1 /*
2 * reserved comment block
3 * DO NOT REMOVE OR ALTER!
4 */
5 /*
6 * Copyright 1999-2004 The Apache Software Foundation.
7 *
8 * Licensed under the Apache License, Version 2.0 (the "License");
9 * you may not use this file except in compliance with the License.
10 * You may obtain a copy of the License at
11 *
12 * http://www.apache.org/licenses/LICENSE-2.0
13 *
14 * Unless required by applicable law or agreed to in writing, software
15 * distributed under the License is distributed on an "AS IS" BASIS,
16 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
17 * See the License for the specific language governing permissions and
18 * limitations under the License.
19 */
20 /*
21 * $Id: URI.java,v 1.2.4.1 2005/09/15 08:16:00 suresh_emailid Exp $
22 */
23 package com.sun.org.apache.xml.internal.utils;
24
25 import java.io.IOException;
26 import java.io.Serializable;
27
28 import com.sun.org.apache.xml.internal.res.XMLErrorResources;
29 import com.sun.org.apache.xml.internal.res.XMLMessages;
30
31 /**
32 * A class to represent a Uniform Resource Identifier (URI). This class
33 * is designed to handle the parsing of URIs and provide access to
34 * the various components (scheme, host, port, userinfo, path, query
35 * string and fragment) that may constitute a URI.
36 * <p>
37 * Parsing of a URI specification is done according to the URI
38 * syntax described in RFC 2396
39 * <http://www.ietf.org/rfc/rfc2396.txt?number=2396>. Every URI consists
40 * of a scheme, followed by a colon (':'), followed by a scheme-specific
41 * part. For URIs that follow the "generic URI" syntax, the scheme-
42 * specific part begins with two slashes ("//") and may be followed
43 * by an authority segment (comprised of user information, host, and
44 * port), path segment, query segment and fragment. Note that RFC 2396
45 * no longer specifies the use of the parameters segment and excludes
46 * the "user:password" syntax as part of the authority segment. If
47 * "user:password" appears in a URI, the entire user/password string
48 * is stored as userinfo.
49 * <p>
50 * For URIs that do not follow the "generic URI" syntax (e.g. mailto),
51 * the entire scheme-specific part is treated as the "path" portion
52 * of the URI.
53 * <p>
54 * Note that, unlike the java.net.URL class, this class does not provide
55 * any built-in network access functionality nor does it provide any
56 * scheme-specific functionality (for example, it does not know a
57 * default port for a specific scheme). Rather, it only knows the
58 * grammar and basic set of operations that can be applied to a URI.
59 *
60 *
61 */
62 public class URI implements Serializable
63 {
64 static final long serialVersionUID = 7096266377907081897L;
65
66 /**
67 * MalformedURIExceptions are thrown in the process of building a URI
68 * or setting fields on a URI when an operation would result in an
69 * invalid URI specification.
70 *
71 */
72 public static class MalformedURIException extends IOException
73 {
74
75 /**
76 * Constructs a <code>MalformedURIException</code> with no specified
77 * detail message.
78 */
79 public MalformedURIException()
80 {
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 {
92 super(p_msg);
93 }
94 }
95
96 /** reserved characters */
97 private static final String RESERVED_CHARACTERS = ";/?:@&=+$,";
98
99 /**
100 * URI punctuation mark characters - these, combined with
101 * alphanumerics, constitute the "unreserved" characters
102 */
103 private static final String MARK_CHARACTERS = "-_.!~*'() ";
104
105 /** scheme can be composed of alphanumerics and these characters */
106 private static final String SCHEME_CHARACTERS = "+-.";
107
108 /**
109 * userinfo can be composed of unreserved, escaped and these
110 * characters
111 */
112 private static final String USERINFO_CHARACTERS = ";:&=+$,";
113
114 /** Stores the scheme (usually the protocol) for this URI.
115 * @serial */
116 private String m_scheme = null;
117
118 /** If specified, stores the userinfo for this URI; otherwise null.
119 * @serial */
120 private String m_userinfo = null;
121
122 /** If specified, stores the host for this URI; otherwise null.
123 * @serial */
124 private String m_host = null;
125
126 /** If specified, stores the port for this URI; otherwise -1.
127 * @serial */
128 private int m_port = -1;
129
130 /** If specified, stores the path for this URI; otherwise null.
131 * @serial */
132 private String m_path = null;
133
134 /**
135 * If specified, stores the query string for this URI; otherwise
136 * null.
137 * @serial
138 */
139 private String m_queryString = null;
140
141 /** If specified, stores the fragment for this URI; otherwise null.
142 * @serial */
143 private String m_fragment = null;
144
145 /** Indicate whether in DEBUG mode */
146 private static boolean DEBUG = false;
147
148 /**
149 * Construct a new and uninitialized URI.
150 */
151 public URI(){}
152
153 /**
154 * Construct a new URI from another URI. All fields for this URI are
155 * set equal to the fields of the URI passed in.
156 *
157 * @param p_other the URI to copy (cannot be null)
158 */
159 public URI(URI p_other)
160 {
161 initialize(p_other);
162 }
163
164 /**
165 * Construct a new URI from a URI specification string. If the
166 * specification follows the "generic URI" syntax, (two slashes
167 * following the first colon), the specification will be parsed
168 * accordingly - setting the scheme, userinfo, host,port, path, query
169 * string and fragment fields as necessary. If the specification does
170 * not follow the "generic URI" syntax, the specification is parsed
171 * into a scheme and scheme-specific part (stored as the path) only.
172 *
173 * @param p_uriSpec the URI specification string (cannot be null or
174 * empty)
175 *
176 * @throws MalformedURIException if p_uriSpec violates any syntax
177 * rules
178 */
179 public URI(String p_uriSpec) throws MalformedURIException
180 {
181 this((URI) null, p_uriSpec);
182 }
183
184 /**
185 * Construct a new URI from a base URI and a URI specification string.
186 * The URI specification string may be a relative URI.
187 *
188 * @param p_base the base URI (cannot be null if p_uriSpec is null or
189 * empty)
190 * @param p_uriSpec the URI specification string (cannot be null or
191 * empty if p_base is null)
192 *
193 * @throws MalformedURIException if p_uriSpec violates any syntax
194 * rules
195 */
196 public URI(URI p_base, String p_uriSpec) throws MalformedURIException
197 {
198 initialize(p_base, p_uriSpec);
199 }
200
201 /**
202 * Construct a new URI that does not follow the generic URI syntax.
