mimeutility.java

java Email you can use it to send email to others

	return defaultJavaCharset;
    }

    /*
     * Get the default MIME charset for this locale.
     */
    static String getDefaultMIMECharset() {
	if (defaultMIMECharset == null) {
	    try {
		defaultMIMECharset = System.getProperty("mail.mime.charset");
	    } catch (SecurityException ex) { }	// ignore it
	}
	if (defaultMIMECharset == null)
	    defaultMIMECharset = mimeCharset(getDefaultJavaCharset());
	return defaultMIMECharset;
    }

    // Tables to map MIME charset names to Java names and vice versa.
    // XXX - Should eventually use J2SE 1.4 java.nio.charset.Charset
    private static Hashtable mime2java;
    private static Hashtable java2mime;

    static {
	java2mime = new Hashtable(40);
	mime2java = new Hashtable(10);

	try {
	    // Use this class's classloader to load the mapping file
	    // XXX - we should use SecuritySupport, but it's in another package
	    InputStream is = 
		    javax.mail.internet.MimeUtility.class.getResourceAsStream(
		    "/META-INF/javamail.charset.map");

	    if (is != null) {
		try {
		    is = new LineInputStream(is);

		    // Load the JDK-to-MIME charset mapping table
		    loadMappings((LineInputStream)is, java2mime);

		    // Load the MIME-to-JDK charset mapping table
		    loadMappings((LineInputStream)is, mime2java);
		} finally {
		    try {
			is.close();
		    } catch (Exception cex) {
			// ignore
		    }
		}
	    }
	} catch (Exception ex) { }

	// If we didn't load the tables, e.g., because we didn't have
	// permission, load them manually.  The entries here should be
	// the same as the default javamail.charset.map.
	if (java2mime.isEmpty()) {
	    java2mime.put("8859_1", "ISO-8859-1");
	    java2mime.put("iso8859_1", "ISO-8859-1");
	    java2mime.put("iso8859-1", "ISO-8859-1");

	    java2mime.put("8859_2", "ISO-8859-2");
	    java2mime.put("iso8859_2", "ISO-8859-2");
	    java2mime.put("iso8859-2", "ISO-8859-2");

	    java2mime.put("8859_3", "ISO-8859-3");
	    java2mime.put("iso8859_3", "ISO-8859-3");
	    java2mime.put("iso8859-3", "ISO-8859-3");

	    java2mime.put("8859_4", "ISO-8859-4");
	    java2mime.put("iso8859_4", "ISO-8859-4");
	    java2mime.put("iso8859-4", "ISO-8859-4");

	    java2mime.put("8859_5", "ISO-8859-5");
	    java2mime.put("iso8859_5", "ISO-8859-5");
	    java2mime.put("iso8859-5", "ISO-8859-5");

	    java2mime.put("8859_6", "ISO-8859-6");
	    java2mime.put("iso8859_6", "ISO-8859-6");
	    java2mime.put("iso8859-6", "ISO-8859-6");

	    java2mime.put("8859_7", "ISO-8859-7");
	    java2mime.put("iso8859_7", "ISO-8859-7");
	    java2mime.put("iso8859-7", "ISO-8859-7");

	    java2mime.put("8859_8", "ISO-8859-8");
	    java2mime.put("iso8859_8", "ISO-8859-8");
	    java2mime.put("iso8859-8", "ISO-8859-8");

	    java2mime.put("8859_9", "ISO-8859-9");
	    java2mime.put("iso8859_9", "ISO-8859-9");
	    java2mime.put("iso8859-9", "ISO-8859-9");

	    java2mime.put("sjis", "Shift_JIS");
	    java2mime.put("jis", "ISO-2022-JP");
	    java2mime.put("iso2022jp", "ISO-2022-JP");
	    java2mime.put("euc_jp", "euc-jp");
	    java2mime.put("koi8_r", "koi8-r");
	    java2mime.put("euc_cn", "euc-cn");
	    java2mime.put("euc_tw", "euc-tw");
	    java2mime.put("euc_kr", "euc-kr");
	}
	if (mime2java.isEmpty()) {
	    mime2java.put("iso-2022-cn", "ISO2022CN");
	    mime2java.put("iso-2022-kr", "ISO2022KR");
	    mime2java.put("utf-8", "UTF8");
	    mime2java.put("utf8", "UTF8");
	    mime2java.put("ja_jp.iso2022-7", "ISO2022JP");
	    mime2java.put("ja_jp.eucjp", "EUCJIS");
	    mime2java.put("euc-kr", "KSC5601");
	    mime2java.put("euckr", "KSC5601");
	    mime2java.put("us-ascii", "ISO-8859-1");
	    mime2java.put("x-us-ascii", "ISO-8859-1");
	}
    }

    private static void loadMappings(LineInputStream is, Hashtable table) {
	String currLine;

	while (true) {
	    try {
		currLine = is.readLine();
	    } catch (IOException ioex) {
		break; // error in reading, stop
	    }

	    if (currLine == null) // end of file, stop
		break;
	    if (currLine.startsWith("--") && currLine.endsWith("--"))
		// end of this table
		break;	

