1907 * Returns the IANA encoding name that is auto-detected from
1908 * the bytes specified, with the endian-ness of that encoding where appropriate.
1909 *
1910 * @param b4 The first four bytes of the input.
1911 * @param count The number of bytes actually read.
1912 * @return a 2-element array: the first element, an IANA-encoding string,
1913 * the second element a Boolean which is true iff the document is big endian, false
1914 * if it's little-endian, and null if the distinction isn't relevant.
1915 */
1916 protected Object[] getEncodingName(byte[] b4, int count) {
1917
1918 if (count < 2) {
1919 return new Object[]{"UTF-8", null};
1920 }
1921
1922 // UTF-16, with BOM
1923 int b0 = b4[0] & 0xFF;
1924 int b1 = b4[1] & 0xFF;
1925 if (b0 == 0xFE && b1 == 0xFF) {
1926 // UTF-16, big-endian
1927 return new Object [] {"UTF-16BE", new Boolean(true)};
1928 }
1929 if (b0 == 0xFF && b1 == 0xFE) {
1930 // UTF-16, little-endian
1931 return new Object [] {"UTF-16LE", new Boolean(false)};
1932 }
1933
1934 // default to UTF-8 if we don't have enough bytes to make a
1935 // good determination of the encoding
1936 if (count < 3) {
1937 return new Object [] {"UTF-8", null};
1938 }
1939
1940 // UTF-8 with a BOM
1941 int b2 = b4[2] & 0xFF;
1942 if (b0 == 0xEF && b1 == 0xBB && b2 == 0xBF) {
1943 return new Object [] {"UTF-8", null};
1944 }
1945
1946 // default to UTF-8 if we don't have enough bytes to make a
1947 // good determination of the encoding
1948 if (count < 4) {
1949 return new Object [] {"UTF-8", null};
1950 }
1951
1952 // other encodings
1953 int b3 = b4[3] & 0xFF;
1954 if (b0 == 0x00 && b1 == 0x00 && b2 == 0x00 && b3 == 0x3C) {
1955 // UCS-4, big endian (1234)
1956 return new Object [] {"ISO-10646-UCS-4", new Boolean(true)};
1957 }
1958 if (b0 == 0x3C && b1 == 0x00 && b2 == 0x00 && b3 == 0x00) {
1959 // UCS-4, little endian (4321)
1960 return new Object [] {"ISO-10646-UCS-4", new Boolean(false)};
1961 }
1962 if (b0 == 0x00 && b1 == 0x00 && b2 == 0x3C && b3 == 0x00) {
1963 // UCS-4, unusual octet order (2143)
1964 // REVISIT: What should this be?
1965 return new Object [] {"ISO-10646-UCS-4", null};
1966 }
1967 if (b0 == 0x00 && b1 == 0x3C && b2 == 0x00 && b3 == 0x00) {
1968 // UCS-4, unusual octect order (3412)
1969 // REVISIT: What should this be?
1970 return new Object [] {"ISO-10646-UCS-4", null};
1971 }
1972 if (b0 == 0x00 && b1 == 0x3C && b2 == 0x00 && b3 == 0x3F) {
1973 // UTF-16, big-endian, no BOM
1974 // (or could turn out to be UCS-2...
1975 // REVISIT: What should this be?
1976 return new Object [] {"UTF-16BE", new Boolean(true)};
1977 }
1978 if (b0 == 0x3C && b1 == 0x00 && b2 == 0x3F && b3 == 0x00) {
1979 // UTF-16, little-endian, no BOM
1980 // (or could turn out to be UCS-2...
1981 return new Object [] {"UTF-16LE", new Boolean(false)};
1982 }
1983 if (b0 == 0x4C && b1 == 0x6F && b2 == 0xA7 && b3 == 0x94) {
1984 // EBCDIC
1985 // a la xerces1, return CP037 instead of EBCDIC here
1986 return new Object [] {"CP037", null};
1987 }
1988
1989 // default encoding
1990 return new Object [] {"UTF-8", null};
1991
1992 } // getEncodingName(byte[],int):Object[]
1993
1994 /**
1995 * xxx not removing endEntity() so that i remember that we need to implement it.
1996 * Ends an entity.
1997 *
1998 * @throws XNIException Thrown by entity handler to signal an error.
