1 // Copyright (C) 2003 Adam Megacz <adam@ibex.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[] { '"' };
73 private int base; // base+off == distance into the stream
76 private Element current;
78 // used in readEntity() to process a single character without creating a new array
79 private char[] singlechar = new char[1];
82 public XML() { this(BUFFER_SIZE); }
84 public XML(int bSize) {
85 buf = new char[bSize];
87 current = (Element)elements.remove(false);
88 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; }
97 /** Returns the global file offset at the beginning of the last process call. */
98 public int getGlobalOffset() { return base + off; }
101 * Parse given input and call the abstract event functions.
103 * Careful with threading, as this function is not synchronized.
105 public final void parse(Reader reader) throws IOException, Exn {
110 clear(); // clean up possible mid-way linked-list element
113 // process the stream
116 if (current.qName == null) break;
117 throw new Exn("reached eof without closing <"+current.qName+"> element", Exn.WFC, getLine(), getCol());
120 if (buf[off] == '<') readTag();
121 readChars(current.qName != null);
123 } finally { clear(); } // clean up elements
126 /** remove any leftover elements from the linked list and queue them */
127 private final void clear() {
128 for (Element last = current; current.parent != null; ) {
129 current = current.parent;
131 elements.append(last);
136 /** reads in a tag. expects <tt>buf[off] == '<'</tt> */
137 private final void readTag() throws IOException, Exn {
138 // Start Tag '<' Name (S Attribute)* S? '>'
139 boolean starttag = true;
141 // End Tag '</' Name S? '>'
142 boolean endtag = false;
144 // if (starttag & endtag) then: EmptyElemTag '<' Name (S Attribute)* S? '/>'
146 // Position in the name of the ':' namespace prefix
152 if (!buffer(1)) throw new EOFException("Unexpected EOF processing element tag");
154 // work out what we can from the beginning of the tag
157 // definitions here don't necessarily conform to xml spec (as DTDs not yet implemented)
159 if (!buffer(4)) throw new EOFException("Unexpected EOF processing <! element");
164 if (buf[off+1] != '-') { bad = true; break; }
165 col += 2; off += 2; len -= 2;
167 // Comment '<!--' ((Char - '-') | ('-' (Char - '-')))* '-->'
168 readChars(false, "-->", false);
169 col += 3; off += 3; len -= 3;
172 // we don't care about the following definitions
176 || buf[off+1] != 'T' || buf[off+2] != 'T' || buf[off+3] != 'L'
177 || buf[off+4] != 'I' || buf[off+5] != 'S' || buf[off+6] != 'T') {
180 col += 7; off += 7; len -= 7;
182 // ATTLIST '<!ATTLIST' (Char* - '>') '>'
183 readChars(false, ">", true);
188 || buf[off+1] != 'O' || buf[off+2] != 'C' || buf[off+3] != 'T'
189 || buf[off+4] != 'Y' || buf[off+5] != 'P' || buf[off+6] != 'E') {
192 col += 7; off += 7; len -= 7;
194 // DTD '<!DOCTYPE' (Char* - '>') '>'
195 readChars(false, ">", true);
201 } else if (buf[off+1] == 'L' && buf[off+2] == 'E' && buf[off+3] == 'M'
202 && buf[off+4] == 'E' && buf[off+5] == 'N' && buf[off+6] == 'T') {
203 // ELEMENT '<!ELEMENT' (Char* - '>') '>'
204 readChars(false, ">", true);
207 } else if (buf[off+1] == 'N' && buf[off+2] == 'T' && buf[off+3] == 'I'
208 && buf[off+4] == 'T' && buf[off+5] == 'Y') {
209 // ENTITY '<!ENTITY' (Char* - '>') '>'
210 readChars(false, ">", true);
220 || buf[off+1] != 'O' || buf[off+2] != 'T' || buf[off+3] != 'A' || buf[off+4] != 'T'
221 || buf[off+5] != 'I' || buf[off+6] != 'O' || buf[off+7] != 'N') {
224 col += 8; off += 8; len -= 8;
225 // NOTATION '<!NOTATION' (Char* - '>') '>'
226 readChars(false, ">", true);
233 if (bad) throw new Exn("element tag start character is invalid", Exn.MARKUP, getLine(), getCol());
235 } else if (s == '?') {
236 // PI (Ignored) '<?' (Char* - (Char* '?>' Char*)) '?>'
238 readChars(false, "?>", true);
239 if (!buffer(2)) throw new EOFException("Unexpected EOF at end of Processing Instruction");
240 col += 2; off += 2; len -= 2;
242 } else if (s == '[') {
244 || buf[off+1] != 'C' || buf[off+2] != 'D' || buf[off+3] != 'A'
245 || buf[off+4] != 'T' || buf[off+5] != 'A' || buf[off+6] != '[') {
247 // Conditional '<![' (Char* - (Char* ']]>' Char*)) ']]>'
248 readChars(false, "]]>", false);
250 col += 7; off += 7; len -=7;
251 // CDATA '<![CDATA[' (Char* - (Char* ']]>' Char*)) ']]>'
252 readChars(true, "]]>", false);
254 col += 3; off += 3; len -= 3;
257 // End Tag '</' Name S? '>'
262 if (!buffer(1)) throw new EOFException("Unexpected EOF processing end tag");
266 if (!Name(s)) throw new Exn("invalid starting character in element name", Exn.MARKUP, getLine(), getCol());
268 // find the element name (defined in XML Spec: section 2.3)
269 for (namelen = 0; ; namelen++) {
270 if (!buffer(namelen+1)) throw new EOFException("Unexpected EOF in element tag name");
272 s = buf[off+namelen];
274 if (S(s) || s == '>') {
