1 /*
2 * Copyright (c) 1997, 2017, 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">
43 * RFC 2396</a>. 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 **********************************************************************/
65 public class JaxmURI implements Serializable {
66
67 /*******************************************************************
68 * MalformedURIExceptions are thrown in the process of building a URI
69 * or setting fields on a URI when an operation would result in an
70 * invalid URI specification.
71 *
72 ********************************************************************/
73 public static class MalformedURIException extends IOException {
74
75 /******************************************************************
76 * Constructs a <code>MalformedURIException</code> with no specified
77 * detail message.
78 ******************************************************************/
79 public MalformedURIException() {
80 super();
81 }
82
83 /*****************************************************************
84 * Constructs a <code>MalformedURIException</code> with the
85 * specified detail message.
86 *
87 * @param p_msg the detail message.
88 ******************************************************************/
89 public MalformedURIException(String p_msg) {
90 super(p_msg);
91 }
92 }
93
94 /** reserved characters */
95 private static final String RESERVED_CHARACTERS = ";/?:@&=+$,";
96
97 /** URI punctuation mark characters - these, combined with
98 alphanumerics, constitute the "unreserved" characters */
99 private static final String MARK_CHARACTERS = "-_.!~*'() ";
100
101 /** scheme can be composed of alphanumerics and these characters */
102 private static final String SCHEME_CHARACTERS = "+-.";
103
104 /** userinfo can be composed of unreserved, escaped and these
105 characters */
106 private static final String USERINFO_CHARACTERS = ";:&=+$,";
107
108 /** Stores the scheme (usually the protocol) for this URI. */
109 private String m_scheme = null;
110
111 /** If specified, stores the userinfo for this URI; otherwise null */
112 private String m_userinfo = null;
113
114 /** If specified, stores the host for this URI; otherwise null */
115 private String m_host = null;
116
117 /** If specified, stores the port for this URI; otherwise -1 */
118 private int m_port = -1;
119
120 /** If specified, stores the path for this URI; otherwise null */
121 private String m_path = null;
122
123 /** If specified, stores the query string for this URI; otherwise
124 null. */
125 private String m_queryString = null;
126
127 /** If specified, stores the fragment for this URI; otherwise null */
128 private String m_fragment = null;
129
130 /**
131 * Construct a new and uninitialized URI.
132 */
133 public JaxmURI() {
134 }
135
136 /**
137 * Construct a new URI from another URI. All fields for this URI are
138 * set equal to the fields of the URI passed in.
139 *
140 * @param p_other the URI to copy (cannot be null)
141 */
142 public JaxmURI(JaxmURI p_other) {
143 initialize(p_other);
144 }
145
146 /**
147 * Construct a new URI from a URI specification string. If the
148 * specification follows the "generic URI" syntax, (two slashes
149 * following the first colon), the specification will be parsed
150 * accordingly - setting the scheme, userinfo, host,port, path, query
151 * string and fragment fields as necessary. If the specification does
152 * not follow the "generic URI" syntax, the specification is parsed
153 * into a scheme and scheme-specific part (stored as the path) only.
154 *
155 * @param p_uriSpec the URI specification string (cannot be null or
156 * empty)
157 *
158 * @exception MalformedURIException if p_uriSpec violates any syntax
159 * rules
160 */
161 public JaxmURI(String p_uriSpec) throws MalformedURIException {
162 this((JaxmURI)null, p_uriSpec);
163 }
164
165 /**
166 * Construct a new URI from a base URI and a URI specification string.
167 * The URI specification string may be a relative URI.
168 *
169 * @param p_base the base URI (cannot be null if p_uriSpec is null or
170 * empty)
171 * @param p_uriSpec the URI specification string (cannot be null or
172 * empty if p_base is null)
173 *
174 * @exception MalformedURIException if p_uriSpec violates any syntax
175 * rules
176 */
177 public JaxmURI(JaxmURI p_base, String p_uriSpec) throws MalformedURIException {
178 initialize(p_base, p_uriSpec);
179 }
180
181 /**
182 * Construct a new URI that does not follow the generic URI syntax.
183 * Only the scheme and scheme-specific part (stored as the path) are
184 * initialized.
185 *
186 * @param p_scheme the URI scheme (cannot be null or empty)
187 * @param p_schemeSpecificPart the scheme-specific part (cannot be
188 * null or empty)
189 *
190 * @exception MalformedURIException if p_scheme violates any
191 * syntax rules
192 */
193 public JaxmURI(String p_scheme, String p_schemeSpecificPart)
194 throws MalformedURIException {
195 if (p_scheme == null || p_scheme.trim().length() == 0) {
196 throw new MalformedURIException(
197 "Cannot construct URI with null/empty scheme!");
198 }
199 if (p_schemeSpecificPart == null ||
200 p_schemeSpecificPart.trim().length() == 0) {
201 throw new MalformedURIException(
202 "Cannot construct URI with null/empty scheme-specific part!");
203 }
204 setScheme(p_scheme);
205 setPath(p_schemeSpecificPart);
206 }
207
208 /**
209 * Construct a new URI that follows the generic URI syntax from its
210 * component parts. Each component is validated for syntax and some
211 * basic semantic checks are performed as well. See the individual
212 * setter methods for specifics.
