0001: /*
0002: * The Apache Software License, Version 1.1
0003: *
0004: *
0005: * Copyright (c) 1999,2001 The Apache Software Foundation. All rights
0006: * reserved.
0007: *
0008: * Redistribution and use in source and binary forms, with or without
0009: * modification, are permitted provided that the following conditions
0010: * are met:
0011: *
0012: * 1. Redistributions of source code must retain the above copyright
0013: * notice, this list of conditions and the following disclaimer.
0014: *
0015: * 2. Redistributions in binary form must reproduce the above copyright
0016: * notice, this list of conditions and the following disclaimer in
0017: * the documentation and/or other materials provided with the
0018: * distribution.
0019: *
0020: * 3. The end-user documentation included with the redistribution,
0021: * if any, must include the following acknowledgment:
0022: * "This product includes software developed by the
0023: * Apache Software Foundation (http://www.apache.org/)."
0024: * Alternately, this acknowledgment may appear in the software itself,
0025: * if and wherever such third-party acknowledgments normally appear.
0026: *
0027: * 4. The names "Xerces" and "Apache Software Foundation" must
0028: * not be used to endorse or promote products derived from this
0029: * software without prior written permission. For written
0030: * permission, please contact apache@apache.org.
0031: *
0032: * 5. Products derived from this software may not be called "Apache",
0033: * nor may "Apache" appear in their name, without prior written
0034: * permission of the Apache Software Foundation.
0035: *
0036: * THIS SOFTWARE IS PROVIDED ``AS IS'' AND ANY EXPRESSED OR IMPLIED
0037: * WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
0038: * OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
0039: * DISCLAIMED. IN NO EVENT SHALL THE APACHE SOFTWARE FOUNDATION OR
0040: * ITS CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
0041: * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
0042: * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF
0043: * USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
0044: * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY,
0045: * OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT
0046: * OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
0047: * SUCH DAMAGE.
0048: * ====================================================================
0049: *
0050: * This software consists of voluntary contributions made by many
0051: * individuals on behalf of the Apache Software Foundation and was
0052: * originally based on software copyright (c) 1999, iClick Inc.,
0053: * http://www.apache.org. For more information on the Apache Software
0054: * Foundation, please see <http://www.apache.org/>.
0055: */
0056:
0057: package org.apache.xerces.utils;
0058:
0059: import java.io.IOException;
0060: import java.io.Serializable;
0061:
0062: /**********************************************************************
0063: * A class to represent a Uniform Resource Identifier (URI). This class
0064: * is designed to handle the parsing of URIs and provide access to
0065: * the various components (scheme, host, port, userinfo, path, query
0066: * string and fragment) that may constitute a URI.
0067: * <p>
0068: * Parsing of a URI specification is done according to the URI
0069: * syntax described in RFC 2396
0070: * <http://www.ietf.org/rfc/rfc2396.txt?number=2396>. Every URI consists
0071: * of a scheme, followed by a colon (':'), followed by a scheme-specific
0072: * part. For URIs that follow the "generic URI" syntax, the scheme-
0073: * specific part begins with two slashes ("//") and may be followed
0074: * by an authority segment (comprised of user information, host, and
0075: * port), path segment, query segment and fragment. Note that RFC 2396
0076: * no longer specifies the use of the parameters segment and excludes
0077: * the "user:password" syntax as part of the authority segment. If
0078: * "user:password" appears in a URI, the entire user/password string
0079: * is stored as userinfo.
0080: * <p>
0081: * For URIs that do not follow the "generic URI" syntax (e.g. mailto),
0082: * the entire scheme-specific part is treated as the "path" portion
0083: * of the URI.
0084: * <p>
0085: * Note that, unlike the java.net.URL class, this class does not provide
0086: * any built-in network access functionality nor does it provide any
0087: * scheme-specific functionality (for example, it does not know a
0088: * default port for a specific scheme). Rather, it only knows the
0089: * grammar and basic set of operations that can be applied to a URI.
0090: *
0091: * @version $Id: URI.java,v 1.8.2.1 2001/11/12 20:24:15 sandygao Exp $
0092: *
0093: **********************************************************************/
0094: public class URI implements Serializable {
0095:
0096: /*******************************************************************
0097: * MalformedURIExceptions are thrown in the process of building a URI
0098: * or setting fields on a URI when an operation would result in an
0099: * invalid URI specification.
0100: *
0101: ********************************************************************/
0102: public static class MalformedURIException extends IOException {
0103:
0104: /******************************************************************
0105: * Constructs a <code>MalformedURIException</code> with no specified
0106: * detail message.
0107: ******************************************************************/
0108: public MalformedURIException() {
0109: super ();
0110: }
0111:
0112: /*****************************************************************
0113: * Constructs a <code>MalformedURIException</code> with the
0114: * specified detail message.
0115: *
0116: * @param p_msg the detail message.
0117: ******************************************************************/
0118: public MalformedURIException(String p_msg) {
0119: super (p_msg);
0120: }
0121: }
0122:
0123: /** reserved characters */
0124: private static final String RESERVED_CHARACTERS = ";/?:@&=+$,";
0125:
0126: /** URI punctuation mark characters - these, combined with
0127: alphanumerics, constitute the "unreserved" characters */
0128: private static final String MARK_CHARACTERS = "-_.!~*'() ";
0129:
0130: /** scheme can be composed of alphanumerics and these characters */
0131: private static final String SCHEME_CHARACTERS = "+-.";
0132:
0133: /** userinfo can be composed of unreserved, escaped and these
0134: characters */
0135: private static final String USERINFO_CHARACTERS = ";:&=+$,";
0136:
0137: /** Stores the scheme (usually the protocol) for this URI. */
0138: private String m_scheme = null;
0139:
0140: /** If specified, stores the userinfo for this URI; otherwise null */
0141: private String m_userinfo = null;
0142:
0143: /** If specified, stores the host for this URI; otherwise null */
0144: private String m_host = null;
0145:
0146: /** If specified, stores the port for this URI; otherwise -1 */
0147: private int m_port = -1;
0148:
0149: /** If specified, stores the path for this URI; otherwise null */
0150: private String m_path = null;
0151:
0152: /** If specified, stores the query string for this URI; otherwise
0153: null. */
0154: private String m_queryString = null;
0155:
0156: /** If specified, stores the fragment for this URI; otherwise null */
0157: private String m_fragment = null;
0158:
0159: private static boolean DEBUG = false;
0160:
0161: /**
0162: * Construct a new and uninitialized URI.
