0001: /*
0002: * Licensed to the Apache Software Foundation (ASF) under one or more
0003: * contributor license agreements. See the NOTICE file distributed with
0004: * this work for additional information regarding copyright ownership.
0005: * The ASF licenses this file to You under the Apache License, Version 2.0
0006: * (the "License"); you may not use this file except in compliance with
0007: * the License. You may obtain a copy of the License at
0008: *
0009: * http://www.apache.org/licenses/LICENSE-2.0
0010: *
0011: * Unless required by applicable law or agreed to in writing, software
0012: * distributed under the License is distributed on an "AS IS" BASIS,
0013: * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
0014: * See the License for the specific language governing permissions and
0015: * limitations under the License.
0016: */
0017:
0018: package org.apache.xerces.util;
0019:
0020: import java.io.IOException;
0021: import java.io.Serializable;
0022:
0023: /**********************************************************************
0024: * A class to represent a Uniform Resource Identifier (URI). This class
0025: * is designed to handle the parsing of URIs and provide access to
0026: * the various components (scheme, host, port, userinfo, path, query
0027: * string and fragment) that may constitute a URI.
0028: * <p>
0029: * Parsing of a URI specification is done according to the URI
0030: * syntax described in
0031: * <a href="http://www.ietf.org/rfc/rfc2396.txt?number=2396">RFC 2396</a>,
0032: * and amended by
0033: * <a href="http://www.ietf.org/rfc/rfc2732.txt?number=2732">RFC 2732</a>.
0034: * <p>
0035: * Every absolute URI consists of a scheme, followed by a colon (':'),
0036: * followed by a scheme-specific part. For URIs that follow the
0037: * "generic URI" syntax, the scheme-specific part begins with two
0038: * slashes ("//") and may be followed by an authority segment (comprised
0039: * of user information, host, and port), path segment, query segment
0040: * and fragment. Note that RFC 2396 no longer specifies the use of the
0041: * parameters segment and excludes the "user:password" syntax as part of
0042: * the authority segment. If "user:password" appears in a URI, the entire
0043: * user/password string is stored as userinfo.
0044: * <p>
0045: * For URIs that do not follow the "generic URI" syntax (e.g. mailto),
0046: * the entire scheme-specific part is treated as the "path" portion
0047: * of the URI.
0048: * <p>
0049: * Note that, unlike the java.net.URL class, this class does not provide
0050: * any built-in network access functionality nor does it provide any
0051: * scheme-specific functionality (for example, it does not know a
0052: * default port for a specific scheme). Rather, it only knows the
0053: * grammar and basic set of operations that can be applied to a URI.
0054: *
0055: * @version $Id: URI.java 447241 2006-09-18 05:12:57Z mrglavas $
0056: *
0057: **********************************************************************/
0058: public class URI implements Serializable {
0059:
0060: /*******************************************************************
0061: * MalformedURIExceptions are thrown in the process of building a URI
0062: * or setting fields on a URI when an operation would result in an
0063: * invalid URI specification.
0064: *
0065: ********************************************************************/
0066: public static class MalformedURIException extends IOException {
0067:
0068: /** Serialization version. */
0069: static final long serialVersionUID = -6695054834342951930L;
0070:
0071: /******************************************************************
0072: * Constructs a <code>MalformedURIException</code> with no specified
0073: * detail message.
0074: ******************************************************************/
0075: public MalformedURIException() {
0076: super ();
0077: }
0078:
0079: /*****************************************************************
0080: * Constructs a <code>MalformedURIException</code> with the
0081: * specified detail message.
0082: *
0083: * @param p_msg the detail message.
0084: ******************************************************************/
0085: public MalformedURIException(String p_msg) {
0086: super (p_msg);
0087: }
0088: }
0089:
0090: /** Serialization version. */
0091: static final long serialVersionUID = 1601921774685357214L;
0092:
0093: private static final byte[] fgLookupTable = new byte[128];
0094:
0095: /**
0096: * Character Classes
0097: */
0098:
0099: /** reserved characters ;/?:@&=+$,[] */
0100: //RFC 2732 added '[' and ']' as reserved characters
0101: private static final int RESERVED_CHARACTERS = 0x01;
0102:
0103: /** URI punctuation mark characters: -_.!~*'() - these, combined with
0104: alphanumerics, constitute the "unreserved" characters */
0105: private static final int MARK_CHARACTERS = 0x02;
0106:
0107: /** scheme can be composed of alphanumerics and these characters: +-. */
0108: private static final int SCHEME_CHARACTERS = 0x04;
0109:
0110: /** userinfo can be composed of unreserved, escaped and these
0111: characters: ;:&=+$, */
0112: private static final int USERINFO_CHARACTERS = 0x08;
0113:
0114: /** ASCII letter characters */
0115: private static final int ASCII_ALPHA_CHARACTERS = 0x10;
0116:
0117: /** ASCII digit characters */
0118: private static final int ASCII_DIGIT_CHARACTERS = 0x20;
0119:
0120: /** ASCII hex characters */
0121: private static final int ASCII_HEX_CHARACTERS = 0x40;
0122:
0123: /** Path characters */
0124: private static final int PATH_CHARACTERS = 0x80;
0125:
0126: /** Mask for alpha-numeric characters */
0127: private static final int MASK_ALPHA_NUMERIC = ASCII_ALPHA_CHARACTERS
0128: | ASCII_DIGIT_CHARACTERS;
0129:
0130: /** Mask for unreserved characters */
0131: private static final int MASK_UNRESERVED_MASK = MASK_ALPHA_NUMERIC
0132: | MARK_CHARACTERS;
0133:
0134: /** Mask for URI allowable characters except for % */
0135: private static final int MASK_URI_CHARACTER = MASK_UNRESERVED_MASK
0136: | RESERVED_CHARACTERS;
0137:
0138: /** Mask for scheme characters */
0139: private static final int MASK_SCHEME_CHARACTER = MASK_ALPHA_NUMERIC
0140: | SCHEME_CHARACTERS;
0141:
0142: /** Mask for userinfo characters */
0143: private static final int MASK_USERINFO_CHARACTER = MASK_UNRESERVED_MASK
0144: | USERINFO_CHARACTERS;
0145:
0146: /** Mask for path characters */
0147: private static final int MASK_PATH_CHARACTER = MASK_UNRESERVED_MASK
0148: | PATH_CHARACTERS;
0149:
0150: static {
0151: // Add ASCII Digits and ASCII Hex Numbers
0152: for (int i = '0'; i <= '9'; ++i) {
0153: fgLookupTable[i] |= ASCII_DIGIT_CHARACTERS
0154: | ASCII_HEX_CHARACTERS;
0155: }
0156:
0157: // Add ASCII Letters and ASCII Hex Numbers
0158: for (int i = 'A'; i <= 'F'; ++i) {
0159: fgLookupTable[i] |= ASCII_ALPHA_CHARACTERS
0160: | ASCII_HEX_CHARACTERS;
0161: fgLookupTable[i + 0x00000020] |= ASCII_ALPHA_CHARACTERS
0162: | ASCII_HEX_CHARACTERS;
0163: }
0164:
0165: // Add ASCII Letters
0166: for (int i = 'G'; i <= 'Z'; ++i) {
0167: fgLookupTable[i] |= ASCII_ALPHA_CHARACTERS;
0168: fgLookupTable[i + 0x00000020] |= ASCII_ALPHA_CHARACTERS;
0169: }
0170:
0171: // Add Reserved Characters
0172: fgLookupTable[';'] |= RESERVED_CHARACTERS;
0173: fgLookupTable['/'] |= RESERVED_CHARACTERS;
0174: fgLookupTable['?'] |= RESERVED_CHARACTERS;
0175: fgLookupTable[':'] |= RESERVED_CHARACTERS;
0176: fgLookupTable['@'] |= RESERVED_CHARACTERS;
0177: fgLookupTable['&'] |= RESERVED_CHARACTERS;
0178: fgLookupTable['='] |= RESERVED_CHARACTERS;
0179: fgLookupTable['+'] |= RESERVED_CHARACTERS;
0180: fgLookupTable['$'] |= RESERVED_CHARACTERS;
0181: fgLookupTable[','] |= RESERVED_CHARACTERS;
0182: fgLookupTable['['] |= RESERVED_CHARACTERS;
0183: fgLookupTable[']'] |= RESERVED_CHARACTERS;
0184:
0185: // Add Mark Characters
0186: fgLookupTable['-'] |= MARK_CHARACTERS;
0187: fgLookupTable['_'] |= MARK_CHARACTERS;
0188: fgLookupTable['.'] |= MARK_CHARACTERS;
0189: fgLookupTable['!'] |= MARK_CHARACTERS;
0190: fgLookupTable['~'] |= MARK_CHARACTERS;
0191: fgLookupTable['*'] |= MARK_CHARACTERS;
0192: fgLookupTable['\''] |= MARK_CHARACTERS;
0193: fgLookupTable['('] |= MARK_CHARACTERS;
0194: fgLookupTable[')'] |= MARK_CHARACTERS;
0195:
0196: // Add Scheme Characters
0197: fgLookupTable['+'] |= SCHEME_CHARACTERS;
0198: fgLookupTable['-'] |= SCHEME_CHARACTERS;
0199: fgLookupTable['.'] |= SCHEME_CHARACTERS;
0200:
0201: // Add Userinfo Characters
0202: fgLookupTable[';'] |= USERINFO_CHARACTERS;
0203: fgLookupTable[':'] |= USERINFO_CHARACTERS;
0204: fgLookupTable['&'] |= USERINFO_CHARACTERS;
0205: fgLookupTable['='] |= USERINFO_CHARACTERS;
0206: fgLookupTable['+'] |= USERINFO_CHARACTERS;
0207: fgLookupTable['$'] |= USERINFO_CHARACTERS;
0208: fgLookupTable[','] |= USERINFO_CHARACTERS;
0209:
0210: // Add Path Characters
0211: fgLookupTable[';'] |= PATH_CHARACTERS;
0212: fgLookupTable['/'] |= PATH_CHARACTERS;
0213: fgLookupTable[':'] |= PATH_CHARACTERS;
0214: fgLookupTable['@'] |= PATH_CHARACTERS;
0215: fgLookupTable['&'] |= PATH_CHARACTERS;
0216: fgLookupTable['='] |= PATH_CHARACTERS;
0217: fgLookupTable['+'] |= PATH_CHARACTERS;
0218: fgLookupTable['$'] |= PATH_CHARACTERS;
0219: fgLookupTable[','] |= PATH_CHARACTERS;
0220: }
0221:
0222: /** Stores the scheme (usually the protocol) for this URI. */
0223: private String m_scheme = null;
0224:
0225: /** If specified, stores the userinfo for this URI; otherwise null */
0226: private String m_userinfo = null;
0227:
0228: /** If specified, stores the host for this URI; otherwise null */
0229: private String m_host = null;
0230:
0231: /** If specified, stores the port for this URI; otherwise -1 */
0232: private int m_port = -1;
0233:
0234: /** If specified, stores the registry based authority for this URI; otherwise -1 */
0235: private String m_regAuthority = null;
0236:
0237: /** If specified, stores the path for this URI; otherwise null */
0238: private String m_path = null;
0239:
0240: /** If specified, stores the query string for this URI; otherwise
0241: null. */
0242: private String m_queryString = null;
0243:
0244: /** If specified, stores the fragment for this URI; otherwise null */
0245: private String m_fragment = null;
0246:
0247: private static boolean DEBUG = false;
0248:
0249: /**
0250: * Construct a new and uninitialized URI.
