1 /*
2 * Copyright (c) 1997, 2013, Oracle and/or its affiliates. All rights reserved.
3 * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
4 *
5 * This code is free software; you can redistribute it and/or modify it
6 * under the terms of the GNU General Public License version 2 only, as
7 * published by the Free Software Foundation. Oracle designates this
8 * particular file as subject to the "Classpath" exception as provided
9 * by Oracle in the LICENSE file that accompanied this code.
10 *
11 * This code is distributed in the hope that it will be useful, but WITHOUT
12 * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
13 * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
14 * version 2 for more details (a copy is included in the LICENSE file that
15 * accompanied this code).
16 *
17 * You should have received a copy of the GNU General Public License version
18 * 2 along with this work; if not, write to the Free Software Foundation,
19 * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA.
20 *
21 * Please contact Oracle, 500 Oracle Parkway, Redwood Shores, CA 94065 USA
22 * or visit www.oracle.com if you need additional information or have any
23 * questions.
24 */
25
26 package com.sun.xml.internal.messaging.saaj.util;
27
28 // Imported from: org.apache.xerces.util
29 // Needed to work around differences in JDK1.2 and 1.3 and deal with userInfo
30
31 import java.io.IOException;
32 import java.io.Serializable;
33
34
35 /**********************************************************************
36 * A class to represent a Uniform Resource Identifier (URI). This class
37 * is designed to handle the parsing of URIs and provide access to
38 * the various components (scheme, host, port, userinfo, path, query
39 * string and fragment) that may constitute a URI.
40 * <p>
41 * Parsing of a URI specification is done according to the URI
42 * syntax described in RFC 2396
43 * <http://www.ietf.org/rfc/rfc2396.txt?number=2396>. Every URI consists
44 * of a scheme, followed by a colon (':'), followed by a scheme-specific
45 * part. For URIs that follow the "generic URI" syntax, the scheme-
46 * specific part begins with two slashes ("//") and may be followed
47 * by an authority segment (comprised of user information, host, and
48 * port), path segment, query segment and fragment. Note that RFC 2396
49 * no longer specifies the use of the parameters segment and excludes
50 * the "user:password" syntax as part of the authority segment. If
51 * "user:password" appears in a URI, the entire user/password string
52 * is stored as userinfo.
53 * <p>
54 * For URIs that do not follow the "generic URI" syntax (e.g. mailto),
55 * the entire scheme-specific part is treated as the "path" portion
56 * of the URI.
57 * <p>
58 * Note that, unlike the java.net.URL class, this class does not provide
59 * any built-in network access functionality nor does it provide any
60 * scheme-specific functionality (for example, it does not know a
61 * default port for a specific scheme). Rather, it only knows the
62 * grammar and basic set of operations that can be applied to a URI.
63 *
64 * @version
65 *
66 **********************************************************************/
67 public class JaxmURI implements Serializable {
68
69 /*******************************************************************
70 * MalformedURIExceptions are thrown in the process of building a URI
71 * or setting fields on a URI when an operation would result in an
72 * invalid URI specification.
73 *
74 ********************************************************************/
75 public static class MalformedURIException extends IOException {
76
77 /******************************************************************
78 * Constructs a <code>MalformedURIException</code> with no specified
79 * detail message.
80 ******************************************************************/
81 public MalformedURIException() {
82 super();
83 }
84
85 /*****************************************************************
86 * Constructs a <code>MalformedURIException</code> with the
87 * specified detail message.
88 *
89 * @param p_msg the detail message.
90 ******************************************************************/
91 public MalformedURIException(String p_msg) {
92 super(p_msg);
93 }
94 }
95
96 /** reserved characters */
97 private static final String RESERVED_CHARACTERS = ";/?:@&=+$,";
98
99 /** URI punctuation mark characters - these, combined with
100 alphanumerics, constitute the "unreserved" characters */
101 private static final String MARK_CHARACTERS = "-_.!~*'() ";
102
103 /** scheme can be composed of alphanumerics and these characters */
104 private static final String SCHEME_CHARACTERS = "+-.";
105
106 /** userinfo can be composed of unreserved, escaped and these
107 characters */
108 private static final String USERINFO_CHARACTERS = ";:&=+$,";
109
110 /** Stores the scheme (usually the protocol) for this URI. */
111 private String m_scheme = null;
112
113 /** If specified, stores the userinfo for this URI; otherwise null */
114 private String m_userinfo = null;
115
116 /** If specified, stores the host for this URI; otherwise null */
117 private String m_host = null;
118
119 /** If specified, stores the port for this URI; otherwise -1 */
120 private int m_port = -1;
121
122 /** If specified, stores the path for this URI; otherwise null */
123 private String m_path = null;
124
125 /** If specified, stores the query string for this URI; otherwise
126 null. */
127 private String m_queryString = null;
128
129 /** If specified, stores the fragment for this URI; otherwise null */
130 private String m_fragment = null;
131
132 private static boolean DEBUG = false;
133
134 /**
135 * Construct a new and uninitialized URI.
136 */
137 public JaxmURI() {
138 }
139
140 /**
141 * Construct a new URI from another URI. All fields for this URI are
142 * set equal to the fields of the URI passed in.
143 *
144 * @param p_other the URI to copy (cannot be null)
145 */
146 public JaxmURI(JaxmURI p_other) {
147 initialize(p_other);
148 }
149
150 /**
151 * Construct a new URI from a URI specification string. If the
152 * specification follows the "generic URI" syntax, (two slashes
153 * following the first colon), the specification will be parsed
154 * accordingly - setting the scheme, userinfo, host,port, path, query
155 * string and fragment fields as necessary. If the specification does
156 * not follow the "generic URI" syntax, the specification is parsed
157 * into a scheme and scheme-specific part (stored as the path) only.
