1 // Copyright (C) 2003 Adam Megacz <adam@xwt.org> all rights reserved.
3 // You may modify, copy, and redistribute this code under the terms of
4 // the GNU Library Public License version 2.1, with the exception of
5 // the portion of clause 6a after the semicolon (aka the "obnoxious
10 import java.io.Reader;
11 import java.io.IOException;
12 import java.io.EOFException;
15 * An Event-Driving, Non-Validating XML Parser with Namespace support.
17 * A subclass can implement the abstract functions for receiving details
18 * about an xml file as it is parsed. To initate a parse, use the parse()
21 * <h3>Implementation Notes</h3>
22 * <p>As the parser traverses into an element, it adds it to the linked list
23 * called <tt>elements</tt>. However, <tt>elements</tt> has been pre-filled
24 * with instances of the Element inner class. So in the vast majority of
25 * cases, the pointer current is moved along one, and the values for the
26 * new element are filled into the current object.</p>
28 * <p>This parser supports all the unicode ranges required by the XML
29 * Specification. However, it is optimised for well-formed ASCII documents.
30 * Documents containing unicode Names and Attributes will take much longer
31 * to process, and invalid documents (badly formed Names or invalid attributes)
32 * will be run through a test on every single unicode character range before
33 * being declared invalid.</p>
36 * <li>Each time the buffer offset <tt>off</tt> is moved, the length
37 * <tt>len</tt> must be decreased.</li>
38 * <li>Each time the buffer length is decreased, it must be checked to make
39 * sure it is >0.</li>
40 * <li><i>error</i> is defined as a Validity Constraint Violation and
42 * <li><i>fatal error</i> is defined as a Well-formedness Constraint
43 * Violation and is not recoverable</li>
46 * @author David Crawshaw
47 * @see <a href="http://w3.org/TR/REC-xml">XML Specification</a>
48 * @see <a href="http://w3.org/TR/REC-xml-names">XML Namespaces</a>
50 public abstract class XML
52 /////////////////////////////////////////////////////////////////////////////////////////////
54 /////////////////////////////////////////////////////////////////////////////////////////////
56 public static final int BUFFER_SIZE = 255;
58 /** static pool of XML.Element instances shared by all XML Parsers. */
59 private static final Queue elements = new Queue(30);
61 private static final char[] single_amp = new char[] { '&' };
62 private static final char[] single_apos = new char[] { '\'' };
63 private static final char[] single_gt = new char[] { '>' };
64 private static final char[] single_lt = new char[] { '<' };
65 private static final char[] single_quot = new char[] { '"' };
75 private Element current;
77 // used in readEntity() to process a single character without creating a new array
78 private char[] singlechar = new char[1];
81 public XML() { this(BUFFER_SIZE); }
83 public XML(int bSize) {
84 buf = new char[bSize];
86 current = (Element)elements.remove(false);
87 if (current == null) current = new Element();
91 /** Returns the line number at the beginning of the last process call. */
92 public int getLine() { return line; }
94 /** Returns the column number at the beginning of the last process call. */
95 public int getCol() { return col; }
98 * Parse given input and call the abstract event functions.
100 * Careful with threading, as this function is not synchronized.
102 public final void parse(Reader reader) throws IOException, Exn {
107 clear(); // clean up possible mid-way linked-list element
110 // process the stream
113 if (current.qName == null) break;
114 throw new Exn("reached eof without closing <"+current.qName+"> element", Exn.WFC, getLine(), getCol());
117 if (buf[off] == '<') readTag();
118 readChars(current.qName != null);
120 } finally { clear(); } // clean up elements
123 /** remove any leftover elements from the linked list and queue them */
124 private final void clear() {
125 for (Element last = current; current.parent != null; ) {
126 current = current.parent;
128 elements.append(last);
133 /** reads in a tag. expects <tt>buf[off] == '<'</tt> */
134 private final void readTag() throws IOException, Exn {
135 // Start Tag '<' Name (S Attribute)* S? '>'
136 boolean starttag = true;
138 // End Tag '</' Name S? '>'
139 boolean endtag = false;
141 // if (starttag & endtag) then: EmptyElemTag '<' Name (S Attribute)* S? '/>'
143 // Position in the name of the ':' namespace prefix
149 if (!buffer(1)) throw new EOFException("Unexpected EOF processing element tag");
151 // work out what we can from the beginning of the tag
154 // definitions here don't necessarily conform to xml spec (as DTDs not yet implemented)
156 if (!buffer(4)) throw new EOFException("Unexpected EOF processing <! element");
161 if (buf[off+1] != '-') { bad = true; break; }
162 col += 2; off += 2; len -= 2;
