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.data;
michael@0: 
michael@0: import java.io.IOException;
michael@0: import java.io.InputStream;
michael@0: import java.io.InputStreamReader;
michael@0: import java.io.Reader;
michael@0: import java.io.StreamTokenizer;
michael@0: import java.net.URISyntaxException;
michael@0: import java.text.MessageFormat;
michael@0: import java.text.ParseException;
michael@0: 
michael@0: import net.fortuna.ical4j.model.Calendar;
michael@0: import net.fortuna.ical4j.model.Component;
michael@0: 
michael@0: import org.apache.commons.logging.Log;
michael@0: import org.apache.commons.logging.LogFactory;
michael@0: 
michael@0: /**
michael@0:  * <pre>
michael@0:  * $Id$
michael@0:  * 
michael@0:  *  Created [Nov 5, 2004]
michael@0:  * </pre>
michael@0:  *
michael@0:  * The default implementation of a calendar parser.
michael@0:  * @author Ben Fortuna
michael@0:  */
michael@0: public class CalendarParserImpl implements CalendarParser {
michael@0: 
michael@0:     private static final int WORD_CHAR_START = 32;
michael@0: 
michael@0:     private static final int WORD_CHAR_END = 255;
michael@0: 
michael@0:     private static final int WHITESPACE_CHAR_START = 0;
michael@0: 
michael@0:     private static final int WHITESPACE_CHAR_END = 20;
michael@0:     
michael@0:     private static final String UNEXPECTED_TOKEN_MESSAGE = "Expected [{0}], read [{1}]";
michael@0: 
michael@0:     private Log log = LogFactory.getLog(CalendarParserImpl.class);
michael@0: 
michael@0:     private final ComponentListParser componentListParser = new ComponentListParser();
michael@0: 
michael@0:     private final ComponentParser componentParser = new ComponentParser();
michael@0: 
michael@0:     private final PropertyListParser propertyListParser = new PropertyListParser();
michael@0: 
michael@0:     private final PropertyParser propertyParser = new PropertyParser();
michael@0: 
michael@0:     private final ParameterListParser paramListParser = new ParameterListParser();
michael@0: 
michael@0:     private final ParameterParser paramParser = new ParameterParser();
michael@0:     
michael@0:     /**
michael@0:      * {@inheritDoc}
michael@0:      */
michael@0:     public final void parse(final InputStream in, final ContentHandler handler)
michael@0:             throws IOException, ParserException {
michael@0:         parse(new InputStreamReader(in), handler);
michael@0:     }
michael@0: 
michael@0:     /**
michael@0:      * {@inheritDoc}
michael@0:      */
michael@0:     public final void parse(final Reader in, final ContentHandler handler)
michael@0:             throws IOException, ParserException {
michael@0: 
michael@0:         final StreamTokenizer tokeniser = new StreamTokenizer(in);
michael@0:         try {
michael@0:             tokeniser.resetSyntax();
michael@0:             tokeniser.wordChars(WORD_CHAR_START, WORD_CHAR_END);
michael@0:             tokeniser.whitespaceChars(WHITESPACE_CHAR_START,
michael@0:                     WHITESPACE_CHAR_END);
michael@0:             tokeniser.ordinaryChar(':');
michael@0:             tokeniser.ordinaryChar(';');
michael@0:             tokeniser.ordinaryChar('=');
michael@0:             tokeniser.ordinaryChar('\t');
michael@0:             tokeniser.eolIsSignificant(true);
michael@0:             tokeniser.whitespaceChars(0, 0);
michael@0:             tokeniser.quoteChar('"');
michael@0: 
michael@0:             // BEGIN:VCALENDAR
michael@0:             assertToken(tokeniser, in, Calendar.BEGIN);
michael@0: 
michael@0:             assertToken(tokeniser, in, ':');
michael@0: 
michael@0:             assertToken(tokeniser, in, Calendar.VCALENDAR, true);
michael@0: 
michael@0:             assertToken(tokeniser, in, StreamTokenizer.TT_EOL);
michael@0: 
michael@0:             handler.startCalendar();
michael@0: 
michael@0:             // parse calendar properties..
michael@0:             propertyListParser.parse(tokeniser, in, handler);
michael@0: 
michael@0:             // parse components..