158 *
159 * @param p_uriSpec the URI specification string (cannot be null or
160 * empty)
161 *
162 * @exception MalformedURIException if p_uriSpec violates any syntax
163 * rules
164 */
165 public JaxmURI(String p_uriSpec) throws MalformedURIException {
166 this((JaxmURI)null, p_uriSpec);
167 }
168
169 /**
170 * Construct a new URI from a base URI and a URI specification string.
171 * The URI specification string may be a relative URI.
172 *
173 * @param p_base the base URI (cannot be null if p_uriSpec is null or
174 * empty)
175 * @param p_uriSpec the URI specification string (cannot be null or
176 * empty if p_base is null)
177 *
178 * @exception MalformedURIException if p_uriSpec violates any syntax
179 * rules
180 */
181 public JaxmURI(JaxmURI p_base, String p_uriSpec) throws MalformedURIException {
182 initialize(p_base, p_uriSpec);
183 }
184
185 /**
186 * Construct a new URI that does not follow the generic URI syntax.
187 * Only the scheme and scheme-specific part (stored as the path) are
188 * initialized.
189 *
190 * @param p_scheme the URI scheme (cannot be null or empty)
191 * @param p_schemeSpecificPart the scheme-specific part (cannot be
192 * null or empty)
193 *
194 * @exception MalformedURIException if p_scheme violates any
195 * syntax rules
196 */
197 public JaxmURI(String p_scheme, String p_schemeSpecificPart)
198 throws MalformedURIException {
199 if (p_scheme == null || p_scheme.trim().length() == 0) {
200 throw new MalformedURIException(
201 "Cannot construct URI with null/empty scheme!");
202 }
203 if (p_schemeSpecificPart == null ||
204 p_schemeSpecificPart.trim().length() == 0) {
205 throw new MalformedURIException(
206 "Cannot construct URI with null/empty scheme-specific part!");
207 }
208 setScheme(p_scheme);
209 setPath(p_schemeSpecificPart);
210 }
211
212 /**
213 * Construct a new URI that follows the generic URI syntax from its
214 * component parts. Each component is validated for syntax and some
215 * basic semantic checks are performed as well. See the individual
216 * setter methods for specifics.
217 *
218 * @param p_scheme the URI scheme (cannot be null or empty)
219 * @param p_host the hostname or IPv4 address for the URI
220 * @param p_path the URI path - if the path contains '?' or '#',
221 * then the query string and/or fragment will be
222 * set from the path; however, if the query and
223 * fragment are specified both in the path and as
224 * separate parameters, an exception is thrown
225 * @param p_queryString the URI query string (cannot be specified
226 * if path is null)
227 * @param p_fragment the URI fragment (cannot be specified if path
228 * is null)
229 *
230 * @exception MalformedURIException if any of the parameters violates
231 * syntax rules or semantic rules
232 */
233 public JaxmURI(String p_scheme, String p_host, String p_path,
234 String p_queryString, String p_fragment)
235 throws MalformedURIException {
236 this(p_scheme, null, p_host, -1, p_path, p_queryString, p_fragment);
237 }
238
239 /**
240 * Construct a new URI that follows the generic URI syntax from its
241 * component parts. Each component is validated for syntax and some
242 * basic semantic checks are performed as well. See the individual
243 * setter methods for specifics.
244 *
245 * @param p_scheme the URI scheme (cannot be null or empty)
246 * @param p_userinfo the URI userinfo (cannot be specified if host
247 * is null)
248 * @param p_host the hostname or IPv4 address for the URI
249 * @param p_port the URI port (may be -1 for "unspecified"; cannot
250 * be specified if host is null)
251 * @param p_path the URI path - if the path contains '?' or '#',
252 * then the query string and/or fragment will be
253 * set from the path; however, if the query and
254 * fragment are specified both in the path and as
255 * separate parameters, an exception is thrown
256 * @param p_queryString the URI query string (cannot be specified
257 * if path is null)
258 * @param p_fragment the URI fragment (cannot be specified if path
259 * is null)
260 *
261 * @exception MalformedURIException if any of the parameters violates
262 * syntax rules or semantic rules
263 */
264 public JaxmURI(String p_scheme, String p_userinfo,
265 String p_host, int p_port, String p_path,
266 String p_queryString, String p_fragment)
267 throws MalformedURIException {
268 if (p_scheme == null || p_scheme.trim().length() == 0) {
269 throw new MalformedURIException("Scheme is required!");
270 }
271
272 if (p_host == null) {
273 if (p_userinfo != null) {
274 throw new MalformedURIException(
275 "Userinfo may not be specified if host is not specified!");
276 }
277 if (p_port != -1) {
278 throw new MalformedURIException(
279 "Port may not be specified if host is not specified!");
280 }
281 }
282
283 if (p_path != null) {
284 if (p_path.indexOf('?') != -1 && p_queryString != null) {
285 throw new MalformedURIException(
286 "Query string cannot be specified in path and query string!");
287 }
288
289 if (p_path.indexOf('#') != -1 && p_fragment != null) {
290 throw new MalformedURIException(
291 "Fragment cannot be specified in both the path and fragment!");
292 }
293 }
294
295 setScheme(p_scheme);
296 setHost(p_host);
297 setPort(p_port);
298 setUserinfo(p_userinfo);
299 setPath(p_path);
300 setQueryString(p_queryString);
301 setFragment(p_fragment);
302 }
303
304 /**
305 * Initialize all fields of this URI from another URI.
