--- /dev/null
+package org.xwt.util;
+
+import java.io.Reader;
+import java.io.IOException;
+import java.io.EOFException;
+
+/**
+ * An Event-Driving, Non-Validating XML Parser with Namespace support.
+ *
+ * A subclass can implement the abstract functions for receiving details
+ * about an xml file as it is parsed. To initate a parse, use the parse()
+ * function.
+ *
+ * <h3>IMPLEMENTATION NOTES</h3>
+ * <p>As the parser traverses into an element, it adds it to the linked list
+ * called <tt>elements</tt>. However, <tt>elements</tt> has been pre-filled
+ * with instances of the Element inner class. So in the vast majority of
+ * cases, the pointer current is moved along one, and the values for the
+ * new element are filled into the current object.</p>
+ *
+ * <p>This parser supports all the unicode ranges required by the XML
+ * Specification. However, it is optimised for well-formed ASCII documents.
+ * Documents containing unicode Names and Attributes will take much longer
+ * to process, and invalid documents (badly formed Names or invalid attributes)
+ * will be run through a test on every single unicode character range before
+ * being declared invalid.</p>
+ *
+ * <h3>IMPLEMENTATION RULES</h3>
+ * <ul>
+ * <li>Each time the buffer offset <tt>off</tt> is moved, the length
+ * <tt>len</tt> must be decreased.</li>
+ * <li>Each time the buffer length is decreased, it must be checked to make
+ * sure it is >0.</li>
+ * </ul>
+ *
+ * <h3>Other Notes</h3>
+ * <ul>
+ * <li><i>error</i> is defined as a Validity Constraint Violation and is recoverable</li>
+ * <li><i>fatal error</i> is defined as a Well-formedness Constraint Violation and is not recoverable</li>
+ * </ul>
+ *
+ * @author David Crawshaw
+ * @see XML-Specification-1.0 http://w3.org/TR/REC-xml
+ */
+public abstract class XML
+{
+ /////////////////////////////////////////////////////////////////////////////////////////////
+ // XML Parser
+ /////////////////////////////////////////////////////////////////////////////////////////////
+
+ public static final int BUFFER_SIZE = 255;
+
+ /** static pool of XML.Element instances shared by all XML Parsers.
+ * elements in the queue have dirty prev and next references, that need cleaning before use. */
+ private static final Queue elements = new Queue(30);
+
+ private static final char[] single_amp = new char[] { '&' };
+ private static final char[] single_apos = new char[] { '\'' };
+ private static final char[] single_gt = new char[] { '>' };
+ private static final char[] single_lt = new char[] { '<' };
+ private static final char[] single_quot = new char[] { '"' };
+
+ private int line;
+ private int col;
+
+ private Reader in;
+ private char[] buf;
+ private int off;
+ private int len;
+
+ private Element current;
+
+ // used in readEntity() to process a single character without creating a new array
+ private char[] singlechar = new char[1];
+
+
+ public XML() { this(BUFFER_SIZE); }
+
+ public XML(int bSize) {
+ buf = new char[bSize];
+
+ current = (Element)elements.remove(false);
+ if (current == null) current = new Element();
+ current.prev = current.next = null;
+ }
+
+
+ /** Returns the line number at the beginning of the last process call. */
+ public int getLine() { return line; }
+
+ /** Returns the column number at the beginning of the last process call. */
+ public int getCol() { return col; }
+
+ /**
+ * Parse given input and call the abstract event functions.
+ *
+ * Careful with threading, as this function is not synchronized.
+ */
+ public final void parse(Reader reader) throws IOException, XMLException {
+ in = reader;
+ off = len = 0;
+ line = col = 1;
+
+ clean(); // clean up possible mid-way linked-list element
+
+ try {
+ // process the stream
+ while (true) {
+ if (!buffer(1)) {
+ if (current.qName == null) break;
+ throw new WFCException("reached eof without closing <"+current.qName+"> element", getLine(), getCol());
+ }
+
+ if (buf[off] == '<') readTag();
+ readChars(current.qName != null);
+ }
+ } finally { clean(); } // clean up elements
+ }
+
+ /** remove any leftover elements from the linked list and queue them */
+ private final void clean() {
+ while (current.prev != null) elements.append((current = current.prev).next);
+ current.next = null;
+ current.qName = null;
+ }
+
+ /** reads in a tag. expects <tt>buf[off] == '<'</tt> */
+ private final void readTag() throws IOException, XMLException {
+ // Start Tag '<' Name (S Attribute)* S? '>'
+ boolean starttag = true;
+
+ // End Tag '</' Name S? '>'
+ boolean endtag = false;
+
+ // if (starttag & endtag) then: EmptyElemTag '<' Name (S Attribute)* S? '/>'
+
+ // Position in the name of the ':' namespace prefix
+ int prefix = -1;
+
+ int namelen = 0;
+
+ col++; off++; len--;
+ if (!buffer(1)) throw new EOFException("Unexpected EOF processing element tag");
+
+ // work out what we can from the beginning of the tag
+ char s = buf[off];
+ if (s == '!') {
+ // definitions here don't necessarily conform to xml spec (as DTDs not yet implemented)
+ col++; off++; len--;
+ if (!buffer(4)) throw new EOFException("Unexpected EOF processing <! element");
+
+ boolean bad = false;
+ switch (buf[off]) {
+ case '-':
+ if (buf[off+1] != '-') { bad = true; break; }
+ col += 2; off += 2; len -= 2;
+
+ // Comment '<!--' ((Char - '-') | ('-' (Char - '-')))* '-->'
+ readChars(false, "-->", false);
