4 import java.io.IOException;
5 import java.io.EOFException;
8 * An Event-Driving, Non-Validating XML Parser with Namespace support.
10 * A subclass can implement the abstract functions for receiving details
11 * about an xml file as it is parsed. To initate a parse, use the parse()
14 * <h3>IMPLEMENTATION NOTES</h3>
15 * <p>As the parser traverses into an element, it adds it to the linked list
16 * called <tt>elements</tt>. However, <tt>elements</tt> has been pre-filled
17 * with instances of the Element inner class. So in the vast majority of
18 * cases, the pointer current is moved along one, and the values for the
19 * new element are filled into the current object.</p>
21 * <p>This parser supports all the unicode ranges required by the XML
22 * Specification. However, it is optimised for well-formed ASCII documents.
23 * Documents containing unicode Names and Attributes will take much longer
24 * to process, and invalid documents (badly formed Names or invalid attributes)
25 * will be run through a test on every single unicode character range before
26 * being declared invalid.</p>
28 * <h3>IMPLEMENTATION RULES</h3>
30 * <li>Each time the buffer offset <tt>off</tt> is moved, the length
31 * <tt>len</tt> must be decreased.</li>
32 * <li>Each time the buffer length is decreased, it must be checked to make
33 * sure it is >0.</li>
36 * <h3>Other Notes</h3>
38 * <li><i>error</i> is defined as a Validity Constraint Violation and is recoverable</li>
39 * <li><i>fatal error</i> is defined as a Well-formedness Constraint Violation and is not recoverable</li>
42 * @author David Crawshaw
43 * @see XML-Specification-1.0 http://w3.org/TR/REC-xml
45 public abstract class XML
47 /////////////////////////////////////////////////////////////////////////////////////////////
49 /////////////////////////////////////////////////////////////////////////////////////////////
51 public static final int BUFFER_SIZE = 255;
53 /** static pool of XML.Element instances shared by all XML Parsers.
54 * elements in the queue have dirty prev and next references, that need cleaning before use. */
55 private static final Queue elements = new Queue(30);
57 private static final char[] single_amp = new char[] { '&' };
58 private static final char[] single_apos = new char[] { '\'' };
59 private static final char[] single_gt = new char[] { '>' };
60 private static final char[] single_lt = new char[] { '<' };
61 private static final char[] single_quot = new char[] { '"' };
71 private Element current;
73 // used in readEntity() to process a single character without creating a new array
74 private char[] singlechar = new char[1];
77 public XML() { this(BUFFER_SIZE); }
79 public XML(int bSize) {
80 buf = new char[bSize];
82 current = (Element)elements.remove(false);
83 if (current == null) current = new Element();
84 current.prev = current.next = null;
88 /** Returns the line number at the beginning of the last process call. */
89 public int getLine() { return line; }
91 /** Returns the column number at the beginning of the last process call. */
92 public int getCol() { return col; }
95 * Parse given input and call the abstract event functions.
97 * Careful with threading, as this function is not synchronized.
99 public final void parse(Reader reader) throws IOException, XMLException {
104 clean(); // clean up possible mid-way linked-list element
107 // process the stream
110 if (current.qName == null) break;
111 throw new WFCException("reached eof without closing <"+current.qName+"> element", getLine(), getCol());
114 if (buf[off] == '<') readTag();
115 readChars(current.qName != null);
117 } finally { clean(); } // clean up elements
120 /** remove any leftover elements from the linked list and queue them */
121 private final void clean() {
122 while (current.prev != null) elements.append((current = current.prev).next);
124 current.qName = null;
127 /** reads in a tag. expects <tt>buf[off] == '<'</tt> */
128 private final void readTag() throws IOException, XMLException {
129 // Start Tag '<' Name (S Attribute)* S? '>'
130 boolean starttag = true;
132 // End Tag '</' Name S? '>'
133 boolean endtag = false;
135 // if (starttag & endtag) then: EmptyElemTag '<' Name (S Attribute)* S? '/>'
137 // Position in the name of the ':' namespace prefix
143 if (!buffer(1)) throw new EOFException("Unexpected EOF processing element tag");
145 // work out what we can from the beginning of the tag
148 // definitions here don't necessarily conform to xml spec (as DTDs not yet implemented)
150 if (!buffer(4)) throw new EOFException("Unexpected EOF processing <! element");
155 if (buf[off+1] != '-') { bad = true; break; }
156 col += 2; off += 2; len -= 2;
158 // Comment '<!--' ((Char - '-') | ('-' (Char - '-')))* '-->'
159 readChars(false, "-->", false);
160 col += 3; off += 3; len -= 3;
163 // we don't care about the following definitions
167 || buf[off+1] != 'T' || buf[off+2] != 'T' || buf[off+3] != 'L'
168 || buf[off+4] != 'I' || buf[off+5] != 'S' || buf[off+6] != 'T') {
171 col += 7; off += 7; len -= 7;
173 // ATTLIST '<!ATTLIST' (Char* - '>') '>'
