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 1.1.4.1 2005/09/08 11:03:20 suresh_emailid Exp $ 22 */ 23 package com.sun.org.apache.xml.internal.serializer.utils; 24 25 import java.io.IOException; 26 import java.io.Serializable; 27 28 29 /** 30 * A class to represent a Uniform Resource Identifier (URI). This class 31 * is designed to handle the parsing of URIs and provide access to 32 * the various components (scheme, host, port, userinfo, path, query 33 * string and fragment) that may constitute a URI. 34 * <p> 35 * Parsing of a URI specification is done according to the URI 36 * syntax described in RFC 2396 37 * <http://www.ietf.org/rfc/rfc2396.txt?number=2396>. Every URI consists 38 * of a scheme, followed by a colon (':'), followed by a scheme-specific 39 * part. For URIs that follow the "generic URI" syntax, the scheme- 40 * specific part begins with two slashes ("//") and may be followed 41 * by an authority segment (comprised of user information, host, and 42 * port), path segment, query segment and fragment. Note that RFC 2396 43 * no longer specifies the use of the parameters segment and excludes 44 * the "user:password" syntax as part of the authority segment. If 45 * "user:password" appears in a URI, the entire user/password string 46 * is stored as userinfo. 47 * <p> 48 * For URIs that do not follow the "generic URI" syntax (e.g. mailto), 49 * the entire scheme-specific part is treated as the "path" portion 50 * of the URI. 51 * <p> 52 * Note that, unlike the java.net.URL class, this class does not provide 53 * any built-in network access functionality nor does it provide any 54 * scheme-specific functionality (for example, it does not know a 55 * default port for a specific scheme). Rather, it only knows the 56 * grammar and basic set of operations that can be applied to a URI. 57 * 58 * This class is a copy of the one in com.sun.org.apache.xml.internal.utils. 59 * It exists to cut the serializers dependancy on that package. 60 * 61 * A minor change from the original is that this class no longer implements 62 * Serializable, and the serialVersionUID magic field is dropped, and 63 * the class is no longer "public". 64 * 65 * @xsl.usage internal 66 */ 67 final class URI 68 { 69 /** 70 * MalformedURIExceptions are thrown in the process of building a URI 71 * or setting fields on a URI when an operation would result in an 72 * invalid URI specification. 73 * 74 */ 75 public static class MalformedURIException extends IOException 76 { 77 78 /** 79 * Constructs a <code>MalformedURIException</code> with no specified 80 * detail message. 81 */ 82 public MalformedURIException() 83 { 84 super(); 85 } 86 87 /** 88 * Constructs a <code>MalformedURIException</code> with the 89 * specified detail message. 90 * 91 * @param p_msg the detail message. 92 */ 93 public MalformedURIException(String p_msg) 94 { 95 super(p_msg); 96 } 97 } 98 99 /** reserved characters */ 100 private static final String RESERVED_CHARACTERS = ";/?:@&=+$,"; 101 102 /** 103 * URI punctuation mark characters - these, combined with 104 * alphanumerics, constitute the "unreserved" characters 105 */ 106 private static final String MARK_CHARACTERS = "-_.!~*'() "; 107 108 /** scheme can be composed of alphanumerics and these characters */ 109 private static final String SCHEME_CHARACTERS = "+-."; 110 111 /** 112 * userinfo can be composed of unreserved, escaped and these 113 * characters 114 */ 115 private static final String USERINFO_CHARACTERS = ";:&=+$,"; 116 117 /** Stores the scheme (usually the protocol) for this URI. 118 * @serial */ 119 private String m_scheme = null; 120 121 /** If specified, stores the userinfo for this URI; otherwise null. 122 * @serial */ 123 private String m_userinfo = null; 124 125 /** If specified, stores the host for this URI; otherwise null. 126 * @serial */ 127 private String m_host = null; 128 129 /** If specified, stores the port for this URI; otherwise -1. 130 * @serial */ 131 private int m_port = -1; 132 133 /** If specified, stores the path for this URI; otherwise null. 134 * @serial */ 135 private String m_path = null; 136 137 /** 138 * If specified, stores the query string for this URI; otherwise 139 * null. 140 * @serial 141 */ 142 private String m_queryString = null; 143 144 /** If specified, stores the fragment for this URI; otherwise null. 145 * @serial */ 146 private String m_fragment = null; 147 148 /** Indicate whether in DEBUG mode */ 149 private static boolean DEBUG = false; 150 151 /** 152 * Construct a new and uninitialized URI. 153 */ 154 public URI(){} 155 156 /** 157 * Construct a new URI from another URI. All fields for this URI are 158 * set equal to the fields of the URI passed in. 159 * 160 * @param p_other the URI to copy (cannot be null) 161 */ 162 public URI(URI p_other) 163 { 164 initialize(p_other); 165 } 166 167 /** 168 * Construct a new URI from a URI specification string. If the 169 * specification follows the "generic URI" syntax, (two slashes 170 * following the first colon), the specification will be parsed 171 * accordingly - setting the scheme, userinfo, host,port, path, query 172 * string and fragment fields as necessary. If the specification does 173 * not follow the "generic URI" syntax, the specification is parsed 174 * into a scheme and scheme-specific part (stored as the path) only. 175 * 176 * @param p_uriSpec the URI specification string (cannot be null or 177 * empty) 178 * 179 * @throws MalformedURIException if p_uriSpec violates any syntax 180 * rules 181 */ 182 public URI(String p_uriSpec) throws MalformedURIException 183 { 184 this((URI) null, p_uriSpec); 185 } 186 187 /** 188 * Construct a new URI from a base URI and a URI specification string. 