xmlentityscanner.java

来自「JAVA 所有包」· Java 代码 · 共 1,674 行 · 第 1/5 页

     * @see com.sun.org.apache.xerces.internal.xni.XMLResourceIdentifier#setExpandedSystemId(String)
     */
    public void setExpandedSystemId(String systemId) {
        //no-op
    }
    
    /** Returns the literal system identifier.  */
    public String getLiteralSystemId() {
        return (fCurrentEntity != null && fCurrentEntity.entityLocation != null) ? fCurrentEntity.entityLocation.getLiteralSystemId() : null;
    }
    
    /**
     * @see com.sun.org.apache.xerces.internal.xni.XMLResourceIdentifier#setLiteralSystemId(String)
     */
    public void setLiteralSystemId(String systemId) {
        //no-op
    }
    
    /** Returns the public identifier.  */
    public String getPublicId() {
        return (fCurrentEntity != null && fCurrentEntity.entityLocation != null) ? fCurrentEntity.entityLocation.getPublicId() : null;
    }
    
    /**
     * @see com.sun.org.apache.xerces.internal.xni.XMLResourceIdentifier#setPublicId(String)
     */
    public void setPublicId(String publicId) {
        //no-op
    }
    
    ///////////////// Locator methods finished.
    
    /** the version of the current entity being scanned */
    public void setVersion(String version){
        fCurrentEntity.version = version;
    }
    
    public String getVersion(){
        return fCurrentEntity.version ;
    }
    
    public String getEncoding(){
        return fCurrentEntity.encoding ;
    }
    /**
     * Sets the encoding of the scanner. This method is used by the
     * scanners if the XMLDecl or TextDecl line contains an encoding
     * pseudo-attribute.
     * <p>
     * <strong>Note:</strong> The underlying character reader on the
     * current entity will be changed to accomodate the new encoding.
     * However, the new encoding is ignored if the current reader was
     * not constructed from an input stream (e.g. an external entity
     * that is resolved directly to the appropriate java.io.Reader
     * object).
     *
     * @param encoding The IANA encoding name of the new encoding.
     *
     * @throws IOException Thrown if the new encoding is not supported.
     *
     * @see com.sun.org.apache.xerces.internal.util.EncodingMap
     */
    public void setEncoding(String encoding) throws IOException {
        
        if (DEBUG_ENCODINGS) {
            System.out.println("$$$ setEncoding: "+encoding);
        }
        
        if (fCurrentEntity.stream != null) {
            // if the encoding is the same, don't change the reader and
            // re-use the original reader used by the OneCharReader
            // NOTE: Besides saving an object, this overcomes deficiencies
            //       in the UTF-16 reader supplied with the standard Java
            //       distribution (up to and including 1.3). The UTF-16
            //       decoder buffers 8K blocks even when only asked to read
            //       a single char! -Ac
            if (fCurrentEntity.encoding == null ||
                    !fCurrentEntity.encoding.equals(encoding)) {
                // UTF-16 is a bit of a special case.  If the encoding is UTF-16,
                // and we know the endian-ness, we shouldn't change readers.
                // If it's ISO-10646-UCS-(2|4), then we'll have to deduce
                // the endian-ness from the encoding we presently have.
                if(fCurrentEntity.encoding != null && fCurrentEntity.encoding.startsWith("UTF-16")) {
                    String ENCODING = encoding.toUpperCase(Locale.ENGLISH);
                    if(ENCODING.equals("UTF-16")) return;
                    if(ENCODING.equals("ISO-10646-UCS-4")) {
                        if(fCurrentEntity.encoding.equals("UTF-16BE")) {
                            fCurrentEntity.reader = new UCSReader(fCurrentEntity.stream, UCSReader.UCS4BE);
                        } else {
                            fCurrentEntity.reader = new UCSReader(fCurrentEntity.stream, UCSReader.UCS4LE);
                        }
                        return;
                    }
                    if(ENCODING.equals("ISO-10646-UCS-2")) {
                        if(fCurrentEntity.encoding.equals("UTF-16BE")) {
                            fCurrentEntity.reader = new UCSReader(fCurrentEntity.stream, UCSReader.UCS2BE);
                        } else {
                            fCurrentEntity.reader = new UCSReader(fCurrentEntity.stream, UCSReader.UCS2LE);
                        }
                        return;
                    }
                }
                // wrap a new reader around the input stream, changing
                // the encoding
                if (DEBUG_ENCODINGS) {
                    System.out.println("$$$ creating new reader from stream: "+
                            fCurrentEntity.stream);
                }
                //fCurrentEntity.stream.reset();
                fCurrentEntity.reader = createReader(fCurrentEntity.stream, encoding, null);
                fCurrentEntity.encoding = encoding;
                     
            } else {
                if (DEBUG_ENCODINGS)
                    System.out.println("$$$ reusing old reader on stream");
            }
        }
        
    } // setEncoding(String)
    
    /** Returns true if the current entity being scanned is external. */
    public boolean isExternal() {
        return fCurrentEntity.isExternal();
    } // isExternal():boolean
    
    public int getChar(int relative) throws IOException{
        if(arrangeCapacity(relative + 1, false)){
            return fCurrentEntity.ch[fCurrentEntity.position + relative];
        }else{
            return -1;
        }
    }//getChar()
    
    /**
     * Returns the next character on the input.
     * <p>
     * <strong>Note:</strong> The character is <em>not</em> consumed.
     *
     * @throws IOException  Thrown if i/o error occurs.
     * @throws EOFException Thrown on end of file.
     */
    public int peekChar() throws IOException {
        if (DEBUG_BUFFER) {
            System.out.print("(peekChar: ");
            print();
            System.out.println();
        }
        