+ col += 3; off += 3; len -= 3;
+ break;
+ case '[':
+ if (!buffer(7)
+ || buf[off+1] != 'C' || buf[off+2] != 'D' || buf[off+3] != 'A'
+ || buf[off+4] != 'T' || buf[off+5] != 'A' || buf[off+6] != '[') {
+ col++; off--; len++;
+ // Conditional '<![' (Char* - (Char* ']]>' Char*)) ']]>'
+ readChars(false, "]]>", false);
+ } else {
+ col += 7; off += 7; len -=7;
+ // CDATA '<![CDATA[' (Char* - (Char* ']]>' Char*)) ']]>'
+ readChars(true, "]]>", false);
+ }
+ col += 3; off += 3; len -= 3;
+ break;
+
+ // we don't care about the following definitions
+
+ case 'A':
+ if (!buffer(7)
+ || buf[off+1] != 'T' || buf[off+2] != 'T' || buf[off+3] != 'L'
+ || buf[off+4] != 'I' || buf[off+5] != 'S' || buf[off+6] != 'T') {
+ bad = true; break;
+ }
+ col += 7; off += 7; len -= 7;
+
+ // ATTLIST '<!ATTLIST' (Char* - '>') '>'
+ readChars(false, ">", true);
+ col++; off++; len--;
+ break;
+ case 'D':
+ if (!buffer(7)
+ || buf[off+1] != 'O' || buf[off+2] != 'C' || buf[off+3] != 'T'
+ || buf[off+4] != 'Y' || buf[off+5] != 'P' || buf[off+6] != 'E') {
+ bad = true; break;
+ }
+ col += 7; off += 7; len -= 7;
+
+ // DTD '<!DOCTYPE' (Char* - '>') '>'
+ readChars(false, ">", true);
+ col++; off++; len--;
+ break;
+ case 'E':
+ if (!buffer(7)) {
+ bad = true;
+ } else if (buf[off+1] == 'L' && buf[off+2] == 'E' && buf[off+3] == 'M'
+ && buf[off+4] == 'E' && buf[off+5] == 'N' && buf[off+6] == 'T') {
+ // ELEMENT '<!ELEMENT' (Char* - '>') '>'
+ readChars(false, ">", true);
+ col++; off++; len--;
+
+ } else if (buf[off+1] == 'N' && buf[off+2] == 'T' && buf[off+3] == 'I'
+ && buf[off+4] == 'T' && buf[off+5] == 'Y') {
+ // ENTITY '<!ENTITY' (Char* - '>') '>'
+ readChars(false, ">", true);
+ col++; off++; len--;
+
+ } else {
+ bad = true;
+ }
+ break;
+
+ case 'N':
+ if (!buffer(8)
+ || buf[off+1] != 'O' || buf[off+2] != 'T' || buf[off+3] != 'A' || buf[off+4] != 'T'
+ || buf[off+5] != 'I' || buf[off+6] != 'O' || buf[off+7] != 'N') {
+ bad = true; break;
+ }
+ col += 8; off += 8; len -= 8;
+ // NOTATION '<!NOTATION' (Char* - '>') '>'
+ readChars(false, ">", true);
+ col++; off++; len--;
+
+ break;
+ default: bad = true;
+ }
+
+ if (bad) throw new MarkupException("element tag start character is invalid", getLine(), getCol());
+
+ } else if (s == '?') {
+ // PI (Ignored) '<?' (Char* - (Char* '?>' Char*)) '?>'
+ col++; off++; len--;
+ readChars(false, "?>", true);
+ if (!buffer(2)) throw new EOFException("Unexpected EOF at end of Processing Instruction");
+ col += 2; off += 2; len -= 2;
+
+ } else {
+ if (s == '/') {
+ // End Tag '</' Name S? '>'
+ starttag = false;
+ endtag = true;
+
+ col++; off++; len--;
+ if (!buffer(1)) throw new EOFException("Unexpected EOF processing end tag");
+ s = buf[off];
+ }
+
+ if (!Name(s)) throw new MarkupException("invalid starting character in element name", getLine(), getCol());
+
+ // find the element name (defined in XML Spec: section 2.3)
+ for (namelen = 0; ; namelen++) {
+ if (!buffer(namelen+1)) throw new EOFException("Unexpected EOF in element tag name");
+
+ s = buf[off+namelen];
+
+ if (S(s) || s == '>') {
+ break;
+ } else if (s == '/') {
+ endtag = true;
+ break;
+ } else if (s == ':' && namelen > 0 && prefix < 1) {
+ // we have a definition of the prefix range available
+ prefix = namelen;
+ } else if (!NameChar(s)) {
+ throw new MarkupException("element name contains invalid character", getLine(), getCol());
+ }
+ }
+
+ // process name (based on calculated region)
+ if (namelen < 1) throw new MarkupException("element name is null", getLine(), getCol());
+
+ // we have marked out the name region, so turn it into a string and move on
+ String qName = new String(buf, off, namelen);
+
+ col += namelen; off += namelen; len -= namelen;
+
+ if (starttag) {
+ // create the in-memory element representation of this beast
+ // if current.qName == null then this is the root element we're dealing with
+ if (current.qName != null) {
+ if (current.next == null) {
+ // we're at the end of the default element depth
+ current.next = (Element)elements.remove(false);
+ if (current.next == null) current.next = new Element();
+ current.next.prev = current;
+ current.next.next = null;
+ }
+ current = current.next;
+ }
+
+ current.clear();
+ current.qName = qName;
+ current.defaultUri = current.uri = null;
+
+ if (prefix > 0) {
+ current.prefix = current.qName.substring(0, prefix);
+ current.localName = current.qName.substring(prefix+1);
+ } else {
+ current.prefix = null;
+ current.localName = current.qName;
+ }
+
+ // process attributes
+ readWhitespace();
+ if (!buffer(1)) throw new EOFException("Unexpected EOF - processing attributes part 1");
+ while (buf[off] != '/' && buf[off] != '>') {
+ readAttribute();
+ if (!buffer(1)) throw new EOFException("Unexpected EOF - processing attributes part 2");
+ readWhitespace();
+ }
+
+ // inherit namespace default uri if attribute was not provided
+ if (current.defaultUri == null) {
+ current.defaultUri = (current.prev != null) ? current.prev.defaultUri : null;
+ }
+
+ // work out the uri of this element
+ if (current.prefix == null) {
+ // element has no prefix, therefore is the default uri
+ current.uri = current.defaultUri;
+ } else {
+ // work back through the hashtables until uri is found
+ for (Element e = current; e != null && current.uri == null; e = e.prev) {
+ current.uri = (String)e.urimap.get(current.prefix);
+ }
