michael@0: /**
michael@0:  * Copyright (c) 2012, Ben Fortuna
michael@0:  * All rights reserved.
michael@0:  *
michael@0:  * Redistribution and use in source and binary forms, with or without
michael@0:  * modification, are permitted provided that the following conditions
michael@0:  * are met:
michael@0:  *
michael@0:  *  o Redistributions of source code must retain the above copyright
michael@0:  * notice, this list of conditions and the following disclaimer.
michael@0:  *
michael@0:  *  o Redistributions in binary form must reproduce the above copyright
michael@0:  * notice, this list of conditions and the following disclaimer in the
michael@0:  * documentation and/or other materials provided with the distribution.
michael@0:  *
michael@0:  *  o Neither the name of Ben Fortuna nor the names of any other contributors
michael@0:  * may be used to endorse or promote products derived from this software
michael@0:  * without specific prior written permission.
michael@0:  *
michael@0:  * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
michael@0:  * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
michael@0:  * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
michael@0:  * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR
michael@0:  * CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL,
michael@0:  * EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
michael@0:  * PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR
michael@0:  * PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF
michael@0:  * LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING
michael@0:  * NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
michael@0:  * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
michael@0:  */
michael@0: package net.fortuna.ical4j.util;
michael@0: 
michael@0: import java.net.URI;
michael@0: import java.net.URISyntaxException;
michael@0: import java.util.regex.Pattern;
michael@0: 
michael@0: /**
michael@0:  * $Id$
michael@0:  *
michael@0:  * Created on 11/09/2005
michael@0:  *
michael@0:  * Utility methods for working with URIs.
michael@0:  * @author Ben Fortuna
michael@0:  */
michael@0: public final class Uris {
michael@0: 
michael@0:     /**
michael@0:      * URI Scheme used when relaxed parsing is enabled and the given input stream lead to an invalid URI.
michael@0:      */
michael@0:     public static final String INVALID_SCHEME = "net.fortunal.ical4j.invalid";
michael@0:     
michael@0:     private static final Pattern CID_PATTERN = Pattern.compile("(?i)^cid:.*");
michael@0:     private static final Pattern NOTES_CID_REPLACEMENT_PATTERN = Pattern.compile("[<>]");
michael@0: 
michael@0:     /**
michael@0:      * Constructor made private to enforce static nature.
michael@0:      */
michael@0:     private Uris() {
michael@0:     }
michael@0: 
michael@0:     /**
michael@0:      * Encodes the specified URI string using the UTF-8 charset. In the event that an exception is thrown, the specifed
michael@0:      * URI string is returned unmodified.
michael@0:      * @param s a URI string
michael@0:      * @return an encoded URI string
michael@0:      */
michael@0:     public static String encode(final String s) {
michael@0:         /*
michael@0:          * try { return URLEncoder.encode(s, ENCODING_CHARSET); } catch (UnsupportedEncodingException use) {
michael@0:          * log.error("Error ocurred encoding URI [" + s + "]", use); }
michael@0:          */
michael@0: 
michael@0:         /*
michael@0:          * Lotus Notes does not correctly strip angle brackets from cid uris. From RFC2392: A "cid" URL is converted to
michael@0:          * the corresponding Content-ID message header [MIME] by removing the "cid:" prefix, converting the % encoded
michael@0:          * character to their equivalent US-ASCII characters, and enclosing the remaining parts with an angle bracket
michael@0:          * pair, "<" and ">". For example, "cid:foo4%25foo1@bar.net" corresponds to Content-ID: <foo4%25foo1@bar.net>
michael@0:          * Reversing the process and converting URL special characters to their % encodings produces the original cid. A
michael@0:          * "mid" URL is converted to a Message-ID or Message-ID/Content-ID pair in a similar fashion.
michael@0:          */
michael@0:         if (CompatibilityHints.isHintEnabled(CompatibilityHints.KEY_NOTES_COMPATIBILITY)
michael@0:         		&& CID_PATTERN.matcher(s).matches()) {
michael@0:         	
michael@0:             return NOTES_CID_REPLACEMENT_PATTERN.matcher(s).replaceAll("");
michael@0:         }
michael@0:         return s;
michael@0:     }
michael@0: 
michael@0:     /**
michael@0:      * Decodes the specified URI string using the UTF-8 charset. In the event that an exception is thrown, the specifed
michael@0:      * URI string is returned unmodified.
michael@0:      * @param s a URI string
michael@0:      * @return an encoded URI string
michael@0:      */
michael@0:     public static String decode(final String s) {
michael@0:         /*
michael@0:          * try { return URLDecoder.decode(s, ENCODING_CHARSET); } catch (UnsupportedEncodingException use) {
michael@0:          * log.error("Error ocurred decoding URI [" + s + "]", use); }
michael@0:          */
michael@0:         return s;
michael@0:     }
michael@0: 
michael@0:     /**
michael@0:      * Attempts to create a URI instance and will optionally swallow any resulting URISyntaxException depending on
michael@0:      * configured {@link CompatibilityHints}. Will also automatically attempt encoding of the string representation for
michael@0:      * greater compatibility.
michael@0:      * <p>When relaxed parsing is enabled and if the string representation is not valid, a second URI creation attempt is made
michael@0:      *  by extracting the scheme from the scheme specific part and URI encoding that later part. For example,
michael@0:      *  "mailto: joe smith@example.com" becomes "mailto:joe%20smith@example.com".<p>
michael@0:      * <p>If the second attempts also leads to a {@code URISyntaxException}, an opaque URI is constructed with a scheme
michael@0:      *  of {@code Uris.INVALID_SCHEME} and a value corresponding to the initial representation.</p>
michael@0:      * 
michael@0:      * @param s a string representation of a URI.
michael@0:      * @return a URI instance, which may not correspond to the URI string if a valid
michael@0:      *  URI string is not specified and relaxed parsing is enabled.
michael@0:      * @throws URISyntaxException if a valid URI string is not specified and relaxed parsing is disabled
michael@0:      */
michael@0:     public static URI create(final String s) throws URISyntaxException {
michael@0:         try {
michael@0:             return new URI(encode(s));
michael@0:         }
michael@0:         catch (URISyntaxException use) {
michael@0:             if (CompatibilityHints
michael@0:                     .isHintEnabled(CompatibilityHints.KEY_RELAXED_PARSING)) {
michael@0:                 String encoded = encode(s);
michael@0:                 int index = encoded.indexOf(':');
michael@0:                 if (index != -1 && index < encoded.length() -1) {
michael@0:                     try {
michael@0:                         return new URI(encoded.substring(0, index), encoded.substring(index + 1), null);
michael@0:                     } catch (URISyntaxException use2) {
michael@0:                     }
michael@0:                 }
michael@0:                 try {
michael@0:                     return new URI(INVALID_SCHEME, s, null);
michael@0:                 } catch (URISyntaxException use2) {
michael@0:                     // should not happen as we are building an opaque URI
michael@0:                     throw new IllegalArgumentException("Could not build URI from " + s);
michael@0:                 }
michael@0:             }
michael@0:             throw use;
michael@0:         }
michael@0:     }
michael@0: }