xincludetextreader.java

来自「JAVA 所有包」· Java 代码 · 共 514 行 · 第 1/2 页

    }

    /** 
     * XMLEntityManager cares about endian-ness, since it creates its own optimized
     * readers. Since we're just using generic Java readers for now, we're not caring
     * about endian-ness.  If this changes, even more code needs to be copied from
     * XMLEntity manager. -- PJM
     */
    protected String getEncodingName(InputStream stream) throws IOException {
        final byte[] b4 = new byte[4];
        String encoding = null;

        // this has the potential to throw an exception
        // it will be fixed when we ensure the stream is rewindable (see note above)
        stream.mark(4);
        int count = stream.read(b4, 0, 4);
        stream.reset();
        if (count == 4) {
            encoding = getEncodingName(b4);
        }

        return encoding;
    }

    /**
     * Removes the byte order mark from the stream, if
     * it exists and returns the encoding name.
     * 
     * @param stream
     * @param encoding
     * @throws IOException
     */
    protected String consumeBOM(InputStream stream, String encoding)
        throws IOException {

        byte[] b = new byte[3];
        int count = 0;
        stream.mark(3);
        if (encoding.equals("UTF-8")) {
            count = stream.read(b, 0, 3);
            if (count == 3) {
                final int b0 = b[0] & 0xFF; 
                final int b1 = b[1] & 0xFF;
                final int b2 = b[2] & 0xFF;
                if (b0 != 0xEF || b1 != 0xBB || b2 != 0xBF) {
                    // First three bytes are not BOM, so reset.
                    stream.reset();
                }
            }
            else {
                stream.reset();
            }
        }
        else if (encoding.startsWith("UTF-16")) {
            count = stream.read(b, 0, 2);
            if (count == 2) {
                final int b0 = b[0] & 0xFF;
                final int b1 = b[1] & 0xFF;
                if (b0 == 0xFE && b1 == 0xFF) {
                    return "UTF-16BE";
                }
                else if (b0 == 0xFF && b1 == 0xFE) {
                    return "UTF-16LE";
                }
            }
            // First two bytes are not BOM, so reset.
            stream.reset();
        }
        // We could do UTF-32, but since the getEncodingName() doesn't support that
        // we won't support it here.
        // To implement UTF-32, look for:  00 00 FE FF for big-endian
        //                             or  FF FE 00 00 for little-endian
        return encoding;
    }

    /**
     * REVISIT: This code is taken from com.sun.org.apache.xerces.internal.impl.XMLEntityManager.
     *          Is there any way we can share the code, without having it implemented twice?
     *          I think we should make it public and static in XMLEntityManager. --PJM
     *
     * Returns the IANA encoding name that is auto-detected from
     * the bytes specified, with the endian-ness of that encoding where appropriate.
     *
     * @param b4    The first four bytes of the input.
     * @return the encoding name, or null if no encoding could be detected
     */
    protected String getEncodingName(byte[] b4) {

        // UTF-16, with BOM
        int b0 = b4[0] & 0xFF;
        int b1 = b4[1] & 0xFF;
        if (b0 == 0xFE && b1 == 0xFF) {
            // UTF-16, big-endian
            return "UTF-16BE";
        }
        if (b0 == 0xFF && b1 == 0xFE) {
            // UTF-16, little-endian
            return "UTF-16LE";
        }

        // UTF-8 with a BOM
        int b2 = b4[2] & 0xFF;
        if (b0 == 0xEF && b1 == 0xBB && b2 == 0xBF) {
            return "UTF-8";
        }

        // other encodings
        int b3 = b4[3] & 0xFF;
        if (b0 == 0x00 && b1 == 0x00 && b2 == 0x00 && b3 == 0x3C) {
            // UCS-4, big endian (1234)
            return "ISO-10646-UCS-4";
        }
        if (b0 == 0x3C && b1 == 0x00 && b2 == 0x00 && b3 == 0x00) {
            // UCS-4, little endian (4321)
            return "ISO-10646-UCS-4";
        }
        if (b0 == 0x00 && b1 == 0x00 && b2 == 0x3C && b3 == 0x00) {
            // UCS-4, unusual octet order (2143)
            return "ISO-10646-UCS-4";
        }
        if (b0 == 0x00 && b1 == 0x3C && b2 == 0x00 && b3 == 0x00) {
            // UCS-4, unusual octect order (3412)
            return "ISO-10646-UCS-4";
        }
        if (b0 == 0x00 && b1 == 0x3C && b2 == 0x00 && b3 == 0x3F) {
            // UTF-16, big-endian, no BOM
            // (or could turn out to be UCS-2...
            return "UTF-16BE";
        }
        if (b0 == 0x3C && b1 == 0x00 && b2 == 0x3F && b3 == 0x00) {
            // UTF-16, little-endian, no BOM
            // (or could turn out to be UCS-2...
            return "UTF-16LE";
        }
        if (b0 == 0x4C && b1 == 0x6F && b2 == 0xA7 && b3 == 0x94) {
            // EBCDIC
            // a la xerces1, return CP037 instead of EBCDIC here
            return "CP037";
        }