306 *
307 * @param p_other the URI to copy (cannot be null)
308 */
309 private void initialize(JaxmURI p_other) {
310 m_scheme = p_other.getScheme();
311 m_userinfo = p_other.getUserinfo();
312 m_host = p_other.getHost();
313 m_port = p_other.getPort();
314 m_path = p_other.getPath();
315 m_queryString = p_other.getQueryString();
316 m_fragment = p_other.getFragment();
317 }
318
319 /**
320 * Initializes this URI from a base URI and a URI specification string.
321 * See RFC 2396 Section 4 and Appendix B for specifications on parsing
322 * the URI and Section 5 for specifications on resolving relative URIs
323 * and relative paths.
324 *
325 * @param p_base the base URI (may be null if p_uriSpec is an absolute
326 * URI)
327 * @param p_uriSpec the URI spec string which may be an absolute or
328 * relative URI (can only be null/empty if p_base
329 * is not null)
330 *
331 * @exception MalformedURIException if p_base is null and p_uriSpec
332 * is not an absolute URI or if
333 * p_uriSpec violates syntax rules
334 */
335 private void initialize(JaxmURI p_base, String p_uriSpec)
336 throws MalformedURIException {
337 if (p_base == null &&
338 (p_uriSpec == null || p_uriSpec.trim().length() == 0)) {
339 throw new MalformedURIException(
340 "Cannot initialize URI with empty parameters.");
341 }
342
343 // just make a copy of the base if spec is empty
344 if (p_uriSpec == null || p_uriSpec.trim().length() == 0) {
345 initialize(p_base);
346 return;
347 }
348
349 String uriSpec = p_uriSpec.trim();
350 int uriSpecLen = uriSpec.length();
351 int index = 0;
352
353 // Check for scheme, which must be before `/'. Also handle names with
354 // DOS drive letters ('D:'), so 1-character schemes are not allowed.
355 int colonIdx = uriSpec.indexOf(':');
356 int slashIdx = uriSpec.indexOf('/');
357 if ((colonIdx < 2) || (colonIdx > slashIdx && slashIdx != -1)) {
358 int fragmentIdx = uriSpec.indexOf('#');
359 // A standalone base is a valid URI according to spec
360 if (p_base == null && fragmentIdx != 0 ) {
361 throw new MalformedURIException("No scheme found in URI.");
362 }
363 }
364 else {
365 initializeScheme(uriSpec);
366 index = m_scheme.length()+1;
367 }
368
369 // two slashes means generic URI syntax, so we get the authority
370 if (((index+1) < uriSpecLen) &&
371 (uriSpec.substring(index).startsWith("//"))) {
372 index += 2;
373 int startPos = index;
374
375 // get authority - everything up to path, query or fragment
376 char testChar = '\0';
377 while (index < uriSpecLen) {
378 testChar = uriSpec.charAt(index);
379 if (testChar == '/' || testChar == '?' || testChar == '#') {
380 break;
381 }
382 index++;
383 }
384
385 // if we found authority, parse it out, otherwise we set the
386 // host to empty string
387 if (index > startPos) {
388 initializeAuthority(uriSpec.substring(startPos, index));
389 }
390 else {
391 m_host = "";
392 }
393 }
394
395 initializePath(uriSpec.substring(index));
396
397 // Resolve relative URI to base URI - see RFC 2396 Section 5.2
398 // In some cases, it might make more sense to throw an exception
399 // (when scheme is specified is the string spec and the base URI
400 // is also specified, for example), but we're just following the
401 // RFC specifications
402 if (p_base != null) {
403
404 // check to see if this is the current doc - RFC 2396 5.2 #2
405 // note that this is slightly different from the RFC spec in that
406 // we don't include the check for query string being null
407 // - this handles cases where the urispec is just a query
408 // string or a fragment (e.g. "?y" or "#s") -
409 // see <http://www.ics.uci.edu/~fielding/url/test1.html> which
410 // identified this as a bug in the RFC
411 if (m_path.length() == 0 && m_scheme == null &&
412 m_host == null) {
413 m_scheme = p_base.getScheme();
414 m_userinfo = p_base.getUserinfo();
415 m_host = p_base.getHost();
416 m_port = p_base.getPort();
417 m_path = p_base.getPath();
418
419 if (m_queryString == null) {
420 m_queryString = p_base.getQueryString();
421 }
422 return;
423 }
424
425 // check for scheme - RFC 2396 5.2 #3
426 // if we found a scheme, it means absolute URI, so we're done
427 if (m_scheme == null) {
428 m_scheme = p_base.getScheme();
429 }
430 else {
431 return;
432 }
433
434 // check for authority - RFC 2396 5.2 #4
435 // if we found a host, then we've got a network path, so we're done
436 if (m_host == null) {
437 m_userinfo = p_base.getUserinfo();
438 m_host = p_base.getHost();
439 m_port = p_base.getPort();
440 }
441 else {
442 return;
443 }
444
445 // check for absolute path - RFC 2396 5.2 #5
446 if (m_path.length() > 0 &&
447 m_path.startsWith("/")) {
448 return;
449 }
450
451 // if we get to this point, we need to resolve relative path
452 // RFC 2396 5.2 #6
453 String path = "";
454 String basePath = p_base.getPath();
455
456 // 6a - get all but the last segment of the base URI path
457 if (basePath != null) {
458 int lastSlash = basePath.lastIndexOf('/');
459 if (lastSlash != -1) {
460 path = basePath.substring(0, lastSlash+1);
461 }
462 }
463
464 // 6b - append the relative URI path
465 path = path.concat(m_path);
466
467 // 6c - remove all "./" where "." is a complete path segment
468 index = -1;
469 while ((index = path.indexOf("/./")) != -1) {
470 path = path.substring(0, index+1).concat(path.substring(index+3));
471 }
472
473 // 6d - remove "." if path ends with "." as a complete path segment
474 if (path.endsWith("/.")) {
475 path = path.substring(0, path.length()-1);
476 }
477
478 // 6e - remove all "<segment>/../" where "<segment>" is a complete
479 // path segment not equal to ".."