203 * Only the scheme and scheme-specific part (stored as the path) are
204 * initialized.
205 *
206 * @param p_scheme the URI scheme (cannot be null or empty)
207 * @param p_schemeSpecificPart the scheme-specific part (cannot be
208 * null or empty)
209 *
210 * @throws MalformedURIException if p_scheme violates any
211 * syntax rules
212 */
213 public URI(String p_scheme, String p_schemeSpecificPart)
214 throws MalformedURIException
215 {
216
217 if (p_scheme == null || p_scheme.trim().length() == 0)
218 {
219 throw new MalformedURIException(
220 "Cannot construct URI with null/empty scheme!");
221 }
222
223 if (p_schemeSpecificPart == null
224 || p_schemeSpecificPart.trim().length() == 0)
225 {
226 throw new MalformedURIException(
227 "Cannot construct URI with null/empty scheme-specific part!");
228 }
229
230 setScheme(p_scheme);
231 setPath(p_schemeSpecificPart);
232 }
233
234 /**
235 * Construct a new URI that follows the generic URI syntax from its
236 * component parts. Each component is validated for syntax and some
237 * basic semantic checks are performed as well. See the individual
238 * setter methods for specifics.
239 *
240 * @param p_scheme the URI scheme (cannot be null or empty)
241 * @param p_host the hostname or IPv4 address for the URI
242 * @param p_path the URI path - if the path contains '?' or '#',
243 * then the query string and/or fragment will be
244 * set from the path; however, if the query and
245 * fragment are specified both in the path and as
246 * separate parameters, an exception is thrown
247 * @param p_queryString the URI query string (cannot be specified
248 * if path is null)
249 * @param p_fragment the URI fragment (cannot be specified if path
250 * is null)
251 *
252 * @throws MalformedURIException if any of the parameters violates
253 * syntax rules or semantic rules
254 */
255 public URI(String p_scheme, String p_host, String p_path, String p_queryString, String p_fragment)
256 throws MalformedURIException
257 {
258 this(p_scheme, null, p_host, -1, p_path, p_queryString, p_fragment);
259 }
260
261 /**
262 * Construct a new URI that follows the generic URI syntax from its
263 * component parts. Each component is validated for syntax and some
264 * basic semantic checks are performed as well. See the individual
265 * setter methods for specifics.
266 *
267 * @param p_scheme the URI scheme (cannot be null or empty)
268 * @param p_userinfo the URI userinfo (cannot be specified if host
269 * is null)
270 * @param p_host the hostname or IPv4 address for the URI
271 * @param p_port the URI port (may be -1 for "unspecified"; cannot
272 * be specified if host is null)
273 * @param p_path the URI path - if the path contains '?' or '#',
274 * then the query string and/or fragment will be
275 * set from the path; however, if the query and
276 * fragment are specified both in the path and as
277 * separate parameters, an exception is thrown
278 * @param p_queryString the URI query string (cannot be specified
279 * if path is null)
280 * @param p_fragment the URI fragment (cannot be specified if path
281 * is null)
282 *
283 * @throws MalformedURIException if any of the parameters violates
284 * syntax rules or semantic rules
285 */
286 public URI(String p_scheme, String p_userinfo, String p_host, int p_port, String p_path, String p_queryString, String p_fragment)
287 throws MalformedURIException
288 {
289
290 if (p_scheme == null || p_scheme.trim().length() == 0)
291 {
292 throw new MalformedURIException(XMLMessages.createXMLMessage(XMLErrorResources.ER_SCHEME_REQUIRED, null)); //"Scheme is required!");
293 }
294
295 if (p_host == null)
296 {
297 if (p_userinfo != null)
298 {
299 throw new MalformedURIException(
300 XMLMessages.createXMLMessage(XMLErrorResources.ER_NO_USERINFO_IF_NO_HOST, null)); //"Userinfo may not be specified if host is not specified!");
301 }
302
303 if (p_port != -1)
304 {
305 throw new MalformedURIException(
306 XMLMessages.createXMLMessage(XMLErrorResources.ER_NO_PORT_IF_NO_HOST, null)); //"Port may not be specified if host is not specified!");
307 }
308 }
309
310 if (p_path != null)
311 {
312 if (p_path.indexOf('?') != -1 && p_queryString != null)
313 {
314 throw new MalformedURIException(
315 XMLMessages.createXMLMessage(XMLErrorResources.ER_NO_QUERY_STRING_IN_PATH, null)); //"Query string cannot be specified in path and query string!");
316 }
317
318 if (p_path.indexOf('#') != -1 && p_fragment != null)
319 {
320 throw new MalformedURIException(
321 XMLMessages.createXMLMessage(XMLErrorResources.ER_NO_FRAGMENT_STRING_IN_PATH, null)); //"Fragment cannot be specified in both the path and fragment!");
322 }
323 }
324
325 setScheme(p_scheme);
326 setHost(p_host);
327 setPort(p_port);
328 setUserinfo(p_userinfo);
329 setPath(p_path);
330 setQueryString(p_queryString);
331 setFragment(p_fragment);
332 }
333
334 /**
335 * Initialize all fields of this URI from another URI.
336 *
337 * @param p_other the URI to copy (cannot be null)
338 */
339 private void initialize(URI p_other)
340 {
341
342 m_scheme = p_other.getScheme();
343 m_userinfo = p_other.getUserinfo();
344 m_host = p_other.getHost();
345 m_port = p_other.getPort();
346 m_path = p_other.getPath();
347 m_queryString = p_other.getQueryString();
348 m_fragment = p_other.getFragment();
349 }
350
351 /**
352 * Initializes this URI from a base URI and a URI specification string.
353 * See RFC 2396 Section 4 and Appendix B for specifications on parsing
354 * the URI and Section 5 for specifications on resolving relative URIs
355 * and relative paths.
