0001: /*
0002: * Copyright 1999-2004 The Apache Software Foundation.
0003: *
0004: * Licensed under the Apache License, Version 2.0 (the "License");
0005: * you may not use this file except in compliance with the License.
0006: * You may obtain a copy of the License at
0007: *
0008: * http://www.apache.org/licenses/LICENSE-2.0
0009: *
0010: * Unless required by applicable law or agreed to in writing, software
0011: * distributed under the License is distributed on an "AS IS" BASIS,
0012: * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
0013: * See the License for the specific language governing permissions and
0014: * limitations under the License.
0015: */
0016: /*
0017: * $Id: URI.java,v 1.13 2005/02/22 21:35:41 ytalwar Exp $
0018: */
0019: package org.apache.xml.utils;
0020:
0021: import java.io.IOException;
0022: import java.io.Serializable;
0023:
0024: import org.apache.xml.res.XMLErrorResources;
0025: import org.apache.xml.res.XMLMessages;
0026:
0027: /**
0028: * A class to represent a Uniform Resource Identifier (URI). This class
0029: * is designed to handle the parsing of URIs and provide access to
0030: * the various components (scheme, host, port, userinfo, path, query
0031: * string and fragment) that may constitute a URI.
0032: * <p>
0033: * Parsing of a URI specification is done according to the URI
0034: * syntax described in RFC 2396
0035: * <http://www.ietf.org/rfc/rfc2396.txt?number=2396>. Every URI consists
0036: * of a scheme, followed by a colon (':'), followed by a scheme-specific
0037: * part. For URIs that follow the "generic URI" syntax, the scheme-
0038: * specific part begins with two slashes ("//") and may be followed
0039: * by an authority segment (comprised of user information, host, and
0040: * port), path segment, query segment and fragment. Note that RFC 2396
0041: * no longer specifies the use of the parameters segment and excludes
0042: * the "user:password" syntax as part of the authority segment. If
0043: * "user:password" appears in a URI, the entire user/password string
0044: * is stored as userinfo.
0045: * <p>
0046: * For URIs that do not follow the "generic URI" syntax (e.g. mailto),
0047: * the entire scheme-specific part is treated as the "path" portion
0048: * of the URI.
0049: * <p>
0050: * Note that, unlike the java.net.URL class, this class does not provide
0051: * any built-in network access functionality nor does it provide any
0052: * scheme-specific functionality (for example, it does not know a
0053: * default port for a specific scheme). Rather, it only knows the
0054: * grammar and basic set of operations that can be applied to a URI.
0055: *
0056: *
0057: */
0058: public class URI implements Serializable {
0059: static final long serialVersionUID = 7096266377907081897L;
0060:
0061: /**
0062: * MalformedURIExceptions are thrown in the process of building a URI
0063: * or setting fields on a URI when an operation would result in an
0064: * invalid URI specification.
0065: *
0066: */
0067: public static class MalformedURIException extends IOException {
0068:
0069: /**
0070: * Constructs a <code>MalformedURIException</code> with no specified
0071: * detail message.
0072: */
0073: public MalformedURIException() {
0074: super ();
0075: }
0076:
0077: /**
0078: * Constructs a <code>MalformedURIException</code> with the
0079: * specified detail message.
0080: *
0081: * @param p_msg the detail message.
0082: */
0083: public MalformedURIException(String p_msg) {
0084: super (p_msg);
0085: }
0086: }
0087:
0088: /** reserved characters */
0089: private static final String RESERVED_CHARACTERS = ";/?:@&=+$,";
0090:
0091: /**
0092: * URI punctuation mark characters - these, combined with
0093: * alphanumerics, constitute the "unreserved" characters
0094: */
0095: private static final String MARK_CHARACTERS = "-_.!~*'() ";
0096:
0097: /** scheme can be composed of alphanumerics and these characters */
0098: private static final String SCHEME_CHARACTERS = "+-.";
0099:
0100: /**
0101: * userinfo can be composed of unreserved, escaped and these
0102: * characters
0103: */
0104: private static final String USERINFO_CHARACTERS = ";:&=+$,";
0105:
0106: /** Stores the scheme (usually the protocol) for this URI.
0107: * @serial */
0108: private String m_scheme = null;
0109:
0110: /** If specified, stores the userinfo for this URI; otherwise null.
0111: * @serial */
0112: private String m_userinfo = null;
0113:
0114: /** If specified, stores the host for this URI; otherwise null.
0115: * @serial */
0116: private String m_host = null;
0117:
0118: /** If specified, stores the port for this URI; otherwise -1.
0119: * @serial */
0120: private int m_port = -1;
0121:
0122: /** If specified, stores the path for this URI; otherwise null.
0123: * @serial */
0124: private String m_path = null;
0125:
0126: /**
0127: * If specified, stores the query string for this URI; otherwise
0128: * null.
0129: * @serial
0130: */
0131: private String m_queryString = null;
0132:
0133: /** If specified, stores the fragment for this URI; otherwise null.
0134: * @serial */
0135: private String m_fragment = null;
0136:
0137: /** Indicate whether in DEBUG mode */
0138: private static boolean DEBUG = false;
0139:
0140: /**
0141: * Construct a new and uninitialized URI.
0142: */
0143: public URI() {
0144: }
0145:
0146: /**
0147: * Construct a new URI from another URI. All fields for this URI are
0148: * set equal to the fields of the URI passed in.
0149: *
0150: * @param p_other the URI to copy (cannot be null)
0151: */
0152: public URI(URI p_other) {
0153: initialize(p_other);
0154: }
0155:
0156: /**
0157: * Construct a new URI from a URI specification string. If the
0158: * specification follows the "generic URI" syntax, (two slashes
0159: * following the first colon), the specification will be parsed
0160: * accordingly - setting the scheme, userinfo, host,port, path, query
0161: * string and fragment fields as necessary. If the specification does
0162: * not follow the "generic URI" syntax, the specification is parsed
0163: * into a scheme and scheme-specific part (stored as the path) only.
0164: *
0165: * @param p_uriSpec the URI specification string (cannot be null or
0166: * empty)
0167: *
0168: * @throws MalformedURIException if p_uriSpec violates any syntax
0169: * rules
0170: */
0171: public URI(String p_uriSpec) throws MalformedURIException {
0172: this ((URI) null, p_uriSpec);
0173: }
0174:
0175: /**
0176: * Construct a new URI from a base URI and a URI specification string.
0177: * The URI specification string may be a relative URI.