0163: */
0164: public URI() {
0165: }
0166:
0167: /**
0168: * Construct a new URI from another URI. All fields for this URI are
0169: * set equal to the fields of the URI passed in.
0170: *
0171: * @param p_other the URI to copy (cannot be null)
0172: */
0173: public URI(URI p_other) {
0174: initialize(p_other);
0175: }
0176:
0177: /**
0178: * Construct a new URI from a URI specification string. If the
0179: * specification follows the "generic URI" syntax, (two slashes
0180: * following the first colon), the specification will be parsed
0181: * accordingly - setting the scheme, userinfo, host,port, path, query
0182: * string and fragment fields as necessary. If the specification does
0183: * not follow the "generic URI" syntax, the specification is parsed
0184: * into a scheme and scheme-specific part (stored as the path) only.
0185: *
0186: * @param p_uriSpec the URI specification string (cannot be null or
0187: * empty)
0188: *
0189: * @exception MalformedURIException if p_uriSpec violates any syntax
0190: * rules
0191: */
0192: public URI(String p_uriSpec) throws MalformedURIException {
0193: this ((URI) null, p_uriSpec);
0194: }
0195:
0196: /**
0197: * Construct a new URI from a base URI and a URI specification string.
0198: * The URI specification string may be a relative URI.
0199: *
0200: * @param p_base the base URI (cannot be null if p_uriSpec is null or
0201: * empty)
0202: * @param p_uriSpec the URI specification string (cannot be null or
0203: * empty if p_base is null)
0204: *
0205: * @exception MalformedURIException if p_uriSpec violates any syntax
0206: * rules
0207: */
0208: public URI(URI p_base, String p_uriSpec)
0209: throws MalformedURIException {
0210: initialize(p_base, p_uriSpec);
0211: }
0212:
0213: /**
0214: * Construct a new URI that does not follow the generic URI syntax.
0215: * Only the scheme and scheme-specific part (stored as the path) are
0216: * initialized.
0217: *
0218: * @param p_scheme the URI scheme (cannot be null or empty)
0219: * @param p_schemeSpecificPart the scheme-specific part (cannot be
0220: * null or empty)
0221: *
0222: * @exception MalformedURIException if p_scheme violates any
0223: * syntax rules
0224: */
0225: public URI(String p_scheme, String p_schemeSpecificPart)
0226: throws MalformedURIException {
0227: if (p_scheme == null || p_scheme.trim().length() == 0) {
0228: throw new MalformedURIException(
0229: "Cannot construct URI with null/empty scheme!");
0230: }
0231: if (p_schemeSpecificPart == null
0232: || p_schemeSpecificPart.trim().length() == 0) {
0233: throw new MalformedURIException(
0234: "Cannot construct URI with null/empty scheme-specific part!");
0235: }
0236: setScheme(p_scheme);
0237: setPath(p_schemeSpecificPart);
0238: }
0239:
0240: /**
0241: * Construct a new URI that follows the generic URI syntax from its
0242: * component parts. Each component is validated for syntax and some
0243: * basic semantic checks are performed as well. See the individual
0244: * setter methods for specifics.
0245: *
0246: * @param p_scheme the URI scheme (cannot be null or empty)
0247: * @param p_host the hostname or IPv4 address for the URI
0248: * @param p_path the URI path - if the path contains '?' or '#',
0249: * then the query string and/or fragment will be
0250: * set from the path; however, if the query and
0251: * fragment are specified both in the path and as
0252: * separate parameters, an exception is thrown
0253: * @param p_queryString the URI query string (cannot be specified
0254: * if path is null)
0255: * @param p_fragment the URI fragment (cannot be specified if path
0256: * is null)
0257: *
0258: * @exception MalformedURIException if any of the parameters violates
0259: * syntax rules or semantic rules
0260: */
0261: public URI(String p_scheme, String p_host, String p_path,
0262: String p_queryString, String p_fragment)
0263: throws MalformedURIException {
0264: this (p_scheme, null, p_host, -1, p_path, p_queryString,
0265: p_fragment);
0266: }
0267:
0268: /**
0269: * Construct a new URI that follows the generic URI syntax from its
0270: * component parts. Each component is validated for syntax and some
0271: * basic semantic checks are performed as well. See the individual
0272: * setter methods for specifics.
0273: *
0274: * @param p_scheme the URI scheme (cannot be null or empty)
0275: * @param p_userinfo the URI userinfo (cannot be specified if host
0276: * is null)
0277: * @param p_host the hostname or IPv4 address for the URI
0278: * @param p_port the URI port (may be -1 for "unspecified"; cannot
0279: * be specified if host is null)
0280: * @param p_path the URI path - if the path contains '?' or '#',
0281: * then the query string and/or fragment will be
0282: * set from the path; however, if the query and
0283: * fragment are specified both in the path and as
0284: * separate parameters, an exception is thrown
0285: * @param p_queryString the URI query string (cannot be specified
0286: * if path is null)
0287: * @param p_fragment the URI fragment (cannot be specified if path
0288: * is null)
0289: *
0290: * @exception MalformedURIException if any of the parameters violates
0291: * syntax rules or semantic rules
0292: */
0293: public URI(String p_scheme, String p_userinfo, String p_host,
0294: int p_port, String p_path, String p_queryString,
0295: String p_fragment) throws MalformedURIException {
0296: if (p_scheme == null || p_scheme.trim().length() == 0) {
0297: throw new MalformedURIException("Scheme is required!");
0298: }
0299:
0300: if (p_host == null) {
0301: if (p_userinfo != null) {
0302: throw new MalformedURIException(
0303: "Userinfo may not be specified if host is not specified!");
0304: }
0305: if (p_port != -1) {
0306: throw new MalformedURIException(
0307: "Port may not be specified if host is not specified!");
0308: }
0309: }
0310:
0311: if (p_path != null) {
0312: if (p_path.indexOf('?') != -1 && p_queryString != null) {
0313: throw new MalformedURIException(
0314: "Query string cannot be specified in path and query string!");
0315: }
0316:
0317: if (p_path.indexOf('#') != -1 && p_fragment != null) {
0318: throw new MalformedURIException(
0319: "Fragment cannot be specified in both the path and fragment!");
0320: }
0321: }
0322:
0323: setScheme(p_scheme);
0324: setHost(p_host);
0325: setPort(p_port);
0326: setUserinfo(p_userinfo);
0327: setPath(p_path);
0328: setQueryString(p_queryString);
0329: setFragment(p_fragment);
0330: }
0331:
0332: /**
0333: * Initialize all fields of this URI from another URI.