356 *
357 * @param p_base the base URI (may be null if p_uriSpec is an absolute
358 * URI)
359 * @param p_uriSpec the URI spec string which may be an absolute or
360 * relative URI (can only be null/empty if p_base
361 * is not null)
362 *
363 * @throws MalformedURIException if p_base is null and p_uriSpec
364 * is not an absolute URI or if
365 * p_uriSpec violates syntax rules
366 */
367 private void initialize(URI p_base, String p_uriSpec)
368 throws MalformedURIException
369 {
370
371 if (p_base == null
372 && (p_uriSpec == null || p_uriSpec.trim().length() == 0))
373 {
374 throw new MalformedURIException(
375 XMLMessages.createXMLMessage(XMLErrorResources.ER_CANNOT_INIT_URI_EMPTY_PARMS, null)); //"Cannot initialize URI with empty parameters.");
376 }
377
378 // just make a copy of the base if spec is empty
379 if (p_uriSpec == null || p_uriSpec.trim().length() == 0)
380 {
381 initialize(p_base);
382
383 return;
384 }
385
386 String uriSpec = p_uriSpec.trim();
387 int uriSpecLen = uriSpec.length();
388 int index = 0;
389
390 // check for scheme
391 int colonIndex = uriSpec.indexOf(':');
392 if (colonIndex < 0)
393 {
394 if (p_base == null)
395 {
396 throw new MalformedURIException(XMLMessages.createXMLMessage(XMLErrorResources.ER_NO_SCHEME_IN_URI, new Object[]{uriSpec})); //"No scheme found in URI: "+uriSpec);
397 }
398 }
399 else
400 {
401 initializeScheme(uriSpec);
402 uriSpec = uriSpec.substring(colonIndex+1);
403 // This is a fix for XALANJ-2059.
404 if(m_scheme != null && p_base != null)
405 {
406 // a) If <uriSpec> starts with a slash (/), it means <uriSpec> is absolute
407 // and p_base can be ignored.
408 // For example,
409 // uriSpec = file:/myDIR/myXSLFile.xsl
410 // p_base = file:/myWork/
411 //
412 // Here, uriSpec has absolute path after scheme file and :
413 // Hence p_base can be ignored.
414 //
415 // b) Similarily, according to RFC 2396, uri is resolved for <uriSpec> relative to <p_base>
416 // if scheme in <uriSpec> is same as scheme in <p_base>, else p_base can be ignored.
417 //
418 // c) if <p_base> is not hierarchical, it can be ignored.
419 //
420 if(uriSpec.startsWith("/") || !m_scheme.equals(p_base.m_scheme) || !p_base.getSchemeSpecificPart().startsWith("/"))
421 {
422 p_base = null;
423 }
424 }
425 // Fix for XALANJ-2059
426 uriSpecLen = uriSpec.length();
427 }
428
429 // two slashes means generic URI syntax, so we get the authority
430 if (((index + 1) < uriSpecLen)
431 && (uriSpec.substring(index).startsWith("//")))
432 {
433 index += 2;
434
435 int startPos = index;
436
437 // get authority - everything up to path, query or fragment
438 char testChar = '\0';
439
440 while (index < uriSpecLen)
441 {
442 testChar = uriSpec.charAt(index);
443
444 if (testChar == '/' || testChar == '?' || testChar == '#')
445 {
446 break;
447 }
448
449 index++;
450 }
451
452 // if we found authority, parse it out, otherwise we set the
453 // host to empty string
454 if (index > startPos)
455 {
456 initializeAuthority(uriSpec.substring(startPos, index));
457 }
458 else
459 {
460 m_host = "";
461 }
462 }
463
464 initializePath(uriSpec.substring(index));
465
466 // Resolve relative URI to base URI - see RFC 2396 Section 5.2
467 // In some cases, it might make more sense to throw an exception
468 // (when scheme is specified is the string spec and the base URI
469 // is also specified, for example), but we're just following the
470 // RFC specifications
471 if (p_base != null)
472 {
473
474 // check to see if this is the current doc - RFC 2396 5.2 #2
475 // note that this is slightly different from the RFC spec in that
476 // we don't include the check for query string being null
477 // - this handles cases where the urispec is just a query
478 // string or a fragment (e.g. "?y" or "#s") -
479 // see <http://www.ics.uci.edu/~fielding/url/test1.html> which
480 // identified this as a bug in the RFC
481 if (m_path.length() == 0 && m_scheme == null && m_host == null)
482 {
483 m_scheme = p_base.getScheme();
484 m_userinfo = p_base.getUserinfo();
485 m_host = p_base.getHost();
486 m_port = p_base.getPort();
487 m_path = p_base.getPath();
488
489 if (m_queryString == null)
490 {
491 m_queryString = p_base.getQueryString();
492 }
493
494 return;
495 }
496
497 // check for scheme - RFC 2396 5.2 #3
498 // if we found a scheme, it means absolute URI, so we're done
499 if (m_scheme == null)
500 {
501 m_scheme = p_base.getScheme();
502 }
503
504 // check for authority - RFC 2396 5.2 #4
505 // if we found a host, then we've got a network path, so we're done
506 if (m_host == null)
507 {
508 m_userinfo = p_base.getUserinfo();
509 m_host = p_base.getHost();
510 m_port = p_base.getPort();
511 }
512 else
513 {
514 return;
515 }
516
517 // check for absolute path - RFC 2396 5.2 #5
518 if (m_path.length() > 0 && m_path.startsWith("/"))
519 {
520 return;
521 }
522
523 // if we get to this point, we need to resolve relative path
524 // RFC 2396 5.2 #6
525 String path = "";
526 String basePath = p_base.getPath();
527
528 // 6a - get all but the last segment of the base URI path
529 if (basePath != null)
530 {
531 int lastSlash = basePath.lastIndexOf('/');
532
533 if (lastSlash != -1)
534 {
535 path = basePath.substring(0, lastSlash + 1);
536 }
537 }
538
539 // 6b - append the relative URI path
540 path = path.concat(m_path);
541
542 // 6c - remove all "./" where "." is a complete path segment
543 index = -1;
544
545 while ((index = path.indexOf("/./")) != -1)
546 {
547 path = path.substring(0, index + 1).concat(path.substring(index + 3));
548 }
549
550 // 6d - remove "." if path ends with "." as a complete path segment
551 if (path.endsWith("/."))
552 {
553 path = path.substring(0, path.length() - 1);
554 }
555
556 // 6e - remove all "<segment>/../" where "<segment>" is a complete
557 // path segment not equal to ".."
558 index = -1;
559
560 int segIndex = -1;
561 String tempString = null;
562
563 while ((index = path.indexOf("/../")) > 0)
564 {
565 tempString = path.substring(0, path.indexOf("/../"));
566 segIndex = tempString.lastIndexOf('/');
567
568 if (segIndex != -1)
569 {
570 if (!tempString.substring(segIndex++).equals(".."))