0178: *
0179: * @param p_base the base URI (cannot be null if p_uriSpec is null or
0180: * empty)
0181: * @param p_uriSpec the URI specification string (cannot be null or
0182: * empty if p_base is null)
0183: *
0184: * @throws MalformedURIException if p_uriSpec violates any syntax
0185: * rules
0186: */
0187: public URI(URI p_base, String p_uriSpec)
0188: throws MalformedURIException {
0189: initialize(p_base, p_uriSpec);
0190: }
0191:
0192: /**
0193: * Construct a new URI that does not follow the generic URI syntax.
0194: * Only the scheme and scheme-specific part (stored as the path) are
0195: * initialized.
0196: *
0197: * @param p_scheme the URI scheme (cannot be null or empty)
0198: * @param p_schemeSpecificPart the scheme-specific part (cannot be
0199: * null or empty)
0200: *
0201: * @throws MalformedURIException if p_scheme violates any
0202: * syntax rules
0203: */
0204: public URI(String p_scheme, String p_schemeSpecificPart)
0205: throws MalformedURIException {
0206:
0207: if (p_scheme == null || p_scheme.trim().length() == 0) {
0208: throw new MalformedURIException(
0209: "Cannot construct URI with null/empty scheme!");
0210: }
0211:
0212: if (p_schemeSpecificPart == null
0213: || p_schemeSpecificPart.trim().length() == 0) {
0214: throw new MalformedURIException(
0215: "Cannot construct URI with null/empty scheme-specific part!");
0216: }
0217:
0218: setScheme(p_scheme);
0219: setPath(p_schemeSpecificPart);
0220: }
0221:
0222: /**
0223: * Construct a new URI that follows the generic URI syntax from its
0224: * component parts. Each component is validated for syntax and some
0225: * basic semantic checks are performed as well. See the individual
0226: * setter methods for specifics.
0227: *
0228: * @param p_scheme the URI scheme (cannot be null or empty)
0229: * @param p_host the hostname or IPv4 address for the URI
0230: * @param p_path the URI path - if the path contains '?' or '#',
0231: * then the query string and/or fragment will be
0232: * set from the path; however, if the query and
0233: * fragment are specified both in the path and as
0234: * separate parameters, an exception is thrown
0235: * @param p_queryString the URI query string (cannot be specified
0236: * if path is null)
0237: * @param p_fragment the URI fragment (cannot be specified if path
0238: * is null)
0239: *
0240: * @throws MalformedURIException if any of the parameters violates
0241: * syntax rules or semantic rules
0242: */
0243: public URI(String p_scheme, String p_host, String p_path,
0244: String p_queryString, String p_fragment)
0245: throws MalformedURIException {
0246: this (p_scheme, null, p_host, -1, p_path, p_queryString,
0247: p_fragment);
0248: }
0249:
0250: /**
0251: * Construct a new URI that follows the generic URI syntax from its
0252: * component parts. Each component is validated for syntax and some
0253: * basic semantic checks are performed as well. See the individual
0254: * setter methods for specifics.
0255: *
0256: * @param p_scheme the URI scheme (cannot be null or empty)
0257: * @param p_userinfo the URI userinfo (cannot be specified if host
0258: * is null)
0259: * @param p_host the hostname or IPv4 address for the URI
0260: * @param p_port the URI port (may be -1 for "unspecified"; cannot
0261: * be specified if host is null)
0262: * @param p_path the URI path - if the path contains '?' or '#',
0263: * then the query string and/or fragment will be
0264: * set from the path; however, if the query and
0265: * fragment are specified both in the path and as
0266: * separate parameters, an exception is thrown
0267: * @param p_queryString the URI query string (cannot be specified
0268: * if path is null)
0269: * @param p_fragment the URI fragment (cannot be specified if path
0270: * is null)
0271: *
0272: * @throws MalformedURIException if any of the parameters violates
0273: * syntax rules or semantic rules
0274: */
0275: public URI(String p_scheme, String p_userinfo, String p_host,
0276: int p_port, String p_path, String p_queryString,
0277: String p_fragment) throws MalformedURIException {
0278:
0279: if (p_scheme == null || p_scheme.trim().length() == 0) {
0280: throw new MalformedURIException(XMLMessages
0281: .createXMLMessage(
0282: XMLErrorResources.ER_SCHEME_REQUIRED, null)); //"Scheme is required!");
0283: }
0284:
0285: if (p_host == null) {
0286: if (p_userinfo != null) {
0287: throw new MalformedURIException(
0288: XMLMessages
0289: .createXMLMessage(
0290: XMLErrorResources.ER_NO_USERINFO_IF_NO_HOST,
0291: null)); //"Userinfo may not be specified if host is not specified!");
0292: }
0293:
0294: if (p_port != -1) {
0295: throw new MalformedURIException(
0296: XMLMessages
0297: .createXMLMessage(
0298: XMLErrorResources.ER_NO_PORT_IF_NO_HOST,
0299: null)); //"Port may not be specified if host is not specified!");
0300: }
0301: }
0302:
0303: if (p_path != null) {
0304: if (p_path.indexOf('?') != -1 && p_queryString != null) {
0305: throw new MalformedURIException(
0306: XMLMessages
0307: .createXMLMessage(
0308: XMLErrorResources.ER_NO_QUERY_STRING_IN_PATH,
0309: null)); //"Query string cannot be specified in path and query string!");
0310: }
0311:
0312: if (p_path.indexOf('#') != -1 && p_fragment != null) {
0313: throw new MalformedURIException(
0314: XMLMessages
0315: .createXMLMessage(
0316: XMLErrorResources.ER_NO_FRAGMENT_STRING_IN_PATH,
0317: null)); //"Fragment cannot be specified in both the path and fragment!");
0318: }
0319: }
0320:
0321: setScheme(p_scheme);
0322: setHost(p_host);
0323: setPort(p_port);
0324: setUserinfo(p_userinfo);
0325: setPath(p_path);
0326: setQueryString(p_queryString);
0327: setFragment(p_fragment);
0328: }
0329:
0330: /**
0331: * Initialize all fields of this URI from another URI.
0332: *
0333: * @param p_other the URI to copy (cannot be null)
0334: */
0335: private void initialize(URI p_other) {
0336:
0337: m_scheme = p_other.getScheme();
0338: m_userinfo = p_other.getUserinfo();
0339: m_host = p_other.getHost();
0340: m_port = p_other.getPort();
0341: m_path = p_other.getPath();
0342: m_queryString = p_other.getQueryString();
0343: m_fragment = p_other.getFragment();
0344: }
0345:
0346: /**
0347: * Initializes this URI from a base URI and a URI specification string.