michael@0:             componentListParser.parse(tokeniser, in, handler);
michael@0: 
michael@0:             // END:VCALENDAR
michael@0:             // assertToken(tokeniser,Calendar.END);
michael@0: 
michael@0:             assertToken(tokeniser, in, ':');
michael@0: 
michael@0:             assertToken(tokeniser, in, Calendar.VCALENDAR, true);
michael@0: 
michael@0:             handler.endCalendar();
michael@0:         }
michael@0:         catch (Exception e) {
michael@0: 
michael@0:             if (e instanceof IOException) {
michael@0:                 throw (IOException) e;
michael@0:             }
michael@0:             if (e instanceof ParserException) {
michael@0:                 throw (ParserException) e;
michael@0:             }
michael@0:             else {
michael@0:                 throw new ParserException(e.getMessage(), getLineNumber(tokeniser, in), e);
michael@0:             }
michael@0:         }
michael@0:     }
michael@0: 
michael@0:     /**
michael@0:      * Parses an iCalendar property list from the specified stream tokeniser.
michael@0:      * @param tokeniser
michael@0:      * @throws IOException
michael@0:      * @throws ParseException
michael@0:      * @throws URISyntaxException
michael@0:      * @throws URISyntaxException
michael@0:      * @throws ParserException
michael@0:      */
michael@0:     private class PropertyListParser {
michael@0:         
michael@0:         public void parse(final StreamTokenizer tokeniser, Reader in,
michael@0:                 final ContentHandler handler) throws IOException, ParseException,
michael@0:                 URISyntaxException, ParserException {
michael@0: 
michael@0:             assertToken(tokeniser, in, StreamTokenizer.TT_WORD);
michael@0: 
michael@0:             while (/*
michael@0:                      * !Component.BEGIN.equals(tokeniser.sval) &&
michael@0:                      */!Component.END.equals(tokeniser.sval)) {
michael@0:                 // check for timezones observances or vevent/vtodo alarms..
michael@0:                 if (Component.BEGIN.equals(tokeniser.sval)) {
michael@0:                     componentParser.parse(tokeniser, in, handler);
michael@0:                 }
michael@0:                 else {
michael@0:                     propertyParser.parse(tokeniser, in, handler);
michael@0:                 }
michael@3:                 absorbWhitespace(tokeniser, in);
michael@0:                 // assertToken(tokeniser, StreamTokenizer.TT_WORD);
michael@0:             }
michael@0:         }
michael@0:     }
michael@0: 
michael@0:     /**
michael@0:      * Parses an iCalendar property from the specified stream tokeniser.
michael@0:      * @param tokeniser
michael@0:      * @throws IOException
michael@0:      * @throws ParserException
michael@0:      * @throws URISyntaxException
michael@0:      * @throws ParseException
michael@0:      */
michael@0:     private class PropertyParser {
michael@0:         
michael@0:         private static final String PARSE_DEBUG_MESSAGE = "Property [{0}]";
michael@0:         
michael@0:         private static final String PARSE_EXCEPTION_MESSAGE = "Property [{0}]";
michael@0:         
michael@0:         private void parse(final StreamTokenizer tokeniser, Reader in,
michael@0:                 final ContentHandler handler) throws IOException, ParserException,
michael@0:                 URISyntaxException, ParseException {
michael@0: 
michael@0:             final String name = tokeniser.sval;
michael@0: 
michael@0:             // debugging..
michael@0:             if (log.isDebugEnabled()) {
michael@0:                 log.debug(MessageFormat.format(PARSE_DEBUG_MESSAGE, new Object[] {name}));
michael@0:             }
michael@0: 
michael@0:             handler.startProperty(name);
michael@0: 
michael@0:             paramListParser.parse(tokeniser, in, handler);
michael@0: 
michael@0:             // it appears that control tokens (ie. ':') are allowed
michael@0:             // after the first instance on a line is used.. as such
michael@0:             // we must continue appending to value until EOL is
michael@0:             // reached..