276 } else if (s == '/') {
279 } else if (s == ':' && namelen > 0 && prefix < 1) {
280 // we have a definition of the prefix range available
282 } else if (!NameChar(s)) {
283 throw new Exn("element name contains invalid character", Exn.MARKUP, getLine(), getCol());
287 // process name (based on calculated region)
288 if (namelen < 1) throw new Exn("element name is null", Exn.MARKUP, getLine(), getCol());
290 // we have marked out the name region, so turn it into a string and move on
291 String qName = new String(buf, off, namelen);
293 col += namelen; off += namelen; len -= namelen;
296 // create the in-memory element representation of this beast
297 // if current.qName == null then this is the root element we're dealing with
298 if (current.qName != null) {
299 Element next = (Element)elements.remove(false);
300 if (next == null) next = new Element();
301 //next.clear(); // TODO: remove as elements now checked as they're added to the queue
302 next.parent = current;
306 current.qName = qName;
309 current.prefix = current.qName.substring(0, prefix);
310 current.localName = current.qName.substring(prefix+1);
312 current.prefix = null;
313 current.localName = current.qName;
316 // process attributes
318 if (!buffer(1)) throw new EOFException("Unexpected EOF - processing attributes part 1");
319 while (buf[off] != '/' && buf[off] != '>') {
321 if (!buffer(1)) throw new EOFException("Unexpected EOF - processing attributes part 2");
325 // work out the uri of this element
326 current.uri = current.getUri(current.getPrefix());
327 if (current.getUri().equals("") && current.getPrefix() != null)
328 current.addError(new Exn("undefined prefix '"+current.getPrefix()+"'", Exn.NC, getLine(), getCol()));
331 // this is an end-of-element tag
332 if (!qName.equals(current.getQName())) throw new Exn(
333 "end tag </"+qName+"> does not line up with start tag <"+current.getQName()+">", Exn.WFC, getLine(), getCol()
337 // deal with whitespace
341 if (!buffer(1)) throw new EOFException("Unexpected EOF before end of tag");
342 if (buf[off] == '/') {
346 if (!buffer(1)) throw new EOFException("Unexpected EOF before end of endtag");
347 if (buf[off] == '>') {
350 throw new Exn("missing '>' character from element '"+qName+"'", Exn.MARKUP, getLine(), getCol());
353 // send element signals
354 if (starttag) startElement(current);
358 // we just closed an element, so remove it from the element 'stack'
359 if (current.getParent() == null) {
360 // we just finished the root element
363 Element last = current;
364 current = current.parent;
366 elements.append(last);
372 /** reads in an attribute of an element. expects Name(buf[off]) */
373 private final void readAttribute() throws IOException, Exn {
376 String n, v, p, u; // attribute name, value, prefix and uri respectively
377 n = v = p = u = null;
380 // find the element name (defined in XML Spec: section 2.3)
381 for (ref= 0; ; ref++) {
382 if (!buffer(ref+1)) throw new EOFException("Unexpected EOF in read attribute loop part 1");
386 if (s == '=' || S(s)) {
388 } else if (s == ':' && ref > 0 && prefix < 1) {
389 // we have a definition of the prefix range available
391 } else if (!NameChar(s)) {
392 throw new Exn("attribute name contains invalid characters", Exn.MARKUP, getLine(), getCol());
396 // determine prefix and key name
398 p = new String(buf, off, prefix-1);
399 col += prefix; off += prefix; len -= prefix; ref -= prefix;
401 n = new String(buf, off, ref);
402 col += ref; off += ref; len -= ref;
404 // find name/value divider ('=')
406 if (!buffer(1)) throw new EOFException("Unexpected EOF before attribute '=' divider");
407 if (buf[off] != '=') throw new Exn("attribute name not followed by '=' sign", Exn.MARKUP, getLine(), getCol());
412 if (!buffer(1)) throw new EOFException("Unexpected EOF after attribute '=' divider");
415 if (buf[off] == '\'' || buf[off] == '"') {
418 throw new Exn("attribute '"+n+"' must have attribute wrapped in ' or \"", Exn.MARKUP, getLine(), getCol());
422 // find the attribute value
423 attval: for (ref = 0; ; ref++) {
424 if (!buffer(ref+1)) throw new EOFException("Unexpected EOF in attribute value");
426 if (buf[off+ref] == wrap) {
428 } else if (buf[off+ref] == '<') {
429 throw new Exn("attribute value for '"+n+"' must not contain '<'", Exn.WFC, getLine(), getCol());
433 v = new String(buf, off, ref);
434 col += ref; off += ref; len -= ref;
436 // remove end wrapper character
440 if (p != null && p.equals("xmlns")) {
441 current.addUri(n, v);
442 } else if (n.equals("xmlns")) {
443 if (current.getUri().equals("")) {
444 current.addUri("", v);
446 current.addError(new Exn("default namespace definition repeated", Exn.NC, getLine(), getCol()));
449 // find attribute uri
450 u = current.getUri(p);
451 if (p != null && u.equals("")) current.addError(new Exn("undefined attribute prefix '"+p+"'", Exn.NC, getLine(), getCol()));