+ if (current.uri == null) current.addError(new NCException("undefined prefix '"+current.prefix+"'", getLine(), getCol()));
+ }
+
+ } else {
+ // this is an end-of-element tag
+ if (!qName.equals(current.qName)) throw new WFCException(
+ "end tag </"+qName+"> does not line up with start tag <"+current.qName+">", getLine(), getCol()
+ );
+ }
+
+ // deal with whitespace
+ readWhitespace();
+
+ // process tag close
+ if (!buffer(1)) throw new EOFException("Unexpected EOF before end of tag");
+ if (buf[off] == '/') {
+ endtag = true;
+ off++; len--; col++;
+ }
+ if (!buffer(1)) throw new EOFException("Unexpected EOF before end of endtag");
+ if (buf[off] == '>') {
+ off++; len--; col++;
+ } else {
+ throw new MarkupException("missing '>' character from element '"+qName+"'", getLine(), getCol());
+ }
+
+ // send element signals
+ if (starttag) startElement(current);
+ if (endtag) {
+ endElement(current);
+
+ // we just closed an element, so remove it from the element 'stack'
+ if (current.prev == null) {
+ // we just finished the root element
+ current.qName = null;
+ } else {
+ elements.append((current = current.prev).next);
+ current.next = null;
+ }
+ }
+ }
+ }
+
+ /** reads in an attribute of an element. expects Name(buf[off]) */
+ private final void readAttribute() throws IOException, XMLException {
+ int ref = 0;
+ int prefix = 0;
+ boolean xmlns = false;
+ String n, v;
+ n = v = null;
+ char s;
+
+ // find the element name (defined in XML Spec: section 2.3)
+ for (ref= 0; ; ref++) {
+ if (!buffer(ref+1)) throw new EOFException("Unexpected EOF in read attribute loop part 1");
+
+ s = buf[off+ref];
+
+ if (s == '=' || S(s)) {
+ break;
+ } else if (s == ':' && ref > 0 && prefix < 1) {
+ // we have a definition of the prefix range available
+ prefix = ref+1;
+ } else if (!NameChar(s)) {
+ throw new MarkupException("attribute name contains invalid characters", getLine(), getCol());
+ }
+ }
+
+ // work out if attribute is related to XML Namespacing
+ if (ref > 4 && buf[off] == 'x' && buf[off+1] == 'm' && buf[off+2] == 'l' && buf[off+3] == 'n' && buf[off+4] == 's') {
+ if (ref == 5) {
+ xmlns = true;
+ } else if (prefix > 0) {
+ xmlns = true;
+ n = new String(buf, off+prefix, ref-prefix);
+ } else {
+ n = new String(buf, off, ref);
+ }
+ } else {
+ n = new String(buf, off, ref);
+ }
+ col += ref; off += ref; len -= ref;
+
+ readWhitespace();
+ if (!buffer(1)) throw new EOFException("Unexpected EOF before attribute '=' divider");
+ if (buf[off] != '=') throw new MarkupException("attribute name not followed by '=' sign", getLine(), getCol());
+
+ col++; off++; len--;
+ readWhitespace();
+
+ if (!buffer(1)) throw new EOFException("Unexpected EOF after attribute '=' divider");
+
+ char wrap;
+ if (buf[off] == '\'' || buf[off] == '"') {
+ wrap = buf[off];
+ } else {
+ throw new MarkupException("attribute '"+n+"' must have attribute wrapped in ' or \"", getLine(), getCol());
+ }
+ col++; off++; len--;
+
+ // find the attribute value
+ attval: for (ref = 0; ; ref++) {
+ if (!buffer(ref+1)) throw new EOFException("Unexpected EOF in attribute value");
+
+ if (buf[off+ref] == wrap) {
+ break attval;
+ } else if (buf[off+ref] == '<') {
+ throw new WFCException("attribute value for '"+n+"' must not contain '<'", getLine(), getCol());
+ }
+ }
+
+ v = new String(buf, off, ref);
+ col += ref; off += ref; len -= ref;
+
+ // remove end wrapper character
+ col++; off++; len--;
+
+ if (xmlns) {
+ if (n == null) {
+ if (current.defaultUri != null) {
+ current.addError(new NCException("default namespace definition repeated", getLine(), getCol()));
+ } else {
+ current.defaultUri = v;
+ }
+ } else {
+ current.urimap.put(n, v);
+ }
+ } else {
+ // check to see if attribute is a repeat
+ for (int i=0; current.len > i; i++) if (n.equals(current.keys[i])) throw new WFCException(
+ "attribute name '"+n+"' may not appear more than once in the same element tag", getLine(), getCol()
+ );
+
+ // add attribute to the attribute arrays
+ if (current.len == current.keys.length) current.morekeys();
+ current.keys[current.len] = n;
+ current.vals[current.len] = v;
+ current.len++;
+ }
+ }
+
+ /** reads an entity and processes out its value. expects buf[off] == '&' */
+ private final void readEntity() throws IOException, XMLException {
+ off++; len--;
+ if (!buffer(2)) throw new EOFException("Unexpected EOF reading entity");
+
+ boolean unknown = false;
+ switch (buf[off]) {
+ case '#':
+ off++; len--;
+
+ int radix;
+ if (buf[off] == 'x') { off++; len--; radix = 16; } else { radix = 10; }
+ int c = 0;
+
+ // read in each char, then shift total value to the left and add the extra
+ // style of loop is slightly different from all the others, as this should run a limited number of times
+ findchar: while (true) {
+ if (!buffer(1)) throw new EOFException("Unexpected EOF reading entity");
+ int d = Character.digit(buf[off], radix);
+ if (d == -1) {
+ if (buf[off] != ';') throw new WFCException("illegal characters in entity reference", getLine(), getCol());
+ off++; len--; col++;
+ break findchar;
+ }
+ c = (c * radix) + d;
+
+ off++; len--;
+ }
+
+ singlechar[0] = Character.forDigit(c, radix);
+ characters(singlechar, 0, 1);
+ break;
+
+ case 'a':
+ if (buffer(4) && buf[off+1] == 'm' && buf[off+2] == 'p' && buf[off+3] == ';') {
+ characters(single_amp, 0, 1); // &
+ off += 4; len -= 4; col++;
