Detik Parser

I have some leisure time this Saturday so I think it'll be good if I wrote a useful program. So here it is, a program that will read an HTML source of Detik News website, parse all the titles, save all the titles in a list, and finally return the list. The title will be saved in an entity-like Java class so it'll be much easier to save it using ORM mechanism using Hibernate, etc.

Here's the NewsEntity.java:

[sourcecode language='java']
package org.gandhim.news.model;

/**
*
* @author gandhim
*/
public class NewsEntity {
private Long id;

private String datePublished;
private String link;
private String subTitle;
private String title;

public void setId(Long id) {
this.id = id;
}

public Long getId() {
return id;
}

public String getLink() {
return link;
}

public void setDatePublished(String datePublished) {
this.datePublished = datePublished;
}

public void setLink(String link) {
this.link = link;
}

public void setSubTitle(String subTitle) {
this.subTitle = subTitle;
}

public void setTitle(String title) {
this.title = title;
}

public NewsEntity(String datePublished, String link, String subTitle, String title) {
this.datePublished = datePublished;
this.link = link;
this.subTitle = subTitle;
this.title = title;
}

public String getDatePublished() {
return datePublished;
}

public String getSubTitle() {
return subTitle;
}

public String getTitle() {
return title;
}

}
[/sourcecode]

And this is the parser, DetikParser.java:

[sourcecode language='java']
package org.gandhim.news.parser;

import java.io.*;
import java.nio.*;
import java.net.*;
import java.util.List;
import java.util.LinkedList;
import org.gandhim.news.model.NewsEntity;

/**
*
* @author gandhim
*/
public class DetikParser {

private static List newsList = new LinkedList();
private static String url = "http://www.detik.com/indexberita/index.php?fuseaction=indeks.berita&idkanal=10";

public static List doParse() {
setupProxy();
return parseHtml(getRawHtml());
}

public static void setupProxy() {
System.setProperty("http.proxyHost", "your_proxy_server_ip");
System.setProperty("http.proxyPort", "8080");
}

public static String getRawHtml() {
String rawHtml = "";

try {
URL detik = new URL(url);

URLConnection detikConnection = detik.openConnection();

BufferedReader in = new BufferedReader(new InputStreamReader(detikConnection.getInputStream()));
String readLine;
while ((readLine = in.readLine()) != null) {
rawHtml += readLine + "\n";
}

} catch (Exception e) {
System.out.println("Error occurred: " + e.getMessage());
}

return rawHtml;
}

public static List parseHtml(String rawHtml) {
if (rawHtml == null || rawHtml.trim().equals("")) {
System.exit(-1);
}

String startNews = "namakanalindex";
int startPos, endPos;

startPos = rawHtml.indexOf(startNews);
startPos = rawHtml.indexOf("

", startPos) + "
".length();
endPos = rawHtml.indexOf("
", startPos);

String[] news = rawHtml.substring(startPos, endPos).trim().split("
• ");
  
  int awal, akhir;
  String datePublished, link, subTitle, title;
  for (int i = 1; i < news.length; i++) {
  datePublished = "";
  link = "";
  subTitle = "";
  title = "";
  
  // untuk tanggal publikasi
  awal = news[i].indexOf("") +
  "".length();
  akhir = news[i].indexOf("");
  datePublished = news[i].substring(awal, akhir);
  
  // untuk link
  awal = news[i].indexOf(" " akhir = news[i].indexOf("\" class=\"judulindex\">");
  link = news[i].substring(awal, akhir);
  
  // untuk subjudul (jika ada)
  awal = news[i].indexOf("") +
  "".length();
  akhir = news[i].indexOf("");
  
  if (akhir != -1) {
  subTitle = news[i].substring(awal, akhir);
  }
  
  // untuk judul
  awal = news[i].indexOf("class=\"judulindex\">") +
  "class=\"judulindex\">".length();
  akhir = news[i].indexOf("");
  title = news[i].substring(awal, akhir);
  NewsEntity aNews = new NewsEntity(datePublished, link, subTitle, title);
  
  // simpan ke list untuk ditampilkan di Readers
  newsList.add(aNews);
  }
  
  return newsList;
  }
  }
  [/sourcecode]
  
  Now you'll just need to write a code to store each of the NewsEntity from the list to a database. I suggest to use Hibernate.