0348: * See RFC 2396 Section 4 and Appendix B for specifications on parsing
0349: * the URI and Section 5 for specifications on resolving relative URIs
0350: * and relative paths.
0351: *
0352: * @param p_base the base URI (may be null if p_uriSpec is an absolute
0353: * URI)
0354: * @param p_uriSpec the URI spec string which may be an absolute or
0355: * relative URI (can only be null/empty if p_base
0356: * is not null)
0357: *
0358: * @throws MalformedURIException if p_base is null and p_uriSpec
0359: * is not an absolute URI or if
0360: * p_uriSpec violates syntax rules
0361: */
0362: private void initialize(URI p_base, String p_uriSpec)
0363: throws MalformedURIException {
0364:
0365: if (p_base == null
0366: && (p_uriSpec == null || p_uriSpec.trim().length() == 0)) {
0367: throw new MalformedURIException(
0368: XMLMessages
0369: .createXMLMessage(
0370: XMLErrorResources.ER_CANNOT_INIT_URI_EMPTY_PARMS,
0371: null)); //"Cannot initialize URI with empty parameters.");
0372: }
0373:
0374: // just make a copy of the base if spec is empty
0375: if (p_uriSpec == null || p_uriSpec.trim().length() == 0) {
0376: initialize(p_base);
0377:
0378: return;
0379: }
0380:
0381: String uriSpec = p_uriSpec.trim();
0382: int uriSpecLen = uriSpec.length();
0383: int index = 0;
0384:
0385: // check for scheme
0386: int colonIndex = uriSpec.indexOf(':');
0387: if (colonIndex < 0) {
0388: if (p_base == null) {
0389: throw new MalformedURIException(XMLMessages
0390: .createXMLMessage(
0391: XMLErrorResources.ER_NO_SCHEME_IN_URI,
0392: new Object[] { uriSpec })); //"No scheme found in URI: "+uriSpec);
0393: }
0394: } else {
0395: initializeScheme(uriSpec);
0396: uriSpec = uriSpec.substring(colonIndex + 1);
0397: // This is a fix for XALANJ-2059.
0398: if (m_scheme != null && p_base != null) {
0399: // a) If <uriSpec> starts with a slash (/), it means <uriSpec> is absolute
0400: // and p_base can be ignored.
0401: // For example,
0402: // uriSpec = file:/myDIR/myXSLFile.xsl
0403: // p_base = file:/myWork/
0404: //
0405: // Here, uriSpec has absolute path after scheme file and :
0406: // Hence p_base can be ignored.
0407: //
0408: // b) Similarily, according to RFC 2396, uri is resolved for <uriSpec> relative to <p_base>
0409: // if scheme in <uriSpec> is same as scheme in <p_base>, else p_base can be ignored.
0410: //
0411: // c) if <p_base> is not hierarchical, it can be ignored.
0412: //
0413: if (uriSpec.startsWith("/")
0414: || !m_scheme.equals(p_base.m_scheme)
0415: || !p_base.getSchemeSpecificPart().startsWith(
0416: "/")) {
0417: p_base = null;
0418: }
0419: }
0420: // Fix for XALANJ-2059
0421: uriSpecLen = uriSpec.length();
0422: }
0423:
0424: // two slashes means generic URI syntax, so we get the authority
0425: if (((index + 1) < uriSpecLen)
0426: && (uriSpec.substring(index).startsWith("//"))) {
0427: index += 2;
0428:
0429: int startPos = index;
0430:
0431: // get authority - everything up to path, query or fragment
0432: char testChar = '\0';
0433:
0434: while (index < uriSpecLen) {
0435: testChar = uriSpec.charAt(index);
0436:
0437: if (testChar == '/' || testChar == '?'
0438: || testChar == '#') {
0439: break;
0440: }
0441:
0442: index++;
0443: }
0444:
0445: // if we found authority, parse it out, otherwise we set the
0446: // host to empty string
0447: if (index > startPos) {
0448: initializeAuthority(uriSpec.substring(startPos, index));
0449: } else {
0450: m_host = "";
0451: }
0452: }
0453:
0454: initializePath(uriSpec.substring(index));
0455:
0456: // Resolve relative URI to base URI - see RFC 2396 Section 5.2
0457: // In some cases, it might make more sense to throw an exception
0458: // (when scheme is specified is the string spec and the base URI
0459: // is also specified, for example), but we're just following the
0460: // RFC specifications
0461: if (p_base != null) {
0462:
0463: // check to see if this is the current doc - RFC 2396 5.2 #2
0464: // note that this is slightly different from the RFC spec in that
0465: // we don't include the check for query string being null
0466: // - this handles cases where the urispec is just a query
0467: // string or a fragment (e.g. "?y" or "#s") -
0468: // see <http://www.ics.uci.edu/~fielding/url/test1.html> which
0469: // identified this as a bug in the RFC
0470: if (m_path.length() == 0 && m_scheme == null
0471: && m_host == null) {
0472: m_scheme = p_base.getScheme();
0473: m_userinfo = p_base.getUserinfo();
0474: m_host = p_base.getHost();
0475: m_port = p_base.getPort();
0476: m_path = p_base.getPath();
0477:
0478: if (m_queryString == null) {
0479: m_queryString = p_base.getQueryString();
0480: }
0481:
0482: return;
0483: }
0484:
0485: // check for scheme - RFC 2396 5.2 #3
0486: // if we found a scheme, it means absolute URI, so we're done
0487: if (m_scheme == null) {
0488: m_scheme = p_base.getScheme();
0489: }
0490:
0491: // check for authority - RFC 2396 5.2 #4
0492: // if we found a host, then we've got a network path, so we're done
0493: if (m_host == null) {
0494: m_userinfo = p_base.getUserinfo();
0495: m_host = p_base.getHost();
0496: m_port = p_base.getPort();
0497: } else {
0498: return;
0499: }
0500:
0501: // check for absolute path - RFC 2396 5.2 #5
0502: if (m_path.length() > 0 && m_path.startsWith("/")) {
0503: return;
0504: }
0505:
0506: // if we get to this point, we need to resolve relative path
0507: // RFC 2396 5.2 #6
0508: String path = new String();
0509: String basePath = p_base.getPath();
0510:
0511: // 6a - get all but the last segment of the base URI path
0512: if (basePath != null) {
0513: int lastSlash = basePath.lastIndexOf('/');
0514:
0515: if (lastSlash != -1) {
0516: path = basePath.substring(0, lastSlash + 1);
0517: }
0518: }
0519:
0520: // 6b - append the relative URI path
0521: path = path.concat(m_path);
0522:
0523: // 6c - remove all "./" where "." is a complete path segment
0524: index = -1;
0525:
0526: while ((index = path.indexOf("/./")) != -1) {
0527: path = path.substring(0, index + 1).concat(
0528: path.substring(index + 3));
0529: }
0530:
0531: // 6d - remove "." if path ends with "." as a complete path segment
0532: if (path.endsWith("/.")) {
0533: path = path.substring(0, path.length() - 1);
0534: }
0535:
0536: // 6e - remove all "<segment>/../" where "<segment>" is a complete
0537: // path segment not equal to ".."