        // this signals us to use the value from the encoding attribute
        return null;

    } // getEncodingName(byte[]):Object[]

    /**
     * Read the input stream as text, and pass the text on to the XIncludeHandler
     * using calls to characters().  This will read all of the text it can from the
     * resource.
     * 
     * @throws IOException
     */
    public void parse() throws IOException {
        
        fReader = getReader(fSource);
        fSource = null;
        int readSize = fReader.read(fTempString.ch, 0, fTempString.ch.length - 1);
        while (readSize != -1) {
            for (int i = 0; i < readSize; ++i) {
                char ch = fTempString.ch[i];
                if (!isValid(ch)) {
                    if (XMLChar.isHighSurrogate(ch)) {
                        int ch2;
                        // retrieve next character
                        if (++i < readSize) {
                            ch2 = fTempString.ch[i];
                        }
                        // handle rare boundary case
                        else {
                            ch2 = fReader.read();
                            if (ch2 != -1) {
                                fTempString.ch[readSize++] = (char) ch2;
                            }
                        }
                        if (XMLChar.isLowSurrogate(ch2)) {
                            // convert surrogates to a supplemental character
                            int sup = XMLChar.supplemental(ch, (char)ch2);
                            if (!isValid(sup)) {
                                fErrorReporter.reportError(XMLMessageFormatter.XML_DOMAIN,
                                                           "InvalidCharInContent", 
                                                           new Object[] { Integer.toString(sup, 16) },
                                                           XMLErrorReporter.SEVERITY_FATAL_ERROR);
                            }
                        }
                        else {
                            fErrorReporter.reportError(XMLMessageFormatter.XML_DOMAIN,
                                                       "InvalidCharInContent", 
                                                       new Object[] { Integer.toString(ch2, 16) },
                                                       XMLErrorReporter.SEVERITY_FATAL_ERROR);
                        }
                    }
                    else {
                        fErrorReporter.reportError(XMLMessageFormatter.XML_DOMAIN,
                                                   "InvalidCharInContent", 
                                                   new Object[] { Integer.toString(ch, 16) },
                                                   XMLErrorReporter.SEVERITY_FATAL_ERROR);
                    }
                }
            }
            if (fHandler != null && readSize > 0) {
                fTempString.offset = 0;
                fTempString.length = readSize;
                fHandler.characters(
                    fTempString,
                    fHandler.modifyAugmentations(null, true));
            }
            readSize = fReader.read(fTempString.ch, 0, fTempString.ch.length - 1);
        }
        
    }
    
    /**
     * Sets the input source on this text reader.
     * 
     * @param source The XMLInputSource to use.
     */
    public void setInputSource(XMLInputSource source) {
        fSource = source;
    }
    
    /**
     * Closes the stream.  Call this after parse(), or when there is no longer any need
     * for this object.
     * 
     * @throws IOException
     */
    public void close() throws IOException {
        if (fReader != null) {
            fReader.close();
            fReader = null;
        }
    }
    
    /**
     * Returns true if the specified character is a valid XML character
     * as per the rules of XML 1.0.
     *
     * @param ch The character to check.
     */
    protected boolean isValid(int ch) {
        return XMLChar.isValid(ch);
    }
    
    /**
     * Sets the buffer size property for the reader which decides the chunk sizes that are parsed
     * by the reader at a time and passed to the handler
     * 
     * @param bufferSize The size of the buffer desired
     */
    protected void setBufferSize(int bufferSize) {
        if (fTempString.ch.length != ++bufferSize) {
            fTempString.ch = new char[bufferSize];
        }
    }
 
}