/**
* Scans surrogates and append them to the specified buffer.
*
* <p><strong>Note:</strong> This assumes the current char has already been identified as a high
* surrogate.
*
* @param buf The StringBuffer to append the read surrogates to.
* @return True if it succeeded.
*/
protected boolean scanSurrogates(XMLStringBuffer buf) throws IOException, XNIException {
int high = fEntityScanner.scanChar();
int low = fEntityScanner.peekChar();
if (!XMLChar.isLowSurrogate(low)) {
reportFatalError("InvalidCharInContent", new Object[] {Integer.toString(high, 16)});
return false;
}
fEntityScanner.scanChar();
// convert surrogates to supplemental character
int c = XMLChar.supplemental((char) high, (char) low);
// supplemental character must be a valid XML character
if (isInvalid(c)) {
reportFatalError("InvalidCharInContent", new Object[] {Integer.toString(c, 16)});
return false;
}
// fill in the buffer
buf.append((char) high);
buf.append((char) low);
return true;
} // scanSurrogates():boolean
/**
* Scans External ID and return the public and system IDs.
*
* @param identifiers An array of size 2 to return the system id, and public id (in that order).
* @param optionalSystemId Specifies whether the system id is optional.
* <p><strong>Note:</strong> This method uses fString and fStringBuffer, anything in them at
* the time of calling is lost.
*/
protected void scanExternalID(String[] identifiers, boolean optionalSystemId)
throws IOException, XNIException {
String systemId = null;
String publicId = null;
if (fEntityScanner.skipString("PUBLIC")) {
if (!fEntityScanner.skipSpaces()) {
reportFatalError("SpaceRequiredAfterPUBLIC", null);
}
scanPubidLiteral(fString);
publicId = fString.toString();
if (!fEntityScanner.skipSpaces() && !optionalSystemId) {
reportFatalError("SpaceRequiredBetweenPublicAndSystem", null);
}
}
if (publicId != null || fEntityScanner.skipString("SYSTEM")) {
if (publicId == null && !fEntityScanner.skipSpaces()) {
reportFatalError("SpaceRequiredAfterSYSTEM", null);
}
int quote = fEntityScanner.peekChar();
if (quote != '\'' && quote != '"') {
if (publicId != null && optionalSystemId) {
// looks like we don't have any system id
// simply return the public id
identifiers[0] = null;
identifiers[1] = publicId;
return;
}
reportFatalError("QuoteRequiredInSystemID", null);
}
fEntityScanner.scanChar();
XMLString ident = fString;
if (fEntityScanner.scanLiteral(quote, ident) != quote) {
fStringBuffer.clear();
do {
fStringBuffer.append(ident);
int c = fEntityScanner.peekChar();
if (XMLChar.isMarkup(c) || c == ']') {
fStringBuffer.append((char) fEntityScanner.scanChar());
}
} while (fEntityScanner.scanLiteral(quote, ident) != quote);
fStringBuffer.append(ident);
ident = fStringBuffer;
}
systemId = ident.toString();
if (!fEntityScanner.skipChar(quote)) {
reportFatalError("SystemIDUnterminated", null);
}
}
// store result in array
identifiers[0] = systemId;
identifiers[1] = publicId;
}
/**
* Scans a processing data. This is needed to handle the situation where a document starts with a
* processing instruction whose target name <em>starts with</em> "xml". (e.g. xmlfoo)
*
* <p><strong>Note:</strong> This method uses fStringBuffer, anything in it at the time of calling
* is lost.
