1 /*
2 * Copyright (c) 1997, 2012, 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 /**
133 * Construct a new and uninitialized URI.
134 */
135 public JaxmURI() {
136 }
137
138 /**
139 * Construct a new URI from another URI. All fields for this URI are
140 * set equal to the fields of the URI passed in.
141 *
142 * @param p_other the URI to copy (cannot be null)
143 */
144 public JaxmURI(JaxmURI p_other) {
145 initialize(p_other);
146 }
147
148 /**
149 * Construct a new URI from a URI specification string. If the
150 * specification follows the "generic URI" syntax, (two slashes
151 * following the first colon), the specification will be parsed
152 * accordingly - setting the scheme, userinfo, host,port, path, query
153 * string and fragment fields as necessary. If the specification does
154 * not follow the "generic URI" syntax, the specification is parsed
155 * into a scheme and scheme-specific part (stored as the path) only.
156 *
157 * @param p_uriSpec the URI specification string (cannot be null or
158 * empty)
159 *
160 * @exception MalformedURIException if p_uriSpec violates any syntax
161 * rules
162 */
163 public JaxmURI(String p_uriSpec) throws MalformedURIException {
164 this((JaxmURI)null, p_uriSpec);
165 }
166
167 /**
168 * Construct a new URI from a base URI and a URI specification string.
169 * The URI specification string may be a relative URI.
170 *
171 * @param p_base the base URI (cannot be null if p_uriSpec is null or
172 * empty)
173 * @param p_uriSpec the URI specification string (cannot be null or
174 * empty if p_base is null)
175 *
176 * @exception MalformedURIException if p_uriSpec violates any syntax
177 * rules
178 */
179 public JaxmURI(JaxmURI p_base, String p_uriSpec) throws MalformedURIException {
180 initialize(p_base, p_uriSpec);
181 }
182
183 /**
184 * Construct a new URI that does not follow the generic URI syntax.
185 * Only the scheme and scheme-specific part (stored as the path) are
186 * initialized.
187 *
188 * @param p_scheme the URI scheme (cannot be null or empty)
189 * @param p_schemeSpecificPart the scheme-specific part (cannot be
190 * null or empty)
191 *
192 * @exception MalformedURIException if p_scheme violates any
193 * syntax rules
194 */
195 public JaxmURI(String p_scheme, String p_schemeSpecificPart)
196 throws MalformedURIException {
197 if (p_scheme == null || p_scheme.trim().length() == 0) {
198 throw new MalformedURIException(
199 "Cannot construct URI with null/empty scheme!");
200 }
201 if (p_schemeSpecificPart == null ||
202 p_schemeSpecificPart.trim().length() == 0) {
203 throw new MalformedURIException(
204 "Cannot construct URI with null/empty scheme-specific part!");
205 }
206 setScheme(p_scheme);
207 setPath(p_schemeSpecificPart);
208 }
209
210 /**
211 * Construct a new URI that follows the generic URI syntax from its
212 * component parts. Each component is validated for syntax and some
213 * basic semantic checks are performed as well. See the individual
214 * setter methods for specifics.
215 *
216 * @param p_scheme the URI scheme (cannot be null or empty)
217 * @param p_host the hostname or IPv4 address for the URI
218 * @param p_path the URI path - if the path contains '?' or '#',
219 * then the query string and/or fragment will be
220 * set from the path; however, if the query and
221 * fragment are specified both in the path and as
222 * separate parameters, an exception is thrown
223 * @param p_queryString the URI query string (cannot be specified
224 * if path is null)
225 * @param p_fragment the URI fragment (cannot be specified if path
226 * is null)
227 *
228 * @exception MalformedURIException if any of the parameters violates
229 * syntax rules or semantic rules
230 */
231 public JaxmURI(String p_scheme, String p_host, String p_path,
232 String p_queryString, String p_fragment)
233 throws MalformedURIException {
234 this(p_scheme, null, p_host, -1, p_path, p_queryString, p_fragment);
235 }
236
237 /**
238 * Construct a new URI that follows the generic URI syntax from its
239 * component parts. Each component is validated for syntax and some
240 * basic semantic checks are performed as well. See the individual
241 * setter methods for specifics.