571 {
572 path = path.substring(0, segIndex).concat(path.substring(index
573 + 4));
574 }
575 }
576 }
577
578 // 6f - remove ending "<segment>/.." where "<segment>" is a
579 // complete path segment
580 if (path.endsWith("/.."))
581 {
582 tempString = path.substring(0, path.length() - 3);
583 segIndex = tempString.lastIndexOf('/');
584
585 if (segIndex != -1)
586 {
587 path = path.substring(0, segIndex + 1);
588 }
589 }
590
591 m_path = path;
592 }
593 }
594
595 /**
596 * Initialize the scheme for this URI from a URI string spec.
597 *
598 * @param p_uriSpec the URI specification (cannot be null)
599 *
600 * @throws MalformedURIException if URI does not have a conformant
601 * scheme
602 */
603 private void initializeScheme(String p_uriSpec) throws MalformedURIException
604 {
605
606 int uriSpecLen = p_uriSpec.length();
607 int index = 0;
608 String scheme = null;
609 char testChar = '\0';
610
611 while (index < uriSpecLen)
612 {
613 testChar = p_uriSpec.charAt(index);
614
615 if (testChar == ':' || testChar == '/' || testChar == '?'
616 || testChar == '#')
617 {
618 break;
619 }
620
621 index++;
622 }
623
624 scheme = p_uriSpec.substring(0, index);
625
626 if (scheme.length() == 0)
627 {
628 throw new MalformedURIException(XMLMessages.createXMLMessage(XMLErrorResources.ER_NO_SCHEME_INURI, null)); //"No scheme found in URI.");
629 }
630 else
631 {
632 setScheme(scheme);
633 }
634 }
635
636 /**
637 * Initialize the authority (userinfo, host and port) for this
638 * URI from a URI string spec.
639 *
640 * @param p_uriSpec the URI specification (cannot be null)
641 *
642 * @throws MalformedURIException if p_uriSpec violates syntax rules
643 */
644 private void initializeAuthority(String p_uriSpec)
645 throws MalformedURIException
646 {
647
648 int index = 0;
649 int start = 0;
650 int end = p_uriSpec.length();
651 char testChar = '\0';
652 String userinfo = null;
653
654 // userinfo is everything up @
655 if (p_uriSpec.indexOf('@', start) != -1)
656 {
657 while (index < end)
658 {
659 testChar = p_uriSpec.charAt(index);
660
661 if (testChar == '@')
662 {
663 break;
664 }
665
666 index++;
667 }
668
669 userinfo = p_uriSpec.substring(start, index);
670
671 index++;
672 }
673
674 // host is everything up to ':'
675 String host = null;
676
677 start = index;
678
679 while (index < end)
680 {
681 testChar = p_uriSpec.charAt(index);
682
683 if (testChar == ':')
684 {
685 break;
686 }
687
688 index++;
689 }
690
691 host = p_uriSpec.substring(start, index);
692
693 int port = -1;
694
695 if (host.length() > 0)
696 {
697
698 // port
699 if (testChar == ':')
700 {
701 index++;
702
703 start = index;
704
705 while (index < end)
706 {
707 index++;
708 }
709
710 String portStr = p_uriSpec.substring(start, index);
711
712 if (portStr.length() > 0)
713 {
714 for (int i = 0; i < portStr.length(); i++)
715 {
716 if (!isDigit(portStr.charAt(i)))
717 {
718 throw new MalformedURIException(
719 portStr + " is invalid. Port should only contain digits!");
720 }
721 }
722
723 try
724 {
725 port = Integer.parseInt(portStr);
726 }
727 catch (NumberFormatException nfe)
728 {
729
730 // can't happen
731 }
732 }
733 }
734 }
735
736 setHost(host);
737 setPort(port);
738 setUserinfo(userinfo);
739 }
740
741 /**
742 * Initialize the path for this URI from a URI string spec.
743 *
744 * @param p_uriSpec the URI specification (cannot be null)
745 *
746 * @throws MalformedURIException if p_uriSpec violates syntax rules
747 */
748 private void initializePath(String p_uriSpec) throws MalformedURIException
749 {
750
751 if (p_uriSpec == null)
752 {
753 throw new MalformedURIException(
754 "Cannot initialize path from null string!");
755 }
756
757 int index = 0;
758 int start = 0;
759 int end = p_uriSpec.length();
760 char testChar = '\0';
761
762 // path - everything up to query string or fragment
763 while (index < end)
764 {
765 testChar = p_uriSpec.charAt(index);
766
767 if (testChar == '?' || testChar == '#')
768 {
769 break;
770 }
771
772 // check for valid escape sequence
773 if (testChar == '%')
774 {
775 if (index + 2 >= end ||!isHex(p_uriSpec.charAt(index + 1))
776 ||!isHex(p_uriSpec.charAt(index + 2)))
777 {
778 throw new MalformedURIException(
779 XMLMessages.createXMLMessage(XMLErrorResources.ER_PATH_CONTAINS_INVALID_ESCAPE_SEQUENCE, null)); //"Path contains invalid escape sequence!");
780 }
781 }
782 else if (!isReservedCharacter(testChar)
783 &&!isUnreservedCharacter(testChar))
784 {
785 if ('\\' != testChar)
786 throw new MalformedURIException(XMLMessages.createXMLMessage(XMLErrorResources.ER_PATH_INVALID_CHAR, new Object[]{String.valueOf(testChar)})); //"Path contains invalid character: "
787 //+ testChar);
788 }
789
790 index++;
791 }
792
793 m_path = p_uriSpec.substring(start, index);
794
795 // query - starts with ? and up to fragment or end
796 if (testChar == '?')
