001 /*
002 * Licensed to the Apache Software Foundation (ASF) under one
003 * or more contributor license agreements. See the NOTICE file
004 * distributed with this work for additional information
005 * regarding copyright ownership. The ASF licenses this file
006 * to you under the Apache License, Version 2.0 (the "License");
007 * you may not use this file except in compliance with the License.
008 * You may obtain a copy of the License at
009 *
010 * http://www.apache.org/licenses/LICENSE-2.0
011 *
012 * Unless required by applicable law or agreed to in writing, software
013 * distributed under the License is distributed on an "AS IS" BASIS,
014 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
015 * See the License for the specific language governing permissions and
016 * limitations under the License.
017 */
018 /*
019 * $Id: URI.java 468655 2006-10-28 07:12:06Z minchau $
020 */
021 package org.apache.xml.utils;
022
023 import java.io.IOException;
024 import java.io.Serializable;
025
026 import org.apache.xml.res.XMLErrorResources;
027 import org.apache.xml.res.XMLMessages;
028
029 /**
030 * A class to represent a Uniform Resource Identifier (URI). This class
031 * is designed to handle the parsing of URIs and provide access to
032 * the various components (scheme, host, port, userinfo, path, query
033 * string and fragment) that may constitute a URI.
034 * <p>
035 * Parsing of a URI specification is done according to the URI
036 * syntax described in RFC 2396
037 * <http://www.ietf.org/rfc/rfc2396.txt?number=2396>. Every URI consists
038 * of a scheme, followed by a colon (':'), followed by a scheme-specific
039 * part. For URIs that follow the "generic URI" syntax, the scheme-
040 * specific part begins with two slashes ("//") and may be followed
041 * by an authority segment (comprised of user information, host, and
042 * port), path segment, query segment and fragment. Note that RFC 2396
043 * no longer specifies the use of the parameters segment and excludes
044 * the "user:password" syntax as part of the authority segment. If
045 * "user:password" appears in a URI, the entire user/password string
046 * is stored as userinfo.
047 * <p>
048 * For URIs that do not follow the "generic URI" syntax (e.g. mailto),
049 * the entire scheme-specific part is treated as the "path" portion
050 * of the URI.
051 * <p>
052 * Note that, unlike the java.net.URL class, this class does not provide
053 * any built-in network access functionality nor does it provide any
054 * scheme-specific functionality (for example, it does not know a
055 * default port for a specific scheme). Rather, it only knows the
056 * grammar and basic set of operations that can be applied to a URI.
057 *
058 *
059 */
060 public class URI implements Serializable
061 {
062 static final long serialVersionUID = 7096266377907081897L;
063
064 /**
065 * MalformedURIExceptions are thrown in the process of building a URI
066 * or setting fields on a URI when an operation would result in an
067 * invalid URI specification.
068 *
069 */
070 public static class MalformedURIException extends IOException
071 {
072
073 /**
074 * Constructs a <code>MalformedURIException</code> with no specified
075 * detail message.
076 */
077 public MalformedURIException()
078 {
079 super();
080 }
081
082 /**
083 * Constructs a <code>MalformedURIException</code> with the
084 * specified detail message.
085 *
086 * @param p_msg the detail message.
087 */
088 public MalformedURIException(String p_msg)
089 {
090 super(p_msg);
091 }
092 }
093
094 /** reserved characters */
095 private static final String RESERVED_CHARACTERS = ";/?:@&=+$,";
096
097 /**
098 * URI punctuation mark characters - these, combined with
099 * alphanumerics, constitute the "unreserved" characters
100 */
101 private static final String MARK_CHARACTERS = "-_.!~*'() ";
102
103 /** scheme can be composed of alphanumerics and these characters */
104 private static final String SCHEME_CHARACTERS = "+-.";
105
106 /**
107 * userinfo can be composed of unreserved, escaped and these
108 * characters
109 */
110 private static final String USERINFO_CHARACTERS = ";:&=+$,";
111
112 /** Stores the scheme (usually the protocol) for this URI.
113 * @serial */
114 private String m_scheme = null;
115
116 /** If specified, stores the userinfo for this URI; otherwise null.
117 * @serial */
118 private String m_userinfo = null;
119
120 /** If specified, stores the host for this URI; otherwise null.
121 * @serial */
122 private String m_host = null;
123
124 /** If specified, stores the port for this URI; otherwise -1.
125 * @serial */
126 private int m_port = -1;
127
128 /** If specified, stores the path for this URI; otherwise null.
129 * @serial */
130 private String m_path = null;
131
132 /**
133 * If specified, stores the query string for this URI; otherwise
134 * null.
135 * @serial
136 */
137 private String m_queryString = null;
138
139 /** If specified, stores the fragment for this URI; otherwise null.
140 * @serial */
141 private String m_fragment = null;
142
143 /** Indicate whether in DEBUG mode */
144 private static boolean DEBUG = false;
145
146 /**
147 * Construct a new and uninitialized URI.
148 */
149 public URI(){}
150
151 /**
152 * Construct a new URI from another URI. All fields for this URI are
153 * set equal to the fields of the URI passed in.
154 *
155 * @param p_other the URI to copy (cannot be null)
156 */
157 public URI(URI p_other)
158 {
159 initialize(p_other);
160 }
161
162 /**
163 * Construct a new URI from a URI specification string. If the
164 * specification follows the "generic URI" syntax, (two slashes
165 * following the first colon), the specification will be parsed
166 * accordingly - setting the scheme, userinfo, host,port, path, query
167 * string and fragment fields as necessary. If the specification does
168 * not follow the "generic URI" syntax, the specification is parsed
169 * into a scheme and scheme-specific part (stored as the path) only.
170 *
171 * @param p_uriSpec the URI specification string (cannot be null or
172 * empty)
173 *
174 * @throws MalformedURIException if p_uriSpec violates any syntax
175 * rules
176 */
177 public URI(String p_uriSpec) throws MalformedURIException
178 {
179 this((URI) null, p_uriSpec);
180 }
181
182 /**
183 * Construct a new URI from a base URI and a URI specification string.
184 * The URI specification string may be a relative URI.