+ } else if (buffer(5) && buf[off+1] == 'p' && buf[off+2] == 'o' && buf[off+3] == 's' && buf[off+4] == ';') {
+ characters(single_apos, 0, 1); // '
+ off += 5; len -= 5; col++;
+ } else {
+ unknown = true;
+ }
+ break;
+
+ case 'g':
+ if (buffer(3) && buf[off+1] == 't' && buf[off+2] == ';') {
+ characters(single_gt, 0, 1); // >
+ off += 3; len -= 3; col++;
+ } else {
+ unknown = true;
+ }
+ break;
+
+ case 'l':
+ if (buffer(3) && buf[off+1] == 't' && buf[off+2] == ';') {
+ characters(single_lt, 0, 1); // <
+ off += 3; len -= 3; col++;
+ } else {
+ unknown = true;
+ }
+ break;
+
+ case 'q':
+ if (buffer(5) && buf[off+1] == 'u' && buf[off+2] == 'o' && buf[off+3] == 't' && buf[off+4] == ';') {
+ characters(single_quot, 0, 1); // "
+ off += 5; len -= 5; col++;
+ } else {
+ unknown = true;
+ }
+ break;
+
+ // TODO: check a parser-level Hash of defined entities
+ }
+
+ if (unknown) throw new WFCException("unknown entity (<!ENTITY> not supported)", getLine(), getCol());
+ }
+
+ /** reads until the passed string is encountered. */
+ private final void readChars(boolean p, String match, boolean entities) throws IOException, XMLException {
+ int ref;
+ char[] end = match.toCharArray();
+
+ for (boolean more = true; more;) {
+ if (!buffer(1)) return;
+
+ buf: for (ref = 0; ref < len; ref++) {
+ switch (buf[off+ref]) {
+ case '\r': // windows or macos9 newline
+ // normalise and process
+ buf[off+ref] = '\n'; ref++;
+ if (p) characters(buf, off, ref);
+ off += ref; len -= ref; ref = -1;
+ line++; col = 1;
+
+ // windows double-char newline; skip the next char
+ if (!buffer(1)) return;
+ if (buf[off] == '\n') { off++; len--; }
+ break;
+
+ case '\n': // unix newline
+ ref++;
+ if (p) characters(buf, off, ref);
+ off += ref; len -= ref; ref = -1;
+ line++; col = 1;
+ break;
+
+ case '&': // entity
+ if (entities) {
+ if (p) {
+ if (ref > 0) characters(buf, off, ref);
+ off += ref; len -= ref; ref = -1;
+ readEntity();
+ }
+ break;
+ }
+
+ default:
+ if (!buffer(ref+end.length)) continue buf;
+ for (int i=0; end.length > i; i++) if (end[i] != buf[off+ref+i]) continue buf;
+ more = false;
+ break buf;
+ }
+ }
+
+ if (p && ref > 0) characters(buf, off, ref);
+ off += ref; len -= ref; col += ref;
+ }
+ }
+
+ /**
+ * reads until a <tt><</tt> symbol is encountered
+ * @param p If true call the characters(char[],int,int) funciton for the processed characters
+ */
+ private final void readChars(boolean p) throws IOException, XMLException {
+ int ref;
+
+ for (boolean more = true; more;) {
+ if (!buffer(1)) return;
+
+ buf: for (ref = 0; ref < len; ref++) {
+ switch (buf[off+ref]) {
+ case '\r': // windows or macos9 newline
+ // normalise and process
+ buf[off+ref] = '\n'; ref++;
+ if (p) characters(buf, off, ref);
+ off += ref; len -= ref; ref = -1;
+ line++; col = 1;
+
+ // windows double-char newline; skip the next char
+ if (!buffer(1)) return;
+ if (buf[off] == '\n') { off++; len--; }
+ break;
+
+ case '\n': // unix newline
+ ref++;
+ if (p) characters(buf, off, ref);
+ off += ref; len -= ref; ref = -1;
+ line++; col = 1;
+ break;
+
+ case '&': // entity
+ if (p) {
+ if (ref > 0) characters(buf, off, ref);
+ off += ref; len -= ref; ref = -1;
+ readEntity();
+ }
+ break;
+
+ case '<': // end of chars section
+ more = false;
+ break buf;
+ }
+ }
+
+ if (p && ref > 0) characters(buf, off, ref);
+ off += ref; len -= ref; col += ref;
+ }
+ }
+
+ /** reads until a non-whitespace symbol is encountered */
+ private final void readWhitespace() throws IOException, XMLException {
+ int ref;
+
+ for (boolean more = true; more;) {
+ if (!buffer(1)) return;
+
+ buf: for (ref = 0; ref < len; ref++) {
+ switch (buf[off+ref]) {
+ case '\r': // windows or macos9 newline
+ // normalise and process
+ buf[off+ref] = '\n';
+ whitespace(buf, off, ++ref);
+ off += ref; len -= ref; ref = -1;
+ line++; col = 1;
+
+ // windows double-char newline; skip the next char
+ if (!buffer(1)) return;
+ if (buf[off] == '\n') { off++; len--; }
+ break;
+
+ case '\n': // unix newline
+ whitespace(buf, off, ++ref);
+ off += ref; len -= ref; ref = -1;
+ line++; col = 1;
+ break;
+
+ case ' ': // space
+ case '\t': // tab
+ break;
+
+ default: // end of whitespace
+ more = false;
+ break buf;
+ }
+ }
+
+ off += ref; len -= ref; col += ref;
+ }
+ }
+
+ /**
+ * attempt to fill the buffer.
+ *
+ * @param min Minimum number of characters to read (even if we have to block to do it).
+ * @return return false if min can't be reached.
+ */
+ private final boolean buffer(int min) throws IOException {
+ if (len > min) return true;
+
+ if (buf.length - (off+len) >= min) {
+ // plenty of space left on the end of the buffer
+ } else if (off >= min) {
+ // moving offset data to start will leave enough free space on the end
+ System.arraycopy(buf, off, buf, 0, len);
+ off = 0;
+ } else {
+ // buffer size will have to be increased
+ char[] newbuf = new char[buf.length * 2];
+ System.arraycopy(buf, off, newbuf, 0, len);
+ buf = newbuf;
+ off = 0;
+ }
+
+ while (min > len) {
+ int newlen = in.read(buf, off+len, buf.length-(off+len));
+ if (newlen < 0) return false;
+ len += newlen;
+ }
+
+ return true;
+ }
+
+
+ /////////////////////////////////////////////////////////////////////////////////////////////
+ // Abstract SAX-Like Interface
+ /////////////////////////////////////////////////////////////////////////////////////////////
+
+ /**
+ * Called when the start of an element is processed.