0334: *
0335: * @param p_other the URI to copy (cannot be null)
0336: */
0337: private void initialize(URI p_other) {
0338: m_scheme = p_other.getScheme();
0339: m_userinfo = p_other.getUserinfo();
0340: m_host = p_other.getHost();
0341: m_port = p_other.getPort();
0342: m_path = p_other.getPath();
0343: m_queryString = p_other.getQueryString();
0344: m_fragment = p_other.getFragment();
0345: }
0346:
0347: /**
0348: * Initializes this URI from a base URI and a URI specification string.
0349: * See RFC 2396 Section 4 and Appendix B for specifications on parsing
0350: * the URI and Section 5 for specifications on resolving relative URIs
0351: * and relative paths.
0352: *
0353: * @param p_base the base URI (may be null if p_uriSpec is an absolute
0354: * URI)
0355: * @param p_uriSpec the URI spec string which may be an absolute or
0356: * relative URI (can only be null/empty if p_base
0357: * is not null)
0358: *
0359: * @exception MalformedURIException if p_base is null and p_uriSpec
0360: * is not an absolute URI or if
0361: * p_uriSpec violates syntax rules
0362: */
0363: private void initialize(URI p_base, String p_uriSpec)
0364: throws MalformedURIException {
0365: if (p_base == null
0366: && (p_uriSpec == null || p_uriSpec.trim().length() == 0)) {
0367: throw new MalformedURIException(
0368: "Cannot initialize URI with empty parameters.");
0369: }
0370:
0371: // just make a copy of the base if spec is empty
0372: if (p_uriSpec == null || p_uriSpec.trim().length() == 0) {
0373: initialize(p_base);
0374: return;
0375: }
0376:
0377: String uriSpec = p_uriSpec.trim();
0378: int uriSpecLen = uriSpec.length();
0379: int index = 0;
0380:
0381: // Check for scheme, which must be before '/', '?' or '#'. Also handle
0382: // names with DOS drive letters ('D:'), so 1-character schemes are not
0383: // allowed.
0384: int colonIdx = uriSpec.indexOf(':');
0385: int slashIdx = uriSpec.indexOf('/');
0386: int queryIdx = uriSpec.indexOf('?');
0387: int fragmentIdx = uriSpec.indexOf('#');
0388:
0389: if ((colonIdx < 2) || (colonIdx > slashIdx && slashIdx != -1)
0390: || (colonIdx > queryIdx && queryIdx != -1)
0391: || (colonIdx > fragmentIdx && fragmentIdx != -1)) {
0392: // A standalone base is a valid URI according to spec
0393: if (p_base == null && fragmentIdx != 0) {
0394: throw new MalformedURIException(
0395: "No scheme found in URI.");
0396: }
0397: } else {
0398: initializeScheme(uriSpec);
0399: index = m_scheme.length() + 1;
0400: }
0401:
0402: // two slashes means generic URI syntax, so we get the authority
0403: if (((index + 1) < uriSpecLen)
0404: && (uriSpec.substring(index).startsWith("//"))) {
0405: index += 2;
0406: int startPos = index;
0407:
0408: // get authority - everything up to path, query or fragment
0409: char testChar = '\0';
0410: while (index < uriSpecLen) {
0411: testChar = uriSpec.charAt(index);
0412: if (testChar == '/' || testChar == '?'
0413: || testChar == '#') {
0414: break;
0415: }
0416: index++;
0417: }
0418:
0419: // if we found authority, parse it out, otherwise we set the
0420: // host to empty string
0421: if (index > startPos) {
0422: initializeAuthority(uriSpec.substring(startPos, index));
0423: } else {
0424: m_host = "";
0425: }
0426: }
0427:
0428: initializePath(uriSpec.substring(index));
0429:
0430: // Resolve relative URI to base URI - see RFC 2396 Section 5.2
0431: // In some cases, it might make more sense to throw an exception
0432: // (when scheme is specified is the string spec and the base URI
0433: // is also specified, for example), but we're just following the
0434: // RFC specifications
0435: if (p_base != null) {
0436:
0437: // check to see if this is the current doc - RFC 2396 5.2 #2
0438: // note that this is slightly different from the RFC spec in that
0439: // we don't include the check for query string being null
0440: // - this handles cases where the urispec is just a query
0441: // string or a fragment (e.g. "?y" or "#s") -
0442: // see <http://www.ics.uci.edu/~fielding/url/test1.html> which
0443: // identified this as a bug in the RFC
0444: if (m_path.length() == 0 && m_scheme == null
0445: && m_host == null) {
0446: m_scheme = p_base.getScheme();
0447: m_userinfo = p_base.getUserinfo();
0448: m_host = p_base.getHost();
0449: m_port = p_base.getPort();
0450: m_path = p_base.getPath();
0451:
0452: if (m_queryString == null) {
0453: m_queryString = p_base.getQueryString();
0454: }
0455: return;
0456: }
0457:
0458: // check for scheme - RFC 2396 5.2 #3
0459: // if we found a scheme, it means absolute URI, so we're done
0460: if (m_scheme == null) {
0461: m_scheme = p_base.getScheme();
0462: } else {
0463: return;
0464: }
0465:
0466: // check for authority - RFC 2396 5.2 #4
0467: // if we found a host, then we've got a network path, so we're done
0468: if (m_host == null) {
0469: m_userinfo = p_base.getUserinfo();
0470: m_host = p_base.getHost();
0471: m_port = p_base.getPort();
0472: } else {
0473: return;
0474: }
0475:
0476: // check for absolute path - RFC 2396 5.2 #5
0477: if (m_path.length() > 0 && m_path.startsWith("/")) {
0478: return;
0479: }
0480:
0481: // if we get to this point, we need to resolve relative path
0482: // RFC 2396 5.2 #6
0483: String path = new String();
0484: String basePath = p_base.getPath();
0485:
0486: // 6a - get all but the last segment of the base URI path
0487: if (basePath != null) {
0488: int lastSlash = basePath.lastIndexOf('/');
0489: if (lastSlash != -1) {
0490: path = basePath.substring(0, lastSlash + 1);
0491: }
0492: }
0493:
0494: // 6b - append the relative URI path
0495: path = path.concat(m_path);
0496:
0497: // 6c - remove all "./" where "." is a complete path segment
0498: index = -1;
0499: while ((index = path.indexOf("/./")) != -1) {
0500: path = path.substring(0, index + 1).concat(
0501: path.substring(index + 3));
0502: }
0503:
0504: // 6d - remove "." if path ends with "." as a complete path segment
0505: if (path.endsWith("/.")) {
0506: path = path.substring(0, path.length() - 1);
0507: }
0508:
0509: // 6e - remove all "<segment>/../" where "<segment>" is a complete
0510: // path segment not equal to ".."
