import java.lang.String;

import java.util.ArrayList;

import java.util.Map;

import java.util.HashMap;

import java.io.FileInputStream;

import java.io.InputStreamReader;

import java.io.BufferedReader;

import java.io.FileNotFoundException;

import java.io.IOException;

import java.nio.charset.Charset;

import java.sql.Connection;

import java.sql.DriverManager;

import java.sql.ResultSet;

import java.sql.Statement;

import javax.xml.stream.XMLInputFactory;

import javax.xml.stream.XMLStreamConstants;

import javax.xml.stream.XMLStreamReader;

// Question Analysis

// To deal with large data in XML format, StAX is the best way to read file

// If we want to do some join between different file, we need to store the parsed data.

// As the data is very large, my thought is to use database.

// But if we have a huge memory and don't need to worry about heap limitation, the hashMap will be a good choice. ("memory mode")

// I am using SQLite, which is embedded into the program and is easy to setup.

// Since using DB, we don't need class for user and poster anymore.

// I still remained the User and Post class, in order to quick change to "memory mode" when we have the ability.

// Also, we don't need to maintain the arrays to store the result.

// Database Table

// QUESTION TABLE

// ID INT PRIMARY KEY NOT NULL

// POSTTYPEID INT NOT NULL

// OWNERUSERID INT NOT NULL

//

// ANSWER TABLE

// ID INT PRIMARY KEY NOT NULL

// POSTTYPEID INT NOT NULL

// OWNERUSERID INT NOT NULL

//

// USER TABLE

// ID PRIMARY KEY NOT NULL

// USERNAME VCHAR NOT NULL

// Future Improvement

// 1.

// Make the table more subtle

// For example: we can split question/answer table into sub tables by owneruserid

// Owneruserid start with the same digit go to one table.

// Then when we select from the question/answer table, the table size will be smaller.

// 2.

// Rewrite the sql to make it more effective.

// 2.

// Redesign database table, make it more effective

// For example1: for question and answer table, we can use owneruserid as primary key

// Precalculate the user's post numbers and join with user table to get the username

// (Like the HashMap design in "memory mode")

// 3.

// Change to more effective database, like mySql or even Oracle, SQL server

// 4.

// Use Hadoop or similar implementations to design MapReduce programming mode

// Make the program run parallel and distributed on a cluster

// Use HBase or similar distribute database to store data

public class ParseXml {

}