185 *
186 * @param p_base the base URI (cannot be null if p_uriSpec is null or
187 * empty)
188 * @param p_uriSpec the URI specification string (cannot be null or
189 * empty if p_base is null)
190 *
191 * @throws MalformedURIException if p_uriSpec violates any syntax
192 * rules
193 */
194 public URI(URI p_base, String p_uriSpec) throws MalformedURIException
195 {
196 initialize(p_base, p_uriSpec);
197 }
198
199 /**
200 * Construct a new URI that does not follow the generic URI syntax.
201 * Only the scheme and scheme-specific part (stored as the path) are
202 * initialized.
203 *
204 * @param p_scheme the URI scheme (cannot be null or empty)
205 * @param p_schemeSpecificPart the scheme-specific part (cannot be
206 * null or empty)
207 *
208 * @throws MalformedURIException if p_scheme violates any
209 * syntax rules
210 */
211 public URI(String p_scheme, String p_schemeSpecificPart)
212 throws MalformedURIException
213 {
214
215 if (p_scheme == null || p_scheme.trim().length() == 0)
216 {
217 throw new MalformedURIException(
218 "Cannot construct URI with null/empty scheme!");
219 }
220
221 if (p_schemeSpecificPart == null
222 || p_schemeSpecificPart.trim().length() == 0)
223 {
224 throw new MalformedURIException(
225 "Cannot construct URI with null/empty scheme-specific part!");
226 }
227
228 setScheme(p_scheme);
229 setPath(p_schemeSpecificPart);
230 }
231
232 /**
233 * Construct a new URI that follows the generic URI syntax from its
234 * component parts. Each component is validated for syntax and some
235 * basic semantic checks are performed as well. See the individual
236 * setter methods for specifics.
237 *
238 * @param p_scheme the URI scheme (cannot be null or empty)
239 * @param p_host the hostname or IPv4 address for the URI
240 * @param p_path the URI path - if the path contains '?' or '#',
241 * then the query string and/or fragment will be
242 * set from the path; however, if the query and
243 * fragment are specified both in the path and as
244 * separate parameters, an exception is thrown
245 * @param p_queryString the URI query string (cannot be specified
246 * if path is null)
247 * @param p_fragment the URI fragment (cannot be specified if path
248 * is null)
249 *
250 * @throws MalformedURIException if any of the parameters violates
251 * syntax rules or semantic rules
252 */
253 public URI(String p_scheme, String p_host, String p_path, String p_queryString, String p_fragment)
254 throws MalformedURIException
255 {
256 this(p_scheme, null, p_host, -1, p_path, p_queryString, p_fragment);
257 }
258
259 /**
260 * Construct a new URI that follows the generic URI syntax from its
261 * component parts. Each component is validated for syntax and some
262 * basic semantic checks are performed as well. See the individual
263 * setter methods for specifics.
264 *
265 * @param p_scheme the URI scheme (cannot be null or empty)
266 * @param p_userinfo the URI userinfo (cannot be specified if host
267 * is null)
268 * @param p_host the hostname or IPv4 address for the URI
269 * @param p_port the URI port (may be -1 for "unspecified"; cannot
270 * be specified if host is null)
271 * @param p_path the URI path - if the path contains '?' or '#',
272 * then the query string and/or fragment will be
273 * set from the path; however, if the query and
274 * fragment are specified both in the path and as
275 * separate parameters, an exception is thrown
276 * @param p_queryString the URI query string (cannot be specified
277 * if path is null)
278 * @param p_fragment the URI fragment (cannot be specified if path
279 * is null)
280 *
281 * @throws MalformedURIException if any of the parameters violates
282 * syntax rules or semantic rules
283 */
284 public URI(String p_scheme, String p_userinfo, String p_host, int p_port, String p_path, String p_queryString, String p_fragment)
285 throws MalformedURIException
286 {
287
288 if (p_scheme == null || p_scheme.trim().length() == 0)
289 {
290 throw new MalformedURIException(XMLMessages.createXMLMessage(XMLErrorResources.ER_SCHEME_REQUIRED, null)); //"Scheme is required!");
291 }
292
293 if (p_host == null)
294 {
295 if (p_userinfo != null)
296 {
297 throw new MalformedURIException(
298 XMLMessages.createXMLMessage(XMLErrorResources.ER_NO_USERINFO_IF_NO_HOST, null)); //"Userinfo may not be specified if host is not specified!");
299 }
300
301 if (p_port != -1)
302 {
303 throw new MalformedURIException(
304 XMLMessages.createXMLMessage(XMLErrorResources.ER_NO_PORT_IF_NO_HOST, null)); //"Port may not be specified if host is not specified!");
305 }
306 }
307
308 if (p_path != null)
309 {
310 if (p_path.indexOf('?') != -1 && p_queryString != null)
311 {
312 throw new MalformedURIException(
313 XMLMessages.createXMLMessage(XMLErrorResources.ER_NO_QUERY_STRING_IN_PATH, null)); //"Query string cannot be specified in path and query string!");
314 }
315
316 if (p_path.indexOf('#') != -1 && p_fragment != null)
317 {
318 throw new MalformedURIException(
319 XMLMessages.createXMLMessage(XMLErrorResources.ER_NO_FRAGMENT_STRING_IN_PATH, null)); //"Fragment cannot be specified in both the path and fragment!");
320 }
321 }
322
323 setScheme(p_scheme);
324 setHost(p_host);
325 setPort(p_port);
326 setUserinfo(p_userinfo);
327 setPath(p_path);
328 setQueryString(p_queryString);
329 setFragment(p_fragment);
330 }
331
332 /**
333 * Initialize all fields of this URI from another URI.
334 *
335 * @param p_other the URI to copy (cannot be null)
336 */
337 private void initialize(URI p_other)
338 {
339
340 m_scheme = p_other.getScheme();
341 m_userinfo = p_other.getUserinfo();
342 m_host = p_other.getHost();
343 m_port = p_other.getPort();
344 m_path = p_other.getPath();
345 m_queryString = p_other.getQueryString();
346 m_fragment = p_other.getFragment();
347 }
348
349 /**
350 * Initializes this URI from a base URI and a URI specification string.
351 * See RFC 2396 Section 4 and Appendix B for specifications on parsing
352 * the URI and Section 5 for specifications on resolving relative URIs
353 * and relative paths.