189 * The URI specification string may be a relative URI. 190 * 191 * @param p_base the base URI (cannot be null if p_uriSpec is null or 192 * empty) 193 * @param p_uriSpec the URI specification string (cannot be null or 194 * empty if p_base is null) 195 * 196 * @throws MalformedURIException if p_uriSpec violates any syntax 197 * rules 198 */ 199 public URI(URI p_base, String p_uriSpec) throws MalformedURIException 200 { 201 initialize(p_base, p_uriSpec); 202 } 203 204 /** 205 * Construct a new URI that does not follow the generic URI syntax. 206 * Only the scheme and scheme-specific part (stored as the path) are 207 * initialized. 208 * 209 * @param p_scheme the URI scheme (cannot be null or empty) 210 * @param p_schemeSpecificPart the scheme-specific part (cannot be 211 * null or empty) 212 * 213 * @throws MalformedURIException if p_scheme violates any 214 * syntax rules 215 */ 216 public URI(String p_scheme, String p_schemeSpecificPart) 217 throws MalformedURIException 218 { 219 220 if (p_scheme == null || p_scheme.trim().length() == 0) 221 { 222 throw new MalformedURIException( 223 "Cannot construct URI with null/empty scheme!"); 224 } 225 226 if (p_schemeSpecificPart == null 227 || p_schemeSpecificPart.trim().length() == 0) 228 { 229 throw new MalformedURIException( 230 "Cannot construct URI with null/empty scheme-specific part!"); 231 } 232 233 setScheme(p_scheme); 234 setPath(p_schemeSpecificPart); 235 } 236 237 /** 238 * Construct a new URI that follows the generic URI syntax from its 239 * component parts. Each component is validated for syntax and some 240 * basic semantic checks are performed as well. See the individual 241 * setter methods for specifics. 242 * 243 * @param p_scheme the URI scheme (cannot be null or empty) 244 * @param p_host the hostname or IPv4 address for the URI 245 * @param p_path the URI path - if the path contains '?' or '#', 246 * then the query string and/or fragment will be 247 * set from the path; however, if the query and 248 * fragment are specified both in the path and as 249 * separate parameters, an exception is thrown 250 * @param p_queryString the URI query string (cannot be specified 251 * if path is null) 252 * @param p_fragment the URI fragment (cannot be specified if path 253 * is null) 254 * 255 * @throws MalformedURIException if any of the parameters violates 256 * syntax rules or semantic rules 257 */ 258 public URI(String p_scheme, String p_host, String p_path, String p_queryString, String p_fragment) 259 throws MalformedURIException 260 { 261 this(p_scheme, null, p_host, -1, p_path, p_queryString, p_fragment); 262 } 263 264 /** 265 * Construct a new URI that follows the generic URI syntax from its 266 * component parts. Each component is validated for syntax and some 267 * basic semantic checks are performed as well. See the individual 268 * setter methods for specifics. 269 * 270 * @param p_scheme the URI scheme (cannot be null or empty) 271 * @param p_userinfo the URI userinfo (cannot be specified if host 272 * is null) 273 * @param p_host the hostname or IPv4 address for the URI 274 * @param p_port the URI port (may be -1 for "unspecified"; cannot 275 * be specified if host is null) 276 * @param p_path the URI path - if the path contains '?' or '#', 277 * then the query string and/or fragment will be 278 * set from the path; however, if the query and 279 * fragment are specified both in the path and as 280 * separate parameters, an exception is thrown 281 * @param p_queryString the URI query string (cannot be specified 282 * if path is null) 283 * @param p_fragment the URI fragment (cannot be specified if path 284 * is null) 285 * 286 * @throws MalformedURIException if any of the parameters violates 287 * syntax rules or semantic rules 288 */ 289 public URI(String p_scheme, String p_userinfo, String p_host, int p_port, String p_path, String p_queryString, String p_fragment) 290 throws MalformedURIException 291 { 292 293 if (p_scheme == null || p_scheme.trim().length() == 0) 294 { 295 throw new MalformedURIException(Utils.messages.createMessage(MsgKey.ER_SCHEME_REQUIRED, null)); //"Scheme is required!"); 296 } 297 298 if (p_host == null) 299 { 300 if (p_userinfo != null) 301 { 302 throw new MalformedURIException( 303 Utils.messages.createMessage(MsgKey.ER_NO_USERINFO_IF_NO_HOST, null)); //"Userinfo may not be specified if host is not specified!"); 304 } 305 306 if (p_port != -1) 307 { 308 throw new MalformedURIException( 309 Utils.messages.createMessage(MsgKey.ER_NO_PORT_IF_NO_HOST, null)); //"Port may not be specified if host is not specified!"); 310 } 311 } 312 313 if (p_path != null) 314 { 315 if (p_path.indexOf('?') != -1 && p_queryString != null) 316 { 317 throw new MalformedURIException( 318 Utils.messages.createMessage(MsgKey.ER_NO_QUERY_STRING_IN_PATH, null)); //"Query string cannot be specified in path and query string!"); 319 } 320 321 if (p_path.indexOf('#') != -1 && p_fragment != null) 322 { 323 throw new MalformedURIException( 324 Utils.messages.createMessage(MsgKey.ER_NO_FRAGMENT_STRING_IN_PATH, null)); //"Fragment cannot be specified in both the path and fragment!"); 325 } 326 } 327 328 setScheme(p_scheme); 329 setHost(p_host); 330 setPort(p_port); 331 setUserinfo(p_userinfo); 332 setPath(p_path); 333 setQueryString(p_queryString); 334 setFragment(p_fragment); 335 } 336 337 /** 338 * Initialize all fields of this URI from another URI. 339 * 340 * @param p_other the URI to copy (cannot be null) 341 */ 342 private void initialize(URI p_other) 343 { 344 345 m_scheme = p_other.getScheme(); 346 m_userinfo = p_other.getUserinfo(); 347 m_host = p_other.getHost(); 348 m_port = p_other.getPort(); 349 m_path = p_other.getPath(); 350 m_queryString = p_other.getQueryString(); 351 m_fragment = p_other.getFragment(); 352 } 353 354 /** 355 * Initializes this URI from a base URI and a URI specification string. 