174 readChars(false, ">", true);
179 || buf[off+1] != 'O' || buf[off+2] != 'C' || buf[off+3] != 'T'
180 || buf[off+4] != 'Y' || buf[off+5] != 'P' || buf[off+6] != 'E') {
183 col += 7; off += 7; len -= 7;
185 // DTD '<!DOCTYPE' (Char* - '>') '>'
186 readChars(false, ">", true);
192 } else if (buf[off+1] == 'L' && buf[off+2] == 'E' && buf[off+3] == 'M'
193 && buf[off+4] == 'E' && buf[off+5] == 'N' && buf[off+6] == 'T') {
194 // ELEMENT '<!ELEMENT' (Char* - '>') '>'
195 readChars(false, ">", true);
198 } else if (buf[off+1] == 'N' && buf[off+2] == 'T' && buf[off+3] == 'I'
199 && buf[off+4] == 'T' && buf[off+5] == 'Y') {
200 // ENTITY '<!ENTITY' (Char* - '>') '>'
201 readChars(false, ">", true);
211 || buf[off+1] != 'O' || buf[off+2] != 'T' || buf[off+3] != 'A' || buf[off+4] != 'T'
212 || buf[off+5] != 'I' || buf[off+6] != 'O' || buf[off+7] != 'N') {
215 col += 8; off += 8; len -= 8;
216 // NOTATION '<!NOTATION' (Char* - '>') '>'
217 readChars(false, ">", true);
224 if (bad) throw new MarkupException("element tag start character is invalid", getLine(), getCol());
226 } else if (s == '?') {
227 // PI (Ignored) '<?' (Char* - (Char* '?>' Char*)) '?>'
229 readChars(false, "?>", true);
230 if (!buffer(2)) throw new EOFException("Unexpected EOF at end of Processing Instruction");
231 col += 2; off += 2; len -= 2;
233 } else if (s == '[') {
235 || buf[off+1] != 'C' || buf[off+2] != 'D' || buf[off+3] != 'A'
236 || buf[off+4] != 'T' || buf[off+5] != 'A' || buf[off+6] != '[') {
238 // Conditional '<![' (Char* - (Char* ']]>' Char*)) ']]>'
239 readChars(false, "]]>", false);
241 col += 7; off += 7; len -=7;
242 // CDATA '<![CDATA[' (Char* - (Char* ']]>' Char*)) ']]>'
243 readChars(true, "]]>", false);
245 col += 3; off += 3; len -= 3;
248 // End Tag '</' Name S? '>'
253 if (!buffer(1)) throw new EOFException("Unexpected EOF processing end tag");
257 if (!Name(s)) throw new MarkupException("invalid starting character in element name", getLine(), getCol());
259 // find the element name (defined in XML Spec: section 2.3)
260 for (namelen = 0; ; namelen++) {
261 if (!buffer(namelen+1)) throw new EOFException("Unexpected EOF in element tag name");
263 s = buf[off+namelen];
265 if (S(s) || s == '>') {
267 } else if (s == '/') {
270 } else if (s == ':' && namelen > 0 && prefix < 1) {
271 // we have a definition of the prefix range available
273 } else if (!NameChar(s)) {
274 throw new MarkupException("element name contains invalid character", getLine(), getCol());
278 // process name (based on calculated region)
279 if (namelen < 1) throw new MarkupException("element name is null", getLine(), getCol());
281 // we have marked out the name region, so turn it into a string and move on
282 String qName = new String(buf, off, namelen);
284 col += namelen; off += namelen; len -= namelen;
287 // create the in-memory element representation of this beast
288 // if current.qName == null then this is the root element we're dealing with
289 if (current.qName != null) {
290 if (current.next == null) {
291 // we're at the end of the default element depth
292 current.next = (Element)elements.remove(false);
293 if (current.next == null) current.next = new Element();
294 current.next.prev = current;
295 current.next.next = null;
297 current = current.next;
301 current.qName = qName;
302 current.defaultUri = current.uri = null;
305 current.prefix = current.qName.substring(0, prefix);
306 current.localName = current.qName.substring(prefix+1);
308 current.prefix = null;
309 current.localName = current.qName;
312 // process attributes
314 if (!buffer(1)) throw new EOFException("Unexpected EOF - processing attributes part 1");
315 while (buf[off] != '/' && buf[off] != '>') {
317 if (!buffer(1)) throw new EOFException("Unexpected EOF - processing attributes part 2");
321 // inherit namespace default uri if attribute was not provided
322 if (current.defaultUri == null) {
323 current.defaultUri = (current.prev != null) ? current.prev.defaultUri : null;
326 // work out the uri of this element
327 if (current.prefix == null) {
328 // element has no prefix, therefore is the default uri
329 current.uri = current.defaultUri;
331 // work back through the hashtables until uri is found
332 for (Element e = current; e != null && current.uri == null; e = e.prev) {
333 current.uri = (String)e.urimap.get(current.prefix);
335 if (current.uri == null) current.addError(new NCException("undefined prefix '"+current.prefix+"'", getLine(), getCol()));
339 // this is an end-of-element tag
340 if (!qName.equals(current.qName)) throw new WFCException(
341 "end tag </"+qName+"> does not line up with start tag <"+current.qName+">", getLine(), getCol()
345 // deal with whitespace
349 if (!buffer(1)) throw new EOFException("Unexpected EOF before end of tag");
350 if (buf[off] == '/') {
354 if (!buffer(1)) throw new EOFException("Unexpected EOF before end of endtag");
355 if (buf[off] == '>') {
358 throw new MarkupException("missing '>' character from element '"+qName+"'", getLine(), getCol());
361 // send element signals
362 if (starttag) startElement(current);