0538: index = -1;
0539:
0540: int segIndex = -1;
0541: String tempString = null;
0542:
0543: while ((index = path.indexOf("/../")) > 0) {
0544: tempString = path.substring(0, path.indexOf("/../"));
0545: segIndex = tempString.lastIndexOf('/');
0546:
0547: if (segIndex != -1) {
0548: if (!tempString.substring(segIndex++).equals("..")) {
0549: path = path.substring(0, segIndex).concat(
0550: path.substring(index + 4));
0551: }
0552: }
0553: }
0554:
0555: // 6f - remove ending "<segment>/.." where "<segment>" is a
0556: // complete path segment
0557: if (path.endsWith("/..")) {
0558: tempString = path.substring(0, path.length() - 3);
0559: segIndex = tempString.lastIndexOf('/');
0560:
0561: if (segIndex != -1) {
0562: path = path.substring(0, segIndex + 1);
0563: }
0564: }
0565:
0566: m_path = path;
0567: }
0568: }
0569:
0570: /**
0571: * Initialize the scheme for this URI from a URI string spec.
0572: *
0573: * @param p_uriSpec the URI specification (cannot be null)
0574: *
0575: * @throws MalformedURIException if URI does not have a conformant
0576: * scheme
0577: */
0578: private void initializeScheme(String p_uriSpec)
0579: throws MalformedURIException {
0580:
0581: int uriSpecLen = p_uriSpec.length();
0582: int index = 0;
0583: String scheme = null;
0584: char testChar = '\0';
0585:
0586: while (index < uriSpecLen) {
0587: testChar = p_uriSpec.charAt(index);
0588:
0589: if (testChar == ':' || testChar == '/' || testChar == '?'
0590: || testChar == '#') {
0591: break;
0592: }
0593:
0594: index++;
0595: }
0596:
0597: scheme = p_uriSpec.substring(0, index);
0598:
0599: if (scheme.length() == 0) {
0600: throw new MalformedURIException(XMLMessages
0601: .createXMLMessage(
0602: XMLErrorResources.ER_NO_SCHEME_INURI, null)); //"No scheme found in URI.");
0603: } else {
0604: setScheme(scheme);
0605: }
0606: }
0607:
0608: /**
0609: * Initialize the authority (userinfo, host and port) for this
0610: * URI from a URI string spec.
0611: *
0612: * @param p_uriSpec the URI specification (cannot be null)
0613: *
0614: * @throws MalformedURIException if p_uriSpec violates syntax rules
0615: */
0616: private void initializeAuthority(String p_uriSpec)
0617: throws MalformedURIException {
0618:
0619: int index = 0;
0620: int start = 0;
0621: int end = p_uriSpec.length();
0622: char testChar = '\0';
0623: String userinfo = null;
0624:
0625: // userinfo is everything up @
0626: if (p_uriSpec.indexOf('@', start) != -1) {
0627: while (index < end) {
0628: testChar = p_uriSpec.charAt(index);
0629:
0630: if (testChar == '@') {
0631: break;
0632: }
0633:
0634: index++;
0635: }
0636:
0637: userinfo = p_uriSpec.substring(start, index);
0638:
0639: index++;
0640: }
0641:
0642: // host is everything up to ':'
0643: String host = null;
0644:
0645: start = index;
0646:
0647: while (index < end) {
0648: testChar = p_uriSpec.charAt(index);
0649:
0650: if (testChar == ':') {
0651: break;
0652: }
0653:
0654: index++;
0655: }
0656:
0657: host = p_uriSpec.substring(start, index);
0658:
0659: int port = -1;
0660:
0661: if (host.length() > 0) {
0662:
0663: // port
0664: if (testChar == ':') {
0665: index++;
0666:
0667: start = index;
0668:
0669: while (index < end) {
0670: index++;
0671: }
0672:
0673: String portStr = p_uriSpec.substring(start, index);
0674:
0675: if (portStr.length() > 0) {
0676: for (int i = 0; i < portStr.length(); i++) {
0677: if (!isDigit(portStr.charAt(i))) {
0678: throw new MalformedURIException(
0679: portStr
0680: + " is invalid. Port should only contain digits!");
0681: }
0682: }
0683:
0684: try {
0685: port = Integer.parseInt(portStr);
0686: } catch (NumberFormatException nfe) {
0687:
0688: // can't happen
0689: }
0690: }
0691: }
0692: }
0693:
0694: setHost(host);
0695: setPort(port);
0696: setUserinfo(userinfo);
0697: }
0698:
0699: /**
0700: * Initialize the path for this URI from a URI string spec.