453 // check to see if attribute is a repeat
454 for (int i=0; current.len > i; i++) if (n.equals(current.getAttrKey(i)) && u.equals(current.getAttrUri(i))) throw new Exn(
455 "attribute name '"+n+"' may not appear more than once in the same element tag", Exn.WFC, getLine(), getCol()
458 current.addAttr(n, v, u);
462 /** reads an entity and processes out its value. expects buf[off] == '&' */
463 private final void readEntity() throws IOException, Exn {
465 if (!buffer(2)) throw new EOFException("Unexpected EOF reading entity");
467 boolean unknown = false;
473 if (buf[off] == 'x') { off++; len--; radix = 16; } else { radix = 10; }
476 // read in each char, then shift total value to the left and add the extra
477 // style of loop is slightly different from all the others, as this should run a limited number of times
478 findchar: while (true) {
479 if (!buffer(1)) throw new EOFException("Unexpected EOF reading entity");
480 int d = Character.digit(buf[off], radix);
482 if (buf[off] != ';') throw new Exn("illegal characters in entity reference", Exn.WFC, getLine(), getCol());
491 singlechar[0] = Character.forDigit(c, radix);
492 characters(singlechar, 0, 1);
496 if (buffer(4) && buf[off+1] == 'm' && buf[off+2] == 'p' && buf[off+3] == ';') {
497 characters(single_amp, 0, 1); // &
498 off += 4; len -= 4; col++;
499 } else if (buffer(5) && buf[off+1] == 'p' && buf[off+2] == 'o' && buf[off+3] == 's' && buf[off+4] == ';') {
500 characters(single_apos, 0, 1); // '
501 off += 5; len -= 5; col++;
508 if (buffer(3) && buf[off+1] == 't' && buf[off+2] == ';') {
509 characters(single_gt, 0, 1); // >
510 off += 3; len -= 3; col++;
517 if (buffer(3) && buf[off+1] == 't' && buf[off+2] == ';') {
518 characters(single_lt, 0, 1); // <
519 off += 3; len -= 3; col++;
526 if (buffer(5) && buf[off+1] == 'u' && buf[off+2] == 'o' && buf[off+3] == 't' && buf[off+4] == ';') {
527 characters(single_quot, 0, 1); // "
528 off += 5; len -= 5; col++;
534 // TODO: check a parser-level Hash of defined entities
537 if (unknown) throw new Exn("unknown entity (<!ENTITY> not supported)", Exn.WFC, getLine(), getCol());
540 /** reads until the passed string is encountered. */
541 private final void readChars(boolean p, String match, boolean entities) throws IOException, Exn {
543 char[] end = match.toCharArray();
545 for (boolean more = true; more;) {
546 if (!buffer(1)) return;
548 buf: for (ref = 0; ref < len; ref++) {
549 switch (buf[off+ref]) {
550 case '\r': // windows or macos9 newline
551 // normalise and process
552 buf[off+ref] = '\n'; ref++;
553 if (p) characters(buf, off, ref);
554 off += ref; len -= ref; ref = -1;
557 // windows double-char newline; skip the next char
558 if (!buffer(1)) return;
559 if (buf[off] == '\n') { off++; len--; }
562 case '\n': // unix newline
564 if (p) characters(buf, off, ref);
565 off += ref; len -= ref; ref = -1;
572 if (ref > 0) characters(buf, off, ref);
573 off += ref; len -= ref; ref = -1;
580 if (!buffer(ref+end.length)) continue buf;
581 for (int i=0; end.length > i; i++) if (end[i] != buf[off+ref+i]) continue buf;
587 if (p && ref > 0) characters(buf, off, ref);
588 off += ref; len -= ref; col += ref;
593 * reads until a <tt><</tt> symbol is encountered
594 * @param p If true call the characters(char[],int,int) funciton for the processed characters
596 private final void readChars(boolean p) throws IOException, Exn {
599 for (boolean more = true; more;) {
600 if (!buffer(1)) return;
602 buf: for (ref = 0; ref < len; ref++) {
603 switch (buf[off+ref]) {
604 case '\r': // windows or macos9 newline
605 // normalise and process
606 buf[off+ref] = '\n'; ref++;
607 if (p) characters(buf, off, ref);
608 off += ref; len -= ref; ref = -1;
611 // windows double-char newline; skip the next char
612 if (!buffer(1)) return;
613 if (buf[off] == '\n') { off++; len--; }
616 case '\n': // unix newline
618 if (p) characters(buf, off, ref);
619 off += ref; len -= ref; ref = -1;
625 if (ref > 0) characters(buf, off, ref);
626 off += ref; len -= ref; ref = -1;
631 case '<': // end of chars section
637 if (p && ref > 0) characters(buf, off, ref);
638 off += ref; len -= ref; col += ref;
642 /** reads until a non-whitespace symbol is encountered */
643 private final void readWhitespace() throws IOException, Exn {
646 for (boolean more = true; more;) {
647 if (!buffer(1)) return;
649 buf: for (ref = 0; ref < len; ref++) {
650 switch (buf[off+ref]) {
651 case '\r': // windows or macos9 newline
652 // normalise and process
654 whitespace(buf, off, ++ref);
655 off += ref; len -= ref; ref = -1;
658 // windows double-char newline; skip the next char
659 if (!buffer(1)) return;
660 if (buf[off] == '\n') { off++; len--; }
663 case '\n': // unix newline
664 whitespace(buf, off, ++ref);
665 off += ref; len -= ref; ref = -1;
673 default: // end of whitespace
679 off += ref; len -= ref; col += ref;
684 * attempt to fill the buffer.
