版权声明：本文为博主原创文章，未经博主允许不得转载 https://blog.csdn.net/kzcming/article/details/83383909
1 需求:
某个接口的纪录在ftp 服务器上,以类别/日期/时间.来存放文件,而一天可能会产生几百个文件,需要下载文件进行保存
问题:
1. 这个时候如果同时,要拿几个类别,某个时间段的数据,就要疯狂下载了,如果是单线程的
2. ftp 一般只允许同一个用户名,同时有几个进程连接
3. ftp 多个读取文件循环读取的时候,经常会出现,第一个文件得到字节流,而后面的文件的字节流都是null, 不管是sun的ftp,还是apache得common得ftp 包
4. 读取完文件,多线程如何通知,或者得到结果
解决办法:
1. 第一个问题,使用多线程去爬取,先取出符合条件得文件名,然后放到文件名集合,然后取根据文件名集合下载文件
2. 第二个问题,给每个线程都有一个连接
3.第三个问题: ,读取完一个文件,就关流,并执行
completePending 方法,使连接可以持续读取文件,不然读取一个文件后,读取其他文件就一直获取null,ftp.enablePassiveMode(true);//被动模式,读取文件时有时候ftp 时区和本地时区不一样,读取文件之前,需要将连接设置为被动模式,不然有时也会报错4.使用 Executors 类 和 Future来获得通知,并使用
future.get(); 方法,这个方法会阻塞当前线程,直至线程执行完毕具体代码:
package com.my;
import java.io.BufferedReader;
import java.io.File;
import java.io.FileOutputStream;
import java.io.IOException;
import java.io.InputStream;
import java.io.InputStreamReader;
import java.net.InetSocketAddress;
import java.net.SocketAddress;
import java.util.ArrayList;
import java.util.HashMap;
import java.util.List;
import java.util.Map;
import java.util.concurrent.*;
import org.dom4j.*;
import sun.net.ftp.FtpClient;
import sun.net.ftp.FtpProtocolException;
public class FtpUtil {
public static void main(String[] args) {
long startTime = System.currentTimeMillis();
List list = history("201810230000","201810231200");
long endTime = System.currentTimeMillis();
System.out.println((endTime - startTime) + "毫秒");
System.out.println(list.size());
}
public static FtpClient connectFTP(String url, int port, String username, String password) {
//创建ftp
FtpClient ftp = null;
try {
//创建地址
SocketAddress addr = new InetSocketAddress(url, port);
//连接
ftp = FtpClient.create();
ftp.connect(addr);
//登陆
ftp.login(username, password.toCharArray());
ftp.setBinaryType();
} catch (FtpProtocolException e) {
e.printStackTrace();
} catch (IOException e) {
e.printStackTrace();
}
return ftp;
}
public static void close (FtpClient ftp) {
if(ftp != null) {
try {
ftp.close();
} catch (IOException e) {
e.printStackTrace();
}
}
}
// ======================================== 对 ag 的文件编译的字符串进行 解析成Map ========================================================
//对时间进行转换 转换为美东时间
// 找出对应的日期的文件有哪些
//对四个顶层目录,分别遍历查找对应时间的文件有哪些,将这些文件名放入 对应的四个map中
// 对每个map 中的文件,分别进行解析,并放入map,然后将map 放入list
//最后将四个list 放入list
//顶层目录有四个
private static String [] parentFile = {"文件夹1","文件夹2","文件夹3", "文件夹4"};
private static final String url = "ftp网址";
private static final int port = 23;
private static final String username = "用户名";
private static final String password = "密码";
/**
* 时间为UTC-4 美东时间,与北京时间相差8个小时
* @param fromDate
* @param toDate
*/
public static List history(String fromDate,String toDate) {
FtpClient ftp = FtpUtil.connectFTP(url, port, username, password);//获得ftp 连接
List<List> list = new ArrayList<>();
List<List> nameLists = new ArrayList<>();
if(ftp.isConnected() && ftp.isLoggedIn()) {
for (int i = 0; i < parentFile.length; i++) {
String filePath = "/";
filePath = filePath + parentFile[i] +"/" + fromDate.substring(0,8);
List<String> nameList = getChildFileName(filePath, ftp, fromDate.substring(0,12), toDate.substring(0,12));
if(nameList != null && nameList.size() > 0) {
nameLists.add(nameList);
}
}
}
close(ftp);
if(nameLists != null && nameLists.size() > 0) {
for (List nameList: nameLists ) {
FtpUtil ftpUtil = new FtpUtil();
List<Future<List<Map<String, String>>>> futureList = ftpUtil.threadSearch(nameList,10);
List<Map<String, String>> childList = new ArrayList<>();
list.add(childList);
for (Future<List<Map<String, String>>> future : futureList) {
try {
System.out.println(future.isDone() + "---" + future.toString());
List<Map<String, String>> reList = future.get();
System.out.println(future.toString() + "---单个线程所得的结果---" + reList.size());
if (reList != null) {
childList.addAll(reList);
}