213 *
214 * @param p_scheme the URI scheme (cannot be null or empty)
215 * @param p_host the hostname or IPv4 address for the URI
216 * @param p_path the URI path - if the path contains '?' or '#',
217 * then the query string and/or fragment will be
218 * set from the path; however, if the query and
219 * fragment are specified both in the path and as
220 * separate parameters, an exception is thrown
221 * @param p_queryString the URI query string (cannot be specified
222 * if path is null)
223 * @param p_fragment the URI fragment (cannot be specified if path
224 * is null)
225 *
226 * @exception MalformedURIException if any of the parameters violates
227 * syntax rules or semantic rules
228 */
229 public JaxmURI(String p_scheme, String p_host, String p_path,
230 String p_queryString, String p_fragment)
231 throws MalformedURIException {
232 this(p_scheme, null, p_host, -1, p_path, p_queryString, p_fragment);
233 }
234
235 /**
236 * Construct a new URI that follows the generic URI syntax from its
237 * component parts. Each component is validated for syntax and some
238 * basic semantic checks are performed as well. See the individual
239 * setter methods for specifics.
240 *
241 * @param p_scheme the URI scheme (cannot be null or empty)
242 * @param p_userinfo the URI userinfo (cannot be specified if host
243 * is null)
244 * @param p_host the hostname or IPv4 address for the URI
245 * @param p_port the URI port (may be -1 for "unspecified"; cannot
246 * be specified if host is null)
247 * @param p_path the URI path - if the path contains '?' or '#',
248 * then the query string and/or fragment will be
249 * set from the path; however, if the query and
250 * fragment are specified both in the path and as
251 * separate parameters, an exception is thrown
252 * @param p_queryString the URI query string (cannot be specified
253 * if path is null)
254 * @param p_fragment the URI fragment (cannot be specified if path
255 * is null)
256 *
257 * @exception MalformedURIException if any of the parameters violates
258 * syntax rules or semantic rules
259 */
260 public JaxmURI(String p_scheme, String p_userinfo,
261 String p_host, int p_port, String p_path,
262 String p_queryString, String p_fragment)
263 throws MalformedURIException {
264 if (p_scheme == null || p_scheme.trim().length() == 0) {
265 throw new MalformedURIException("Scheme is required!");
266 }
267
268 if (p_host == null) {
269 if (p_userinfo != null) {
270 throw new MalformedURIException(
271 "Userinfo may not be specified if host is not specified!");
272 }
273 if (p_port != -1) {
274 throw new MalformedURIException(
275 "Port may not be specified if host is not specified!");
276 }
277 }
278
279 if (p_path != null) {
280 if (p_path.indexOf('?') != -1 && p_queryString != null) {
281 throw new MalformedURIException(
282 "Query string cannot be specified in path and query string!");
283 }
284
285 if (p_path.indexOf('#') != -1 && p_fragment != null) {
286 throw new MalformedURIException(
287 "Fragment cannot be specified in both the path and fragment!");
288 }
289 }
290
291 setScheme(p_scheme);
292 setHost(p_host);
293 setPort(p_port);
294 setUserinfo(p_userinfo);
295 setPath(p_path);
296 setQueryString(p_queryString);
297 setFragment(p_fragment);
298 }
299
300 /**
301 * Initialize all fields of this URI from another URI.
302 *
303 * @param p_other the URI to copy (cannot be null)
304 */
305 private void initialize(JaxmURI p_other) {
306 m_scheme = p_other.getScheme();
307 m_userinfo = p_other.getUserinfo();
308 m_host = p_other.getHost();
309 m_port = p_other.getPort();
310 m_path = p_other.getPath();
311 m_queryString = p_other.getQueryString();
312 m_fragment = p_other.getFragment();
313 }
314
315 /**
316 * Initializes this URI from a base URI and a URI specification string.
317 * See RFC 2396 Section 4 and Appendix B for specifications on parsing
318 * the URI and Section 5 for specifications on resolving relative URIs
319 * and relative paths.
320 *
321 * @param p_base the base URI (may be null if p_uriSpec is an absolute
322 * URI)
323 * @param p_uriSpec the URI spec string which may be an absolute or
324 * relative URI (can only be null/empty if p_base
325 * is not null)
326 *
327 * @exception MalformedURIException if p_base is null and p_uriSpec
328 * is not an absolute URI or if
329 * p_uriSpec violates syntax rules
330 */
331 private void initialize(JaxmURI p_base, String p_uriSpec)
332 throws MalformedURIException {
333 if (p_base == null &&
334 (p_uriSpec == null || p_uriSpec.trim().length() == 0)) {
335 throw new MalformedURIException(
336 "Cannot initialize URI with empty parameters.");
337 }
338
339 // just make a copy of the base if spec is empty
340 if (p_uriSpec == null || p_uriSpec.trim().length() == 0) {
341 initialize(p_base);
342 return;
343 }
344
345 String uriSpec = p_uriSpec.trim();
346 int uriSpecLen = uriSpec.length();
347 int index = 0;
348
349 // Check for scheme, which must be before `/'. Also handle names with
350 // DOS drive letters ('D:'), so 1-character schemes are not allowed.