797 {
798 index++;
799
800 start = index;
801
802 while (index < end)
803 {
804 testChar = p_uriSpec.charAt(index);
805
806 if (testChar == '#')
807 {
808 break;
809 }
810
811 if (testChar == '%')
812 {
813 if (index + 2 >= end ||!isHex(p_uriSpec.charAt(index + 1))
814 ||!isHex(p_uriSpec.charAt(index + 2)))
815 {
816 throw new MalformedURIException(
817 "Query string contains invalid escape sequence!");
818 }
819 }
820 else if (!isReservedCharacter(testChar)
821 &&!isUnreservedCharacter(testChar))
822 {
823 throw new MalformedURIException(
824 "Query string contains invalid character:" + testChar);
825 }
826
827 index++;
828 }
829
830 m_queryString = p_uriSpec.substring(start, index);
831 }
832
833 // fragment - starts with #
834 if (testChar == '#')
835 {
836 index++;
837
838 start = index;
839
840 while (index < end)
841 {
842 testChar = p_uriSpec.charAt(index);
843
844 if (testChar == '%')
845 {
846 if (index + 2 >= end ||!isHex(p_uriSpec.charAt(index + 1))
847 ||!isHex(p_uriSpec.charAt(index + 2)))
848 {
849 throw new MalformedURIException(
850 "Fragment contains invalid escape sequence!");
851 }
852 }
853 else if (!isReservedCharacter(testChar)
854 &&!isUnreservedCharacter(testChar))
855 {
856 throw new MalformedURIException(
857 "Fragment contains invalid character:" + testChar);
858 }
859
860 index++;
861 }
862
863 m_fragment = p_uriSpec.substring(start, index);
864 }
865 }
866
867 /**
868 * Get the scheme for this URI.
869 *
870 * @return the scheme for this URI
871 */
872 public String getScheme()
873 {
874 return m_scheme;
875 }
876
877 /**
878 * Get the scheme-specific part for this URI (everything following the
879 * scheme and the first colon). See RFC 2396 Section 5.2 for spec.
880 *
881 * @return the scheme-specific part for this URI
882 */
883 public String getSchemeSpecificPart()
884 {
885
886 StringBuffer schemespec = new StringBuffer();
887
888 if (m_userinfo != null || m_host != null || m_port != -1)
889 {
890 schemespec.append("//");
891 }
892
893 if (m_userinfo != null)
894 {
895 schemespec.append(m_userinfo);
896 schemespec.append('@');
897 }
898
899 if (m_host != null)
900 {
901 schemespec.append(m_host);
902 }
903
904 if (m_port != -1)
905 {
906 schemespec.append(':');
907 schemespec.append(m_port);
908 }
909
910 if (m_path != null)
911 {
912 schemespec.append((m_path));
913 }
914
915 if (m_queryString != null)
916 {
917 schemespec.append('?');
918 schemespec.append(m_queryString);
919 }
920
921 if (m_fragment != null)
922 {
923 schemespec.append('#');
924 schemespec.append(m_fragment);
925 }
926
927 return schemespec.toString();
928 }
929
930 /**
931 * Get the userinfo for this URI.
932 *
933 * @return the userinfo for this URI (null if not specified).
934 */
935 public String getUserinfo()
936 {
937 return m_userinfo;
938 }
939
940 /**
941 * Get the host for this URI.
942 *
943 * @return the host for this URI (null if not specified).
944 */
945 public String getHost()
946 {
947 return m_host;
948 }
949
950 /**
951 * Get the port for this URI.
952 *
953 * @return the port for this URI (-1 if not specified).
954 */
955 public int getPort()
956 {
957 return m_port;
958 }
959
960 /**
961 * Get the path for this URI (optionally with the query string and
962 * fragment).
963 *
964 * @param p_includeQueryString if true (and query string is not null),
965 * then a "?" followed by the query string
966 * will be appended
967 * @param p_includeFragment if true (and fragment is not null),
968 * then a "#" followed by the fragment
969 * will be appended
970 *
971 * @return the path for this URI possibly including the query string
972 * and fragment
973 */
974 public String getPath(boolean p_includeQueryString,
975 boolean p_includeFragment)
976 {
977
978 StringBuffer pathString = new StringBuffer(m_path);
979
980 if (p_includeQueryString && m_queryString != null)
981 {
982 pathString.append('?');
983 pathString.append(m_queryString);
984 }
985
986 if (p_includeFragment && m_fragment != null)
987 {
988 pathString.append('#');
989 pathString.append(m_fragment);
990 }
991
992 return pathString.toString();
993 }
994
995 /**
996 * Get the path for this URI. Note that the value returned is the path
997 * only and does not include the query string or fragment.
998 *
999 * @return the path for this URI.
1000 */
1001 public String getPath()
1002 {
1003 return m_path;
1004 }
1005
1006 /**
1007 * Get the query string for this URI.
1008 *
1009 * @return the query string for this URI. Null is returned if there
1010 * was no "?" in the URI spec, empty string if there was a
1011 * "?" but no query string following it.
1012 */
1013 public String getQueryString()
1014 {
1015 return m_queryString;
1016 }
1017
1018 /**
1019 * Get the fragment for this URI.
1020 *
1021 * @return the fragment for this URI. Null is returned if there
1022 * was no "#" in the URI spec, empty string if there was a
1023 * "#" but no fragment following it.
1024 */
1025 public String getFragment()
1026 {
1027 return m_fragment;
1028 }
1029
1030 /**
1031 * Set the scheme for this URI. The scheme is converted to lowercase
1032 * before it is set.
1033 *
1034 * @param p_scheme the scheme for this URI (cannot be null)
1035 *
1036 * @throws MalformedURIException if p_scheme is not a conformant
1037 * scheme name
1038 */
1039 public void setScheme(String p_scheme) throws MalformedURIException
1040 {
1041
1042 if (p_scheme == null)
1043 {
1044 throw new MalformedURIException(XMLMessages.createXMLMessage(XMLErrorResources.ER_SCHEME_FROM_NULL_STRING, null)); //"Cannot set scheme from null string!");
1045 }
1046
1047 if (!isConformantSchemeName(p_scheme))
1048 {
1049 throw new MalformedURIException(XMLMessages.createXMLMessage(XMLErrorResources.ER_SCHEME_NOT_CONFORMANT, null)); //"The scheme is not conformant.");
1050 }
1051
1052 m_scheme = p_scheme.toLowerCase();
1053 }
1054
1055 /**
1056 * Set the userinfo for this URI. If a non-null value is passed in and
1057 * the host value is null, then an exception is thrown.