686 * @param min Minimum number of characters to read (even if we have to block to do it).
687 * @return return false if min can't be reached.
689 private final boolean buffer(int min) throws IOException {
690 if (len > min) return true;
692 if (buf.length - (off+len) >= min) {
693 // plenty of space left on the end of the buffer
694 } else if (off >= min) {
695 // moving offset data to start will leave enough free space on the end
696 System.arraycopy(buf, off, buf, 0, len);
700 // buffer size will have to be increased
701 char[] newbuf = new char[buf.length * 2];
702 System.arraycopy(buf, off, newbuf, 0, len);
709 int newlen = in.read(buf, off+len, buf.length-(off+len));
710 if (newlen < 0) return false;
718 /////////////////////////////////////////////////////////////////////////////////////////////
719 // Abstract SAX-Like Interface
720 /////////////////////////////////////////////////////////////////////////////////////////////
723 * Called when the start of an element is processed.
725 * <p><b>DO NOT</b> store a reference to the Element object, as
726 * they are reused by XML Parser.</p>
728 public abstract void startElement(Element e) throws Exn;
731 * Represents up to a line of character data.
733 * <p>Newlines are all normalised to the Unix \n as per the XML Spec,
734 * and a newline will only appear as the last character in the passed
737 * <p>XML.getLine() and XML.getCol() report the position at the
738 * beginning of this character segment, which can be processed in a
739 * line-by-line fashion due to the above newline restriction.</p>
741 public abstract void characters(char[] ch, int start, int length) throws Exn, IOException;
743 /** Represents up to a line of ignorable whitespace. */
744 public abstract void whitespace(char[] ch, int start, int length) throws Exn, IOException;
746 /** Represents the end of an Element. */
747 public abstract void endElement(Element e) throws Exn, IOException;
750 /////////////////////////////////////////////////////////////////////////////////////////////
751 // Inner Classes for Parser Support
752 /////////////////////////////////////////////////////////////////////////////////////////////
755 * Represents an element in an XML document. Stores a reference to its
756 * parent, forming a one-way linked list.
758 * Element objects are reused, so client code making use of them must
759 * drop their references after the specific element process function
762 public static final class Element {
764 private static final int DEFAULT_ATTR_SIZE = 10;
766 protected Element parent = null;
768 protected String uri = null;
769 protected String localName = null;
770 protected String qName = null;
771 protected String prefix = null;
773 protected Hash urimap = new Hash(3,3);
775 protected String[] keys = new String[DEFAULT_ATTR_SIZE];
776 protected String[] vals = new String[DEFAULT_ATTR_SIZE];
777 protected String[] uris = new String[DEFAULT_ATTR_SIZE];
778 protected int len = 0;
780 protected Exn[] errors = new Exn[] {};
782 /** Parent of current element. */
783 public Element getParent() { return parent; }
785 /** Qualified Name of current element. XML Namespace Spec 14-Jan-1999 [6] */
786 public String getQName() { return qName; }
788 /** LocalPart of current element. XML Namespace Spec 14-Jan-1999 [8] */
789 public String getLocalName() { return localName; }
791 /** Prefix of current element. Substring of qName. XML Namespace Spec 14-Jan-1999 [7] */
792 public String getPrefix() { return prefix; }
795 public Hash getUriMap() {
796 Hash map = new Hash();
797 for (Element e = this; e != null; e = e.getParent()) {
798 java.util.Enumeration en = e.urimap.keys();
799 while(en.hasMoreElements()) {
800 String key = (String)en.nextElement();