164 // Comment '<!--' ((Char - '-') | ('-' (Char - '-')))* '-->'
165 readChars(false, "-->", false);
166 col += 3; off += 3; len -= 3;
169 // we don't care about the following definitions
173 || buf[off+1] != 'T' || buf[off+2] != 'T' || buf[off+3] != 'L'
174 || buf[off+4] != 'I' || buf[off+5] != 'S' || buf[off+6] != 'T') {
177 col += 7; off += 7; len -= 7;
179 // ATTLIST '<!ATTLIST' (Char* - '>') '>'
180 readChars(false, ">", true);
185 || buf[off+1] != 'O' || buf[off+2] != 'C' || buf[off+3] != 'T'
186 || buf[off+4] != 'Y' || buf[off+5] != 'P' || buf[off+6] != 'E') {
189 col += 7; off += 7; len -= 7;
191 // DTD '<!DOCTYPE' (Char* - '>') '>'
192 readChars(false, ">", true);
198 } else if (buf[off+1] == 'L' && buf[off+2] == 'E' && buf[off+3] == 'M'
199 && buf[off+4] == 'E' && buf[off+5] == 'N' && buf[off+6] == 'T') {
200 // ELEMENT '<!ELEMENT' (Char* - '>') '>'
201 readChars(false, ">", true);
204 } else if (buf[off+1] == 'N' && buf[off+2] == 'T' && buf[off+3] == 'I'
205 && buf[off+4] == 'T' && buf[off+5] == 'Y') {
206 // ENTITY '<!ENTITY' (Char* - '>') '>'
207 readChars(false, ">", true);
217 || buf[off+1] != 'O' || buf[off+2] != 'T' || buf[off+3] != 'A' || buf[off+4] != 'T'
218 || buf[off+5] != 'I' || buf[off+6] != 'O' || buf[off+7] != 'N') {
221 col += 8; off += 8; len -= 8;
222 // NOTATION '<!NOTATION' (Char* - '>') '>'
223 readChars(false, ">", true);
230 if (bad) throw new Exn("element tag start character is invalid", Exn.MARKUP, getLine(), getCol());
232 } else if (s == '?') {
233 // PI (Ignored) '<?' (Char* - (Char* '?>' Char*)) '?>'
235 readChars(false, "?>", true);
236 if (!buffer(2)) throw new EOFException("Unexpected EOF at end of Processing Instruction");
237 col += 2; off += 2; len -= 2;
239 } else if (s == '[') {
241 || buf[off+1] != 'C' || buf[off+2] != 'D' || buf[off+3] != 'A'
242 || buf[off+4] != 'T' || buf[off+5] != 'A' || buf[off+6] != '[') {
244 // Conditional '<![' (Char* - (Char* ']]>' Char*)) ']]>'
245 readChars(false, "]]>", false);
247 col += 7; off += 7; len -=7;
248 // CDATA '<![CDATA[' (Char* - (Char* ']]>' Char*)) ']]>'
249 readChars(true, "]]>", false);
251 col += 3; off += 3; len -= 3;
254 // End Tag '</' Name S? '>'
259 if (!buffer(1)) throw new EOFException("Unexpected EOF processing end tag");
263 if (!Name(s)) throw new Exn("invalid starting character in element name", Exn.MARKUP, getLine(), getCol());
265 // find the element name (defined in XML Spec: section 2.3)
266 for (namelen = 0; ; namelen++) {
267 if (!buffer(namelen+1)) throw new EOFException("Unexpected EOF in element tag name");
269 s = buf[off+namelen];
271 if (S(s) || s == '>') {
273 } else if (s == '/') {
276 } else if (s == ':' && namelen > 0 && prefix < 1) {
277 // we have a definition of the prefix range available
279 } else if (!NameChar(s)) {
280 throw new Exn("element name contains invalid character", Exn.MARKUP, getLine(), getCol());
284 // process name (based on calculated region)
285 if (namelen < 1) throw new Exn("element name is null", Exn.MARKUP, getLine(), getCol());
287 // we have marked out the name region, so turn it into a string and move on
288 String qName = new String(buf, off, namelen);
290 col += namelen; off += namelen; len -= namelen;
293 // create the in-memory element representation of this beast
294 // if current.qName == null then this is the root element we're dealing with
295 if (current.qName != null) {
296 Element next = (Element)elements.remove(false);
297 if (next == null) next = new Element();
298 //next.clear(); // TODO: remove as elements now checked as they're added to the queue
299 next.parent = current;
303 current.qName = qName;
306 current.prefix = current.qName.substring(0, prefix);
307 current.localName = current.qName.substring(prefix+1);
309 current.prefix = null;
310 current.localName = current.qName;
313 // process attributes
315 if (!buffer(1)) throw new EOFException("Unexpected EOF - processing attributes part 1");
316 while (buf[off] != '/' && buf[off] != '>') {
318 if (!buffer(1)) throw new EOFException("Unexpected EOF - processing attributes part 2");
322 // work out the uri of this element
323 current.uri = current.getUri(current.getPrefix());
324 if (current.getUri().equals("") && current.getPrefix() != null)
325 current.addError(new Exn("undefined prefix '"+current.getPrefix()+"'", Exn.NC, getLine(), getCol()));
328 // this is an end-of-element tag
329 if (!qName.equals(current.getQName())) throw new Exn(
330 "end tag </"+qName+"> does not line up with start tag <"+current.getQName()+">", Exn.WFC, getLine(), getCol()
334 // deal with whitespace
338 if (!buffer(1)) throw new EOFException("Unexpected EOF before end of tag");
339 if (buf[off] == '/') {
343 if (!buffer(1)) throw new EOFException("Unexpected EOF before end of endtag");
344 if (buf[off] == '>') {
347 throw new Exn("missing '>' character from element '"+qName+"'", Exn.MARKUP, getLine(), getCol());