366 // we just closed an element, so remove it from the element 'stack'
367 if (current.prev == null) {
368 // we just finished the root element
369 current.qName = null;
371 elements.append((current = current.prev).next);
378 /** reads in an attribute of an element. expects Name(buf[off]) */
379 private final void readAttribute() throws IOException, XMLException {
382 String n, v, p, u; // attribute name, value, prefix and uri respectively
383 n = v = p = u = null;
386 // find the element name (defined in XML Spec: section 2.3)
387 for (ref= 0; ; ref++) {
388 if (!buffer(ref+1)) throw new EOFException("Unexpected EOF in read attribute loop part 1");
392 if (s == '=' || S(s)) {
394 } else if (s == ':' && ref > 0 && prefix < 1) {
395 // we have a definition of the prefix range available
397 } else if (!NameChar(s)) {
398 throw new MarkupException("attribute name contains invalid characters", getLine(), getCol());
402 // determine prefix and key name
404 p = new String(buf, off, prefix-1);
405 col += prefix; off += prefix; len -= prefix; ref -= prefix;
407 n = new String(buf, off, ref);
408 col += ref; off += ref; len -= ref;
410 // find name/value divider ('=')
412 if (!buffer(1)) throw new EOFException("Unexpected EOF before attribute '=' divider");
413 if (buf[off] != '=') throw new MarkupException("attribute name not followed by '=' sign", getLine(), getCol());
418 if (!buffer(1)) throw new EOFException("Unexpected EOF after attribute '=' divider");
421 if (buf[off] == '\'' || buf[off] == '"') {
424 throw new MarkupException("attribute '"+n+"' must have attribute wrapped in ' or \"", getLine(), getCol());
428 // find the attribute value
429 attval: for (ref = 0; ; ref++) {
430 if (!buffer(ref+1)) throw new EOFException("Unexpected EOF in attribute value");
432 if (buf[off+ref] == wrap) {
434 } else if (buf[off+ref] == '<') {
435 throw new WFCException("attribute value for '"+n+"' must not contain '<'", getLine(), getCol());
439 v = new String(buf, off, ref);
440 col += ref; off += ref; len -= ref;
442 // remove end wrapper character
446 if (p != null && p.equals("xmlns")) {
447 current.urimap.put(n, v);
448 } else if (n.equals("xmlns")) {
449 if (current.defaultUri != null) {
450 current.addError(new NCException("default namespace definition repeated", getLine(), getCol()));
452 current.defaultUri = v;
455 // find attribute uri
457 for (Element e = current; e != null && u == null; e = e.prev) { u = e.uri; }
459 for (Element e = current; e != null && u == null; e = e.prev) { u = (String)e.urimap.get(p); }
460 if (u == null) current.addError(new NCException("undefined attribute prefix '"+current.prefix+"'", getLine(), getCol()));
463 // check to see if attribute is a repeat
464 for (int i=0; current.len > i; i++) if (n.equals(current.keys[i]) && u.equals(current.uris[i])) throw new WFCException(
465 "attribute name '"+n+"' may not appear more than once in the same element tag", getLine(), getCol()
468 // add attribute to the attribute arrays
469 if (current.len == current.keys.length) current.morekeys();
470 current.keys[current.len] = n;
471 current.vals[current.len] = v;
472 current.uris[current.len] = u;
477 /** reads an entity and processes out its value. expects buf[off] == '&' */
478 private final void readEntity() throws IOException, XMLException {
480 if (!buffer(2)) throw new EOFException("Unexpected EOF reading entity");
482 boolean unknown = false;
488 if (buf[off] == 'x') { off++; len--; radix = 16; } else { radix = 10; }
491 // read in each char, then shift total value to the left and add the extra
492 // style of loop is slightly different from all the others, as this should run a limited number of times
493 findchar: while (true) {
494 if (!buffer(1)) throw new EOFException("Unexpected EOF reading entity");
495 int d = Character.digit(buf[off], radix);
497 if (buf[off] != ';') throw new WFCException("illegal characters in entity reference", getLine(), getCol());
506 singlechar[0] = Character.forDigit(c, radix);
507 characters(singlechar, 0, 1);
511 if (buffer(4) && buf[off+1] == 'm' && buf[off+2] == 'p' && buf[off+3] == ';') {
512 characters(single_amp, 0, 1); // &
513 off += 4; len -= 4; col++;
514 } else if (buffer(5) && buf[off+1] == 'p' && buf[off+2] == 'o' && buf[off+3] == 's' && buf[off+4] == ';') {
515 characters(single_apos, 0, 1); // '
516 off += 5; len -= 5; col++;
523 if (buffer(3) && buf[off+1] == 't' && buf[off+2] == ';') {
524 characters(single_gt, 0, 1); // >
525 off += 3; len -= 3; col++;
532 if (buffer(3) && buf[off+1] == 't' && buf[off+2] == ';') {
533 characters(single_lt, 0, 1); // <
534 off += 3; len -= 3; col++;
541 if (buffer(5) && buf[off+1] == 'u' && buf[off+2] == 'o' && buf[off+3] == 't' && buf[off+4] == ';') {
542 characters(single_quot, 0, 1); // "
543 off += 5; len -= 5; col++;
549 // TODO: check a parser-level Hash of defined entities
552 if (unknown) throw new WFCException("unknown entity (<!ENTITY> not supported)", getLine(), getCol());
555 /** reads until the passed string is encountered. */
556 private final void readChars(boolean p, String match, boolean entities) throws IOException, XMLException {
