001    /**
002     * Licensed to the Apache Software Foundation (ASF) under one or more
003     * contributor license agreements.  See the NOTICE file distributed with
004     * this work for additional information regarding copyright ownership.
005     * The ASF licenses this file to You under the Apache License, Version 2.0
006     * (the "License"); you may not use this file except in compliance with
007     * the License.  You may obtain a copy of the License at
008     *
009     *      http://www.apache.org/licenses/LICENSE-2.0
010     *
011     * Unless required by applicable law or agreed to in writing, software
012     * distributed under the License is distributed on an "AS IS" BASIS,
013     * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
014     * See the License for the specific language governing permissions and
015     * limitations under the License.
016     */
017    package org.apache.camel.support;
018    
019    import java.io.InputStream;
020    import java.util.Iterator;
021    import java.util.LinkedHashMap;
022    import java.util.Map;
023    import java.util.Scanner;
024    import java.util.regex.Matcher;
025    import java.util.regex.Pattern;
026    
027    import org.apache.camel.util.ObjectHelper;
028    
029    /**
030     * {@link org.apache.camel.Expression} to walk a {@link org.apache.camel.Message} XML body
031     * using an {@link java.util.Iterator}, which grabs the content between a XML start and end token.
032     * <p/>
033     * The message body must be able to convert to {@link java.io.InputStream} type which is used as stream
034     * to access the message body.
035     * <p/>
036     * Can be used to split big XML files.
037     * <p/>
038     * This implementation supports inheriting namespaces from a parent/root tag.
039     */
040    public class TokenXMLPairExpressionIterator extends TokenPairExpressionIterator {
041    
042        private static final Pattern NAMESPACE_PATTERN = Pattern.compile("xmlns(:\\w+|)=\\\"(.*?)\\\"");
043        private static final String SCAN_TOKEN_REGEX = "(\\s+.*?|)>";
044        protected final String inheritNamespaceToken;
045    
046        public TokenXMLPairExpressionIterator(String startToken, String endToken, String inheritNamespaceToken) {
047            super(startToken, endToken, true);
048            // namespace token is optional
049            this.inheritNamespaceToken = inheritNamespaceToken;
050    
051            // must be XML tokens
052            if (!startToken.startsWith("<") || !startToken.endsWith(">")) {
053                throw new IllegalArgumentException("Start token must be a valid XML token, was: " + startToken);
054            }
055            if (!endToken.startsWith("<") || !endToken.endsWith(">")) {
056                throw new IllegalArgumentException("End token must be a valid XML token, was: " + endToken);
057            }
058            if (inheritNamespaceToken != null && (!inheritNamespaceToken.startsWith("<") || !inheritNamespaceToken.endsWith(">"))) {
059                throw new IllegalArgumentException("Namespace token must be a valid XML token, was: " + inheritNamespaceToken);
060            }
061        }
062    
063        @Override
064        protected Iterator<?> createIterator(InputStream in, String charset) {
065            XMLTokenPairIterator iterator = new XMLTokenPairIterator(startToken, endToken, inheritNamespaceToken, in, charset);
066            iterator.init();
067            return iterator;
068        }
069    
070        /**
071         * Iterator to walk the input stream
072         */
073        static class XMLTokenPairIterator extends TokenPairIterator {
074    
075            private final Pattern startTokenPattern;
076            private final String scanEndToken;
077            private final String inheritNamespaceToken;
078            private Pattern inheritNamespaceTokenPattern;
079            private String rootTokenNamespaces;
080    
081            XMLTokenPairIterator(String startToken, String endToken, String inheritNamespaceToken, InputStream in, String charset) {
082                super(startToken, endToken, true, in, charset);
083    
084                // remove any ending > as we need to support attributes on the tags, so we need to use a reg exp pattern
085                String token = startToken.substring(0, startToken.length() - 1) + SCAN_TOKEN_REGEX;
086                this.startTokenPattern = Pattern.compile(token);
087                this.scanEndToken = endToken.substring(0, endToken.length() - 1) + SCAN_TOKEN_REGEX;
088                this.inheritNamespaceToken = inheritNamespaceToken;
089                if (inheritNamespaceToken != null) {
090                    token = inheritNamespaceToken.substring(0, inheritNamespaceToken.length() - 1) + SCAN_TOKEN_REGEX;