*
* @param target The PI target
* @param data The string to fill in with the data
*/
protected void scanPIData(String target, XMLString data) throws IOException, XNIException {
// check target
if (target.length() == 3) {
char c0 = Character.toLowerCase(target.charAt(0));
char c1 = Character.toLowerCase(target.charAt(1));
char c2 = Character.toLowerCase(target.charAt(2));
if (c0 == 'x' && c1 == 'm' && c2 == 'l') {
reportFatalError("ReservedPITarget", null);
}
}
// spaces
if (!fEntityScanner.skipSpaces()) {
if (fEntityScanner.skipString("?>")) {
// we found the end, there is no data
data.clear();
return;
} else {
if (fNamespaces && fEntityScanner.peekChar() == ':') {
fEntityScanner.scanChar();
XMLStringBuffer colonName = new XMLStringBuffer(target);
colonName.append(":");
String str = fEntityScanner.scanName();
if (str != null) colonName.append(str);
reportFatalError("ColonNotLegalWithNS", new Object[] {colonName.toString()});
fEntityScanner.skipSpaces();
} else {
// if there is data there should be some space
reportFatalError("SpaceRequiredInPI", null);
}
}
}
fStringBuffer.clear();
// data
if (fEntityScanner.scanData("?>", fStringBuffer)) {
do {
int c = fEntityScanner.peekChar();
if (c != -1) {
if (XMLChar.isHighSurrogate(c)) {
scanSurrogates(fStringBuffer);
} else if (isInvalidLiteral(c)) {
reportFatalError("InvalidCharInPI", new Object[] {Integer.toHexString(c)});
fEntityScanner.scanChar();
}
}
} while (fEntityScanner.scanData("?>", fStringBuffer));
}
data.setValues(fStringBuffer);
} // scanPIData(String,XMLString)
/**
* Scans public ID literal.
*
* <p>[12] PubidLiteral ::= '"' PubidChar* '"' | "'" (PubidChar - "'")* "'" [13] PubidChar::= #x20
* | #xD | #xA | [a-zA-Z0-9] | [-'()+,./:=?;!*#@$_%]
*
* <p>The returned string is normalized according to the following rule, from
* http://www.w3.org/TR/REC-xml#dt-pubid:
*
* <p>Before a match is attempted, all strings of white space in the public identifier must be
* normalized to single space characters (#x20), and leading and trailing white space must be
* removed.
*
* @param literal The string to fill in with the public ID literal.
* @return True on success.
* <p><strong>Note:</strong> This method uses fStringBuffer, anything in it at the time of
* calling is lost.
*/
protected boolean scanPubidLiteral(XMLString literal) throws IOException, XNIException {
int quote = fEntityScanner.scanChar();
if (quote != '\'' && quote != '"') {
reportFatalError("QuoteRequiredInPublicID", null);
return false;
}
fStringBuffer.clear();
// skip leading whitespace
boolean skipSpace = true;
boolean dataok = true;
while (true) {
int c = fEntityScanner.scanChar();
if (c == ' ' || c == '\n' || c == '\r') {
if (!skipSpace) {
// take the first whitespace as a space and skip the others
fStringBuffer.append(' ');
skipSpace = true;
}
} else if (c == quote) {
if (skipSpace) {
// if we finished on a space let's trim it
fStringBuffer.length--;
}
literal.setValues(fStringBuffer);
break;
} else if (XMLChar.isPubid(c)) {
fStringBuffer.append((char) c);
skipSpace = false;
} else if (c == -1) {
reportFatalError("PublicIDUnterminated", null);
return false;
} else {
dataok = false;
reportFatalError("InvalidCharInPublicID", new Object[] {Integer.toHexString(c)});
}
}
return dataok;
}
/**
* Scans an attribute value and normalizes whitespace converting all whitespace characters to
* space characters.
*
* <p>[10] AttValue ::= '"' ([^<&"] | Reference)* '"' | "'" ([^<&'] | Reference)* "'"
*
* @param value The XMLString to fill in with the value.
* @param nonNormalizedValue The XMLString to fill in with the non-normalized value.
* @param atName The name of the attribute being parsed (for error msgs).
* @param checkEntities true if undeclared entities should be reported as VC violation, false if
* undeclared entities should be reported as WFC violation.
* @param eleName The name of element to which this attribute belongs.
* @return true if the non-normalized and normalized value are the same
* <p><strong>Note:</strong> This method uses fStringBuffer2, anything in it at the time of
* calling is lost.