0251: */
0252: public URI() {
0253: }
0254:
0255: /**
0256: * Construct a new URI from another URI. All fields for this URI are
0257: * set equal to the fields of the URI passed in.
0258: *
0259: * @param p_other the URI to copy (cannot be null)
0260: */
0261: public URI(URI p_other) {
0262: initialize(p_other);
0263: }
0264:
0265: /**
0266: * Construct a new URI from a URI specification string. If the
0267: * specification follows the "generic URI" syntax, (two slashes
0268: * following the first colon), the specification will be parsed
0269: * accordingly - setting the scheme, userinfo, host,port, path, query
0270: * string and fragment fields as necessary. If the specification does
0271: * not follow the "generic URI" syntax, the specification is parsed
0272: * into a scheme and scheme-specific part (stored as the path) only.
0273: *
0274: * @param p_uriSpec the URI specification string (cannot be null or
0275: * empty)
0276: *
0277: * @exception MalformedURIException if p_uriSpec violates any syntax
0278: * rules
0279: */
0280: public URI(String p_uriSpec) throws MalformedURIException {
0281: this ((URI) null, p_uriSpec);
0282: }
0283:
0284: /**
0285: * Construct a new URI from a URI specification string. If the
0286: * specification follows the "generic URI" syntax, (two slashes
0287: * following the first colon), the specification will be parsed
0288: * accordingly - setting the scheme, userinfo, host,port, path, query
0289: * string and fragment fields as necessary. If the specification does
0290: * not follow the "generic URI" syntax, the specification is parsed
0291: * into a scheme and scheme-specific part (stored as the path) only.
0292: * Construct a relative URI if boolean is assigned to "true"
0293: * and p_uriSpec is not valid absolute URI, instead of throwing an exception.
0294: *
0295: * @param p_uriSpec the URI specification string (cannot be null or
0296: * empty)
0297: * @param allowNonAbsoluteURI true to permit non-absolute URIs,
0298: * false otherwise.
0299: *
0300: * @exception MalformedURIException if p_uriSpec violates any syntax
0301: * rules
0302: */
0303: public URI(String p_uriSpec, boolean allowNonAbsoluteURI)
0304: throws MalformedURIException {
0305: this ((URI) null, p_uriSpec, allowNonAbsoluteURI);
0306: }
0307:
0308: /**
0309: * Construct a new URI from a base URI and a URI specification string.
0310: * The URI specification string may be a relative URI.
0311: *
0312: * @param p_base the base URI (cannot be null if p_uriSpec is null or
0313: * empty)
0314: * @param p_uriSpec the URI specification string (cannot be null or
0315: * empty if p_base is null)
0316: *
0317: * @exception MalformedURIException if p_uriSpec violates any syntax
0318: * rules
0319: */
0320: public URI(URI p_base, String p_uriSpec)
0321: throws MalformedURIException {
0322: initialize(p_base, p_uriSpec);
0323: }
0324:
0325: /**
0326: * Construct a new URI from a base URI and a URI specification string.
0327: * The URI specification string may be a relative URI.
0328: * Construct a relative URI if boolean is assigned to "true"
0329: * and p_uriSpec is not valid absolute URI and p_base is null
0330: * instead of throwing an exception.
0331: *
0332: * @param p_base the base URI (cannot be null if p_uriSpec is null or
0333: * empty)
0334: * @param p_uriSpec the URI specification string (cannot be null or
0335: * empty if p_base is null)
0336: * @param allowNonAbsoluteURI true to permit non-absolute URIs,
0337: * false otherwise.
0338: *
0339: * @exception MalformedURIException if p_uriSpec violates any syntax
0340: * rules
0341: */
0342: public URI(URI p_base, String p_uriSpec, boolean allowNonAbsoluteURI)
0343: throws MalformedURIException {
0344: initialize(p_base, p_uriSpec, allowNonAbsoluteURI);
0345: }
0346:
0347: /**
0348: * Construct a new URI that does not follow the generic URI syntax.
0349: * Only the scheme and scheme-specific part (stored as the path) are
0350: * initialized.
0351: *
0352: * @param p_scheme the URI scheme (cannot be null or empty)
0353: * @param p_schemeSpecificPart the scheme-specific part (cannot be
0354: * null or empty)
0355: *
0356: * @exception MalformedURIException if p_scheme violates any
0357: * syntax rules
0358: */
0359: public URI(String p_scheme, String p_schemeSpecificPart)
0360: throws MalformedURIException {
0361: if (p_scheme == null || p_scheme.trim().length() == 0) {
0362: throw new MalformedURIException(
0363: "Cannot construct URI with null/empty scheme!");
0364: }
0365: if (p_schemeSpecificPart == null
0366: || p_schemeSpecificPart.trim().length() == 0) {
0367: throw new MalformedURIException(
0368: "Cannot construct URI with null/empty scheme-specific part!");
0369: }
0370: setScheme(p_scheme);
0371: setPath(p_schemeSpecificPart);
0372: }
0373:
0374: /**
0375: * Construct a new URI that follows the generic URI syntax from its
0376: * component parts. Each component is validated for syntax and some
0377: * basic semantic checks are performed as well. See the individual
0378: * setter methods for specifics.
0379: *
0380: * @param p_scheme the URI scheme (cannot be null or empty)
0381: * @param p_host the hostname, IPv4 address or IPv6 reference for the URI
0382: * @param p_path the URI path - if the path contains '?' or '#',
0383: * then the query string and/or fragment will be
0384: * set from the path; however, if the query and
0385: * fragment are specified both in the path and as
0386: * separate parameters, an exception is thrown
0387: * @param p_queryString the URI query string (cannot be specified
0388: * if path is null)
0389: * @param p_fragment the URI fragment (cannot be specified if path
0390: * is null)
0391: *
0392: * @exception MalformedURIException if any of the parameters violates
0393: * syntax rules or semantic rules
0394: */
0395: public URI(String p_scheme, String p_host, String p_path,
0396: String p_queryString, String p_fragment)
0397: throws MalformedURIException {
0398: this (p_scheme, null, p_host, -1, p_path, p_queryString,
0399: p_fragment);
0400: }
0401:
0402: /**
0403: * Construct a new URI that follows the generic URI syntax from its
0404: * component parts. Each component is validated for syntax and some
0405: * basic semantic checks are performed as well. See the individual
0406: * setter methods for specifics.
0407: *
0408: * @param p_scheme the URI scheme (cannot be null or empty)
0409: * @param p_userinfo the URI userinfo (cannot be specified if host
0410: * is null)
0411: * @param p_host the hostname, IPv4 address or IPv6 reference for the URI
0412: * @param p_port the URI port (may be -1 for "unspecified"; cannot
0413: * be specified if host is null)
0414: * @param p_path the URI path - if the path contains '?' or '#',
0415: * then the query string and/or fragment will be
0416: * set from the path; however, if the query and
0417: * fragment are specified both in the path and as
0418: * separate parameters, an exception is thrown
0419: * @param p_queryString the URI query string (cannot be specified
0420: * if path is null)
0421: * @param p_fragment the URI fragment (cannot be specified if path
0422: * is null)
0423: *
0424: * @exception MalformedURIException if any of the parameters violates
0425: * syntax rules or semantic rules
0426: */
0427: public URI(String p_scheme, String p_userinfo, String p_host,
0428: int p_port, String p_path, String p_queryString,
0429: String p_fragment) throws MalformedURIException {
0430: if (p_scheme == null || p_scheme.trim().length() == 0) {
0431: throw new MalformedURIException("Scheme is required!");
0432: }
0433:
0434: if (p_host == null) {
0435: if (p_userinfo != null) {
0436: throw new MalformedURIException(
0437: "Userinfo may not be specified if host is not specified!");
0438: }
0439: if (p_port != -1) {
0440: throw new MalformedURIException(
0441: "Port may not be specified if host is not specified!");
0442: }
0443: }
0444:
0445: if (p_path != null) {
0446: if (p_path.indexOf('?') != -1 && p_queryString != null) {
0447: throw new MalformedURIException(
0448: "Query string cannot be specified in path and query string!");
0449: }
0450:
0451: if (p_path.indexOf('#') != -1 && p_fragment != null) {
0452: throw new MalformedURIException(
0453: "Fragment cannot be specified in both the path and fragment!");
0454: }
0455: }
0456:
0457: setScheme(p_scheme);
0458: setHost(p_host);
0459: setPort(p_port);
0460: setUserinfo(p_userinfo);
0461: setPath(p_path);
0462: setQueryString(p_queryString);
0463: setFragment(p_fragment);
0464: }
0465:
0466: /**
0467: * Initialize all fields of this URI from another URI.