1999 */
2000 //
2001 /** Prints the contents of the buffer. */
|
1907 * Returns the IANA encoding name that is auto-detected from
1908 * the bytes specified, with the endian-ness of that encoding where appropriate.
1909 *
1910 * @param b4 The first four bytes of the input.
1911 * @param count The number of bytes actually read.
1912 * @return a 2-element array: the first element, an IANA-encoding string,
1913 * the second element a Boolean which is true iff the document is big endian, false
1914 * if it's little-endian, and null if the distinction isn't relevant.
1915 */
1916 protected Object[] getEncodingName(byte[] b4, int count) {
1917
1918 if (count < 2) {
1919 return new Object[]{"UTF-8", null};
1920 }
1921
1922 // UTF-16, with BOM
1923 int b0 = b4[0] & 0xFF;
1924 int b1 = b4[1] & 0xFF;
1925 if (b0 == 0xFE && b1 == 0xFF) {
1926 // UTF-16, big-endian
1927 return new Object [] {"UTF-16BE", Boolean.TRUE};
1928 }
1929 if (b0 == 0xFF && b1 == 0xFE) {
1930 // UTF-16, little-endian
1931 return new Object [] {"UTF-16LE", Boolean.FALSE};
1932 }
1933
1934 // default to UTF-8 if we don't have enough bytes to make a
1935 // good determination of the encoding
1936 if (count < 3) {
1937 return new Object [] {"UTF-8", null};
1938 }
1939
1940 // UTF-8 with a BOM
1941 int b2 = b4[2] & 0xFF;
1942 if (b0 == 0xEF && b1 == 0xBB && b2 == 0xBF) {
1943 return new Object [] {"UTF-8", null};
1944 }
1945
1946 // default to UTF-8 if we don't have enough bytes to make a
1947 // good determination of the encoding
1948 if (count < 4) {
1949 return new Object [] {"UTF-8", null};
1950 }
1951
1952 // other encodings
1953 int b3 = b4[3] & 0xFF;
1954 if (b0 == 0x00 && b1 == 0x00 && b2 == 0x00 && b3 == 0x3C) {
1955 // UCS-4, big endian (1234)
1956 return new Object [] {"ISO-10646-UCS-4", Boolean.TRUE};
1957 }
1958 if (b0 == 0x3C && b1 == 0x00 && b2 == 0x00 && b3 == 0x00) {
1959 // UCS-4, little endian (4321)
1960 return new Object [] {"ISO-10646-UCS-4", Boolean.FALSE};
1961 }
1962 if (b0 == 0x00 && b1 == 0x00 && b2 == 0x3C && b3 == 0x00) {
1963 // UCS-4, unusual octet order (2143)
1964 // REVISIT: What should this be?
1965 return new Object [] {"ISO-10646-UCS-4", null};
1966 }
1967 if (b0 == 0x00 && b1 == 0x3C && b2 == 0x00 && b3 == 0x00) {
1968 // UCS-4, unusual octect order (3412)
1969 // REVISIT: What should this be?
1970 return new Object [] {"ISO-10646-UCS-4", null};
1971 }
1972 if (b0 == 0x00 && b1 == 0x3C && b2 == 0x00 && b3 == 0x3F) {
1973 // UTF-16, big-endian, no BOM
1974 // (or could turn out to be UCS-2...
1975 // REVISIT: What should this be?
1976 return new Object [] {"UTF-16BE", Boolean.TRUE};
1977 }
1978 if (b0 == 0x3C && b1 == 0x00 && b2 == 0x3F && b3 == 0x00) {
1979 // UTF-16, little-endian, no BOM
1980 // (or could turn out to be UCS-2...
1981 return new Object [] {"UTF-16LE", Boolean.FALSE};
1982 }
1983 if (b0 == 0x4C && b1 == 0x6F && b2 == 0xA7 && b3 == 0x94) {
1984 // EBCDIC
1985 // a la xerces1, return CP037 instead of EBCDIC here
1986 return new Object [] {"CP037", null};
1987 }
1988
1989 // default encoding
1990 return new Object [] {"UTF-8", null};
1991
1992 } // getEncodingName(byte[],int):Object[]
1993
1994 /**
1995 * xxx not removing endEntity() so that i remember that we need to implement it.
1996 * Ends an entity.
1997 *
1998 * @throws XNIException Thrown by entity handler to signal an error.
1999 */
2000 //
2001 /** Prints the contents of the buffer. */
|