242 *
243 * @param p_scheme the URI scheme (cannot be null or empty)
244 * @param p_userinfo the URI userinfo (cannot be specified if host
245 * is null)
246 * @param p_host the hostname or IPv4 address for the URI
247 * @param p_port the URI port (may be -1 for "unspecified"; cannot
248 * be specified if host is null)
249 * @param p_path the URI path - if the path contains '?' or '#',
250 * then the query string and/or fragment will be
251 * set from the path; however, if the query and
252 * fragment are specified both in the path and as
253 * separate parameters, an exception is thrown
254 * @param p_queryString the URI query string (cannot be specified
255 * if path is null)
256 * @param p_fragment the URI fragment (cannot be specified if path
257 * is null)
258 *
259 * @exception MalformedURIException if any of the parameters violates
260 * syntax rules or semantic rules
261 */
262 public JaxmURI(String p_scheme, String p_userinfo,
263 String p_host, int p_port, String p_path,
264 String p_queryString, String p_fragment)
265 throws MalformedURIException {
266 if (p_scheme == null || p_scheme.trim().length() == 0) {
267 throw new MalformedURIException("Scheme is required!");
268 }
269
270 if (p_host == null) {
271 if (p_userinfo != null) {
272 throw new MalformedURIException(
273 "Userinfo may not be specified if host is not specified!");
274 }
275 if (p_port != -1) {
276 throw new MalformedURIException(
277 "Port may not be specified if host is not specified!");
278 }
279 }
280
281 if (p_path != null) {
282 if (p_path.indexOf('?') != -1 && p_queryString != null) {
283 throw new MalformedURIException(
284 "Query string cannot be specified in path and query string!");
285 }
286
287 if (p_path.indexOf('#') != -1 && p_fragment != null) {
288 throw new MalformedURIException(
289 "Fragment cannot be specified in both the path and fragment!");
290 }
291 }
292
293 setScheme(p_scheme);
294 setHost(p_host);
295 setPort(p_port);
296 setUserinfo(p_userinfo);
297 setPath(p_path);
298 setQueryString(p_queryString);
299 setFragment(p_fragment);
300 }
301
302 /**
303 * Initialize all fields of this URI from another URI.
304 *
305 * @param p_other the URI to copy (cannot be null)
306 */
307 private void initialize(JaxmURI p_other) {
308 m_scheme = p_other.getScheme();
309 m_userinfo = p_other.getUserinfo();
310 m_host = p_other.getHost();
311 m_port = p_other.getPort();
312 m_path = p_other.getPath();
313 m_queryString = p_other.getQueryString();
314 m_fragment = p_other.getFragment();
315 }
316
317 /**
318 * Initializes this URI from a base URI and a URI specification string.
319 * See RFC 2396 Section 4 and Appendix B for specifications on parsing
320 * the URI and Section 5 for specifications on resolving relative URIs
321 * and relative paths.
322 *
323 * @param p_base the base URI (may be null if p_uriSpec is an absolute
324 * URI)
325 * @param p_uriSpec the URI spec string which may be an absolute or
326 * relative URI (can only be null/empty if p_base
327 * is not null)
328 *
329 * @exception MalformedURIException if p_base is null and p_uriSpec
330 * is not an absolute URI or if
331 * p_uriSpec violates syntax rules
332 */
333 private void initialize(JaxmURI p_base, String p_uriSpec)
334 throws MalformedURIException {
335 if (p_base == null &&
336 (p_uriSpec == null || p_uriSpec.trim().length() == 0)) {
337 throw new MalformedURIException(
338 "Cannot initialize URI with empty parameters.");
339 }
340
341 // just make a copy of the base if spec is empty
342 if (p_uriSpec == null || p_uriSpec.trim().length() == 0) {
343 initialize(p_base);
344 return;
345 }
346
347 String uriSpec = p_uriSpec.trim();
348 int uriSpecLen = uriSpec.length();
349 int index = 0;
350
351 // Check for scheme, which must be before `/'. Also handle names with
352 // DOS drive letters ('D:'), so 1-character schemes are not allowed.
353 int colonIdx = uriSpec.indexOf(':');
354 int slashIdx = uriSpec.indexOf('/');
355 if ((colonIdx < 2) || (colonIdx > slashIdx && slashIdx != -1)) {
356 int fragmentIdx = uriSpec.indexOf('#');
357 // A standalone base is a valid URI according to spec
358 if (p_base == null && fragmentIdx != 0 ) {
359 throw new MalformedURIException("No scheme found in URI.");
360 }
361 }
362 else {
363 initializeScheme(uriSpec);
364 index = m_scheme.length()+1;
365 }
366
367 // two slashes means generic URI syntax, so we get the authority
368 if (((index+1) < uriSpecLen) &&
369 (uriSpec.substring(index).startsWith("//"))) {
370 index += 2;
371 int startPos = index;
372
373 // get authority - everything up to path, query or fragment
374 char testChar = '\0';
375 while (index < uriSpecLen) {
376 testChar = uriSpec.charAt(index);
377 if (testChar == '/' || testChar == '?' || testChar == '#') {