0468: *
0469: * @param p_other the URI to copy (cannot be null)
0470: */
0471: private void initialize(URI p_other) {
0472: m_scheme = p_other.getScheme();
0473: m_userinfo = p_other.getUserinfo();
0474: m_host = p_other.getHost();
0475: m_port = p_other.getPort();
0476: m_regAuthority = p_other.getRegBasedAuthority();
0477: m_path = p_other.getPath();
0478: m_queryString = p_other.getQueryString();
0479: m_fragment = p_other.getFragment();
0480: }
0481:
0482: /**
0483: * Initializes this URI from a base URI and a URI specification string.
0484: * See RFC 2396 Section 4 and Appendix B for specifications on parsing
0485: * the URI and Section 5 for specifications on resolving relative URIs
0486: * and relative paths.
0487: *
0488: * @param p_base the base URI (may be null if p_uriSpec is an absolute
0489: * URI)
0490: * @param p_uriSpec the URI spec string which may be an absolute or
0491: * relative URI (can only be null/empty if p_base
0492: * is not null)
0493: * @param allowNonAbsoluteURI true to permit non-absolute URIs,
0494: * in case of relative URI, false otherwise.
0495: *
0496: * @exception MalformedURIException if p_base is null and p_uriSpec
0497: * is not an absolute URI or if
0498: * p_uriSpec violates syntax rules
0499: */
0500: private void initialize(URI p_base, String p_uriSpec,
0501: boolean allowNonAbsoluteURI) throws MalformedURIException {
0502:
0503: String uriSpec = p_uriSpec;
0504: int uriSpecLen = (uriSpec != null) ? uriSpec.length() : 0;
0505:
0506: if (p_base == null && uriSpecLen == 0) {
0507: if (allowNonAbsoluteURI) {
0508: m_path = "";
0509: return;
0510: }
0511: throw new MalformedURIException(
0512: "Cannot initialize URI with empty parameters.");
0513: }
0514:
0515: // just make a copy of the base if spec is empty
0516: if (uriSpecLen == 0) {
0517: initialize(p_base);
0518: return;
0519: }
0520:
0521: int index = 0;
0522:
0523: // Check for scheme, which must be before '/', '?' or '#'.
0524: int colonIdx = uriSpec.indexOf(':');
0525: if (colonIdx != -1) {
0526: final int searchFrom = colonIdx - 1;
0527: // search backwards starting from character before ':'.
0528: int slashIdx = uriSpec.lastIndexOf('/', searchFrom);
0529: int queryIdx = uriSpec.lastIndexOf('?', searchFrom);
0530: int fragmentIdx = uriSpec.lastIndexOf('#', searchFrom);
0531:
0532: if (colonIdx == 0 || slashIdx != -1 || queryIdx != -1
0533: || fragmentIdx != -1) {
0534: // A standalone base is a valid URI according to spec
0535: if (colonIdx == 0
0536: || (p_base == null && fragmentIdx != 0 && !allowNonAbsoluteURI)) {
0537: throw new MalformedURIException(
0538: "No scheme found in URI.");
0539: }
0540: } else {
0541: initializeScheme(uriSpec);
0542: index = m_scheme.length() + 1;
0543:
0544: // Neither 'scheme:' or 'scheme:#fragment' are valid URIs.
0545: if (colonIdx == uriSpecLen - 1
0546: || uriSpec.charAt(colonIdx + 1) == '#') {
0547: throw new MalformedURIException(
0548: "Scheme specific part cannot be empty.");
0549: }
0550: }
0551: } else if (p_base == null && uriSpec.indexOf('#') != 0
0552: && !allowNonAbsoluteURI) {
0553: throw new MalformedURIException("No scheme found in URI.");
0554: }
0555:
0556: // Two slashes means we may have authority, but definitely means we're either
0557: // matching net_path or abs_path. These two productions are ambiguous in that
0558: // every net_path (except those containing an IPv6Reference) is an abs_path.
0559: // RFC 2396 resolves this ambiguity by applying a greedy left most matching rule.
0560: // Try matching net_path first, and if that fails we don't have authority so
0561: // then attempt to match abs_path.
0562: //
0563: // net_path = "//" authority [ abs_path ]
0564: // abs_path = "/" path_segments
0565: if (((index + 1) < uriSpecLen)
0566: && (uriSpec.charAt(index) == '/' && uriSpec
0567: .charAt(index + 1) == '/')) {
0568: index += 2;
0569: int startPos = index;
0570:
0571: // Authority will be everything up to path, query or fragment
0572: char testChar = '\0';
0573: while (index < uriSpecLen) {
0574: testChar = uriSpec.charAt(index);
0575: if (testChar == '/' || testChar == '?'
0576: || testChar == '#') {
0577: break;
0578: }
0579: index++;
0580: }
0581:
0582: // Attempt to parse authority. If the section is an empty string
0583: // this is a valid server based authority, so set the host to this
0584: // value.
0585: if (index > startPos) {
0586: // If we didn't find authority we need to back up. Attempt to
0587: // match against abs_path next.
0588: if (!initializeAuthority(uriSpec.substring(startPos,
0589: index))) {
0590: index = startPos - 2;
0591: }
0592: } else {
0593: m_host = "";
0594: }
0595: }
0596:
0597: initializePath(uriSpec, index);
0598:
0599: // Resolve relative URI to base URI - see RFC 2396 Section 5.2
0600: // In some cases, it might make more sense to throw an exception
0601: // (when scheme is specified is the string spec and the base URI
0602: // is also specified, for example), but we're just following the
0603: // RFC specifications
0604: if (p_base != null) {
0605: absolutize(p_base);
0606: }
0607: }
0608:
0609: /**
0610: * Initializes this URI from a base URI and a URI specification string.
0611: * See RFC 2396 Section 4 and Appendix B for specifications on parsing
0612: * the URI and Section 5 for specifications on resolving relative URIs
0613: * and relative paths.
0614: *
0615: * @param p_base the base URI (may be null if p_uriSpec is an absolute
0616: * URI)
0617: * @param p_uriSpec the URI spec string which may be an absolute or
0618: * relative URI (can only be null/empty if p_base
0619: * is not null)
0620: *
0621: * @exception MalformedURIException if p_base is null and p_uriSpec
0622: * is not an absolute URI or if
0623: * p_uriSpec violates syntax rules
0624: */
0625: private void initialize(URI p_base, String p_uriSpec)
0626: throws MalformedURIException {
0627:
0628: String uriSpec = p_uriSpec;
0629: int uriSpecLen = (uriSpec != null) ? uriSpec.length() : 0;
0630:
0631: if (p_base == null && uriSpecLen == 0) {
0632: throw new MalformedURIException(
0633: "Cannot initialize URI with empty parameters.");
0634: }
0635:
0636: // just make a copy of the base if spec is empty
0637: if (uriSpecLen == 0) {
0638: initialize(p_base);
0639: return;
0640: }
0641:
0642: int index = 0;
0643:
0644: // Check for scheme, which must be before '/', '?' or '#'.
0645: int colonIdx = uriSpec.indexOf(':');
0646: if (colonIdx != -1) {
0647: final int searchFrom = colonIdx - 1;
0648: // search backwards starting from character before ':'.
0649: int slashIdx = uriSpec.lastIndexOf('/', searchFrom);
0650: int queryIdx = uriSpec.lastIndexOf('?', searchFrom);
0651: int fragmentIdx = uriSpec.lastIndexOf('#', searchFrom);
0652:
0653: if (colonIdx == 0 || slashIdx != -1 || queryIdx != -1
0654: || fragmentIdx != -1) {
0655: // A standalone base is a valid URI according to spec
0656: if (colonIdx == 0
0657: || (p_base == null && fragmentIdx != 0)) {
0658: throw new MalformedURIException(
0659: "No scheme found in URI.");
0660: }
0661: } else {
0662: initializeScheme(uriSpec);
0663: index = m_scheme.length() + 1;
0664:
0665: // Neither 'scheme:' or 'scheme:#fragment' are valid URIs.
0666: if (colonIdx == uriSpecLen - 1
0667: || uriSpec.charAt(colonIdx + 1) == '#') {
0668: throw new MalformedURIException(
0669: "Scheme specific part cannot be empty.");
0670: }
0671: }
0672: } else if (p_base == null && uriSpec.indexOf('#') != 0) {
0673: throw new MalformedURIException("No scheme found in URI.");
0674: }
0675:
0676: // Two slashes means we may have authority, but definitely means we're either
0677: // matching net_path or abs_path. These two productions are ambiguous in that
0678: // every net_path (except those containing an IPv6Reference) is an abs_path.
0679: // RFC 2396 resolves this ambiguity by applying a greedy left most matching rule.
0680: // Try matching net_path first, and if that fails we don't have authority so
0681: // then attempt to match abs_path.
0682: //
0683: // net_path = "//" authority [ abs_path ]
0684: // abs_path = "/" path_segments
0685: if (((index + 1) < uriSpecLen)
0686: && (uriSpec.charAt(index) == '/' && uriSpec
0687: .charAt(index + 1) == '/')) {
0688: index += 2;
0689: int startPos = index;
0690:
0691: // Authority will be everything up to path, query or fragment
0692: char testChar = '\0';
0693: while (index < uriSpecLen) {
0694: testChar = uriSpec.charAt(index);
0695: if (testChar == '/' || testChar == '?'
0696: || testChar == '#') {
0697: break;
0698: }
0699: index++;
0700: }
0701:
0702: // Attempt to parse authority. If the section is an empty string
0703: // this is a valid server based authority, so set the host to this
0704: // value.
0705: if (index > startPos) {
0706: // If we didn't find authority we need to back up. Attempt to
0707: // match against abs_path next.