350 // send element signals
351 if (starttag) startElement(current);
355 // we just closed an element, so remove it from the element 'stack'
356 if (current.getParent() == null) {
357 // we just finished the root element
360 Element last = current;
361 current = current.parent;
363 elements.append(last);
369 /** reads in an attribute of an element. expects Name(buf[off]) */
370 private final void readAttribute() throws IOException, Exn {
373 String n, v, p, u; // attribute name, value, prefix and uri respectively
374 n = v = p = u = null;
377 // find the element name (defined in XML Spec: section 2.3)
378 for (ref= 0; ; ref++) {
379 if (!buffer(ref+1)) throw new EOFException("Unexpected EOF in read attribute loop part 1");
383 if (s == '=' || S(s)) {
385 } else if (s == ':' && ref > 0 && prefix < 1) {
386 // we have a definition of the prefix range available
388 } else if (!NameChar(s)) {
389 throw new Exn("attribute name contains invalid characters", Exn.MARKUP, getLine(), getCol());
393 // determine prefix and key name
395 p = new String(buf, off, prefix-1);
396 col += prefix; off += prefix; len -= prefix; ref -= prefix;
398 n = new String(buf, off, ref);
399 col += ref; off += ref; len -= ref;
401 // find name/value divider ('=')
403 if (!buffer(1)) throw new EOFException("Unexpected EOF before attribute '=' divider");
404 if (buf[off] != '=') throw new Exn("attribute name not followed by '=' sign", Exn.MARKUP, getLine(), getCol());
409 if (!buffer(1)) throw new EOFException("Unexpected EOF after attribute '=' divider");
412 if (buf[off] == '\'' || buf[off] == '"') {
415 throw new Exn("attribute '"+n+"' must have attribute wrapped in ' or \"", Exn.MARKUP, getLine(), getCol());
419 // find the attribute value
420 attval: for (ref = 0; ; ref++) {
421 if (!buffer(ref+1)) throw new EOFException("Unexpected EOF in attribute value");
423 if (buf[off+ref] == wrap) {
425 } else if (buf[off+ref] == '<') {
426 throw new Exn("attribute value for '"+n+"' must not contain '<'", Exn.WFC, getLine(), getCol());
430 v = new String(buf, off, ref);
431 col += ref; off += ref; len -= ref;
433 // remove end wrapper character
437 if (p != null && p.equals("xmlns")) {
438 current.addUri(n, v);
439 } else if (n.equals("xmlns")) {
440 if (current.getUri().equals("")) {
441 current.addUri("", v);
443 current.addError(new Exn("default namespace definition repeated", Exn.NC, getLine(), getCol()));
446 // find attribute uri
447 u = current.getUri(p);
448 if (p != null && u.equals("")) current.addError(new Exn("undefined attribute prefix '"+p+"'", Exn.NC, getLine(), getCol()));
450 // check to see if attribute is a repeat
451 for (int i=0; current.len > i; i++) if (n.equals(current.getAttrKey(i)) && u.equals(current.getAttrUri(i))) throw new Exn(
452 "attribute name '"+n+"' may not appear more than once in the same element tag", Exn.WFC, getLine(), getCol()
455 current.addAttr(n, v, u);
459 /** reads an entity and processes out its value. expects buf[off] == '&' */
460 private final void readEntity() throws IOException, Exn {
462 if (!buffer(2)) throw new EOFException("Unexpected EOF reading entity");
464 boolean unknown = false;
470 if (buf[off] == 'x') { off++; len--; radix = 16; } else { radix = 10; }
473 // read in each char, then shift total value to the left and add the extra
474 // style of loop is slightly different from all the others, as this should run a limited number of times
475 findchar: while (true) {
476 if (!buffer(1)) throw new EOFException("Unexpected EOF reading entity");
477 int d = Character.digit(buf[off], radix);
479 if (buf[off] != ';') throw new Exn("illegal characters in entity reference", Exn.WFC, getLine(), getCol());
488 singlechar[0] = Character.forDigit(c, radix);
489 characters(singlechar, 0, 1);
493 if (buffer(4) && buf[off+1] == 'm' && buf[off+2] == 'p' && buf[off+3] == ';') {
494 characters(single_amp, 0, 1); // &
495 off += 4; len -= 4; col++;
496 } else if (buffer(5) && buf[off+1] == 'p' && buf[off+2] == 'o' && buf[off+3] == 's' && buf[off+4] == ';') {
497 characters(single_apos, 0, 1); // '
498 off += 5; len -= 5; col++;
505 if (buffer(3) && buf[off+1] == 't' && buf[off+2] == ';') {
506 characters(single_gt, 0, 1); // >
507 off += 3; len -= 3; col++;
514 if (buffer(3) && buf[off+1] == 't' && buf[off+2] == ';') {
515 characters(single_lt, 0, 1); // <
516 off += 3; len -= 3; col++;
523 if (buffer(5) && buf[off+1] == 'u' && buf[off+2] == 'o' && buf[off+3] == 't' && buf[off+4] == ';') {
524 characters(single_quot, 0, 1); // "
525 off += 5; len -= 5; col++;
531 // TODO: check a parser-level Hash of defined entities
534 if (unknown) throw new Exn("unknown entity (<!ENTITY> not supported)", Exn.WFC, getLine(), getCol());
537 /** reads until the passed string is encountered. */
538 private final void readChars(boolean p, String match, boolean entities) throws IOException, Exn {
540 char[] end = match.toCharArray();
