xmlparser.java

linux下建立JAVA虚拟机的源码KAFFE

                stream.close();
              }
            catch (IOException e)
              {
                /* ignore */
              }
          }
        if (is != null)
          {
            try
              {
                is.close();
              }
            catch (IOException e)
              {
                /* ignore */
              }
          }
        scratch = null;
      }
  }

  //////////////////////////////////////////////////////////////////////
  // Error reporting.
  //////////////////////////////////////////////////////////////////////
    
  /**
   * Report an error.
   * @param message The error message.
   * @param textFound The text that caused the error (or null).
   * @see SAXDriver#error
   * @see #line
   */
  private void error(String message, String textFound, String textExpected)
    throws SAXException
  {
    if (textFound != null)
      {
        message = message + " (found \"" + textFound + "\")";
      }
    if (textExpected != null)
      {
        message = message + " (expected \"" + textExpected + "\")";
      }
    handler.fatal(message);
    
    // "can't happen"
    throw new SAXException(message);
  }

  /**
   * Report a serious error.
   * @param message The error message.
   * @param textFound The text that caused the error (or null).
   */
  private void error(String message, char textFound, String textExpected)
    throws SAXException
  {
    error(message, new Character(textFound).toString(), textExpected);
  }

  /**
   * Report typical case fatal errors.
   */
  private void error(String message)
    throws SAXException
  {
    handler.fatal(message);
  }

  //////////////////////////////////////////////////////////////////////
  // Major syntactic productions.
  //////////////////////////////////////////////////////////////////////

  /**
   * Parse an XML document.
   * <pre>
   * [1] document ::= prolog element Misc*
   * </pre>
   * <p>This is the top-level parsing function for a single XML
   * document.  As a minimum, a well-formed document must have
   * a document element, and a valid document must have a prolog
   * (one with doctype) as well.
   */
  private void parseDocument()
    throws Exception
  {
    try
      {                                       // added by MHK
        boolean sawDTD = parseProlog();
        require('<');
        parseElement(!sawDTD);
      }
    catch (EOFException ee)
      {                 // added by MHK
        error("premature end of file", "[EOF]", null);
      }
    
    try
      {
        parseMisc();   //skip all white, PIs, and comments
        char c = readCh();    //if this doesn't throw an exception...
        error("unexpected characters after document end", c, null);
      }
    catch (EOFException e)
      {
        return;
      }
  }
  
  static final char[] startDelimComment = { '<', '!', '-', '-' };
  static final char[] endDelimComment = { '-', '-' };

  /**
   * Skip a comment.
   * <pre>
   * [15] Comment ::= '&lt;!--' ((Char - '-') | ('-' (Char - '-')))* "-->"
   * </pre>
   * <p> (The <code>&lt;!--</code> has already been read.)
   */
  private void parseComment()
    throws Exception
  {
    char c;
    boolean saved = expandPE;
    
    expandPE = false;
    parseUntil(endDelimComment);
    require('>');
    expandPE = saved;
    handler.comment(dataBuffer, 0, dataBufferPos);
    dataBufferPos = 0;
  }
  
  static final char[] startDelimPI = { '<', '?' };
  static final char[] endDelimPI = { '?', '>' };

  /**
   * Parse a processing instruction and do a call-back.
   * <pre>
   * [16] PI ::= '&lt;?' PITarget
   *    (S (Char* - (Char* '?&gt;' Char*)))?
   *    '?&gt;'
   * [17] PITarget ::= Name - ( ('X'|'x') ('M'|m') ('L'|l') )
   * </pre>
   * <p> (The <code>&lt;?</code> has already been read.)
   */
  private void parsePI()
    throws SAXException, IOException
  {
    String name;
    boolean saved = expandPE;
    
    expandPE = false;
    name = readNmtoken(true);
    //NE08
    if (name.indexOf(':') >= 0)
      {
        error("Illegal character(':') in processing instruction name ",
              name, null);
      }
    if ("xml".equalsIgnoreCase(name))
      {
        error("Illegal processing instruction target", name, null);
      }
    if (!tryRead(endDelimPI))
      {
        requireWhitespace();
        parseUntil(endDelimPI);
      }
    expandPE = saved;
    handler.processingInstruction(name, dataBufferToString());
  }
  
  static final char[] endDelimCDATA = { ']', ']', '>' };

  private boolean isDirtyCurrentElement;