351 int colonIdx = uriSpec.indexOf(':');
352 int slashIdx = uriSpec.indexOf('/');
353 if ((colonIdx < 2) || (colonIdx > slashIdx && slashIdx != -1)) {
354 int fragmentIdx = uriSpec.indexOf('#');
355 // A standalone base is a valid URI according to spec
356 if (p_base == null && fragmentIdx != 0 ) {
357 throw new MalformedURIException("No scheme found in URI.");
358 }
359 }
360 else {
361 initializeScheme(uriSpec);
362 index = m_scheme.length()+1;
363 }
364
365 // two slashes means generic URI syntax, so we get the authority
366 if (((index+1) < uriSpecLen) &&
367 (uriSpec.substring(index).startsWith("//"))) {
368 index += 2;
369 int startPos = index;
370
371 // get authority - everything up to path, query or fragment
372 char testChar = '\0';
373 while (index < uriSpecLen) {
374 testChar = uriSpec.charAt(index);
375 if (testChar == '/' || testChar == '?' || testChar == '#') {
376 break;
377 }
378 index++;
379 }
380
381 // if we found authority, parse it out, otherwise we set the
382 // host to empty string
383 if (index > startPos) {
384 initializeAuthority(uriSpec.substring(startPos, index));
385 }
386 else {
387 m_host = "";
388 }
389 }
390
391 initializePath(uriSpec.substring(index));
392
393 // Resolve relative URI to base URI - see RFC 2396 Section 5.2
394 // In some cases, it might make more sense to throw an exception
395 // (when scheme is specified is the string spec and the base URI
396 // is also specified, for example), but we're just following the
397 // RFC specifications
398 if (p_base != null) {
399
400 // check to see if this is the current doc - RFC 2396 5.2 #2
401 // note that this is slightly different from the RFC spec in that
402 // we don't include the check for query string being null
403 // - this handles cases where the urispec is just a query
404 // string or a fragment (e.g. "?y" or "#s") -
405 // see <http://www.ics.uci.edu/~fielding/url/test1.html> which
406 // identified this as a bug in the RFC
407 if (m_path.length() == 0 && m_scheme == null &&
408 m_host == null) {
409 m_scheme = p_base.getScheme();
410 m_userinfo = p_base.getUserinfo();
411 m_host = p_base.getHost();
412 m_port = p_base.getPort();
413 m_path = p_base.getPath();
414
415 if (m_queryString == null) {
416 m_queryString = p_base.getQueryString();
417 }
418 return;
419 }
420
421 // check for scheme - RFC 2396 5.2 #3
422 // if we found a scheme, it means absolute URI, so we're done
423 if (m_scheme == null) {
424 m_scheme = p_base.getScheme();
425 }
426 else {
427 return;
428 }
429
430 // check for authority - RFC 2396 5.2 #4
431 // if we found a host, then we've got a network path, so we're done
432 if (m_host == null) {
433 m_userinfo = p_base.getUserinfo();
434 m_host = p_base.getHost();
435 m_port = p_base.getPort();
436 }
437 else {
438 return;
439 }
440
441 // check for absolute path - RFC 2396 5.2 #5
442 if (m_path.length() > 0 &&
443 m_path.startsWith("/")) {
444 return;
445 }
446
447 // if we get to this point, we need to resolve relative path
448 // RFC 2396 5.2 #6
449 String path = "";
450 String basePath = p_base.getPath();
451
452 // 6a - get all but the last segment of the base URI path
453 if (basePath != null) {
454 int lastSlash = basePath.lastIndexOf('/');
455 if (lastSlash != -1) {
456 path = basePath.substring(0, lastSlash+1);
457 }
458 }
459
460 // 6b - append the relative URI path
461 path = path.concat(m_path);
462
463 // 6c - remove all "./" where "." is a complete path segment
464 index = -1;
465 while ((index = path.indexOf("/./")) != -1) {
466 path = path.substring(0, index+1).concat(path.substring(index+3));
467 }
468
469 // 6d - remove "." if path ends with "." as a complete path segment
470 if (path.endsWith("/.")) {
471 path = path.substring(0, path.length()-1);
472 }
473
474 // 6e - remove all "<segment>/../" where "<segment>" is a complete
475 // path segment not equal to ".."
476 index = 1;
477 int segIndex = -1;
478 String tempString = null;
479
480 while ((index = path.indexOf("/../", index)) > 0) {
481 tempString = path.substring(0, path.indexOf("/../"));
482 segIndex = tempString.lastIndexOf('/');
483 if (segIndex != -1) {
484 if (!tempString.substring(segIndex++).equals("..")) {
485 path = path.substring(0, segIndex).concat(path.substring(index+4));
486 }
487 else
488 index += 4;
489 }
490 else
491 index += 4;
492 }
493
494 // 6f - remove ending "<segment>/.." where "<segment>" is a
495 // complete path segment
496 if (path.endsWith("/..")) {
497 tempString = path.substring(0, path.length()-3);
498 segIndex = tempString.lastIndexOf('/');
499 if (segIndex != -1) {
500 path = path.substring(0, segIndex+1);
501 }
502 }
503 m_path = path;
504 }
505 }
506
507 /**
508 * Initialize the scheme for this URI from a URI string spec.