System.out.println(future.isDone() + "---" + future.toString());
} catch (InterruptedException | ExecutionException e) {
e.printStackTrace();
}
}
}
}
return list;
}
/**
* 根据父文件名,获得列出其下的文件的字节流
* 读取字节流的同时,根据 开始时间和结束时间判断是否是需要的文件,
* 将需要的文件放入list中
* @param ftpFile
* @param ftp
* @param fromDate
* @param toDate
* @return
*/
public static List<String> getChildFileName(String ftpFile,FtpClient ftp,String fromDate,String toDate) {
long startTime = Long.valueOf(fromDate);
long endTime = Long.valueOf(toDate);
InputStream is = null;
List<String> list = new ArrayList<>();
try {
// 获取ftp上的文件
is = ftp.nameList(ftpFile);
if(is != null) {
BufferedReader in = new BufferedReader(new InputStreamReader(is,"UTF-8"));
while (true) {
String line = in.readLine();
if (line == null) {
break;
}
else {
long dateNumber = Long.valueOf(line.substring(line.lastIndexOf("/") + 1,line.length()).replace(".xml",""));
if(startTime <= dateNumber && dateNumber <= endTime ) {
list.add(line);
}
}
}
in.close();
}
} catch (FtpProtocolException | IOException e) {
e.printStackTrace();
} finally {
try {
if(is!=null){
is.close();
}
} catch (IOException e) {
e.printStackTrace();
}
}
return list;
}
public static String parseXmlFile(String ftpFile,FtpClient ftp) {
InputStream is = null;
StringBuffer sbf = new StringBuffer();
ftp.enablePassiveMode(true);//被动模式
try {
// 获取ftp上的文件
is = ftp.getFileStream(ftpFile);
if (is != null) {
BufferedReader in = new BufferedReader(new InputStreamReader(is, "UTF-8"));
while (true) {
String line = in.readLine();
if (line == null) {
break;
} else {
sbf.append(line);
sbf.append("\r\n");
// valList.add(parseXml(line));
}
}
in.close();
ftp.completePending();
}
} catch (FtpProtocolException | IOException e) {
e.printStackTrace();
}
return sbf.toString();
}
/**
* 多线程分配
*/
public List<Future<List<Map<String,String>>>> threadSearch(List<String> list,int threadCount) {
int len = threadCount;//定义分割多少份线程
if(list.size() < len){
len = list.size();
}
List<List<String>> splitList = splitList(list, len);//分割一个线程执行多少个类型
TimerThread tt = null;
ExecutorService exec = Executors.newCachedThreadPool();//工头
ArrayList<Future<List<Map<String,String>>>> results = new ArrayList<>();//结果通知
for (int i = 0; i < len; i++) {
TimerThread timerThread = new TimerThread(splitList.get(i),connectFTP(url,port,username,password));
results.add(exec.submit(timerThread));//submit返回一个Future，代表了即将要返回的结果
}
return results;
}
/**
* 多线程具体类
* @author 12198
*
*/
public class TimerThread implements
Callable<List<Map<String,String>>> {
public List<String> li;
public FtpClient ftp;
public List<Map<String,String>> list;
public TimerThread(List<String> li,FtpClient ftp){
this.li=li;
this.ftp = ftp;
this.list = new ArrayList<>();
}
@Override
public List<Map<String, String>> call() throws Exception {
StringBuffer sbf = new StringBuffer();
if(null != li && li.size() > 0){
for(String filePath : li){
System.out.println(Thread.currentThread().getName() + "--" + ftp.toString());
// list.addAll(parseXmlFile(filePath,ftp));
sbf.append(parseXmlFile(filePath,ftp));
}
}
close(ftp);
//最后一起解析字符串
String [] str = sbf.toString().split("\r\n");
for (String xmlString:str) {
Map<String,String> map = parseXml(xmlString);
if(map != null && map.size() > 0) list.add(map);
}
return this.list;
}
}
// ========================= 引用的其他类方法 ====================================
public static Map<String,String> parseXml(String str) {
Map<String,String> xmlMap = new HashMap<>();
Document doc = null;
try {
doc = DocumentHelper.parseText(str);
} catch (DocumentException e) {
e.printStackTrace();
}
if(doc != null) {
Element rootElement = doc.getRootElement();
List<Attribute> list = rootElement.attributes();
if(list != null && list.size() > 0) {
if(list != null && list.size() > 0 ) {
for (Attribute attr : list) {