+ *
+ * <p>The array of Attribute names and values may be longer than the
+ * number of entries they contain, but all the entries will be
+ * packed at the top.</p>
+ *
+ * <p><b>DO NOT</b> store a reference to the attribute arrays, as
+ * they are reused by other elements.</p>
+ */
+ public abstract void startElement(Element e) throws SchemaException;
+
+ /**
+ * Represents a line of character data.
+ *
+ * <p>Newlines are all normalised to the Unix \n as per the XML Spec,
+ * and a newline will only appear as the last character in the passed
+ * array segment.</p>
+ *
+ * <p>XML.getLine() and XML.getCol() report the position at the
+ * beginning of this character segment, which can be processed in a
+ * line-by-line fashion due to the above newline restriction.</p>
+ */
+ public abstract void characters(char[] ch, int start, int length) throws SchemaException;
+
+ /** Represents a line of ignorable whitespace. */
+ public abstract void whitespace(char[] ch, int start, int length) throws SchemaException;
+
+ /** Represents the end of an Element. */
+ public abstract void endElement(Element e) throws SchemaException;
+
+
+ /////////////////////////////////////////////////////////////////////////////////////////////
+ // Inner Classes for Parser Support
+ /////////////////////////////////////////////////////////////////////////////////////////////
+
+ /**
+ * Used as a struct for holding information about a current element,
+ * and acts as a linked list entry.
+ *
+ * <p>Each element stores a hashtable of namespace definitions against
+ * their respective prefix, and a variable holding their default
+ * uri. If they did not specify a default uri, their
+ * parent's uri is copied in to keep up the sembelence of speedy
+ * parsing.</p>
+ *
+ * <h3>SLOWEST PART OF THE XML PARSER</h3>
+ * <p>To implement the Namespace Specification exactly, we have to
+ * store prefix mappings for elements away from its parents and
+ * siblings. This means if a child of a child of-a child uses
+ * a prefix defined in the root, we have to search each Hashtable
+ * in each Element until we get to the root.</p>
+ *
+ * <p>Unfortunetally, every other solution I can think of requires
+ * more work than this one, shifted to different parts of the
+ * parser.</p>
+ */
+ public static final class Element
+ {
+ public Element next, prev;
+
+ /** A hashtable of all namespace prefixes that are defined by this element. */
+ public Hash urimap;
+
+ /** An array of attribute names. */
+ public String[] keys;
+
+ /** An array of attribute values. */
+ public String[] vals;
+
+ /** An array of non-fatal errors related to this element. */
+ public XMLException[] errors;
+
+ /** Current number of attributes in the <tt>keys</tt> and <tt>vals</tt> arrays. */
+ public int len;
+
+ /** Default URI for this element and its children with no prefix. */
+ public String defaultUri;
+
+ /** URI of current tag. XML Namespace Spec 14-Jan-1999 section 1 */
+ public String uri;
+
+ /** LocalPart of current element. XML Namespace Spec 14-Jan-1999 [8] */
+ public String localName;
+
+ /** Qualified Name of current element. XML Namespace Spec 14-Jan-1999 [6] */
+ public String qName;
+
+ /** Prefix of current element. Substring of qName. XML Namespace Spec 14-Jan-1999 [7] */
+ public String prefix;
+
+ public Element() {
+ defaultUri = uri = prefix = localName = qName = null;
+ urimap = new Hash(3,3);
+ keys = new String[10];
+ vals = new String[10];
+ errors = new XMLException[] {};
+ len = 0;
+ }
+
+ /** increase the size of the attributes arrays */
+ void morekeys() {
+ String[] newkeys = new String[keys.length+5];
+ String[] newvals = new String[vals.length+5];
+ System.arraycopy(keys, 0, newkeys, 0, keys.length);
+ System.arraycopy(vals, 0, newvals, 0, vals.length);
+ keys = newkeys; vals = newvals;
+ }
+
+ /** empty out the arrays */
+ void clear() {
+ for (int i=0; len > i; i++) { keys[i] = null; vals[i] = null; }; len = 0;
+ errors = new XMLException[] {};
+ }
+
+ /** add an error to the errors array */
+ void addError(XMLException e) {
+ // it doesn't really matter about continually expanding the array, as this case is quite rare
+ XMLException[] newe = new XMLException[errors.length+1];
+ System.arraycopy(errors, 0, newe, 0, errors.length);
+ newe[errors.length] = e;
+ errors = newe;
+ }
+ }
+
+ /** Parse or Structural Error */
+ public static class XMLException extends Exception
+ {
+ private int line;
+ private int col;
+ private String error;
+
+ public XMLException(String e) { this(e, -1, -1); }
+
+ public XMLException(String e, int l, int c) {
+ this.error = e;
+ this.line = l;
+ this.col = c;
+ }
+
+ public int getLine() { return this.line; }
+ public int getCol() { return this.col; }
+ public String getMessage() { return this.error; }
+ }
+
+ /** Violation of Markup restrictions in XML Specification - Fatal Error */
+ public static class MarkupException extends XMLException { public MarkupException(String e, int l, int c) { super(e,l,c); } }
+
+ /** Well-Formedness Constraint Violation - Fatal Error */
+ public static final class WFCException extends MarkupException { public WFCException(String e, int l, int c) { super(e,l,c); } }
+
+ /** Namespace Constraint Violation - Recoverable Error */
+ public static final class NCException extends XMLException { public NCException(String e, int l, int c) { super(e,l,c); } }
+
+ /** Schema Violation - Fatal Error */
+ public static class SchemaException extends XMLException {
+ public SchemaException(String e) { this(e, -1, -1); }
+ public SchemaException(String e, int l, int c) { super(e,l,c); }
+ }
+
+
+ /////////////////////////////////////////////////////////////////////////////////////////////
+ // Static Support Functions for the XML Specification
+ /////////////////////////////////////////////////////////////////////////////////////////////
+
+ // attempt to avoid these functions unless you *expect* the input to fall in the given range.