542 for (boolean more = true; more;) {
543 if (!buffer(1)) return;
545 buf: for (ref = 0; ref < len; ref++) {
546 switch (buf[off+ref]) {
547 case '\r': // windows or macos9 newline
548 // normalise and process
549 buf[off+ref] = '\n'; ref++;
550 if (p) characters(buf, off, ref);
551 off += ref; len -= ref; ref = -1;
554 // windows double-char newline; skip the next char
555 if (!buffer(1)) return;
556 if (buf[off] == '\n') { off++; len--; }
559 case '\n': // unix newline
561 if (p) characters(buf, off, ref);
562 off += ref; len -= ref; ref = -1;
569 if (ref > 0) characters(buf, off, ref);
570 off += ref; len -= ref; ref = -1;
577 if (!buffer(ref+end.length)) continue buf;
578 for (int i=0; end.length > i; i++) if (end[i] != buf[off+ref+i]) continue buf;
584 if (p && ref > 0) characters(buf, off, ref);
585 off += ref; len -= ref; col += ref;
590 * reads until a <tt><</tt> symbol is encountered
591 * @param p If true call the characters(char[],int,int) funciton for the processed characters
593 private final void readChars(boolean p) throws IOException, Exn {
596 for (boolean more = true; more;) {
597 if (!buffer(1)) return;
599 buf: for (ref = 0; ref < len; ref++) {
600 switch (buf[off+ref]) {
601 case '\r': // windows or macos9 newline
602 // normalise and process
603 buf[off+ref] = '\n'; ref++;
604 if (p) characters(buf, off, ref);
605 off += ref; len -= ref; ref = -1;
608 // windows double-char newline; skip the next char
609 if (!buffer(1)) return;
610 if (buf[off] == '\n') { off++; len--; }
613 case '\n': // unix newline
615 if (p) characters(buf, off, ref);
616 off += ref; len -= ref; ref = -1;
622 if (ref > 0) characters(buf, off, ref);
623 off += ref; len -= ref; ref = -1;
628 case '<': // end of chars section
634 if (p && ref > 0) characters(buf, off, ref);
635 off += ref; len -= ref; col += ref;
639 /** reads until a non-whitespace symbol is encountered */
640 private final void readWhitespace() throws IOException, Exn {
643 for (boolean more = true; more;) {
644 if (!buffer(1)) return;
646 buf: for (ref = 0; ref < len; ref++) {
647 switch (buf[off+ref]) {
648 case '\r': // windows or macos9 newline
649 // normalise and process
651 whitespace(buf, off, ++ref);
652 off += ref; len -= ref; ref = -1;
655 // windows double-char newline; skip the next char
656 if (!buffer(1)) return;
657 if (buf[off] == '\n') { off++; len--; }
660 case '\n': // unix newline
661 whitespace(buf, off, ++ref);
662 off += ref; len -= ref; ref = -1;
670 default: // end of whitespace
676 off += ref; len -= ref; col += ref;
681 * attempt to fill the buffer.
683 * @param min Minimum number of characters to read (even if we have to block to do it).
684 * @return return false if min can't be reached.
686 private final boolean buffer(int min) throws IOException {
687 if (len > min) return true;
689 if (buf.length - (off+len) >= min) {
690 // plenty of space left on the end of the buffer
691 } else if (off >= min) {
692 // moving offset data to start will leave enough free space on the end
693 System.arraycopy(buf, off, buf, 0, len);
696 // buffer size will have to be increased
697 char[] newbuf = new char[buf.length * 2];
698 System.arraycopy(buf, off, newbuf, 0, len);
704 int newlen = in.read(buf, off+len, buf.length-(off+len));
705 if (newlen < 0) return false;
713 /////////////////////////////////////////////////////////////////////////////////////////////
714 // Abstract SAX-Like Interface
715 /////////////////////////////////////////////////////////////////////////////////////////////
718 * Called when the start of an element is processed.
720 * <p><b>DO NOT</b> store a reference to the Element object, as
721 * they are reused by XML Parser.</p>
723 public abstract void startElement(Element e) throws Exn;
726 * Represents up to a line of character data.
728 * <p>Newlines are all normalised to the Unix \n as per the XML Spec,
729 * and a newline will only appear as the last character in the passed
732 * <p>XML.getLine() and XML.getCol() report the position at the
733 * beginning of this character segment, which can be processed in a
734 * line-by-line fashion due to the above newline restriction.</p>
736 public abstract void characters(char[] ch, int start, int length) throws Exn, IOException;
738 /** Represents up to a line of ignorable whitespace. */
739 public abstract void whitespace(char[] ch, int start, int length) throws Exn, IOException;
741 /** Represents the end of an Element. */
742 public abstract void endElement(Element e) throws Exn, IOException;
745 /////////////////////////////////////////////////////////////////////////////////////////////
746 // Inner Classes for Parser Support
747 /////////////////////////////////////////////////////////////////////////////////////////////
750 * Represents an element in an XML document. Stores a reference to its
751 * parent, forming a one-way linked list.