509 *
510 * @param p_uriSpec the URI specification (cannot be null)
511 *
512 * @exception MalformedURIException if URI does not have a conformant
513 * scheme
514 */
515 private void initializeScheme(String p_uriSpec)
516 throws MalformedURIException {
517 int uriSpecLen = p_uriSpec.length();
518 int index = 0;
519 String scheme = null;
520 char testChar = '\0';
521
522 while (index < uriSpecLen) {
523 testChar = p_uriSpec.charAt(index);
524 if (testChar == ':' || testChar == '/' ||
525 testChar == '?' || testChar == '#') {
526 break;
527 }
528 index++;
529 }
530 scheme = p_uriSpec.substring(0, index);
531
532 if (scheme.length() == 0) {
533 throw new MalformedURIException("No scheme found in URI.");
534 }
535 else {
536 setScheme(scheme);
537 }
538 }
539
540 /**
541 * Initialize the authority (userinfo, host and port) for this
542 * URI from a URI string spec.
543 *
544 * @param p_uriSpec the URI specification (cannot be null)
545 *
546 * @exception MalformedURIException if p_uriSpec violates syntax rules
547 */
548 private void initializeAuthority(String p_uriSpec)
549 throws MalformedURIException {
550 int index = 0;
551 int start = 0;
552 int end = p_uriSpec.length();
553 char testChar = '\0';
554 String userinfo = null;
555
556 // userinfo is everything up @
557 if (p_uriSpec.indexOf('@', start) != -1) {
558 while (index < end) {
559 testChar = p_uriSpec.charAt(index);
560 if (testChar == '@') {
561 break;
562 }
563 index++;
564 }
565 userinfo = p_uriSpec.substring(start, index);
566 index++;
567 }
568
569 // host is everything up to ':'
570 String host = null;
571 start = index;
572 while (index < end) {
573 testChar = p_uriSpec.charAt(index);
574 if (testChar == ':') {
575 break;
576 }
577 index++;
578 }
579 host = p_uriSpec.substring(start, index);
580 int port = -1;
581 if (host.length() > 0) {
582 // port
583 if (testChar == ':') {
584 index++;
585 start = index;
586 while (index < end) {
587 index++;
588 }
589 String portStr = p_uriSpec.substring(start, index);
590 if (portStr.length() > 0) {
591 for (int i = 0; i < portStr.length(); i++) {
592 if (!isDigit(portStr.charAt(i))) {
593 throw new MalformedURIException(
594 portStr +
595 " is invalid. Port should only contain digits!");
596 }
597 }
598 try {
599 port = Integer.parseInt(portStr);
600 }
601 catch (NumberFormatException nfe) {
602 // can't happen
603 }
604 }
605 }
606 }
607 setHost(host);
608 setPort(port);
609 setUserinfo(userinfo);
610 }
611
612 /**
613 * Initialize the path for this URI from a URI string spec.
614 *
615 * @param p_uriSpec the URI specification (cannot be null)
616 *
617 * @exception MalformedURIException if p_uriSpec violates syntax rules
618 */
619 private void initializePath(String p_uriSpec)
620 throws MalformedURIException {
621 if (p_uriSpec == null) {
622 throw new MalformedURIException(
623 "Cannot initialize path from null string!");
624 }
625
626 int index = 0;
627 int start = 0;
628 int end = p_uriSpec.length();
629 char testChar = '\0';
630
631 // path - everything up to query string or fragment
632 while (index < end) {
633 testChar = p_uriSpec.charAt(index);
634 if (testChar == '?' || testChar == '#') {
635 break;
636 }
637 // check for valid escape sequence
638 if (testChar == '%') {
639 if (index+2 >= end ||
640 !isHex(p_uriSpec.charAt(index+1)) ||
641 !isHex(p_uriSpec.charAt(index+2))) {
642 throw new MalformedURIException(
643 "Path contains invalid escape sequence!");
644 }
645 }
646 else if (!isReservedCharacter(testChar) &&
647 !isUnreservedCharacter(testChar)) {
648 throw new MalformedURIException(
649 "Path contains invalid character: " + testChar);
650 }
651 index++;
652 }
653 m_path = p_uriSpec.substring(start, index);
654
655 // query - starts with ? and up to fragment or end
656 if (testChar == '?') {
657 index++;
658 start = index;
659 while (index < end) {
660 testChar = p_uriSpec.charAt(index);
661 if (testChar == '#') {
662 break;
663 }
664 if (testChar == '%') {
665 if (index+2 >= end ||
666 !isHex(p_uriSpec.charAt(index+1)) ||
667 !isHex(p_uriSpec.charAt(index+2))) {
668 throw new MalformedURIException(
669 "Query string contains invalid escape sequence!");
670 }
671 }
672 else if (!isReservedCharacter(testChar) &&
673 !isUnreservedCharacter(testChar)) {
674 throw new MalformedURIException(
675 "Query string contains invalid character:" + testChar);
676 }
677 index++;
678 }
679 m_queryString = p_uriSpec.substring(start, index);
680 }
681
682 // fragment - starts with #
683 if (testChar == '#') {
684 index++;
685 start = index;
686 while (index < end) {
687 testChar = p_uriSpec.charAt(index);
688
689 if (testChar == '%') {
690 if (index+2 >= end ||
691 !isHex(p_uriSpec.charAt(index+1)) ||
692 !isHex(p_uriSpec.charAt(index+2))) {
693 throw new MalformedURIException(
694 "Fragment contains invalid escape sequence!");
695 }
696 }
697 else if (!isReservedCharacter(testChar) &&
698 !isUnreservedCharacter(testChar)) {
699 throw new MalformedURIException(
700 "Fragment contains invalid character:"+testChar);
701 }
702 index++;
703 }
704 m_fragment = p_uriSpec.substring(start, index);
705 }
706 }
707
708 /**
709 * Get the scheme for this URI.