558 char[] end = match.toCharArray();
560 for (boolean more = true; more;) {
561 if (!buffer(1)) return;
563 buf: for (ref = 0; ref < len; ref++) {
564 switch (buf[off+ref]) {
565 case '\r': // windows or macos9 newline
566 // normalise and process
567 buf[off+ref] = '\n'; ref++;
568 if (p) characters(buf, off, ref);
569 off += ref; len -= ref; ref = -1;
572 // windows double-char newline; skip the next char
573 if (!buffer(1)) return;
574 if (buf[off] == '\n') { off++; len--; }
577 case '\n': // unix newline
579 if (p) characters(buf, off, ref);
580 off += ref; len -= ref; ref = -1;
587 if (ref > 0) characters(buf, off, ref);
588 off += ref; len -= ref; ref = -1;
595 if (!buffer(ref+end.length)) continue buf;
596 for (int i=0; end.length > i; i++) if (end[i] != buf[off+ref+i]) continue buf;
602 if (p && ref > 0) characters(buf, off, ref);
603 off += ref; len -= ref; col += ref;
608 * reads until a <tt><</tt> symbol is encountered
609 * @param p If true call the characters(char[],int,int) funciton for the processed characters
611 private final void readChars(boolean p) throws IOException, XMLException {
614 for (boolean more = true; more;) {
615 if (!buffer(1)) return;
617 buf: for (ref = 0; ref < len; ref++) {
618 switch (buf[off+ref]) {
619 case '\r': // windows or macos9 newline
620 // normalise and process
621 buf[off+ref] = '\n'; ref++;
622 if (p) characters(buf, off, ref);
623 off += ref; len -= ref; ref = -1;
626 // windows double-char newline; skip the next char
627 if (!buffer(1)) return;
628 if (buf[off] == '\n') { off++; len--; }
631 case '\n': // unix newline
633 if (p) characters(buf, off, ref);
634 off += ref; len -= ref; ref = -1;
640 if (ref > 0) characters(buf, off, ref);
641 off += ref; len -= ref; ref = -1;
646 case '<': // end of chars section
652 if (p && ref > 0) characters(buf, off, ref);
653 off += ref; len -= ref; col += ref;
657 /** reads until a non-whitespace symbol is encountered */
658 private final void readWhitespace() throws IOException, XMLException {
661 for (boolean more = true; more;) {
662 if (!buffer(1)) return;
664 buf: for (ref = 0; ref < len; ref++) {
665 switch (buf[off+ref]) {
666 case '\r': // windows or macos9 newline
667 // normalise and process
669 whitespace(buf, off, ++ref);
670 off += ref; len -= ref; ref = -1;
673 // windows double-char newline; skip the next char
674 if (!buffer(1)) return;
675 if (buf[off] == '\n') { off++; len--; }
678 case '\n': // unix newline
679 whitespace(buf, off, ++ref);
680 off += ref; len -= ref; ref = -1;
688 default: // end of whitespace
694 off += ref; len -= ref; col += ref;
699 * attempt to fill the buffer.
701 * @param min Minimum number of characters to read (even if we have to block to do it).
702 * @return return false if min can't be reached.
704 private final boolean buffer(int min) throws IOException {
705 if (len > min) return true;
707 if (buf.length - (off+len) >= min) {
708 // plenty of space left on the end of the buffer
709 } else if (off >= min) {
710 // moving offset data to start will leave enough free space on the end
711 System.arraycopy(buf, off, buf, 0, len);
714 // buffer size will have to be increased
715 char[] newbuf = new char[buf.length * 2];
716 System.arraycopy(buf, off, newbuf, 0, len);
722 int newlen = in.read(buf, off+len, buf.length-(off+len));
723 if (newlen < 0) return false;
731 /////////////////////////////////////////////////////////////////////////////////////////////
732 // Abstract SAX-Like Interface
733 /////////////////////////////////////////////////////////////////////////////////////////////
736 * Called when the start of an element is processed.
738 * <p>The array of Attribute names and values may be longer than the
739 * number of entries they contain, but all the entries will be
740 * packed at the top.</p>
742 * <p><b>DO NOT</b> store a reference to the attribute arrays, as
743 * they are reused by other elements.</p>
745 public abstract void startElement(Element e) throws SchemaException;
748 * Represents a line of character data.
750 * <p>Newlines are all normalised to the Unix \n as per the XML Spec,
751 * and a newline will only appear as the last character in the passed
754 * <p>XML.getLine() and XML.getCol() report the position at the
755 * beginning of this character segment, which can be processed in a
756 * line-by-line fashion due to the above newline restriction.</p>
758 public abstract void characters(char[] ch, int start, int length) throws SchemaException;
760 /** Represents a line of ignorable whitespace. */
761 public abstract void whitespace(char[] ch, int start, int length) throws SchemaException;
763 /** Represents the end of an Element. */
764 public abstract void endElement(Element e) throws SchemaException;
767 /////////////////////////////////////////////////////////////////////////////////////////////
768 // Inner Classes for Parser Support
769 /////////////////////////////////////////////////////////////////////////////////////////////
772 * Used as a struct for holding information about a current element,
773 * and acts as a linked list entry.