1058 *
1059 * @param p_userinfo the userinfo for this URI
1060 *
1061 * @throws MalformedURIException if p_userinfo contains invalid
1062 * characters
1063 */
1064 public void setUserinfo(String p_userinfo) throws MalformedURIException
1065 {
1066
1067 if (p_userinfo == null)
1068 {
1069 m_userinfo = null;
1070 }
1071 else
1072 {
1073 if (m_host == null)
1074 {
1075 throw new MalformedURIException(
1076 "Userinfo cannot be set when host is null!");
1077 }
1078
1079 // userinfo can contain alphanumerics, mark characters, escaped
1080 // and ';',':','&','=','+','$',','
1081 int index = 0;
1082 int end = p_userinfo.length();
1083 char testChar = '\0';
1084
1085 while (index < end)
1086 {
1087 testChar = p_userinfo.charAt(index);
1088
1089 if (testChar == '%')
1090 {
1091 if (index + 2 >= end ||!isHex(p_userinfo.charAt(index + 1))
1092 ||!isHex(p_userinfo.charAt(index + 2)))
1093 {
1094 throw new MalformedURIException(
1095 "Userinfo contains invalid escape sequence!");
1096 }
1097 }
1098 else if (!isUnreservedCharacter(testChar)
1099 && USERINFO_CHARACTERS.indexOf(testChar) == -1)
1100 {
1101 throw new MalformedURIException(
1102 "Userinfo contains invalid character:" + testChar);
1103 }
1104
1105 index++;
1106 }
1107 }
1108
1109 m_userinfo = p_userinfo;
1110 }
1111
1112 /**
1113 * Set the host for this URI. If null is passed in, the userinfo
1114 * field is also set to null and the port is set to -1.
1115 *
1116 * @param p_host the host for this URI
1117 *
1118 * @throws MalformedURIException if p_host is not a valid IP
1119 * address or DNS hostname.
1120 */
1121 public void setHost(String p_host) throws MalformedURIException
1122 {
1123
1124 if (p_host == null || p_host.trim().length() == 0)
1125 {
1126 m_host = p_host;
1127 m_userinfo = null;
1128 m_port = -1;
1129 }
1130 else if (!isWellFormedAddress(p_host))
1131 {
1132 throw new MalformedURIException(XMLMessages.createXMLMessage(XMLErrorResources.ER_HOST_ADDRESS_NOT_WELLFORMED, null)); //"Host is not a well formed address!");
1133 }
1134
1135 m_host = p_host;
1136 }
1137
1138 /**
1139 * Set the port for this URI. -1 is used to indicate that the port is
1140 * not specified, otherwise valid port numbers are between 0 and 65535.
1141 * If a valid port number is passed in and the host field is null,
1142 * an exception is thrown.
1143 *
1144 * @param p_port the port number for this URI
1145 *
1146 * @throws MalformedURIException if p_port is not -1 and not a
1147 * valid port number
1148 */
1149 public void setPort(int p_port) throws MalformedURIException
1150 {
1151
1152 if (p_port >= 0 && p_port <= 65535)
1153 {
1154 if (m_host == null)
1155 {
1156 throw new MalformedURIException(
1157 XMLMessages.createXMLMessage(XMLErrorResources.ER_PORT_WHEN_HOST_NULL, null)); //"Port cannot be set when host is null!");
1158 }
1159 }
1160 else if (p_port != -1)
1161 {
1162 throw new MalformedURIException(XMLMessages.createXMLMessage(XMLErrorResources.ER_INVALID_PORT, null)); //"Invalid port number!");
1163 }
1164
1165 m_port = p_port;
1166 }
1167
1168 /**
1169 * Set the path for this URI. If the supplied path is null, then the
1170 * query string and fragment are set to null as well. If the supplied
1171 * path includes a query string and/or fragment, these fields will be
1172 * parsed and set as well. Note that, for URIs following the "generic
1173 * URI" syntax, the path specified should start with a slash.
1174 * For URIs that do not follow the generic URI syntax, this method
1175 * sets the scheme-specific part.
1176 *
1177 * @param p_path the path for this URI (may be null)
1178 *
1179 * @throws MalformedURIException if p_path contains invalid
1180 * characters
1181 */
1182 public void setPath(String p_path) throws MalformedURIException
1183 {
1184
1185 if (p_path == null)
1186 {
1187 m_path = null;
1188 m_queryString = null;
1189 m_fragment = null;
1190 }
1191 else
1192 {
1193 initializePath(p_path);
1194 }
1195 }
1196
1197 /**
1198 * Append to the end of the path of this URI. If the current path does
1199 * not end in a slash and the path to be appended does not begin with
1200 * a slash, a slash will be appended to the current path before the
1201 * new segment is added. Also, if the current path ends in a slash
1202 * and the new segment begins with a slash, the extra slash will be
1203 * removed before the new segment is appended.
1204 *
1205 * @param p_addToPath the new segment to be added to the current path
1206 *
1207 * @throws MalformedURIException if p_addToPath contains syntax
1208 * errors
1209 */
1210 public void appendPath(String p_addToPath) throws MalformedURIException
1211 {
1212
1213 if (p_addToPath == null || p_addToPath.trim().length() == 0)
1214 {
1215 return;
1216 }
1217
1218 if (!isURIString(p_addToPath))
1219 {
1220 throw new MalformedURIException(XMLMessages.createXMLMessage(XMLErrorResources.ER_PATH_INVALID_CHAR, new Object[]{p_addToPath})); //"Path contains invalid character!");
1221 }
1222
1223 if (m_path == null || m_path.trim().length() == 0)
1224 {
1225 if (p_addToPath.startsWith("/"))
1226 {
1227 m_path = p_addToPath;
1228 }
1229 else
1230 {
1231 m_path = "/" + p_addToPath;
1232 }
1233 }
1234 else if (m_path.endsWith("/"))
1235 {
1236 if (p_addToPath.startsWith("/"))
1237 {
1238 m_path = m_path.concat(p_addToPath.substring(1));
1239 }
1240 else
1241 {
1242 m_path = m_path.concat(p_addToPath);
1243 }
1244 }
1245 else
1246 {
1247 if (p_addToPath.startsWith("/"))
1248 {
1249 m_path = m_path.concat(p_addToPath);
1250 }
1251 else
1252 {
1253 m_path = m_path.concat("/" + p_addToPath);
1254 }
1255 }
1256 }
1257
1258 /**
1259 * Set the query string for this URI. A non-null value is valid only
1260 * if this is an URI conforming to the generic URI syntax and
1261 * the path value is not null.