378 break;
379 }
380 index++;
381 }
382
383 // if we found authority, parse it out, otherwise we set the
384 // host to empty string
385 if (index > startPos) {
386 initializeAuthority(uriSpec.substring(startPos, index));
387 }
388 else {
389 m_host = "";
390 }
391 }
392
393 initializePath(uriSpec.substring(index));
394
395 // Resolve relative URI to base URI - see RFC 2396 Section 5.2
396 // In some cases, it might make more sense to throw an exception
397 // (when scheme is specified is the string spec and the base URI
398 // is also specified, for example), but we're just following the
399 // RFC specifications
400 if (p_base != null) {
401
402 // check to see if this is the current doc - RFC 2396 5.2 #2
403 // note that this is slightly different from the RFC spec in that
404 // we don't include the check for query string being null
405 // - this handles cases where the urispec is just a query
406 // string or a fragment (e.g. "?y" or "#s") -
407 // see <http://www.ics.uci.edu/~fielding/url/test1.html> which
408 // identified this as a bug in the RFC
409 if (m_path.length() == 0 && m_scheme == null &&
410 m_host == null) {
411 m_scheme = p_base.getScheme();
412 m_userinfo = p_base.getUserinfo();
413 m_host = p_base.getHost();
414 m_port = p_base.getPort();
415 m_path = p_base.getPath();
416
417 if (m_queryString == null) {
418 m_queryString = p_base.getQueryString();
419 }
420 return;
421 }
422
423 // check for scheme - RFC 2396 5.2 #3
424 // if we found a scheme, it means absolute URI, so we're done
425 if (m_scheme == null) {
426 m_scheme = p_base.getScheme();
427 }
428 else {
429 return;
430 }
431
432 // check for authority - RFC 2396 5.2 #4
433 // if we found a host, then we've got a network path, so we're done
434 if (m_host == null) {
435 m_userinfo = p_base.getUserinfo();
436 m_host = p_base.getHost();
437 m_port = p_base.getPort();
438 }
439 else {
440 return;
441 }
442
443 // check for absolute path - RFC 2396 5.2 #5
444 if (m_path.length() > 0 &&
445 m_path.startsWith("/")) {
446 return;
447 }
448
449 // if we get to this point, we need to resolve relative path
450 // RFC 2396 5.2 #6
451 String path = "";
452 String basePath = p_base.getPath();
453
454 // 6a - get all but the last segment of the base URI path
455 if (basePath != null) {
456 int lastSlash = basePath.lastIndexOf('/');
457 if (lastSlash != -1) {
458 path = basePath.substring(0, lastSlash+1);
459 }
460 }
461
462 // 6b - append the relative URI path
463 path = path.concat(m_path);
464
465 // 6c - remove all "./" where "." is a complete path segment
466 index = -1;
467 while ((index = path.indexOf("/./")) != -1) {
468 path = path.substring(0, index+1).concat(path.substring(index+3));
469 }
470
471 // 6d - remove "." if path ends with "." as a complete path segment
472 if (path.endsWith("/.")) {
473 path = path.substring(0, path.length()-1);
474 }
475
476 // 6e - remove all "<segment>/../" where "<segment>" is a complete
477 // path segment not equal to ".."
478 index = 1;
479 int segIndex = -1;
480 String tempString = null;
481
482 while ((index = path.indexOf("/../", index)) > 0) {
483 tempString = path.substring(0, path.indexOf("/../"));
484 segIndex = tempString.lastIndexOf('/');
485 if (segIndex != -1) {
486 if (!tempString.substring(segIndex++).equals("..")) {
487 path = path.substring(0, segIndex).concat(path.substring(index+4));
488 }
489 else
490 index += 4;
491 }
492 else
493 index += 4;
494 }
495
496 // 6f - remove ending "<segment>/.." where "<segment>" is a
497 // complete path segment
498 if (path.endsWith("/..")) {
499 tempString = path.substring(0, path.length()-3);
500 segIndex = tempString.lastIndexOf('/');
501 if (segIndex != -1) {
502 path = path.substring(0, segIndex+1);
503 }
504 }
505 m_path = path;
506 }
507 }
508
509 /**
510 * Initialize the scheme for this URI from a URI string spec.
511 *
512 * @param p_uriSpec the URI specification (cannot be null)
513 *
514 * @exception MalformedURIException if URI does not have a conformant
515 * scheme
516 */
517 private void initializeScheme(String p_uriSpec)
518 throws MalformedURIException {
519 int uriSpecLen = p_uriSpec.length();
520 int index = 0;
521 String scheme = null;
522 char testChar = '\0';
523
524 while (index < uriSpecLen) {
525 testChar = p_uriSpec.charAt(index);
526 if (testChar == ':' || testChar == '/' ||
527 testChar == '?' || testChar == '#') {
528 break;
529 }
530 index++;
531 }
532 scheme = p_uriSpec.substring(0, index);
533
534 if (scheme.length() == 0) {
535 throw new MalformedURIException("No scheme found in URI.");
536 }
537 else {
538 setScheme(scheme);
539 }
540 }
541
542 /**
543 * Initialize the authority (userinfo, host and port) for this
544 * URI from a URI string spec.