0511: index = 1;
0512: int segIndex = -1;
0513:
0514: while ((index = path.indexOf("/../", index)) > 0) {
0515: segIndex = path.lastIndexOf('/', index - 1);
0516: if (segIndex != -1
0517: && !path.substring(segIndex + 1, index).equals(
0518: "..")) {
0519: path = path.substring(0, segIndex).concat(
0520: path.substring(index + 3));
0521: index = segIndex;
0522: } else {
0523: index += 4;
0524: }
0525: }
0526:
0527: // 6f - remove ending "<segment>/.." where "<segment>" is a
0528: // complete path segment
0529: if (path.endsWith("/..")) {
0530: index = path.length() - 3;
0531: segIndex = path.lastIndexOf('/', index - 1);
0532: if (segIndex != -1
0533: && !path.substring(segIndex + 1, index).equals(
0534: "..")) {
0535: path = path.substring(0, segIndex + 1);
0536: }
0537: }
0538:
0539: m_path = path;
0540: }
0541: }
0542:
0543: /**
0544: * Initialize the scheme for this URI from a URI string spec.
0545: *
0546: * @param p_uriSpec the URI specification (cannot be null)
0547: *
0548: * @exception MalformedURIException if URI does not have a conformant
0549: * scheme
0550: */
0551: private void initializeScheme(String p_uriSpec)
0552: throws MalformedURIException {
0553: int uriSpecLen = p_uriSpec.length();
0554: int index = 0;
0555: String scheme = null;
0556: char testChar = '\0';
0557:
0558: while (index < uriSpecLen) {
0559: testChar = p_uriSpec.charAt(index);
0560: if (testChar == ':' || testChar == '/' || testChar == '?'
0561: || testChar == '#') {
0562: break;
0563: }
0564: index++;
0565: }
0566: scheme = p_uriSpec.substring(0, index);
0567:
0568: if (scheme.length() == 0) {
0569: throw new MalformedURIException("No scheme found in URI.");
0570: } else {
0571: setScheme(scheme);
0572: }
0573: }
0574:
0575: /**
0576: * Initialize the authority (userinfo, host and port) for this
0577: * URI from a URI string spec.
0578: *
0579: * @param p_uriSpec the URI specification (cannot be null)
0580: *
0581: * @exception MalformedURIException if p_uriSpec violates syntax rules
0582: */
0583: private void initializeAuthority(String p_uriSpec)
0584: throws MalformedURIException {
0585: int index = 0;
0586: int start = 0;
0587: int end = p_uriSpec.length();
0588: char testChar = '\0';
0589: String userinfo = null;
0590:
0591: // userinfo is everything up @
0592: if (p_uriSpec.indexOf('@', start) != -1) {
0593: while (index < end) {
0594: testChar = p_uriSpec.charAt(index);
0595: if (testChar == '@') {
0596: break;
0597: }
0598: index++;
0599: }
0600: userinfo = p_uriSpec.substring(start, index);
0601: index++;
0602: }
0603:
0604: // host is everything up to ':'
0605: String host = null;
0606: start = index;
0607: while (index < end) {
0608: testChar = p_uriSpec.charAt(index);
0609: if (testChar == ':') {
0610: break;
0611: }
0612: index++;
0613: }
0614: host = p_uriSpec.substring(start, index);
0615: int port = -1;
0616: if (host.length() > 0) {
0617: // port
0618: if (testChar == ':') {
0619: index++;
0620: start = index;
0621: while (index < end) {
0622: index++;
0623: }
0624: String portStr = p_uriSpec.substring(start, index);
0625: if (portStr.length() > 0) {
0626: for (int i = 0; i < portStr.length(); i++) {
0627: if (!isDigit(portStr.charAt(i))) {
0628: throw new MalformedURIException(
0629: portStr
0630: + " is invalid. Port should only contain digits!");
0631: }
0632: }
0633: try {
0634: port = Integer.parseInt(portStr);
0635: } catch (NumberFormatException nfe) {
0636: // can't happen
0637: }
0638: }
0639: }
0640: }
0641: setHost(host);
0642: setPort(port);
0643: setUserinfo(userinfo);
0644: }
0645:
0646: /**
0647: * Initialize the path for this URI from a URI string spec.