	    // ignore empty lines and comments
	    if (currLine.trim().length() == 0 || currLine.startsWith("#"))
		continue;
	    
	    // A valid entry is of the form <key><separator><value>
	    // where, <separator> := SPACE | HT. Parse this
	    StringTokenizer tk = new StringTokenizer(currLine, " \t");
	    try {
		String key = tk.nextToken();
		String value = tk.nextToken();
		table.put(key.toLowerCase(Locale.ENGLISH), value);
	    } catch (NoSuchElementException nex) { }
	}
    }

    static final int ALL_ASCII 		= 1;
    static final int MOSTLY_ASCII 	= 2;
    static final int MOSTLY_NONASCII 	= 3;

    /** 
     * Check if the given string contains non US-ASCII characters.
     * @param	s	string
     * @return		ALL_ASCII if all characters in the string 
     *			belong to the US-ASCII charset. MOSTLY_ASCII
     *			if more than half of the available characters
     *			are US-ASCII characters. Else MOSTLY_NONASCII.
     */
    static int checkAscii(String s) {
	int ascii = 0, non_ascii = 0;
	int l = s.length();

	for (int i = 0; i < l; i++) {
	    if (nonascii((int)s.charAt(i))) // non-ascii
		non_ascii++;
	    else
		ascii++;
	}

	if (non_ascii == 0)
	    return ALL_ASCII;
	if (ascii > non_ascii)
	    return MOSTLY_ASCII;

	return MOSTLY_NONASCII;
    }

    /** 
     * Check if the given byte array contains non US-ASCII characters.
     * @param	b	byte array
     * @return		ALL_ASCII if all characters in the string 
     *			belong to the US-ASCII charset. MOSTLY_ASCII
     *			if more than half of the available characters
     *			are US-ASCII characters. Else MOSTLY_NONASCII.
     *
     * XXX - this method is no longer used
     */
    static int checkAscii(byte[] b) {
	int ascii = 0, non_ascii = 0;

	for (int i=0; i < b.length; i++) {
	    // The '&' operator automatically causes b[i] to be promoted
	    // to an int, and we mask out the higher bytes in the int 
	    // so that the resulting value is not a negative integer.
	    if (nonascii(b[i] & 0xff)) // non-ascii
		non_ascii++;
	    else
		ascii++;
	}
	
	if (non_ascii == 0)
	    return ALL_ASCII;
	if (ascii > non_ascii)
	    return MOSTLY_ASCII;
	
	return MOSTLY_NONASCII;
    }

    /** 
     * Check if the given input stream contains non US-ASCII characters.
     * Upto <code>max</code> bytes are checked. If <code>max</code> is
     * set to <code>ALL</code>, then all the bytes available in this
     * input stream are checked. If <code>breakOnNonAscii</code> is true
     * the check terminates when the first non-US-ASCII character is
     * found and MOSTLY_NONASCII is returned. Else, the check continues
     * till <code>max</code> bytes or till the end of stream.
     *
     * @param	is	the input stream
     * @param	max	maximum bytes to check for. The special value
     *			ALL indicates that all the bytes in this input
     *			stream must be checked.
     * @param	breakOnNonAscii if <code>true</code>, then terminate the
     *			the check when the first non-US-ASCII character
     *			is found.
     * @return		ALL_ASCII if all characters in the string 
     *			belong to the US-ASCII charset. MOSTLY_ASCII
     *			if more than half of the available characters
     *			are US-ASCII characters. Else MOSTLY_NONASCII.
     */
    static int checkAscii(InputStream is, int max, boolean breakOnNonAscii) {
	int ascii = 0, non_ascii = 0;
	int len;
	int block = 4096;
	int linelen = 0;
	boolean longLine = false, badEOL = false;
	boolean checkEOL = encodeEolStrict && breakOnNonAscii;
	byte buf[] = null;
	if (max != 0) {
	    block = (max == ALL) ? 4096 : Math.min(max, 4096);
	    buf = new byte[block]; 
	}
	while (max != 0) {
	    try {
		if ((len = is.read(buf, 0, block)) == -1)
		    break;
		int lastb = 0;
		for (int i = 0; i < len; i++) {
	    	    // The '&' operator automatically causes b[i] to 
		    // be promoted to an int, and we mask out the higher
		    // bytes in the int so that the resulting value is 
		    // not a negative integer.
		    int b = buf[i] & 0xff;
		    if (checkEOL &&
			    ((lastb == '\r' && b != '\n') ||
			    (lastb != '\r' && b == '\n')))
			badEOL = true;
		    if (b == '\r' || b == '\n')
			linelen = 0;
		    else {
			linelen++;
			if (linelen > 998)	// 1000 - CRLF
			    longLine = true;
		    }
		    if (nonascii(b)) {	// non-ascii
		        if (breakOnNonAscii) // we are done
			    return MOSTLY_NONASCII;
		        else
			    non_ascii++;
		    } else
		        ascii++;
		    lastb = b;
		}
	    } catch (IOException ioex) {
		break;
	    }
	    if (max != ALL)
		max -= len;
	}