480 index = 1;
481 int segIndex = -1;
482 String tempString = null;
483
484 while ((index = path.indexOf("/../", index)) > 0) {
485 tempString = path.substring(0, path.indexOf("/../"));
486 segIndex = tempString.lastIndexOf('/');
487 if (segIndex != -1) {
488 if (!tempString.substring(segIndex++).equals("..")) {
489 path = path.substring(0, segIndex).concat(path.substring(index+4));
490 }
491 else
492 index += 4;
493 }
494 else
495 index += 4;
496 }
497
498 // 6f - remove ending "<segment>/.." where "<segment>" is a
499 // complete path segment
500 if (path.endsWith("/..")) {
501 tempString = path.substring(0, path.length()-3);
502 segIndex = tempString.lastIndexOf('/');
503 if (segIndex != -1) {
504 path = path.substring(0, segIndex+1);
505 }
506 }
507 m_path = path;
508 }
509 }
510
511 /**
512 * Initialize the scheme for this URI from a URI string spec.
513 *
514 * @param p_uriSpec the URI specification (cannot be null)
515 *
516 * @exception MalformedURIException if URI does not have a conformant
517 * scheme
518 */
519 private void initializeScheme(String p_uriSpec)
520 throws MalformedURIException {
521 int uriSpecLen = p_uriSpec.length();
522 int index = 0;
523 String scheme = null;
524 char testChar = '\0';
525
526 while (index < uriSpecLen) {
527 testChar = p_uriSpec.charAt(index);
528 if (testChar == ':' || testChar == '/' ||
529 testChar == '?' || testChar == '#') {
530 break;
531 }
532 index++;
533 }
534 scheme = p_uriSpec.substring(0, index);
535
536 if (scheme.length() == 0) {
537 throw new MalformedURIException("No scheme found in URI.");
538 }
539 else {
540 setScheme(scheme);
541 }
542 }
543
544 /**
545 * Initialize the authority (userinfo, host and port) for this
546 * URI from a URI string spec.
547 *
548 * @param p_uriSpec the URI specification (cannot be null)
549 *
550 * @exception MalformedURIException if p_uriSpec violates syntax rules
551 */
552 private void initializeAuthority(String p_uriSpec)
553 throws MalformedURIException {
554 int index = 0;
555 int start = 0;
556 int end = p_uriSpec.length();
557 char testChar = '\0';
558 String userinfo = null;
559
560 // userinfo is everything up @
561 if (p_uriSpec.indexOf('@', start) != -1) {
562 while (index < end) {
563 testChar = p_uriSpec.charAt(index);
564 if (testChar == '@') {
565 break;
566 }
567 index++;
568 }
569 userinfo = p_uriSpec.substring(start, index);
570 index++;
571 }
572
573 // host is everything up to ':'
574 String host = null;
575 start = index;
576 while (index < end) {
577 testChar = p_uriSpec.charAt(index);
578 if (testChar == ':') {
579 break;
580 }
581 index++;
582 }
583 host = p_uriSpec.substring(start, index);
584 int port = -1;
585 if (host.length() > 0) {
586 // port
587 if (testChar == ':') {
588 index++;
589 start = index;
590 while (index < end) {
591 index++;
592 }
593 String portStr = p_uriSpec.substring(start, index);
594 if (portStr.length() > 0) {
595 for (int i = 0; i < portStr.length(); i++) {
596 if (!isDigit(portStr.charAt(i))) {
597 throw new MalformedURIException(
598 portStr +
599 " is invalid. Port should only contain digits!");
600 }
601 }
602 try {
603 port = Integer.parseInt(portStr);
604 }
605 catch (NumberFormatException nfe) {
606 // can't happen
607 }
608 }
609 }
610 }
611 setHost(host);
612 setPort(port);
613 setUserinfo(userinfo);
614 }
615
616 /**
617 * Initialize the path for this URI from a URI string spec.
618 *
619 * @param p_uriSpec the URI specification (cannot be null)
620 *
621 * @exception MalformedURIException if p_uriSpec violates syntax rules
622 */
623 private void initializePath(String p_uriSpec)
624 throws MalformedURIException {
625 if (p_uriSpec == null) {
626 throw new MalformedURIException(
627 "Cannot initialize path from null string!");
628 }
629
630 int index = 0;
631 int start = 0;
632 int end = p_uriSpec.length();
633 char testChar = '\0';
634
635 // path - everything up to query string or fragment
636 while (index < end) {
637 testChar = p_uriSpec.charAt(index);
638 if (testChar == '?' || testChar == '#') {
639 break;
640 }
641 // check for valid escape sequence
642 if (testChar == '%') {
643 if (index+2 >= end ||
644 !isHex(p_uriSpec.charAt(index+1)) ||
645 !isHex(p_uriSpec.charAt(index+2))) {
646 throw new MalformedURIException(
647 "Path contains invalid escape sequence!");
648 }
649 }
650 else if (!isReservedCharacter(testChar) &&
651 !isUnreservedCharacter(testChar)) {
652 throw new MalformedURIException(
653 "Path contains invalid character: " + testChar);
654 }
655 index++;
656 }
657 m_path = p_uriSpec.substring(start, index);
658
659 // query - starts with ? and up to fragment or end
660 if (testChar == '?') {
661 index++;
662 start = index;
663 while (index < end) {
664 testChar = p_uriSpec.charAt(index);
665 if (testChar == '#') {
666 break;
667 }
668 if (testChar == '%') {
669 if (index+2 >= end ||
670 !isHex(p_uriSpec.charAt(index+1)) ||
671 !isHex(p_uriSpec.charAt(index+2))) {
672 throw new MalformedURIException(
673 "Query string contains invalid escape sequence!");
674 }
675 }
676 else if (!isReservedCharacter(testChar) &&
677 !isUnreservedCharacter(testChar)) {
678 throw new MalformedURIException(
679 "Query string contains invalid character:" + testChar);
680 }
681 index++;
682 }
683 m_queryString = p_uriSpec.substring(start, index);
684 }
685
686 // fragment - starts with #
687 if (testChar == '#') {
688 index++;
689 start = index;
690 while (index < end) {
691 testChar = p_uriSpec.charAt(index);
692
693 if (testChar == '%') {
694 if (index+2 >= end ||
695 !isHex(p_uriSpec.charAt(index+1)) ||
696 !isHex(p_uriSpec.charAt(index+2))) {
697 throw new MalformedURIException(
698 "Fragment contains invalid escape sequence!");
699 }
700 }
701 else if (!isReservedCharacter(testChar) &&
702 !isUnreservedCharacter(testChar)) {
703 throw new MalformedURIException(
704 "Fragment contains invalid character:"+testChar);
705 }
706 index++;
707 }
708 m_fragment = p_uriSpec.substring(start, index);
709 }
710 }
711
712 /**
713 * Get the scheme for this URI.