  /**
   * Parse a CDATA section.
   * <pre>
   * [18] CDSect ::= CDStart CData CDEnd
   * [19] CDStart ::= '&lt;![CDATA['
   * [20] CData ::= (Char* - (Char* ']]&gt;' Char*))
   * [21] CDEnd ::= ']]&gt;'
   * </pre>
   * <p> (The '&lt;![CDATA[' has already been read.)
   */
  private void parseCDSect()
    throws Exception
  {
    parseUntil(endDelimCDATA);
    dataBufferFlush();
  }

  /**
   * Parse the prolog of an XML document.
   * <pre>
   * [22] prolog ::= XMLDecl? Misc* (Doctypedecl Misc*)?
   * </pre>
   * <p>We do not look for the XML declaration here, because it was
   * handled by pushURL ().
   * @see pushURL
   * @return true if a DTD was read.
   */
  private boolean parseProlog()
    throws Exception
  {
    parseMisc();

    if (tryRead("<!DOCTYPE"))
      {
        parseDoctypedecl();
        parseMisc();
        return true;
      }
    return false;
  }

  private void checkLegalVersion(String version)
    throws SAXException
  {
    int len = version.length();
    for (int i = 0; i < len; i++)
      {
        char c = version.charAt(i);
        if ('0' <= c && c <= '9')
          {
            continue;
          }
        if (c == '_' || c == '.' || c == ':' || c == '-')
          {
            continue;
          }
        if ('a' <= c && c <= 'z')
          {
            continue;
          }
        if ('A' <= c && c <= 'Z')
          {
            continue;
          }
        error ("illegal character in version", version, "1.0");
      }
  }

  /**
   * Parse the XML declaration.
   * <pre>
   * [23] XMLDecl ::= '&lt;?xml' VersionInfo EncodingDecl? SDDecl? S? '?&gt;'
   * [24] VersionInfo ::= S 'version' Eq
   *    ("'" VersionNum "'" | '"' VersionNum '"' )
   * [26] VersionNum ::= ([a-zA-Z0-9_.:] | '-')*
   * [32] SDDecl ::= S 'standalone' Eq
   *    ( "'"" ('yes' | 'no') "'"" | '"' ("yes" | "no") '"' )
   * [80] EncodingDecl ::= S 'encoding' Eq
   *    ( "'" EncName "'" | "'" EncName "'" )
   * [81] EncName ::= [A-Za-z] ([A-Za-z0-9._] | '-')*
   * </pre>
   * <p> (The <code>&lt;?xml</code> and whitespace have already been read.)
   * @return the encoding in the declaration, uppercased; or null
   * @see #parseTextDecl
   * @see #setupDecoding
   */
  private String parseXMLDecl(boolean ignoreEncoding)
    throws SAXException, IOException
  {
    String version;
    String encodingName = null;
    String standalone = null;
    int flags = LIT_DISABLE_CREF | LIT_DISABLE_PE | LIT_DISABLE_EREF;
    String inputEncoding = null;
        
    switch (this.encoding)
      {
      case ENCODING_EXTERNAL:
      case ENCODING_UTF_8:
        inputEncoding = "UTF-8";
        break;
      case ENCODING_ISO_8859_1:
        inputEncoding = "ISO-8859-1";
        break;
      case ENCODING_UCS_2_12:
        inputEncoding = "UTF-16BE";
        break;
      case ENCODING_UCS_2_21:
        inputEncoding = "UTF-16LE";
        break;
      }
    
    // Read the version.
    require("version");
    parseEq();
    checkLegalVersion(version = readLiteral(flags));
    if (!version.equals("1.0"))
      {
        if (version.equals("1.1"))
          {
            handler.warn("expected XML version 1.0, not: " + version);
            xmlVersion = XML_11;
          }
        else
          {
            error("illegal XML version", version, "1.0 or 1.1");
          }
      }
    else
      {
        xmlVersion = XML_10;
      }
    // Try reading an encoding declaration.
    boolean white = tryWhitespace();
    
    if (tryRead("encoding"))
      {
        if (!white)
          {
            error("whitespace required before 'encoding='");
          }
        parseEq();
        encodingName = readLiteral(flags);
        if (!ignoreEncoding)
          {
            setupDecoding(encodingName);
          }
      }
    