	if (max == 0 && breakOnNonAscii)
	    // We have been told to break on the first non-ascii character.
	    // We haven't got any non-ascii character yet, but then we
	    // have not checked all of the available bytes either. So we
	    // cannot say for sure that this input stream is ALL_ASCII,
	    // and hence we must play safe and return MOSTLY_NONASCII

	    return MOSTLY_NONASCII;

	if (non_ascii == 0) { // no non-us-ascii characters so far
	    // If we're looking at non-text data, and we saw CR without LF
	    // or vice versa, consider this mostly non-ASCII so that it
	    // will be base64 encoded (since the quoted-printable encoder
	    // doesn't encode this case properly).
	    if (badEOL)
		return MOSTLY_NONASCII;
	    // if we've seen a long line, we degrade to mostly ascii
	    else if (longLine)
		return MOSTLY_ASCII;
	    else
		return ALL_ASCII;
	}
	if (ascii > non_ascii) // mostly ascii
	    return MOSTLY_ASCII;
	return MOSTLY_NONASCII;
    }

    static final boolean nonascii(int b) {
	return b >= 0177 || (b < 040 && b != '\r' && b != '\n' && b != '\t');
    }
}

/**
 * An OutputStream that determines whether the data written to
 * it is all ASCII, mostly ASCII, or mostly non-ASCII.
 */
class AsciiOutputStream extends OutputStream {
    private boolean breakOnNonAscii;
    private int ascii = 0, non_ascii = 0;
    private int linelen = 0;
    private boolean longLine = false;
    private boolean badEOL = false;
    private boolean checkEOL = false;
    private int lastb = 0;
    private int ret = 0;

    public AsciiOutputStream(boolean breakOnNonAscii, boolean encodeEolStrict) {
	this.breakOnNonAscii = breakOnNonAscii;
	checkEOL = encodeEolStrict && breakOnNonAscii;
    }

    public void write(int b) throws IOException {
	check(b);
    }

    public void write(byte b[]) throws IOException {
	write(b, 0, b.length);
    }

    public void write(byte b[], int off, int len) throws IOException {
	len += off;
	for (int i = off; i < len ; i++)
	    check(b[i]);
    }

    private final void check(int b) throws IOException {
	b &= 0xff;
	if (checkEOL &&
		((lastb == '\r' && b != '\n') || (lastb != '\r' && b == '\n')))
	    badEOL = true;
	if (b == '\r' || b == '\n')
	    linelen = 0;
	else {
	    linelen++;
	    if (linelen > 998)	// 1000 - CRLF
		longLine = true;
	}
	if (MimeUtility.nonascii(b)) { // non-ascii
	    non_ascii++;
	    if (breakOnNonAscii) {	// we are done
		ret = MimeUtility.MOSTLY_NONASCII;
		throw new EOFException();
	    }
	} else
	    ascii++;
	lastb = b;
    }

    /**
     * Return ASCII-ness of data stream.
     */
    public int getAscii() {
	if (ret != 0)
	    return ret;
	// If we're looking at non-text data, and we saw CR without LF
	// or vice versa, consider this mostly non-ASCII so that it
	// will be base64 encoded (since the quoted-printable encoder
	// doesn't encode this case properly).
	if (badEOL)
	    return MimeUtility.MOSTLY_NONASCII;
	else if (non_ascii == 0) { // no non-us-ascii characters so far
	    // if we've seen a long line, we degrade to mostly ascii
	    if (longLine)
		return MimeUtility.MOSTLY_ASCII;
	    else
		return MimeUtility.ALL_ASCII;
	}
	if (ascii > non_ascii) // mostly ascii
	    return MimeUtility.MOSTLY_ASCII;
	return MimeUtility.MOSTLY_NONASCII;
    }
}