714 *
715 * @return the scheme for this URI
716 */
717 public String getScheme() {
718 return m_scheme;
719 }
720
721 /**
722 * Get the scheme-specific part for this URI (everything following the
723 * scheme and the first colon). See RFC 2396 Section 5.2 for spec.
724 *
725 * @return the scheme-specific part for this URI
726 */
727 public String getSchemeSpecificPart() {
728 StringBuffer schemespec = new StringBuffer();
729
730 if (m_userinfo != null || m_host != null || m_port != -1) {
731 schemespec.append("//");
732 }
733
734 if (m_userinfo != null) {
735 schemespec.append(m_userinfo);
736 schemespec.append('@');
737 }
738
739 if (m_host != null) {
740 schemespec.append(m_host);
741 }
742
743 if (m_port != -1) {
744 schemespec.append(':');
745 schemespec.append(m_port);
746 }
747
748 if (m_path != null) {
749 schemespec.append((m_path));
750 }
751
752 if (m_queryString != null) {
753 schemespec.append('?');
754 schemespec.append(m_queryString);
755 }
756
757 if (m_fragment != null) {
758 schemespec.append('#');
759 schemespec.append(m_fragment);
760 }
761
762 return schemespec.toString();
763 }
764
765 /**
766 * Get the userinfo for this URI.
767 *
768 * @return the userinfo for this URI (null if not specified).
769 */
770 public String getUserinfo() {
771 return m_userinfo;
772 }
773
774 /**
775 * Get the host for this URI.
776 *
777 * @return the host for this URI (null if not specified).
778 */
779 public String getHost() {
780 return m_host;
781 }
782
783 /**
784 * Get the port for this URI.
785 *
786 * @return the port for this URI (-1 if not specified).
787 */
788 public int getPort() {
789 return m_port;
790 }
791
792 /**
793 * Get the path for this URI (optionally with the query string and
794 * fragment).
795 *
796 * @param p_includeQueryString if true (and query string is not null),
797 * then a "?" followed by the query string
798 * will be appended
799 * @param p_includeFragment if true (and fragment is not null),
800 * then a "#" followed by the fragment
801 * will be appended
802 *
803 * @return the path for this URI possibly including the query string
804 * and fragment
805 */
806 public String getPath(boolean p_includeQueryString,
807 boolean p_includeFragment) {
808 StringBuffer pathString = new StringBuffer(m_path);
809
810 if (p_includeQueryString && m_queryString != null) {
811 pathString.append('?');
812 pathString.append(m_queryString);
813 }
814
815 if (p_includeFragment && m_fragment != null) {
816 pathString.append('#');
817 pathString.append(m_fragment);
818 }
819 return pathString.toString();
820 }
821
822 /**
823 * Get the path for this URI. Note that the value returned is the path
824 * only and does not include the query string or fragment.
825 *
826 * @return the path for this URI.
827 */
828 public String getPath() {
829 return m_path;
830 }
831
832 /**
833 * Get the query string for this URI.
834 *
835 * @return the query string for this URI. Null is returned if there
836 * was no "?" in the URI spec, empty string if there was a
837 * "?" but no query string following it.
838 */
839 public String getQueryString() {
840 return m_queryString;
841 }
842
843 /**
844 * Get the fragment for this URI.
845 *
846 * @return the fragment for this URI. Null is returned if there
847 * was no "#" in the URI spec, empty string if there was a
848 * "#" but no fragment following it.
849 */
850 public String getFragment() {
851 return m_fragment;
852 }
853
854 /**
855 * Set the scheme for this URI. The scheme is converted to lowercase
856 * before it is set.
857 *
858 * @param p_scheme the scheme for this URI (cannot be null)
859 *
860 * @exception MalformedURIException if p_scheme is not a conformant
861 * scheme name
862 */
863 public void setScheme(String p_scheme) throws MalformedURIException {
864 if (p_scheme == null) {
865 throw new MalformedURIException(
866 "Cannot set scheme from null string!");
867 }
868 if (!isConformantSchemeName(p_scheme)) {
869 throw new MalformedURIException("The scheme is not conformant.");
870 }
871
872 m_scheme = p_scheme.toLowerCase();
873 }
874
875 /**
876 * Set the userinfo for this URI. If a non-null value is passed in and
877 * the host value is null, then an exception is thrown.