*/
protected boolean scanAttributeValue(
XMLString value,
XMLString nonNormalizedValue,
String atName,
boolean checkEntities,
String eleName)
throws IOException, XNIException {
// quote
int quote = fEntityScanner.peekChar();
if (quote != '\'' && quote != '"') {
reportFatalError("OpenQuoteExpected", new Object[] {eleName, atName});
}
fEntityScanner.scanChar();
int entityDepth = fEntityDepth;
int c = fEntityScanner.scanLiteral(quote, value);
if (DEBUG_ATTR_NORMALIZATION) {
System.out.println("** scanLiteral -> \"" + value.toString() + "\"");
}
int fromIndex = 0;
if (c == quote && (fromIndex = isUnchangedByNormalization(value)) == -1) {
/** Both the non-normalized and normalized attribute values are equal. * */
nonNormalizedValue.setValues(value);
int cquote = fEntityScanner.scanChar();
if (cquote != quote) {
reportFatalError("CloseQuoteExpected", new Object[] {eleName, atName});
}
return true;
}
fStringBuffer2.clear();
fStringBuffer2.append(value);
normalizeWhitespace(value, fromIndex);
if (DEBUG_ATTR_NORMALIZATION) {
System.out.println("** normalizeWhitespace -> \"" + value.toString() + "\"");
}
if (c != quote) {
fScanningAttribute = true;
fStringBuffer.clear();
do {
fStringBuffer.append(value);
if (DEBUG_ATTR_NORMALIZATION) {
System.out.println("** value2: \"" + fStringBuffer.toString() + "\"");
}
if (c == '&') {
fEntityScanner.skipChar('&');
if (entityDepth == fEntityDepth) {
fStringBuffer2.append('&');
}
if (fEntityScanner.skipChar('#')) {
if (entityDepth == fEntityDepth) {
fStringBuffer2.append('#');
}
int ch = scanCharReferenceValue(fStringBuffer, fStringBuffer2);
if (ch != -1) {
if (DEBUG_ATTR_NORMALIZATION) {
System.out.println("** value3: \"" + fStringBuffer.toString() + "\"");
}
}
} else {
String entityName = fEntityScanner.scanName();
if (entityName == null) {
reportFatalError("NameRequiredInReference", null);
} else if (entityDepth == fEntityDepth) {
fStringBuffer2.append(entityName);
}
if (!fEntityScanner.skipChar(';')) {
reportFatalError("SemicolonRequiredInReference", new Object[] {entityName});
} else if (entityDepth == fEntityDepth) {
fStringBuffer2.append(';');
}
if (entityName == fAmpSymbol) {
fStringBuffer.append('&');
if (DEBUG_ATTR_NORMALIZATION) {
System.out.println("** value5: \"" + fStringBuffer.toString() + "\"");
}
} else if (entityName == fAposSymbol) {
fStringBuffer.append('\'');
if (DEBUG_ATTR_NORMALIZATION) {
System.out.println("** value7: \"" + fStringBuffer.toString() + "\"");
}
} else if (entityName == fLtSymbol) {
fStringBuffer.append('<');
if (DEBUG_ATTR_NORMALIZATION) {
System.out.println("** value9: \"" + fStringBuffer.toString() + "\"");
}
} else if (entityName == fGtSymbol) {
fStringBuffer.append('>');
if (DEBUG_ATTR_NORMALIZATION) {
System.out.println("** valueB: \"" + fStringBuffer.toString() + "\"");
}
} else if (entityName == fQuotSymbol) {
fStringBuffer.append('"');
if (DEBUG_ATTR_NORMALIZATION) {
System.out.println("** valueD: \"" + fStringBuffer.toString() + "\"");
}
} else {
if (fEntityManager.isExternalEntity(entityName)) {
reportFatalError("ReferenceToExternalEntity", new Object[] {entityName});
} else {
if (!fEntityManager.isDeclaredEntity(entityName)) {
// WFC & VC: Entity Declared
if (checkEntities) {
if (fValidation) {
fErrorReporter.reportError(
XMLMessageFormatter.XML_DOMAIN,
"EntityNotDeclared",
new Object[] {entityName},
XMLErrorReporter.SEVERITY_ERROR);
}
} else {
reportFatalError("EntityNotDeclared", new Object[] {entityName});
}
}
fEntityManager.startEntity(entityName, true);
}
}
}
} else if (c == '<') {
reportFatalError("LessthanInAttValue", new Object[] {eleName, atName});
fEntityScanner.scanChar();
if (entityDepth == fEntityDepth) {
fStringBuffer2.append((char) c);
}
} else if (c == '%' || c == ']') {
fEntityScanner.scanChar();
fStringBuffer.append((char) c);
if (entityDepth == fEntityDepth) {
fStringBuffer2.append((char) c);
}
if (DEBUG_ATTR_NORMALIZATION) {
System.out.println("** valueF: \"" + fStringBuffer.toString() + "\"");
}
} else if (c == '\n' || c == '\r') {
fEntityScanner.scanChar();
fStringBuffer.append(' ');
if (entityDepth == fEntityDepth) {
fStringBuffer2.append('\n');
}
} else if (c != -1 && XMLChar.isHighSurrogate(c)) {