michael@0:             // assertToken(tokeniser, StreamTokenizer.TT_WORD);
michael@0: 
michael@0:             // String value = tokeniser.sval;
michael@0:             final StringBuffer value = new StringBuffer();
michael@0: 
michael@0:             // assertToken(tokeniser,StreamTokenizer.TT_EOL);
michael@0: 
michael@0:             // DQUOTE is ordinary char for property value
michael@0:             // From sec 4.3.11 of rfc-2445:
michael@0:             // text       = *(TSAFE-CHAR / ":" / DQUOTE / ESCAPED-CHAR)
michael@0:             //
michael@0:             tokeniser.ordinaryChar('"');
michael@3:             int nextToken = nextToken(tokeniser, in);
michael@0: 
michael@3:             while (nextToken != StreamTokenizer.TT_EOL) {
michael@0: 
michael@0:                 if (tokeniser.ttype == StreamTokenizer.TT_WORD) {
michael@0:                     value.append(tokeniser.sval);
michael@0:                 }
michael@0:                 else {
michael@0:                     value.append((char) tokeniser.ttype);
michael@0:                 }
michael@0: 
michael@3:                 nextToken = nextToken(tokeniser, in);
michael@0:             }
michael@0:             
michael@0:             // reset DQUOTE to be quote char
michael@0:             tokeniser.quoteChar('"');
michael@0:             
michael@0:             try {
michael@0:                 handler.propertyValue(value.toString());
michael@0:             }
michael@0:             catch (ParseException e) {
michael@0:                 final ParseException eNew = new ParseException("[" + name + "] "
michael@0:                         + e.getMessage(), e.getErrorOffset());
michael@0:                 eNew.initCause(e);
michael@0:                 throw eNew;
michael@0:             }
michael@0: 
michael@0:             handler.endProperty(name);
michael@0: 
michael@0:         }
michael@0:     }
michael@0: 
michael@0:     /**
michael@0:      * Parses a list of iCalendar parameters by parsing the specified stream tokeniser.
michael@0:      * @param tokeniser
michael@0:      * @throws IOException
michael@0:      * @throws ParserException
michael@0:      * @throws URISyntaxException
michael@0:      */
michael@0:     private class ParameterListParser {
michael@0:         
michael@0:         public void parse(final StreamTokenizer tokeniser, Reader in,
michael@0:                 final ContentHandler handler) throws IOException, ParserException,
michael@0:                 URISyntaxException {
michael@0: 
michael@3:             while (nextToken(tokeniser, in) == ';') {
michael@0:                 paramParser.parse(tokeniser, in, handler);
michael@0:             }
michael@0:         }
michael@0:     }
michael@0: 
michael@0:     /**
michael@0:      * @param tokeniser
michael@0:      * @param handler
michael@0:      * @throws IOException
michael@0:      * @throws ParserException
michael@0:      * @throws URISyntaxException
michael@0:      */
michael@0:     private class ParameterParser {
michael@0:         
michael@0:         private void parse(final StreamTokenizer tokeniser, Reader in,
michael@0:                 final ContentHandler handler) throws IOException, ParserException,
michael@0:                 URISyntaxException {
michael@0: 
michael@0:             assertToken(tokeniser, in, StreamTokenizer.TT_WORD);
michael@0: 
michael@0:             final String paramName = tokeniser.sval;
michael@0: 
michael@0:             // debugging..
michael@0:             if (log.isDebugEnabled()) {
michael@0:                 log.debug("Parameter [" + paramName + "]");
michael@0:             }
michael@0: 
michael@0:             assertToken(tokeniser, in, '=');
michael@0: 
michael@0:             final StringBuffer paramValue = new StringBuffer();
michael@0: 
michael@0:             // preserve quote chars..
michael@3:             if (nextToken(tokeniser, in) == '"') {
michael@0:                 paramValue.append('"');
michael@0:                 paramValue.append(tokeniser.sval);
michael@0:                 paramValue.append('"');
michael@0:             }
michael@0:             else if (tokeniser.sval != null) {
michael@0:                 paramValue.append(tokeniser.sval);
michael@0:                 // check for additional words to account for equals (=) in param-value
michael@3:                 int nextToken = nextToken(tokeniser, in);
michael@0: 
michael@0:                 while (nextToken != ';' && nextToken != ':' && nextToken != ',') {
michael@0: 
michael@0:                     if (tokeniser.ttype == StreamTokenizer.TT_WORD) {
michael@0:                         paramValue.append(tokeniser.sval);
michael@0:                     }
michael@0:                     else {
michael@0:                     	paramValue.append((char) tokeniser.ttype);
michael@0:                     }
michael@0: 
michael@3:                     nextToken = nextToken(tokeniser, in);
michael@0:                 }
michael@0:                 tokeniser.pushBack();
michael@0:             } else if(tokeniser.sval == null) { 
michael@0:             	tokeniser.pushBack();
michael@0:             }
michael@0: 
michael@0:             try {
michael@0:                 handler.parameter(paramName, paramValue.toString());
michael@0:             }
michael@0:             catch (ClassCastException cce) {
michael@0:                 throw new ParserException("Error parsing parameter", getLineNumber(tokeniser, in), cce);
michael@0:             }
michael@0:         }
michael@0:     }
michael@0: 
michael@0:     /**
michael@0:      * Parses an iCalendar component list from the specified stream tokeniser.