710 *
711 * @return the scheme for this URI
712 */
713 public String getScheme() {
714 return m_scheme;
715 }
716
717 /**
718 * Get the scheme-specific part for this URI (everything following the
719 * scheme and the first colon). See RFC 2396 Section 5.2 for spec.
720 *
721 * @return the scheme-specific part for this URI
722 */
723 public String getSchemeSpecificPart() {
724 StringBuilder schemespec = new StringBuilder();
725
726 if (m_userinfo != null || m_host != null || m_port != -1) {
727 schemespec.append("//");
728 }
729
730 if (m_userinfo != null) {
731 schemespec.append(m_userinfo);
732 schemespec.append('@');
733 }
734
735 if (m_host != null) {
736 schemespec.append(m_host);
737 }
738
739 if (m_port != -1) {
740 schemespec.append(':');
741 schemespec.append(m_port);
742 }
743
744 if (m_path != null) {
745 schemespec.append((m_path));
746 }
747
748 if (m_queryString != null) {
749 schemespec.append('?');
750 schemespec.append(m_queryString);
751 }
752
753 if (m_fragment != null) {
754 schemespec.append('#');
755 schemespec.append(m_fragment);
756 }
757
758 return schemespec.toString();
759 }
760
761 /**
762 * Get the userinfo for this URI.
763 *
764 * @return the userinfo for this URI (null if not specified).
765 */
766 public String getUserinfo() {
767 return m_userinfo;
768 }
769
770 /**
771 * Get the host for this URI.
772 *
773 * @return the host for this URI (null if not specified).
774 */
775 public String getHost() {
776 return m_host;
777 }
778
779 /**
780 * Get the port for this URI.
781 *
782 * @return the port for this URI (-1 if not specified).
783 */
784 public int getPort() {
785 return m_port;
786 }
787
788 /**
789 * Get the path for this URI (optionally with the query string and
790 * fragment).
791 *
792 * @param p_includeQueryString if true (and query string is not null),
793 * then a "?" followed by the query string
794 * will be appended
795 * @param p_includeFragment if true (and fragment is not null),
796 * then a "#" followed by the fragment
797 * will be appended
798 *
799 * @return the path for this URI possibly including the query string
800 * and fragment
801 */
802 public String getPath(boolean p_includeQueryString,
803 boolean p_includeFragment) {
804 StringBuilder pathString = new StringBuilder(m_path);
805
806 if (p_includeQueryString && m_queryString != null) {
807 pathString.append('?');
808 pathString.append(m_queryString);
809 }
810
811 if (p_includeFragment && m_fragment != null) {
812 pathString.append('#');
813 pathString.append(m_fragment);
814 }
815 return pathString.toString();
816 }
817
818 /**
819 * Get the path for this URI. Note that the value returned is the path
820 * only and does not include the query string or fragment.
821 *
822 * @return the path for this URI.
823 */
824 public String getPath() {
825 return m_path;
826 }
827
828 /**
829 * Get the query string for this URI.
830 *
831 * @return the query string for this URI. Null is returned if there
832 * was no "?" in the URI spec, empty string if there was a
833 * "?" but no query string following it.
834 */
835 public String getQueryString() {
836 return m_queryString;
837 }
838
839 /**
840 * Get the fragment for this URI.
841 *
842 * @return the fragment for this URI. Null is returned if there
843 * was no "#" in the URI spec, empty string if there was a
844 * "#" but no fragment following it.
845 */
846 public String getFragment() {
847 return m_fragment;
848 }
849
850 /**
851 * Set the scheme for this URI. The scheme is converted to lowercase
852 * before it is set.
853 *
854 * @param p_scheme the scheme for this URI (cannot be null)
855 *
856 * @exception MalformedURIException if p_scheme is not a conformant
857 * scheme name
858 */
859 public void setScheme(String p_scheme) throws MalformedURIException {
860 if (p_scheme == null) {
861 throw new MalformedURIException(
862 "Cannot set scheme from null string!");
863 }
864 if (!isConformantSchemeName(p_scheme)) {
865 throw new MalformedURIException("The scheme is not conformant.");
866 }
867
868 m_scheme = p_scheme.toLowerCase();
869 }
870
871 /**
872 * Set the userinfo for this URI. If a non-null value is passed in and
873 * the host value is null, then an exception is thrown.
