2 * Copyright 2005 the original author or authors.
4 * Licensed under the Apache License, Version 2.0 (the "License");
5 * you may not use this file except in compliance with the License.
6 * You may obtain a copy of the License at
8 * http://www.apache.org/licenses/LICENSE-2.0
10 * Unless required by applicable law or agreed to in writing, software
11 * distributed under the License is distributed on an "AS IS" BASIS,
12 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
13 * See the License for the specific language governing permissions and
14 * limitations under the License.
17 package org.wamblee.crawler.kiss.main;
19 import java.io.InputStream;
20 import java.util.ArrayList;
21 import java.util.Iterator;
22 import java.util.List;
24 import org.dom4j.Attribute;
25 import org.dom4j.Document;
26 import org.dom4j.DocumentException;
27 import org.dom4j.Element;
28 import org.dom4j.io.SAXReader;
29 import org.wamblee.conditions.AndCondition;
30 import org.wamblee.conditions.Condition;
31 import org.wamblee.conditions.PropertyRegexCondition;
32 import org.wamblee.crawler.kiss.guide.Program;
35 * Parse the configuration of desired programs.
37 * @author Erik Brakkee
39 class ProgramConfigurationParser {
40 private static final int DEFAULT_PRIORITY = 1;
42 // Configuration of interesting programs.
44 private static final String ELEM_PROGRAM = "program";
46 private static final String ELEM_PRIORITY = "priority";
48 private static final String ELEM_PATTERN = "match";
50 private static final String ELEM_ACTION = "action";
52 private static final String ELEM_CATEGORY = "category";
54 private static final String ACTION_NOTIFY = "notify";
56 private List<ProgramFilter> _filters;
58 ProgramConfigurationParser() {
63 * Parses the condition used to match the desired programs.
66 * Input stream to parse from.
69 void parse(InputStream aStream) {
70 List<ProgramFilter> filters = new ArrayList<ProgramFilter>();
72 SAXReader reader = new SAXReader();
73 Document document = reader.read(aStream);
75 Element root = document.getRootElement();
77 for (Iterator i = root.elementIterator(ELEM_PROGRAM); i.hasNext();) {
78 Element program = (Element) i.next();
80 Element categoryElem = program.element(ELEM_CATEGORY);
82 if (categoryElem != null) {
83 category = categoryElem.getText().trim();
86 Element actionElem = program.element(ELEM_ACTION);
87 int priority = DEFAULT_PRIORITY;
88 String priorityString = program.elementTextTrim(ELEM_PRIORITY);
89 if ( priorityString != null ) {
90 priority = Integer.valueOf(priorityString);
92 ProgramAction action = new RecordProgramAction(priority);
93 if (actionElem != null) {
94 if (actionElem.getText().equals(ACTION_NOTIFY)) {
95 action = new InterestingProgramAction(category);
99 List<Condition<Program>> regexConditions = new ArrayList<Condition<Program>>();
100 for (Iterator j = program.elementIterator(ELEM_PATTERN); j
102 Element patternElem = (Element) j.next();
103 String fieldName = "name";
104 Attribute fieldAttribute = patternElem.attribute("field");
105 if (fieldAttribute != null) {
106 fieldName = fieldAttribute.getText();
108 String pattern = ".*(" + patternElem.getText() + ").*";
109 regexConditions.add(new PropertyRegexCondition<Program>(
110 fieldName, pattern, true));
112 Condition<Program> condition = new AndCondition<Program>(
114 filters.add(new ProgramFilter(condition, action));
117 } catch (DocumentException e) {
118 throw new RuntimeException("Error parsing program configuraiton", e);
123 * Returns the list of program filters.
125 * @return Filter list.
127 public List<ProgramFilter> getFilters() {