0701: *
0702: * @param p_uriSpec the URI specification (cannot be null)
0703: *
0704: * @throws MalformedURIException if p_uriSpec violates syntax rules
0705: */
0706: private void initializePath(String p_uriSpec)
0707: throws MalformedURIException {
0708:
0709: if (p_uriSpec == null) {
0710: throw new MalformedURIException(
0711: "Cannot initialize path from null string!");
0712: }
0713:
0714: int index = 0;
0715: int start = 0;
0716: int end = p_uriSpec.length();
0717: char testChar = '\0';
0718:
0719: // path - everything up to query string or fragment
0720: while (index < end) {
0721: testChar = p_uriSpec.charAt(index);
0722:
0723: if (testChar == '?' || testChar == '#') {
0724: break;
0725: }
0726:
0727: // check for valid escape sequence
0728: if (testChar == '%') {
0729: if (index + 2 >= end
0730: || !isHex(p_uriSpec.charAt(index + 1))
0731: || !isHex(p_uriSpec.charAt(index + 2))) {
0732: throw new MalformedURIException(
0733: XMLMessages
0734: .createXMLMessage(
0735: XMLErrorResources.ER_PATH_CONTAINS_INVALID_ESCAPE_SEQUENCE,
0736: null)); //"Path contains invalid escape sequence!");
0737: }
0738: } else if (!isReservedCharacter(testChar)
0739: && !isUnreservedCharacter(testChar)) {
0740: if ('\\' != testChar)
0741: throw new MalformedURIException(
0742: XMLMessages
0743: .createXMLMessage(
0744: XMLErrorResources.ER_PATH_INVALID_CHAR,
0745: new Object[] { String
0746: .valueOf(testChar) })); //"Path contains invalid character: "
0747: //+ testChar);
0748: }
0749:
0750: index++;
0751: }
0752:
0753: m_path = p_uriSpec.substring(start, index);
0754:
0755: // query - starts with ? and up to fragment or end
0756: if (testChar == '?') {
0757: index++;
0758:
0759: start = index;
0760:
0761: while (index < end) {
0762: testChar = p_uriSpec.charAt(index);
0763:
0764: if (testChar == '#') {
0765: break;
0766: }
0767:
0768: if (testChar == '%') {
0769: if (index + 2 >= end
0770: || !isHex(p_uriSpec.charAt(index + 1))
0771: || !isHex(p_uriSpec.charAt(index + 2))) {
0772: throw new MalformedURIException(
0773: "Query string contains invalid escape sequence!");
0774: }
0775: } else if (!isReservedCharacter(testChar)
0776: && !isUnreservedCharacter(testChar)) {
0777: throw new MalformedURIException(
0778: "Query string contains invalid character:"
0779: + testChar);
0780: }
0781:
0782: index++;
0783: }
0784:
0785: m_queryString = p_uriSpec.substring(start, index);
0786: }
0787:
0788: // fragment - starts with #
0789: if (testChar == '#') {
0790: index++;
0791:
0792: start = index;
0793:
0794: while (index < end) {
0795: testChar = p_uriSpec.charAt(index);
0796:
0797: if (testChar == '%') {
0798: if (index + 2 >= end
0799: || !isHex(p_uriSpec.charAt(index + 1))
0800: || !isHex(p_uriSpec.charAt(index + 2))) {
0801: throw new MalformedURIException(
0802: "Fragment contains invalid escape sequence!");
0803: }
0804: } else if (!isReservedCharacter(testChar)
0805: && !isUnreservedCharacter(testChar)) {
0806: throw new MalformedURIException(
0807: "Fragment contains invalid character:"
0808: + testChar);
0809: }
0810:
0811: index++;
0812: }
0813:
0814: m_fragment = p_uriSpec.substring(start, index);
0815: }
0816: }
0817:
0818: /**
0819: * Get the scheme for this URI.
0820: *
0821: * @return the scheme for this URI
0822: */
0823: public String getScheme() {
0824: return m_scheme;
0825: }
0826:
0827: /**
0828: * Get the scheme-specific part for this URI (everything following the
0829: * scheme and the first colon). See RFC 2396 Section 5.2 for spec.
0830: *
0831: * @return the scheme-specific part for this URI
0832: */
0833: public String getSchemeSpecificPart() {
0834:
0835: StringBuffer schemespec = new StringBuffer();
0836:
0837: if (m_userinfo != null || m_host != null || m_port != -1) {
0838: schemespec.append("//");
0839: }
0840:
0841: if (m_userinfo != null) {
0842: schemespec.append(m_userinfo);
0843: schemespec.append('@');
0844: }
0845:
0846: if (m_host != null) {
0847: schemespec.append(m_host);
0848: }
0849:
0850: if (m_port != -1) {
0851: schemespec.append(':');
0852: schemespec.append(m_port);
0853: }
0854:
0855: if (m_path != null) {
0856: schemespec.append((m_path));
0857: }
0858:
0859: if (m_queryString != null) {
0860: schemespec.append('?');
0861: schemespec.append(m_queryString);
0862: }
0863:
0864: if (m_fragment != null) {
0865: schemespec.append('#');
0866: schemespec.append(m_fragment);
0867: }
0868:
0869: return schemespec.toString();
0870: }
0871:
0872: /**
0873: * Get the userinfo for this URI.
0874: *
0875: * @return the userinfo for this URI (null if not specified).
0876: */
0877: public String getUserinfo() {
0878: return m_userinfo;
0879: }
0880:
0881: /**
0882: * Get the host for this URI.
0883: *
0884: * @return the host for this URI (null if not specified).
0885: */
0886: public String getHost() {
0887: return m_host;
0888: }
0889:
0890: /**
0891: * Get the port for this URI.
0892: *
0893: * @return the port for this URI (-1 if not specified).
0894: */
0895: public int getPort() {
0896: return m_port;
0897: }
0898:
0899: /**
0900: * Get the path for this URI (optionally with the query string and
0901: * fragment).
0902: *
0903: * @param p_includeQueryString if true (and query string is not null),
0904: * then a "?" followed by the query string
0905: * will be appended
0906: * @param p_includeFragment if true (and fragment is not null),
0907: * then a "#" followed by the fragment
0908: * will be appended
0909: *
0910: * @return the path for this URI possibly including the query string
0911: * and fragment
0912: */
0913: public String getPath(boolean p_includeQueryString,
0914: boolean p_includeFragment) {
0915:
0916: StringBuffer pathString = new StringBuffer(m_path);
0917:
0918: if (p_includeQueryString && m_queryString != null) {
0919: pathString.append('?');
0920: pathString.append(m_queryString);
0921: }
0922:
0923: if (p_includeFragment && m_fragment != null) {
0924: pathString.append('#');
0925: pathString.append(m_fragment);
0926: }
0927:
0928: return pathString.toString();
0929: }
0930:
0931: /**
0932: * Get the path for this URI. Note that the value returned is the path
0933: * only and does not include the query string or fragment.
0934: *
0935: * @return the path for this URI.
0936: */
0937: public String getPath() {
0938: return m_path;
0939: }
0940:
0941: /**
0942: * Get the query string for this URI.
0943: *
0944: * @return the query string for this URI. Null is returned if there
0945: * was no "?" in the URI spec, empty string if there was a
0946: * "?" but no query string following it.
0947: */
0948: public String getQueryString() {
0949: return m_queryString;
0950: }
0951:
0952: /**
0953: * Get the fragment for this URI.
0954: *
0955: * @return the fragment for this URI. Null is returned if there
0956: * was no "#" in the URI spec, empty string if there was a
0957: * "#" but no fragment following it.
0958: */
0959: public String getFragment() {
0960: return m_fragment;
0961: }
0962:
0963: /**
0964: * Set the scheme for this URI. The scheme is converted to lowercase
0965: * before it is set.
