1、從mysql讀取海量數據,將其寫入到文件中,並拆分文件進行排序。下面是測試用例。代碼展示:
/**
* Created by xxx on 2016/8/29.
*/
import java.io.BufferedReader;
import java.io.BufferedWriter;
import java.io.FileNotFoundException;
import java.io.FileReader;
import java.io.FileWriter;
import java.io.IOException;
import java.util.Collections;
import java.util.Iterator;
import java.util.LinkedList;
import java.util.Random;
public class CombinationSortWrite {
/**
* 大數據排序合併
* 從msql數據庫讀取數據,寫入文件,並將文件拆分
*/
public static void main(String[] args) throws IOException {
// 寫入文件的路徑
String filePath = "D:\\456";
// 切分文件的路徑
String sqlitFilePath = "D:\\456\\123";
//數據的個數
int CountNumbers = 1000000;
//子文件的個數
int CountFile = 10;
//精度
int countAccuracy = 30*CountFile;
long startNumber=System.currentTimeMillis();
// 寫入大數據文件
WriteData(filePath,CountNumbers);
System.out.println("存儲完畢");
// 將大數據文件切分到另外的十個小文件中
sqlitFileDate(filePath, sqlitFilePath,CountFile);
System.out.println("文件切割完畢!");
// 把每個文件的數據進行排序
singleFileDataSort(sqlitFilePath,CountFile);
System.out.println("每個子文件排序完畢!");
//精度調整,十個文件數據進行比較整合
deathDataFile(filePath,sqlitFilePath,countAccuracy,CountFile);
System.out.println("整合完畢");
long stopNumber=System.currentTimeMillis();
System.out.println("耗時"+(stopNumber-startNumber)/1000+"毫秒");
}
// 寫入大數據文件
public static void WriteData(String path,int CountNumbers) throws IOException {
path = path + "\\12114.txt";
FileWriter fs = new FileWriter(path);
BufferedWriter fw = new BufferedWriter(fs);
for (int i = 0; i < CountNumbers; i++) {
fw.write(new Random().nextInt(Integer.MAX_VALUE) + "\r\n");
}
fw.close();
fs.close();
}
// 將大數據文件切分到另外的十個小文件中
public static void sqlitFileDate(String filepath, String sqlitPath,
int CountFile) throws IOException {
FileWriter fs = null;
BufferedWriter fw=null;
FileReader fr = new FileReader(filepath + "\\12114.txt");
BufferedReader br = new BufferedReader(fr); // 讀取獲取整行數據
int i = 1;
LinkedList WriterLists=new LinkedList(); //初始化文件流對象集合
LinkedList fwLists=new LinkedList();
for (int j = 1; j <= CountFile; j++) {
//聲明對象
fs = new FileWriter(sqlitPath + "\\12" + j + ".txt",false);
fw=new BufferedWriter(fs);
//將對象裝入集合
WriterLists.add(fs);
fwLists.add(fw);
}
//判斷是文件流中是否還有數據返回
while (br.ready()) {
int count=1;//初始化第一文件流
for (Iterator iterator = fwLists.iterator(); iterator.hasNext();) {
BufferedWriter type = (BufferedWriter) iterator.next();
if(i==count)//判斷輪到第幾個文件流寫入數據了
{
//寫入數據,跳出,進行下一個文件流,下一個數據的寫入
type.write(br.readLine() + "\r\n");
break;
}
count++;
}
//判斷是否到了最後一個文件流了
if (i >= CountFile) {
i = 1;
} else
i++;
}
br.close();
fr.close();
for (Iterator iterator = fwLists.iterator(); iterator.hasNext();) {
BufferedWriter object = (BufferedWriter) iterator.next();
object.close();
}
//遍歷關閉所有子文件流
for (Iterator iterator = WriterLists.iterator(); iterator.hasNext();) {
FileWriter object = (FileWriter) iterator.next();
object.close();
}
}
// 把每個文件的數據進行排序
public static void singleFileDataSort(String path1,int CountFile) throws IOException {
LinkedList nums = null;
for (int i = 1; i <= CountFile; i++) {
nums = new LinkedList();
String path = path1 + "\\12" + i + ".txt";
try {
FileReader fr = new FileReader(path);
BufferedReader br = new BufferedReader(fr);
while (br.ready()) {
// 將讀取的單個數據加入到集合裏面
nums.add(Integer.parseInt(br.readLine()));
}
// 對集合進行排序
Collections.sort(nums);
// 將排序好的數據寫入源文件
numberSort(nums, path);
br.close();
fr.close();
} catch (NumberFormatException e) {
e.printStackTrace();
} catch (FileNotFoundException e) {
e.printStackTrace();