801 String val = getUri(key);
808 /** URI of current tag. XML Namespace Spec 14-Jan-1999 section 1 */
809 public String getUri() { return getUri(prefix); }
811 /** URI of a given prefix. Never returns null, instead gives "". */
812 public String getUri(String p) {
814 for (Element e = this; e != null && ret == null; e = e.getParent()) {
815 ret = (String)e.urimap.get(p == null ? "" : p);
817 return ret == null ? "" : ret;
820 /** An array of attribute names. */
821 public String getAttrKey(int pos) { return len > pos ? keys[pos] : null; }
823 /** An array of attribute values. */
824 public String getAttrVal(int pos) { return len > pos ? vals[pos] : null; }
826 /** An array of attribute uris. */
827 public String getAttrUri(int pos) { return len > pos ? uris[pos] : null; }
829 /** Current number of attributes in the element. */
830 public int getAttrLen() { return len; }
832 /** Poor performance, but easier to use when speed is not a concern */
833 public Hash getAttrHash() {
834 Hash ret = new Hash(getAttrLen() * 2, 3);
835 for(int i=0; i<len; i++)
836 ret.put(getAttrKey(i), getAttrVal(i));
840 /** Poor performance, but easier to use */
841 public String getAttrVal(String key) {
842 for(int i=0; i<len; i++) if (keys[i].equals(key)) return vals[i];
846 /** An array of non-fatal errors related to this element. */
847 public Exn[] getErrors() { return errors; }
849 protected Element() { }
851 /** Add (replace if exists in current element) a Namespace prefix/uri map. */
852 public void addUri(String name, String value) {
853 urimap.put(name, value);
856 /** Add an attribute. */
857 protected void addAttr(String key, String val, String uri) {
858 if (len == keys.length) {
859 // increase the size of the attributes arrays
860 String[] newkeys = new String[keys.length*2];
861 String[] newvals = new String[vals.length*2];
862 String[] newuris = new String[uris.length*2];
863 System.arraycopy(keys, 0, newkeys, 0, keys.length);
864 System.arraycopy(vals, 0, newvals, 0, vals.length);
865 System.arraycopy(uris, 0, newuris, 0, uris.length);
866 keys = newkeys; vals = newvals; uris = newuris;
876 protected void addError(Exn e) {
877 // it doesn't really matter about continually expanding the array, as this case is quite rare
878 Exn[] newe = new Exn[errors.length+1];
879 System.arraycopy(errors, 0, newe, 0, errors.length);
880 newe[errors.length] = e;
884 /** Empty out all the data from the Element. */
885 protected void clear() {
887 uri = localName = qName = prefix = null;
890 if (keys.length != vals.length || vals.length != uris.length) {
891 keys = new String[DEFAULT_ATTR_SIZE];
892 vals = new String[DEFAULT_ATTR_SIZE];
893 uris = new String[DEFAULT_ATTR_SIZE];
895 for (int i=0; keys.length > i; i++) { keys[i] = null; vals[i] = null; uris[i] = null; };
899 errors = new Exn[] {};
903 /** Parse or Structural Error */
904 public static class Exn extends Exception {
905 /** Violation of Markup restrictions in XML Specification - Fatal Error */
906 public static final int MARKUP = 1;
908 /** Well-Formedness Constraint Violation - Fatal Error */
909 public static final int WFC = 2;
911 /** Namespace Constraint Violation - Recoverable Error */
912 public static final int NC = 3;
914 /** Schema Violation - Fatal Error */
915 public static final int SCHEMA = 4;
917 private String error;
922 public Exn(String e) { this(e, MARKUP, -1, -1); }
924 public Exn(String e, int type, int line, int col) {
931 public int getType() { return this.type; }
932 public int getLine() { return this.line; }
933 public int getCol() { return this.col; }
934 public String getMessage() { return this.error + (line >= 0 && col >= 0 ? " at " + line + ":" + col: ""); }
938 /////////////////////////////////////////////////////////////////////////////////////////////