xmlMap.put(attr.getName(), attr.getValue());
}
}
}
}
return xmlMap;
}
public static <T> List<List<T>> splitList(List<T> list, int n) {
List<List<T>> strList = new ArrayList<>();
if (list == null) return strList;
int size = list.size();
int quotient = size / n; // 商数
int remainder = size % n; // 余数
int offset; // 偏移量
int len = quotient > 0 ? n : remainder; // 循环长度
int start = 0; // 起始下标
int end; // 结束下标
List<T> tempList;
for (int i = 0; i < len; i++) {
if (remainder != 0) {
remainder--;
offset = 1;
} else {
offset = 0;
}
end = start + quotient + offset;
tempList = list.subList(start, end);
start = end;
strList.add(tempList);
}
return strList;
}
}
apache commons ftp 版得
package com.my;
import org.apache.commons.net.ftp.FTPClient;
import org.dom4j.*;
import java.io.BufferedReader;
import java.io.IOException;
import java.io.InputStream;
import java.io.InputStreamReader;
import java.util.ArrayList;
import java.util.HashMap;
import java.util.List;
import java.util.Map;
import java.util.concurrent.*;
/**
* 使用apache-commons-net 包的ftp 连接工具
*
* 指定每个线程获得一个FtpClient 连接实例
*
*/
public class ApacheFtpUtils {
public static void main(String[] args) {
List list = history("201810170200","201810170300");
System.out.println(list.size());
}
public static FTPClient connectFTP(String hostName,int port,String username,String password) {
FTPClient ftp = new FTPClient();
boolean isLogin = false;
try {
ftp.connect(url,port);
isLogin = ftp.login(username,password);
} catch (IOException e) {
e.printStackTrace();
}
if(isLogin) return ftp;
else return null;
}
/**
* 根据父文件名,获得列出其下的文件的字节流
* 读取字节流的同时,根据 开始时间和结束时间判断是否是需要的文件,
* 将需要的文件放入list中
* @param ftpFile
* @param ftp
* @param fromDate
* @param toDate
* @return
*/
public static List<String> getChildFileName(String ftpFile,FTPClient ftp,String fromDate,String toDate) {
long startTime = Long.valueOf(fromDate);
long endTime = Long.valueOf(toDate);
List<String> list = new ArrayList<>();
String [] nameStr = null;
ftp.enterLocalPassiveMode();//解决 net 包因中文 年/月,无法匹配,而导致的文件查找为空问题
try {
nameStr = ftp.listNames(ftpFile);
} catch (IOException e) {
e.printStackTrace();
}
if(nameStr != null && nameStr.length >0 ) {
for (int i = 0; i < nameStr.length; i++) {
String line = nameStr[i];
long dateNumber = Long.valueOf(line.substring(line.lastIndexOf("/") + 1,line.length()).replace(".xml",""));
if(startTime <= dateNumber && dateNumber <= endTime ) {
list.add(line);
}
}
}
ftp.enterLocalActiveMode();
return list;
}
/**
* 下载ftp 文件,并隔行解析,放入list
* @param ftpFile
* @param ftp
* @return
*/
public static List<Map<String,String>> parseXmlFile(String ftpFile,FTPClient ftp) {
List<Map<String,String>> valList = new ArrayList<>();
InputStream is = null;
StringBuffer sbf = new StringBuffer();
try {
// 获取ftp上的文件
ftp.enterLocalPassiveMode();
is = ftp.retrieveFileStream(ftpFile);
if (is != null) {
BufferedReader in = new BufferedReader(new InputStreamReader(is, "UTF-8"));
while (true) {
String line = in.readLine();
if (line == null) {
break;
} else {
valList.add(parseXml(line));
}
}
in.close();
ftp.completePendingCommand();//完成等待方法 ,如果一个连接中要连续读写多个文件,需要强行关流,在执行completePendingCommand 方法
}
} catch ( IOException e) {
e.printStackTrace();
}
ftp.enterLocalActiveMode();
return valList;
}
private static String [] parentFile = {"文件名1","文件名2","文件名3", "文件名4"};
private static final String url = "ftp网址";
private static final int port = 23;
private static final String username = "用户名";
private static final String password = "密码";
/**
* 时间为UTC-4 美东时间,与北京时间相差8个小时
* @param fromDate
* @param toDate
*/
public static List history(String fromDate,String toDate) {
FTPClient ftp = ApacheFtpUtils.connectFTP(url, port, username, password);//获得ftp 连接
List<List> list = new ArrayList<>();// 具体的List<List<Map<String,String>>> 集合
List<List> nameLists = new ArrayList<>();// 文件名列集合