354 *
355 * @param p_base the base URI (may be null if p_uriSpec is an absolute
356 * URI)
357 * @param p_uriSpec the URI spec string which may be an absolute or
358 * relative URI (can only be null/empty if p_base
359 * is not null)
360 *
361 * @throws MalformedURIException if p_base is null and p_uriSpec
362 * is not an absolute URI or if
363 * p_uriSpec violates syntax rules
364 */
365 private void initialize(URI p_base, String p_uriSpec)
366 throws MalformedURIException
367 {
368
369 if (p_base == null
370 && (p_uriSpec == null || p_uriSpec.trim().length() == 0))
371 {
372 throw new MalformedURIException(
373 XMLMessages.createXMLMessage(XMLErrorResources.ER_CANNOT_INIT_URI_EMPTY_PARMS, null)); //"Cannot initialize URI with empty parameters.");
374 }
375
376 // just make a copy of the base if spec is empty
377 if (p_uriSpec == null || p_uriSpec.trim().length() == 0)
378 {
379 initialize(p_base);
380
381 return;
382 }
383
384 String uriSpec = p_uriSpec.trim();
385 int uriSpecLen = uriSpec.length();
386 int index = 0;
387
388 // check for scheme
389 int colonIndex = uriSpec.indexOf(':');
390 if (colonIndex < 0)
391 {
392 if (p_base == null)
393 {
394 throw new MalformedURIException(XMLMessages.createXMLMessage(XMLErrorResources.ER_NO_SCHEME_IN_URI, new Object[]{uriSpec})); //"No scheme found in URI: "+uriSpec);
395 }
396 }
397 else
398 {
399 initializeScheme(uriSpec);
400 uriSpec = uriSpec.substring(colonIndex+1);
401 // This is a fix for XALANJ-2059.
402 if(m_scheme != null && p_base != null)
403 {
404 // a) If <uriSpec> starts with a slash (/), it means <uriSpec> is absolute
405 // and p_base can be ignored.
406 // For example,
407 // uriSpec = file:/myDIR/myXSLFile.xsl
408 // p_base = file:/myWork/
409 //
410 // Here, uriSpec has absolute path after scheme file and :
411 // Hence p_base can be ignored.
412 //
413 // b) Similarily, according to RFC 2396, uri is resolved for <uriSpec> relative to <p_base>
414 // if scheme in <uriSpec> is same as scheme in <p_base>, else p_base can be ignored.
415 //
416 // c) if <p_base> is not hierarchical, it can be ignored.
417 //
418 if(uriSpec.startsWith("/") || !m_scheme.equals(p_base.m_scheme) || !p_base.getSchemeSpecificPart().startsWith("/"))
419 {
420 p_base = null;
421 }
422 }
423 // Fix for XALANJ-2059
424 uriSpecLen = uriSpec.length();
425 }
426
427 // two slashes means generic URI syntax, so we get the authority
428 if (uriSpec.startsWith("//"))
429 {
430 index += 2;
431
432 int startPos = index;
433
434 // get authority - everything up to path, query or fragment
435 char testChar = '\0';
436
437 while (index < uriSpecLen)
438 {
439 testChar = uriSpec.charAt(index);
440
441 if (testChar == '/' || testChar == '?' || testChar == '#')
442 {
443 break;
444 }
445
446 index++;
447 }
448
449 // if we found authority, parse it out, otherwise we set the
450 // host to empty string
451 if (index > startPos)
452 {
453 initializeAuthority(uriSpec.substring(startPos, index));
454 }
455 else
456 {
457 m_host = "";
458 }
459 }
460
461 initializePath(uriSpec.substring(index));
462
463 // Resolve relative URI to base URI - see RFC 2396 Section 5.2
464 // In some cases, it might make more sense to throw an exception
465 // (when scheme is specified is the string spec and the base URI
466 // is also specified, for example), but we're just following the
467 // RFC specifications
468 if (p_base != null)
469 {
470
471 // check to see if this is the current doc - RFC 2396 5.2 #2
472 // note that this is slightly different from the RFC spec in that
473 // we don't include the check for query string being null
474 // - this handles cases where the urispec is just a query
475 // string or a fragment (e.g. "?y" or "#s") -
476 // see <http://www.ics.uci.edu/~fielding/url/test1.html> which
477 // identified this as a bug in the RFC
478 if (m_path.length() == 0 && m_scheme == null && m_host == null)
479 {
480 m_scheme = p_base.getScheme();
481 m_userinfo = p_base.getUserinfo();
482 m_host = p_base.getHost();
483 m_port = p_base.getPort();
484 m_path = p_base.getPath();
485
486 if (m_queryString == null)
487 {
488 m_queryString = p_base.getQueryString();
489 }
490
491 return;
492 }
493
494 // check for scheme - RFC 2396 5.2 #3
495 // if we found a scheme, it means absolute URI, so we're done
496 if (m_scheme == null)
497 {
498 m_scheme = p_base.getScheme();
499 }
500
501 // check for authority - RFC 2396 5.2 #4
502 // if we found a host, then we've got a network path, so we're done
503 if (m_host == null)
504 {
505 m_userinfo = p_base.getUserinfo();
506 m_host = p_base.getHost();
507 m_port = p_base.getPort();
508 }
509 else
510 {
511 return;
512 }
513
514 // check for absolute path - RFC 2396 5.2 #5
515 if (m_path.length() > 0 && m_path.startsWith("/"))
516 {
517 return;
518 }
519
520 // if we get to this point, we need to resolve relative path
521 // RFC 2396 5.2 #6
522 String path = new String();
523 String basePath = p_base.getPath();
524
525 // 6a - get all but the last segment of the base URI path
526 if (basePath != null)
527 {
528 int lastSlash = basePath.lastIndexOf('/');
529
530 if (lastSlash != -1)
531 {
532 path = basePath.substring(0, lastSlash + 1);
533 }
534 }
535
536 // 6b - append the relative URI path
537 path = path.concat(m_path);
538
539 // 6c - remove all "./" where "." is a complete path segment
540 index = -1;
541
542 while ((index = path.indexOf("/./")) != -1)
543 {
544 path = path.substring(0, index + 1).concat(path.substring(index + 3));
545 }
546
547 // 6d - remove "." if path ends with "." as a complete path segment
548 if (path.endsWith("/."))
549 {
550 path = path.substring(0, path.length() - 1);
551 }
552
553 // 6e - remove all "<segment>/../" where "<segment>" is a complete
554 // path segment not equal to ".."