0966: *
0967: * @param p_scheme the scheme for this URI (cannot be null)
0968: *
0969: * @throws MalformedURIException if p_scheme is not a conformant
0970: * scheme name
0971: */
0972: public void setScheme(String p_scheme) throws MalformedURIException {
0973:
0974: if (p_scheme == null) {
0975: throw new MalformedURIException(
0976: XMLMessages
0977: .createXMLMessage(
0978: XMLErrorResources.ER_SCHEME_FROM_NULL_STRING,
0979: null)); //"Cannot set scheme from null string!");
0980: }
0981:
0982: if (!isConformantSchemeName(p_scheme)) {
0983: throw new MalformedURIException(XMLMessages
0984: .createXMLMessage(
0985: XMLErrorResources.ER_SCHEME_NOT_CONFORMANT,
0986: null)); //"The scheme is not conformant.");
0987: }
0988:
0989: m_scheme = p_scheme.toLowerCase();
0990: }
0991:
0992: /**
0993: * Set the userinfo for this URI. If a non-null value is passed in and
0994: * the host value is null, then an exception is thrown.
0995: *
0996: * @param p_userinfo the userinfo for this URI
0997: *
0998: * @throws MalformedURIException if p_userinfo contains invalid
0999: * characters
1000: */
1001: public void setUserinfo(String p_userinfo)
1002: throws MalformedURIException {
1003:
1004: if (p_userinfo == null) {
1005: m_userinfo = null;
1006: } else {
1007: if (m_host == null) {
1008: throw new MalformedURIException(
1009: "Userinfo cannot be set when host is null!");
1010: }
1011:
1012: // userinfo can contain alphanumerics, mark characters, escaped
1013: // and ';',':','&','=','+','$',','
1014: int index = 0;
1015: int end = p_userinfo.length();
1016: char testChar = '\0';
1017:
1018: while (index < end) {
1019: testChar = p_userinfo.charAt(index);
1020:
1021: if (testChar == '%') {
1022: if (index + 2 >= end
1023: || !isHex(p_userinfo.charAt(index + 1))
1024: || !isHex(p_userinfo.charAt(index + 2))) {
1025: throw new MalformedURIException(
1026: "Userinfo contains invalid escape sequence!");
1027: }
1028: } else if (!isUnreservedCharacter(testChar)
1029: && USERINFO_CHARACTERS.indexOf(testChar) == -1) {
1030: throw new MalformedURIException(
1031: "Userinfo contains invalid character:"
1032: + testChar);
1033: }
1034:
1035: index++;
1036: }
1037: }
1038:
1039: m_userinfo = p_userinfo;
1040: }
1041:
1042: /**
1043: * Set the host for this URI. If null is passed in, the userinfo
1044: * field is also set to null and the port is set to -1.
1045: *
1046: * @param p_host the host for this URI
1047: *
1048: * @throws MalformedURIException if p_host is not a valid IP
1049: * address or DNS hostname.
1050: */
1051: public void setHost(String p_host) throws MalformedURIException {
1052:
1053: if (p_host == null || p_host.trim().length() == 0) {
1054: m_host = p_host;
1055: m_userinfo = null;
1056: m_port = -1;
1057: } else if (!isWellFormedAddress(p_host)) {
1058: throw new MalformedURIException(
1059: XMLMessages
1060: .createXMLMessage(
1061: XMLErrorResources.ER_HOST_ADDRESS_NOT_WELLFORMED,
1062: null)); //"Host is not a well formed address!");
1063: }
1064:
1065: m_host = p_host;
1066: }
1067:
1068: /**
1069: * Set the port for this URI. -1 is used to indicate that the port is
1070: * not specified, otherwise valid port numbers are between 0 and 65535.
1071: * If a valid port number is passed in and the host field is null,
1072: * an exception is thrown.
1073: *
1074: * @param p_port the port number for this URI
1075: *
1076: * @throws MalformedURIException if p_port is not -1 and not a
1077: * valid port number
1078: */
1079: public void setPort(int p_port) throws MalformedURIException {
1080:
1081: if (p_port >= 0 && p_port <= 65535) {
1082: if (m_host == null) {
1083: throw new MalformedURIException(
1084: XMLMessages
1085: .createXMLMessage(
1086: XMLErrorResources.ER_PORT_WHEN_HOST_NULL,
1087: null)); //"Port cannot be set when host is null!");
1088: }
1089: } else if (p_port != -1) {
1090: throw new MalformedURIException(XMLMessages
1091: .createXMLMessage(
1092: XMLErrorResources.ER_INVALID_PORT, null)); //"Invalid port number!");
1093: }
1094:
1095: m_port = p_port;
1096: }
1097:
1098: /**
1099: * Set the path for this URI. If the supplied path is null, then the
1100: * query string and fragment are set to null as well. If the supplied
1101: * path includes a query string and/or fragment, these fields will be
1102: * parsed and set as well. Note that, for URIs following the "generic
1103: * URI" syntax, the path specified should start with a slash.
1104: * For URIs that do not follow the generic URI syntax, this method
1105: * sets the scheme-specific part.
1106: *
1107: * @param p_path the path for this URI (may be null)
1108: *
1109: * @throws MalformedURIException if p_path contains invalid
1110: * characters
1111: */
1112: public void setPath(String p_path) throws MalformedURIException {
1113:
1114: if (p_path == null) {
1115: m_path = null;
1116: m_queryString = null;
1117: m_fragment = null;
1118: } else {
1119: initializePath(p_path);
1120: }
1121: }
1122:
1123: /**
1124: * Append to the end of the path of this URI. If the current path does
1125: * not end in a slash and the path to be appended does not begin with
1126: * a slash, a slash will be appended to the current path before the
1127: * new segment is added. Also, if the current path ends in a slash
1128: * and the new segment begins with a slash, the extra slash will be
1129: * removed before the new segment is appended.
1130: *
1131: * @param p_addToPath the new segment to be added to the current path
1132: *
1133: * @throws MalformedURIException if p_addToPath contains syntax
1134: * errors
1135: */
1136: public void appendPath(String p_addToPath)
1137: throws MalformedURIException {
1138:
1139: if (p_addToPath == null || p_addToPath.trim().length() == 0) {
1140: return;
1141: }
1142:
1143: if (!isURIString(p_addToPath)) {
1144: throw new MalformedURIException(XMLMessages
1145: .createXMLMessage(
1146: XMLErrorResources.ER_PATH_INVALID_CHAR,
1147: new Object[] { p_addToPath })); //"Path contains invalid character!");
1148: }
1149:
1150: if (m_path == null || m_path.trim().length() == 0) {
1151: if (p_addToPath.startsWith("/")) {
1152: m_path = p_addToPath;
1153: } else {
1154: m_path = "/" + p_addToPath;
1155: }
1156: } else if (m_path.endsWith("/")) {
1157: if (p_addToPath.startsWith("/")) {
1158: m_path = m_path.concat(p_addToPath.substring(1));
1159: } else {
1160: m_path = m_path.concat(p_addToPath);
1161: }
1162: } else {
1163: if (p_addToPath.startsWith("/")) {
1164: m_path = m_path.concat(p_addToPath);
1165: } else {
1166: m_path = m_path.concat("/" + p_addToPath);
1167: }
1168: }
1169: }
1170:
1171: /**
1172: * Set the query string for this URI. A non-null value is valid only
1173: * if this is an URI conforming to the generic URI syntax and
1174: * the path value is not null.