356 * See RFC 2396 Section 4 and Appendix B for specifications on parsing 357 * the URI and Section 5 for specifications on resolving relative URIs 358 * and relative paths. 359 * 360 * @param p_base the base URI (may be null if p_uriSpec is an absolute 361 * URI) 362 * @param p_uriSpec the URI spec string which may be an absolute or 363 * relative URI (can only be null/empty if p_base 364 * is not null) 365 * 366 * @throws MalformedURIException if p_base is null and p_uriSpec 367 * is not an absolute URI or if 368 * p_uriSpec violates syntax rules 369 */ 370 private void initialize(URI p_base, String p_uriSpec) 371 throws MalformedURIException 372 { 373 374 if (p_base == null 375 && (p_uriSpec == null || p_uriSpec.trim().length() == 0)) 376 { 377 throw new MalformedURIException( 378 Utils.messages.createMessage(MsgKey.ER_CANNOT_INIT_URI_EMPTY_PARMS, null)); //"Cannot initialize URI with empty parameters."); 379 } 380 381 // just make a copy of the base if spec is empty 382 if (p_uriSpec == null || p_uriSpec.trim().length() == 0) 383 { 384 initialize(p_base); 385 386 return; 387 } 388 389 String uriSpec = p_uriSpec.trim(); 390 int uriSpecLen = uriSpec.length(); 391 int index = 0; 392 393 // check for scheme 394 int colonIndex = uriSpec.indexOf(':'); 395 if (colonIndex < 0) 396 { 397 if (p_base == null) 398 { 399 throw new MalformedURIException(Utils.messages.createMessage(MsgKey.ER_NO_SCHEME_IN_URI, new Object[]{uriSpec})); //"No scheme found in URI: "+uriSpec); 400 } 401 } 402 else 403 { 404 initializeScheme(uriSpec); 405 uriSpec = uriSpec.substring(colonIndex+1); 406 uriSpecLen = uriSpec.length(); 407 } 408 409 // two slashes means generic URI syntax, so we get the authority 410 if (((index + 1) < uriSpecLen) 411 && (uriSpec.substring(index).startsWith("//"))) 412 { 413 index += 2; 414 415 int startPos = index; 416 417 // get authority - everything up to path, query or fragment 418 char testChar = '\0'; 419 420 while (index < uriSpecLen) 421 { 422 testChar = uriSpec.charAt(index); 423 424 if (testChar == '/' || testChar == '?' || testChar == '#') 425 { 426 break; 427 } 428 429 index++; 430 } 431 432 // if we found authority, parse it out, otherwise we set the 433 // host to empty string 434 if (index > startPos) 435 { 436 initializeAuthority(uriSpec.substring(startPos, index)); 437 } 438 else 439 { 440 m_host = ""; 441 } 442 } 443 444 initializePath(uriSpec.substring(index)); 445 446 // Resolve relative URI to base URI - see RFC 2396 Section 5.2 447 // In some cases, it might make more sense to throw an exception 448 // (when scheme is specified is the string spec and the base URI 449 // is also specified, for example), but we're just following the 450 // RFC specifications 451 if (p_base != null) 452 { 453 454 // check to see if this is the current doc - RFC 2396 5.2 #2 455 // note that this is slightly different from the RFC spec in that 456 // we don't include the check for query string being null 457 // - this handles cases where the urispec is just a query 458 // string or a fragment (e.g. "?y" or "#s") - 459 // see <http://www.ics.uci.edu/~fielding/url/test1.html> which 460 // identified this as a bug in the RFC 461 if (m_path.length() == 0 && m_scheme == null && m_host == null) 462 { 463 m_scheme = p_base.getScheme(); 464 m_userinfo = p_base.getUserinfo(); 465 m_host = p_base.getHost(); 466 m_port = p_base.getPort(); 467 m_path = p_base.getPath(); 468 469 if (m_queryString == null) 470 { 471 m_queryString = p_base.getQueryString(); 472 } 473 474 return; 475 } 476 477 // check for scheme - RFC 2396 5.2 #3 478 // if we found a scheme, it means absolute URI, so we're done 479 if (m_scheme == null) 480 { 481 m_scheme = p_base.getScheme(); 482 } 483 484 // check for authority - RFC 2396 5.2 #4 485 // if we found a host, then we've got a network path, so we're done 486 if (m_host == null) 487 { 488 m_userinfo = p_base.getUserinfo(); 489 m_host = p_base.getHost(); 490 m_port = p_base.getPort(); 491 } 492 else 493 { 494 return; 495 } 496 497 // check for absolute path - RFC 2396 5.2 #5 498 if (m_path.length() > 0 && m_path.startsWith("/")) 499 { 500 return; 501 } 502 503 // if we get to this point, we need to resolve relative path 504 // RFC 2396 5.2 #6 505 String path = ""; 506 String basePath = p_base.getPath(); 507 508 // 6a - get all but the last segment of the base URI path 509 if (basePath != null) 510 { 511 int lastSlash = basePath.lastIndexOf('/'); 512 513 if (lastSlash != -1) 514 { 515 path = basePath.substring(0, lastSlash + 1); 516 } 517 } 518 519 // 6b - append the relative URI path 520 path = path.concat(m_path); 521 522 // 6c - remove all "./" where "." is a complete path segment 523 index = -1; 524 525 while ((index = path.indexOf("/./")) != -1) 526 { 527 path = path.substring(0, index + 1).concat(path.substring(index + 3)); 528 } 529 530 // 6d - remove "." if path ends with "." as a complete path segment 531 if (path.endsWith("/.")) 532 { 533 path = path.substring(0, path.length() - 1); 534 } 535 536 // 6e - remove all "<segment>/../" where "<segment>" is a complete 537 // path segment not equal to ".." 