555 index = -1;
556
557 int segIndex = -1;
558 String tempString = null;
559
560 while ((index = path.indexOf("/../")) > 0)
561 {
562 tempString = path.substring(0, path.indexOf("/../"));
563 segIndex = tempString.lastIndexOf('/');
564
565 if (segIndex != -1)
566 {
567 if (!tempString.substring(segIndex++).equals(".."))
568 {
569 path = path.substring(0, segIndex).concat(path.substring(index
570 + 4));
571 }
572 }
573 }
574
575 // 6f - remove ending "<segment>/.." where "<segment>" is a
576 // complete path segment
577 if (path.endsWith("/.."))
578 {
579 tempString = path.substring(0, path.length() - 3);
580 segIndex = tempString.lastIndexOf('/');
581
582 if (segIndex != -1)
583 {
584 path = path.substring(0, segIndex + 1);
585 }
586 }
587
588 m_path = path;
589 }
590 }
591
592 /**
593 * Initialize the scheme for this URI from a URI string spec.
594 *
595 * @param p_uriSpec the URI specification (cannot be null)
596 *
597 * @throws MalformedURIException if URI does not have a conformant
598 * scheme
599 */
600 private void initializeScheme(String p_uriSpec) throws MalformedURIException
601 {
602
603 int uriSpecLen = p_uriSpec.length();
604 int index = 0;
605 String scheme = null;
606 char testChar = '\0';
607
608 while (index < uriSpecLen)
609 {
610 testChar = p_uriSpec.charAt(index);
611
612 if (testChar == ':' || testChar == '/' || testChar == '?'
613 || testChar == '#')
614 {
615 break;
616 }
617
618 index++;
619 }
620
621 scheme = p_uriSpec.substring(0, index);
622
623 if (scheme.length() == 0)
624 {
625 throw new MalformedURIException(XMLMessages.createXMLMessage(XMLErrorResources.ER_NO_SCHEME_INURI, null)); //"No scheme found in URI.");
626 }
627 else
628 {
629 setScheme(scheme);
630 }
631 }
632
633 /**
634 * Initialize the authority (userinfo, host and port) for this
635 * URI from a URI string spec.
636 *
637 * @param p_uriSpec the URI specification (cannot be null)
638 *
639 * @throws MalformedURIException if p_uriSpec violates syntax rules
640 */
641 private void initializeAuthority(String p_uriSpec)
642 throws MalformedURIException
643 {
644
645 int index = 0;
646 int start = 0;
647 int end = p_uriSpec.length();
648 char testChar = '\0';
649 String userinfo = null;
650
651 // userinfo is everything up @
652 if (p_uriSpec.indexOf('@', start) != -1)
653 {
654 while (index < end)
655 {
656 testChar = p_uriSpec.charAt(index);
657
658 if (testChar == '@')
659 {
660 break;
661 }
662
663 index++;
664 }
665
666 userinfo = p_uriSpec.substring(start, index);
667
668 index++;
669 }
670
671 // host is everything up to ':'
672 String host = null;
673
674 start = index;
675
676 while (index < end)
677 {
678 testChar = p_uriSpec.charAt(index);
679
680 if (testChar == ':')
681 {
682 break;
683 }
684
685 index++;
686 }
687
688 host = p_uriSpec.substring(start, index);
689
690 int port = -1;
691
692 if (host.length() > 0)
693 {
694
695 // port
696 if (testChar == ':')
697 {
698 index++;
699
700 start = index;
701
702 while (index < end)
703 {
704 index++;
705 }
706
707 String portStr = p_uriSpec.substring(start, index);
708
709 if (portStr.length() > 0)
710 {
711 for (int i = 0; i < portStr.length(); i++)
712 {
713 if (!isDigit(portStr.charAt(i)))
714 {
715 throw new MalformedURIException(
716 portStr + " is invalid. Port should only contain digits!");
717 }
718 }
719
720 try
721 {
722 port = Integer.parseInt(portStr);
723 }
724 catch (NumberFormatException nfe)
725 {
726
727 // can't happen
728 }
729 }
730 }
731 }
732
733 setHost(host);
734 setPort(port);
735 setUserinfo(userinfo);
736 }
737
738 /**
739 * Initialize the path for this URI from a URI string spec.
740 *
741 * @param p_uriSpec the URI specification (cannot be null)
742 *
743 * @throws MalformedURIException if p_uriSpec violates syntax rules
744 */
745 private void initializePath(String p_uriSpec) throws MalformedURIException
746 {
747
748 if (p_uriSpec == null)
749 {
750 throw new MalformedURIException(
751 "Cannot initialize path from null string!");
752 }
753
754 int index = 0;
755 int start = 0;
756 int end = p_uriSpec.length();
757 char testChar = '\0';
758
759 // path - everything up to query string or fragment
760 while (index < end)
761 {
762 testChar = p_uriSpec.charAt(index);
763
764 if (testChar == '?' || testChar == '#')
765 {
766 break;
767 }
768
769 // check for valid escape sequence
770 if (testChar == '%')
771 {
772 if (index + 2 >= end ||!isHex(p_uriSpec.charAt(index + 1))
773 ||!isHex(p_uriSpec.charAt(index + 2)))
774 {
775 throw new MalformedURIException(
776 XMLMessages.createXMLMessage(XMLErrorResources.ER_PATH_CONTAINS_INVALID_ESCAPE_SEQUENCE, null)); //"Path contains invalid escape sequence!");
777 }
778 }
779 else if (!isReservedCharacter(testChar)
780 &&!isUnreservedCharacter(testChar))
781 {
782 if ('\\' != testChar)
783 throw new MalformedURIException(XMLMessages.createXMLMessage(XMLErrorResources.ER_PATH_INVALID_CHAR, new Object[]{String.valueOf(testChar)})); //"Path contains invalid character: "
784 //+ testChar);
785 }
786
787 index++;
788 }
789
790 m_path = p_uriSpec.substring(start, index);
791
792 // query - starts with ? and up to fragment or end
793 if (testChar == '?')