if(ftp != null ) {
for (int i = 0; i < parentFile.length; i++) {
String filePath = "/";
filePath = filePath + parentFile[i] +"/" + fromDate.substring(0,8);//HUNTER","AGIN","XIN", "YOPLAY
List<String> nameList = getChildFileName(filePath, ftp, fromDate.substring(0,12), toDate.substring(0,12));
//nameList 为0 判断
if(nameList != null && nameList.size() > 0) {
nameLists.add(nameList);
}
}
}
try {
ftp.logout();
} catch (IOException e) {
e.printStackTrace();
}
//将得到的符合查询要求的文件名集合,进行遍历,每个元素,循环处理,每个元素,分多个线程去下载解析
if(nameLists != null && nameLists.size() > 0) {
for (List nameList:nameLists) {
ApacheFtpUtils ftpUtils = new ApacheFtpUtils();
List<Future<List<Map<String, String>>>> futureList = ftpUtils.threadSearch(nameList);
List<Map<String,String>> childList = new ArrayList<>();
list.add(childList);
for (Future<List<Map<String, String>>> future : futureList) {
try {
System.out.println(future.isDone() + "---" + future.toString());
List<Map<String, String>> reList = future.get();
System.out.println(future.toString() + "---单个线程所得的结果---" + reList.size());
if (reList != null) {
childList.addAll(reList);
}
System.out.println(future.isDone() + "---" + future.toString());
} catch (InterruptedException | ExecutionException e) {
e.printStackTrace();
}
}
}
}
return list;
}
/**
* 多线程分配
* @param list
* @return
*/
public List<Future<List<Map<String,String>>>> threadSearch(List<String> list) {
int len = 5;//定义分割多少份线程
if(list.size() < len){
len = list.size();
}
List<List<String>> splitList = splitList(list, len);//分割一个线程执行多少个类型
TimerThread tt = null;
ExecutorService exec = Executors.newCachedThreadPool();//工头
ArrayList<Future<List<Map<String,String>>>> results = new ArrayList<>();//结果通知
for (int i = 0; i < len; i++) {
TimerThread timerThread = new TimerThread(splitList.get(i),connectFTP(url,port,username,password));
results.add(exec.submit(timerThread));//submit返回一个Future，代表了即将要返回的结果
}
return results;
}
/**
* 多线程 具体类
*/
public class TimerThread implements
Callable<List<Map<String,String>>> {
public List<String> li;
public FTPClient ftp;
public List<Map<String,String>> list;
public TimerThread(List<String> li,FTPClient ftp){
this.li=li;
this.ftp = ftp;
this.list = new ArrayList<>();
}
@Override
public List<Map<String, String>> call() throws Exception {
if(null != li && li.size() > 0){
// for(String filePath : li){
for (int i = 0; i < li.size(); i++) {
System.out.println(Thread.currentThread().getName() + "--" + ftp.toString());
list.addAll(parseXmlFile(li.get(i),ftp));
}
}
//执行完毕,关闭连接
ftp.logout();
return this.list;
}
}
//====================================== 引入的其他类的方法 ============================================================
/**
* 解析xml 字符串
* @param str
* @return
*/
public static Map<String,String> parseXml(String str) {
Map<String,String> xmlMap = new HashMap<>();
Document doc = null;
try {
doc = DocumentHelper.parseText(str);
} catch (DocumentException e) {
e.printStackTrace();
}
if(doc != null) {
Element rootElement = doc.getRootElement();
List<Attribute> list = rootElement.attributes();
if(list != null && list.size() > 0) {
if(list != null && list.size() > 0 ) {
for (Attribute attr : list) {
xmlMap.put(attr.getName(), attr.getValue());
}
}
}
}
return xmlMap;
}
public static <T> List<List<T>> splitList(List<T> list, int n) {
List<List<T>> strList = new ArrayList<>();
if (list == null) return strList;
int size = list.size();
int quotient = size / n; // 商数
int remainder = size % n; // 余数
int offset; // 偏移量
int len = quotient > 0 ? n : remainder; // 循环长度
int start = 0; // 起始下标
int end; // 结束下标
List<T> tempList;
for (int i = 0; i < len; i++) {
if (remainder != 0) {
remainder--;
offset = 1;
} else {
offset = 0;
}
end = start + quotient + offset;
tempList = list.subList(start, end);
start = end;
strList.add(tempList);
}
return strList;
}
}