538 index = -1; 539 540 int segIndex = -1; 541 String tempString = null; 542 543 while ((index = path.indexOf("/../")) > 0) 544 { 545 tempString = path.substring(0, path.indexOf("/../")); 546 segIndex = tempString.lastIndexOf('/'); 547 548 if (segIndex != -1) 549 { 550 if (!tempString.substring(segIndex++).equals("..")) 551 { 552 path = path.substring(0, segIndex).concat(path.substring(index 553 + 4)); 554 } 555 } 556 } 557 558 // 6f - remove ending "<segment>/.." where "<segment>" is a 559 // complete path segment 560 if (path.endsWith("/..")) 561 { 562 tempString = path.substring(0, path.length() - 3); 563 segIndex = tempString.lastIndexOf('/'); 564 565 if (segIndex != -1) 566 { 567 path = path.substring(0, segIndex + 1); 568 } 569 } 570 571 m_path = path; 572 } 573 } 574 575 /** 576 * Initialize the scheme for this URI from a URI string spec. 577 * 578 * @param p_uriSpec the URI specification (cannot be null) 579 * 580 * @throws MalformedURIException if URI does not have a conformant 581 * scheme 582 */ 583 private void initializeScheme(String p_uriSpec) throws MalformedURIException 584 { 585 586 int uriSpecLen = p_uriSpec.length(); 587 int index = 0; 588 String scheme = null; 589 char testChar = '\0'; 590 591 while (index < uriSpecLen) 592 { 593 testChar = p_uriSpec.charAt(index); 594 595 if (testChar == ':' || testChar == '/' || testChar == '?' 596 || testChar == '#') 597 { 598 break; 599 } 600 601 index++; 602 } 603 604 scheme = p_uriSpec.substring(0, index); 605 606 if (scheme.length() == 0) 607 { 608 throw new MalformedURIException(Utils.messages.createMessage(MsgKey.ER_NO_SCHEME_INURI, null)); //"No scheme found in URI."); 609 } 610 else 611 { 612 setScheme(scheme); 613 } 614 } 615 616 /** 617 * Initialize the authority (userinfo, host and port) for this 618 * URI from a URI string spec. 619 * 620 * @param p_uriSpec the URI specification (cannot be null) 621 * 622 * @throws MalformedURIException if p_uriSpec violates syntax rules 623 */ 624 private void initializeAuthority(String p_uriSpec) 625 throws MalformedURIException 626 { 627 628 int index = 0; 629 int start = 0; 630 int end = p_uriSpec.length(); 631 char testChar = '\0'; 632 String userinfo = null; 633 634 // userinfo is everything up @ 635 if (p_uriSpec.indexOf('@', start) != -1) 636 { 637 while (index < end) 638 { 639 testChar = p_uriSpec.charAt(index); 640 641 if (testChar == '@') 642 { 643 break; 644 } 645 646 index++; 647 } 648 649 userinfo = p_uriSpec.substring(start, index); 650 651 index++; 652 } 653 654 // host is everything up to ':' 655 String host = null; 656 657 start = index; 658 659 while (index < end) 660 { 661 testChar = p_uriSpec.charAt(index); 662 663 if (testChar == ':') 664 { 665 break; 666 } 667 668 index++; 669 } 670 671 host = p_uriSpec.substring(start, index); 672 673 int port = -1; 674 675 if (host.length() > 0) 676 { 677 678 // port 679 if (testChar == ':') 680 { 681 index++; 682 683 start = index; 684 685 while (index < end) 686 { 687 index++; 688 } 689 690 String portStr = p_uriSpec.substring(start, index); 691 692 if (portStr.length() > 0) 693 { 694 for (int i = 0; i < portStr.length(); i++) 695 { 696 if (!isDigit(portStr.charAt(i))) 697 { 698 throw new MalformedURIException( 699 portStr + " is invalid. Port should only contain digits!"); 700 } 701 } 702 703 try 704 { 705 port = Integer.parseInt(portStr); 706 } 707 catch (NumberFormatException nfe) 708 { 709 710 // can't happen 711 } 712 } 713 } 714 } 715 716 setHost(host); 717 setPort(port); 718 setUserinfo(userinfo); 719 } 720 721 /** 722 * Initialize the path for this URI from a URI string spec. 723 * 724 * @param p_uriSpec the URI specification (cannot be null) 725 * 726 * @throws MalformedURIException if p_uriSpec violates syntax rules 727 */ 728 private void initializePath(String p_uriSpec) throws MalformedURIException 729 { 730 731 if (p_uriSpec == null) 732 { 733 throw new MalformedURIException( 734 "Cannot initialize path from null string!"); 735 } 736 737 int index = 0; 738 int start = 0; 739 int end = p_uriSpec.length(); 740 char testChar = '\0'; 741 742 // path - everything up to query string or fragment 743 while (index < end) 744 { 745 testChar = p_uriSpec.charAt(index); 746 747 if (testChar == '?' || testChar == '#') 748 { 749 break; 750 } 751 752 // check for valid escape sequence 753 if (testChar == '%') 754 { 755 if (index + 2 >= end ||!isHex(p_uriSpec.charAt(index + 1)) 756 ||!isHex(p_uriSpec.charAt(index + 2))) 757 { 758 throw new MalformedURIException( 759 Utils.messages.createMessage(MsgKey.ER_PATH_CONTAINS_INVALID_ESCAPE_SEQUENCE, null)); //"Path contains invalid escape sequence!"); 760 } 761 } 762 else if (!isReservedCharacter(testChar) 763 &&!isUnreservedCharacter(testChar)) 764 { 765 if ('\\' != testChar) 766 throw new MalformedURIException(Utils.messages.createMessage(MsgKey.ER_PATH_INVALID_CHAR, new Object[]{String.valueOf(testChar)})); //"Path contains invalid character: " 767 //+ testChar); 768 } 769 770 index++; 771 } 772 773 m_path = p_uriSpec.substring(start, index); 774 775 // query - starts with ? and up to fragment or end 776 if (testChar == '?') 