    // Try reading a standalone declaration
    if (encodingName != null)
      {
        white = tryWhitespace();
      }
    if (tryRead("standalone"))
      {
        if (!white)
          {
            error("whitespace required before 'standalone='");
          }
        parseEq();
        standalone = readLiteral(flags);
        if ("yes".equals(standalone))
          {
            docIsStandalone = true;
          }
        else if (!"no".equals(standalone))
          {
            error("standalone flag must be 'yes' or 'no'");
          }
      }

    skipWhitespace();
    require("?>");

    if (inputEncoding == null)
      {
        inputEncoding = encodingName;
      }
    return encodingName;
  }

  /**
   * Parse a text declaration.
   * <pre>
   * [79] TextDecl ::= '&lt;?xml' VersionInfo? EncodingDecl S? '?&gt;'
   * [80] EncodingDecl ::= S 'encoding' Eq
   *    ( '"' EncName '"' | "'" EncName "'" )
   * [81] EncName ::= [A-Za-z] ([A-Za-z0-9._] | '-')*
   * </pre>
   * <p> (The <code>&lt;?xml</code>' and whitespace have already been read.)
   * @return the encoding in the declaration, uppercased; or null
   * @see #parseXMLDecl
   * @see #setupDecoding
   */
  private String parseTextDecl(boolean ignoreEncoding)
    throws SAXException, IOException
  {
    String encodingName = null;
    int flags = LIT_DISABLE_CREF | LIT_DISABLE_PE | LIT_DISABLE_EREF;

    // Read an optional version.
    if (tryRead ("version"))
      {
        String version;
        parseEq();
        checkLegalVersion(version = readLiteral(flags));
        
        if (version.equals("1.1"))
          {
            if (xmlVersion == XML_10)
              {
                error("external subset has later version number.", "1.0",
                      version);    
              }
            handler.warn("expected XML version 1.0, not: " + version);
            xmlVersion = XML_11;
          }
        else if (!version.equals("1.0"))
          {
            error("illegal XML version", version, "1.0 or 1.1");
          }
        requireWhitespace();
      }
    
    // Read the encoding.
    require("encoding");
    parseEq();
    encodingName = readLiteral(flags);
    if (!ignoreEncoding)
      {
        setupDecoding(encodingName);
      }
    skipWhitespace();
    require("?>");
    
    return encodingName;
  }

  /**
   * Sets up internal state so that we can decode an entity using the
   * specified encoding.  This is used when we start to read an entity
   * and we have been given knowledge of its encoding before we start to
   * read any data (e.g. from a SAX input source or from a MIME type).
   *
   * <p> It is also used after autodetection, at which point only very
   * limited adjustments to the encoding may be used (switching between
   * related builtin decoders).
   *
   * @param encodingName The name of the encoding specified by the user.
   * @exception IOException if the encoding isn't supported either
   *  internally to this parser, or by the hosting JVM.
   * @see #parseXMLDecl
   * @see #parseTextDecl
     */
  private void setupDecoding(String encodingName)
    throws SAXException, IOException
  {
    encodingName = encodingName.toUpperCase();
    
    // ENCODING_EXTERNAL indicates an encoding that wasn't
    // autodetected ... we can use builtin decoders, or
    // ones from the JVM (InputStreamReader).
    
    // Otherwise we can only tweak what was autodetected, and
    // only for single byte (ASCII derived) builtin encodings.
    
    // ASCII-derived encodings
    if (encoding == ENCODING_UTF_8 || encoding == ENCODING_EXTERNAL)
      {
        if (encodingName.equals("ISO-8859-1")
            || encodingName.equals("8859_1")
            || encodingName.equals("ISO8859_1"))
          {
            encoding = ENCODING_ISO_8859_1;
            return;
          }
        else if (encodingName.equals("US-ASCII")
                 || encodingName.equals("ASCII"))
          {
            encoding = ENCODING_ASCII;
            return;
          }
        else if (encodingName.equals("UTF-8")
                 || encodingName.equals("UTF8"))
          {
            encoding = ENCODING_UTF_8;
            return;
          }
        else if (encoding != ENCODING_EXTERNAL)
          {
            // used to start with a new reader ...
            throw new UnsupportedEncodingException(encodingName);
          }
        // else fallthrough ...
        // it's ASCII-ish and something other than a builtin
      }
    
    // Unicode and such
    if (encoding == ENCODING_UCS_2_12 || encoding == ENCODING_UCS_2_21)
      {
        if (!(encodingName.equals("ISO-10646-UCS-2")
              || encodingName.equals("UTF-16")
              || encodingName.equals("UTF-16BE")
              || encodingName.equals("UTF-16LE")))
          {
            error("unsupported Unicode encoding", encodingName, "UTF-16");
          }