794 {
795 index++;
796
797 start = index;
798
799 while (index < end)
800 {
801 testChar = p_uriSpec.charAt(index);
802
803 if (testChar == '#')
804 {
805 break;
806 }
807
808 if (testChar == '%')
809 {
810 if (index + 2 >= end ||!isHex(p_uriSpec.charAt(index + 1))
811 ||!isHex(p_uriSpec.charAt(index + 2)))
812 {
813 throw new MalformedURIException(
814 "Query string contains invalid escape sequence!");
815 }
816 }
817 else if (!isReservedCharacter(testChar)
818 &&!isUnreservedCharacter(testChar))
819 {
820 throw new MalformedURIException(
821 "Query string contains invalid character:" + testChar);
822 }
823
824 index++;
825 }
826
827 m_queryString = p_uriSpec.substring(start, index);
828 }
829
830 // fragment - starts with #
831 if (testChar == '#')
832 {
833 index++;
834
835 start = index;
836
837 while (index < end)
838 {
839 testChar = p_uriSpec.charAt(index);
840
841 if (testChar == '%')
842 {
843 if (index + 2 >= end ||!isHex(p_uriSpec.charAt(index + 1))
844 ||!isHex(p_uriSpec.charAt(index + 2)))
845 {
846 throw new MalformedURIException(
847 "Fragment contains invalid escape sequence!");
848 }
849 }
850 else if (!isReservedCharacter(testChar)
851 &&!isUnreservedCharacter(testChar))
852 {
853 throw new MalformedURIException(
854 "Fragment contains invalid character:" + testChar);
855 }
856
857 index++;
858 }
859
860 m_fragment = p_uriSpec.substring(start, index);
861 }
862 }
863
864 /**
865 * Get the scheme for this URI.
866 *
867 * @return the scheme for this URI
868 */
869 public String getScheme()
870 {
871 return m_scheme;
872 }
873
874 /**
875 * Get the scheme-specific part for this URI (everything following the
876 * scheme and the first colon). See RFC 2396 Section 5.2 for spec.
877 *
878 * @return the scheme-specific part for this URI
879 */
880 public String getSchemeSpecificPart()
881 {
882
883 StringBuffer schemespec = new StringBuffer();
884
885 if (m_userinfo != null || m_host != null || m_port != -1)
886 {
887 schemespec.append("//");
888 }
889
890 if (m_userinfo != null)
891 {
892 schemespec.append(m_userinfo);
893 schemespec.append('@');
894 }
895
896 if (m_host != null)
897 {
898 schemespec.append(m_host);
899 }
900
901 if (m_port != -1)
902 {
903 schemespec.append(':');
904 schemespec.append(m_port);
905 }
906
907 if (m_path != null)
908 {
909 schemespec.append((m_path));
910 }
911
912 if (m_queryString != null)
913 {
914 schemespec.append('?');
915 schemespec.append(m_queryString);
916 }
917
918 if (m_fragment != null)
919 {
920 schemespec.append('#');
921 schemespec.append(m_fragment);
922 }
923
924 return schemespec.toString();
925 }
926
927 /**
928 * Get the userinfo for this URI.
929 *
930 * @return the userinfo for this URI (null if not specified).
931 */
932 public String getUserinfo()
933 {
934 return m_userinfo;
935 }
936
937 /**
938 * Get the host for this URI.
939 *
940 * @return the host for this URI (null if not specified).
941 */
942 public String getHost()
943 {
944 return m_host;
945 }
946
947 /**
948 * Get the port for this URI.
949 *
950 * @return the port for this URI (-1 if not specified).
951 */
952 public int getPort()
953 {
954 return m_port;
955 }
956
957 /**
958 * Get the path for this URI (optionally with the query string and
959 * fragment).
960 *
961 * @param p_includeQueryString if true (and query string is not null),
962 * then a "?" followed by the query string
963 * will be appended
964 * @param p_includeFragment if true (and fragment is not null),
965 * then a "#" followed by the fragment
966 * will be appended
967 *
968 * @return the path for this URI possibly including the query string
969 * and fragment
970 */
971 public String getPath(boolean p_includeQueryString,
972 boolean p_includeFragment)
973 {
974
975 StringBuffer pathString = new StringBuffer(m_path);
976
977 if (p_includeQueryString && m_queryString != null)
978 {
979 pathString.append('?');
980 pathString.append(m_queryString);
981 }
982
983 if (p_includeFragment && m_fragment != null)
984 {
985 pathString.append('#');
986 pathString.append(m_fragment);
987 }
988
989 return pathString.toString();
990 }
991
992 /**
993 * Get the path for this URI. Note that the value returned is the path
994 * only and does not include the query string or fragment.
995 *
996 * @return the path for this URI.
997 */
998 public String getPath()
999 {
1000 return m_path;
1001 }
1002
1003 /**
1004 * Get the query string for this URI.
1005 *
1006 * @return the query string for this URI. Null is returned if there
1007 * was no "?" in the URI spec, empty string if there was a
1008 * "?" but no query string following it.
1009 */
1010 public String getQueryString()
1011 {
1012 return m_queryString;
1013 }
1014
1015 /**
1016 * Get the fragment for this URI.
1017 *
1018 * @return the fragment for this URI. Null is returned if there
1019 * was no "#" in the URI spec, empty string if there was a
1020 * "#" but no fragment following it.
1021 */
1022 public String getFragment()
1023 {
1024 return m_fragment;
1025 }
1026
1027 /**
1028 * Set the scheme for this URI. The scheme is converted to lowercase
1029 * before it is set.
1030 *
1031 * @param p_scheme the scheme for this URI (cannot be null)
1032 *
1033 * @throws MalformedURIException if p_scheme is not a conformant
1034 * scheme name
1035 */
1036 public void setScheme(String p_scheme) throws MalformedURIException
1037 {
1038
1039 if (p_scheme == null)
1040 {
1041 throw new MalformedURIException(XMLMessages.createXMLMessage(XMLErrorResources.ER_SCHEME_FROM_NULL_STRING, null)); //"Cannot set scheme from null string!");
1042 }
1043
1044 if (!isConformantSchemeName(p_scheme))
1045 {
1046 throw new MalformedURIException(XMLMessages.createXMLMessage(XMLErrorResources.ER_SCHEME_NOT_CONFORMANT, null)); //"The scheme is not conformant.");
1047 }
1048
1049 m_scheme = p_scheme.toLowerCase();
1050 }
1051
1052 /**
1053 * Set the userinfo for this URI. If a non-null value is passed in and
1054 * the host value is null, then an exception is thrown.