777 { 778 index++; 779 780 start = index; 781 782 while (index < end) 783 { 784 testChar = p_uriSpec.charAt(index); 785 786 if (testChar == '#') 787 { 788 break; 789 } 790 791 if (testChar == '%') 792 { 793 if (index + 2 >= end ||!isHex(p_uriSpec.charAt(index + 1)) 794 ||!isHex(p_uriSpec.charAt(index + 2))) 795 { 796 throw new MalformedURIException( 797 "Query string contains invalid escape sequence!"); 798 } 799 } 800 else if (!isReservedCharacter(testChar) 801 &&!isUnreservedCharacter(testChar)) 802 { 803 throw new MalformedURIException( 804 "Query string contains invalid character:" + testChar); 805 } 806 807 index++; 808 } 809 810 m_queryString = p_uriSpec.substring(start, index); 811 } 812 813 // fragment - starts with # 814 if (testChar == '#') 815 { 816 index++; 817 818 start = index; 819 820 while (index < end) 821 { 822 testChar = p_uriSpec.charAt(index); 823 824 if (testChar == '%') 825 { 826 if (index + 2 >= end ||!isHex(p_uriSpec.charAt(index + 1)) 827 ||!isHex(p_uriSpec.charAt(index + 2))) 828 { 829 throw new MalformedURIException( 830 "Fragment contains invalid escape sequence!"); 831 } 832 } 833 else if (!isReservedCharacter(testChar) 834 &&!isUnreservedCharacter(testChar)) 835 { 836 throw new MalformedURIException( 837 "Fragment contains invalid character:" + testChar); 838 } 839 840 index++; 841 } 842 843 m_fragment = p_uriSpec.substring(start, index); 844 } 845 } 846 847 /** 848 * Get the scheme for this URI. 849 * 850 * @return the scheme for this URI 851 */ 852 public String getScheme() 853 { 854 return m_scheme; 855 } 856 857 /** 858 * Get the scheme-specific part for this URI (everything following the 859 * scheme and the first colon). See RFC 2396 Section 5.2 for spec. 860 * 861 * @return the scheme-specific part for this URI 862 */ 863 public String getSchemeSpecificPart() 864 { 865 866 StringBuffer schemespec = new StringBuffer(); 867 868 if (m_userinfo != null || m_host != null || m_port != -1) 869 { 870 schemespec.append("//"); 871 } 872 873 if (m_userinfo != null) 874 { 875 schemespec.append(m_userinfo); 876 schemespec.append('@'); 877 } 878 879 if (m_host != null) 880 { 881 schemespec.append(m_host); 882 } 883 884 if (m_port != -1) 885 { 886 schemespec.append(':'); 887 schemespec.append(m_port); 888 } 889 890 if (m_path != null) 891 { 892 schemespec.append((m_path)); 893 } 894 895 if (m_queryString != null) 896 { 897 schemespec.append('?'); 898 schemespec.append(m_queryString); 899 } 900 901 if (m_fragment != null) 902 { 903 schemespec.append('#'); 904 schemespec.append(m_fragment); 905 } 906 907 return schemespec.toString(); 908 } 909 910 /** 911 * Get the userinfo for this URI. 912 * 913 * @return the userinfo for this URI (null if not specified). 914 */ 915 public String getUserinfo() 916 { 917 return m_userinfo; 918 } 919 920 /** 921 * Get the host for this URI. 922 * 923 * @return the host for this URI (null if not specified). 924 */ 925 public String getHost() 926 { 927 return m_host; 928 } 929 930 /** 931 * Get the port for this URI. 932 * 933 * @return the port for this URI (-1 if not specified). 934 */ 935 public int getPort() 936 { 937 return m_port; 938 } 939 940 /** 941 * Get the path for this URI (optionally with the query string and 942 * fragment). 943 * 944 * @param p_includeQueryString if true (and query string is not null), 945 * then a "?" followed by the query string 946 * will be appended 947 * @param p_includeFragment if true (and fragment is not null), 948 * then a "#" followed by the fragment 949 * will be appended 950 * 951 * @return the path for this URI possibly including the query string 952 * and fragment 953 */ 954 public String getPath(boolean p_includeQueryString, 955 boolean p_includeFragment) 956 { 957 958 StringBuffer pathString = new StringBuffer(m_path); 959 960 if (p_includeQueryString && m_queryString != null) 961 { 962 pathString.append('?'); 963 pathString.append(m_queryString); 964 } 965 966 if (p_includeFragment && m_fragment != null) 967 { 968 pathString.append('#'); 969 pathString.append(m_fragment); 970 } 971 972 return pathString.toString(); 973 } 974 975 /** 976 * Get the path for this URI. Note that the value returned is the path 977 * only and does not include the query string or fragment. 978 * 979 * @return the path for this URI. 980 */ 981 public String getPath() 982 { 983 return m_path; 984 } 985 986 /** 987 * Get the query string for this URI. 988 * 989 * @return the query string for this URI. Null is returned if there 990 * was no "?" in the URI spec, empty string if there was a 991 * "?" but no query string following it. 992 */ 993 public String getQueryString() 994 { 995 return m_queryString; 996 } 997 998 /** 999 * Get the fragment for this URI. 1000 * 1001 * @return the fragment for this URI. Null is returned if there 1002 * was no "#" in the URI spec, empty string if there was a 1003 * "#" but no fragment following it. 1004 */ 1005 public String getFragment() 1006 { 1007 return m_fragment; 1008 } 1009 1010 /** 1011 * Set the scheme for this URI. The scheme is converted to lowercase 1012 * before it is set. 1013 * 1014 * @param p_scheme the scheme for this URI (cannot be null) 1015 * 1016 * @throws MalformedURIException if p_scheme is not a conformant 1017 * scheme name 1018 */ 1019 public void setScheme(String p_scheme) throws MalformedURIException 1020 { 1021 1022 if (p_scheme == null) 1023 { 1024 throw new MalformedURIException(Utils.messages.createMessage(MsgKey.ER_SCHEME_FROM_NULL_STRING, null)); //"Cannot set scheme from null string!"); 1025 } 1026 1027 if (!isConformantSchemeName(p_scheme)) 1028 { 1029 throw new MalformedURIException(Utils.messages.createMessage(MsgKey.ER_SCHEME_NOT_CONFORMANT, null)); //"The scheme is not conformant."); 1030 } 1031 1032 m_scheme = p_scheme.toLowerCase(); 1033 } 1034 1035 /** 1036 * Set the userinfo for this URI. If a non-null value is passed in and 1037 * the host value is null, then an exception is thrown. 