0648: *
0649: * @param p_uriSpec the URI specification (cannot be null)
0650: *
0651: * @exception MalformedURIException if p_uriSpec violates syntax rules
0652: */
0653: private void initializePath(String p_uriSpec)
0654: throws MalformedURIException {
0655: if (p_uriSpec == null) {
0656: throw new MalformedURIException(
0657: "Cannot initialize path from null string!");
0658: }
0659:
0660: int index = 0;
0661: int start = 0;
0662: int end = p_uriSpec.length();
0663: char testChar = '\0';
0664:
0665: // path - everything up to query string or fragment
0666: while (index < end) {
0667: testChar = p_uriSpec.charAt(index);
0668: if (testChar == '?' || testChar == '#') {
0669: break;
0670: }
0671: // check for valid escape sequence
0672: if (testChar == '%') {
0673: if (index + 2 >= end
0674: || !isHex(p_uriSpec.charAt(index + 1))
0675: || !isHex(p_uriSpec.charAt(index + 2))) {
0676: throw new MalformedURIException(
0677: "Path contains invalid escape sequence!");
0678: }
0679: } else if (!isReservedCharacter(testChar)
0680: && !isUnreservedCharacter(testChar)) {
0681: throw new MalformedURIException(
0682: "Path contains invalid character: " + testChar);
0683: }
0684: index++;
0685: }
0686: m_path = p_uriSpec.substring(start, index);
0687:
0688: // query - starts with ? and up to fragment or end
0689: if (testChar == '?') {
0690: index++;
0691: start = index;
0692: while (index < end) {
0693: testChar = p_uriSpec.charAt(index);
0694: if (testChar == '#') {
0695: break;
0696: }
0697: if (testChar == '%') {
0698: if (index + 2 >= end
0699: || !isHex(p_uriSpec.charAt(index + 1))
0700: || !isHex(p_uriSpec.charAt(index + 2))) {
0701: throw new MalformedURIException(
0702: "Query string contains invalid escape sequence!");
0703: }
0704: } else if (!isReservedCharacter(testChar)
0705: && !isUnreservedCharacter(testChar)) {
0706: throw new MalformedURIException(
0707: "Query string contains invalid character:"
0708: + testChar);
0709: }
0710: index++;
0711: }
0712: m_queryString = p_uriSpec.substring(start, index);
0713: }
0714:
0715: // fragment - starts with #
0716: if (testChar == '#') {
0717: index++;
0718: start = index;
0719: while (index < end) {
0720: testChar = p_uriSpec.charAt(index);
0721:
0722: if (testChar == '%') {
0723: if (index + 2 >= end
0724: || !isHex(p_uriSpec.charAt(index + 1))
0725: || !isHex(p_uriSpec.charAt(index + 2))) {
0726: throw new MalformedURIException(
0727: "Fragment contains invalid escape sequence!");
0728: }
0729: } else if (!isReservedCharacter(testChar)
0730: && !isUnreservedCharacter(testChar)) {
0731: throw new MalformedURIException(
0732: "Fragment contains invalid character:"
0733: + testChar);
0734: }
0735: index++;
0736: }
0737: m_fragment = p_uriSpec.substring(start, index);
0738: }
0739: }
0740:
0741: /**
0742: * Get the scheme for this URI.
0743: *
0744: * @return the scheme for this URI
0745: */
0746: public String getScheme() {
0747: return m_scheme;
0748: }
0749:
0750: /**
0751: * Get the scheme-specific part for this URI (everything following the
0752: * scheme and the first colon). See RFC 2396 Section 5.2 for spec.
0753: *
0754: * @return the scheme-specific part for this URI
0755: */
0756: public String getSchemeSpecificPart() {
0757: StringBuffer schemespec = new StringBuffer();
0758:
0759: if (m_userinfo != null || m_host != null || m_port != -1) {
0760: schemespec.append("//");
0761: }
0762:
0763: if (m_userinfo != null) {
0764: schemespec.append(m_userinfo);
0765: schemespec.append('@');
0766: }
0767:
0768: if (m_host != null) {
0769: schemespec.append(m_host);
0770: }
0771:
0772: if (m_port != -1) {
0773: schemespec.append(':');
0774: schemespec.append(m_port);
0775: }
0776:
0777: if (m_path != null) {
0778: schemespec.append((m_path));
0779: }
0780:
0781: if (m_queryString != null) {
0782: schemespec.append('?');
0783: schemespec.append(m_queryString);
0784: }
0785:
0786: if (m_fragment != null) {
0787: schemespec.append('#');
0788: schemespec.append(m_fragment);
0789: }
0790:
0791: return schemespec.toString();
0792: }
0793:
0794: /**
0795: * Get the userinfo for this URI.
0796: *
0797: * @return the userinfo for this URI (null if not specified).
0798: */
0799: public String getUserinfo() {
0800: return m_userinfo;
0801: }
0802:
0803: /**
0804: * Get the host for this URI.
0805: *
0806: * @return the host for this URI (null if not specified).
0807: */
0808: public String getHost() {
0809: return m_host;
0810: }
0811:
0812: /**
0813: * Get the port for this URI.
0814: *
0815: * @return the port for this URI (-1 if not specified).
0816: */
0817: public int getPort() {
0818: return m_port;
0819: }
0820:
0821: /**
0822: * Get the path for this URI (optionally with the query string and
0823: * fragment).
0824: *
0825: * @param p_includeQueryString if true (and query string is not null),
0826: * then a "?" followed by the query string
0827: * will be appended
0828: * @param p_includeFragment if true (and fragment is not null),
0829: * then a "#" followed by the fragment
0830: * will be appended
0831: *
0832: * @return the path for this URI possibly including the query string
0833: * and fragment
0834: */
0835: public String getPath(boolean p_includeQueryString,
0836: boolean p_includeFragment) {
0837: StringBuffer pathString = new StringBuffer(m_path);
0838:
0839: if (p_includeQueryString && m_queryString != null) {
0840: pathString.append('?');
0841: pathString.append(m_queryString);
0842: }
0843:
0844: if (p_includeFragment && m_fragment != null) {
0845: pathString.append('#');
0846: pathString.append(m_fragment);
0847: }
0848: return pathString.toString();
0849: }
0850:
0851: /**
0852: * Get the path for this URI. Note that the value returned is the path
0853: * only and does not include the query string or fragment.
0854: *
0855: * @return the path for this URI.
0856: */
0857: public String getPath() {
0858: return m_path;
0859: }
0860:
0861: /**
0862: * Get the query string for this URI.