1055 *
1056 * @param p_userinfo the userinfo for this URI
1057 *
1058 * @throws MalformedURIException if p_userinfo contains invalid
1059 * characters
1060 */
1061 public void setUserinfo(String p_userinfo) throws MalformedURIException
1062 {
1063
1064 if (p_userinfo == null)
1065 {
1066 m_userinfo = null;
1067 }
1068 else
1069 {
1070 if (m_host == null)
1071 {
1072 throw new MalformedURIException(
1073 "Userinfo cannot be set when host is null!");
1074 }
1075
1076 // userinfo can contain alphanumerics, mark characters, escaped
1077 // and ';',':','&','=','+','$',','
1078 int index = 0;
1079 int end = p_userinfo.length();
1080 char testChar = '\0';
1081
1082 while (index < end)
1083 {
1084 testChar = p_userinfo.charAt(index);
1085
1086 if (testChar == '%')
1087 {
1088 if (index + 2 >= end ||!isHex(p_userinfo.charAt(index + 1))
1089 ||!isHex(p_userinfo.charAt(index + 2)))
1090 {
1091 throw new MalformedURIException(
1092 "Userinfo contains invalid escape sequence!");
1093 }
1094 }
1095 else if (!isUnreservedCharacter(testChar)
1096 && USERINFO_CHARACTERS.indexOf(testChar) == -1)
1097 {
1098 throw new MalformedURIException(
1099 "Userinfo contains invalid character:" + testChar);
1100 }
1101
1102 index++;
1103 }
1104 }
1105
1106 m_userinfo = p_userinfo;
1107 }
1108
1109 /**
1110 * Set the host for this URI. If null is passed in, the userinfo
1111 * field is also set to null and the port is set to -1.
1112 *
1113 * @param p_host the host for this URI
1114 *
1115 * @throws MalformedURIException if p_host is not a valid IP
1116 * address or DNS hostname.
1117 */
1118 public void setHost(String p_host) throws MalformedURIException
1119 {
1120
1121 if (p_host == null || p_host.trim().length() == 0)
1122 {
1123 m_host = p_host;
1124 m_userinfo = null;
1125 m_port = -1;
1126 }
1127 else if (!isWellFormedAddress(p_host))
1128 {
1129 throw new MalformedURIException(XMLMessages.createXMLMessage(XMLErrorResources.ER_HOST_ADDRESS_NOT_WELLFORMED, null)); //"Host is not a well formed address!");
1130 }
1131
1132 m_host = p_host;
1133 }
1134
1135 /**
1136 * Set the port for this URI. -1 is used to indicate that the port is
1137 * not specified, otherwise valid port numbers are between 0 and 65535.
1138 * If a valid port number is passed in and the host field is null,
1139 * an exception is thrown.
1140 *
1141 * @param p_port the port number for this URI
1142 *
1143 * @throws MalformedURIException if p_port is not -1 and not a
1144 * valid port number
1145 */
1146 public void setPort(int p_port) throws MalformedURIException
1147 {
1148
1149 if (p_port >= 0 && p_port <= 65535)
1150 {
1151 if (m_host == null)
1152 {
1153 throw new MalformedURIException(
1154 XMLMessages.createXMLMessage(XMLErrorResources.ER_PORT_WHEN_HOST_NULL, null)); //"Port cannot be set when host is null!");
1155 }
1156 }
1157 else if (p_port != -1)
1158 {
1159 throw new MalformedURIException(XMLMessages.createXMLMessage(XMLErrorResources.ER_INVALID_PORT, null)); //"Invalid port number!");
1160 }
1161
1162 m_port = p_port;
1163 }
1164
1165 /**
1166 * Set the path for this URI. If the supplied path is null, then the
1167 * query string and fragment are set to null as well. If the supplied
1168 * path includes a query string and/or fragment, these fields will be
1169 * parsed and set as well. Note that, for URIs following the "generic
1170 * URI" syntax, the path specified should start with a slash.
1171 * For URIs that do not follow the generic URI syntax, this method
1172 * sets the scheme-specific part.
1173 *
1174 * @param p_path the path for this URI (may be null)
1175 *
1176 * @throws MalformedURIException if p_path contains invalid
1177 * characters
1178 */
1179 public void setPath(String p_path) throws MalformedURIException
1180 {
1181
1182 if (p_path == null)
1183 {
1184 m_path = null;
1185 m_queryString = null;
1186 m_fragment = null;
1187 }
1188 else
1189 {
1190 initializePath(p_path);
1191 }
1192 }
1193
1194 /**
1195 * Append to the end of the path of this URI. If the current path does
1196 * not end in a slash and the path to be appended does not begin with
1197 * a slash, a slash will be appended to the current path before the
1198 * new segment is added. Also, if the current path ends in a slash
1199 * and the new segment begins with a slash, the extra slash will be
1200 * removed before the new segment is appended.
1201 *
1202 * @param p_addToPath the new segment to be added to the current path
1203 *
1204 * @throws MalformedURIException if p_addToPath contains syntax
1205 * errors
1206 */
1207 public void appendPath(String p_addToPath) throws MalformedURIException
1208 {
1209
1210 if (p_addToPath == null || p_addToPath.trim().length() == 0)
1211 {
1212 return;
1213 }
1214
1215 if (!isURIString(p_addToPath))
1216 {
1217 throw new MalformedURIException(XMLMessages.createXMLMessage(XMLErrorResources.ER_PATH_INVALID_CHAR, new Object[]{p_addToPath})); //"Path contains invalid character!");
1218 }
1219
1220 if (m_path == null || m_path.trim().length() == 0)
1221 {
1222 if (p_addToPath.startsWith("/"))
1223 {
1224 m_path = p_addToPath;
1225 }
1226 else
1227 {
1228 m_path = "/" + p_addToPath;
1229 }
1230 }
1231 else if (m_path.endsWith("/"))
1232 {
1233 if (p_addToPath.startsWith("/"))
1234 {
1235 m_path = m_path.concat(p_addToPath.substring(1));
1236 }
1237 else
1238 {
1239 m_path = m_path.concat(p_addToPath);
1240 }
1241 }
1242 else
1243 {
1244 if (p_addToPath.startsWith("/"))
1245 {
1246 m_path = m_path.concat(p_addToPath);
1247 }
1248 else
1249 {
1250 m_path = m_path.concat("/" + p_addToPath);
1251 }
1252 }
1253 }
1254
1255 /**
1256 * Set the query string for this URI. A non-null value is valid only
1257 * if this is an URI conforming to the generic URI syntax and
1258 * the path value is not null.
