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