939 // Static Support Functions for the XML Specification
940 /////////////////////////////////////////////////////////////////////////////////////////////
942 // attempt to avoid these functions unless you *expect* the input to fall in the given range.
944 /** First Character of Name - XML Specification 1.0 [5] */
945 private static final boolean Name(char c) {
946 return BaseCharAscii(c) || c == '_' || c == ':' || Letter(c);
949 /** NameChar - XML Specification 1.0 [4] */
950 private static final boolean NameChar(char c) {
951 return BaseCharAscii(c) || c == '.' || c == '-' || c == '_' || c == ':'
952 || Digit(c) || Letter(c) || Extender(c); // TODO: || CombiningChar(c);
955 /** BaseChar - XMl Specification 1.0 [84] */
956 private static final boolean Letter(char c) {
957 return BaseChar(c) || Ideographic(c);
960 /** Elements of BaseChar that exist in ASCII. */
961 private static final boolean BaseCharAscii(char c) {
962 return (c >= '\u0041' && c <= '\u005A') || (c >= '\u0061' && c <= '\u007A');
965 /** Char - XML Specification 1.0 [2] */
966 private static final boolean Char(char c) {
967 // u000A == r and u000D == n, but the javac compiler can't handle the \ u form
968 return c == '\u0009' || c == '\r' || c == '\n'
969 || (c >= '\u0020' && c <= '\uD7FF')
970 || (c >= '\uE000' && c <= '\uFFFD');
973 /** BaseChar - XML Specification 1.0 [85] */
974 private static final boolean BaseChar(char c) {
975 return BaseCharAscii(c) || (c >= '\u00C0' && c <= '\u00D6')
976 || (c >= '\u00D8' && c <= '\u00F6') || (c >= '\u00F8' && c <= '\u00FF') || (c >= '\u0100' && c <= '\u0131')
977 || (c >= '\u0134' && c <= '\u013E') || (c >= '\u0141' && c <= '\u0148') || (c >= '\u014A' && c <= '\u017E')
978 || (c >= '\u0180' && c <= '\u01C3') || (c >= '\u01CD' && c <= '\u01F0') || (c >= '\u01F4' && c <= '\u01F5')
979 || (c >= '\u01FA' && c <= '\u0217') || (c >= '\u0250' && c <= '\u02A8') || (c >= '\u02BB' && c <= '\u02C1')
980 || (c == '\u0386') || (c >= '\u0388' && c <= '\u038A') || (c == '\u038C')
981 || (c >= '\u038E' && c <= '\u03A1') || (c >= '\u03A3' && c <= '\u03CE') || (c >= '\u03D0' && c <= '\u03D6')
982 || (c == '\u03DA') || (c == '\u03DC') || (c == '\u03DE')
984 || (c >= '\u03E2' && c <= '\u03F3') || (c >= '\u0401' && c <= '\u040C') || (c >= '\u040E' && c <= '\u044F')
985 || (c >= '\u0451' && c <= '\u045C') || (c >= '\u045E' && c <= '\u0481') || (c >= '\u0490' && c <= '\u04C4')
986 || (c >= '\u04C7' && c <= '\u04C8') || (c >= '\u04CB' && c <= '\u04CC') || (c >= '\u04D0' && c <= '\u04EB')
987 || (c >= '\u04EE' && c <= '\u04F5') || (c >= '\u04F8' && c <= '\u04F9') || (c >= '\u0531' && c <= '\u0556')
989 || (c >= '\u0561' && c <= '\u0586') || (c >= '\u05D0' && c <= '\u05EA') || (c >= '\u05F0' && c <= '\u05F2')
990 || (c >= '\u0621' && c <= '\u063A') || (c >= '\u0641' && c <= '\u064A') || (c >= '\u0671' && c <= '\u06B7')
991 || (c >= '\u06BA' && c <= '\u06BE') || (c >= '\u06C0' && c <= '\u06CE') || (c >= '\u06D0' && c <= '\u06D3')
993 || (c >= '\u06E5' && c <= '\u06E6') || (c >= '\u0905' && c <= '\u0939')
995 || (c >= '\u0958' && c <= '\u0961') || (c >= '\u0985' && c <= '\u098C') || (c >= '\u098F' && c <= '\u0990')
996 || (c >= '\u0993' && c <= '\u09A8') || (c >= '\u09AA' && c <= '\u09B0')
998 || (c >= '\u09B6' && c <= '\u09B9') || (c >= '\u09DF' && c <= '\u09E1') || (c >= '\u09F0' && c <= '\u09F1')
999 || (c >= '\u0A05' && c <= '\u0A0A') || (c >= '\u0A0F' && c <= '\u0A10') || (c >= '\u0A13' && c <= '\u0A28')
1000 || (c >= '\u0A2A' && c <= '\u0A30') || (c >= '\u0A32' && c <= '\u0A33') || (c >= '\u0A35' && c <= '\u0A36')
1001 || (c >= '\u0A38' && c <= '\u0A39') || (c >= '\u0A59' && c <= '\u0A5C')
1003 || (c >= '\u0A72' && c <= '\u0A74') || (c >= '\u0A85' && c <= '\u0A8B')