0708: if (!initializeAuthority(uriSpec.substring(startPos,
0709: index))) {
0710: index = startPos - 2;
0711: }
0712: } else {
0713: m_host = "";
0714: }
0715: }
0716:
0717: initializePath(uriSpec, index);
0718:
0719: // Resolve relative URI to base URI - see RFC 2396 Section 5.2
0720: // In some cases, it might make more sense to throw an exception
0721: // (when scheme is specified is the string spec and the base URI
0722: // is also specified, for example), but we're just following the
0723: // RFC specifications
0724: if (p_base != null) {
0725: absolutize(p_base);
0726: }
0727: }
0728:
0729: /**
0730: * Absolutize URI with given base URI.
0731: *
0732: * @param p_base base URI for absolutization
0733: */
0734: public void absolutize(URI p_base) {
0735:
0736: // check to see if this is the current doc - RFC 2396 5.2 #2
0737: // note that this is slightly different from the RFC spec in that
0738: // we don't include the check for query string being null
0739: // - this handles cases where the urispec is just a query
0740: // string or a fragment (e.g. "?y" or "#s") -
0741: // see <http://www.ics.uci.edu/~fielding/url/test1.html> which
0742: // identified this as a bug in the RFC
0743: if (m_path.length() == 0 && m_scheme == null && m_host == null
0744: && m_regAuthority == null) {
0745: m_scheme = p_base.getScheme();
0746: m_userinfo = p_base.getUserinfo();
0747: m_host = p_base.getHost();
0748: m_port = p_base.getPort();
0749: m_regAuthority = p_base.getRegBasedAuthority();
0750: m_path = p_base.getPath();
0751:
0752: if (m_queryString == null) {
0753: m_queryString = p_base.getQueryString();
0754:
0755: if (m_fragment == null) {
0756: m_fragment = p_base.getFragment();
0757: }
0758: }
0759: return;
0760: }
0761:
0762: // check for scheme - RFC 2396 5.2 #3
0763: // if we found a scheme, it means absolute URI, so we're done
0764: if (m_scheme == null) {
0765: m_scheme = p_base.getScheme();
0766: } else {
0767: return;
0768: }
0769:
0770: // check for authority - RFC 2396 5.2 #4
0771: // if we found a host, then we've got a network path, so we're done
0772: if (m_host == null && m_regAuthority == null) {
0773: m_userinfo = p_base.getUserinfo();
0774: m_host = p_base.getHost();
0775: m_port = p_base.getPort();
0776: m_regAuthority = p_base.getRegBasedAuthority();
0777: } else {
0778: return;
0779: }
0780:
0781: // check for absolute path - RFC 2396 5.2 #5
0782: if (m_path.length() > 0 && m_path.startsWith("/")) {
0783: return;
0784: }
0785:
0786: // if we get to this point, we need to resolve relative path
0787: // RFC 2396 5.2 #6
0788: String path = "";
0789: String basePath = p_base.getPath();
0790:
0791: // 6a - get all but the last segment of the base URI path
0792: if (basePath != null && basePath.length() > 0) {
0793: int lastSlash = basePath.lastIndexOf('/');
0794: if (lastSlash != -1) {
0795: path = basePath.substring(0, lastSlash + 1);
0796: }
0797: } else if (m_path.length() > 0) {
0798: path = "/";
0799: }
0800:
0801: // 6b - append the relative URI path
0802: path = path.concat(m_path);
0803:
0804: // 6c - remove all "./" where "." is a complete path segment
0805: int index = -1;
0806: while ((index = path.indexOf("/./")) != -1) {
0807: path = path.substring(0, index + 1).concat(
0808: path.substring(index + 3));
0809: }
0810:
0811: // 6d - remove "." if path ends with "." as a complete path segment
0812: if (path.endsWith("/.")) {
0813: path = path.substring(0, path.length() - 1);
0814: }
0815:
0816: // 6e - remove all "<segment>/../" where "<segment>" is a complete
0817: // path segment not equal to ".."
0818: index = 1;
0819: int segIndex = -1;
0820: String tempString = null;
0821:
0822: while ((index = path.indexOf("/../", index)) > 0) {
0823: tempString = path.substring(0, path.indexOf("/../"));
0824: segIndex = tempString.lastIndexOf('/');
0825: if (segIndex != -1) {
0826: if (!tempString.substring(segIndex).equals("..")) {
0827: path = path.substring(0, segIndex + 1).concat(
0828: path.substring(index + 4));
0829: index = segIndex;
0830: } else {
0831: index += 4;
0832: }
0833: } else {
0834: index += 4;
0835: }
0836: }
0837:
0838: // 6f - remove ending "<segment>/.." where "<segment>" is a
0839: // complete path segment
0840: if (path.endsWith("/..")) {
0841: tempString = path.substring(0, path.length() - 3);
0842: segIndex = tempString.lastIndexOf('/');
0843: if (segIndex != -1) {
0844: path = path.substring(0, segIndex + 1);
0845: }
0846: }
0847: m_path = path;
0848: }
0849:
0850: /**
0851: * Initialize the scheme for this URI from a URI string spec.
0852: *
0853: * @param p_uriSpec the URI specification (cannot be null)
0854: *
0855: * @exception MalformedURIException if URI does not have a conformant
0856: * scheme
0857: */
0858: private void initializeScheme(String p_uriSpec)
0859: throws MalformedURIException {
0860: int uriSpecLen = p_uriSpec.length();
0861: int index = 0;
0862: String scheme = null;
0863: char testChar = '\0';
0864:
0865: while (index < uriSpecLen) {
0866: testChar = p_uriSpec.charAt(index);
0867: if (testChar == ':' || testChar == '/' || testChar == '?'
0868: || testChar == '#') {
0869: break;
0870: }
0871: index++;
0872: }
0873: scheme = p_uriSpec.substring(0, index);
0874:
0875: if (scheme.length() == 0) {
0876: throw new MalformedURIException("No scheme found in URI.");
0877: } else {
0878: setScheme(scheme);
0879: }
0880: }
0881:
0882: /**
0883: * Initialize the authority (either server or registry based)
0884: * for this URI from a URI string spec.
0885: *
0886: * @param p_uriSpec the URI specification (cannot be null)
0887: *
0888: * @return true if the given string matched server or registry
0889: * based authority
0890: */
0891: private boolean initializeAuthority(String p_uriSpec) {
0892:
0893: int index = 0;
0894: int start = 0;
0895: int end = p_uriSpec.length();
0896:
0897: char testChar = '\0';
0898: String userinfo = null;
0899:
0900: // userinfo is everything up to @
0901: if (p_uriSpec.indexOf('@', start) != -1) {
0902: while (index < end) {
0903: testChar = p_uriSpec.charAt(index);
0904: if (testChar == '@') {
0905: break;
0906: }
0907: index++;
0908: }
0909: userinfo = p_uriSpec.substring(start, index);
0910: index++;
0911: }
0912:
0913: // host is everything up to last ':', or up to
0914: // and including ']' if followed by ':'.
0915: String host = null;
0916: start = index;
0917: boolean hasPort = false;
0918: if (index < end) {
0919: if (p_uriSpec.charAt(start) == '[') {
0920: int bracketIndex = p_uriSpec.indexOf(']', start);
0921: index = (bracketIndex != -1) ? bracketIndex : end;
0922: if (index + 1 < end
0923: && p_uriSpec.charAt(index + 1) == ':') {
0924: ++index;
0925: hasPort = true;
0926: } else {
0927: index = end;
0928: }
0929: } else {
0930: int colonIndex = p_uriSpec.lastIndexOf(':', end);
0931: index = (colonIndex > start) ? colonIndex : end;
0932: hasPort = (index != end);
0933: }
0934: }
0935: host = p_uriSpec.substring(start, index);
0936: int port = -1;
0937: if (host.length() > 0) {
0938: // port
0939: if (hasPort) {
0940: index++;
0941: start = index;
0942: while (index < end) {
0943: index++;
0944: }
0945: String portStr = p_uriSpec.substring(start, index);
0946: if (portStr.length() > 0) {
0947: // REVISIT: Remove this code.
0948: /** for (int i = 0; i < portStr.length(); i++) {
0949: if (!isDigit(portStr.charAt(i))) {
0950: throw new MalformedURIException(
0951: portStr +
0952: " is invalid. Port should only contain digits!");
0953: }
0954: }**/
0955: // REVISIT: Remove this code.
0956: // Store port value as string instead of integer.
0957: try {
0958: port = Integer.parseInt(portStr);
0959: if (port == -1)
0960: --port;
0961: } catch (NumberFormatException nfe) {
0962: port = -2;
0963: }
0964: }
0965: }
0966: }
0967:
0968: if (isValidServerBasedAuthority(host, port, userinfo)) {
0969: m_host = host;
0970: m_port = port;
0971: m_userinfo = userinfo;
0972: return true;
0973: }
0974: // Note: Registry based authority is being removed from a
0975: // new spec for URI which would obsolete RFC 2396. If the
0976: // spec is added to XML errata, processing of reg_name
0977: // needs to be removed. - mrglavas.
0978: else if (isValidRegistryBasedAuthority(p_uriSpec)) {
0979: m_regAuthority = p_uriSpec;
0980: return true;
0981: }
0982: return false;
0983: }
0984:
0985: /**
0986: * Determines whether the components host, port, and user info
0987: * are valid as a server authority.
0988: *
0989: * @param host the host component of authority
0990: * @param port the port number component of authority
0991: * @param userinfo the user info component of authority
0992: *
0993: * @return true if the given host, port, and userinfo compose
0994: * a valid server authority
0995: */
0996: private boolean isValidServerBasedAuthority(String host, int port,
0997: String userinfo) {
0998:
0999: // Check if the host is well formed.
1000: if (!isWellFormedAddress(host)) {
1001: return false;
1002: }
1003:
1004: // Check that port is well formed if it exists.
1005: // REVISIT: There's no restriction on port value ranges, but
1006: // perform the same check as in setPort to be consistent. Pass
1007: // in a string to this method instead of an integer.
1008: if (port < -1 || port > 65535) {
1009: return false;
1010: }
1011:
1012: // Check that userinfo is well formed if it exists.