0863: *
0864: * @return the query string for this URI. Null is returned if there
0865: * was no "?" in the URI spec, empty string if there was a
0866: * "?" but no query string following it.
0867: */
0868: public String getQueryString() {
0869: return m_queryString;
0870: }
0871:
0872: /**
0873: * Get the fragment for this URI.
0874: *
0875: * @return the fragment for this URI. Null is returned if there
0876: * was no "#" in the URI spec, empty string if there was a
0877: * "#" but no fragment following it.
0878: */
0879: public String getFragment() {
0880: return m_fragment;
0881: }
0882:
0883: /**
0884: * Set the scheme for this URI. The scheme is converted to lowercase
0885: * before it is set.
0886: *
0887: * @param p_scheme the scheme for this URI (cannot be null)
0888: *
0889: * @exception MalformedURIException if p_scheme is not a conformant
0890: * scheme name
0891: */
0892: public void setScheme(String p_scheme) throws MalformedURIException {
0893: if (p_scheme == null) {
0894: throw new MalformedURIException(
0895: "Cannot set scheme from null string!");
0896: }
0897: if (!isConformantSchemeName(p_scheme)) {
0898: throw new MalformedURIException(
0899: "The scheme is not conformant.");
0900: }
0901:
0902: m_scheme = p_scheme.toLowerCase();
0903: }
0904:
0905: /**
0906: * Set the userinfo for this URI. If a non-null value is passed in and
0907: * the host value is null, then an exception is thrown.
0908: *
0909: * @param p_userinfo the userinfo for this URI
0910: *
0911: * @exception MalformedURIException if p_userinfo contains invalid
0912: * characters
0913: */
0914: public void setUserinfo(String p_userinfo)
0915: throws MalformedURIException {
0916: if (p_userinfo == null) {
0917: m_userinfo = null;
0918: } else {
0919: if (m_host == null) {
0920: throw new MalformedURIException(
0921: "Userinfo cannot be set when host is null!");
0922: }
0923:
0924: // userinfo can contain alphanumerics, mark characters, escaped
0925: // and ';',':','&','=','+','$',','
0926: int index = 0;
0927: int end = p_userinfo.length();
0928: char testChar = '\0';
0929: while (index < end) {
0930: testChar = p_userinfo.charAt(index);
0931: if (testChar == '%') {
0932: if (index + 2 >= end
0933: || !isHex(p_userinfo.charAt(index + 1))
0934: || !isHex(p_userinfo.charAt(index + 2))) {
0935: throw new MalformedURIException(
0936: "Userinfo contains invalid escape sequence!");
0937: }
0938: } else if (!isUnreservedCharacter(testChar)
0939: && USERINFO_CHARACTERS.indexOf(testChar) == -1) {
0940: throw new MalformedURIException(
0941: "Userinfo contains invalid character:"
0942: + testChar);
0943: }
0944: index++;
0945: }
0946: }
0947: m_userinfo = p_userinfo;
0948: }
0949:
0950: /**
0951: * Set the host for this URI. If null is passed in, the userinfo
0952: * field is also set to null and the port is set to -1.
0953: *
0954: * @param p_host the host for this URI
0955: *
0956: * @exception MalformedURIException if p_host is not a valid IP
0957: * address or DNS hostname.
0958: */
0959: public void setHost(String p_host) throws MalformedURIException {
0960: if (p_host == null || p_host.trim().length() == 0) {
0961: m_host = p_host;
0962: m_userinfo = null;
0963: m_port = -1;
0964: } else if (!isWellFormedAddress(p_host)) {
0965: throw new MalformedURIException(
0966: "Host is not a well formed address!");
0967: }
0968: m_host = p_host;
0969: }
0970:
0971: /**
0972: * Set the port for this URI. -1 is used to indicate that the port is
0973: * not specified, otherwise valid port numbers are between 0 and 65535.
0974: * If a valid port number is passed in and the host field is null,
0975: * an exception is thrown.
0976: *
0977: * @param p_port the port number for this URI
0978: *
0979: * @exception MalformedURIException if p_port is not -1 and not a
0980: * valid port number
0981: */
0982: public void setPort(int p_port) throws MalformedURIException {
0983: if (p_port >= 0 && p_port <= 65535) {
0984: if (m_host == null) {
0985: throw new MalformedURIException(
0986: "Port cannot be set when host is null!");
0987: }
0988: } else if (p_port != -1) {
0989: throw new MalformedURIException("Invalid port number!");
0990: }
0991: m_port = p_port;
0992: }
0993:
0994: /**
0995: * Set the path for this URI. If the supplied path is null, then the
0996: * query string and fragment are set to null as well. If the supplied
0997: * path includes a query string and/or fragment, these fields will be
0998: * parsed and set as well. Note that, for URIs following the "generic
0999: * URI" syntax, the path specified should start with a slash.
1000: * For URIs that do not follow the generic URI syntax, this method
1001: * sets the scheme-specific part.
1002: *
1003: * @param p_path the path for this URI (may be null)
1004: *
1005: * @exception MalformedURIException if p_path contains invalid
1006: * characters
1007: */
1008: public void setPath(String p_path) throws MalformedURIException {
1009: if (p_path == null) {
1010: m_path = null;
1011: m_queryString = null;
1012: m_fragment = null;
1013: } else {
1014: initializePath(p_path);
1015: }
1016: }
1017:
1018: /**
1019: * Append to the end of the path of this URI. If the current path does
1020: * not end in a slash and the path to be appended does not begin with
1021: * a slash, a slash will be appended to the current path before the
1022: * new segment is added. Also, if the current path ends in a slash
1023: * and the new segment begins with a slash, the extra slash will be
1024: * removed before the new segment is appended.