michael@0:      * @param tokeniser
michael@0:      * @throws IOException
michael@0:      * @throws ParseException
michael@0:      * @throws URISyntaxException
michael@0:      * @throws ParserException
michael@0:      */
michael@0:     private class ComponentListParser {
michael@0:         
michael@0:         private void parse(final StreamTokenizer tokeniser, Reader in,
michael@0:                 final ContentHandler handler) throws IOException, ParseException,
michael@0:                 URISyntaxException, ParserException {
michael@0: 
michael@0:             while (Component.BEGIN.equals(tokeniser.sval)) {
michael@0:                 componentParser.parse(tokeniser, in, handler);
michael@3:                 absorbWhitespace(tokeniser, in);
michael@0:                 // assertToken(tokeniser, StreamTokenizer.TT_WORD);
michael@0:             }
michael@0:         }
michael@0:     }
michael@0: 
michael@0:     /**
michael@0:      * Parses an iCalendar component from the specified stream tokeniser.
michael@0:      * @param tokeniser
michael@0:      * @throws IOException
michael@0:      * @throws ParseException
michael@0:      * @throws URISyntaxException
michael@0:      * @throws ParserException
michael@0:      */
michael@0:     private class ComponentParser {
michael@0:         
michael@0:         private void parse(final StreamTokenizer tokeniser, Reader in,
michael@0:                 final ContentHandler handler) throws IOException, ParseException,
michael@0:                 URISyntaxException, ParserException {
michael@0: 
michael@0:             assertToken(tokeniser, in, ':');
michael@0: 
michael@0:             assertToken(tokeniser, in, StreamTokenizer.TT_WORD);
michael@0: 
michael@0:             final String name = tokeniser.sval;
michael@0: 
michael@0:             handler.startComponent(name);
michael@0: 
michael@0:             assertToken(tokeniser, in, StreamTokenizer.TT_EOL);
michael@0: 
michael@0:             propertyListParser.parse(tokeniser, in, handler);
michael@0: 
michael@0:             /*
michael@0:              * // a special case for VTIMEZONE component which contains
michael@0:              * // sub-components.. 
michael@0:              * if (Component.VTIMEZONE.equals(name)) {
michael@0:              *     parseComponentList(tokeniser, handler);
michael@0:              * }
michael@0:              * // VEVENT/VTODO components may optionally have embedded VALARM
michael@0:              * // components.. 
michael@0:              * else if ((Component.VEVENT.equals(name) || Component.VTODO.equals(name))
michael@0:              *         &amp;&amp; Component.BEGIN.equals(tokeniser.sval)) {
michael@0:              *     parseComponentList(tokeniser, handler);
michael@0:              * }
michael@0:              */
michael@0: 
michael@0:             assertToken(tokeniser, in, ':');
michael@0: 
michael@0:             assertToken(tokeniser, in, name);
michael@0: 
michael@0:             assertToken(tokeniser, in, StreamTokenizer.TT_EOL);
michael@0: 
michael@0:             handler.endComponent(name);
michael@0:         }
michael@0:     }
michael@0: 
michael@0:     /**
michael@0:      * Asserts that the next token in the stream matches the specified token.
michael@0:      * @param tokeniser stream tokeniser to perform assertion on
michael@0:      * @param token expected token
michael@0:      * @throws IOException when unable to read from stream
michael@0:      * @throws ParserException when next token in the stream does not match the expected token
michael@0:      */
michael@0:     private void assertToken(final StreamTokenizer tokeniser, Reader in, final int token)
michael@0:             throws IOException, ParserException {
michael@0: 
michael@3:         if (nextToken(tokeniser, in) != token) {
michael@0:             throw new ParserException(MessageFormat.format(UNEXPECTED_TOKEN_MESSAGE, new Object[] {
michael@0:                     new Integer(token), new Integer(tokeniser.ttype),
michael@0:             }), getLineNumber(tokeniser, in));
michael@0:         }
michael@0: 
michael@0:         if (log.isDebugEnabled()) {
michael@0:             log.debug("[" + token + "]");
michael@0:         }
michael@0:     }
michael@0: 
michael@0:     /**
michael@0:      * Asserts that the next token in the stream matches the specified token. This method is case-sensitive.