1175: *
1176: * @param p_queryString the query string for this URI
1177: *
1178: * @throws MalformedURIException if p_queryString is not null and this
1179: * URI does not conform to the generic
1180: * URI syntax or if the path is null
1181: */
1182: public void setQueryString(String p_queryString)
1183: throws MalformedURIException {
1184:
1185: if (p_queryString == null) {
1186: m_queryString = null;
1187: } else if (!isGenericURI()) {
1188: throw new MalformedURIException(
1189: "Query string can only be set for a generic URI!");
1190: } else if (getPath() == null) {
1191: throw new MalformedURIException(
1192: "Query string cannot be set when path is null!");
1193: } else if (!isURIString(p_queryString)) {
1194: throw new MalformedURIException(
1195: "Query string contains invalid character!");
1196: } else {
1197: m_queryString = p_queryString;
1198: }
1199: }
1200:
1201: /**
1202: * Set the fragment for this URI. A non-null value is valid only
1203: * if this is a URI conforming to the generic URI syntax and
1204: * the path value is not null.
1205: *
1206: * @param p_fragment the fragment for this URI
1207: *
1208: * @throws MalformedURIException if p_fragment is not null and this
1209: * URI does not conform to the generic
1210: * URI syntax or if the path is null
1211: */
1212: public void setFragment(String p_fragment)
1213: throws MalformedURIException {
1214:
1215: if (p_fragment == null) {
1216: m_fragment = null;
1217: } else if (!isGenericURI()) {
1218: throw new MalformedURIException(XMLMessages
1219: .createXMLMessage(
1220: XMLErrorResources.ER_FRAG_FOR_GENERIC_URI,
1221: null)); //"Fragment can only be set for a generic URI!");
1222: } else if (getPath() == null) {
1223: throw new MalformedURIException(XMLMessages
1224: .createXMLMessage(
1225: XMLErrorResources.ER_FRAG_WHEN_PATH_NULL,
1226: null)); //"Fragment cannot be set when path is null!");
1227: } else if (!isURIString(p_fragment)) {
1228: throw new MalformedURIException(XMLMessages
1229: .createXMLMessage(
1230: XMLErrorResources.ER_FRAG_INVALID_CHAR,
1231: null)); //"Fragment contains invalid character!");
1232: } else {
1233: m_fragment = p_fragment;
1234: }
1235: }
1236:
1237: /**
1238: * Determines if the passed-in Object is equivalent to this URI.
1239: *
1240: * @param p_test the Object to test for equality.
1241: *
1242: * @return true if p_test is a URI with all values equal to this
1243: * URI, false otherwise
1244: */
1245: public boolean equals(Object p_test) {
1246:
1247: if (p_test instanceof URI) {
1248: URI testURI = (URI) p_test;
1249:
1250: if (((m_scheme == null && testURI.m_scheme == null) || (m_scheme != null
1251: && testURI.m_scheme != null && m_scheme
1252: .equals(testURI.m_scheme)))
1253: && ((m_userinfo == null && testURI.m_userinfo == null) || (m_userinfo != null
1254: && testURI.m_userinfo != null && m_userinfo
1255: .equals(testURI.m_userinfo)))
1256: && ((m_host == null && testURI.m_host == null) || (m_host != null
1257: && testURI.m_host != null && m_host
1258: .equals(testURI.m_host)))
1259: && m_port == testURI.m_port
1260: && ((m_path == null && testURI.m_path == null) || (m_path != null
1261: && testURI.m_path != null && m_path
1262: .equals(testURI.m_path)))
1263: && ((m_queryString == null && testURI.m_queryString == null) || (m_queryString != null
1264: && testURI.m_queryString != null && m_queryString
1265: .equals(testURI.m_queryString)))
1266: && ((m_fragment == null && testURI.m_fragment == null) || (m_fragment != null
1267: && testURI.m_fragment != null && m_fragment
1268: .equals(testURI.m_fragment)))) {
1269: return true;
1270: }
1271: }
1272:
1273: return false;
1274: }
1275:
1276: /**
1277: * Get the URI as a string specification. See RFC 2396 Section 5.2.
1278: *
1279: * @return the URI string specification
1280: */
1281: public String toString() {
1282:
1283: StringBuffer uriSpecString = new StringBuffer();
1284:
1285: if (m_scheme != null) {
1286: uriSpecString.append(m_scheme);
1287: uriSpecString.append(':');
1288: }
1289:
1290: uriSpecString.append(getSchemeSpecificPart());
1291:
1292: return uriSpecString.toString();
1293: }
1294:
1295: /**
1296: * Get the indicator as to whether this URI uses the "generic URI"
1297: * syntax.
1298: *
1299: * @return true if this URI uses the "generic URI" syntax, false
1300: * otherwise
1301: */
1302: public boolean isGenericURI() {
1303:
1304: // presence of the host (whether valid or empty) means
1305: // double-slashes which means generic uri
1306: return (m_host != null);
1307: }
1308:
1309: /**
1310: * Determine whether a scheme conforms to the rules for a scheme name.
1311: * A scheme is conformant if it starts with an alphanumeric, and
1312: * contains only alphanumerics, '+','-' and '.'.
1313: *
1314: *
1315: * @param p_scheme The sheme name to check
1316: * @return true if the scheme is conformant, false otherwise
1317: */
1318: public static boolean isConformantSchemeName(String p_scheme) {
1319:
1320: if (p_scheme == null || p_scheme.trim().length() == 0) {
1321: return false;
1322: }
1323:
1324: if (!isAlpha(p_scheme.charAt(0))) {
1325: return false;
1326: }
1327:
1328: char testChar;
1329:
1330: for (int i = 1; i < p_scheme.length(); i++) {
1331: testChar = p_scheme.charAt(i);
1332:
1333: if (!isAlphanum(testChar)
1334: && SCHEME_CHARACTERS.indexOf(testChar) == -1) {
1335: return false;
1336: }
1337: }
1338:
1339: return true;
1340: }
1341:
1342: /**
1343: * Determine whether a string is syntactically capable of representing
1344: * a valid IPv4 address or the domain name of a network host. A valid
1345: * IPv4 address consists of four decimal digit groups separated by a
1346: * '.'. A hostname consists of domain labels (each of which must
1347: * begin and end with an alphanumeric but may contain '-') separated
1348: * & by a '.'. See RFC 2396 Section 3.2.2.