753 * Element objects are reused, so client code making use of them must
754 * drop their references after the specific element process function
757 public static final class Element {
759 private static final int DEFAULT_ATTR_SIZE = 10;
761 protected Element parent = null;
763 protected String uri = null;
764 protected String localName = null;
765 protected String qName = null;
766 protected String prefix = null;
768 protected Hash urimap = new Hash(3,3);
770 protected String[] keys = new String[DEFAULT_ATTR_SIZE];
771 protected String[] vals = new String[DEFAULT_ATTR_SIZE];
772 protected String[] uris = new String[DEFAULT_ATTR_SIZE];
773 protected int len = 0;
775 protected Exn[] errors = new Exn[] {};
778 /** Parent of current element. */
779 public Element getParent() { return parent; }
781 /** Qualified Name of current element. XML Namespace Spec 14-Jan-1999 [6] */
782 public String getQName() { return qName; }
784 /** LocalPart of current element. XML Namespace Spec 14-Jan-1999 [8] */
785 public String getLocalName() { return localName; }
787 /** Prefix of current element. Substring of qName. XML Namespace Spec 14-Jan-1999 [7] */
788 public String getPrefix() { return prefix; }
790 /** URI of current tag. XML Namespace Spec 14-Jan-1999 section 1 */
791 public String getUri() { return getUri(prefix); }
793 /** URI of a given prefix. Never returns null, instead gives "". */
794 public String getUri(String p) {
796 for (Element e = this; e != null && ret == null; e = e.getParent()) {
797 ret = (String)e.urimap.get(p == null ? "" : p);
799 return ret == null ? "" : ret;
802 /** An array of attribute names. */
803 public String getAttrKey(int pos) { return len > pos ? keys[pos] : null; }
805 /** An array of attribute values. */
806 public String getAttrVal(int pos) { return len > pos ? vals[pos] : null; }
808 /** An array of attribute uris. */
809 public String getAttrUri(int pos) { return len > pos ? uris[pos] : null; }
811 /** Current number of attributes in the element. */
812 public int getAttrLen() { return len; }
814 /** An array of non-fatal errors related to this element. */
815 public Exn[] getErrors() { return errors; }
818 protected Element() { }
820 /** Add (replace if exists in current element) a Namespace prefix/uri map. */
821 protected void addUri(String name, String value) {
822 urimap.put(name, value);
825 /** Add an attribute. */
826 protected void addAttr(String key, String val, String uri) {
827 if (len == keys.length) {
828 // increase the size of the attributes arrays
829 String[] newkeys = new String[keys.length*2];
830 String[] newvals = new String[vals.length*2];
831 String[] newuris = new String[uris.length*2];
832 System.arraycopy(keys, 0, newkeys, 0, keys.length);
833 System.arraycopy(vals, 0, newvals, 0, vals.length);
834 System.arraycopy(uris, 0, newuris, 0, uris.length);
835 keys = newkeys; vals = newvals; uris = newuris;
845 protected void addError(Exn e) {
846 // it doesn't really matter about continually expanding the array, as this case is quite rare
847 Exn[] newe = new Exn[errors.length+1];
848 System.arraycopy(errors, 0, newe, 0, errors.length);
849 newe[errors.length] = e;
853 /** Empty out all the data from the Element. */
854 protected void clear() {
856 uri = localName = qName = prefix = null;
859 if (keys.length != vals.length || vals.length != uris.length) {
860 keys = new String[DEFAULT_ATTR_SIZE];
861 vals = new String[DEFAULT_ATTR_SIZE];
862 uris = new String[DEFAULT_ATTR_SIZE];
864 for (int i=0; keys.length > i; i++) { keys[i] = null; vals[i] = null; uris[i] = null; };
868 errors = new Exn[] {};
872 /** Parse or Structural Error */
873 public static class Exn extends Exception {
874 /** Violation of Markup restrictions in XML Specification - Fatal Error */
875 public static final int MARKUP = 1;
877 /** Well-Formedness Constraint Violation - Fatal Error */
878 public static final int WFC = 2;
880 /** Namespace Constraint Violation - Recoverable Error */
881 public static final int NC = 3;
883 /** Schema Violation - Fatal Error */
884 public static final int SCHEMA = 4;
886 private String error;
891 public Exn(String e) { this(e, MARKUP, -1, -1); }
893 public Exn(String e, int type, int line, int col) {
900 public int getType() { return this.type; }
901 public int getLine() { return this.line; }
902 public int getCol() { return this.col; }
903 public String getMessage() { return this.error + (line >= 0 && col >= 0 ? " at " + line + ":" + col: ""); }
907 /////////////////////////////////////////////////////////////////////////////////////////////
908 // Static Support Functions for the XML Specification
909 /////////////////////////////////////////////////////////////////////////////////////////////
911 // attempt to avoid these functions unless you *expect* the input to fall in the given range.