1013: if (userinfo != null) {
1014: // Userinfo can contain alphanumerics, mark characters, escaped
1015: // and ';',':','&','=','+','$',','
1016: int index = 0;
1017: int end = userinfo.length();
1018: char testChar = '\0';
1019: while (index < end) {
1020: testChar = userinfo.charAt(index);
1021: if (testChar == '%') {
1022: if (index + 2 >= end
1023: || !isHex(userinfo.charAt(index + 1))
1024: || !isHex(userinfo.charAt(index + 2))) {
1025: return false;
1026: }
1027: index += 2;
1028: } else if (!isUserinfoCharacter(testChar)) {
1029: return false;
1030: }
1031: ++index;
1032: }
1033: }
1034: return true;
1035: }
1036:
1037: /**
1038: * Determines whether the given string is a registry based authority.
1039: *
1040: * @param authority the authority component of a URI
1041: *
1042: * @return true if the given string is a registry based authority
1043: */
1044: private boolean isValidRegistryBasedAuthority(String authority) {
1045: int index = 0;
1046: int end = authority.length();
1047: char testChar;
1048:
1049: while (index < end) {
1050: testChar = authority.charAt(index);
1051:
1052: // check for valid escape sequence
1053: if (testChar == '%') {
1054: if (index + 2 >= end
1055: || !isHex(authority.charAt(index + 1))
1056: || !isHex(authority.charAt(index + 2))) {
1057: return false;
1058: }
1059: index += 2;
1060: }
1061: // can check against path characters because the set
1062: // is the same except for '/' which we've already excluded.
1063: else if (!isPathCharacter(testChar)) {
1064: return false;
1065: }
1066: ++index;
1067: }
1068: return true;
1069: }
1070:
1071: /**
1072: * Initialize the path for this URI from a URI string spec.
1073: *
1074: * @param p_uriSpec the URI specification (cannot be null)
1075: * @param p_nStartIndex the index to begin scanning from
1076: *
1077: * @exception MalformedURIException if p_uriSpec violates syntax rules
1078: */
1079: private void initializePath(String p_uriSpec, int p_nStartIndex)
1080: throws MalformedURIException {
1081: if (p_uriSpec == null) {
1082: throw new MalformedURIException(
1083: "Cannot initialize path from null string!");
1084: }
1085:
1086: int index = p_nStartIndex;
1087: int start = p_nStartIndex;
1088: int end = p_uriSpec.length();
1089: char testChar = '\0';
1090:
1091: // path - everything up to query string or fragment
1092: if (start < end) {
1093: // RFC 2732 only allows '[' and ']' to appear in the opaque part.
1094: if (getScheme() == null || p_uriSpec.charAt(start) == '/') {
1095:
1096: // Scan path.
1097: // abs_path = "/" path_segments
1098: // rel_path = rel_segment [ abs_path ]
1099: while (index < end) {
1100: testChar = p_uriSpec.charAt(index);
1101:
1102: // check for valid escape sequence
1103: if (testChar == '%') {
1104: if (index + 2 >= end
1105: || !isHex(p_uriSpec.charAt(index + 1))
1106: || !isHex(p_uriSpec.charAt(index + 2))) {
1107: throw new MalformedURIException(
1108: "Path contains invalid escape sequence!");
1109: }
1110: index += 2;
1111: }
1112: // Path segments cannot contain '[' or ']' since pchar
1113: // production was not changed by RFC 2732.
1114: else if (!isPathCharacter(testChar)) {
1115: if (testChar == '?' || testChar == '#') {
1116: break;
1117: }
1118: throw new MalformedURIException(
1119: "Path contains invalid character: "
1120: + testChar);
1121: }
1122: ++index;
1123: }
1124: } else {
1125:
1126: // Scan opaque part.
1127: // opaque_part = uric_no_slash *uric
1128: while (index < end) {
1129: testChar = p_uriSpec.charAt(index);
1130:
1131: if (testChar == '?' || testChar == '#') {
1132: break;
1133: }
1134:
1135: // check for valid escape sequence
1136: if (testChar == '%') {
1137: if (index + 2 >= end
1138: || !isHex(p_uriSpec.charAt(index + 1))
1139: || !isHex(p_uriSpec.charAt(index + 2))) {
1140: throw new MalformedURIException(
1141: "Opaque part contains invalid escape sequence!");
1142: }
1143: index += 2;
1144: }
1145: // If the scheme specific part is opaque, it can contain '['
1146: // and ']'. uric_no_slash wasn't modified by RFC 2732, which
1147: // I've interpreted as an error in the spec, since the
1148: // production should be equivalent to (uric - '/'), and uric
1149: // contains '[' and ']'. - mrglavas
1150: else if (!isURICharacter(testChar)) {
1151: throw new MalformedURIException(
1152: "Opaque part contains invalid character: "
1153: + testChar);
1154: }
1155: ++index;
1156: }
1157: }
1158: }
1159: m_path = p_uriSpec.substring(start, index);
1160:
1161: // query - starts with ? and up to fragment or end
1162: if (testChar == '?') {
1163: index++;
1164: start = index;
1165: while (index < end) {
1166: testChar = p_uriSpec.charAt(index);
1167: if (testChar == '#') {
1168: break;
1169: }
1170: if (testChar == '%') {
1171: if (index + 2 >= end
1172: || !isHex(p_uriSpec.charAt(index + 1))
1173: || !isHex(p_uriSpec.charAt(index + 2))) {
1174: throw new MalformedURIException(
1175: "Query string contains invalid escape sequence!");
1176: }
1177: index += 2;
1178: } else if (!isURICharacter(testChar)) {
1179: throw new MalformedURIException(
1180: "Query string contains invalid character: "
1181: + testChar);
1182: }
1183: index++;
1184: }
1185: m_queryString = p_uriSpec.substring(start, index);
1186: }
1187:
1188: // fragment - starts with #
1189: if (testChar == '#') {
1190: index++;
1191: start = index;
1192: while (index < end) {
1193: testChar = p_uriSpec.charAt(index);
1194:
1195: if (testChar == '%') {
1196: if (index + 2 >= end
1197: || !isHex(p_uriSpec.charAt(index + 1))
1198: || !isHex(p_uriSpec.charAt(index + 2))) {
1199: throw new MalformedURIException(
1200: "Fragment contains invalid escape sequence!");
1201: }
1202: index += 2;
1203: } else if (!isURICharacter(testChar)) {
1204: throw new MalformedURIException(
1205: "Fragment contains invalid character: "
1206: + testChar);
1207: }
1208: index++;
1209: }
1210: m_fragment = p_uriSpec.substring(start, index);
1211: }
1212: }
1213:
1214: /**
1215: * Get the scheme for this URI.
1216: *
1217: * @return the scheme for this URI
1218: */
1219: public String getScheme() {
1220: return m_scheme;
1221: }
1222:
1223: /**
1224: * Get the scheme-specific part for this URI (everything following the
1225: * scheme and the first colon). See RFC 2396 Section 5.2 for spec.
1226: *
1227: * @return the scheme-specific part for this URI
1228: */
1229: public String getSchemeSpecificPart() {
1230: StringBuffer schemespec = new StringBuffer();
1231:
1232: if (m_host != null || m_regAuthority != null) {
1233: schemespec.append("//");
1234:
1235: // Server based authority.
1236: if (m_host != null) {
1237:
1238: if (m_userinfo != null) {
1239: schemespec.append(m_userinfo);
1240: schemespec.append('@');
1241: }
1242:
1243: schemespec.append(m_host);
1244:
1245: if (m_port != -1) {
1246: schemespec.append(':');
1247: schemespec.append(m_port);
1248: }
1249: }
1250: // Registry based authority.
1251: else {
1252: schemespec.append(m_regAuthority);
1253: }
1254: }
1255:
1256: if (m_path != null) {
1257: schemespec.append((m_path));
1258: }
1259:
1260: if (m_queryString != null) {
1261: schemespec.append('?');
1262: schemespec.append(m_queryString);
1263: }
1264:
1265: if (m_fragment != null) {
1266: schemespec.append('#');
1267: schemespec.append(m_fragment);
1268: }
1269:
1270: return schemespec.toString();
1271: }
1272:
1273: /**
1274: * Get the userinfo for this URI.
1275: *
1276: * @return the userinfo for this URI (null if not specified).
1277: */
1278: public String getUserinfo() {
1279: return m_userinfo;
1280: }
1281:
1282: /**
1283: * Get the host for this URI.
1284: *
1285: * @return the host for this URI (null if not specified).
1286: */
1287: public String getHost() {
1288: return m_host;
1289: }
1290:
1291: /**
1292: * Get the port for this URI.
1293: *
1294: * @return the port for this URI (-1 if not specified).
1295: */
1296: public int getPort() {
1297: return m_port;
1298: }
1299:
1300: /**
1301: * Get the registry based authority for this URI.
1302: *
1303: * @return the registry based authority (null if not specified).
1304: */
1305: public String getRegBasedAuthority() {
1306: return m_regAuthority;
1307: }
1308:
1309: /**
1310: * Get the authority for this URI.
1311: *
1312: * @return the authority
1313: */
1314: public String getAuthority() {
1315: StringBuffer authority = new StringBuffer();
1316: if (m_host != null || m_regAuthority != null) {
1317: authority.append("//");
1318:
1319: // Server based authority.
1320: if (m_host != null) {
1321:
1322: if (m_userinfo != null) {
1323: authority.append(m_userinfo);
1324: authority.append('@');
1325: }
1326:
1327: authority.append(m_host);
1328:
1329: if (m_port != -1) {
1330: authority.append(':');
1331: authority.append(m_port);
1332: }
1333: }
1334: // Registry based authority.
1335: else {
1336: authority.append(m_regAuthority);
1337: }
1338: }
1339: return authority.toString();
1340: }
1341:
1342: /**
1343: * Get the path for this URI (optionally with the query string and
1344: * fragment).