1038 * 1039 * @param p_userinfo the userinfo for this URI 1040 * 1041 * @throws MalformedURIException if p_userinfo contains invalid 1042 * characters 1043 */ 1044 public void setUserinfo(String p_userinfo) throws MalformedURIException 1045 { 1046 1047 if (p_userinfo == null) 1048 { 1049 m_userinfo = null; 1050 } 1051 else 1052 { 1053 if (m_host == null) 1054 { 1055 throw new MalformedURIException( 1056 "Userinfo cannot be set when host is null!"); 1057 } 1058 1059 // userinfo can contain alphanumerics, mark characters, escaped 1060 // and ';',':','&','=','+','$',',' 1061 int index = 0; 1062 int end = p_userinfo.length(); 1063 char testChar = '\0'; 1064 1065 while (index < end) 1066 { 1067 testChar = p_userinfo.charAt(index); 1068 1069 if (testChar == '%') 1070 { 1071 if (index + 2 >= end ||!isHex(p_userinfo.charAt(index + 1)) 1072 ||!isHex(p_userinfo.charAt(index + 2))) 1073 { 1074 throw new MalformedURIException( 1075 "Userinfo contains invalid escape sequence!"); 1076 } 1077 } 1078 else if (!isUnreservedCharacter(testChar) 1079 && USERINFO_CHARACTERS.indexOf(testChar) == -1) 1080 { 1081 throw new MalformedURIException( 1082 "Userinfo contains invalid character:" + testChar); 1083 } 1084 1085 index++; 1086 } 1087 } 1088 1089 m_userinfo = p_userinfo; 1090 } 1091 1092 /** 1093 * Set the host for this URI. If null is passed in, the userinfo 1094 * field is also set to null and the port is set to -1. 1095 * 1096 * @param p_host the host for this URI 1097 * 1098 * @throws MalformedURIException if p_host is not a valid IP 1099 * address or DNS hostname. 1100 */ 1101 public void setHost(String p_host) throws MalformedURIException 1102 { 1103 1104 if (p_host == null || p_host.trim().length() == 0) 1105 { 1106 m_host = p_host; 1107 m_userinfo = null; 1108 m_port = -1; 1109 } 1110 else if (!isWellFormedAddress(p_host)) 1111 { 1112 throw new MalformedURIException(Utils.messages.createMessage(MsgKey.ER_HOST_ADDRESS_NOT_WELLFORMED, null)); //"Host is not a well formed address!"); 1113 } 1114 1115 m_host = p_host; 1116 } 1117 1118 /** 1119 * Set the port for this URI. -1 is used to indicate that the port is 1120 * not specified, otherwise valid port numbers are between 0 and 65535. 1121 * If a valid port number is passed in and the host field is null, 1122 * an exception is thrown. 1123 * 1124 * @param p_port the port number for this URI 1125 * 1126 * @throws MalformedURIException if p_port is not -1 and not a 1127 * valid port number 1128 */ 1129 public void setPort(int p_port) throws MalformedURIException 1130 { 1131 1132 if (p_port >= 0 && p_port <= 65535) 1133 { 1134 if (m_host == null) 1135 { 1136 throw new MalformedURIException( 1137 Utils.messages.createMessage(MsgKey.ER_PORT_WHEN_HOST_NULL, null)); //"Port cannot be set when host is null!"); 1138 } 1139 } 1140 else if (p_port != -1) 1141 { 1142 throw new MalformedURIException(Utils.messages.createMessage(MsgKey.ER_INVALID_PORT, null)); //"Invalid port number!"); 1143 } 1144 1145 m_port = p_port; 1146 } 1147 1148 /** 1149 * Set the path for this URI. If the supplied path is null, then the 1150 * query string and fragment are set to null as well. If the supplied 1151 * path includes a query string and/or fragment, these fields will be 1152 * parsed and set as well. Note that, for URIs following the "generic 1153 * URI" syntax, the path specified should start with a slash. 1154 * For URIs that do not follow the generic URI syntax, this method 1155 * sets the scheme-specific part. 1156 * 1157 * @param p_path the path for this URI (may be null) 1158 * 1159 * @throws MalformedURIException if p_path contains invalid 1160 * characters 1161 */ 1162 public void setPath(String p_path) throws MalformedURIException 1163 { 1164 1165 if (p_path == null) 1166 { 1167 m_path = null; 1168 m_queryString = null; 1169 m_fragment = null; 1170 } 1171 else 1172 { 1173 initializePath(p_path); 1174 } 1175 } 1176 1177 /** 1178 * Append to the end of the path of this URI. If the current path does 1179 * not end in a slash and the path to be appended does not begin with 1180 * a slash, a slash will be appended to the current path before the 1181 * new segment is added. Also, if the current path ends in a slash 1182 * and the new segment begins with a slash, the extra slash will be 1183 * removed before the new segment is appended. 1184 * 1185 * @param p_addToPath the new segment to be added to the current path 1186 * 1187 * @throws MalformedURIException if p_addToPath contains syntax 1188 * errors 1189 */ 1190 public void appendPath(String p_addToPath) throws MalformedURIException 1191 { 1192 1193 if (p_addToPath == null || p_addToPath.trim().length() == 0) 1194 { 1195 return; 1196 } 1197 1198 if (!isURIString(p_addToPath)) 1199 { 1200 throw new MalformedURIException(Utils.messages.createMessage(MsgKey.ER_PATH_INVALID_CHAR, new Object[]{p_addToPath})); //"Path contains invalid character!"); 1201 } 1202 1203 if (m_path == null || m_path.trim().length() == 0) 1204 { 1205 if (p_addToPath.startsWith("/")) 1206 { 1207 m_path = p_addToPath; 1208 } 1209 else 1210 { 1211 m_path = "/" + p_addToPath; 1212 } 1213 } 1214 else if (m_path.endsWith("/")) 1215 { 1216 if (p_addToPath.startsWith("/")) 1217 { 1218 m_path = m_path.concat(p_addToPath.substring(1)); 1219 } 1220 else 1221 { 1222 m_path = m_path.concat(p_addToPath); 1223 } 1224 } 1225 else 1226 { 1227 if (p_addToPath.startsWith("/")) 1228 { 1229 m_path = m_path.concat(p_addToPath); 1230 } 1231 else 1232 { 1233 m_path = m_path.concat("/" + p_addToPath); 1234 } 1235 } 1236 } 1237 1238 /** 1239 * Set the query string for this URI. A non-null value is valid