091                    // the namespaces on the parent tag can be in multi line, so we need to instruct the dot to support multilines
092                    this.inheritNamespaceTokenPattern = Pattern.compile(token, Pattern.MULTILINE | Pattern.DOTALL);
093                }
094            }
095    
096            @Override
097            void init() {
098                // use scan end token as delimiter which supports attributes/namespaces
099                this.scanner = new Scanner(in, charset).useDelimiter(scanEndToken);
100                // this iterator will do look ahead as we may have data
101                // after the last end token, which the scanner would find
102                // so we need to be one step ahead of the scanner
103                this.image = scanner.hasNext() ? (String) next(true) : null;
104            }
105    
106            @Override
107            String getNext(boolean first) {
108                String next = scanner.next();
109                if (next == null) {
110                    return null;
111                }
112    
113                // initialize inherited namespaces on first
114                if (first && inheritNamespaceToken != null) {
115                    rootTokenNamespaces = getNamespacesFromNamespaceToken(next);
116                }
117    
118                // make sure next is positioned at start token as we can have leading data
119                // or we reached EOL and there is no more start tags
120                Matcher matcher = startTokenPattern.matcher(next);
121                if (!matcher.find()) {
122                    return null;
123                } else {
124                    int index = matcher.start();
125                    next = next.substring(index);
126                }
127    
128                // build answer accordingly to whether namespaces should be inherited or not
129                StringBuilder sb = new StringBuilder();
130                if (inheritNamespaceToken != null && rootTokenNamespaces != null) {
131                    // append root namespaces to local start token
132                    String tag = ObjectHelper.before(next, ">");
133                    // grab the text
134                    String text = ObjectHelper.after(next, ">");
135                    // build result with inherited namespaces
136                    next = sb.append(tag).append(rootTokenNamespaces).append(">").append(text).append(endToken).toString();
137                } else {
138                    next = sb.append(next).append(endToken).toString();
139                }
140    
141                return next;
142            }
143    
144            private String getNamespacesFromNamespaceToken(String text) {
145                if (text == null) {
146                    return null;
147                }
148    
149                // grab the namespace tag
150                Matcher mat = inheritNamespaceTokenPattern.matcher(text);
151                if (mat.find()) {
152                    text = mat.group(0);
153                } else {
154                    // cannot find namespace tag
155                    return null;
156                }
157    
158                // find namespaces (there can be attributes mixed, so we should only grab the namespaces)
159                Map<String, String> namespaces = new LinkedHashMap<String, String>();
160                Matcher matcher = NAMESPACE_PATTERN.matcher(text);
161                while (matcher.find()) {
162                    String prefix = matcher.group(1);
163                    String url = matcher.group(2);
164                    if (ObjectHelper.isEmpty(prefix)) {
165                        prefix = "_DEFAULT_";
166                    } else {
167                        // skip leading :
168                        prefix = prefix.substring(1);
169                    }
170                    namespaces.put(prefix, url);
171                }
172    
173                // did we find any namespaces
174                if (namespaces.isEmpty()) {
175                    return null;
176                }
177    
178                // build namespace String
179                StringBuilder sb = new StringBuilder();
180                for (Map.Entry<String, String> entry : namespaces.entrySet()) {
181                    String key = entry.getKey();
182                    String value = entry.getValue();
183                    if ("_DEFAULT_".equals(key)) {
184                        sb.append(" xmlns=\"").append(value).append("\"");
185                    } else {
186                        sb.append(" xmlns:").append(key).append("=\"").append(value).append("\"");
187                    }
188                }
189    
190                return sb.toString();
191            }
192        }
193    
194    }