1345: *
1346: * @param p_includeQueryString if true (and query string is not null),
1347: * then a "?" followed by the query string
1348: * will be appended
1349: * @param p_includeFragment if true (and fragment is not null),
1350: * then a "#" followed by the fragment
1351: * will be appended
1352: *
1353: * @return the path for this URI possibly including the query string
1354: * and fragment
1355: */
1356: public String getPath(boolean p_includeQueryString,
1357: boolean p_includeFragment) {
1358: StringBuffer pathString = new StringBuffer(m_path);
1359:
1360: if (p_includeQueryString && m_queryString != null) {
1361: pathString.append('?');
1362: pathString.append(m_queryString);
1363: }
1364:
1365: if (p_includeFragment && m_fragment != null) {
1366: pathString.append('#');
1367: pathString.append(m_fragment);
1368: }
1369: return pathString.toString();
1370: }
1371:
1372: /**
1373: * Get the path for this URI. Note that the value returned is the path
1374: * only and does not include the query string or fragment.
1375: *
1376: * @return the path for this URI.
1377: */
1378: public String getPath() {
1379: return m_path;
1380: }
1381:
1382: /**
1383: * Get the query string for this URI.
1384: *
1385: * @return the query string for this URI. Null is returned if there
1386: * was no "?" in the URI spec, empty string if there was a
1387: * "?" but no query string following it.
1388: */
1389: public String getQueryString() {
1390: return m_queryString;
1391: }
1392:
1393: /**
1394: * Get the fragment for this URI.
1395: *
1396: * @return the fragment for this URI. Null is returned if there
1397: * was no "#" in the URI spec, empty string if there was a
1398: * "#" but no fragment following it.
1399: */
1400: public String getFragment() {
1401: return m_fragment;
1402: }
1403:
1404: /**
1405: * Set the scheme for this URI. The scheme is converted to lowercase
1406: * before it is set.
1407: *
1408: * @param p_scheme the scheme for this URI (cannot be null)
1409: *
1410: * @exception MalformedURIException if p_scheme is not a conformant
1411: * scheme name
1412: */
1413: public void setScheme(String p_scheme) throws MalformedURIException {
1414: if (p_scheme == null) {
1415: throw new MalformedURIException(
1416: "Cannot set scheme from null string!");
1417: }
1418: if (!isConformantSchemeName(p_scheme)) {
1419: throw new MalformedURIException(
1420: "The scheme is not conformant.");
1421: }
1422:
1423: m_scheme = p_scheme.toLowerCase();
1424: }
1425:
1426: /**
1427: * Set the userinfo for this URI. If a non-null value is passed in and
1428: * the host value is null, then an exception is thrown.
1429: *
1430: * @param p_userinfo the userinfo for this URI
1431: *
1432: * @exception MalformedURIException if p_userinfo contains invalid
1433: * characters
1434: */
1435: public void setUserinfo(String p_userinfo)
1436: throws MalformedURIException {
1437: if (p_userinfo == null) {
1438: m_userinfo = null;
1439: return;
1440: } else {
1441: if (m_host == null) {
1442: throw new MalformedURIException(
1443: "Userinfo cannot be set when host is null!");
1444: }
1445:
1446: // userinfo can contain alphanumerics, mark characters, escaped
1447: // and ';',':','&','=','+','$',','
1448: int index = 0;
1449: int end = p_userinfo.length();
1450: char testChar = '\0';
1451: while (index < end) {
1452: testChar = p_userinfo.charAt(index);
1453: if (testChar == '%') {
1454: if (index + 2 >= end
1455: || !isHex(p_userinfo.charAt(index + 1))
1456: || !isHex(p_userinfo.charAt(index + 2))) {
1457: throw new MalformedURIException(
1458: "Userinfo contains invalid escape sequence!");
1459: }
1460: } else if (!isUserinfoCharacter(testChar)) {
1461: throw new MalformedURIException(
1462: "Userinfo contains invalid character:"
1463: + testChar);
1464: }
1465: index++;
1466: }
1467: }
1468: m_userinfo = p_userinfo;
1469: }
1470:
1471: /**
1472: * <p>Set the host for this URI. If null is passed in, the userinfo
1473: * field is also set to null and the port is set to -1.</p>
1474: *
1475: * <p>Note: This method overwrites registry based authority if it
1476: * previously existed in this URI.</p>
1477: *
1478: * @param p_host the host for this URI
1479: *
1480: * @exception MalformedURIException if p_host is not a valid IP
1481: * address or DNS hostname.
1482: */
1483: public void setHost(String p_host) throws MalformedURIException {
1484: if (p_host == null || p_host.length() == 0) {
1485: if (p_host != null) {
1486: m_regAuthority = null;
1487: }
1488: m_host = p_host;
1489: m_userinfo = null;
1490: m_port = -1;
1491: return;
1492: } else if (!isWellFormedAddress(p_host)) {
1493: throw new MalformedURIException(
1494: "Host is not a well formed address!");
1495: }
1496: m_host = p_host;
1497: m_regAuthority = null;
1498: }
1499:
1500: /**
1501: * Set the port for this URI. -1 is used to indicate that the port is
1502: * not specified, otherwise valid port numbers are between 0 and 65535.
1503: * If a valid port number is passed in and the host field is null,
1504: * an exception is thrown.
1505: *
1506: * @param p_port the port number for this URI
1507: *
1508: * @exception MalformedURIException if p_port is not -1 and not a
1509: * valid port number
1510: */
1511: public void setPort(int p_port) throws MalformedURIException {
1512: if (p_port >= 0 && p_port <= 65535) {
1513: if (m_host == null) {
1514: throw new MalformedURIException(
1515: "Port cannot be set when host is null!");
1516: }
1517: } else if (p_port != -1) {
1518: throw new MalformedURIException("Invalid port number!");
1519: }
1520: m_port = p_port;
1521: }
1522:
1523: /**
1524: * <p>Sets the registry based authority for this URI.</p>
1525: *
1526: * <p>Note: This method overwrites server based authority
1527: * if it previously existed in this URI.</p>
1528: *
1529: * @param authority the registry based authority for this URI
1530: *
1531: * @exception MalformedURIException it authority is not a
1532: * well formed registry based authority
1533: */
1534: public void setRegBasedAuthority(String authority)
1535: throws MalformedURIException {
1536:
1537: if (authority == null) {
1538: m_regAuthority = null;
1539: return;
1540: }
1541: // reg_name = 1*( unreserved | escaped | "$" | "," |
1542: // ";" | ":" | "@" | "&" | "=" | "+" )
1543: else if (authority.length() < 1
1544: || !isValidRegistryBasedAuthority(authority)
1545: || authority.indexOf('/') != -1) {
1546: throw new MalformedURIException(
1547: "Registry based authority is not well formed.");
1548: }
1549: m_regAuthority = authority;
1550: m_host = null;
1551: m_userinfo = null;
1552: m_port = -1;
1553: }
1554:
1555: /**
1556: * Set the path for this URI. If the supplied path is null, then the
1557: * query string and fragment are set to null as well. If the supplied
1558: * path includes a query string and/or fragment, these fields will be
1559: * parsed and set as well. Note that, for URIs following the "generic
1560: * URI" syntax, the path specified should start with a slash.
1561: * For URIs that do not follow the generic URI syntax, this method
1562: * sets the scheme-specific part.
1563: *
1564: * @param p_path the path for this URI (may be null)
1565: *
1566: * @exception MalformedURIException if p_path contains invalid
1567: * characters
1568: */
1569: public void setPath(String p_path) throws MalformedURIException {
1570: if (p_path == null) {
1571: m_path = null;
1572: m_queryString = null;
1573: m_fragment = null;
1574: } else {
1575: initializePath(p_path, 0);
1576: }
1577: }
1578:
1579: /**
1580: * Append to the end of the path of this URI. If the current path does
1581: * not end in a slash and the path to be appended does not begin with
1582: * a slash, a slash will be appended to the current path before the
1583: * new segment is added. Also, if the current path ends in a slash
1584: * and the new segment begins with a slash, the extra slash will be
1585: * removed before the new segment is appended.
1586: *
1587: * @param p_addToPath the new segment to be added to the current path
1588: *
1589: * @exception MalformedURIException if p_addToPath contains syntax
1590: * errors
1591: */
1592: public void appendPath(String p_addToPath)
1593: throws MalformedURIException {
1594: if (p_addToPath == null || p_addToPath.trim().length() == 0) {
1595: return;
1596: }
1597:
1598: if (!isURIString(p_addToPath)) {
1599: throw new MalformedURIException(
1600: "Path contains invalid character!");
1601: }
1602:
1603: if (m_path == null || m_path.trim().length() == 0) {
1604: if (p_addToPath.startsWith("/")) {
1605: m_path = p_addToPath;
1606: } else {
1607: m_path = "/" + p_addToPath;
1608: }
1609: } else if (m_path.endsWith("/")) {
1610: if (p_addToPath.startsWith("/")) {
1611: m_path = m_path.concat(p_addToPath.substring(1));
1612: } else {
1613: m_path = m_path.concat(p_addToPath);
1614: }
1615: } else {
1616: if (p_addToPath.startsWith("/")) {
1617: m_path = m_path.concat(p_addToPath);
1618: } else {
1619: m_path = m_path.concat("/" + p_addToPath);
1620: }
1621: }
1622: }
1623:
1624: /**
1625: * Set the query string for this URI. A non-null value is valid only
1626: * if this is an URI conforming to the generic URI syntax and
1627: * the path value is not null.
1628: *
1629: * @param p_queryString the query string for this URI
1630: *
1631: * @exception MalformedURIException if p_queryString is not null and this
1632: * URI does not conform to the generic
1633: * URI syntax or if the path is null
1634: */
1635: public void setQueryString(String p_queryString)
1636: throws MalformedURIException {
1637: if (p_queryString == null) {
1638: m_queryString = null;
1639: } else if (!isGenericURI()) {
1640: throw new MalformedURIException(
1641: "Query string can only be set for a generic URI!");
1642: } else if (getPath() == null) {
1643: throw new MalformedURIException(
1644: "Query string cannot be set when path is null!");
1645: } else if (!isURIString(p_queryString)) {
1646: throw new MalformedURIException(
1647: "Query string contains invalid character!");
1648: } else {
1649: m_queryString = p_queryString;
1650: }
1651: }
1652:
1653: /**
1654: * Set the fragment for this URI. A non-null value is valid only
1655: * if this is a URI conforming to the generic URI syntax and
1656: * the path value is not null.