1262 *
1263 * @param p_queryString the query string for this URI
1264 *
1265 * @throws MalformedURIException if p_queryString is not null and this
1266 * URI does not conform to the generic
1267 * URI syntax or if the path is null
1268 */
1269 public void setQueryString(String p_queryString)
1270 throws MalformedURIException
1271 {
1272
1273 if (p_queryString == null)
1274 {
1275 m_queryString = null;
1276 }
1277 else if (!isGenericURI())
1278 {
1279 throw new MalformedURIException(
1280 "Query string can only be set for a generic URI!");
1281 }
1282 else if (getPath() == null)
1283 {
1284 throw new MalformedURIException(
1285 "Query string cannot be set when path is null!");
1286 }
1287 else if (!isURIString(p_queryString))
1288 {
1289 throw new MalformedURIException(
1290 "Query string contains invalid character!");
1291 }
1292 else
1293 {
1294 m_queryString = p_queryString;
1295 }
1296 }
1297
1298 /**
1299 * Set the fragment for this URI. A non-null value is valid only
1300 * if this is a URI conforming to the generic URI syntax and
1301 * the path value is not null.
1302 *
1303 * @param p_fragment the fragment for this URI
1304 *
1305 * @throws MalformedURIException if p_fragment is not null and this
1306 * URI does not conform to the generic
1307 * URI syntax or if the path is null
1308 */
1309 public void setFragment(String p_fragment) throws MalformedURIException
1310 {
1311
1312 if (p_fragment == null)
1313 {
1314 m_fragment = null;
1315 }
1316 else if (!isGenericURI())
1317 {
1318 throw new MalformedURIException(
1319 XMLMessages.createXMLMessage(XMLErrorResources.ER_FRAG_FOR_GENERIC_URI, null)); //"Fragment can only be set for a generic URI!");
1320 }
1321 else if (getPath() == null)
1322 {
1323 throw new MalformedURIException(
1324 XMLMessages.createXMLMessage(XMLErrorResources.ER_FRAG_WHEN_PATH_NULL, null)); //"Fragment cannot be set when path is null!");
1325 }
1326 else if (!isURIString(p_fragment))
1327 {
1328 throw new MalformedURIException(XMLMessages.createXMLMessage(XMLErrorResources.ER_FRAG_INVALID_CHAR, null)); //"Fragment contains invalid character!");
1329 }
1330 else
1331 {
1332 m_fragment = p_fragment;
1333 }
1334 }
1335
1336 /**
1337 * Determines if the passed-in Object is equivalent to this URI.
1338 *
1339 * @param p_test the Object to test for equality.
1340 *
1341 * @return true if p_test is a URI with all values equal to this
1342 * URI, false otherwise
1343 */
1344 public boolean equals(Object p_test)
1345 {
1346
1347 if (p_test instanceof URI)
1348 {
1349 URI testURI = (URI) p_test;
1350
1351 if (((m_scheme == null && testURI.m_scheme == null) || (m_scheme != null && testURI.m_scheme != null && m_scheme.equals(
1352 testURI.m_scheme))) && ((m_userinfo == null && testURI.m_userinfo == null) || (m_userinfo != null && testURI.m_userinfo != null && m_userinfo.equals(
1353 testURI.m_userinfo))) && ((m_host == null && testURI.m_host == null) || (m_host != null && testURI.m_host != null && m_host.equals(
1354 testURI.m_host))) && m_port == testURI.m_port && ((m_path == null && testURI.m_path == null) || (m_path != null && testURI.m_path != null && m_path.equals(
1355 testURI.m_path))) && ((m_queryString == null && testURI.m_queryString == null) || (m_queryString != null && testURI.m_queryString != null && m_queryString.equals(
1356 testURI.m_queryString))) && ((m_fragment == null && testURI.m_fragment == null) || (m_fragment != null && testURI.m_fragment != null && m_fragment.equals(
1357 testURI.m_fragment))))
1358 {
1359 return true;
1360 }
1361 }
1362
1363 return false;
1364 }
1365
1366 /**
1367 * Get the URI as a string specification. See RFC 2396 Section 5.2.
1368 *
1369 * @return the URI string specification
1370 */
1371 public String toString()
1372 {
1373
1374 StringBuffer uriSpecString = new StringBuffer();
1375
1376 if (m_scheme != null)
1377 {
1378 uriSpecString.append(m_scheme);
1379 uriSpecString.append(':');
1380 }
1381
1382 uriSpecString.append(getSchemeSpecificPart());
1383
1384 return uriSpecString.toString();
1385 }
1386
1387 /**
1388 * Get the indicator as to whether this URI uses the "generic URI"
1389 * syntax.
1390 *
1391 * @return true if this URI uses the "generic URI" syntax, false
1392 * otherwise
1393 */
1394 public boolean isGenericURI()
1395 {
1396
1397 // presence of the host (whether valid or empty) means
1398 // double-slashes which means generic uri
1399 return (m_host != null);
1400 }
1401
1402 /**
1403 * Determine whether a scheme conforms to the rules for a scheme name.
1404 * A scheme is conformant if it starts with an alphanumeric, and
1405 * contains only alphanumerics, '+','-' and '.'.
1406 *
1407 *
1408 * @param p_scheme The sheme name to check
1409 * @return true if the scheme is conformant, false otherwise
1410 */
1411 public static boolean isConformantSchemeName(String p_scheme)
1412 {
1413
1414 if (p_scheme == null || p_scheme.trim().length() == 0)
1415 {
1416 return false;
1417 }
1418
1419 if (!isAlpha(p_scheme.charAt(0)))
1420 {
1421 return false;
1422 }
1423
1424 char testChar;
1425
1426 for (int i = 1; i < p_scheme.length(); i++)
1427 {
1428 testChar = p_scheme.charAt(i);
1429
1430 if (!isAlphanum(testChar) && SCHEME_CHARACTERS.indexOf(testChar) == -1)
1431 {
1432 return false;
1433 }
1434 }
1435
1436 return true;
1437 }
1438
1439 /**
1440 * Determine whether a string is syntactically capable of representing
1441 * a valid IPv4 address or the domain name of a network host. A valid
1442 * IPv4 address consists of four decimal digit groups separated by a
1443 * '.'. A hostname consists of domain labels (each of which must
1444 * begin and end with an alphanumeric but may contain '-') separated
1445 * & by a '.'. See RFC 2396 Section 3.2.2.