913 /** First Character of Name - XML Specification 1.0 [5] */
914 private static final boolean Name(char c) {
915 return BaseCharAscii(c) || c == '_' || c == ':' || Letter(c);
918 /** NameChar - XML Specification 1.0 [4] */
919 private static final boolean NameChar(char c) {
920 return BaseCharAscii(c) || c == '.' || c == '-' || c == '_' || c == ':'
921 || Digit(c) || Letter(c) || Extender(c); // TODO: || CombiningChar(c);
924 /** BaseChar - XMl Specification 1.0 [84] */
925 private static final boolean Letter(char c) {
926 return BaseChar(c) || Ideographic(c);
929 /** Elements of BaseChar that exist in ASCII. */
930 private static final boolean BaseCharAscii(char c) {
931 return (c >= '\u0041' && c <= '\u005A') || (c >= '\u0061' && c <= '\u007A');
934 /** Char - XML Specification 1.0 [2] */
935 private static final boolean Char(char c) {
936 // u000A == r and u000D == n, but the javac compiler can't handle the \ u form
937 return c == '\u0009' || c == '\r' || c == '\n'
938 || (c >= '\u0020' && c <= '\uD7FF')
939 || (c >= '\uE000' && c <= '\uFFFD');
942 /** BaseChar - XML Specification 1.0 [85] */
943 private static final boolean BaseChar(char c) {
944 return BaseCharAscii(c) || (c >= '\u00C0' && c <= '\u00D6')
945 || (c >= '\u00D8' && c <= '\u00F6') || (c >= '\u00F8' && c <= '\u00FF') || (c >= '\u0100' && c <= '\u0131')
946 || (c >= '\u0134' && c <= '\u013E') || (c >= '\u0141' && c <= '\u0148') || (c >= '\u014A' && c <= '\u017E')
947 || (c >= '\u0180' && c <= '\u01C3') || (c >= '\u01CD' && c <= '\u01F0') || (c >= '\u01F4' && c <= '\u01F5')
948 || (c >= '\u01FA' && c <= '\u0217') || (c >= '\u0250' && c <= '\u02A8') || (c >= '\u02BB' && c <= '\u02C1')
949 || (c == '\u0386') || (c >= '\u0388' && c <= '\u038A') || (c == '\u038C')
950 || (c >= '\u038E' && c <= '\u03A1') || (c >= '\u03A3' && c <= '\u03CE') || (c >= '\u03D0' && c <= '\u03D6')
951 || (c == '\u03DA') || (c == '\u03DC') || (c == '\u03DE')
953 || (c >= '\u03E2' && c <= '\u03F3') || (c >= '\u0401' && c <= '\u040C') || (c >= '\u040E' && c <= '\u044F')
954 || (c >= '\u0451' && c <= '\u045C') || (c >= '\u045E' && c <= '\u0481') || (c >= '\u0490' && c <= '\u04C4')
955 || (c >= '\u04C7' && c <= '\u04C8') || (c >= '\u04CB' && c <= '\u04CC') || (c >= '\u04D0' && c <= '\u04EB')
956 || (c >= '\u04EE' && c <= '\u04F5') || (c >= '\u04F8' && c <= '\u04F9') || (c >= '\u0531' && c <= '\u0556')
958 || (c >= '\u0561' && c <= '\u0586') || (c >= '\u05D0' && c <= '\u05EA') || (c >= '\u05F0' && c <= '\u05F2')
959 || (c >= '\u0621' && c <= '\u063A') || (c >= '\u0641' && c <= '\u064A') || (c >= '\u0671' && c <= '\u06B7')
960 || (c >= '\u06BA' && c <= '\u06BE') || (c >= '\u06C0' && c <= '\u06CE') || (c >= '\u06D0' && c <= '\u06D3')
962 || (c >= '\u06E5' && c <= '\u06E6') || (c >= '\u0905' && c <= '\u0939')
964 || (c >= '\u0958' && c <= '\u0961') || (c >= '\u0985' && c <= '\u098C') || (c >= '\u098F' && c <= '\u0990')
965 || (c >= '\u0993' && c <= '\u09A8') || (c >= '\u09AA' && c <= '\u09B0')
967 || (c >= '\u09B6' && c <= '\u09B9') || (c >= '\u09DF' && c <= '\u09E1') || (c >= '\u09F0' && c <= '\u09F1')
968 || (c >= '\u0A05' && c <= '\u0A0A') || (c >= '\u0A0F' && c <= '\u0A10') || (c >= '\u0A13' && c <= '\u0A28')
969 || (c >= '\u0A2A' && c <= '\u0A30') || (c >= '\u0A32' && c <= '\u0A33') || (c >= '\u0A35' && c <= '\u0A36')
970 || (c >= '\u0A38' && c <= '\u0A39') || (c >= '\u0A59' && c <= '\u0A5C')
972 || (c >= '\u0A72' && c <= '\u0A74') || (c >= '\u0A85' && c <= '\u0A8B')
974 || (c >= '\u0A8F' && c <= '\u0A91') || (c >= '\u0A93' && c <= '\u0AA8') || (c >= '\u0AAA' && c <= '\u0AB0')
975 || (c >= '\u0AB2' && c <= '\u0AB3') || (c >= '\u0AB5' && c <= '\u0AB9')
