001: /*
002: * WebSphinx web-crawling toolkit
003: *
004: * Copyright (c) 1998-2002 Carnegie Mellon University. All rights
005: * reserved.
006: *
007: * Redistribution and use in source and binary forms, with or without
008: * modification, are permitted provided that the following conditions
009: * are met:
010: *
011: * 1. Redistributions of source code must retain the above copyright
012: * notice, this list of conditions and the following disclaimer.
013: *
014: * 2. Redistributions in binary form must reproduce the above copyright
015: * notice, this list of conditions and the following disclaimer in
016: * the documentation and/or other materials provided with the
017: * distribution.
018: *
019: * THIS SOFTWARE IS PROVIDED BY CARNEGIE MELLON UNIVERSITY ``AS IS'' AND
020: * ANY EXPRESSED OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO,
021: * THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
022: * PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL CARNEGIE MELLON UNIVERSITY
023: * NOR ITS EMPLOYEES BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
024: * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
025: * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
026: * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
027: * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
028: * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
029: * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
030: *
031: */
032:
033: package websphinx.searchengine;
034:
035: import websphinx.*;
036: import java.net.URL;
037: import java.net.URLEncoder;
038: import java.net.MalformedURLException;
039:
040: /**
041: * <A href="http://www.metacrawler.com/">MetaCrawler</a> search engine.
042: */
043: public class MetaCrawler implements SearchEngine {
044:
045: static Pattern patCount = new Regexp(
046: "Collated Results: 1 to \\d+ of (\\d+) references");
047: static Pattern patNoHits = new Regexp(
048: "Your search did not produce any results");
049:
050: static Pattern patResult = new Tagexp(
051: "<dt><font color=#000000><b>(?{relevance})</b></font>" // relevance rating
052: + "(?{link}(?{title}<a>.*?</a>))" // title and main link
053: + "(?{description}<dt>.*?<font>)" // description
054: );
055:
056: //static Pattern patMoreLink = new Regexp (
057: // "<a href=\"http://\\w+.metacrawler.com/crawler\\?general.*?\">\\d+</a>"
058: //);
059: static Pattern patMoreLink = new Tagexp(
060: "<a href=http://*.metacrawler.com/crawler\\?general*></a>");
061:
062: /**
063: * Classify a page. Sets the following labels:
064: * <TABLE>
065: * <TR><TH>Name <TH>Type <TH>Meaning
066: * <TR><TD>searchengine.source <TD>Page label <TD>MetaCrawler object that labeled the page
067: * <TR><TD>searchengine.count <TD>Page field <TD>Number of results on page
068: * <TR><TD>searchengine.results <TD>Page fields <TD>Array of results. Each result region
069: * contains subfields: rank, title, description, and link.
070: * <TR><TD>searchengine.more-results <TD>Link label <TD>Link to a page containing more results.
071: * </TABLE>
072: */
073: public void classify(Page page) {
074: String title = page.getTitle();
075: if (title != null && title.startsWith("Metacrawler query:")) {
076: page.setObjectLabel("searchengine.source", this );
077:
078: Region count = patCount.oneMatch(page);
079: if (count != null)
080: page
081: .setField("searchengine.count", count
082: .getField("0"));
083:
084: Region[] results = patResult.allMatches(page);
085: SearchEngineResult[] ser = new SearchEngineResult[results.length];
086: for (int i = 0; i < results.length; ++i)
087: ser[i] = new SearchEngineResult(results[i]);
088: page.setFields("searchengine.results", ser);
089:
090: PatternMatcher m = patMoreLink.match(page);
091: while (m.hasMoreElements()) {
092: Link link = (Link) m.nextMatch();
093: link.setLabel("searchengine.more-results");
094: link.setLabel("hyperlink");
095: }
096: }
097: }
098:
099: /**
100: * Priority of this classifier.
101: */
102: public static final float priority = 0.0F;
103:
104: /**
105: * Get priority of this classifier.
106: * @return priority.
107: */
108: public float getPriority() {
109: return priority;
110: }
111:
112: /**
113: * Make a query URL for MetaCrawler.
114: * @param keywords list of keywords, separated by spaces
115: * @return URL that submits the keywords to MetaCrawler.
116: */
117: public URL makeQuery(String keywords) {
118: try {
119: return new URL(
120: "http://www.metacrawler.com/crawler?general="
121: + URLEncoder.encode(keywords)
122: + "&method=1&format=1®ion=&rpp=20&timeout=15&hpe=10");
123: } catch (MalformedURLException e) {
124: throw new RuntimeException("internal error");
125: }
126: }
127:
128: /**
129: * Get number of results per page for this search engine.
130: * @return typical number of results per page
131: */
132: public int getResultsPerPage() {
133: return 20;
134: }
135:
136: /**
137: * Search MetaCrawler.
138: * @param keywords list of keywords, separated by spaces
139: * @return enumeration of SearchEngineResults returned by an MetaCrawler query constructed from the keywords.
140: */
141: public static Search search(String keywords) {
142: return new Search(new MetaCrawler(), keywords);
143: }
144:
145: /**
146: * Search MetaCrawler.
147: * @param keywords list of keywords, separated by spaces
148: * @param maxResults maximum number of results to return
149: * @return enumeration of SearchEngineResults returned by an MetaCrawler query constructed from the keywords.
150: * The enumeration yields at most maxResults objects.
151: */
152: public static Search search(String keywords, int maxResults) {
153: return new Search(new MetaCrawler(), keywords, maxResults);
154: }
155: }
|