+
+ /** First Character of Name - XML Specification 1.0 [5] */
+ private static final boolean Name(char c) {
+ return BaseCharAscii(c) || c == '_' || c == ':' || Letter(c);
+ }
+
+ /** NameChar - XML Specification 1.0 [4] */
+ private static final boolean NameChar(char c) {
+ return BaseCharAscii(c) || c == '.' || c == '-' || c == '_' || c == ':'
+ || Digit(c) || Letter(c) || Extender(c); // TODO: || CombiningChar(c);
+ }
+
+ /** BaseChar - XMl Specification 1.0 [84] */
+ private static final boolean Letter(char c) {
+ return BaseChar(c) || Ideographic(c);
+ }
+
+ /** Elements of BaseChar that exist in ASCII. */
+ private static final boolean BaseCharAscii(char c) {
+ return (c >= '\u0041' && c <= '\u005A') || (c >= '\u0061' && c <= '\u007A');
+ }
+
+ /** Char - XML Specification 1.0 [2] */
+ private static final boolean Char(char c) {
+ // u000A == r and u000D == n, but the javac compiler can't handle the \ u form
+ return c == '\u0009' || c == '\r' || c == '\n'
+ || (c >= '\u0020' && c <= '\uD7FF')
+ || (c >= '\uE000' && c <= '\uFFFD');
+ }
+
+ /** BaseChar - XML Specification 1.0 [85] */
+ private static final boolean BaseChar(char c) {
+ return BaseCharAscii(c) || (c >= '\u00C0' && c <= '\u00D6')
+ || (c >= '\u00D8' && c <= '\u00F6') || (c >= '\u00F8' && c <= '\u00FF') || (c >= '\u0100' && c <= '\u0131')
+ || (c >= '\u0134' && c <= '\u013E') || (c >= '\u0141' && c <= '\u0148') || (c >= '\u014A' && c <= '\u017E')
+ || (c >= '\u0180' && c <= '\u01C3') || (c >= '\u01CD' && c <= '\u01F0') || (c >= '\u01F4' && c <= '\u01F5')
+ || (c >= '\u01FA' && c <= '\u0217') || (c >= '\u0250' && c <= '\u02A8') || (c >= '\u02BB' && c <= '\u02C1')
+ || (c == '\u0386') || (c >= '\u0388' && c <= '\u038A') || (c == '\u038C')
+ || (c >= '\u038E' && c <= '\u03A1') || (c >= '\u03A3' && c <= '\u03CE') || (c >= '\u03D0' && c <= '\u03D6')
+ || (c == '\u03DA') || (c == '\u03DC') || (c == '\u03DE')
+ || (c == '\u03E0')
+ || (c >= '\u03E2' && c <= '\u03F3') || (c >= '\u0401' && c <= '\u040C') || (c >= '\u040E' && c <= '\u044F')
+ || (c >= '\u0451' && c <= '\u045C') || (c >= '\u045E' && c <= '\u0481') || (c >= '\u0490' && c <= '\u04C4')
+ || (c >= '\u04C7' && c <= '\u04C8') || (c >= '\u04CB' && c <= '\u04CC') || (c >= '\u04D0' && c <= '\u04EB')
+ || (c >= '\u04EE' && c <= '\u04F5') || (c >= '\u04F8' && c <= '\u04F9') || (c >= '\u0531' && c <= '\u0556')
+ || (c == '\u0559')
+ || (c >= '\u0561' && c <= '\u0586') || (c >= '\u05D0' && c <= '\u05EA') || (c >= '\u05F0' && c <= '\u05F2')
+ || (c >= '\u0621' && c <= '\u063A') || (c >= '\u0641' && c <= '\u064A') || (c >= '\u0671' && c <= '\u06B7')
+ || (c >= '\u06BA' && c <= '\u06BE') || (c >= '\u06C0' && c <= '\u06CE') || (c >= '\u06D0' && c <= '\u06D3')
+ || (c == '\u06D5')
+ || (c >= '\u06E5' && c <= '\u06E6') || (c >= '\u0905' && c <= '\u0939')
+ || (c == '\u093D')
+ || (c >= '\u0958' && c <= '\u0961') || (c >= '\u0985' && c <= '\u098C') || (c >= '\u098F' && c <= '\u0990')
+ || (c >= '\u0993' && c <= '\u09A8') || (c >= '\u09AA' && c <= '\u09B0')
+ || (c == '\u09B2')
+ || (c >= '\u09B6' && c <= '\u09B9') || (c >= '\u09DF' && c <= '\u09E1') || (c >= '\u09F0' && c <= '\u09F1')
+ || (c >= '\u0A05' && c <= '\u0A0A') || (c >= '\u0A0F' && c <= '\u0A10') || (c >= '\u0A13' && c <= '\u0A28')
+ || (c >= '\u0A2A' && c <= '\u0A30') || (c >= '\u0A32' && c <= '\u0A33') || (c >= '\u0A35' && c <= '\u0A36')
+ || (c >= '\u0A38' && c <= '\u0A39') || (c >= '\u0A59' && c <= '\u0A5C')
+ || (c == '\u0A5E')
+ || (c >= '\u0A72' && c <= '\u0A74') || (c >= '\u0A85' && c <= '\u0A8B')
+ || (c == '\u0A8D')
+ || (c >= '\u0A8F' && c <= '\u0A91') || (c >= '\u0A93' && c <= '\u0AA8') || (c >= '\u0AAA' && c <= '\u0AB0')
+ || (c >= '\u0AB2' && c <= '\u0AB3') || (c >= '\u0AB5' && c <= '\u0AB9')
+ || (c == '\u0ABD')
+ || (c == '\u0AE0')
+ || (c >= '\u0B05' && c <= '\u0B0C') || (c >= '\u0B0F' && c <= '\u0B10') || (c >= '\u0B13' && c <= '\u0B28')
+ || (c >= '\u0B2A' && c <= '\u0B30') || (c >= '\u0B32' && c <= '\u0B33') || (c >= '\u0B36' && c <= '\u0B39')
+ || (c == '\u0B3D')