978 || (c >= '\u0B05' && c <= '\u0B0C') || (c >= '\u0B0F' && c <= '\u0B10') || (c >= '\u0B13' && c <= '\u0B28')
979 || (c >= '\u0B2A' && c <= '\u0B30') || (c >= '\u0B32' && c <= '\u0B33') || (c >= '\u0B36' && c <= '\u0B39')
981 || (c >= '\u0B5C' && c <= '\u0B5D') || (c >= '\u0B5F' && c <= '\u0B61') || (c >= '\u0B85' && c <= '\u0B8A')
982 || (c >= '\u0B8E' && c <= '\u0B90') || (c >= '\u0B92' && c <= '\u0B95') || (c >= '\u0B99' && c <= '\u0B9A')
984 || (c >= '\u0B9E' && c <= '\u0B9F') || (c >= '\u0BA3' && c <= '\u0BA4') || (c >= '\u0BA8' && c <= '\u0BAA')
985 || (c >= '\u0BAE' && c <= '\u0BB5') || (c >= '\u0BB7' && c <= '\u0BB9') || (c >= '\u0C05' && c <= '\u0C0C')
986 || (c >= '\u0C0E' && c <= '\u0C10') || (c >= '\u0C12' && c <= '\u0C28') || (c >= '\u0C2A' && c <= '\u0C33')
987 || (c >= '\u0C35' && c <= '\u0C39') || (c >= '\u0C60' && c <= '\u0C61') || (c >= '\u0C85' && c <= '\u0C8C')
988 || (c >= '\u0C8E' && c <= '\u0C90') || (c >= '\u0C92' && c <= '\u0CA8') || (c >= '\u0CAA' && c <= '\u0CB3')
989 || (c >= '\u0CB5' && c <= '\u0CB9')
991 || (c >= '\u0CE0' && c <= '\u0CE1') || (c >= '\u0D05' && c <= '\u0D0C') || (c >= '\u0D0E' && c <= '\u0D10')
992 || (c >= '\u0D12' && c <= '\u0D28') || (c >= '\u0D2A' && c <= '\u0D39') || (c >= '\u0D60' && c <= '\u0D61')
993 || (c >= '\u0E01' && c <= '\u0E2E')
995 || (c >= '\u0E32' && c <= '\u0E33') || (c >= '\u0E40' && c <= '\u0E45') || (c >= '\u0E81' && c <= '\u0E82')
997 || (c >= '\u0E87' && c <= '\u0E88')
1000 || (c >= '\u0E94' && c <= '\u0E97') || (c >= '\u0E99' && c <= '\u0E9F') || (c >= '\u0EA1' && c <= '\u0EA3')
1003 || (c >= '\u0EAA' && c <= '\u0EAB') || (c >= '\u0EAD' && c <= '\u0EAE')
1005 || (c >= '\u0EB2' && c <= '\u0EB3')
1007 || (c >= '\u0EC0' && c <= '\u0EC4') || (c >= '\u0F40' && c <= '\u0F47') || (c >= '\u0F49' && c <= '\u0F69')
1008 || (c >= '\u10A0' && c <= '\u10C5') || (c >= '\u10D0' && c <= '\u10F6')
1010 || (c >= '\u1102' && c <= '\u1103') || (c >= '\u1105' && c <= '\u1107')
1012 || (c >= '\u110B' && c <= '\u110C') || (c >= '\u110E' && c <= '\u1112')
1019 || (c >= '\u1154' && c <= '\u1155')
1021 || (c >= '\u115F' && c <= '\u1161')
1026 || (c >= '\u116D' && c <= '\u116E') || (c >= '\u1172' && c <= '\u1173')
1031 || (c >= '\u11AE' && c <= '\u11AF') || (c >= '\u11B7' && c <= '\u11B8')
1033 || (c >= '\u11BC' && c <= '\u11C2')
1037 || (c >= '\u1E00' && c <= '\u1E9B') || (c >= '\u1EA0' && c <= '\u1EF9') || (c >= '\u1F00' && c <= '\u1F15')
1038 || (c >= '\u1F18' && c <= '\u1F1D') || (c >= '\u1F20' && c <= '\u1F45') || (c >= '\u1F48' && c <= '\u1F4D')
1039 || (c >= '\u1F50' && c <= '\u1F57')
1043 || (c >= '\u1F5F' && c <= '\u1F7D') || (c >= '\u1F80' && c <= '\u1FB4') || (c >= '\u1FB6' && c <= '\u1FBC')
1045 || (c >= '\u1FC2' && c <= '\u1FC4') || (c >= '\u1FC6' && c <= '\u1FCC') || (c >= '\u1FD0' && c <= '\u1FD3')
1046 || (c >= '\u1FD6' && c <= '\u1FDB') || (c >= '\u1FE0' && c <= '\u1FEC') || (c >= '\u1FF2' && c <= '\u1FF4')
1047 || (c >= '\u1FF6' && c <= '\u1FFC')
1049 || (c >= '\u212A' && c <= '\u212B')
1051 || (c >= '\u2180' && c <= '\u2182') || (c >= '\u3041' && c <= '\u3094') || (c >= '\u30A1' && c <= '\u30FA')
1052 || (c >= '\u3105' && c <= '\u312C') || (c >= '\uAC00' && c <= '\uD7A3');
1055 /** BaseChar - XMl Specification 1.0 [86] */
1056 private static final boolean Ideographic(char c) {
1057 return (c >= '\u4E00' && c <= '\u9FA5') || c == '\u3007' || (c >= '\u3021' && c <= '\u3029');
1060 /** CombiningChar - XMl Specification 1.0 [87] */
1061 /*private static final boolean CombiningChar(char c) {
1062 return (c >= '\u0300' && c <= '\u0345')