878 *
879 * @param p_userinfo the userinfo for this URI
880 *
881 * @exception MalformedURIException if p_userinfo contains invalid
882 * characters
883 */
884 public void setUserinfo(String p_userinfo) throws MalformedURIException {
885 if (p_userinfo == null) {
886 m_userinfo = null;
887 }
888 else {
889 if (m_host == null) {
890 throw new MalformedURIException(
891 "Userinfo cannot be set when host is null!");
892 }
893
894 // userinfo can contain alphanumerics, mark characters, escaped
895 // and ';',':','&','=','+','$',','
896 int index = 0;
897 int end = p_userinfo.length();
898 char testChar = '\0';
899 while (index < end) {
900 testChar = p_userinfo.charAt(index);
901 if (testChar == '%') {
902 if (index+2 >= end ||
903 !isHex(p_userinfo.charAt(index+1)) ||
904 !isHex(p_userinfo.charAt(index+2))) {
905 throw new MalformedURIException(
906 "Userinfo contains invalid escape sequence!");
907 }
908 }
909 else if (!isUnreservedCharacter(testChar) &&
910 USERINFO_CHARACTERS.indexOf(testChar) == -1) {
911 throw new MalformedURIException(
912 "Userinfo contains invalid character:"+testChar);
913 }
914 index++;
915 }
916 }
917 m_userinfo = p_userinfo;
918 }
919
920 /**
921 * Set the host for this URI. If null is passed in, the userinfo
922 * field is also set to null and the port is set to -1.
923 *
924 * @param p_host the host for this URI
925 *
926 * @exception MalformedURIException if p_host is not a valid IP
927 * address or DNS hostname.
928 */
929 public void setHost(String p_host) throws MalformedURIException {
930 if (p_host == null || p_host.trim().length() == 0) {
931 m_host = p_host;
932 m_userinfo = null;
933 m_port = -1;
934 }
935 else if (!isWellFormedAddress(p_host)) {
936 throw new MalformedURIException("Host is not a well formed address!");
937 }
938 m_host = p_host;
939 }
940
941 /**
942 * Set the port for this URI. -1 is used to indicate that the port is
943 * not specified, otherwise valid port numbers are between 0 and 65535.
944 * If a valid port number is passed in and the host field is null,
945 * an exception is thrown.
946 *
947 * @param p_port the port number for this URI
948 *
949 * @exception MalformedURIException if p_port is not -1 and not a
950 * valid port number
951 */
952 public void setPort(int p_port) throws MalformedURIException {
953 if (p_port >= 0 && p_port <= 65535) {
954 if (m_host == null) {
955 throw new MalformedURIException(
956 "Port cannot be set when host is null!");
957 }
958 }
959 else if (p_port != -1) {
960 throw new MalformedURIException("Invalid port number!");
961 }
962 m_port = p_port;
963 }
964
965 /**
966 * Set the path for this URI. If the supplied path is null, then the
967 * query string and fragment are set to null as well. If the supplied
968 * path includes a query string and/or fragment, these fields will be
969 * parsed and set as well. Note that, for URIs following the "generic
970 * URI" syntax, the path specified should start with a slash.
971 * For URIs that do not follow the generic URI syntax, this method
972 * sets the scheme-specific part.
973 *
974 * @param p_path the path for this URI (may be null)
975 *
976 * @exception MalformedURIException if p_path contains invalid
977 * characters
978 */
979 public void setPath(String p_path) throws MalformedURIException {
980 if (p_path == null) {
981 m_path = null;
982 m_queryString = null;
983 m_fragment = null;
984 }
985 else {
986 initializePath(p_path);
987 }
988 }
989
990 /**
991 * Append to the end of the path of this URI. If the current path does
992 * not end in a slash and the path to be appended does not begin with
993 * a slash, a slash will be appended to the current path before the
994 * new segment is added. Also, if the current path ends in a slash
995 * and the new segment begins with a slash, the extra slash will be
996 * removed before the new segment is appended.
997 *
998 * @param p_addToPath the new segment to be added to the current path
999 *
1000 * @exception MalformedURIException if p_addToPath contains syntax
1001 * errors
1002 */
1003 public void appendPath(String p_addToPath)
1004 throws MalformedURIException {
1005 if (p_addToPath == null || p_addToPath.trim().length() == 0) {
1006 return;
1007 }
1008
1009 if (!isURIString(p_addToPath)) {
1010 throw new MalformedURIException(
1011 "Path contains invalid character!");
1012 }
1013
1014 if (m_path == null || m_path.trim().length() == 0) {
1015 if (p_addToPath.startsWith("/")) {
1016 m_path = p_addToPath;
1017 }
1018 else {
1019 m_path = "/" + p_addToPath;
1020 }
1021 }
1022 else if (m_path.endsWith("/")) {
1023 if (p_addToPath.startsWith("/")) {
1024 m_path = m_path.concat(p_addToPath.substring(1));
1025 }
1026 else {
1027 m_path = m_path.concat(p_addToPath);
1028 }
1029 }
1030 else {
1031 if (p_addToPath.startsWith("/")) {
1032 m_path = m_path.concat(p_addToPath);
1033 }
1034 else {
1035 m_path = m_path.concat("/" + p_addToPath);
1036 }
1037 }
1038 }
1039
1040 /**
1041 * Set the query string for this URI. A non-null value is valid only
1042 * if this is an URI conforming to the generic URI syntax and
1043 * the path value is not null.