only 1240 * if this is an URI conforming to the generic URI syntax and 1241 * the path value is not null. 1242 * 1243 * @param p_queryString the query string for this URI 1244 * 1245 * @throws MalformedURIException if p_queryString is not null and this 1246 * URI does not conform to the generic 1247 * URI syntax or if the path is null 1248 */ 1249 public void setQueryString(String p_queryString) 1250 throws MalformedURIException 1251 { 1252 1253 if (p_queryString == null) 1254 { 1255 m_queryString = null; 1256 } 1257 else if (!isGenericURI()) 1258 { 1259 throw new MalformedURIException( 1260 "Query string can only be set for a generic URI!"); 1261 } 1262 else if (getPath() == null) 1263 { 1264 throw new MalformedURIException( 1265 "Query string cannot be set when path is null!"); 1266 } 1267 else if (!isURIString(p_queryString)) 1268 { 1269 throw new MalformedURIException( 1270 "Query string contains invalid character!"); 1271 } 1272 else 1273 { 1274 m_queryString = p_queryString; 1275 } 1276 } 1277 1278 /** 1279 * Set the fragment for this URI. A non-null value is valid only 1280 * if this is a URI conforming to the generic URI syntax and 1281 * the path value is not null. 1282 * 1283 * @param p_fragment the fragment for this URI 1284 * 1285 * @throws MalformedURIException if p_fragment is not null and this 1286 * URI does not conform to the generic 1287 * URI syntax or if the path is null 1288 */ 1289 public void setFragment(String p_fragment) throws MalformedURIException 1290 { 1291 1292 if (p_fragment == null) 1293 { 1294 m_fragment = null; 1295 } 1296 else if (!isGenericURI()) 1297 { 1298 throw new MalformedURIException( 1299 Utils.messages.createMessage(MsgKey.ER_FRAG_FOR_GENERIC_URI, null)); //"Fragment can only be set for a generic URI!"); 1300 } 1301 else if (getPath() == null) 1302 { 1303 throw new MalformedURIException( 1304 Utils.messages.createMessage(MsgKey.ER_FRAG_WHEN_PATH_NULL, null)); //"Fragment cannot be set when path is null!"); 1305 } 1306 else if (!isURIString(p_fragment)) 1307 { 1308 throw new MalformedURIException(Utils.messages.createMessage(MsgKey.ER_FRAG_INVALID_CHAR, null)); //"Fragment contains invalid character!"); 1309 } 1310 else 1311 { 1312 m_fragment = p_fragment; 1313 } 1314 } 1315 1316 /** 1317 * Determines if the passed-in Object is equivalent to this URI. 1318 * 1319 * @param p_test the Object to test for equality. 1320 * 1321 * @return true if p_test is a URI with all values equal to this 1322 * URI, false otherwise 1323 */ 1324 public boolean equals(Object p_test) 1325 { 1326 1327 if (p_test instanceof URI) 1328 { 1329 URI testURI = (URI) p_test; 1330 1331 if (((m_scheme == null && testURI.m_scheme == null) || (m_scheme != null && testURI.m_scheme != null && m_scheme.equals( 1332 testURI.m_scheme))) && ((m_userinfo == null && testURI.m_userinfo == null) || (m_userinfo != null && testURI.m_userinfo != null && m_userinfo.equals( 1333 testURI.m_userinfo))) && ((m_host == null && testURI.m_host == null) || (m_host != null && testURI.m_host != null && m_host.equals( 1334 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( 1335 testURI.m_path))) && ((m_queryString == null && testURI.m_queryString == null) || (m_queryString != null && testURI.m_queryString != null && m_queryString.equals( 1336 testURI.m_queryString))) && ((m_fragment == null && testURI.m_fragment == null) || (m_fragment != null && testURI.m_fragment != null && m_fragment.equals( 1337 testURI.m_fragment)))) 1338 { 1339 return true; 1340 } 1341 } 1342 1343 return false; 1344 } 1345 1346 /** 1347 * Get the URI as a string specification. See RFC 2396 Section 5.2. 1348 * 1349 * @return the URI string specification 1350 */ 1351 public String toString() 1352 { 1353 1354 StringBuffer uriSpecString = new StringBuffer(); 1355 1356 if (m_scheme != null) 1357 { 1358 uriSpecString.append(m_scheme); 1359 uriSpecString.append(':'); 1360 } 1361 1362 uriSpecString.append(getSchemeSpecificPart()); 1363 1364 return uriSpecString.toString(); 1365 } 1366 1367 /** 1368 * Get the indicator as to whether this URI uses the "generic URI" 1369 * syntax. 1370 * 1371 * @return true if this URI uses the "generic URI" syntax, false 1372 * otherwise 1373 */ 1374 public boolean isGenericURI() 1375 { 1376 1377 // presence of the host (whether valid or empty) means 1378 // double-slashes which means generic uri 1379 return (m_host != null); 1380 } 1381 1382 /** 1383 * Determine whether a scheme conforms to the rules for a scheme name. 1384 * A scheme is conformant if it starts with an alphanumeric, and 1385 * contains only alphanumerics, '+','-' and '.'. 1386 * 1387 * 1388 * @param p_scheme The sheme name to check 1389 * @return true if the scheme is conformant, false otherwise 1390 */ 1391 public static boolean isConformantSchemeName(String p_scheme) 1392 { 1393 1394 if (p_scheme == null || p_scheme.trim().length() == 0) 1395 { 1396 return false; 1397 } 1398 1399 if (!isAlpha(p_scheme.charAt(0))) 1400 { 1401 return false; 1402 } 1403 1404 char testChar; 1405 1406 for (int i = 1; i < p_scheme.length(); i++) 1407 { 1408 testChar = p_scheme.charAt(i); 1409 1410 if (!isAlphanum(testChar) && SCHEME_CHARACTERS.indexOf(testChar) == -1) 1411 { 1412 return false; 1413 } 1414 } 1415 1416 return true; 1417 } 1418 1419 /** 1420 * Determine whether a string is syntactically capable of representing 1421 * a valid IPv4 address or the domain name of a network host. A valid 1422 * IPv4 address consists of four decimal digit groups separated by a 1423 * '.'