1025: *
1026: * @param p_addToPath the new segment to be added to the current path
1027: *
1028: * @exception MalformedURIException if p_addToPath contains syntax
1029: * errors
1030: */
1031: public void appendPath(String p_addToPath)
1032: throws MalformedURIException {
1033: if (p_addToPath == null || p_addToPath.trim().length() == 0) {
1034: return;
1035: }
1036:
1037: if (!isURIString(p_addToPath)) {
1038: throw new MalformedURIException(
1039: "Path contains invalid character!");
1040: }
1041:
1042: if (m_path == null || m_path.trim().length() == 0) {
1043: if (p_addToPath.startsWith("/")) {
1044: m_path = p_addToPath;
1045: } else {
1046: m_path = "/" + p_addToPath;
1047: }
1048: } else if (m_path.endsWith("/")) {
1049: if (p_addToPath.startsWith("/")) {
1050: m_path = m_path.concat(p_addToPath.substring(1));
1051: } else {
1052: m_path = m_path.concat(p_addToPath);
1053: }
1054: } else {
1055: if (p_addToPath.startsWith("/")) {
1056: m_path = m_path.concat(p_addToPath);
1057: } else {
1058: m_path = m_path.concat("/" + p_addToPath);
1059: }
1060: }
1061: }
1062:
1063: /**
1064: * Set the query string for this URI. A non-null value is valid only
1065: * if this is an URI conforming to the generic URI syntax and
1066: * the path value is not null.
1067: *
1068: * @param p_queryString the query string for this URI
1069: *
1070: * @exception MalformedURIException if p_queryString is not null and this
1071: * URI does not conform to the generic
1072: * URI syntax or if the path is null
1073: */
1074: public void setQueryString(String p_queryString)
1075: throws MalformedURIException {
1076: if (p_queryString == null) {
1077: m_queryString = null;
1078: } else if (!isGenericURI()) {
1079: throw new MalformedURIException(
1080: "Query string can only be set for a generic URI!");
1081: } else if (getPath() == null) {
1082: throw new MalformedURIException(
1083: "Query string cannot be set when path is null!");
1084: } else if (!isURIString(p_queryString)) {
1085: throw new MalformedURIException(
1086: "Query string contains invalid character!");
1087: } else {
1088: m_queryString = p_queryString;
1089: }
1090: }
1091:
1092: /**
1093: * Set the fragment for this URI. A non-null value is valid only
1094: * if this is a URI conforming to the generic URI syntax and
1095: * the path value is not null.
1096: *
1097: * @param p_fragment the fragment for this URI
1098: *
1099: * @exception MalformedURIException if p_fragment is not null and this
1100: * URI does not conform to the generic
1101: * URI syntax or if the path is null
1102: */
1103: public void setFragment(String p_fragment)
1104: throws MalformedURIException {
1105: if (p_fragment == null) {
1106: m_fragment = null;
1107: } else if (!isGenericURI()) {
1108: throw new MalformedURIException(
1109: "Fragment can only be set for a generic URI!");
1110: } else if (getPath() == null) {
1111: throw new MalformedURIException(
1112: "Fragment cannot be set when path is null!");
1113: } else if (!isURIString(p_fragment)) {
1114: throw new MalformedURIException(
1115: "Fragment contains invalid character!");
1116: } else {
1117: m_fragment = p_fragment;
1118: }
1119: }
1120:
1121: /**
1122: * Determines if the passed-in Object is equivalent to this URI.
1123: *
1124: * @param p_test the Object to test for equality.
1125: *
1126: * @return true if p_test is a URI with all values equal to this
1127: * URI, false otherwise
1128: */
1129: public boolean equals(Object p_test) {
1130: if (p_test instanceof URI) {
1131: URI testURI = (URI) p_test;
1132: if (((m_scheme == null && testURI.m_scheme == null) || (m_scheme != null
1133: && testURI.m_scheme != null && m_scheme
1134: .equals(testURI.m_scheme)))
1135: && ((m_userinfo == null && testURI.m_userinfo == null) || (m_userinfo != null
1136: && testURI.m_userinfo != null && m_userinfo
1137: .equals(testURI.m_userinfo)))
1138: && ((m_host == null && testURI.m_host == null) || (m_host != null
1139: && testURI.m_host != null && m_host
1140: .equals(testURI.m_host)))
1141: && m_port == testURI.m_port
1142: && ((m_path == null && testURI.m_path == null) || (m_path != null
1143: && testURI.m_path != null && m_path
1144: .equals(testURI.m_path)))
1145: && ((m_queryString == null && testURI.m_queryString == null) || (m_queryString != null
1146: && testURI.m_queryString != null && m_queryString
1147: .equals(testURI.m_queryString)))
1148: && ((m_fragment == null && testURI.m_fragment == null) || (m_fragment != null
1149: && testURI.m_fragment != null && m_fragment
1150: .equals(testURI.m_fragment)))) {
1151: return true;
1152: }
1153: }
1154: return false;
1155: }
1156:
1157: /**
1158: * Get the URI as a string specification. See RFC 2396 Section 5.2.
1159: *
1160: * @return the URI string specification
1161: */
1162: public String toString() {
1163: StringBuffer uriSpecString = new StringBuffer();
1164:
1165: if (m_scheme != null) {
1166: uriSpecString.append(m_scheme);
1167: uriSpecString.append(':');
1168: }
1169: uriSpecString.append(getSchemeSpecificPart());
1170: return uriSpecString.toString();
1171: }
1172:
1173: /**
1174: * Get the indicator as to whether this URI uses the "generic URI"
1175: * syntax.
1176: *
1177: * @return true if this URI uses the "generic URI" syntax, false
1178: * otherwise
1179: */
1180: public boolean isGenericURI() {
1181: // presence of the host (whether valid or empty) means
1182: // double-slashes which means generic uri
1183: return (m_host != null);
1184: }
1185:
1186: /**
1187: * Determine whether a scheme conforms to the rules for a scheme name.
1188: * A scheme is conformant if it starts with an alphanumeric, and
1189: * contains only alphanumerics, '+','-' and '.'.
1190: *
1191: * @return true if the scheme is conformant, false otherwise
1192: */
1193: public static boolean isConformantSchemeName(String p_scheme) {
1194: if (p_scheme == null || p_scheme.trim().length() == 0) {
1195: return false;
1196: }
1197:
1198: if (!isAlpha(p_scheme.charAt(0))) {
1199: return false;
1200: }
1201:
1202: char testChar;
1203: for (int i = 1; i < p_scheme.length(); i++) {
1204: testChar = p_scheme.charAt(i);
1205: if (!isAlphanum(testChar)
1206: && SCHEME_CHARACTERS.indexOf(testChar) == -1) {
1207: return false;
1208: }
1209: }
1210:
1211: return true;
1212: }
1213:
1214: /**
1215: * Determine whether a string is syntactically capable of representing
1216: * a valid IPv4 address or the domain name of a network host. A valid
1217: * IPv4 address consists of four decimal digit groups separated by a
1218: * '.'. A hostname consists of domain labels (each of which must
1219: * begin and end with an alphanumeric but may contain '-') separated
1220: & by a '.'. See RFC 2396 Section 3.2.2.