775 * <p>Each element stores a hashtable of namespace definitions against
776 * their respective prefix, and a variable holding their default
777 * uri. If they did not specify a default uri, their
778 * parent's uri is copied in to keep up the sembelence of speedy
781 * <h3>SLOWEST PART OF THE XML PARSER</h3>
782 * <p>To implement the Namespace Specification exactly, we have to
783 * store prefix mappings for elements away from its parents and
784 * siblings. This means if a child of a child of-a child uses
785 * a prefix defined in the root, we have to search each Hashtable
786 * in each Element until we get to the root.</p>
788 * <p>Unfortunetally, every other solution I can think of requires
789 * more work than this one, shifted to different parts of the
792 public static final class Element
794 public Element next, prev;
796 /** A hashtable of all namespace prefixes that are defined by this element. */
799 /** An array of attribute names. */
800 public String[] keys;
802 /** An array of attribute values. */
803 public String[] vals;
805 /** An array of attribute uris. */
806 public String[] uris;
808 /** An array of non-fatal errors related to this element. */
809 public XMLException[] errors;
811 /** Current number of attributes in the <tt>keys</tt> and <tt>vals</tt> arrays. */
814 /** Default URI for this element and its children with no prefix. */
815 public String defaultUri;
817 /** URI of current tag. XML Namespace Spec 14-Jan-1999 section 1 */
820 /** LocalPart of current element. XML Namespace Spec 14-Jan-1999 [8] */
821 public String localName;
823 /** Qualified Name of current element. XML Namespace Spec 14-Jan-1999 [6] */
826 /** Prefix of current element. Substring of qName. XML Namespace Spec 14-Jan-1999 [7] */
827 public String prefix;
830 defaultUri = uri = prefix = localName = qName = null;
831 urimap = new Hash(3,3);
832 keys = new String[10];
833 vals = new String[10];
834 uris = new String[10];
835 errors = new XMLException[] {};
839 /** increase the size of the attributes arrays */
841 String[] newkeys = new String[keys.length+5];
842 String[] newvals = new String[vals.length+5];
843 String[] newuris = new String[uris.length+5];
844 System.arraycopy(keys, 0, newkeys, 0, keys.length);
845 System.arraycopy(vals, 0, newvals, 0, vals.length);
846 System.arraycopy(uris, 0, newuris, 0, uris.length);
847 keys = newkeys; vals = newvals; uris = newuris;
850 /** empty out the arrays */
852 if (keys.length != vals.length || vals.length != uris.length) {
853 keys = new String[10]; vals = new String[10]; uris = new String[10];
855 for (int i=0; keys.length > i; i++) { keys[i] = null; vals[i] = null; uris[i] = null; }; len = 0;
857 errors = new XMLException[] {};
860 /** add an error to the errors array */
861 void addError(XMLException e) {
862 // it doesn't really matter about continually expanding the array, as this case is quite rare
863 XMLException[] newe = new XMLException[errors.length+1];
864 System.arraycopy(errors, 0, newe, 0, errors.length);
865 newe[errors.length] = e;
870 /** Parse or Structural Error */
871 public static class XMLException extends Exception
875 private String error;
877 public XMLException(String e) { this(e, -1, -1); }
879 public XMLException(String e, int l, int c) {
885 public int getLine() { return this.line; }
886 public int getCol() { return this.col; }
887 public String getMessage() { return this.error; }
890 /** Violation of Markup restrictions in XML Specification - Fatal Error */
891 public static class MarkupException extends XMLException { public MarkupException(String e, int l, int c) { super(e,l,c); } }
893 /** Well-Formedness Constraint Violation - Fatal Error */
894 public static final class WFCException extends MarkupException { public WFCException(String e, int l, int c) { super(e,l,c); } }
896 /** Namespace Constraint Violation - Recoverable Error */
897 public static final class NCException extends XMLException { public NCException(String e, int l, int c) { super(e,l,c); } }
899 /** Schema Violation - Fatal Error */
900 public static class SchemaException extends XMLException {
901 public SchemaException(String e) { this(e, -1, -1); }
902 public SchemaException(String e, int l, int c) { super(e,l,c); }
906 /////////////////////////////////////////////////////////////////////////////////////////////
907 // Static Support JSFunctions for the XML Specification
908 /////////////////////////////////////////////////////////////////////////////////////////////
910 // attempt to avoid these functions unless you *expect* the input to fall in the given range.