. A hostname consists of domain labels (each of which must 1424 * begin and end with an alphanumeric but may contain '-') separated 1425 * & by a '.'. See RFC 2396 Section 3.2.2. 1426 * 1427 * 1428 * @param p_address The address string to check 1429 * @return true if the string is a syntactically valid IPv4 address 1430 * or hostname 1431 */ 1432 public static boolean isWellFormedAddress(String p_address) 1433 { 1434 1435 if (p_address == null) 1436 { 1437 return false; 1438 } 1439 1440 String address = p_address.trim(); 1441 int addrLength = address.length(); 1442 1443 if (addrLength == 0 || addrLength > 255) 1444 { 1445 return false; 1446 } 1447 1448 if (address.startsWith(".") || address.startsWith("-")) 1449 { 1450 return false; 1451 } 1452 1453 // rightmost domain label starting with digit indicates IP address 1454 // since top level domain label can only start with an alpha 1455 // see RFC 2396 Section 3.2.2 1456 int index = address.lastIndexOf('.'); 1457 1458 if (address.endsWith(".")) 1459 { 1460 index = address.substring(0, index).lastIndexOf('.'); 1461 } 1462 1463 if (index + 1 < addrLength && isDigit(p_address.charAt(index + 1))) 1464 { 1465 char testChar; 1466 int numDots = 0; 1467 1468 // make sure that 1) we see only digits and dot separators, 2) that 1469 // any dot separator is preceded and followed by a digit and 1470 // 3) that we find 3 dots 1471 for (int i = 0; i < addrLength; i++) 1472 { 1473 testChar = address.charAt(i); 1474 1475 if (testChar == '.') 1476 { 1477 if (!isDigit(address.charAt(i - 1)) 1478 || (i + 1 < addrLength &&!isDigit(address.charAt(i + 1)))) 1479 { 1480 return false; 1481 } 1482 1483 numDots++; 1484 } 1485 else if (!isDigit(testChar)) 1486 { 1487 return false; 1488 } 1489 } 1490 1491 if (numDots != 3) 1492 { 1493 return false; 1494 } 1495 } 1496 else 1497 { 1498 1499 // domain labels can contain alphanumerics and '-" 1500 // but must start and end with an alphanumeric 1501 char testChar; 1502 1503 for (int i = 0; i < addrLength; i++) 1504 { 1505 testChar = address.charAt(i); 1506 1507 if (testChar == '.') 1508 { 1509 if (!isAlphanum(address.charAt(i - 1))) 1510 { 1511 return false; 1512 } 1513 1514 if (i + 1 < addrLength &&!isAlphanum(address.charAt(i + 1))) 1515 { 1516 return false; 1517 } 1518 } 1519 else if (!isAlphanum(testChar) && testChar != '-') 1520 { 1521 return false; 1522 } 1523 } 1524 } 1525 1526 return true; 1527 } 1528 1529 /** 1530 * Determine whether a char is a digit. 1531 * 1532 * 1533 * @param p_char the character to check 1534 * @return true if the char is betweeen '0' and '9', false otherwise 1535 */ 1536 private static boolean isDigit(char p_char) 1537 { 1538 return p_char >= '0' && p_char <= '9'; 1539 } 1540 1541 /** 1542 * Determine whether a character is a hexadecimal character. 1543 * 1544 * 1545 * @param p_char the character to check 1546 * @return true if the char is betweeen '0' and '9', 'a' and 'f' 1547 * or 'A' and 'F', false otherwise 1548 */ 1549 private static boolean isHex(char p_char) 1550 { 1551 return (isDigit(p_char) || (p_char >= 'a' && p_char <= 'f') 1552 || (p_char >= 'A' && p_char <= 'F')); 1553 } 1554 1555 /** 1556 * Determine whether a char is an alphabetic character: a-z or A-Z 1557 * 1558 * 1559 * @param p_char the character to check 1560 * @return true if the char is alphabetic, false otherwise 1561 */ 1562 private static boolean isAlpha(char p_char) 1563 { 1564 return ((p_char >= 'a' && p_char <= 'z') 1565 || (p_char >= 'A' && p_char <= 'Z')); 1566 } 1567 1568 /** 1569 * Determine whether a char is an alphanumeric: 0-9, a-z or A-Z 1570 * 1571 * 1572 * @param p_char the character to check 1573 * @return true if the char is alphanumeric, false otherwise 1574 */ 1575 private static boolean isAlphanum(char p_char) 1576 { 1577 return (isAlpha(p_char) || isDigit(p_char)); 1578 } 1579 1580 /** 1581 * Determine whether a character is a reserved character: 1582 * ';', '/', '?', ':', '@', '&', '=', '+', '$' or ',' 1583 * 1584 * 1585 * @param p_char the character to check 1586 * @return true if the string contains any reserved characters 1587 */ 1588 private static boolean isReservedCharacter(char p_char) 1589 { 1590 return RESERVED_CHARACTERS.indexOf(p_char) != -1; 1591 } 1592 1593 /** 1594 * Determine whether a char is an unreserved character. 1595 * 1596 * 1597 * @param p_char the character to check 1598 * @return true if the char is unreserved, false otherwise 1599 */ 1600 private static boolean isUnreservedCharacter(char p_char) 1601 { 1602 return (isAlphanum(p_char) || MARK_CHARACTERS.indexOf(p_char) != -1); 1603 } 1604 1605 /** 1606 * Determine whether a given string contains only URI characters (also 1607 * called "uric" in RFC 2396). uric consist of all reserved 1608 * characters, unreserved characters and escaped characters. 1609 * 1610 * 1611 * @param p_uric URI string 1612 * @return true if the string is comprised of uric, false otherwise 1613 */ 1614 private static boolean isURIString(String p_uric) 1615 { 1616 1617 if (p_uric == null) 1618 { 1619 return false; 1620 } 1621 1622 int end = p_uric.length(); 1623 char testChar = '\0'; 1624 1625 for (int i = 0; i < end; i++) 1626 { 1627 testChar = p_uric.charAt(i); 1628 1629 if (testChar == '%') 1630 { 1631 if (i + 2 >= end ||!isHex(p_uric.charAt(i + 1)) 1632 ||!isHex(p_uric.charAt(i + 2))) 1633 { 1634 return false; 1635 } 1636 else 1637 { 1638 i += 2; 1639 1640 continue; 1641 } 1642 } 1643 1644 if (isReservedCharacter(testChar) || isUnreservedCharacter(testChar)) 1645 { 1646 continue; 1647 } 1648 else 1649 { 1650 return false; 1651 } 1652 } 1653 1654 return true; 1655 } 1656 }