1349: *
1350: *
1351: * @param p_address The address string to check
1352: * @return true if the string is a syntactically valid IPv4 address
1353: * or hostname
1354: */
1355: public static boolean isWellFormedAddress(String p_address) {
1356:
1357: if (p_address == null) {
1358: return false;
1359: }
1360:
1361: String address = p_address.trim();
1362: int addrLength = address.length();
1363:
1364: if (addrLength == 0 || addrLength > 255) {
1365: return false;
1366: }
1367:
1368: if (address.startsWith(".") || address.startsWith("-")) {
1369: return false;
1370: }
1371:
1372: // rightmost domain label starting with digit indicates IP address
1373: // since top level domain label can only start with an alpha
1374: // see RFC 2396 Section 3.2.2
1375: int index = address.lastIndexOf('.');
1376:
1377: if (address.endsWith(".")) {
1378: index = address.substring(0, index).lastIndexOf('.');
1379: }
1380:
1381: if (index + 1 < addrLength
1382: && isDigit(p_address.charAt(index + 1))) {
1383: char testChar;
1384: int numDots = 0;
1385:
1386: // make sure that 1) we see only digits and dot separators, 2) that
1387: // any dot separator is preceded and followed by a digit and
1388: // 3) that we find 3 dots
1389: for (int i = 0; i < addrLength; i++) {
1390: testChar = address.charAt(i);
1391:
1392: if (testChar == '.') {
1393: if (!isDigit(address.charAt(i - 1))
1394: || (i + 1 < addrLength && !isDigit(address
1395: .charAt(i + 1)))) {
1396: return false;
1397: }
1398:
1399: numDots++;
1400: } else if (!isDigit(testChar)) {
1401: return false;
1402: }
1403: }
1404:
1405: if (numDots != 3) {
1406: return false;
1407: }
1408: } else {
1409:
1410: // domain labels can contain alphanumerics and '-"
1411: // but must start and end with an alphanumeric
1412: char testChar;
1413:
1414: for (int i = 0; i < addrLength; i++) {
1415: testChar = address.charAt(i);
1416:
1417: if (testChar == '.') {
1418: if (!isAlphanum(address.charAt(i - 1))) {
1419: return false;
1420: }
1421:
1422: if (i + 1 < addrLength
1423: && !isAlphanum(address.charAt(i + 1))) {
1424: return false;
1425: }
1426: } else if (!isAlphanum(testChar) && testChar != '-') {
1427: return false;
1428: }
1429: }
1430: }
1431:
1432: return true;
1433: }
1434:
1435: /**
1436: * Determine whether a char is a digit.
1437: *
1438: *
1439: * @param p_char the character to check
1440: * @return true if the char is betweeen '0' and '9', false otherwise
1441: */
1442: private static boolean isDigit(char p_char) {
1443: return p_char >= '0' && p_char <= '9';
1444: }
1445:
1446: /**
1447: * Determine whether a character is a hexadecimal character.
1448: *
1449: *
1450: * @param p_char the character to check
1451: * @return true if the char is betweeen '0' and '9', 'a' and 'f'
1452: * or 'A' and 'F', false otherwise
1453: */
1454: private static boolean isHex(char p_char) {
1455: return (isDigit(p_char) || (p_char >= 'a' && p_char <= 'f') || (p_char >= 'A' && p_char <= 'F'));
1456: }
1457:
1458: /**
1459: * Determine whether a char is an alphabetic character: a-z or A-Z
1460: *
1461: *
1462: * @param p_char the character to check
1463: * @return true if the char is alphabetic, false otherwise
1464: */
1465: private static boolean isAlpha(char p_char) {
1466: return ((p_char >= 'a' && p_char <= 'z') || (p_char >= 'A' && p_char <= 'Z'));
1467: }
1468:
1469: /**
1470: * Determine whether a char is an alphanumeric: 0-9, a-z or A-Z
1471: *
1472: *
1473: * @param p_char the character to check
1474: * @return true if the char is alphanumeric, false otherwise
1475: */
1476: private static boolean isAlphanum(char p_char) {
1477: return (isAlpha(p_char) || isDigit(p_char));
1478: }
1479:
1480: /**
1481: * Determine whether a character is a reserved character:
1482: * ';', '/', '?', ':', '@', '&', '=', '+', '$' or ','
1483: *
1484: *
1485: * @param p_char the character to check
1486: * @return true if the string contains any reserved characters
1487: */
1488: private static boolean isReservedCharacter(char p_char) {
1489: return RESERVED_CHARACTERS.indexOf(p_char) != -1;
1490: }
1491:
1492: /**
1493: * Determine whether a char is an unreserved character.
1494: *
1495: *
1496: * @param p_char the character to check
1497: * @return true if the char is unreserved, false otherwise
1498: */
1499: private static boolean isUnreservedCharacter(char p_char) {
1500: return (isAlphanum(p_char) || MARK_CHARACTERS.indexOf(p_char) != -1);
1501: }
1502:
1503: /**
1504: * Determine whether a given string contains only URI characters (also
1505: * called "uric" in RFC 2396). uric consist of all reserved
1506: * characters, unreserved characters and escaped characters.
1507: *
1508: *
1509: * @param p_uric URI string
1510: * @return true if the string is comprised of uric, false otherwise
1511: */
1512: private static boolean isURIString(String p_uric) {
1513:
1514: if (p_uric == null) {
1515: return false;
1516: }
1517:
1518: int end = p_uric.length();
1519: char testChar = '\0';
1520:
1521: for (int i = 0; i < end; i++) {
1522: testChar = p_uric.charAt(i);
1523:
1524: if (testChar == '%') {
1525: if (i + 2 >= end || !isHex(p_uric.charAt(i + 1))
1526: || !isHex(p_uric.charAt(i + 2))) {
1527: return false;
1528: } else {
1529: i += 2;
1530:
1531: continue;
1532: }
1533: }
1534:
1535: if (isReservedCharacter(testChar)
1536: || isUnreservedCharacter(testChar)) {
1537: continue;
1538: } else {
1539: return false;
1540: }
1541: }
1542:
1543: return true;
1544: }
1545: }
|