Java源码示例：com.sun.org.apache.xerces.internal.impl.io.UCSReader
示例1
/**
* 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 final 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("$$$ reusing old reader on stream");
}
}
}
示例2
/**
* Sets the encoding of the scanner. This method is used by the
* scanners if the XMLDecl or TextDecl line contains an encoding
* pseudo-attribute.
* <p>
* <strong>Note:</strong> The underlying character reader on the
* current entity will be changed to accomodate the new encoding.
* However, the new encoding is ignored if the current reader was
* not constructed from an input stream (e.g. an external entity
* that is resolved directly to the appropriate java.io.Reader
* object).
*
* @param encoding The IANA encoding name of the new encoding.
*
* @throws IOException Thrown if the new encoding is not supported.
*
* @see com.sun.org.apache.xerces.internal.util.EncodingMap
*/
public final void setEncoding(String encoding) throws IOException {
if (DEBUG_ENCODINGS) {
System.out.println("$$$ 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");
}
}
}
示例3
/**
* 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 final 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("$$$ reusing old reader on stream");
}
}
}
示例4
/**
* Sets the encoding of the scanner. This method is used by the
* scanners if the XMLDecl or TextDecl line contains an encoding
* pseudo-attribute.
* <p>
* <strong>Note:</strong> The underlying character reader on the
* current entity will be changed to accomodate the new encoding.
* However, the new encoding is ignored if the current reader was
* not constructed from an input stream (e.g. an external entity
* that is resolved directly to the appropriate java.io.Reader
* object).
*
* @param encoding The IANA encoding name of the new encoding.
*
* @throws IOException Thrown if the new encoding is not supported.
*
* @see com.sun.org.apache.xerces.internal.util.EncodingMap
*/
public final void setEncoding(String encoding) throws IOException {
if (DEBUG_ENCODINGS) {
System.out.println("$$$ 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");
}
}
}
示例5
/**
* 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 final 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("$$$ reusing old reader on stream");
}
}
}
示例6
/**
* Sets the encoding of the scanner. This method is used by the
* scanners if the XMLDecl or TextDecl line contains an encoding
* pseudo-attribute.
* <p>
* <strong>Note:</strong> The underlying character reader on the
* current entity will be changed to accomodate the new encoding.
* However, the new encoding is ignored if the current reader was
* not constructed from an input stream (e.g. an external entity
* that is resolved directly to the appropriate java.io.Reader
* object).
*
* @param encoding The IANA encoding name of the new encoding.
*
* @throws IOException Thrown if the new encoding is not supported.
*
* @see com.sun.org.apache.xerces.internal.util.EncodingMap
*/
public final void setEncoding(String encoding) throws IOException {
if (DEBUG_ENCODINGS) {
System.out.println("$$$ 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");
}
}
}
示例7
/**
* 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 final 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("$$$ reusing old reader on stream");
}
}
}
示例8
/**
* Sets the encoding of the scanner. This method is used by the
* scanners if the XMLDecl or TextDecl line contains an encoding
* pseudo-attribute.
* <p>
* <strong>Note:</strong> The underlying character reader on the
* current entity will be changed to accomodate the new encoding.
* However, the new encoding is ignored if the current reader was
* not constructed from an input stream (e.g. an external entity
* that is resolved directly to the appropriate java.io.Reader
* object).
*
* @param encoding The IANA encoding name of the new encoding.
*
* @throws IOException Thrown if the new encoding is not supported.
*
* @see com.sun.org.apache.xerces.internal.util.EncodingMap
*/
public final void setEncoding(String encoding) throws IOException {
if (DEBUG_ENCODINGS) {
System.out.println("$$$ 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");
}
}
}
示例9
/**
* 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 final 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("$$$ reusing old reader on stream");
}
}
}
示例10
/**
* Sets the encoding of the scanner. This method is used by the
* scanners if the XMLDecl or TextDecl line contains an encoding
* pseudo-attribute.
* <p>
* <strong>Note:</strong> The underlying character reader on the
* current entity will be changed to accomodate the new encoding.
* However, the new encoding is ignored if the current reader was
* not constructed from an input stream (e.g. an external entity
* that is resolved directly to the appropriate java.io.Reader
* object).
*
* @param encoding The IANA encoding name of the new encoding.
*
* @throws IOException Thrown if the new encoding is not supported.
*
* @see com.sun.org.apache.xerces.internal.util.EncodingMap
*/
public final void setEncoding(String encoding) throws IOException {
if (DEBUG_ENCODINGS) {
System.out.println("$$$ 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");
}
}
}
示例11
/**
* 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 final 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");
}
}
}