1044 *
1045 * @param p_queryString the query string for this URI
1046 *
1047 * @exception MalformedURIException if p_queryString is not null and this
1048 * URI does not conform to the generic
1049 * URI syntax or if the path is null
1050 */
1051 public void setQueryString(String p_queryString) throws MalformedURIException {
1052 if (p_queryString == null) {
1053 m_queryString = null;
1054 }
1055 else if (!isGenericURI()) {
1056 throw new MalformedURIException(
1057 "Query string can only be set for a generic URI!");
1058 }
1059 else if (getPath() == null) {
1060 throw new MalformedURIException(
1061 "Query string cannot be set when path is null!");
1062 }
1063 else if (!isURIString(p_queryString)) {
1064 throw new MalformedURIException(
1065 "Query string contains invalid character!");
1066 }
1067 else {
1068 m_queryString = p_queryString;
1069 }
1070 }
1071
1072 /**
1073 * Set the fragment for this URI. A non-null value is valid only
1074 * if this is a URI conforming to the generic URI syntax and
1075 * the path value is not null.
1076 *
1077 * @param p_fragment the fragment for this URI
1078 *
1079 * @exception MalformedURIException if p_fragment is not null and this
1080 * URI does not conform to the generic
1081 * URI syntax or if the path is null
1082 */
1083 public void setFragment(String p_fragment) throws MalformedURIException {
1084 if (p_fragment == null) {
1085 m_fragment = null;
1086 }
1087 else if (!isGenericURI()) {
1088 throw new MalformedURIException(
1089 "Fragment can only be set for a generic URI!");
1090 }
1091 else if (getPath() == null) {
1092 throw new MalformedURIException(
1093 "Fragment cannot be set when path is null!");
1094 }
1095 else if (!isURIString(p_fragment)) {
1096 throw new MalformedURIException(
1097 "Fragment contains invalid character!");
1098 }
1099 else {
1100 m_fragment = p_fragment;
1101 }
1102 }
1103
1104 /**
1105 * Determines if the passed-in Object is equivalent to this URI.
1106 *
1107 * @param p_test the Object to test for equality.
1108 *
1109 * @return true if p_test is a URI with all values equal to this
1110 * URI, false otherwise
1111 */
1112 public boolean equals(Object p_test) {
1113 if (p_test instanceof JaxmURI) {
1114 JaxmURI testURI = (JaxmURI) p_test;
1115 if (((m_scheme == null && testURI.m_scheme == null) ||
1116 (m_scheme != null && testURI.m_scheme != null &&
1117 m_scheme.equals(testURI.m_scheme))) &&
1118 ((m_userinfo == null && testURI.m_userinfo == null) ||
1119 (m_userinfo != null && testURI.m_userinfo != null &&
1120 m_userinfo.equals(testURI.m_userinfo))) &&
1121 ((m_host == null && testURI.m_host == null) ||
1122 (m_host != null && testURI.m_host != null &&
1123 m_host.equals(testURI.m_host))) &&
1124 m_port == testURI.m_port &&
1125 ((m_path == null && testURI.m_path == null) ||
1126 (m_path != null && testURI.m_path != null &&
1127 m_path.equals(testURI.m_path))) &&
1128 ((m_queryString == null && testURI.m_queryString == null) ||
1129 (m_queryString != null && testURI.m_queryString != null &&
1130 m_queryString.equals(testURI.m_queryString))) &&
1131 ((m_fragment == null && testURI.m_fragment == null) ||
1132 (m_fragment != null && testURI.m_fragment != null &&
1133 m_fragment.equals(testURI.m_fragment)))) {
1134 return true;
1135 }
1136 }
1137 return false;
1138 }
1139
1140 public int hashCode() {
1141 // No members safe to use, just default to a constant.
1142 return 153214;
1143 }
1144
1145 /**
1146 * Get the URI as a string specification. See RFC 2396 Section 5.2.
1147 *
1148 * @return the URI string specification
1149 */
1150 public String toString() {
1151 StringBuffer uriSpecString = new StringBuffer();
1152
1153 if (m_scheme != null) {
1154 uriSpecString.append(m_scheme);
1155 uriSpecString.append(':');
1156 }
1157 uriSpecString.append(getSchemeSpecificPart());
1158 return uriSpecString.toString();
1159 }
1160
1161 /**
1162 * Get the indicator as to whether this URI uses the "generic URI"
1163 * syntax.
1164 *
1165 * @return true if this URI uses the "generic URI" syntax, false
1166 * otherwise
1167 */
1168 public boolean isGenericURI() {
1169 // presence of the host (whether valid or empty) means
1170 // double-slashes which means generic uri
1171 return (m_host != null);
1172 }
1173
1174 /**
1175 * Determine whether a scheme conforms to the rules for a scheme name.
1176 * A scheme is conformant if it starts with an alphanumeric, and
1177 * contains only alphanumerics, '+','-' and '.'.
1178 *
1179 * @return true if the scheme is conformant, false otherwise
1180 */
1181 public static boolean isConformantSchemeName(String p_scheme) {
1182 if (p_scheme == null || p_scheme.trim().length() == 0) {
1183 return false;
1184 }
1185
1186 if (!isAlpha(p_scheme.charAt(0))) {
1187 return false;
1188 }
1189
1190 char testChar;
1191 for (int i = 1; i < p_scheme.length(); i++) {
1192 testChar = p_scheme.charAt(i);
1193 if (!isAlphanum(testChar) &&
1194 SCHEME_CHARACTERS.indexOf(testChar) == -1) {
1195 return false;
1196 }
1197 }
1198
1199 return true;
1200 }
1201
1202 /**
1203 * Determine whether a string is syntactically capable of representing
1204 * a valid IPv4 address or the domain name of a network host. A valid
1205 * IPv4 address consists of four decimal digit groups separated by a
1206 * '.'. A hostname consists of domain labels (each of which must
1207 * begin and end with an alphanumeric but may contain '-') separated
1208 & by a '.'. See RFC 2396 Section 3.2.2.