1005 || (c >= '\u0A8F' && c <= '\u0A91') || (c >= '\u0A93' && c <= '\u0AA8') || (c >= '\u0AAA' && c <= '\u0AB0')
1006 || (c >= '\u0AB2' && c <= '\u0AB3') || (c >= '\u0AB5' && c <= '\u0AB9')
1009 || (c >= '\u0B05' && c <= '\u0B0C') || (c >= '\u0B0F' && c <= '\u0B10') || (c >= '\u0B13' && c <= '\u0B28')
1010 || (c >= '\u0B2A' && c <= '\u0B30') || (c >= '\u0B32' && c <= '\u0B33') || (c >= '\u0B36' && c <= '\u0B39')
1012 || (c >= '\u0B5C' && c <= '\u0B5D') || (c >= '\u0B5F' && c <= '\u0B61') || (c >= '\u0B85' && c <= '\u0B8A')
1013 || (c >= '\u0B8E' && c <= '\u0B90') || (c >= '\u0B92' && c <= '\u0B95') || (c >= '\u0B99' && c <= '\u0B9A')
1015 || (c >= '\u0B9E' && c <= '\u0B9F') || (c >= '\u0BA3' && c <= '\u0BA4') || (c >= '\u0BA8' && c <= '\u0BAA')
1016 || (c >= '\u0BAE' && c <= '\u0BB5') || (c >= '\u0BB7' && c <= '\u0BB9') || (c >= '\u0C05' && c <= '\u0C0C')
1017 || (c >= '\u0C0E' && c <= '\u0C10') || (c >= '\u0C12' && c <= '\u0C28') || (c >= '\u0C2A' && c <= '\u0C33')
1018 || (c >= '\u0C35' && c <= '\u0C39') || (c >= '\u0C60' && c <= '\u0C61') || (c >= '\u0C85' && c <= '\u0C8C')
1019 || (c >= '\u0C8E' && c <= '\u0C90') || (c >= '\u0C92' && c <= '\u0CA8') || (c >= '\u0CAA' && c <= '\u0CB3')
1020 || (c >= '\u0CB5' && c <= '\u0CB9')
1022 || (c >= '\u0CE0' && c <= '\u0CE1') || (c >= '\u0D05' && c <= '\u0D0C') || (c >= '\u0D0E' && c <= '\u0D10')
1023 || (c >= '\u0D12' && c <= '\u0D28') || (c >= '\u0D2A' && c <= '\u0D39') || (c >= '\u0D60' && c <= '\u0D61')
1024 || (c >= '\u0E01' && c <= '\u0E2E')
1026 || (c >= '\u0E32' && c <= '\u0E33') || (c >= '\u0E40' && c <= '\u0E45') || (c >= '\u0E81' && c <= '\u0E82')
1028 || (c >= '\u0E87' && c <= '\u0E88')
1031 || (c >= '\u0E94' && c <= '\u0E97') || (c >= '\u0E99' && c <= '\u0E9F') || (c >= '\u0EA1' && c <= '\u0EA3')
1034 || (c >= '\u0EAA' && c <= '\u0EAB') || (c >= '\u0EAD' && c <= '\u0EAE')
1036 || (c >= '\u0EB2' && c <= '\u0EB3')
1038 || (c >= '\u0EC0' && c <= '\u0EC4') || (c >= '\u0F40' && c <= '\u0F47') || (c >= '\u0F49' && c <= '\u0F69')
1039 || (c >= '\u10A0' && c <= '\u10C5') || (c >= '\u10D0' && c <= '\u10F6')
1041 || (c >= '\u1102' && c <= '\u1103') || (c >= '\u1105' && c <= '\u1107')
1043 || (c >= '\u110B' && c <= '\u110C') || (c >= '\u110E' && c <= '\u1112')
1050 || (c >= '\u1154' && c <= '\u1155')
1052 || (c >= '\u115F' && c <= '\u1161')
1057 || (c >= '\u116D' && c <= '\u116E') || (c >= '\u1172' && c <= '\u1173')
1062 || (c >= '\u11AE' && c <= '\u11AF') || (c >= '\u11B7' && c <= '\u11B8')
1064 || (c >= '\u11BC' && c <= '\u11C2')
1068 || (c >= '\u1E00' && c <= '\u1E9B') || (c >= '\u1EA0' && c <= '\u1EF9') || (c >= '\u1F00' && c <= '\u1F15')
1069 || (c >= '\u1F18' && c <= '\u1F1D') || (c >= '\u1F20' && c <= '\u1F45') || (c >= '\u1F48' && c <= '\u1F4D')
1070 || (c >= '\u1F50' && c <= '\u1F57')
1074 || (c >= '\u1F5F' && c <= '\u1F7D') || (c >= '\u1F80' && c <= '\u1FB4') || (c >= '\u1FB6' && c <= '\u1FBC')
1076 || (c >= '\u1FC2' && c <= '\u1FC4') || (c >= '\u1FC6' && c <= '\u1FCC') || (c >= '\u1FD0' && c <= '\u1FD3')
1077 || (c >= '\u1FD6' && c <= '\u1FDB') || (c >= '\u1FE0' && c <= '\u1FEC') || (c >= '\u1FF2' && c <= '\u1FF4')
1078 || (c >= '\u1FF6' && c <= '\u1FFC')
1080 || (c >= '\u212A' && c <= '\u212B')
1082 || (c >= '\u2180' && c <= '\u2182') || (c >= '\u3041' && c <= '\u3094') || (c >= '\u30A1' && c <= '\u30FA')
1083 || (c >= '\u3105' && c <= '\u312C') || (c >= '\uAC00' && c <= '\uD7A3');
1086 /** BaseChar - XMl Specification 1.0 [86] */
1087 private static final boolean Ideographic(char c) {
1088 return (c >= '\u4E00' && c <= '\u9FA5') || c == '\u3007' || (c >= '\u3021' && c <= '\u3029');
1091 /** CombiningChar - XMl Specification 1.0 [87] */
1092 /*private static final boolean CombiningChar(char c) {