fStringBuffer3.clear();
if (scanSurrogates(fStringBuffer3)) {
fStringBuffer.append(fStringBuffer3);
if (entityDepth == fEntityDepth) {
fStringBuffer2.append(fStringBuffer3);
}
if (DEBUG_ATTR_NORMALIZATION) {
System.out.println("** valueI: \"" + fStringBuffer.toString() + "\"");
}
}
} else if (c != -1 && isInvalidLiteral(c)) {
reportFatalError(
"InvalidCharInAttValue", new Object[] {eleName, atName, Integer.toString(c, 16)});
fEntityScanner.scanChar();
if (entityDepth == fEntityDepth) {
fStringBuffer2.append((char) c);
}
}
c = fEntityScanner.scanLiteral(quote, value);
if (entityDepth == fEntityDepth) {
fStringBuffer2.append(value);
}
normalizeWhitespace(value);
} while (c != quote || entityDepth != fEntityDepth);
fStringBuffer.append(value);
if (DEBUG_ATTR_NORMALIZATION) {
System.out.println("** valueN: \"" + fStringBuffer.toString() + "\"");
}
value.setValues(fStringBuffer);
fScanningAttribute = false;
}
nonNormalizedValue.setValues(fStringBuffer2);
// quote
int cquote = fEntityScanner.scanChar();
if (cquote != quote) {
reportFatalError("CloseQuoteExpected", new Object[] {eleName, atName});
}
return nonNormalizedValue.equals(value.ch, value.offset, value.length);
} // scanAttributeValue()
/**
* Scans a pseudo attribute.
*
* @param scanningTextDecl True if scanning this pseudo-attribute for a TextDecl; false if
* scanning XMLDecl. This flag is needed to report the correct type of error.
* @param value The string to fill in with the attribute value.
* @return The name of the attribute
* <p><strong>Note:</strong> This method uses fStringBuffer2, anything in it at the time of
* calling is lost.
*/
public String scanPseudoAttribute(boolean scanningTextDecl, XMLString value)
throws IOException, XNIException {
// REVISIT: This method is used for generic scanning of
// pseudo attributes, but since there are only three such
// attributes: version, encoding, and standalone there are
// for performant ways of scanning them. Every decl must
// have a version, and in TextDecls this version must
// be followed by an encoding declaration. Also the
// methods we invoke on the scanners allow non-ASCII
// characters to be parsed in the decls, but since
// we don't even know what the actual encoding of the
// document is until we scan the encoding declaration
// you cannot reliably read any characters outside
// of the ASCII range here. -- mrglavas
String name = fEntityScanner.scanName();
XMLEntityManager.print(fEntityManager.getCurrentEntity());
if (name == null) {
reportFatalError("PseudoAttrNameExpected", null);
}
fEntityScanner.skipDeclSpaces();
if (!fEntityScanner.skipChar('=')) {
reportFatalError(
scanningTextDecl ? "EqRequiredInTextDecl" : "EqRequiredInXMLDecl", new Object[] {name});
}
fEntityScanner.skipDeclSpaces();
int quote = fEntityScanner.peekChar();
if (quote != '\'' && quote != '"') {
reportFatalError(
scanningTextDecl ? "QuoteRequiredInTextDecl" : "QuoteRequiredInXMLDecl",
new Object[] {name});
}
fEntityScanner.scanChar();
int c = fEntityScanner.scanLiteral(quote, value);
if (c != quote) {
fStringBuffer2.clear();
do {
fStringBuffer2.append(value);
if (c != -1) {
if (c == '&' || c == '%' || c == '<' || c == ']') {
fStringBuffer2.append((char) fEntityScanner.scanChar());
}
// REVISIT: Even if you could reliably read non-ASCII chars
// why bother scanning for surrogates here? Only ASCII chars
// match the productions in XMLDecls and TextDecls. -- mrglavas
else if (XMLChar.isHighSurrogate(c)) {
scanSurrogates(fStringBuffer2);
} else if (isInvalidLiteral(c)) {
String key = scanningTextDecl ? "InvalidCharInTextDecl" : "InvalidCharInXMLDecl";
reportFatalError(key, new Object[] {Integer.toString(c, 16)});
fEntityScanner.scanChar();
}
}
c = fEntityScanner.scanLiteral(quote, value);
} while (c != quote);
fStringBuffer2.append(value);
value.setValues(fStringBuffer2);
}
if (!fEntityScanner.skipChar(quote)) {
reportFatalError(
scanningTextDecl ? "CloseQuoteMissingInTextDecl" : "CloseQuoteMissingInXMLDecl",
new Object[] {name});
}
// return
return name;
} // scanPseudoAttribute(XMLString):String
/**
* Scans a character reference and append the corresponding chars to the specified buffer.