874 *
875 * @param p_userinfo the userinfo for this URI
876 *
877 * @exception MalformedURIException if p_userinfo contains invalid
878 * characters
879 */
880 public void setUserinfo(String p_userinfo) throws MalformedURIException {
881 if (p_userinfo == null) {
882 m_userinfo = null;
883 }
884 else {
885 if (m_host == null) {
886 throw new MalformedURIException(
887 "Userinfo cannot be set when host is null!");
888 }
889
890 // userinfo can contain alphanumerics, mark characters, escaped
891 // and ';',':','&','=','+','$',','
892 int index = 0;
893 int end = p_userinfo.length();
894 char testChar = '\0';
895 while (index < end) {
896 testChar = p_userinfo.charAt(index);
897 if (testChar == '%') {
898 if (index+2 >= end ||
899 !isHex(p_userinfo.charAt(index+1)) ||
900 !isHex(p_userinfo.charAt(index+2))) {
901 throw new MalformedURIException(
902 "Userinfo contains invalid escape sequence!");
903 }
904 }
905 else if (!isUnreservedCharacter(testChar) &&
906 USERINFO_CHARACTERS.indexOf(testChar) == -1) {
907 throw new MalformedURIException(
908 "Userinfo contains invalid character:"+testChar);
909 }
910 index++;
911 }
912 }
913 m_userinfo = p_userinfo;
914 }
915
916 /**
917 * Set the host for this URI. If null is passed in, the userinfo
918 * field is also set to null and the port is set to -1.
919 *
920 * @param p_host the host for this URI
921 *
922 * @exception MalformedURIException if p_host is not a valid IP
923 * address or DNS hostname.
924 */
925 public void setHost(String p_host) throws MalformedURIException {
926 if (p_host == null || p_host.trim().length() == 0) {
927 m_host = p_host;
928 m_userinfo = null;
929 m_port = -1;
930 }
931 else if (!isWellFormedAddress(p_host)) {
932 throw new MalformedURIException("Host is not a well formed address!");
933 }
934 m_host = p_host;
935 }
936
937 /**
938 * Set the port for this URI. -1 is used to indicate that the port is
939 * not specified, otherwise valid port numbers are between 0 and 65535.
940 * If a valid port number is passed in and the host field is null,
941 * an exception is thrown.
942 *
943 * @param p_port the port number for this URI
944 *
945 * @exception MalformedURIException if p_port is not -1 and not a
946 * valid port number
947 */
948 public void setPort(int p_port) throws MalformedURIException {
949 if (p_port >= 0 && p_port <= 65535) {
950 if (m_host == null) {
951 throw new MalformedURIException(
952 "Port cannot be set when host is null!");
953 }
954 }
955 else if (p_port != -1) {
956 throw new MalformedURIException("Invalid port number!");
957 }
958 m_port = p_port;
959 }
960
961 /**
962 * Set the path for this URI. If the supplied path is null, then the
963 * query string and fragment are set to null as well. If the supplied
964 * path includes a query string and/or fragment, these fields will be
965 * parsed and set as well. Note that, for URIs following the "generic
966 * URI" syntax, the path specified should start with a slash.
967 * For URIs that do not follow the generic URI syntax, this method
968 * sets the scheme-specific part.
969 *
970 * @param p_path the path for this URI (may be null)
971 *
972 * @exception MalformedURIException if p_path contains invalid
973 * characters
974 */
975 public void setPath(String p_path) throws MalformedURIException {
976 if (p_path == null) {
977 m_path = null;
978 m_queryString = null;
979 m_fragment = null;
980 }
981 else {
982 initializePath(p_path);
983 }
984 }
985
986 /**
987 * Append to the end of the path of this URI. If the current path does
988 * not end in a slash and the path to be appended does not begin with
989 * a slash, a slash will be appended to the current path before the
990 * new segment is added. Also, if the current path ends in a slash
991 * and the new segment begins with a slash, the extra slash will be
992 * removed before the new segment is appended.
993 *
994 * @param p_addToPath the new segment to be added to the current path
995 *
996 * @exception MalformedURIException if p_addToPath contains syntax
997 * errors
998 */
999 public void appendPath(String p_addToPath)
1000 throws MalformedURIException {
1001 if (p_addToPath == null || p_addToPath.trim().length() == 0) {
1002 return;
1003 }
1004
1005 if (!isURIString(p_addToPath)) {
1006 throw new MalformedURIException(
1007 "Path contains invalid character!");
1008 }
1009
1010 if (m_path == null || m_path.trim().length() == 0) {
1011 if (p_addToPath.startsWith("/")) {
1012 m_path = p_addToPath;
1013 }
1014 else {
1015 m_path = "/" + p_addToPath;
1016 }
1017 }
1018 else if (m_path.endsWith("/")) {
1019 if (p_addToPath.startsWith("/")) {
1020 m_path = m_path.concat(p_addToPath.substring(1));
1021 }
1022 else {
1023 m_path = m_path.concat(p_addToPath);
1024 }
1025 }
1026 else {
1027 if (p_addToPath.startsWith("/")) {
1028 m_path = m_path.concat(p_addToPath);
1029 }
1030 else {
1031 m_path = m_path.concat("/" + p_addToPath);
1032 }
1033 }
1034 }
1035
1036 /**
1037 * Set the query string for this URI. A non-null value is valid only
1038 * if this is an URI conforming to the generic URI syntax and
1039 * the path value is not null.