} catch (IOException e) {
e.printStackTrace();
}
}
}
// 對每個文件數據進行排序,再寫入源文件
public static void numberSort(LinkedList list, String path) {
try {
FileWriter fs = new FileWriter(path);
BufferedWriter fw=new BufferedWriter(fs);
for (Iterator iterator = list.iterator(); iterator.hasNext();) {
Object object = (Object) iterator.next();
fw.write(object + "\r\n");
}
fw.close();
fs.close();
} catch (IOException e) {
e.printStackTrace();
}
}
// 文件數據最終整合(精度調整)
public static void deathDataFile(String filepath, String sqlitFilePath1,
int countAccuracy, int CountFile) throws IOException {
LinkedList nums = new LinkedList();//添加數據,進行排序
Object temp = null; // 記錄每次排序剩下的最後一個數字
boolean ispass = false;
LinkedList ispasses = null;//記錄數據文件的狀態信息
FileWriter fs = new FileWriter(filepath + "\\Sort.txt", false);//創建文件流,以便整合的數據寫入
BufferedWriter bw=new BufferedWriter(fs);
FileReader fr = null; //聲明讀取文件流
BufferedReader br = null; //聲明BufferedReader
LinkedList WriterLists = new LinkedList(); // 初始化文件流對象集合
LinkedList WriterListFile = new LinkedList();
for (int j = 1; j <= CountFile; j++) {
// 聲明對象,開啓所有子文件流訪問所有子文件的數據
fr = new FileReader(sqlitFilePath1 + "\\12" + j + ".txt");
//開啓所有BufferedReader,方便下次的整行的讀取
br = new BufferedReader(fr);
// 將所有 FileReader對象裝入集合
WriterListFile.add(fr);
// 將所有 BufferedReader對象裝入集合
WriterLists.add(br);
}
for (;;) {
// 將十個源文件的是否有數據情況存入集合,以方便後面做判斷
ispasses = new LinkedList();
// 分別讀取十個源文件的單個數據
for (Iterator iterator = WriterLists.iterator(); iterator.hasNext();) {
BufferedReader object = (BufferedReader) iterator.next();
Object obj = null;
while (object.ready()) {
//添加所有文件流的每次的數據
nums.add(Integer.parseInt(object.readLine().toString()));
break;
}
if (object.ready() == false)
ispasses.add("true"); //將各文件中的數據狀態存入集合中
}
// 決斷是否是第一次進來
if (nums.size() % countAccuracy == 0 && ispass == false) {
// 對集合進行排序
Collections.sort(nums);
// 接收最大的數據,其它的數據寫入總排序文件
temp = numberSortData(nums, filepath, false, countAccuracy, bw);
//重新初始化集合
nums = new LinkedList();
// 添加上一組比較剩下的數據
nums.add(temp);
ispass = true;
// 記錄源文件的數據數量,以便下次的遍歷
continue;
}
if (ispass) {
if (nums.size() % countAccuracy == 1 && nums.size() > 1) {
// 對集合進行排序
Collections.sort(nums);
// 接收最大的數據,其它的數據寫入總排序文件
temp = numberSortData(nums, filepath, true, countAccuracy,
bw);
nums = new LinkedList();
nums.add(temp);
continue;
}
}
// 記錄下一組數據的位置
// 判斷是不是十個文件都沒有數據
if (ispasses.size() == CountFile) {
Collections.sort(nums);
temp = numberSortData(nums, filepath, true, countAccuracy, bw);
nums = new LinkedList();
break;
}
}
bw.close();
//關閉寫入流
fs.close();
//關閉所有的BufferedReader
for (Iterator iterator = WriterLists.iterator(); iterator.hasNext();) {
BufferedReader object2 = (BufferedReader) iterator.next();
object2.close();
}
//關閉所有的FileReader
for (Iterator iterator = WriterListFile.iterator(); iterator.hasNext();) {
FileReader object = (FileReader) iterator.next();
object.close();
}
}
// 對數據進行排序,寫入最終文件中(精度調整)
public static Object numberSortData(LinkedList list, String filePath,
boolean ispass, int countAccuracy,BufferedWriter fs) {
Object temp = 0;//記錄最後一個值
int tempCount = 0;//記錄寫入的數據位置
try {
for (Iterator iterator = list.iterator(); iterator.hasNext();) {
Object object = (Object) iterator.next();
// 判斷是否是最後一個數
if (tempCount == list.size() - 1) {
// 判斷集合裏面不足一百個數了
if (list.size() < countAccuracy + 1 && ispass) {
temp = null;
} else {
temp = object;
break;
}
}
// 寫入數據源
fs.write(object + "\r\n");
// 記錄數據的下標
tempCount++;
}
} catch (IOException e) {
e.printStackTrace();
}
return temp;
}
}