545 *
546 * @param p_uriSpec the URI specification (cannot be null)
547 *
548 * @exception MalformedURIException if p_uriSpec violates syntax rules
549 */
550 private void initializeAuthority(String p_uriSpec)
551 throws MalformedURIException {
552 int index = 0;
553 int start = 0;
554 int end = p_uriSpec.length();
555 char testChar = '\0';
556 String userinfo = null;
557
558 // userinfo is everything up @
559 if (p_uriSpec.indexOf('@', start) != -1) {
560 while (index < end) {
561 testChar = p_uriSpec.charAt(index);
562 if (testChar == '@') {
563 break;
564 }
565 index++;
566 }
567 userinfo = p_uriSpec.substring(start, index);
568 index++;
569 }
570
571 // host is everything up to ':'
572 String host = null;
573 start = index;
574 while (index < end) {
575 testChar = p_uriSpec.charAt(index);
576 if (testChar == ':') {
577 break;
578 }
579 index++;
580 }
581 host = p_uriSpec.substring(start, index);
582 int port = -1;
583 if (host.length() > 0) {
584 // port
585 if (testChar == ':') {
586 index++;
587 start = index;
588 while (index < end) {
589 index++;
590 }
591 String portStr = p_uriSpec.substring(start, index);
592 if (portStr.length() > 0) {
593 for (int i = 0; i < portStr.length(); i++) {
594 if (!isDigit(portStr.charAt(i))) {
595 throw new MalformedURIException(
596 portStr +
597 " is invalid. Port should only contain digits!");
598 }
599 }
600 try {
601 port = Integer.parseInt(portStr);
602 }
603 catch (NumberFormatException nfe) {
604 // can't happen
605 }
606 }
607 }
608 }
609 setHost(host);
610 setPort(port);
611 setUserinfo(userinfo);
612 }
613
614 /**
615 * Initialize the path for this URI from a URI string spec.
616 *
617 * @param p_uriSpec the URI specification (cannot be null)
618 *
619 * @exception MalformedURIException if p_uriSpec violates syntax rules
620 */
621 private void initializePath(String p_uriSpec)
622 throws MalformedURIException {
623 if (p_uriSpec == null) {
624 throw new MalformedURIException(
625 "Cannot initialize path from null string!");
626 }
627
628 int index = 0;
629 int start = 0;
630 int end = p_uriSpec.length();
631 char testChar = '\0';
632
633 // path - everything up to query string or fragment
634 while (index < end) {
635 testChar = p_uriSpec.charAt(index);
636 if (testChar == '?' || testChar == '#') {
637 break;
638 }
639 // check for valid escape sequence
640 if (testChar == '%') {
641 if (index+2 >= end ||
642 !isHex(p_uriSpec.charAt(index+1)) ||
643 !isHex(p_uriSpec.charAt(index+2))) {
644 throw new MalformedURIException(
645 "Path contains invalid escape sequence!");
646 }
647 }
648 else if (!isReservedCharacter(testChar) &&
649 !isUnreservedCharacter(testChar)) {
650 throw new MalformedURIException(
651 "Path contains invalid character: " + testChar);
652 }
653 index++;
654 }
655 m_path = p_uriSpec.substring(start, index);
656
657 // query - starts with ? and up to fragment or end
658 if (testChar == '?') {
659 index++;
660 start = index;
661 while (index < end) {
662 testChar = p_uriSpec.charAt(index);
663 if (testChar == '#') {
664 break;
665 }
666 if (testChar == '%') {
667 if (index+2 >= end ||
668 !isHex(p_uriSpec.charAt(index+1)) ||
669 !isHex(p_uriSpec.charAt(index+2))) {
670 throw new MalformedURIException(
671 "Query string contains invalid escape sequence!");
672 }
673 }
674 else if (!isReservedCharacter(testChar) &&
675 !isUnreservedCharacter(testChar)) {
676 throw new MalformedURIException(
677 "Query string contains invalid character:" + testChar);
678 }
679 index++;
680 }
681 m_queryString = p_uriSpec.substring(start, index);
682 }
683
684 // fragment - starts with #
685 if (testChar == '#') {
686 index++;
687 start = index;
688 while (index < end) {
689 testChar = p_uriSpec.charAt(index);
690
691 if (testChar == '%') {
692 if (index+2 >= end ||
693 !isHex(p_uriSpec.charAt(index+1)) ||
694 !isHex(p_uriSpec.charAt(index+2))) {
695 throw new MalformedURIException(
696 "Fragment contains invalid escape sequence!");
697 }
698 }
699 else if (!isReservedCharacter(testChar) &&
700 !isUnreservedCharacter(testChar)) {
701 throw new MalformedURIException(
702 "Fragment contains invalid character:"+testChar);
703 }
704 index++;
705 }
706 m_fragment = p_uriSpec.substring(start, index);
707 }
708 }
709
710 /**
711 * Get the scheme for this URI.
712 *
713 * @return the scheme for this URI
714 */
715 public String getScheme() {
716 return m_scheme;
717 }
718
719 /**
720 * Get the scheme-specific part for this URI (everything following the
721 * scheme and the first colon). See RFC 2396 Section 5.2 for spec.
722 *
723 * @return the scheme-specific part for this URI
724 */
725 public String getSchemeSpecificPart() {
726 StringBuilder schemespec = new StringBuilder();
727
728 if (m_userinfo != null || m_host != null || m_port != -1) {
729 schemespec.append("//");
730 }
731
732 if (m_userinfo != null) {
733 schemespec.append(m_userinfo);
734 schemespec.append('@');
735 }
736
737 if (m_host != null) {
738 schemespec.append(m_host);
739 }
740
741 if (m_port != -1) {
742 schemespec.append(':');
743 schemespec.append(m_port);
744 }
745
746 if (m_path != null) {
747 schemespec.append((m_path));
748 }
749
750 if (m_queryString != null) {
751 schemespec.append('?');
752 schemespec.append(m_queryString);
753 }
754
755 if (m_fragment != null) {
756 schemespec.append('#');
757 schemespec.append(m_fragment);
758 }
759
760 return schemespec.toString();
761 }
762
763 /**
764 * Get the userinfo for this URI.