1259 *
1260 * @param p_queryString the query string for this URI
1261 *
1262 * @throws MalformedURIException if p_queryString is not null and this
1263 * URI does not conform to the generic
1264 * URI syntax or if the path is null
1265 */
1266 public void setQueryString(String p_queryString)
1267 throws MalformedURIException
1268 {
1269
1270 if (p_queryString == null)
1271 {
1272 m_queryString = null;
1273 }
1274 else if (!isGenericURI())
1275 {
1276 throw new MalformedURIException(
1277 "Query string can only be set for a generic URI!");
1278 }
1279 else if (getPath() == null)
1280 {
1281 throw new MalformedURIException(
1282 "Query string cannot be set when path is null!");
1283 }
1284 else if (!isURIString(p_queryString))
1285 {
1286 throw new MalformedURIException(
1287 "Query string contains invalid character!");
1288 }
1289 else
1290 {
1291 m_queryString = p_queryString;
1292 }
1293 }
1294
1295 /**
1296 * Set the fragment for this URI. A non-null value is valid only
1297 * if this is a URI conforming to the generic URI syntax and
1298 * the path value is not null.
1299 *
1300 * @param p_fragment the fragment for this URI
1301 *
1302 * @throws MalformedURIException if p_fragment is not null and this
1303 * URI does not conform to the generic
1304 * URI syntax or if the path is null
1305 */
1306 public void setFragment(String p_fragment) throws MalformedURIException
1307 {
1308
1309 if (p_fragment == null)
1310 {
1311 m_fragment = null;
1312 }
1313 else if (!isGenericURI())
1314 {
1315 throw new MalformedURIException(
1316 XMLMessages.createXMLMessage(XMLErrorResources.ER_FRAG_FOR_GENERIC_URI, null)); //"Fragment can only be set for a generic URI!");
1317 }
1318 else if (getPath() == null)
1319 {
1320 throw new MalformedURIException(
1321 XMLMessages.createXMLMessage(XMLErrorResources.ER_FRAG_WHEN_PATH_NULL, null)); //"Fragment cannot be set when path is null!");
1322 }
1323 else if (!isURIString(p_fragment))
1324 {
1325 throw new MalformedURIException(XMLMessages.createXMLMessage(XMLErrorResources.ER_FRAG_INVALID_CHAR, null)); //"Fragment contains invalid character!");
1326 }
1327 else
1328 {
1329 m_fragment = p_fragment;
1330 }
1331 }
1332
1333 /**
1334 * Determines if the passed-in Object is equivalent to this URI.
1335 *
1336 * @param p_test the Object to test for equality.
1337 *
1338 * @return true if p_test is a URI with all values equal to this
1339 * URI, false otherwise
1340 */
1341 public boolean equals(Object p_test)
1342 {
1343
1344 if (p_test instanceof URI)
1345 {
1346 URI testURI = (URI) p_test;
1347
1348 if (((m_scheme == null && testURI.m_scheme == null) || (m_scheme != null && testURI.m_scheme != null && m_scheme.equals(
1349 testURI.m_scheme))) && ((m_userinfo == null && testURI.m_userinfo == null) || (m_userinfo != null && testURI.m_userinfo != null && m_userinfo.equals(
1350 testURI.m_userinfo))) && ((m_host == null && testURI.m_host == null) || (m_host != null && testURI.m_host != null && m_host.equals(
1351 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(
1352 testURI.m_path))) && ((m_queryString == null && testURI.m_queryString == null) || (m_queryString != null && testURI.m_queryString != null && m_queryString.equals(
1353 testURI.m_queryString))) && ((m_fragment == null && testURI.m_fragment == null) || (m_fragment != null && testURI.m_fragment != null && m_fragment.equals(
1354 testURI.m_fragment))))
1355 {
1356 return true;
1357 }
1358 }
1359
1360 return false;
1361 }
1362
1363 /**
1364 * Get the URI as a string specification. See RFC 2396 Section 5.2.
1365 *
1366 * @return the URI string specification
1367 */
1368 public String toString()
1369 {
1370
1371 StringBuffer uriSpecString = new StringBuffer();
1372
1373 if (m_scheme != null)
1374 {
1375 uriSpecString.append(m_scheme);
1376 uriSpecString.append(':');
1377 }
1378
1379 uriSpecString.append(getSchemeSpecificPart());
1380
1381 return uriSpecString.toString();
1382 }
1383
1384 /**
1385 * Get the indicator as to whether this URI uses the "generic URI"
1386 * syntax.
1387 *
1388 * @return true if this URI uses the "generic URI" syntax, false
1389 * otherwise
1390 */
1391 public boolean isGenericURI()
1392 {
1393
1394 // presence of the host (whether valid or empty) means
1395 // double-slashes which means generic uri
1396 return (m_host != null);
1397 }
1398
1399 /**
1400 * Determine whether a scheme conforms to the rules for a scheme name.
1401 * A scheme is conformant if it starts with an alphanumeric, and
1402 * contains only alphanumerics, '+','-' and '.'.
1403 *
1404 *
1405 * @param p_scheme The sheme name to check
1406 * @return true if the scheme is conformant, false otherwise
1407 */
1408 public static boolean isConformantSchemeName(String p_scheme)
1409 {
1410
1411 if (p_scheme == null || p_scheme.trim().length() == 0)
1412 {
1413 return false;
1414 }
1415
1416 if (!isAlpha(p_scheme.charAt(0)))
1417 {
1418 return false;
1419 }
1420
1421 char testChar;
1422
1423 for (int i = 1; i < p_scheme.length(); i++)
1424 {
1425 testChar = p_scheme.charAt(i);
1426
1427 if (!isAlphanum(testChar) && SCHEME_CHARACTERS.indexOf(testChar) == -1)
1428 {
1429 return false;
1430 }
1431 }
1432
1433 return true;
1434 }
1435
1436 /**
1437 * Determine whether a string is syntactically capable of representing
1438 * a valid IPv4 address or the domain name of a network host. A valid
1439 * IPv4 address consists of four decimal digit groups separated by a
1440 * '.'. A hostname consists of domain labels (each of which must
1441 * begin and end with an alphanumeric but may contain '-') separated
1442 * & by a '.'. See RFC 2396 Section 3.2.2.