912 /** First Character of Name - XML Specification 1.0 [5] */
913 private static final boolean Name(char c) {
914 return BaseCharAscii(c) || c == '_' || c == ':' || Letter(c);
917 /** NameChar - XML Specification 1.0 [4] */
918 private static final boolean NameChar(char c) {
919 return BaseCharAscii(c) || c == '.' || c == '-' || c == '_' || c == ':'
920 || Digit(c) || Letter(c) || Extender(c); // TODO: || CombiningChar(c);
923 /** BaseChar - XMl Specification 1.0 [84] */
924 private static final boolean Letter(char c) {
925 return BaseChar(c) || Ideographic(c);
928 /** Elements of BaseChar that exist in ASCII. */
929 private static final boolean BaseCharAscii(char c) {
930 return (c >= '\u0041' && c <= '\u005A') || (c >= '\u0061' && c <= '\u007A');
933 /** Char - XML Specification 1.0 [2] */
934 private static final boolean Char(char c) {
935 // u000A == r and u000D == n, but the javac compiler can't handle the \ u form
936 return c == '\u0009' || c == '\r' || c == '\n'
937 || (c >= '\u0020' && c <= '\uD7FF')
938 || (c >= '\uE000' && c <= '\uFFFD');
941 /** BaseChar - XML Specification 1.0 [85] */
942 private static final boolean BaseChar(char c) {
943 return BaseCharAscii(c) || (c >= '\u00C0' && c <= '\u00D6')
944 || (c >= '\u00D8' && c <= '\u00F6') || (c >= '\u00F8' && c <= '\u00FF') || (c >= '\u0100' && c <= '\u0131')
945 || (c >= '\u0134' && c <= '\u013E') || (c >= '\u0141' && c <= '\u0148') || (c >= '\u014A' && c <= '\u017E')
946 || (c >= '\u0180' && c <= '\u01C3') || (c >= '\u01CD' && c <= '\u01F0') || (c >= '\u01F4' && c <= '\u01F5')
947 || (c >= '\u01FA' && c <= '\u0217') || (c >= '\u0250' && c <= '\u02A8') || (c >= '\u02BB' && c <= '\u02C1')
948 || (c == '\u0386') || (c >= '\u0388' && c <= '\u038A') || (c == '\u038C')
949 || (c >= '\u038E' && c <= '\u03A1') || (c >= '\u03A3' && c <= '\u03CE') || (c >= '\u03D0' && c <= '\u03D6')
950 || (c == '\u03DA') || (c == '\u03DC') || (c == '\u03DE')
952 || (c >= '\u03E2' && c <= '\u03F3') || (c >= '\u0401' && c <= '\u040C') || (c >= '\u040E' && c <= '\u044F')
953 || (c >= '\u0451' && c <= '\u045C') || (c >= '\u045E' && c <= '\u0481') || (c >= '\u0490' && c <= '\u04C4')
954 || (c >= '\u04C7' && c <= '\u04C8') || (c >= '\u04CB' && c <= '\u04CC') || (c >= '\u04D0' && c <= '\u04EB')
955 || (c >= '\u04EE' && c <= '\u04F5') || (c >= '\u04F8' && c <= '\u04F9') || (c >= '\u0531' && c <= '\u0556')
957 || (c >= '\u0561' && c <= '\u0586') || (c >= '\u05D0' && c <= '\u05EA') || (c >= '\u05F0' && c <= '\u05F2')
958 || (c >= '\u0621' && c <= '\u063A') || (c >= '\u0641' && c <= '\u064A') || (c >= '\u0671' && c <= '\u06B7')
959 || (c >= '\u06BA' && c <= '\u06BE') || (c >= '\u06C0' && c <= '\u06CE') || (c >= '\u06D0' && c <= '\u06D3')
961 || (c >= '\u06E5' && c <= '\u06E6') || (c >= '\u0905' && c <= '\u0939')
963 || (c >= '\u0958' && c <= '\u0961') || (c >= '\u0985' && c <= '\u098C') || (c >= '\u098F' && c <= '\u0990')
964 || (c >= '\u0993' && c <= '\u09A8') || (c >= '\u09AA' && c <= '\u09B0')
966 || (c >= '\u09B6' && c <= '\u09B9') || (c >= '\u09DF' && c <= '\u09E1') || (c >= '\u09F0' && c <= '\u09F1')
967 || (c >= '\u0A05' && c <= '\u0A0A') || (c >= '\u0A0F' && c <= '\u0A10') || (c >= '\u0A13' && c <= '\u0A28')
968 || (c >= '\u0A2A' && c <= '\u0A30') || (c >= '\u0A32' && c <= '\u0A33') || (c >= '\u0A35' && c <= '\u0A36')
969 || (c >= '\u0A38' && c <= '\u0A39') || (c >= '\u0A59' && c <= '\u0A5C')
971 || (c >= '\u0A72' && c <= '\u0A74') || (c >= '\u0A85' && c <= '\u0A8B')
973 || (c >= '\u0A8F' && c <= '\u0A91') || (c >= '\u0A93' && c <= '\u0AA8') || (c >= '\u0AAA' && c <= '\u0AB0')
974 || (c >= '\u0AB2' && c <= '\u0AB3') || (c >= '\u0AB5' && c <= '\u0AB9')