1657: *
1658: * @param p_fragment the fragment for this URI
1659: *
1660: * @exception MalformedURIException if p_fragment is not null and this
1661: * URI does not conform to the generic
1662: * URI syntax or if the path is null
1663: */
1664: public void setFragment(String p_fragment)
1665: throws MalformedURIException {
1666: if (p_fragment == null) {
1667: m_fragment = null;
1668: } else if (!isGenericURI()) {
1669: throw new MalformedURIException(
1670: "Fragment can only be set for a generic URI!");
1671: } else if (getPath() == null) {
1672: throw new MalformedURIException(
1673: "Fragment cannot be set when path is null!");
1674: } else if (!isURIString(p_fragment)) {
1675: throw new MalformedURIException(
1676: "Fragment contains invalid character!");
1677: } else {
1678: m_fragment = p_fragment;
1679: }
1680: }
1681:
1682: /**
1683: * Determines if the passed-in Object is equivalent to this URI.
1684: *
1685: * @param p_test the Object to test for equality.
1686: *
1687: * @return true if p_test is a URI with all values equal to this
1688: * URI, false otherwise
1689: */
1690: public boolean equals(Object p_test) {
1691: if (p_test instanceof URI) {
1692: URI testURI = (URI) p_test;
1693: if (((m_scheme == null && testURI.m_scheme == null) || (m_scheme != null
1694: && testURI.m_scheme != null && m_scheme
1695: .equals(testURI.m_scheme)))
1696: && ((m_userinfo == null && testURI.m_userinfo == null) || (m_userinfo != null
1697: && testURI.m_userinfo != null && m_userinfo
1698: .equals(testURI.m_userinfo)))
1699: && ((m_host == null && testURI.m_host == null) || (m_host != null
1700: && testURI.m_host != null && m_host
1701: .equals(testURI.m_host)))
1702: && m_port == testURI.m_port
1703: && ((m_path == null && testURI.m_path == null) || (m_path != null
1704: && testURI.m_path != null && m_path
1705: .equals(testURI.m_path)))
1706: && ((m_queryString == null && testURI.m_queryString == null) || (m_queryString != null
1707: && testURI.m_queryString != null && m_queryString
1708: .equals(testURI.m_queryString)))
1709: && ((m_fragment == null && testURI.m_fragment == null) || (m_fragment != null
1710: && testURI.m_fragment != null && m_fragment
1711: .equals(testURI.m_fragment)))) {
1712: return true;
1713: }
1714: }
1715: return false;
1716: }
1717:
1718: /**
1719: * Get the URI as a string specification. See RFC 2396 Section 5.2.
1720: *
1721: * @return the URI string specification
1722: */
1723: public String toString() {
1724: StringBuffer uriSpecString = new StringBuffer();
1725:
1726: if (m_scheme != null) {
1727: uriSpecString.append(m_scheme);
1728: uriSpecString.append(':');
1729: }
1730: uriSpecString.append(getSchemeSpecificPart());
1731: return uriSpecString.toString();
1732: }
1733:
1734: /**
1735: * Get the indicator as to whether this URI uses the "generic URI"
1736: * syntax.
1737: *
1738: * @return true if this URI uses the "generic URI" syntax, false
1739: * otherwise
1740: */
1741: public boolean isGenericURI() {
1742: // presence of the host (whether valid or empty) means
1743: // double-slashes which means generic uri
1744: return (m_host != null);
1745: }
1746:
1747: /**
1748: * Returns whether this URI represents an absolute URI.
1749: *
1750: * @return true if this URI represents an absolute URI, false
1751: * otherwise
1752: */
1753: public boolean isAbsoluteURI() {
1754: // presence of the scheme means absolute uri
1755: return (m_scheme != null);
1756: }
1757:
1758: /**
1759: * Determine whether a scheme conforms to the rules for a scheme name.
1760: * A scheme is conformant if it starts with an alphanumeric, and
1761: * contains only alphanumerics, '+','-' and '.'.
1762: *
1763: * @return true if the scheme is conformant, false otherwise
1764: */
1765: public static boolean isConformantSchemeName(String p_scheme) {
1766: if (p_scheme == null || p_scheme.trim().length() == 0) {
1767: return false;
1768: }
1769:
1770: if (!isAlpha(p_scheme.charAt(0))) {
1771: return false;
1772: }
1773:
1774: char testChar;
1775: int schemeLength = p_scheme.length();
1776: for (int i = 1; i < schemeLength; ++i) {
1777: testChar = p_scheme.charAt(i);
1778: if (!isSchemeCharacter(testChar)) {
1779: return false;
1780: }
1781: }
1782:
1783: return true;
1784: }
1785:
1786: /**
1787: * Determine whether a string is syntactically capable of representing
1788: * a valid IPv4 address, IPv6 reference or the domain name of a network host.
1789: * A valid IPv4 address consists of four decimal digit groups separated by a
1790: * '.'. Each group must consist of one to three digits. See RFC 2732 Section 3,
1791: * and RFC 2373 Section 2.2, for the definition of IPv6 references. A hostname
1792: * consists of domain labels (each of which must begin and end with an alphanumeric
1793: * but may contain '-') separated & by a '.'. See RFC 2396 Section 3.2.2.
1794: *
1795: * @return true if the string is a syntactically valid IPv4 address,
1796: * IPv6 reference or hostname
1797: */
1798: public static boolean isWellFormedAddress(String address) {
1799: if (address == null) {
1800: return false;
1801: }
1802:
1803: int addrLength = address.length();
1804: if (addrLength == 0) {
1805: return false;
1806: }
1807:
1808: // Check if the host is a valid IPv6reference.
1809: if (address.startsWith("[")) {
1810: return isWellFormedIPv6Reference(address);
1811: }
1812:
1813: // Cannot start with a '.', '-', or end with a '-'.
1814: if (address.startsWith(".") || address.startsWith("-")
1815: || address.endsWith("-")) {
1816: return false;
1817: }
1818:
1819: // rightmost domain label starting with digit indicates IP address
1820: // since top level domain label can only start with an alpha
1821: // see RFC 2396 Section 3.2.2
1822: int index = address.lastIndexOf('.');
1823: if (address.endsWith(".")) {
1824: index = address.substring(0, index).lastIndexOf('.');
1825: }
1826:
1827: if (index + 1 < addrLength
1828: && isDigit(address.charAt(index + 1))) {
1829: return isWellFormedIPv4Address(address);
1830: } else {
1831: // hostname = *( domainlabel "." ) toplabel [ "." ]
1832: // domainlabel = alphanum | alphanum *( alphanum | "-" ) alphanum
1833: // toplabel = alpha | alpha *( alphanum | "-" ) alphanum
1834:
1835: // RFC 2396 states that hostnames take the form described in
1836: // RFC 1034 (Section 3) and RFC 1123 (Section 2.1). According
1837: // to RFC 1034, hostnames are limited to 255 characters.
1838: if (addrLength > 255) {
1839: return false;
1840: }
1841:
1842: // domain labels can contain alphanumerics and '-"
1843: // but must start and end with an alphanumeric
1844: char testChar;
1845: int labelCharCount = 0;
1846:
1847: for (int i = 0; i < addrLength; i++) {
1848: testChar = address.charAt(i);
1849: if (testChar == '.') {
1850: if (!isAlphanum(address.charAt(i - 1))) {
1851: return false;
1852: }
1853: if (i + 1 < addrLength
1854: && !isAlphanum(address.charAt(i + 1))) {
1855: return false;
1856: }
1857: labelCharCount = 0;
1858: } else if (!isAlphanum(testChar) && testChar != '-') {
1859: return false;
1860: }
1861: // RFC 1034: Labels must be 63 characters or less.
1862: else if (++labelCharCount > 63) {
1863: return false;
1864: }
1865: }
1866: }
1867: return true;
1868: }
1869:
1870: /**
1871: * <p>Determines whether a string is an IPv4 address as defined by
1872: * RFC 2373, and under the further constraint that it must be a 32-bit
1873: * address. Though not expressed in the grammar, in order to satisfy
1874: * the 32-bit address constraint, each segment of the address cannot
1875: * be greater than 255 (8 bits of information).</p>
1876: *
1877: * <p><code>IPv4address = 1*3DIGIT "." 1*3DIGIT "." 1*3DIGIT "." 1*3DIGIT</code></p>
1878: *
1879: * @return true if the string is a syntactically valid IPv4 address
1880: */
1881: public static boolean isWellFormedIPv4Address(String address) {
1882:
1883: int addrLength = address.length();
1884: char testChar;
1885: int numDots = 0;
1886: int numDigits = 0;
1887:
1888: // make sure that 1) we see only digits and dot separators, 2) that
1889: // any dot separator is preceded and followed by a digit and
1890: // 3) that we find 3 dots
1891: //
1892: // RFC 2732 amended RFC 2396 by replacing the definition
1893: // of IPv4address with the one defined by RFC 2373. - mrglavas
1894: //
1895: // IPv4address = 1*3DIGIT "." 1*3DIGIT "." 1*3DIGIT "." 1*3DIGIT
1896: //
1897: // One to three digits must be in each segment.
1898: for (int i = 0; i < addrLength; i++) {
1899: testChar = address.charAt(i);
1900: if (testChar == '.') {
1901: if ((i > 0 && !isDigit(address.charAt(i - 1)))
1902: || (i + 1 < addrLength && !isDigit(address
1903: .charAt(i + 1)))) {
1904: return false;
1905: }
1906: numDigits = 0;
1907: if (++numDots > 3) {
1908: return false;
1909: }
1910: } else if (!isDigit(testChar)) {
1911: return false;
1912: }
1913: // Check that that there are no more than three digits
1914: // in this segment.
1915: else if (++numDigits > 3) {
1916: return false;
1917: }
1918: // Check that this segment is not greater than 255.