1063 || (c >= '\u0360' && c <= '\u0361') || (c >= '\u0483' && c <= '\u0486') || (c >= '\u0591' && c <= '\u05A1')
1064 || (c >= '\u05A3' && c <= '\u05B9') || (c >= '\u05BB' && c <= '\u05BD')
1066 || (c >= '\u05C1' && c <= '\u05C2')
1068 || (c >= '\u064B' && c <= '\u0652')
1070 || (c >= '\u06D6' && c <= '\u06DC') || (c >= '\u06DD' && c <= '\u06DF') || (c >= '\u06E0' && c <= '\u06E4')
1071 || (c >= '\u06E7' && c <= '\u06E8') || (c >= '\u06EA' && c <= '\u06ED') || (c >= '\u0901' && c <= '\u0903')
1073 || (c >= '\u093E' && c <= '\u094C')
1075 || (c >= '\u0951' && c <= '\u0954') || (c >= '\u0962' && c <= '\u0963') || (c >= '\u0981' && c <= '\u0983')
1079 || (c >= '\u09C0' && c <= '\u09C4') || (c >= '\u09C7' && c <= '\u09C8') || (c >= '\u09CB' && c <= '\u09CD')
1081 || (c >= '\u09E2' && c <= '\u09E3')
1086 || (c >= '\u0A40' && c <= '\u0A42') || (c >= '\u0A47' && c <= '\u0A48') || (c >= '\u0A4B' && c <= '\u0A4D')
1087 || (c >= '\u0A70' && c <= '\u0A71') || (c >= '\u0A81' && c <= '\u0A83')
1089 || (c >= '\u0ABE' && c <= '\u0AC5') || (c >= '\u0AC7' && c <= '\u0AC9') || (c >= '\u0ACB' && c <= '\u0ACD')
1090 || (c >= '\u0B01' && c <= '\u0B03')
1092 || (c >= '\u0B3E' && c <= '\u0B43') || (c >= '\u0B47' && c <= '\u0B48') || (c >= '\u0B4B' && c <= '\u0B4D')
1093 || (c >= '\u0B56' && c <= '\u0B57') || (c >= '\u0B82' && c <= '\u0B83') || (c >= '\u0BBE' && c <= '\u0BC2')
1094 || (c >= '\u0BC6' && c <= '\u0BC8') || (c >= '\u0BCA' && c <= '\u0BCD')
1096 || (c >= '\u0C01' && c <= '\u0C03') || (c >= '\u0C3E' && c <= '\u0C44') || (c >= '\u0C46' && c <= '\u0C48')
1097 || (c >= '\u0C4A' && c <= '\u0C4D') || (c >= '\u0C55' && c <= '\u0C56') || (c >= '\u0C82' && c <= '\u0C83')
1098 || (c >= '\u0CBE' && c <= '\u0CC4') || (c >= '\u0CC6' && c <= '\u0CC8') || (c >= '\u0CCA' && c <= '\u0CCD')
1099 || (c >= '\u0CD5' && c <= '\u0CD6') || (c >= '\u0D02' && c <= '\u0D03') || (c >= '\u0D3E' && c <= '\u0D43')
1100 || (c >= '\u0D46' && c <= '\u0D48') || (c >= '\u0D4A' && c <= '\u0D4D')
1103 || (c >= '\u0E34' && c <= '\u0E3A') || (c >= '\u0E47' && c <= '\u0E4E')
1105 || (c >= '\u0EB4' && c <= '\u0EB9') || (c >= '\u0EBB' && c <= '\u0EBC') || (c >= '\u0EC8' && c <= '\u0ECD')
1106 || (c >= '\u0F18' && c <= '\u0F19')
1112 || (c >= '\u0F71' && c <= '\u0F84') || (c >= '\u0F86' && c <= '\u0F8B') || (c >= '\u0F90' && c <= '\u0F95')
1114 || (c >= '\u0F99' && c <= '\u0FAD') || (c >= '\u0FB1' && c <= '\u0FB7')
1116 || (c >= '\u20D0' && c <= '\u20DC')
1118 || (c >= '\u302A' && c <= '\u302F')
1123 /** Digit - XMl Specification 1.0 [88] */
1124 private static final boolean Digit(char c) {
1125 return (c >= '\u0030' && c <= '\u0039') || (c >= '\u0660' && c <= '\u0669') || (c >= '\u06F0' && c <= '\u06F9')
1126 || (c >= '\u0966' && c <= '\u096F') || (c >= '\u09E6' && c <= '\u09EF') || (c >= '\u0A66' && c <= '\u0A6F')
1127 || (c >= '\u0AE6' && c <= '\u0AEF') || (c >= '\u0B66' && c <= '\u0B6F') || (c >= '\u0BE7' && c <= '\u0BEF')
1128 || (c >= '\u0C66' && c <= '\u0C6F') || (c >= '\u0CE6' && c <= '\u0CEF') || (c >= '\u0D66' && c <= '\u0D6F')
1129 || (c >= '\u0E50' && c <= '\u0E59') || (c >= '\u0ED0' && c <= '\u0ED9') || (c >= '\u0F20' && c <= '\u0F29');
1132 /** Extender - XMl Specification 1.0 [89] */
1133 private static final boolean Extender(char c) {
1134 return c == '\u00B7' || c == '\u02D0' || c == '\u02D1' || c == '\u0387'
1135 || c == '\u0640' || c == '\u0E46' || c == '\u0EC6' || c == '\u3005'
1136 || (c >= '\u3031' && c <= '\u3035') || (c >= '\u309D' && c <= '\u309E') || (c >= '\u30FC' && c <= '\u30FE');
1139 /** Whitespace - XML Specification 1.0 [3] */
1140 private static final boolean S(char c) {
1141 return c == '\u0020' || c == '\u0009' || c == '\r' || c == '\n';