        // load more characters, if needed
        if (fCurrentEntity.position == fCurrentEntity.count) {
            invokeListeners(0);
            load(0, true);
        }
        
        // peek at character
        int c = fCurrentEntity.ch[fCurrentEntity.position];
        
        // return peeked character
        if (DEBUG_BUFFER) {
            System.out.print(")peekChar: ");
            print();
            if (isExternal) {
                System.out.println(" -> '"+(c!='\r'?(char)c:'\n')+"'");
            } else {
                System.out.println(" -> '"+(char)c+"'");
            }
        }
        if (isExternal) {
            return c != '\r' ? c : '\n';
        } else {
            return c;
        }
        
    } // peekChar():int
    
    /**
     * Returns the next character on the input.
     * <p>
     * <strong>Note:</strong> The character is consumed.
     *
     * @throws IOException  Thrown if i/o error occurs.
     * @throws EOFException Thrown on end of file.
     */
    public int scanChar() throws IOException {
        if (DEBUG_BUFFER) {
            System.out.print("(scanChar: ");
            print();
            System.out.println();
        }
        
        // load more characters, if needed
        if (fCurrentEntity.position == fCurrentEntity.count) {
            invokeListeners(0);
            load(0, true);
        }
        
        // scan character
        int c = fCurrentEntity.ch[fCurrentEntity.position++];
        if (c == '\n' ||
                (c == '\r' && isExternal)) {
            fCurrentEntity.lineNumber++;
            fCurrentEntity.columnNumber = 1;
            if (fCurrentEntity.position == fCurrentEntity.count) {
                invokeListeners(1);
                fCurrentEntity.ch[0] = (char)c;
                load(1, false);
            }
            if (c == '\r' && isExternal) {
                if (fCurrentEntity.ch[fCurrentEntity.position++] != '\n') {
                    fCurrentEntity.position--;
                }
                c = '\n';
            }
        }
        
        // return character that was scanned
        if (DEBUG_BUFFER) {
            System.out.print(")scanChar: ");
            print();
            System.out.println(" -> '"+(char)c+"'");
        }
        fCurrentEntity.columnNumber++;
        return c;
        
    } // scanChar():int
    
    /**
     * Returns a string matching the NMTOKEN production appearing immediately
     * on the input as a symbol, or null if NMTOKEN Name string is present.
     * <p>
     * <strong>Note:</strong> The NMTOKEN characters are consumed.
     * <p>
     * <strong>Note:</strong> The string returned must be a symbol. The
     * SymbolTable can be used for this purpose.
     *
     * @throws IOException  Thrown if i/o error occurs.
     * @throws EOFException Thrown on end of file.
     *
     * @see com.sun.org.apache.xerces.internal.util.SymbolTable
     * @see com.sun.org.apache.xerces.internal.util.XMLChar#isName
     */
    public String scanNmtoken() throws IOException {
        if (DEBUG_BUFFER) {
            System.out.print("(scanNmtoken: ");
            print();
            System.out.println();
        }
        
        // load more characters, if needed
        if (fCurrentEntity.position == fCurrentEntity.count) {
            invokeListeners(0);
            load(0, true);
        }
        
        // scan nmtoken
        int offset = fCurrentEntity.position;
        boolean vc = false;
        char c;
        while (true){
            //while (XMLChar.isName(fCurrentEntity.ch[fCurrentEntity.position])) {
            c = fCurrentEntity.ch[fCurrentEntity.position];
            if(c < 127){
                vc = VALID_NAMES[c];
            }else{
                vc = XMLChar.isName(c);
            }
            if(!vc)break;
            
            if (++fCurrentEntity.position == fCurrentEntity.count) {
                int length = fCurrentEntity.position - offset;
                invokeListeners(length);
                if (length == fCurrentEntity.fBufferSize) {
                    // bad luck we have to resize our buffer
                    char[] tmp = new char[fCurrentEntity.fBufferSize * 2];
                    System.arraycopy(fCurrentEntity.ch, offset,
                            tmp, 0, length);
                    fCurrentEntity.ch = tmp;
                    fCurrentEntity.fBufferSize *= 2;
                } else {
                    System.arraycopy(fCurrentEntity.ch, offset,
                            fCurrentEntity.ch, 0, length);
                }
                offset = 0;
                if (load(length, false)) {
                    break;
                }
            }
        }
        int length = fCurrentEntity.position - offset;
        fCurrentEntity.columnNumber += length;
        
        // return nmtoken
        String symbol = null;
        if (length > 0) {
            symbol = fSymbolTable.addSymbol(fCurrentEntity.ch, offset, length);
        }
        if (DEBUG_BUFFER) {
            System.out.print(")scanNmtoken: ");
            print();
            System.out.println(" -> "+String.valueOf(symbol));
        }
        return symbol;
        
    } // scanNmtoken():String
    
    /**
     * Returns a string matching the Name production appearing immediately
     * on the input as a symbol, or null if no Name string is present.
     * <p>
     * <strong>Note:</strong> The Name characters are consumed.
     * <p>
     * <strong>Note:</strong> The string returned must be a symbol. The
     * SymbolTable can be used for this purpose.
     *
     * @throws IOException  Thrown if i/o error occurs.
     * @throws EOFException Thrown on end of file.
     *
     * @see com.sun.org.apache.xerces.internal.util.SymbolTable
     * @see com.sun.org.apache.xerces.internal.util.XMLChar#isName
     * @see com.sun.org.apache.xerces.internal.util.XMLChar#isNameStart
     */
    public String scanName() throws IOException {
        if (DEBUG_BUFFER) {
            System.out.print("(scanName: ");
            print();
            System.out.println();
        }
        
        // load more characters, if needed
        if (fCurrentEntity.position == fCurrentEntity.count) {
            invokeListeners(0);
            load(0, true);
        }
        
        // scan name