搜索引擎与信息检索教程袁津生电子教案部分源代码 示例程序.docVIP

9.6 示例代码 9.6.1 FrontierSchedulerForBjfu类 package org.archive.crawler.postprocessor; import org.archive.crawler.datamodel.CandidateURI; public class FrontierSchedulerForBjfu extends FrontierScheduler { public FrontierSchedulerForBjfu(String name){ super(name); } protected void schedule(CandidateURI caUri) { String uri = caUri.toString(); if(uri.indexOf(dns:) != -1){ getController().getFrontier().schedule(caUri); } else if(uri.indexOf(bjfu) != -1 (uri.indexOf(.html) != -1 || uri.indexOf(.htm) != -1 || uri.indexOf(.jsp) != -1 || uri.indexOf(.asp) != -1 || uri.indexOf(.aspx) != -1)){ System.out.println(uri); getController().getFrontier().schedule(caUri); } } } 9.6.2 Page类 package .bjfu.search.page; public class Page { private String url; private String title; private String summary; private String context; private int score; public Page(){ url = null; title = null; summary = null; context = null; score = 10; } public String getUrl(){ return this.url; } public void setUrl(String url){ this.url = url; } public String getTitle(){ return this.title; } public void setTitle(String title){ this.title = title; } public String getSummary(){ return this.summary; } public void setSummary(String summary){ this.summary = summary; } public String getContext(){ return this.context; } public void SetContext(String context){ this.context = context; } public int getScore(){ return this.score; } public void setScore(int score){ this.score = score; } } 9.6.3 Extractor类 package .bjfu.search.extractor; import org.htmlparser.*; import org.htmlparser.util.*; import org.htmlparser.visitors.*; import org.htmlparser.nodes.*; import org.htmlparser.tags.*; import .bjfu.search.page.*; import .bjfu.search.util.*; public class Extractor implements Runnable{ private String filename; private Parser parser; private Page page; private String encode; public void s