1209 *
1210 * @return true if the string is a syntactically valid IPv4 address
1211 * or hostname
1212 */
1213 public static boolean isWellFormedAddress(String p_address) {
1214 if (p_address == null) {
1215 return false;
1216 }
1217
1218 String address = p_address.trim();
1219 int addrLength = address.length();
1220 if (addrLength == 0 || addrLength > 255) {
1221 return false;
1222 }
1223
1224 if (address.startsWith(".") || address.startsWith("-")) {
1225 return false;
1226 }
1227
1228 // rightmost domain label starting with digit indicates IP address
1229 // since top level domain label can only start with an alpha
1230 // see RFC 2396 Section 3.2.2
1231 int index = address.lastIndexOf('.');
1232 if (address.endsWith(".")) {
1233 index = address.substring(0, index).lastIndexOf('.');
1234 }
1235
1236 if (index+1 < addrLength && isDigit(p_address.charAt(index+1))) {
1237 char testChar;
1238 int numDots = 0;
1239
1240 // make sure that 1) we see only digits and dot separators, 2) that
1241 // any dot separator is preceded and followed by a digit and
1242 // 3) that we find 3 dots
1243 for (int i = 0; i < addrLength; i++) {
1244 testChar = address.charAt(i);
1245 if (testChar == '.') {
1246 if (!isDigit(address.charAt(i-1)) ||
1247 (i+1 < addrLength && !isDigit(address.charAt(i+1)))) {
1248 return false;
1249 }
1250 numDots++;
1251 }
1252 else if (!isDigit(testChar)) {
1253 return false;
1254 }
1255 }
1256 if (numDots != 3) {
1257 return false;
1258 }
1259 }
1260 else {
1261 // domain labels can contain alphanumerics and '-"
1262 // but must start and end with an alphanumeric
1263 char testChar;
1264
1265 for (int i = 0; i < addrLength; i++) {
1266 testChar = address.charAt(i);
1267 if (testChar == '.') {
1268 if (!isAlphanum(address.charAt(i-1))) {
1269 return false;
1270 }
1271 if (i+1 < addrLength && !isAlphanum(address.charAt(i+1))) {
1272 return false;
1273 }
1274 }
1275 else if (!isAlphanum(testChar) && testChar != '-') {
1276 return false;
1277 }
1278 }
1279 }
1280 return true;
1281 }
1282
1283
1284 /**
1285 * Determine whether a char is a digit.
1286 *
1287 * @return true if the char is betweeen '0' and '9', false otherwise
1288 */
1289 private static boolean isDigit(char p_char) {
1290 return p_char >= '0' && p_char <= '9';
1291 }
1292
1293 /**
1294 * Determine whether a character is a hexadecimal character.
1295 *
1296 * @return true if the char is betweeen '0' and '9', 'a' and 'f'
1297 * or 'A' and 'F', false otherwise
1298 */
1299 private static boolean isHex(char p_char) {
1300 return (isDigit(p_char) ||
1301 (p_char >= 'a' && p_char <= 'f') ||
1302 (p_char >= 'A' && p_char <= 'F'));
1303 }
1304
1305 /**
1306 * Determine whether a char is an alphabetic character: a-z or A-Z
1307 *
1308 * @return true if the char is alphabetic, false otherwise
1309 */
1310 private static boolean isAlpha(char p_char) {
1311 return ((p_char >= 'a' && p_char <= 'z') ||
1312 (p_char >= 'A' && p_char <= 'Z' ));
1313 }
1314
1315 /**
1316 * Determine whether a char is an alphanumeric: 0-9, a-z or A-Z
1317 *
1318 * @return true if the char is alphanumeric, false otherwise
1319 */
1320 private static boolean isAlphanum(char p_char) {
1321 return (isAlpha(p_char) || isDigit(p_char));
1322 }
1323
1324 /**
1325 * Determine whether a character is a reserved character:
1326 * ';', '/', '?', ':', '@', '&', '=', '+', '$' or ','
1327 *
1328 * @return true if the string contains any reserved characters
1329 */
1330 private static boolean isReservedCharacter(char p_char) {
1331 return RESERVED_CHARACTERS.indexOf(p_char) != -1;
1332 }
1333
1334 /**
1335 * Determine whether a char is an unreserved character.
1336 *
1337 * @return true if the char is unreserved, false otherwise
1338 */
1339 private static boolean isUnreservedCharacter(char p_char) {
1340 return (isAlphanum(p_char) ||
1341 MARK_CHARACTERS.indexOf(p_char) != -1);
1342 }
1343
1344 /**
1345 * Determine whether a given string contains only URI characters (also
1346 * called "uric" in RFC 2396). uric consist of all reserved
1347 * characters, unreserved characters and escaped characters.
1348 *
1349 * @return true if the string is comprised of uric, false otherwise
1350 */
1351 private static boolean isURIString(String p_uric) {
1352 if (p_uric == null) {
1353 return false;
1354 }
1355 int end = p_uric.length();
1356 char testChar = '\0';
1357 for (int i = 0; i < end; i++) {
1358 testChar = p_uric.charAt(i);
1359 if (testChar == '%') {
1360 if (i+2 >= end ||
1361 !isHex(p_uric.charAt(i+1)) ||
1362 !isHex(p_uric.charAt(i+2))) {
1363 return false;
1364 }
1365 else {
1366 i += 2;
1367 continue;
1368 }
1369 }
1370 if (isReservedCharacter(testChar) ||
1371 isUnreservedCharacter(testChar)) {
1372 continue;
1373 }
1374 else {
1375 return false;
1376 }
1377 }
1378 return true;
1379 }
1380 }