1221: *
1222: * @return true if the string is a syntactically valid IPv4 address
1223: * or hostname
1224: */
1225: public static boolean isWellFormedAddress(String p_address) {
1226: if (p_address == null) {
1227: return false;
1228: }
1229:
1230: String address = p_address.trim();
1231: int addrLength = address.length();
1232: if (addrLength == 0 || addrLength > 255) {
1233: return false;
1234: }
1235:
1236: if (address.startsWith(".") || address.startsWith("-")) {
1237: return false;
1238: }
1239:
1240: // rightmost domain label starting with digit indicates IP address
1241: // since top level domain label can only start with an alpha
1242: // see RFC 2396 Section 3.2.2
1243: int index = address.lastIndexOf('.');
1244: if (address.endsWith(".")) {
1245: index = address.substring(0, index).lastIndexOf('.');
1246: }
1247:
1248: if (index + 1 < addrLength
1249: && isDigit(p_address.charAt(index + 1))) {
1250: char testChar;
1251: int numDots = 0;
1252:
1253: // make sure that 1) we see only digits and dot separators, 2) that
1254: // any dot separator is preceded and followed by a digit and
1255: // 3) that we find 3 dots
1256: for (int i = 0; i < addrLength; i++) {
1257: testChar = address.charAt(i);
1258: if (testChar == '.') {
1259: if (!isDigit(address.charAt(i - 1))
1260: || (i + 1 < addrLength && !isDigit(address
1261: .charAt(i + 1)))) {
1262: return false;
1263: }
1264: numDots++;
1265: } else if (!isDigit(testChar)) {
1266: return false;
1267: }
1268: }
1269: if (numDots != 3) {
1270: return false;
1271: }
1272: } else {
1273: // domain labels can contain alphanumerics and '-"
1274: // but must start and end with an alphanumeric
1275: char testChar;
1276:
1277: for (int i = 0; i < addrLength; i++) {
1278: testChar = address.charAt(i);
1279: if (testChar == '.') {
1280: if (!isAlphanum(address.charAt(i - 1))) {
1281: return false;
1282: }
1283: if (i + 1 < addrLength
1284: && !isAlphanum(address.charAt(i + 1))) {
1285: return false;
1286: }
1287: } else if (!isAlphanum(testChar) && testChar != '-') {
1288: return false;
1289: }
1290: }
1291: }
1292: return true;
1293: }
1294:
1295: /**
1296: * Determine whether a char is a digit.
1297: *
1298: * @return true if the char is betweeen '0' and '9', false otherwise
1299: */
1300: private static boolean isDigit(char p_char) {
1301: return p_char >= '0' && p_char <= '9';
1302: }
1303:
1304: /**
1305: * Determine whether a character is a hexadecimal character.
1306: *
1307: * @return true if the char is betweeen '0' and '9', 'a' and 'f'
1308: * or 'A' and 'F', false otherwise
1309: */
1310: private static boolean isHex(char p_char) {
1311: return (isDigit(p_char) || (p_char >= 'a' && p_char <= 'f') || (p_char >= 'A' && p_char <= 'F'));
1312: }
1313:
1314: /**
1315: * Determine whether a char is an alphabetic character: a-z or A-Z
1316: *
1317: * @return true if the char is alphabetic, false otherwise
1318: */
1319: private static boolean isAlpha(char p_char) {
1320: return ((p_char >= 'a' && p_char <= 'z') || (p_char >= 'A' && p_char <= 'Z'));
1321: }
1322:
1323: /**
1324: * Determine whether a char is an alphanumeric: 0-9, a-z or A-Z
1325: *
1326: * @return true if the char is alphanumeric, false otherwise
1327: */
1328: private static boolean isAlphanum(char p_char) {
1329: return (isAlpha(p_char) || isDigit(p_char));
1330: }
1331:
1332: /**
1333: * Determine whether a character is a reserved character:
1334: * ';', '/', '?', ':', '@', '&', '=', '+', '$' or ','
1335: *
1336: * @return true if the string contains any reserved characters
1337: */
1338: private static boolean isReservedCharacter(char p_char) {
1339: return RESERVED_CHARACTERS.indexOf(p_char) != -1;
1340: }
1341:
1342: /**
1343: * Determine whether a char is an unreserved character.
1344: *
1345: * @return true if the char is unreserved, false otherwise
1346: */
1347: private static boolean isUnreservedCharacter(char p_char) {
1348: return (isAlphanum(p_char) || MARK_CHARACTERS.indexOf(p_char) != -1);
1349: }
1350:
1351: /**
1352: * Determine whether a given string contains only URI characters (also
1353: * called "uric" in RFC 2396). uric consist of all reserved
1354: * characters, unreserved characters and escaped characters.
1355: *
1356: * @return true if the string is comprised of uric, false otherwise
1357: */
1358: private static boolean isURIString(String p_uric) {
1359: if (p_uric == null) {
1360: return false;
1361: }
1362: int end = p_uric.length();
1363: char testChar = '\0';
1364: for (int i = 0; i < end; i++) {
1365: testChar = p_uric.charAt(i);
1366: if (testChar == '%') {
1367: if (i + 2 >= end || !isHex(p_uric.charAt(i + 1))
1368: || !isHex(p_uric.charAt(i + 2))) {
1369: return false;
1370: } else {
1371: i += 2;
1372: continue;
1373: }
1374: }
1375: if (isReservedCharacter(testChar)
1376: || isUnreservedCharacter(testChar)) {
1377: continue;
1378: } else {
1379: return false;
1380: }
1381: }
1382: return true;
1383: }
1384: }
|