765 *
766 * @return the userinfo for this URI (null if not specified).
767 */
768 public String getUserinfo() {
769 return m_userinfo;
770 }
771
772 /**
773 * Get the host for this URI.
774 *
775 * @return the host for this URI (null if not specified).
776 */
777 public String getHost() {
778 return m_host;
779 }
780
781 /**
782 * Get the port for this URI.
783 *
784 * @return the port for this URI (-1 if not specified).
785 */
786 public int getPort() {
787 return m_port;
788 }
789
790 /**
791 * Get the path for this URI (optionally with the query string and
792 * fragment).
793 *
794 * @param p_includeQueryString if true (and query string is not null),
795 * then a "?" followed by the query string
796 * will be appended
797 * @param p_includeFragment if true (and fragment is not null),
798 * then a "#" followed by the fragment
799 * will be appended
800 *
801 * @return the path for this URI possibly including the query string
802 * and fragment
803 */
804 public String getPath(boolean p_includeQueryString,
805 boolean p_includeFragment) {
806 StringBuilder pathString = new StringBuilder(m_path);
807
808 if (p_includeQueryString && m_queryString != null) {
809 pathString.append('?');
810 pathString.append(m_queryString);
811 }
812
813 if (p_includeFragment && m_fragment != null) {
814 pathString.append('#');
815 pathString.append(m_fragment);
816 }
817 return pathString.toString();
818 }
819
820 /**
821 * Get the path for this URI. Note that the value returned is the path
822 * only and does not include the query string or fragment.
823 *
824 * @return the path for this URI.
825 */
826 public String getPath() {
827 return m_path;
828 }
829
830 /**
831 * Get the query string for this URI.
832 *
833 * @return the query string for this URI. Null is returned if there
834 * was no "?" in the URI spec, empty string if there was a
835 * "?" but no query string following it.
836 */
837 public String getQueryString() {
838 return m_queryString;
839 }
840
841 /**
842 * Get the fragment for this URI.
843 *
844 * @return the fragment for this URI. Null is returned if there
845 * was no "#" in the URI spec, empty string if there was a
846 * "#" but no fragment following it.
847 */
848 public String getFragment() {
849 return m_fragment;
850 }
851
852 /**
853 * Set the scheme for this URI. The scheme is converted to lowercase
854 * before it is set.
855 *
856 * @param p_scheme the scheme for this URI (cannot be null)
857 *
858 * @exception MalformedURIException if p_scheme is not a conformant
859 * scheme name
860 */
861 public void setScheme(String p_scheme) throws MalformedURIException {
862 if (p_scheme == null) {
863 throw new MalformedURIException(
864 "Cannot set scheme from null string!");
865 }
866 if (!isConformantSchemeName(p_scheme)) {
867 throw new MalformedURIException("The scheme is not conformant.");
868 }
869
870 m_scheme = p_scheme.toLowerCase();
871 }
872
873 /**
874 * Set the userinfo for this URI. If a non-null value is passed in and
875 * the host value is null, then an exception is thrown.
876 *
877 * @param p_userinfo the userinfo for this URI
878 *
879 * @exception MalformedURIException if p_userinfo contains invalid
880 * characters
881 */
882 public void setUserinfo(String p_userinfo) throws MalformedURIException {
883 if (p_userinfo == null) {
884 m_userinfo = null;
885 }
886 else {
887 if (m_host == null) {
888 throw new MalformedURIException(
889 "Userinfo cannot be set when host is null!");
890 }
891
892 // userinfo can contain alphanumerics, mark characters, escaped
893 // and ';',':','&','=','+','$',','
894 int index = 0;
895 int end = p_userinfo.length();
896 char testChar = '\0';
897 while (index < end) {
898 testChar = p_userinfo.charAt(index);
899 if (testChar == '%') {
900 if (index+2 >= end ||
901 !isHex(p_userinfo.charAt(index+1)) ||
902 !isHex(p_userinfo.charAt(index+2))) {
903 throw new MalformedURIException(
904 "Userinfo contains invalid escape sequence!");
905 }
906 }
907 else if (!isUnreservedCharacter(testChar) &&
908 USERINFO_CHARACTERS.indexOf(testChar) == -1) {
909 throw new MalformedURIException(
910 "Userinfo contains invalid character:"+testChar);
911 }
912 index++;
913 }
914 }
915 m_userinfo = p_userinfo;
916 }
917
918 /**
919 * Set the host for this URI. If null is passed in, the userinfo
920 * field is also set to null and the port is set to -1.
921 *
922 * @param p_host the host for this URI
923 *
924 * @exception MalformedURIException if p_host is not a valid IP
925 * address or DNS hostname.