*
* <p>
*
* <pre>
* [66] CharRef ::= '&#' [0-9]+ ';' | '&#x' [0-9a-fA-F]+ ';'
* </pre>
*
* <strong>Note:</strong> This method uses fStringBuffer, anything in it at the time of calling is
* lost.
*
* @param buf the character buffer to append chars to
* @param buf2 the character buffer to append non-normalized chars to
* @return the character value or (-1) on conversion failure
*/
protected int scanCharReferenceValue(XMLStringBuffer buf, XMLStringBuffer buf2)
throws IOException, XNIException {
// scan hexadecimal value
boolean hex = false;
if (fEntityScanner.skipChar('x')) {
if (buf2 != null) {
buf2.append('x');
}
hex = true;
fStringBuffer3.clear();
boolean digit = true;
int c = fEntityScanner.peekChar();
digit = (c >= '0' && c <= '9') || (c >= 'a' && c <= 'f') || (c >= 'A' && c <= 'F');
if (digit) {
if (buf2 != null) {
buf2.append((char) c);
}
fEntityScanner.scanChar();
fStringBuffer3.append((char) c);
do {
c = fEntityScanner.peekChar();
digit = (c >= '0' && c <= '9') || (c >= 'a' && c <= 'f') || (c >= 'A' && c <= 'F');
if (digit) {
if (buf2 != null) {
buf2.append((char) c);
}
fEntityScanner.scanChar();
fStringBuffer3.append((char) c);
}
} while (digit);
} else {
reportFatalError("HexdigitRequiredInCharRef", null);
}
}
// scan decimal value
else {
fStringBuffer3.clear();
boolean digit = true;
int c = fEntityScanner.peekChar();
digit = c >= '0' && c <= '9';
if (digit) {
if (buf2 != null) {
buf2.append((char) c);
}
fEntityScanner.scanChar();
fStringBuffer3.append((char) c);
do {
c = fEntityScanner.peekChar();
digit = c >= '0' && c <= '9';
if (digit) {
if (buf2 != null) {
buf2.append((char) c);
}
fEntityScanner.scanChar();
fStringBuffer3.append((char) c);
}
} while (digit);
} else {
reportFatalError("DigitRequiredInCharRef", null);
}
}
// end
if (!fEntityScanner.skipChar(';')) {
reportFatalError("SemicolonRequiredInCharRef", null);
}
if (buf2 != null) {
buf2.append(';');
}
// convert string to number
int value = -1;
try {
value = Integer.parseInt(fStringBuffer3.toString(), hex ? 16 : 10);
// character reference must be a valid XML character
if (isInvalid(value)) {
StringBuffer errorBuf = new StringBuffer(fStringBuffer3.length + 1);
if (hex) errorBuf.append('x');
errorBuf.append(fStringBuffer3.ch, fStringBuffer3.offset, fStringBuffer3.length);
reportFatalError("InvalidCharRef", new Object[] {errorBuf.toString()});
}
} catch (NumberFormatException e) {
// Conversion failed, let -1 value drop through.
// If we end up here, the character reference was invalid.
StringBuffer errorBuf = new StringBuffer(fStringBuffer3.length + 1);
if (hex) errorBuf.append('x');
errorBuf.append(fStringBuffer3.ch, fStringBuffer3.offset, fStringBuffer3.length);
reportFatalError("InvalidCharRef", new Object[] {errorBuf.toString()});
}
// append corresponding chars to the given buffer
if (!XMLChar.isSupplemental(value)) {
buf.append((char) value);
} else {
// character is supplemental, split it into surrogate chars
buf.append(XMLChar.highSurrogate(value));
buf.append(XMLChar.lowSurrogate(value));
}
// char refs notification code
if (fNotifyCharRefs && value != -1) {
String literal = "#" + (hex ? "x" : "") + fStringBuffer3.toString();
if (!fScanningAttribute) {
fCharRefLiteral = literal;
}
}
return value;
}