977 || (c >= '\u0B05' && c <= '\u0B0C') || (c >= '\u0B0F' && c <= '\u0B10') || (c >= '\u0B13' && c <= '\u0B28')
978 || (c >= '\u0B2A' && c <= '\u0B30') || (c >= '\u0B32' && c <= '\u0B33') || (c >= '\u0B36' && c <= '\u0B39')
980 || (c >= '\u0B5C' && c <= '\u0B5D') || (c >= '\u0B5F' && c <= '\u0B61') || (c >= '\u0B85' && c <= '\u0B8A')
981 || (c >= '\u0B8E' && c <= '\u0B90') || (c >= '\u0B92' && c <= '\u0B95') || (c >= '\u0B99' && c <= '\u0B9A')
983 || (c >= '\u0B9E' && c <= '\u0B9F') || (c >= '\u0BA3' && c <= '\u0BA4') || (c >= '\u0BA8' && c <= '\u0BAA')
984 || (c >= '\u0BAE' && c <= '\u0BB5') || (c >= '\u0BB7' && c <= '\u0BB9') || (c >= '\u0C05' && c <= '\u0C0C')
985 || (c >= '\u0C0E' && c <= '\u0C10') || (c >= '\u0C12' && c <= '\u0C28') || (c >= '\u0C2A' && c <= '\u0C33')
986 || (c >= '\u0C35' && c <= '\u0C39') || (c >= '\u0C60' && c <= '\u0C61') || (c >= '\u0C85' && c <= '\u0C8C')
987 || (c >= '\u0C8E' && c <= '\u0C90') || (c >= '\u0C92' && c <= '\u0CA8') || (c >= '\u0CAA' && c <= '\u0CB3')
988 || (c >= '\u0CB5' && c <= '\u0CB9')
990 || (c >= '\u0CE0' && c <= '\u0CE1') || (c >= '\u0D05' && c <= '\u0D0C') || (c >= '\u0D0E' && c <= '\u0D10')
991 || (c >= '\u0D12' && c <= '\u0D28') || (c >= '\u0D2A' && c <= '\u0D39') || (c >= '\u0D60' && c <= '\u0D61')
992 || (c >= '\u0E01' && c <= '\u0E2E')
994 || (c >= '\u0E32' && c <= '\u0E33') || (c >= '\u0E40' && c <= '\u0E45') || (c >= '\u0E81' && c <= '\u0E82')
996 || (c >= '\u0E87' && c <= '\u0E88')
999 || (c >= '\u0E94' && c <= '\u0E97') || (c >= '\u0E99' && c <= '\u0E9F') || (c >= '\u0EA1' && c <= '\u0EA3')
1002 || (c >= '\u0EAA' && c <= '\u0EAB') || (c >= '\u0EAD' && c <= '\u0EAE')
1004 || (c >= '\u0EB2' && c <= '\u0EB3')
1006 || (c >= '\u0EC0' && c <= '\u0EC4') || (c >= '\u0F40' && c <= '\u0F47') || (c >= '\u0F49' && c <= '\u0F69')
1007 || (c >= '\u10A0' && c <= '\u10C5') || (c >= '\u10D0' && c <= '\u10F6')
1009 || (c >= '\u1102' && c <= '\u1103') || (c >= '\u1105' && c <= '\u1107')
1011 || (c >= '\u110B' && c <= '\u110C') || (c >= '\u110E' && c <= '\u1112')
1018 || (c >= '\u1154' && c <= '\u1155')
1020 || (c >= '\u115F' && c <= '\u1161')
1025 || (c >= '\u116D' && c <= '\u116E') || (c >= '\u1172' && c <= '\u1173')
1030 || (c >= '\u11AE' && c <= '\u11AF') || (c >= '\u11B7' && c <= '\u11B8')
1032 || (c >= '\u11BC' && c <= '\u11C2')
1036 || (c >= '\u1E00' && c <= '\u1E9B') || (c >= '\u1EA0' && c <= '\u1EF9') || (c >= '\u1F00' && c <= '\u1F15')
1037 || (c >= '\u1F18' && c <= '\u1F1D') || (c >= '\u1F20' && c <= '\u1F45') || (c >= '\u1F48' && c <= '\u1F4D')
1038 || (c >= '\u1F50' && c <= '\u1F57')
1042 || (c >= '\u1F5F' && c <= '\u1F7D') || (c >= '\u1F80' && c <= '\u1FB4') || (c >= '\u1FB6' && c <= '\u1FBC')
1044 || (c >= '\u1FC2' && c <= '\u1FC4') || (c >= '\u1FC6' && c <= '\u1FCC') || (c >= '\u1FD0' && c <= '\u1FD3')
1045 || (c >= '\u1FD6' && c <= '\u1FDB') || (c >= '\u1FE0' && c <= '\u1FEC') || (c >= '\u1FF2' && c <= '\u1FF4')
1046 || (c >= '\u1FF6' && c <= '\u1FFC')
1048 || (c >= '\u212A' && c <= '\u212B')
1050 || (c >= '\u2180' && c <= '\u2182') || (c >= '\u3041' && c <= '\u3094') || (c >= '\u30A1' && c <= '\u30FA')
1051 || (c >= '\u3105' && c <= '\u312C') || (c >= '\uAC00' && c <= '\uD7A3');
1054 /** BaseChar - XMl Specification 1.0 [86] */
1055 private static final boolean Ideographic(char c) {
1056 return (c >= '\u4E00' && c <= '\u9FA5') || c == '\u3007' || (c >= '\u3021' && c <= '\u3029');
1059 /** CombiningChar - XMl Specification 1.0 [87] */
1060 /*private static final boolean CombiningChar(char c) {
1061 return (c >= '\u0300' && c <= '\u0345')