926 */
927 public void setHost(String p_host) throws MalformedURIException {
928 if (p_host == null || p_host.trim().length() == 0) {
929 m_host = p_host;
930 m_userinfo = null;
931 m_port = -1;
932 }
933 else if (!isWellFormedAddress(p_host)) {
934 throw new MalformedURIException("Host is not a well formed address!");
935 }
936 m_host = p_host;
937 }
938
939 /**
940 * Set the port for this URI. -1 is used to indicate that the port is
941 * not specified, otherwise valid port numbers are between 0 and 65535.
942 * If a valid port number is passed in and the host field is null,
943 * an exception is thrown.
944 *
945 * @param p_port the port number for this URI
946 *
947 * @exception MalformedURIException if p_port is not -1 and not a
948 * valid port number
949 */
950 public void setPort(int p_port) throws MalformedURIException {
951 if (p_port >= 0 && p_port <= 65535) {
952 if (m_host == null) {
953 throw new MalformedURIException(
954 "Port cannot be set when host is null!");
955 }
956 }
957 else if (p_port != -1) {
958 throw new MalformedURIException("Invalid port number!");
959 }
960 m_port = p_port;
961 }
962
963 /**
964 * Set the path for this URI. If the supplied path is null, then the
965 * query string and fragment are set to null as well. If the supplied
966 * path includes a query string and/or fragment, these fields will be
967 * parsed and set as well. Note that, for URIs following the "generic
968 * URI" syntax, the path specified should start with a slash.
969 * For URIs that do not follow the generic URI syntax, this method
970 * sets the scheme-specific part.
971 *
972 * @param p_path the path for this URI (may be null)
973 *
974 * @exception MalformedURIException if p_path contains invalid
975 * characters
976 */
977 public void setPath(String p_path) throws MalformedURIException {
978 if (p_path == null) {
979 m_path = null;
980 m_queryString = null;
981 m_fragment = null;
982 }
983 else {
984 initializePath(p_path);
985 }
986 }
987
988 /**
989 * Append to the end of the path of this URI. If the current path does
990 * not end in a slash and the path to be appended does not begin with
991 * a slash, a slash will be appended to the current path before the
992 * new segment is added. Also, if the current path ends in a slash
993 * and the new segment begins with a slash, the extra slash will be
994 * removed before the new segment is appended.
995 *
996 * @param p_addToPath the new segment to be added to the current path
997 *
998 * @exception MalformedURIException if p_addToPath contains syntax
999 * errors
1000 */
1001 public void appendPath(String p_addToPath)
1002 throws MalformedURIException {
1003 if (p_addToPath == null || p_addToPath.trim().length() == 0) {
1004 return;
1005 }
1006
1007 if (!isURIString(p_addToPath)) {
1008 throw new MalformedURIException(
1009 "Path contains invalid character!");
1010 }
1011
1012 if (m_path == null || m_path.trim().length() == 0) {
1013 if (p_addToPath.startsWith("/")) {
1014 m_path = p_addToPath;
1015 }
1016 else {
1017 m_path = "/" + p_addToPath;
1018 }
1019 }
1020 else if (m_path.endsWith("/")) {
1021 if (p_addToPath.startsWith("/")) {
1022 m_path = m_path.concat(p_addToPath.substring(1));
1023 }
1024 else {
1025 m_path = m_path.concat(p_addToPath);
1026 }
1027 }
1028 else {
1029 if (p_addToPath.startsWith("/")) {
1030 m_path = m_path.concat(p_addToPath);
1031 }
1032 else {
1033 m_path = m_path.concat("/" + p_addToPath);
1034 }
1035 }
1036 }
1037
1038 /**
1039 * Set the query string for this URI. A non-null value is valid only
1040 * if this is an URI conforming to the generic URI syntax and
1041 * the path value is not null.
1042 *
1043 * @param p_queryString the query string for this URI
1044 *
1045 * @exception MalformedURIException if p_queryString is not null and this
1046 * URI does not conform to the generic
1047 * URI syntax or if the path is null
1048 */
1049 public void setQueryString(String p_queryString) throws MalformedURIException {
1050 if (p_queryString == null) {
1051 m_queryString = null;
1052 }
1053 else if (!isGenericURI()) {
1054 throw new MalformedURIException(
1055 "Query string can only be set for a generic URI!");
1056 }
1057 else if (getPath() == null) {
1058 throw new MalformedURIException(
1059 "Query string cannot be set when path is null!");
1060 }
1061 else if (!isURIString(p_queryString)) {
1062 throw new MalformedURIException(
1063 "Query string contains invalid character!");
1064 }
1065 else {
1066 m_queryString = p_queryString;
1067 }
1068 }
1069
1070 /**
1071 * Set the fragment for this URI. A non-null value is valid only
1072 * if this is a URI conforming to the generic URI syntax and
1073 * the path value is not null.