1093 return (c >= '\u0300' && c <= '\u0345')
1094 || (c >= '\u0360' && c <= '\u0361') || (c >= '\u0483' && c <= '\u0486') || (c >= '\u0591' && c <= '\u05A1')
1095 || (c >= '\u05A3' && c <= '\u05B9') || (c >= '\u05BB' && c <= '\u05BD')
1097 || (c >= '\u05C1' && c <= '\u05C2')
1099 || (c >= '\u064B' && c <= '\u0652')
1101 || (c >= '\u06D6' && c <= '\u06DC') || (c >= '\u06DD' && c <= '\u06DF') || (c >= '\u06E0' && c <= '\u06E4')
1102 || (c >= '\u06E7' && c <= '\u06E8') || (c >= '\u06EA' && c <= '\u06ED') || (c >= '\u0901' && c <= '\u0903')
1104 || (c >= '\u093E' && c <= '\u094C')
1106 || (c >= '\u0951' && c <= '\u0954') || (c >= '\u0962' && c <= '\u0963') || (c >= '\u0981' && c <= '\u0983')
1110 || (c >= '\u09C0' && c <= '\u09C4') || (c >= '\u09C7' && c <= '\u09C8') || (c >= '\u09CB' && c <= '\u09CD')
1112 || (c >= '\u09E2' && c <= '\u09E3')
1117 || (c >= '\u0A40' && c <= '\u0A42') || (c >= '\u0A47' && c <= '\u0A48') || (c >= '\u0A4B' && c <= '\u0A4D')
1118 || (c >= '\u0A70' && c <= '\u0A71') || (c >= '\u0A81' && c <= '\u0A83')
1120 || (c >= '\u0ABE' && c <= '\u0AC5') || (c >= '\u0AC7' && c <= '\u0AC9') || (c >= '\u0ACB' && c <= '\u0ACD')
1121 || (c >= '\u0B01' && c <= '\u0B03')
1123 || (c >= '\u0B3E' && c <= '\u0B43') || (c >= '\u0B47' && c <= '\u0B48') || (c >= '\u0B4B' && c <= '\u0B4D')
1124 || (c >= '\u0B56' && c <= '\u0B57') || (c >= '\u0B82' && c <= '\u0B83') || (c >= '\u0BBE' && c <= '\u0BC2')
1125 || (c >= '\u0BC6' && c <= '\u0BC8') || (c >= '\u0BCA' && c <= '\u0BCD')
1127 || (c >= '\u0C01' && c <= '\u0C03') || (c >= '\u0C3E' && c <= '\u0C44') || (c >= '\u0C46' && c <= '\u0C48')
1128 || (c >= '\u0C4A' && c <= '\u0C4D') || (c >= '\u0C55' && c <= '\u0C56') || (c >= '\u0C82' && c <= '\u0C83')
1129 || (c >= '\u0CBE' && c <= '\u0CC4') || (c >= '\u0CC6' && c <= '\u0CC8') || (c >= '\u0CCA' && c <= '\u0CCD')
1130 || (c >= '\u0CD5' && c <= '\u0CD6') || (c >= '\u0D02' && c <= '\u0D03') || (c >= '\u0D3E' && c <= '\u0D43')
1131 || (c >= '\u0D46' && c <= '\u0D48') || (c >= '\u0D4A' && c <= '\u0D4D')
1134 || (c >= '\u0E34' && c <= '\u0E3A') || (c >= '\u0E47' && c <= '\u0E4E')
1136 || (c >= '\u0EB4' && c <= '\u0EB9') || (c >= '\u0EBB' && c <= '\u0EBC') || (c >= '\u0EC8' && c <= '\u0ECD')
1137 || (c >= '\u0F18' && c <= '\u0F19')
1143 || (c >= '\u0F71' && c <= '\u0F84') || (c >= '\u0F86' && c <= '\u0F8B') || (c >= '\u0F90' && c <= '\u0F95')
1145 || (c >= '\u0F99' && c <= '\u0FAD') || (c >= '\u0FB1' && c <= '\u0FB7')
1147 || (c >= '\u20D0' && c <= '\u20DC')
1149 || (c >= '\u302A' && c <= '\u302F')
1154 /** Digit - XMl Specification 1.0 [88] */
1155 private static final boolean Digit(char c) {
1156 return (c >= '\u0030' && c <= '\u0039') || (c >= '\u0660' && c <= '\u0669') || (c >= '\u06F0' && c <= '\u06F9')
1157 || (c >= '\u0966' && c <= '\u096F') || (c >= '\u09E6' && c <= '\u09EF') || (c >= '\u0A66' && c <= '\u0A6F')
1158 || (c >= '\u0AE6' && c <= '\u0AEF') || (c >= '\u0B66' && c <= '\u0B6F') || (c >= '\u0BE7' && c <= '\u0BEF')
1159 || (c >= '\u0C66' && c <= '\u0C6F') || (c >= '\u0CE6' && c <= '\u0CEF') || (c >= '\u0D66' && c <= '\u0D6F')
1160 || (c >= '\u0E50' && c <= '\u0E59') || (c >= '\u0ED0' && c <= '\u0ED9') || (c >= '\u0F20' && c <= '\u0F29');
1163 /** Extender - XMl Specification 1.0 [89] */
1164 private static final boolean Extender(char c) {
1165 return c == '\u00B7' || c == '\u02D0' || c == '\u02D1' || c == '\u0387'
1166 || c == '\u0640' || c == '\u0E46' || c == '\u0EC6' || c == '\u3005'
1167 || (c >= '\u3031' && c <= '\u3035') || (c >= '\u309D' && c <= '\u309E') || (c >= '\u30FC' && c <= '\u30FE');
1170 /** Whitespace - XML Specification 1.0 [3] */
1171 private static final boolean S(char c) {
1172 return c == '\u0020' || c == '\u0009' || c == '\r' || c == '\n';