1062 || (c >= '\u0360' && c <= '\u0361') || (c >= '\u0483' && c <= '\u0486') || (c >= '\u0591' && c <= '\u05A1')
1063 || (c >= '\u05A3' && c <= '\u05B9') || (c >= '\u05BB' && c <= '\u05BD')
1065 || (c >= '\u05C1' && c <= '\u05C2')
1067 || (c >= '\u064B' && c <= '\u0652')
1069 || (c >= '\u06D6' && c <= '\u06DC') || (c >= '\u06DD' && c <= '\u06DF') || (c >= '\u06E0' && c <= '\u06E4')
1070 || (c >= '\u06E7' && c <= '\u06E8') || (c >= '\u06EA' && c <= '\u06ED') || (c >= '\u0901' && c <= '\u0903')
1072 || (c >= '\u093E' && c <= '\u094C')
1074 || (c >= '\u0951' && c <= '\u0954') || (c >= '\u0962' && c <= '\u0963') || (c >= '\u0981' && c <= '\u0983')
1078 || (c >= '\u09C0' && c <= '\u09C4') || (c >= '\u09C7' && c <= '\u09C8') || (c >= '\u09CB' && c <= '\u09CD')
1080 || (c >= '\u09E2' && c <= '\u09E3')
1085 || (c >= '\u0A40' && c <= '\u0A42') || (c >= '\u0A47' && c <= '\u0A48') || (c >= '\u0A4B' && c <= '\u0A4D')
1086 || (c >= '\u0A70' && c <= '\u0A71') || (c >= '\u0A81' && c <= '\u0A83')
1088 || (c >= '\u0ABE' && c <= '\u0AC5') || (c >= '\u0AC7' && c <= '\u0AC9') || (c >= '\u0ACB' && c <= '\u0ACD')
1089 || (c >= '\u0B01' && c <= '\u0B03')
1091 || (c >= '\u0B3E' && c <= '\u0B43') || (c >= '\u0B47' && c <= '\u0B48') || (c >= '\u0B4B' && c <= '\u0B4D')
1092 || (c >= '\u0B56' && c <= '\u0B57') || (c >= '\u0B82' && c <= '\u0B83') || (c >= '\u0BBE' && c <= '\u0BC2')
1093 || (c >= '\u0BC6' && c <= '\u0BC8') || (c >= '\u0BCA' && c <= '\u0BCD')
1095 || (c >= '\u0C01' && c <= '\u0C03') || (c >= '\u0C3E' && c <= '\u0C44') || (c >= '\u0C46' && c <= '\u0C48')
1096 || (c >= '\u0C4A' && c <= '\u0C4D') || (c >= '\u0C55' && c <= '\u0C56') || (c >= '\u0C82' && c <= '\u0C83')
1097 || (c >= '\u0CBE' && c <= '\u0CC4') || (c >= '\u0CC6' && c <= '\u0CC8') || (c >= '\u0CCA' && c <= '\u0CCD')
1098 || (c >= '\u0CD5' && c <= '\u0CD6') || (c >= '\u0D02' && c <= '\u0D03') || (c >= '\u0D3E' && c <= '\u0D43')
1099 || (c >= '\u0D46' && c <= '\u0D48') || (c >= '\u0D4A' && c <= '\u0D4D')
1102 || (c >= '\u0E34' && c <= '\u0E3A') || (c >= '\u0E47' && c <= '\u0E4E')
1104 || (c >= '\u0EB4' && c <= '\u0EB9') || (c >= '\u0EBB' && c <= '\u0EBC') || (c >= '\u0EC8' && c <= '\u0ECD')
1105 || (c >= '\u0F18' && c <= '\u0F19')
1111 || (c >= '\u0F71' && c <= '\u0F84') || (c >= '\u0F86' && c <= '\u0F8B') || (c >= '\u0F90' && c <= '\u0F95')
1113 || (c >= '\u0F99' && c <= '\u0FAD') || (c >= '\u0FB1' && c <= '\u0FB7')
1115 || (c >= '\u20D0' && c <= '\u20DC')
1117 || (c >= '\u302A' && c <= '\u302F')
1122 /** Digit - XMl Specification 1.0 [88] */
1123 private static final boolean Digit(char c) {
1124 return (c >= '\u0030' && c <= '\u0039') || (c >= '\u0660' && c <= '\u0669') || (c >= '\u06F0' && c <= '\u06F9')
1125 || (c >= '\u0966' && c <= '\u096F') || (c >= '\u09E6' && c <= '\u09EF') || (c >= '\u0A66' && c <= '\u0A6F')
1126 || (c >= '\u0AE6' && c <= '\u0AEF') || (c >= '\u0B66' && c <= '\u0B6F') || (c >= '\u0BE7' && c <= '\u0BEF')
1127 || (c >= '\u0C66' && c <= '\u0C6F') || (c >= '\u0CE6' && c <= '\u0CEF') || (c >= '\u0D66' && c <= '\u0D6F')
1128 || (c >= '\u0E50' && c <= '\u0E59') || (c >= '\u0ED0' && c <= '\u0ED9') || (c >= '\u0F20' && c <= '\u0F29');
1131 /** Extender - XMl Specification 1.0 [89] */
1132 private static final boolean Extender(char c) {
1133 return c == '\u00B7' || c == '\u02D0' || c == '\u02D1' || c == '\u0387'
1134 || c == '\u0640' || c == '\u0E46' || c == '\u0EC6' || c == '\u3005'
1135 || (c >= '\u3031' && c <= '\u3035') || (c >= '\u309D' && c <= '\u309E') || (c >= '\u30FC' && c <= '\u30FE');
1138 /** Whitespace - XML Specification 1.0 [3] */
1139 private static final boolean S(char c) {
1140 return c == '\u0020' || c == '\u0009' || c == '\r' || c == '\n';