1074 *
1075 * @param p_fragment the fragment for this URI
1076 *
1077 * @exception MalformedURIException if p_fragment is not null and this
1078 * URI does not conform to the generic
1079 * URI syntax or if the path is null
1080 */
1081 public void setFragment(String p_fragment) throws MalformedURIException {
1082 if (p_fragment == null) {
1083 m_fragment = null;
1084 }
1085 else if (!isGenericURI()) {
1086 throw new MalformedURIException(
1087 "Fragment can only be set for a generic URI!");
1088 }
1089 else if (getPath() == null) {
1090 throw new MalformedURIException(
1091 "Fragment cannot be set when path is null!");
1092 }
1093 else if (!isURIString(p_fragment)) {
1094 throw new MalformedURIException(
1095 "Fragment contains invalid character!");
1096 }
1097 else {
1098 m_fragment = p_fragment;
1099 }
1100 }
1101
1102 /**
1103 * Determines if the passed-in Object is equivalent to this URI.
1104 *
1105 * @param p_test the Object to test for equality.
1106 *
1107 * @return true if p_test is a URI with all values equal to this
1108 * URI, false otherwise
1109 */
1110 public boolean equals(Object p_test) {
1111 if (p_test instanceof JaxmURI) {
1112 JaxmURI testURI = (JaxmURI) p_test;
1113 if (((m_scheme == null && testURI.m_scheme == null) ||
1114 (m_scheme != null && testURI.m_scheme != null &&
1115 m_scheme.equals(testURI.m_scheme))) &&
1116 ((m_userinfo == null && testURI.m_userinfo == null) ||
1117 (m_userinfo != null && testURI.m_userinfo != null &&
1118 m_userinfo.equals(testURI.m_userinfo))) &&
1119 ((m_host == null && testURI.m_host == null) ||
1120 (m_host != null && testURI.m_host != null &&
1121 m_host.equals(testURI.m_host))) &&
1122 m_port == testURI.m_port &&
1123 ((m_path == null && testURI.m_path == null) ||
1124 (m_path != null && testURI.m_path != null &&
1125 m_path.equals(testURI.m_path))) &&
1126 ((m_queryString == null && testURI.m_queryString == null) ||
1127 (m_queryString != null && testURI.m_queryString != null &&
1128 m_queryString.equals(testURI.m_queryString))) &&
1129 ((m_fragment == null && testURI.m_fragment == null) ||
1130 (m_fragment != null && testURI.m_fragment != null &&
1131 m_fragment.equals(testURI.m_fragment)))) {
1132 return true;
1133 }
1134 }
1135 return false;
1136 }
1137
1138 public int hashCode() {
1139 // No members safe to use, just default to a constant.
1140 return 153214;
1141 }
1142
1143 /**
1144 * Get the URI as a string specification. See RFC 2396 Section 5.2.
1145 *
1146 * @return the URI string specification
1147 */
1148 public String toString() {
1149 StringBuilder uriSpecString = new StringBuilder();
1150
1151 if (m_scheme != null) {
1152 uriSpecString.append(m_scheme);
1153 uriSpecString.append(':');
1154 }
1155 uriSpecString.append(getSchemeSpecificPart());
1156 return uriSpecString.toString();
1157 }
1158
1159 /**
1160 * Get the indicator as to whether this URI uses the "generic URI"
1161 * syntax.
1162 *
1163 * @return true if this URI uses the "generic URI" syntax, false
1164 * otherwise
1165 */
1166 public boolean isGenericURI() {
1167 // presence of the host (whether valid or empty) means
1168 // double-slashes which means generic uri
1169 return (m_host != null);
1170 }
1171
1172 /**
1173 * Determine whether a scheme conforms to the rules for a scheme name.
1174 * A scheme is conformant if it starts with an alphanumeric, and
1175 * contains only alphanumerics, '+','-' and '.'.
1176 *
1177 * @return true if the scheme is conformant, false otherwise
1178 */
1179 public static boolean isConformantSchemeName(String p_scheme) {
1180 if (p_scheme == null || p_scheme.trim().length() == 0) {
1181 return false;
1182 }
1183
1184 if (!isAlpha(p_scheme.charAt(0))) {
1185 return false;
1186 }
1187
1188 char testChar;
1189 for (int i = 1; i < p_scheme.length(); i++) {
1190 testChar = p_scheme.charAt(i);
1191 if (!isAlphanum(testChar) &&
1192 SCHEME_CHARACTERS.indexOf(testChar) == -1) {
1193 return false;
1194 }
1195 }
1196
1197 return true;
1198 }
1199
1200 /**
1201 * Determine whether a string is syntactically capable of representing
1202 * a valid IPv4 address or the domain name of a network host. A valid
1203 * IPv4 address consists of four decimal digit groups separated by a
1204 * '.'. A hostname consists of domain labels (each of which must
1205 * begin and end with an alphanumeric but may contain '-') separated
1206 & by a '.'. See RFC 2396 Section 3.2.2.