1443 *
1444 *
1445 * @param p_address The address string to check
1446 * @return true if the string is a syntactically valid IPv4 address
1447 * or hostname
1448 */
1449 public static boolean isWellFormedAddress(String p_address)
1450 {
1451
1452 if (p_address == null)
1453 {
1454 return false;
1455 }
1456
1457 String address = p_address.trim();
1458 int addrLength = address.length();
1459
1460 if (addrLength == 0 || addrLength > 255)
1461 {
1462 return false;
1463 }
1464
1465 if (address.startsWith(".") || address.startsWith("-"))
1466 {
1467 return false;
1468 }
1469
1470 // rightmost domain label starting with digit indicates IP address
1471 // since top level domain label can only start with an alpha
1472 // see RFC 2396 Section 3.2.2
1473 int index = address.lastIndexOf('.');
1474
1475 if (address.endsWith("."))
1476 {
1477 index = address.substring(0, index).lastIndexOf('.');
1478 }
1479
1480 if (index + 1 < addrLength && isDigit(p_address.charAt(index + 1)))
1481 {
1482 char testChar;
1483 int numDots = 0;
1484
1485 // make sure that 1) we see only digits and dot separators, 2) that
1486 // any dot separator is preceded and followed by a digit and
1487 // 3) that we find 3 dots
1488 for (int i = 0; i < addrLength; i++)
1489 {
1490 testChar = address.charAt(i);
1491
1492 if (testChar == '.')
1493 {
1494 if (!isDigit(address.charAt(i - 1))
1495 || (i + 1 < addrLength &&!isDigit(address.charAt(i + 1))))
1496 {
1497 return false;
1498 }
1499
1500 numDots++;
1501 }
1502 else if (!isDigit(testChar))
1503 {
1504 return false;
1505 }
1506 }
1507
1508 if (numDots != 3)
1509 {
1510 return false;
1511 }
1512 }
1513 else
1514 {
1515
1516 // domain labels can contain alphanumerics and '-"
1517 // but must start and end with an alphanumeric
1518 char testChar;
1519
1520 for (int i = 0; i < addrLength; i++)
1521 {
1522 testChar = address.charAt(i);
1523
1524 if (testChar == '.')
1525 {
1526 if (!isAlphanum(address.charAt(i - 1)))
1527 {
1528 return false;
1529 }
1530
1531 if (i + 1 < addrLength &&!isAlphanum(address.charAt(i + 1)))
1532 {
1533 return false;
1534 }
1535 }
1536 else if (!isAlphanum(testChar) && testChar != '-')
1537 {
1538 return false;
1539 }
1540 }
1541 }
1542
1543 return true;
1544 }
1545
1546 /**
1547 * Determine whether a char is a digit.
1548 *
1549 *
1550 * @param p_char the character to check
1551 * @return true if the char is betweeen '0' and '9', false otherwise
1552 */
1553 private static boolean isDigit(char p_char)
1554 {
1555 return p_char >= '0' && p_char <= '9';
1556 }
1557
1558 /**
1559 * Determine whether a character is a hexadecimal character.
1560 *
1561 *
1562 * @param p_char the character to check
1563 * @return true if the char is betweeen '0' and '9', 'a' and 'f'
1564 * or 'A' and 'F', false otherwise
1565 */
1566 private static boolean isHex(char p_char)
1567 {
1568 return (isDigit(p_char) || (p_char >= 'a' && p_char <= 'f')
1569 || (p_char >= 'A' && p_char <= 'F'));
1570 }
1571
1572 /**
1573 * Determine whether a char is an alphabetic character: a-z or A-Z
1574 *
1575 *
1576 * @param p_char the character to check
1577 * @return true if the char is alphabetic, false otherwise
1578 */
1579 private static boolean isAlpha(char p_char)
1580 {
1581 return ((p_char >= 'a' && p_char <= 'z')
1582 || (p_char >= 'A' && p_char <= 'Z'));
1583 }
1584
1585 /**
1586 * Determine whether a char is an alphanumeric: 0-9, a-z or A-Z
1587 *
1588 *
1589 * @param p_char the character to check
1590 * @return true if the char is alphanumeric, false otherwise
1591 */
1592 private static boolean isAlphanum(char p_char)
1593 {
1594 return (isAlpha(p_char) || isDigit(p_char));
1595 }
1596
1597 /**
1598 * Determine whether a character is a reserved character:
1599 * ';', '/', '?', ':', '@', '&', '=', '+', '$' or ','
1600 *
1601 *
1602 * @param p_char the character to check
1603 * @return true if the string contains any reserved characters
1604 */
1605 private static boolean isReservedCharacter(char p_char)
1606 {
1607 return RESERVED_CHARACTERS.indexOf(p_char) != -1;
1608 }
1609
1610 /**
1611 * Determine whether a char is an unreserved character.
1612 *
1613 *
1614 * @param p_char the character to check
1615 * @return true if the char is unreserved, false otherwise
1616 */
1617 private static boolean isUnreservedCharacter(char p_char)
1618 {
1619 return (isAlphanum(p_char) || MARK_CHARACTERS.indexOf(p_char) != -1);
1620 }
1621
1622 /**
1623 * Determine whether a given string contains only URI characters (also
1624 * called "uric" in RFC 2396). uric consist of all reserved
1625 * characters, unreserved characters and escaped characters.
1626 *
1627 *
1628 * @param p_uric URI string
1629 * @return true if the string is comprised of uric, false otherwise
1630 */
1631 private static boolean isURIString(String p_uric)
1632 {
1633
1634 if (p_uric == null)
1635 {
1636 return false;
1637 }
1638
1639 int end = p_uric.length();
1640 char testChar = '\0';
1641
1642 for (int i = 0; i < end; i++)
1643 {
1644 testChar = p_uric.charAt(i);
1645
1646 if (testChar == '%')
1647 {
1648 if (i + 2 >= end ||!isHex(p_uric.charAt(i + 1))
1649 ||!isHex(p_uric.charAt(i + 2)))
1650 {
1651 return false;
1652 }
1653 else
1654 {
1655 i += 2;
1656
1657 continue;
1658 }
1659 }
1660
1661 if (isReservedCharacter(testChar) || isUnreservedCharacter(testChar))
1662 {
1663 continue;
1664 }
1665 else
1666 {
1667 return false;
1668 }
1669 }
1670
1671 return true;
1672 }
1673 }