+ || (c >= '\u0B5C' && c <= '\u0B5D') || (c >= '\u0B5F' && c <= '\u0B61') || (c >= '\u0B85' && c <= '\u0B8A')
+ || (c >= '\u0B8E' && c <= '\u0B90') || (c >= '\u0B92' && c <= '\u0B95') || (c >= '\u0B99' && c <= '\u0B9A')
+ || (c == '\u0B9C')
+ || (c >= '\u0B9E' && c <= '\u0B9F') || (c >= '\u0BA3' && c <= '\u0BA4') || (c >= '\u0BA8' && c <= '\u0BAA')
+ || (c >= '\u0BAE' && c <= '\u0BB5') || (c >= '\u0BB7' && c <= '\u0BB9') || (c >= '\u0C05' && c <= '\u0C0C')
+ || (c >= '\u0C0E' && c <= '\u0C10') || (c >= '\u0C12' && c <= '\u0C28') || (c >= '\u0C2A' && c <= '\u0C33')
+ || (c >= '\u0C35' && c <= '\u0C39') || (c >= '\u0C60' && c <= '\u0C61') || (c >= '\u0C85' && c <= '\u0C8C')
+ || (c >= '\u0C8E' && c <= '\u0C90') || (c >= '\u0C92' && c <= '\u0CA8') || (c >= '\u0CAA' && c <= '\u0CB3')
+ || (c >= '\u0CB5' && c <= '\u0CB9')
+ || (c == '\u0CDE')
+ || (c >= '\u0CE0' && c <= '\u0CE1') || (c >= '\u0D05' && c <= '\u0D0C') || (c >= '\u0D0E' && c <= '\u0D10')
+ || (c >= '\u0D12' && c <= '\u0D28') || (c >= '\u0D2A' && c <= '\u0D39') || (c >= '\u0D60' && c <= '\u0D61')
+ || (c >= '\u0E01' && c <= '\u0E2E')
+ || (c == '\u0E30')
+ || (c >= '\u0E32' && c <= '\u0E33') || (c >= '\u0E40' && c <= '\u0E45') || (c >= '\u0E81' && c <= '\u0E82')
+ || (c == '\u0E84')
+ || (c >= '\u0E87' && c <= '\u0E88')
+ || (c == '\u0E8A')
+ || (c == '\u0E8D')
+ || (c >= '\u0E94' && c <= '\u0E97') || (c >= '\u0E99' && c <= '\u0E9F') || (c >= '\u0EA1' && c <= '\u0EA3')
+ || (c == '\u0EA5')
+ || (c == '\u0EA7')
+ || (c >= '\u0EAA' && c <= '\u0EAB') || (c >= '\u0EAD' && c <= '\u0EAE')
+ || (c == '\u0EB0')
+ || (c >= '\u0EB2' && c <= '\u0EB3')
+ || (c == '\u0EBD')
+ || (c >= '\u0EC0' && c <= '\u0EC4') || (c >= '\u0F40' && c <= '\u0F47') || (c >= '\u0F49' && c <= '\u0F69')
+ || (c >= '\u10A0' && c <= '\u10C5') || (c >= '\u10D0' && c <= '\u10F6')
+ || (c == '\u1100')
+ || (c >= '\u1102' && c <= '\u1103') || (c >= '\u1105' && c <= '\u1107')
+ || (c == '\u1109')
+ || (c >= '\u110B' && c <= '\u110C') || (c >= '\u110E' && c <= '\u1112')
+ || (c == '\u113C')
+ || (c == '\u113E')
+ || (c == '\u1140')
+ || (c == '\u114C')
+ || (c == '\u114E')
+ || (c == '\u1150')
+ || (c >= '\u1154' && c <= '\u1155')
+ || (c == '\u1159')
+ || (c >= '\u115F' && c <= '\u1161')
+ || (c == '\u1163')
+ || (c == '\u1165')
+ || (c == '\u1167')
+ || (c == '\u1169')
+ || (c >= '\u116D' && c <= '\u116E') || (c >= '\u1172' && c <= '\u1173')
+ || (c == '\u1175')
+ || (c == '\u119E')
+ || (c == '\u11A8')
+ || (c == '\u11AB')
+ || (c >= '\u11AE' && c <= '\u11AF') || (c >= '\u11B7' && c <= '\u11B8')
+ || (c == '\u11BA')
+ || (c >= '\u11BC' && c <= '\u11C2')
+ || (c == '\u11EB')
+ || (c == '\u11F0')
+ || (c == '\u11F9')
+ || (c >= '\u1E00' && c <= '\u1E9B') || (c >= '\u1EA0' && c <= '\u1EF9') || (c >= '\u1F00' && c <= '\u1F15')
+ || (c >= '\u1F18' && c <= '\u1F1D') || (c >= '\u1F20' && c <= '\u1F45') || (c >= '\u1F48' && c <= '\u1F4D')
+ || (c >= '\u1F50' && c <= '\u1F57')
+ || (c == '\u1F59')
+ || (c == '\u1F5B')
+ || (c == '\u1F5D')
+ || (c >= '\u1F5F' && c <= '\u1F7D') || (c >= '\u1F80' && c <= '\u1FB4') || (c >= '\u1FB6' && c <= '\u1FBC')
+ || (c == '\u1FBE')
+ || (c >= '\u1FC2' && c <= '\u1FC4') || (c >= '\u1FC6' && c <= '\u1FCC') || (c >= '\u1FD0' && c <= '\u1FD3')
+ || (c >= '\u1FD6' && c <= '\u1FDB') || (c >= '\u1FE0' && c <= '\u1FEC') || (c >= '\u1FF2' && c <= '\u1FF4')
+ || (c >= '\u1FF6' && c <= '\u1FFC')
+ || (c == '\u2126')
+ || (c >= '\u212A' && c <= '\u212B')
+ || (c == '\u212E')
+ || (c >= '\u2180' && c <= '\u2182') || (c >= '\u3041' && c <= '\u3094') || (c >= '\u30A1' && c <= '\u30FA')
+ || (c >= '\u3105' && c <= '\u312C') || (c >= '\uAC00' && c <= '\uD7A3');
+ }
+
+ /** BaseChar - XMl Specification 1.0 [86] */
+ private static final boolean Ideographic(char c) {
+ return (c >= '\u4E00' && c <= '\u9FA5') || c == '\u3007' || (c >= '\u3021' && c <= '\u3029');
+ }
+
+ /** CombiningChar - XMl Specification 1.0 [87] */
+ /*private static final boolean CombiningChar(char c) {
+ return (c >= '\u0300' && c <= '\u0345')
+ || (c >= '\u0360' && c <= '\u0361') || (c >= '\u0483' && c <= '\u0486') || (c >= '\u0591' && c <= '\u05A1')