1919: else if (numDigits == 3) {
1920: char first = address.charAt(i - 2);
1921: char second = address.charAt(i - 1);
1922: if (!(first < '2' || (first == '2' && (second < '5' || (second == '5' && testChar <= '5'))))) {
1923: return false;
1924: }
1925: }
1926: }
1927: return (numDots == 3);
1928: }
1929:
1930: /**
1931: * <p>Determines whether a string is an IPv6 reference as defined
1932: * by RFC 2732, where IPv6address is defined in RFC 2373. The
1933: * IPv6 address is parsed according to Section 2.2 of RFC 2373,
1934: * with the additional constraint that the address be composed of
1935: * 128 bits of information.</p>
1936: *
1937: * <p><code>IPv6reference = "[" IPv6address "]"</code></p>
1938: *
1939: * <p>Note: The BNF expressed in RFC 2373 Appendix B does not
1940: * accurately describe section 2.2, and was in fact removed from
1941: * RFC 3513, the successor of RFC 2373.</p>
1942: *
1943: * @return true if the string is a syntactically valid IPv6 reference
1944: */
1945: public static boolean isWellFormedIPv6Reference(String address) {
1946:
1947: int addrLength = address.length();
1948: int index = 1;
1949: int end = addrLength - 1;
1950:
1951: // Check if string is a potential match for IPv6reference.
1952: if (!(addrLength > 2 && address.charAt(0) == '[' && address
1953: .charAt(end) == ']')) {
1954: return false;
1955: }
1956:
1957: // Counter for the number of 16-bit sections read in the address.
1958: int[] counter = new int[1];
1959:
1960: // Scan hex sequence before possible '::' or IPv4 address.
1961: index = scanHexSequence(address, index, end, counter);
1962: if (index == -1) {
1963: return false;
1964: }
1965: // Address must contain 128-bits of information.
1966: else if (index == end) {
1967: return (counter[0] == 8);
1968: }
1969:
1970: if (index + 1 < end && address.charAt(index) == ':') {
1971: if (address.charAt(index + 1) == ':') {
1972: // '::' represents at least one 16-bit group of zeros.
1973: if (++counter[0] > 8) {
1974: return false;
1975: }
1976: index += 2;
1977: // Trailing zeros will fill out the rest of the address.
1978: if (index == end) {
1979: return true;
1980: }
1981: }
1982: // If the second character wasn't ':', in order to be valid,
1983: // the remainder of the string must match IPv4Address,
1984: // and we must have read exactly 6 16-bit groups.
1985: else {
1986: return (counter[0] == 6)
1987: && isWellFormedIPv4Address(address.substring(
1988: index + 1, end));
1989: }
1990: } else {
1991: return false;
1992: }
1993:
1994: // 3. Scan hex sequence after '::'.
1995: int prevCount = counter[0];
1996: index = scanHexSequence(address, index, end, counter);
1997:
1998: // We've either reached the end of the string, the address ends in
1999: // an IPv4 address, or it is invalid. scanHexSequence has already
2000: // made sure that we have the right number of bits.
2001: return (index == end)
2002: || (index != -1 && isWellFormedIPv4Address(address
2003: .substring((counter[0] > prevCount) ? index + 1
2004: : index, end)));
2005: }
2006:
2007: /**
2008: * Helper method for isWellFormedIPv6Reference which scans the
2009: * hex sequences of an IPv6 address. It returns the index of the
2010: * next character to scan in the address, or -1 if the string
2011: * cannot match a valid IPv6 address.
2012: *
2013: * @param address the string to be scanned
2014: * @param index the beginning index (inclusive)
2015: * @param end the ending index (exclusive)
2016: * @param counter a counter for the number of 16-bit sections read
2017: * in the address
2018: *
2019: * @return the index of the next character to scan, or -1 if the
2020: * string cannot match a valid IPv6 address
2021: */
2022: private static int scanHexSequence(String address, int index,
2023: int end, int[] counter) {
2024:
2025: char testChar;
2026: int numDigits = 0;
2027: int start = index;
2028:
2029: // Trying to match the following productions:
2030: // hexseq = hex4 *( ":" hex4)
2031: // hex4 = 1*4HEXDIG
2032: for (; index < end; ++index) {
2033: testChar = address.charAt(index);
2034: if (testChar == ':') {
2035: // IPv6 addresses are 128-bit, so there can be at most eight sections.
2036: if (numDigits > 0 && ++counter[0] > 8) {
2037: return -1;
2038: }
2039: // This could be '::'.
2040: if (numDigits == 0
2041: || ((index + 1 < end) && address
2042: .charAt(index + 1) == ':')) {
2043: return index;
2044: }
2045: numDigits = 0;
2046: }
2047: // This might be invalid or an IPv4address. If it's potentially an IPv4address,
2048: // backup to just after the last valid character that matches hexseq.
2049: else if (!isHex(testChar)) {
2050: if (testChar == '.' && numDigits < 4 && numDigits > 0
2051: && counter[0] <= 6) {
2052: int back = index - numDigits - 1;
2053: return (back >= start) ? back : (back + 1);
2054: }
2055: return -1;
2056: }
2057: // There can be at most 4 hex digits per group.
2058: else if (++numDigits > 4) {
2059: return -1;
2060: }
2061: }
2062: return (numDigits > 0 && ++counter[0] <= 8) ? end : -1;
2063: }
2064:
2065: /**
2066: * Determine whether a char is a digit.
2067: *
2068: * @return true if the char is betweeen '0' and '9', false otherwise
2069: */
2070: private static boolean isDigit(char p_char) {
2071: return p_char >= '0' && p_char <= '9';
2072: }
2073:
2074: /**
2075: * Determine whether a character is a hexadecimal character.
2076: *
2077: * @return true if the char is betweeen '0' and '9', 'a' and 'f'
2078: * or 'A' and 'F', false otherwise
2079: */
2080: private static boolean isHex(char p_char) {
2081: return (p_char <= 'f' && (fgLookupTable[p_char] & ASCII_HEX_CHARACTERS) != 0);
2082: }
2083:
2084: /**
2085: * Determine whether a char is an alphabetic character: a-z or A-Z
2086: *
2087: * @return true if the char is alphabetic, false otherwise
2088: */
2089: private static boolean isAlpha(char p_char) {
2090: return ((p_char >= 'a' && p_char <= 'z') || (p_char >= 'A' && p_char <= 'Z'));
2091: }
2092:
2093: /**
2094: * Determine whether a char is an alphanumeric: 0-9, a-z or A-Z
2095: *
2096: * @return true if the char is alphanumeric, false otherwise
2097: */
2098: private static boolean isAlphanum(char p_char) {
2099: return (p_char <= 'z' && (fgLookupTable[p_char] & MASK_ALPHA_NUMERIC) != 0);
2100: }
2101:
2102: /**
2103: * Determine whether a character is a reserved character:
2104: * ';', '/', '?', ':', '@', '&', '=', '+', '$', ',', '[', or ']'
2105: *
2106: * @return true if the string contains any reserved characters
2107: */
2108: private static boolean isReservedCharacter(char p_char) {
2109: return (p_char <= ']' && (fgLookupTable[p_char] & RESERVED_CHARACTERS) != 0);
2110: }
2111:
2112: /**
2113: * Determine whether a char is an unreserved character.
2114: *
2115: * @return true if the char is unreserved, false otherwise
2116: */
2117: private static boolean isUnreservedCharacter(char p_char) {
2118: return (p_char <= '~' && (fgLookupTable[p_char] & MASK_UNRESERVED_MASK) != 0);
2119: }
2120:
2121: /**
2122: * Determine whether a char is a URI character (reserved or
2123: * unreserved, not including '%' for escaped octets).
2124: *
2125: * @return true if the char is a URI character, false otherwise
2126: */
2127: private static boolean isURICharacter(char p_char) {
2128: return (p_char <= '~' && (fgLookupTable[p_char] & MASK_URI_CHARACTER) != 0);
2129: }
2130:
2131: /**
2132: * Determine whether a char is a scheme character.
2133: *
2134: * @return true if the char is a scheme character, false otherwise
2135: */
2136: private static boolean isSchemeCharacter(char p_char) {
2137: return (p_char <= 'z' && (fgLookupTable[p_char] & MASK_SCHEME_CHARACTER) != 0);
2138: }
2139:
2140: /**
2141: * Determine whether a char is a userinfo character.
2142: *
2143: * @return true if the char is a userinfo character, false otherwise
2144: */
2145: private static boolean isUserinfoCharacter(char p_char) {
2146: return (p_char <= 'z' && (fgLookupTable[p_char] & MASK_USERINFO_CHARACTER) != 0);
2147: }
2148:
2149: /**
2150: * Determine whether a char is a path character.
2151: *
2152: * @return true if the char is a path character, false otherwise
2153: */
2154: private static boolean isPathCharacter(char p_char) {
2155: return (p_char <= '~' && (fgLookupTable[p_char] & MASK_PATH_CHARACTER) != 0);
2156: }
2157:
2158: /**
2159: * Determine whether a given string contains only URI characters (also
2160: * called "uric" in RFC 2396). uric consist of all reserved
2161: * characters, unreserved characters and escaped characters.
2162: *
2163: * @return true if the string is comprised of uric, false otherwise
2164: */
2165: private static boolean isURIString(String p_uric) {
2166: if (p_uric == null) {
2167: return false;
2168: }
2169: int end = p_uric.length();
2170: char testChar = '\0';
2171: for (int i = 0; i < end; i++) {
2172: testChar = p_uric.charAt(i);
2173: if (testChar == '%') {
2174: if (i + 2 >= end || !isHex(p_uric.charAt(i + 1))
2175: || !isHex(p_uric.charAt(i + 2))) {
2176: return false;
2177: } else {
2178: i += 2;
2179: continue;
2180: }
2181: }
2182: if (isURICharacter(testChar)) {
2183: continue;
2184: } else {
2185: return false;
2186: }
2187: }
2188: return true;
2189: }
2190: }
|