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.google.com/">Google</a> search engine.
042: * @author Justin Boitano
043: */
044: public class Google implements SearchEngine {
045:
046: static Pattern patCount = new Regexp(
047: "</b> of approximately <b>\\d+,?(\\d+)</b> for <b>");
048: static Pattern patNoHits = new Regexp(
049: "Your search did not produce any results");
050:
051: static Pattern patResult = new Tagexp(
052: "<p>(?{link}<a>(?{title})</a>)<font>" // title and main link
053: + "<BR>(?{description}.*?)<font color=green>" //description of link
054: );
055:
056: static Pattern patMoreLink = new Tagexp(
057: "<A HREF=/search?q=*><img><br><font>.*?</a>");
058:
059: /**
060: * Classify a page. Sets the following labels:
061: * <TABLE>
062: * <TR><TH>Name <TH>Type <TH>Meaning
063: * <TR><TD>searchengine.source <TD>Page label <TD>Google object that labeled the page
064: * <TR><TD>searchengine.count <TD>Page field <TD>Number of results on page
065: * <TR><TD>searchengine.results <TD>Page fields <TD>Array of results. Each result region
066: * contains subfields: rank, title, description, and link.
067: * <TR><TD>searchengine.more-results <TD>Link label <TD>Link to a page containing more results.
068: * </TABLE>
069: */
070: public void classify(Page page) {
071: String title = page.getTitle();
072: if (title != null && title.startsWith("Google Search:")) {
073: page.setObjectLabel("searchengine.source", this );
074:
075: Region count = patCount.oneMatch(page);
076:
077: if (count != null)
078: page
079: .setField("searchengine.count", count
080: .getField("0"));
081:
082: Region[] results = patResult.allMatches(page);
083: SearchEngineResult[] ser = new SearchEngineResult[results.length];
084:
085: for (int i = 0; i < results.length; ++i)
086: ser[i] = new SearchEngineResult(results[i]);
087: page.setFields("searchengine.results", ser);
088:
089: PatternMatcher m = patMoreLink.match(page);
090:
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 Google.
114: * @param keywords list of keywords, separated by spaces
115: * @return URL that submits the keywords to Google.
116: */
117: public URL makeQuery(String keywords) {
118: try {
119: return new URL("http://www.google.com/search?q="
120: + URLEncoder.encode(keywords));
121: } catch (MalformedURLException e) {
122: throw new RuntimeException("internal error");
123: }
124: }
125:
126: /**
127: * Get number of results per page for this search engine.
128: * @return typical number of results per page
129: */
130: public int getResultsPerPage() {
131: return 10;
132: }
133:
134: /**
135: * Search Google.
136: * @param keywords list of keywords, separated by spaces
137: * @return enumeration of SearchEngineResults returned by an Google query constructed from the keywords.
138: */
139: public static Search search(String keywords) {
140: return new Search(new Google(), keywords);
141: }
142:
143: /**
144: * Search Google.
145: * @param keywords list of keywords, separated by spaces
146: * @param maxResults maximum number of results to return
147: * @return enumeration of SearchEngineResults returned by an Google query constructed from the keywords.
148: * The enumeration yields at most maxResults objects.
149: */
150: public static Search search(String keywords, int maxResults) {
151: return new Search(new Google(), keywords, maxResults);
152: }
153: }
|