michael@0:      * @param tokeniser
michael@0:      * @param token
michael@0:      * @throws IOException
michael@0:      * @throws ParserException
michael@0:      */
michael@0:     private void assertToken(final StreamTokenizer tokeniser, Reader in, final String token)
michael@0:             throws IOException, ParserException {
michael@0:         assertToken(tokeniser, in, token, false);
michael@0:     }
michael@0: 
michael@0:     /**
michael@0:      * Asserts that the next token in the stream matches the specified token.
michael@0:      * @param tokeniser stream tokeniser to perform assertion on
michael@0:      * @param token expected token
michael@0:      * @throws IOException when unable to read from stream
michael@0:      * @throws ParserException when next token in the stream does not match the expected token
michael@0:      */
michael@0:     private void assertToken(final StreamTokenizer tokeniser, Reader in,
michael@0:             final String token, final boolean ignoreCase) throws IOException,
michael@0:             ParserException {
michael@0: 
michael@0:         // ensure next token is a word token..
michael@0:         assertToken(tokeniser, in, StreamTokenizer.TT_WORD);
michael@0: 
michael@0:         if (ignoreCase) {
michael@0:             if (!token.equalsIgnoreCase(tokeniser.sval)) {
michael@0:                 throw new ParserException(MessageFormat.format(UNEXPECTED_TOKEN_MESSAGE, new Object[] {
michael@0:                         token, tokeniser.sval,
michael@0:                 }), getLineNumber(tokeniser, in));
michael@0:             }
michael@0:         }
michael@0:         else if (!token.equals(tokeniser.sval)) {
michael@0:             throw new ParserException(MessageFormat.format(UNEXPECTED_TOKEN_MESSAGE, new Object[] {
michael@0:                     token, tokeniser.sval,
michael@0:             }), getLineNumber(tokeniser, in));
michael@0:         }
michael@0: 
michael@0:         if (log.isDebugEnabled()) {
michael@0:             log.debug("[" + token + "]");
michael@0:         }
michael@0:     }
michael@0: 
michael@0:     /**
michael@0:      * Absorbs extraneous newlines.
michael@0:      * @param tokeniser
michael@0:      * @throws IOException
michael@0:      */
michael@3:     private void absorbWhitespace(final StreamTokenizer tokeniser, Reader in) throws IOException, ParserException {
michael@0:         // HACK: absorb extraneous whitespace between components (KOrganizer)..
michael@3:         while (nextToken(tokeniser, in) == StreamTokenizer.TT_EOL) {
michael@0:             if (log.isTraceEnabled()) {
michael@0:                 log.trace("Absorbing extra whitespace..");
michael@0:             }
michael@0:         }
michael@0:         if (log.isTraceEnabled()) {
michael@0:             log.trace("Aborting: absorbing extra whitespace complete");
michael@0:         }
michael@0:     }
michael@0:     
michael@0:     /**
michael@0:      * @param tokeniser
michael@0:      * @param in
michael@0:      * @return
michael@0:      */
michael@0:     private int getLineNumber(StreamTokenizer tokeniser, Reader in) {
michael@0:         int line = tokeniser.lineno();
michael@0:         if (tokeniser.ttype == StreamTokenizer.TT_EOL) {
michael@0:             line -= 1;
michael@0:         }
michael@0:         if (in instanceof UnfoldingReader) {
michael@0:             // need to take unfolded lines into account
michael@0:             final int unfolded = ((UnfoldingReader) in).getLinesUnfolded();
michael@0:             line += unfolded;
michael@0:         }
michael@0:         return line;
michael@0:     }
michael@3: 
michael@3:     /**
michael@3:      * Reads the next token from the tokeniser.
michael@3:      * This method throws a ParseException when reading EOF.
michael@3:      * @param tokeniser
michael@3:      * @param in
michael@3:      * @return
michael@3:      * @throws ParseException When reading EOF.
michael@3:      */
michael@3:     private int nextToken(StreamTokenizer tokeniser, Reader in) throws IOException, ParserException {
michael@3:         int token = tokeniser.nextToken();
michael@3:         if (token == StreamTokenizer.TT_EOF) {
michael@3:             throw new ParserException("Unexpected end of file", getLineNumber(tokeniser, in));
michael@3:         }
michael@3:         return token;
michael@3:     }
michael@0: }