
  65     //
  66     input->strFactory   = antlr3UCS2StringFactoryNew();
  67
  68     // Install function pointers for an 8 bit ASCII input, which are good for almost
  69     // all input stream functions. We will then override those that won't work for 16 bit characters.
  70     //
  71     antlr3GenericSetupStream    (input, type);
  72
  73     // Intstream API overrides for UCS2
  74     //
  75     input->istream->consume         =  antlr3UCS2Consume;           // Consume the next 16 bit character in the buffer
  76     input->istream->_LA             =  antlr3UCS2LA;                // Return the UTF32 character at offset n (1 based)
  77     input->istream->index           =  antlr3UCS2Index;             // Calculate current index in input stream, 16 bit based
  78     input->istream->seek            =  antlr3UCS2Seek;              // How to seek to a specific point in the stream
  79
  80     // Charstream API overrides for UCS2
  81     //
  82     input->substr                   =  antlr3UCS2Substr;            // Return a string from the input stream
  83
  84         input->charByteSize                             = 2;                            // Size in bytes of characters in this stream.
  85
  86 }
  87
  88 /// \brief Consume the next character in an 8 bit ASCII input stream
  89 ///
  90 /// \param input Input stream context pointer
  91 ///
  92 static void
  93 antlr3UCS2Consume(pANTLR3_INT_STREAM is)
  94 {
  95         pANTLR3_INPUT_STREAM input;
  96
  97         input   = ((pANTLR3_INPUT_STREAM) (is->super));
  98
  99         if      ((pANTLR3_UINT16)(input->nextChar) < (((pANTLR3_UINT16)input->data) + input->sizeBuf))
 100         {
 101                 // Indicate one more character in this line
 102                 //
 103                 input->charPositionInLine++;
 104
 105                 if  ((ANTLR3_UCHAR)(*((pANTLR3_UINT16)input->nextChar)) == input->newlineChar)
 106                 {
 107                         // Reset for start of a new line of input
 108                         //
 109                         input->line++;
 110                         input->charPositionInLine       = 0;
 111                         input->currentLine              = (void *)(((pANTLR3_UINT16)input->nextChar) + 1);
 112                 }
 113
 114                 // Increment to next character position
 115                 //
 116                 input->nextChar = (void *)(((pANTLR3_UINT16)input->nextChar) + 1);
 117         }
 118 }
 119
 120 /// \brief Return the input element assuming an 8 bit ascii input
 121 ///
 122 /// \param[in] input Input stream context pointer
 123 /// \param[in] la 1 based offset of next input stream element
 124 ///
 125 /// \return Next input character in internal ANTLR3 encoding (UTF32)
 126 ///
 127 static ANTLR3_UCHAR
 128 antlr3UCS2LA(pANTLR3_INT_STREAM is, ANTLR3_INT32 la)
 129 {
 130         pANTLR3_INPUT_STREAM input;
 131
 132         input   = ((pANTLR3_INPUT_STREAM) (is->super));
 133
 134         if      (( ((pANTLR3_UINT16)input->nextChar) + la - 1) >= (((pANTLR3_UINT16)input->data) + input->sizeBuf))
 135         {
 136                 return  ANTLR3_CHARSTREAM_EOF;
 137         }
 138         else
 139         {
 140                 return  (ANTLR3_UCHAR)(*((pANTLR3_UINT16)input->nextChar + la - 1));
 141         }
 142 }
 143
 144
 145 /// \brief Calculate the current index in the output stream.
 146 /// \param[in] input Input stream context pointer
 147 ///
 148 static ANTLR3_MARKER
 149 antlr3UCS2Index(pANTLR3_INT_STREAM is)
 150 {
 151     pANTLR3_INPUT_STREAM input;
 152
 153     input   = ((pANTLR3_INPUT_STREAM) (is->super));
 154
 155     return  (ANTLR3_MARKER)(input->nextChar);
 156 }
 157
 158 /// \brief Rewind the lexer input to the state specified by the supplied mark.
 159 ///
 160 /// \param[in] input Input stream context pointer
 161 ///
 162 /// \remark
 163 /// Assumes ASCII (or at least, 8 Bit) input stream.
 164 ///
 165 static void
 166 antlr3UCS2Seek  (pANTLR3_INT_STREAM is, ANTLR3_MARKER seekPoint)
 167 {
 168         ANTLR3_INT32   count;
 169         pANTLR3_INPUT_STREAM input;
 170
 171         input   = ((pANTLR3_INPUT_STREAM) is->super);
 172
 173         // If the requested seek point is less than the current
 174         // input point, then we assume that we are resetting from a mark
 175         // and do not need to scan, but can just set to there.
 176         //
 177         if      (seekPoint <= (ANTLR3_MARKER)(input->nextChar))
 178         {
 179                 input->nextChar = (void *)seekPoint;
 180         }
 181         else
 182         {
 183                 count   = (ANTLR3_UINT32)((seekPoint - (ANTLR3_MARKER)(input->nextChar)) / 2); // 16 bits per character in UCS2
 184
 185                 while (count--)
 186                 {
 187                         is->consume(is);
 188                 }
 189         }
 190 }
 191 /// \brief Return a substring of the ucs2 (16 bit) input stream in
 192 ///  newly allocated memory.
 193 ///
 194 /// \param input Input stream context pointer
 195 /// \param start Offset in input stream where the string starts
 196 /// \param stop  Offset in the input stream where the string ends.
 197 ///
 198 static pANTLR3_STRING
 199 antlr3UCS2Substr                (pANTLR3_INPUT_STREAM input, ANTLR3_MARKER start, ANTLR3_MARKER stop)
 200 {
 201     return  input->strFactory->newPtr(input->strFactory, (pANTLR3_UINT8)start, ((ANTLR3_UINT32_CAST(stop - start))/2) + 1);
 202 }