1207 *
1208 * @return true if the string is a syntactically valid IPv4 address
1209 * or hostname
1210 */
1211 public static boolean isWellFormedAddress(String p_address) {
1212 if (p_address == null) {
1213 return false;
1214 }
1215
1216 String address = p_address.trim();
1217 int addrLength = address.length();
1218 if (addrLength == 0 || addrLength > 255) {
1219 return false;
1220 }
1221
1222 if (address.startsWith(".") || address.startsWith("-")) {
1223 return false;
1224 }
1225
1226 // rightmost domain label starting with digit indicates IP address
1227 // since top level domain label can only start with an alpha
1228 // see RFC 2396 Section 3.2.2
1229 int index = address.lastIndexOf('.');
1230 if (address.endsWith(".")) {
1231 index = address.substring(0, index).lastIndexOf('.');
1232 }
1233
1234 if (index+1 < addrLength && isDigit(p_address.charAt(index+1))) {
1235 char testChar;
1236 int numDots = 0;
1237
1238 // make sure that 1) we see only digits and dot separators, 2) that
1239 // any dot separator is preceded and followed by a digit and
1240 // 3) that we find 3 dots
1241 for (int i = 0; i < addrLength; i++) {
1242 testChar = address.charAt(i);
1243 if (testChar == '.') {
1244 if (!isDigit(address.charAt(i-1)) ||
1245 (i+1 < addrLength && !isDigit(address.charAt(i+1)))) {
1246 return false;
1247 }
1248 numDots++;
1249 }
1250 else if (!isDigit(testChar)) {
1251 return false;
1252 }
1253 }
1254 if (numDots != 3) {
1255 return false;
1256 }
1257 }
1258 else {
1259 // domain labels can contain alphanumerics and '-"
1260 // but must start and end with an alphanumeric
1261 char testChar;
1262
1263 for (int i = 0; i < addrLength; i++) {
1264 testChar = address.charAt(i);
1265 if (testChar == '.') {
1266 if (!isAlphanum(address.charAt(i-1))) {
1267 return false;
1268 }
1269 if (i+1 < addrLength && !isAlphanum(address.charAt(i+1))) {
1270 return false;
1271 }
1272 }
1273 else if (!isAlphanum(testChar) && testChar != '-') {
1274 return false;
1275 }
1276 }
1277 }
1278 return true;
1279 }
1280
1281
1282 /**
1283 * Determine whether a char is a digit.
1284 *
1285 * @return true if the char is betweeen '0' and '9', false otherwise
1286 */
1287 private static boolean isDigit(char p_char) {
1288 return p_char >= '0' && p_char <= '9';
1289 }
1290
1291 /**
1292 * Determine whether a character is a hexadecimal character.
1293 *
1294 * @return true if the char is betweeen '0' and '9', 'a' and 'f'
1295 * or 'A' and 'F', false otherwise
1296 */
1297 private static boolean isHex(char p_char) {
1298 return (isDigit(p_char) ||
1299 (p_char >= 'a' && p_char <= 'f') ||
1300 (p_char >= 'A' && p_char <= 'F'));
1301 }
1302
1303 /**
1304 * Determine whether a char is an alphabetic character: a-z or A-Z
1305 *
1306 * @return true if the char is alphabetic, false otherwise
1307 */
1308 private static boolean isAlpha(char p_char) {
1309 return ((p_char >= 'a' && p_char <= 'z') ||
1310 (p_char >= 'A' && p_char <= 'Z' ));
1311 }
1312
1313 /**
1314 * Determine whether a char is an alphanumeric: 0-9, a-z or A-Z
1315 *
1316 * @return true if the char is alphanumeric, false otherwise
1317 */
1318 private static boolean isAlphanum(char p_char) {
1319 return (isAlpha(p_char) || isDigit(p_char));
1320 }
1321
1322 /**
1323 * Determine whether a character is a reserved character:
1324 * ';', '/', '?', ':', '@', '&', '=', '+', '$' or ','
1325 *
1326 * @return true if the string contains any reserved characters
1327 */
1328 private static boolean isReservedCharacter(char p_char) {
1329 return RESERVED_CHARACTERS.indexOf(p_char) != -1;
1330 }
1331
1332 /**
1333 * Determine whether a char is an unreserved character.
1334 *
1335 * @return true if the char is unreserved, false otherwise
1336 */
1337 private static boolean isUnreservedCharacter(char p_char) {
1338 return (isAlphanum(p_char) ||
1339 MARK_CHARACTERS.indexOf(p_char) != -1);
1340 }
1341
1342 /**
1343 * Determine whether a given string contains only URI characters (also
1344 * called "uric" in RFC 2396). uric consist of all reserved
1345 * characters, unreserved characters and escaped characters.
1346 *
1347 * @return true if the string is comprised of uric, false otherwise
1348 */
1349 private static boolean isURIString(String p_uric) {
1350 if (p_uric == null) {
1351 return false;
1352 }
1353 int end = p_uric.length();
1354 char testChar = '\0';
1355 for (int i = 0; i < end; i++) {
1356 testChar = p_uric.charAt(i);
1357 if (testChar == '%') {
1358 if (i+2 >= end ||
1359 !isHex(p_uric.charAt(i+1)) ||
1360 !isHex(p_uric.charAt(i+2))) {
1361 return false;
1362 }
1363 else {
1364 i += 2;
1365 continue;
1366 }
1367 }
1368 if (isReservedCharacter(testChar) ||
1369 isUnreservedCharacter(testChar)) {
1370 continue;
1371 }
1372 else {
1373 return false;
1374 }
1375 }
1376 return true;
1377 }
1378 }