+ || (c >= '\u05A3' && c <= '\u05B9') || (c >= '\u05BB' && c <= '\u05BD')
+ || (c == '\u05BF')
+ || (c >= '\u05C1' && c <= '\u05C2')
+ || (c == '\u05C4')
+ || (c >= '\u064B' && c <= '\u0652')
+ || (c == '\u0670')
+ || (c >= '\u06D6' && c <= '\u06DC') || (c >= '\u06DD' && c <= '\u06DF') || (c >= '\u06E0' && c <= '\u06E4')
+ || (c >= '\u06E7' && c <= '\u06E8') || (c >= '\u06EA' && c <= '\u06ED') || (c >= '\u0901' && c <= '\u0903')
+ || (c == '\u093C')
+ || (c >= '\u093E' && c <= '\u094C')
+ || (c == '\u094D')
+ || (c >= '\u0951' && c <= '\u0954') || (c >= '\u0962' && c <= '\u0963') || (c >= '\u0981' && c <= '\u0983')
+ || (c == '\u09BC')
+ || (c == '\u09BE')
+ || (c == '\u09BF')
+ || (c >= '\u09C0' && c <= '\u09C4') || (c >= '\u09C7' && c <= '\u09C8') || (c >= '\u09CB' && c <= '\u09CD')
+ || (c == '\u09D7')
+ || (c >= '\u09E2' && c <= '\u09E3')
+ || (c == '\u0A02')
+ || (c == '\u0A3C')
+ || (c == '\u0A3E')
+ || (c == '\u0A3F')
+ || (c >= '\u0A40' && c <= '\u0A42') || (c >= '\u0A47' && c <= '\u0A48') || (c >= '\u0A4B' && c <= '\u0A4D')
+ || (c >= '\u0A70' && c <= '\u0A71') || (c >= '\u0A81' && c <= '\u0A83')
+ || (c == '\u0ABC')
+ || (c >= '\u0ABE' && c <= '\u0AC5') || (c >= '\u0AC7' && c <= '\u0AC9') || (c >= '\u0ACB' && c <= '\u0ACD')
+ || (c >= '\u0B01' && c <= '\u0B03')
+ || (c == '\u0B3C')
+ || (c >= '\u0B3E' && c <= '\u0B43') || (c >= '\u0B47' && c <= '\u0B48') || (c >= '\u0B4B' && c <= '\u0B4D')
+ || (c >= '\u0B56' && c <= '\u0B57') || (c >= '\u0B82' && c <= '\u0B83') || (c >= '\u0BBE' && c <= '\u0BC2')
+ || (c >= '\u0BC6' && c <= '\u0BC8') || (c >= '\u0BCA' && c <= '\u0BCD')
+ || (c == '\u0BD7')
+ || (c >= '\u0C01' && c <= '\u0C03') || (c >= '\u0C3E' && c <= '\u0C44') || (c >= '\u0C46' && c <= '\u0C48')
+ || (c >= '\u0C4A' && c <= '\u0C4D') || (c >= '\u0C55' && c <= '\u0C56') || (c >= '\u0C82' && c <= '\u0C83')
+ || (c >= '\u0CBE' && c <= '\u0CC4') || (c >= '\u0CC6' && c <= '\u0CC8') || (c >= '\u0CCA' && c <= '\u0CCD')
+ || (c >= '\u0CD5' && c <= '\u0CD6') || (c >= '\u0D02' && c <= '\u0D03') || (c >= '\u0D3E' && c <= '\u0D43')
+ || (c >= '\u0D46' && c <= '\u0D48') || (c >= '\u0D4A' && c <= '\u0D4D')
+ || (c == '\u0D57')
+ || (c == '\u0E31')
+ || (c >= '\u0E34' && c <= '\u0E3A') || (c >= '\u0E47' && c <= '\u0E4E')
+ || (c == '\u0EB1')
+ || (c >= '\u0EB4' && c <= '\u0EB9') || (c >= '\u0EBB' && c <= '\u0EBC') || (c >= '\u0EC8' && c <= '\u0ECD')
+ || (c >= '\u0F18' && c <= '\u0F19')
+ || (c == '\u0F35')
+ || (c == '\u0F37')
+ || (c == '\u0F39')
+ || (c == '\u0F3E')
+ || (c == '\u0F3F')
+ || (c >= '\u0F71' && c <= '\u0F84') || (c >= '\u0F86' && c <= '\u0F8B') || (c >= '\u0F90' && c <= '\u0F95')
+ || (c == '\u0F97')
+ || (c >= '\u0F99' && c <= '\u0FAD') || (c >= '\u0FB1' && c <= '\u0FB7')
+ || (c == '\u0FB9')
+ || (c >= '\u20D0' && c <= '\u20DC')
+ || (c == '\u20E1')
+ || (c >= '\u302A' && c <= '\u302F')
+ || (c == '\u3099')
+ || (c == '\u309A');
+ }*/
+
+ /** Digit - XMl Specification 1.0 [88] */
+ private static final boolean Digit(char c) {
+ return (c >= '\u0030' && c <= '\u0039') || (c >= '\u0660' && c <= '\u0669') || (c >= '\u06F0' && c <= '\u06F9')
+ || (c >= '\u0966' && c <= '\u096F') || (c >= '\u09E6' && c <= '\u09EF') || (c >= '\u0A66' && c <= '\u0A6F')
+ || (c >= '\u0AE6' && c <= '\u0AEF') || (c >= '\u0B66' && c <= '\u0B6F') || (c >= '\u0BE7' && c <= '\u0BEF')
+ || (c >= '\u0C66' && c <= '\u0C6F') || (c >= '\u0CE6' && c <= '\u0CEF') || (c >= '\u0D66' && c <= '\u0D6F')
+ || (c >= '\u0E50' && c <= '\u0E59') || (c >= '\u0ED0' && c <= '\u0ED9') || (c >= '\u0F20' && c <= '\u0F29');
+ }
+
+ /** Extender - XMl Specification 1.0 [89] */
+ private static final boolean Extender(char c) {
+ return c == '\u00B7' || c == '\u02D0' || c == '\u02D1' || c == '\u0387'
+ || c == '\u0640' || c == '\u0E46' || c == '\u0EC6' || c == '\u3005'
+ || (c >= '\u3031' && c <= '\u3035') || (c >= '\u309D' && c <= '\u309E') || (c >= '\u30FC' && c <= '\u30FE');
+ }
+
+ /** Whitespace - XML Specification 1.0 [3] */
+ private static final boolean S(char c) {
+ return c == '\u0020' || c == '\u0009' || c == '\r' || c == '\n';
+ }
+}