1040 *
1041 * @param p_queryString the query string for this URI
1042 *
1043 * @exception MalformedURIException if p_queryString is not null and this
1044 * URI does not conform to the generic
1045 * URI syntax or if the path is null
1046 */
1047 public void setQueryString(String p_queryString) throws MalformedURIException {
1048 if (p_queryString == null) {
1049 m_queryString = null;
1050 }
1051 else if (!isGenericURI()) {
1052 throw new MalformedURIException(
1053 "Query string can only be set for a generic URI!");
1054 }
1055 else if (getPath() == null) {
1056 throw new MalformedURIException(
1057 "Query string cannot be set when path is null!");
1058 }
1059 else if (!isURIString(p_queryString)) {
1060 throw new MalformedURIException(
1061 "Query string contains invalid character!");
1062 }
1063 else {
1064 m_queryString = p_queryString;
1065 }
1066 }
1067
1068 /**
1069 * Set the fragment for this URI. A non-null value is valid only
1070 * if this is a URI conforming to the generic URI syntax and
1071 * the path value is not null.
1072 *
1073 * @param p_fragment the fragment for this URI
1074 *
1075 * @exception MalformedURIException if p_fragment is not null and this
1076 * URI does not conform to the generic
1077 * URI syntax or if the path is null
1078 */
1079 public void setFragment(String p_fragment) throws MalformedURIException {
1080 if (p_fragment == null) {
1081 m_fragment = null;
1082 }
1083 else if (!isGenericURI()) {
1084 throw new MalformedURIException(
1085 "Fragment can only be set for a generic URI!");
1086 }
1087 else if (getPath() == null) {
1088 throw new MalformedURIException(
1089 "Fragment cannot be set when path is null!");
1090 }
1091 else if (!isURIString(p_fragment)) {
1092 throw new MalformedURIException(
1093 "Fragment contains invalid character!");
1094 }
1095 else {
1096 m_fragment = p_fragment;
1097 }
1098 }
1099
1100 /**
1101 * Determines if the passed-in Object is equivalent to this URI.
1102 *
1103 * @param p_test the Object to test for equality.
1104 *
1105 * @return true if p_test is a URI with all values equal to this
1106 * URI, false otherwise
1107 */
1108 @Override
1109 public boolean equals(Object p_test) {
1110 if (p_test instanceof JaxmURI) {
1111 JaxmURI testURI = (JaxmURI) p_test;
1112 if (((m_scheme == null && testURI.m_scheme == null) ||
1113 (m_scheme != null && testURI.m_scheme != null &&
1114 m_scheme.equals(testURI.m_scheme))) &&
1115 ((m_userinfo == null && testURI.m_userinfo == null) ||
1116 (m_userinfo != null && testURI.m_userinfo != null &&
1117 m_userinfo.equals(testURI.m_userinfo))) &&
1118 ((m_host == null && testURI.m_host == null) ||
1119 (m_host != null && testURI.m_host != null &&
1120 m_host.equals(testURI.m_host))) &&
1121 m_port == testURI.m_port &&
1122 ((m_path == null && testURI.m_path == null) ||
1123 (m_path != null && testURI.m_path != null &&
1124 m_path.equals(testURI.m_path))) &&
1125 ((m_queryString == null && testURI.m_queryString == null) ||
1126 (m_queryString != null && testURI.m_queryString != null &&
1127 m_queryString.equals(testURI.m_queryString))) &&
1128 ((m_fragment == null && testURI.m_fragment == null) ||
1129 (m_fragment != null && testURI.m_fragment != null &&
1130 m_fragment.equals(testURI.m_fragment)))) {
1131 return true;
1132 }
1133 }
1134 return false;
1135 }
1136
1137 @Override
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 @Override
1149 public String toString() {
1150 StringBuilder uriSpecString = new StringBuilder();
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 * @param p_scheme scheme name
1179 *
1180 * @return true if the scheme is conformant, false otherwise
1181 */
1182 public static boolean isConformantSchemeName(String p_scheme) {
1183 if (p_scheme == null || p_scheme.trim().length() == 0) {
1184 return false;
1185 }
1186
1187 if (!isAlpha(p_scheme.charAt(0))) {
1188 return false;
1189 }
1190
1191 char testChar;
1192 for (int i = 1; i < p_scheme.length(); i++) {
1193 testChar = p_scheme.charAt(i);
1194 if (!isAlphanum(testChar) &&
1195 SCHEME_CHARACTERS.indexOf(testChar) == -1) {
1196 return false;
1197 }
1198 }
1199
1200 return true;
1201 }
1202
1203 /**
1204 * Determine whether a string is syntactically capable of representing
1205 * a valid IPv4 address or the domain name of a network host. A valid
1206 * IPv4 address consists of four decimal digit groups separated by a
1207 * '.'. A hostname consists of domain labels (each of which must
1208 * begin and end with an alphanumeric but may contain '-') separated
1209 * by a '.'. See RFC 2396 Section 3.2.2.