1446 *
1447 *
1448 * @param p_address The address string to check
1449 * @return true if the string is a syntactically valid IPv4 address
1450 * or hostname
1451 */
1452 public static boolean isWellFormedAddress(String p_address)
1453 {
1454
1455 if (p_address == null)
1456 {
1457 return false;
1458 }
1459
1460 String address = p_address.trim();
1461 int addrLength = address.length();
1462
1463 if (addrLength == 0 || addrLength > 255)
1464 {
1465 return false;
1466 }
1467
1468 if (address.startsWith(".") || address.startsWith("-"))
1469 {
1470 return false;
1471 }
1472
1473 // rightmost domain label starting with digit indicates IP address
1474 // since top level domain label can only start with an alpha
1475 // see RFC 2396 Section 3.2.2
1476 int index = address.lastIndexOf('.');
1477
1478 if (address.endsWith("."))
1479 {
1480 index = address.substring(0, index).lastIndexOf('.');
1481 }
1482
1483 if (index + 1 < addrLength && isDigit(p_address.charAt(index + 1)))
1484 {
1485 char testChar;
1486 int numDots = 0;
1487
1488 // make sure that 1) we see only digits and dot separators, 2) that
1489 // any dot separator is preceded and followed by a digit and
1490 // 3) that we find 3 dots
1491 for (int i = 0; i < addrLength; i++)
1492 {
1493 testChar = address.charAt(i);
1494
1495 if (testChar == '.')
1496 {
1497 if (!isDigit(address.charAt(i - 1))
1498 || (i + 1 < addrLength &&!isDigit(address.charAt(i + 1))))
1499 {
1500 return false;
1501 }
1502
1503 numDots++;
1504 }
1505 else if (!isDigit(testChar))
1506 {
1507 return false;
1508 }
1509 }
1510
1511 if (numDots != 3)
1512 {
1513 return false;
1514 }
1515 }
1516 else
1517 {
1518
1519 // domain labels can contain alphanumerics and '-"
1520 // but must start and end with an alphanumeric
1521 char testChar;
1522
1523 for (int i = 0; i < addrLength; i++)
1524 {
1525 testChar = address.charAt(i);
1526
1527 if (testChar == '.')
1528 {
1529 if (!isAlphanum(address.charAt(i - 1)))
1530 {
1531 return false;
1532 }
1533
1534 if (i + 1 < addrLength &&!isAlphanum(address.charAt(i + 1)))
1535 {
1536 return false;
1537 }
1538 }
1539 else if (!isAlphanum(testChar) && testChar != '-')
1540 {
1541 return false;
1542 }
1543 }
1544 }
1545
1546 return true;
1547 }
1548
1549 /**
1550 * Determine whether a char is a digit.
1551 *
1552 *
1553 * @param p_char the character to check
1554 * @return true if the char is betweeen '0' and '9', false otherwise
1555 */
1556 private static boolean isDigit(char p_char)
1557 {
1558 return p_char >= '0' && p_char <= '9';
1559 }
1560
1561 /**
1562 * Determine whether a character is a hexadecimal character.
1563 *
1564 *
1565 * @param p_char the character to check
1566 * @return true if the char is betweeen '0' and '9', 'a' and 'f'
1567 * or 'A' and 'F', false otherwise
1568 */
1569 private static boolean isHex(char p_char)
1570 {
1571 return (isDigit(p_char) || (p_char >= 'a' && p_char <= 'f')
1572 || (p_char >= 'A' && p_char <= 'F'));
1573 }
1574
1575 /**
1576 * Determine whether a char is an alphabetic character: a-z or A-Z
1577 *
1578 *
1579 * @param p_char the character to check
1580 * @return true if the char is alphabetic, false otherwise
1581 */
1582 private static boolean isAlpha(char p_char)
1583 {
1584 return ((p_char >= 'a' && p_char <= 'z')
1585 || (p_char >= 'A' && p_char <= 'Z'));
1586 }
1587
1588 /**
1589 * Determine whether a char is an alphanumeric: 0-9, a-z or A-Z
1590 *
1591 *
1592 * @param p_char the character to check
1593 * @return true if the char is alphanumeric, false otherwise
1594 */
1595 private static boolean isAlphanum(char p_char)
1596 {
1597 return (isAlpha(p_char) || isDigit(p_char));
1598 }
1599
1600 /**
1601 * Determine whether a character is a reserved character:
1602 * ';', '/', '?', ':', '@', '&', '=', '+', '$' or ','
1603 *
1604 *
1605 * @param p_char the character to check
1606 * @return true if the string contains any reserved characters
1607 */
1608 private static boolean isReservedCharacter(char p_char)
1609 {
1610 return RESERVED_CHARACTERS.indexOf(p_char) != -1;
1611 }
1612
1613 /**
1614 * Determine whether a char is an unreserved character.
1615 *
1616 *
1617 * @param p_char the character to check
1618 * @return true if the char is unreserved, false otherwise
1619 */
1620 private static boolean isUnreservedCharacter(char p_char)
1621 {
1622 return (isAlphanum(p_char) || MARK_CHARACTERS.indexOf(p_char) != -1);
1623 }
1624
1625 /**
1626 * Determine whether a given string contains only URI characters (also
1627 * called "uric" in RFC 2396). uric consist of all reserved
1628 * characters, unreserved characters and escaped characters.
1629 *
1630 *
1631 * @param p_uric URI string
1632 * @return true if the string is comprised of uric, false otherwise
1633 */
1634 private static boolean isURIString(String p_uric)
1635 {
1636
1637 if (p_uric == null)
1638 {
1639 return false;
1640 }
1641
1642 int end = p_uric.length();
1643 char testChar = '\0';
1644
1645 for (int i = 0; i < end; i++)
1646 {
1647 testChar = p_uric.charAt(i);
1648
1649 if (testChar == '%')
1650 {
1651 if (i + 2 >= end ||!isHex(p_uric.charAt(i + 1))
1652 ||!isHex(p_uric.charAt(i + 2)))
1653 {
1654 return false;
1655 }
1656 else
1657 {
1658 i += 2;
1659
1660 continue;
1661 }
1662 }
1663
1664 if (isReservedCharacter(testChar) || isUnreservedCharacter(testChar))
1665 {
1666 continue;
1667 }
1668 else
1669 {
1670 return false;
1671 }
1672 }
1673
1674 return true;
1675 }
1676 }