1210 *
1211 * @param p_address address
1212 *
1213 * @return true if the string is a syntactically valid IPv4 address
1214 * or hostname
1215 */
1216 public static boolean isWellFormedAddress(String p_address) {
1217 if (p_address == null) {
1218 return false;
1219 }
1220
1221 String address = p_address.trim();
1222 int addrLength = address.length();
1223 if (addrLength == 0 || addrLength > 255) {
1224 return false;
1225 }
1226
1227 if (address.startsWith(".") || address.startsWith("-")) {
1228 return false;
1229 }
1230
1231 // rightmost domain label starting with digit indicates IP address
1232 // since top level domain label can only start with an alpha
1233 // see RFC 2396 Section 3.2.2
1234 int index = address.lastIndexOf('.');
1235 if (address.endsWith(".")) {
1236 index = address.substring(0, index).lastIndexOf('.');
1237 }
1238
1239 if (index+1 < addrLength && isDigit(p_address.charAt(index+1))) {
1240 char testChar;
1241 int numDots = 0;
1242
1243 // make sure that 1) we see only digits and dot separators, 2) that
1244 // any dot separator is preceded and followed by a digit and
1245 // 3) that we find 3 dots
1246 for (int i = 0; i < addrLength; i++) {
1247 testChar = address.charAt(i);
1248 if (testChar == '.') {
1249 if (!isDigit(address.charAt(i-1)) ||
1250 (i+1 < addrLength && !isDigit(address.charAt(i+1)))) {
1251 return false;
1252 }
1253 numDots++;
1254 }
1255 else if (!isDigit(testChar)) {
1256 return false;
1257 }
1258 }
1259 if (numDots != 3) {
1260 return false;
1261 }
1262 }
1263 else {
1264 // domain labels can contain alphanumerics and '-"
1265 // but must start and end with an alphanumeric
1266 char testChar;
1267
1268 for (int i = 0; i < addrLength; i++) {
1269 testChar = address.charAt(i);
1270 if (testChar == '.') {
1271 if (!isAlphanum(address.charAt(i-1))) {
1272 return false;
1273 }
1274 if (i+1 < addrLength && !isAlphanum(address.charAt(i+1))) {
1275 return false;
1276 }
1277 }
1278 else if (!isAlphanum(testChar) && testChar != '-') {
1279 return false;
1280 }
1281 }
1282 }
1283 return true;
1284 }
1285
1286
1287 /**
1288 * Determine whether a char is a digit.
1289 *
1290 * @return true if the char is betweeen '0' and '9', false otherwise
1291 */
1292 private static boolean isDigit(char p_char) {
1293 return p_char >= '0' && p_char <= '9';
1294 }
1295
1296 /**
1297 * Determine whether a character is a hexadecimal character.
1298 *
1299 * @return true if the char is betweeen '0' and '9', 'a' and 'f'
1300 * or 'A' and 'F', false otherwise
1301 */
1302 private static boolean isHex(char p_char) {
1303 return (isDigit(p_char) ||
1304 (p_char >= 'a' && p_char <= 'f') ||
1305 (p_char >= 'A' && p_char <= 'F'));
1306 }
1307
1308 /**
1309 * Determine whether a char is an alphabetic character: a-z or A-Z
1310 *
1311 * @return true if the char is alphabetic, false otherwise
1312 */
1313 private static boolean isAlpha(char p_char) {
1314 return ((p_char >= 'a' && p_char <= 'z') ||
1315 (p_char >= 'A' && p_char <= 'Z' ));
1316 }
1317
1318 /**
1319 * Determine whether a char is an alphanumeric: 0-9, a-z or A-Z
1320 *
1321 * @return true if the char is alphanumeric, false otherwise
1322 */
1323 private static boolean isAlphanum(char p_char) {
1324 return (isAlpha(p_char) || isDigit(p_char));
1325 }
1326
1327 /**
1328 * Determine whether a character is a reserved character:
1329 * ';', '/', '?', ':', '@', '&', '=', '+', '$' or ','
1330 *
1331 * @return true if the string contains any reserved characters
1332 */
1333 private static boolean isReservedCharacter(char p_char) {
1334 return RESERVED_CHARACTERS.indexOf(p_char) != -1;
1335 }
1336
1337 /**
1338 * Determine whether a char is an unreserved character.
1339 *
1340 * @return true if the char is unreserved, false otherwise
1341 */
1342 private static boolean isUnreservedCharacter(char p_char) {
1343 return (isAlphanum(p_char) ||
1344 MARK_CHARACTERS.indexOf(p_char) != -1);
1345 }
1346
1347 /**
1348 * Determine whether a given string contains only URI characters (also
1349 * called "uric" in RFC 2396). uric consist of all reserved
1350 * characters, unreserved characters and escaped characters.
1351 *
1352 * @return true if the string is comprised of uric, false otherwise
1353 */
1354 private static boolean isURIString(String p_uric) {
1355 if (p_uric == null) {
1356 return false;
1357 }
1358 int end = p_uric.length();
1359 char testChar = '\0';
1360 for (int i = 0; i < end; i++) {
1361 testChar = p_uric.charAt(i);
1362 if (testChar == '%') {
1363 if (i+2 >= end ||
1364 !isHex(p_uric.charAt(i+1)) ||
1365 !isHex(p_uric.charAt(i+2))) {
1366 return false;
1367 }
1368 else {
1369 i += 2;
1370 continue;
1371 }
1372 }
1373 if (isReservedCharacter(testChar) ||
1374 isUnreservedCharacter(testChar)) {
1375 continue;
1376 }
1377 else {
1378 return false;
1379 }
1380 }
1381 return true;
1382 }
1383 }