/**
* (C) Copyright 2012 ZWW
*/
package test;
import java.io.BufferedReader;
import java.io.File;
import java.io.FileNotFoundException;
import java.io.FileOutputStream;
import java.io.FileReader;
import java.io.IOException;
import java.io.InputStream;
import java.io.OutputStream;
import java.lang.reflect.InvocationTargetException;
import java.net.URL;
import java.util.ArrayList;
import java.util.List;
import java.util.Scanner;
import java.util.UUID;
import jsoup.ConnectionManager;
import net.sf.json.JSONObject;
import org.apache.commons.lang.StringUtils;
import org.jsoup.nodes.Document;
import org.jsoup.nodes.Element;
import org.jsoup.select.Elements;
import common.Config;
import cputils.XmlUtils;
/**
* @author ZWW
* @date 2014-3-22 下午5:09:37
* @version V1.0
*/
public class XxhhPic {
private static String PIC_PATH = "E:\\xxhh-pic\\";
/**
* @param args
*/
public static void main(String[] args) {
//System.out.println(Integer.parseInt(s));
//System.out.println(UUID.randomUUID().toString().replace("-", ""));
//http://www.xxhh.com/page/4
//getLogoFromWeb("", "temp/");
XxhhPic pic = new XxhhPic();
while(true){
System.out.print("Input keyword:");
Scanner sca = new Scanner(System.in);
String queryStr = sca.next();
if ("0".equals(queryStr)) {
System.exit(0);
sca.close();
}else if("del".equals(queryStr)) {
pic.deletePic(PIC_PATH);
}else if("down".equals(queryStr)){
pic.downPic("http://www.xxhh.com/page/", PIC_PATH);
}
}
// pic.getAllPic();
}
public boolean deletePic(String filePath){
File file = new File(filePath);
if (file.isFile()) {
deleteFile(file);
}else {
deleteDirectory(file.getAbsolutePath());
}
// for (File f : file.listFiles()) {
// f.delete();
// }
return true;
}
/**
* 删除单个文件
* @param sPath 被删除文件的文件名
* @return 单个文件删除成功返回true，否则返回false
*/
public boolean deleteFile(File file) {
// 路径为文件且不为空则进行删除
if (file.isFile() && file.exists()) {
file.delete();
System.out.println("delete file:"+file.getAbsolutePath());
}
return true;
}
/**
* 删除目录（文件夹）以及目录下的文件
* @param sPath 被删除目录的文件路径
* @return 目录删除成功返回true，否则返回false
*/
public boolean deleteDirectory(String sPath) {
//如果sPath不以文件分隔符结尾，自动添加文件分隔符
if (!sPath.endsWith(File.separator)) {
sPath = sPath + File.separator;
}
File dirFile = new File(sPath);
//如果dir对应的文件不存在，或者不是一个目录，则退出
if (!dirFile.exists() || !dirFile.isDirectory()) {
return false;
}
//删除文件夹下的所有文件(包括子目录)
File[] files = dirFile.listFiles();
for (int i = 0; i < files.length; i++) {
//删除子文件
if (files[i].isFile()) {
deleteFile(files[i]);
} //删除子目录
else {
deleteDirectory(files[i].getAbsolutePath());
}
}
//删除当前目录
if (dirFile.delete()) {
System.out.println("delete dir:"+dirFile.getAbsolutePath());
return true;
} else {
return false;
}
}
public void downPic(String url,String filePath) {
int pageNo = 1;
try {
Document doc = ConnectionManager.getInstance().getDocument(url+pageNo);
do {
System.out.println("**********"+doc.baseUri()+"**********");
// Elements elements = doc.select("div[class=container] > div[class=content]");
// for (Element element : elements) {
// System.out.println(element.html());
// }
Elements elements = doc.select("div[class^=bdshare_t]");
JSONObject obj = null;
for (Element element : elements) {
String data = element.attr("data");
try {
if (data.indexOf("'title'") == -1) {
continue;
}
data = data.replaceAll("\r\n", "");
String title = data.substring(data.indexOf("'title'")+9,data.indexOf("'comment'")-2).trim();
title = title.substring(0, title.length()-2);
String pic = data.substring(data.indexOf("'pic'")+7,data.indexOf("'desc'")-2).trim();
pic = pic.substring(0, pic.length()-2);
data = "{"+data.substring(data.indexOf("'title'"));
// System.out.println(data);
// obj = JSONObject.fromObject(data);
// if (obj.getString("pic").endsWith("jpg")||obj.getString("pic").endsWith("gif")) {
// getLogoFromWeb(obj.getString("pic"), filePath+"\\"+pageNo+"\\",obj.getString("title"));
// }
if (pic.endsWith("jpg")||pic.endsWith("gif")) {
getLogoFromWeb(pic, filePath,title);
// getLogoFromWeb(pic, filePath+"\\"+pageNo+"\\",title);
}
} catch (Exception e) {
System.out.println(data);
e.printStackTrace();
}
// System.out.println(obj);
}
// Elements els = doc.getElementsByAttribute("_src");
// //System.out.println(els.size());
// if (els.size()>0) {
// String imgUrl = "";
// for (Element element : els) {
// imgUrl = element.attr("tsrc");
// if (StringUtils.isBlank(imgUrl)) {
// imgUrl = element.attr("_src");
// }
// //System.out.println(imgUrl);
// getLogoFromWeb(imgUrl, filePath);
// }
// }
doc = ConnectionManager.getInstance().getDocument(url+(++pageNo));
} while (doc.text().contains("下一页"));
System.out.println("**********END**********");
} catch (Exception e) {
e.printStackTrace();
}
}
public File getLogoFromWeb(String logourl, String uploadPath,String fileName) {
try {
String fileType = logourl.substring(logourl.lastIndexOf("."), logourl.length());
File dirFile = new File(uploadPath);
if (!dirFile.exists()) {
dirFile.mkdir();
}
File outFile = new File(uploadPath + fileName + fileType);
System.out.println(fileName + fileType);
if (outFile.exists()) {
return outFile;
}
if (!isExisted(fileName + fileType)) {
putFileName(fileName + fileType);
}
outFile.createNewFile();
OutputStream os = new FileOutputStream(outFile);
URL url = new URL(logourl);
InputStream is = url.openStream();
byte[] buff = new byte[1024];
while (true) {
int readed = is.read(buff);
if (readed == -1) {
break;
}
byte[] temp = new byte[readed];
System.arraycopy(buff, 0, temp, 0, readed);
os.write(temp);
}
is.close();
os.close();
return outFile;
} catch (Exception e) {
e.printStackTrace();
}
return null;
}
public File getLogoFromWeb(String logourl, String uploadPath) {
try {
return getLogoFromWeb(logourl, uploadPath, UUID.randomUUID().toString().replace("-", ""));
} catch (Exception e) {
e.printStackTrace();
}
return null;
}
public static boolean isExisted(String content){
boolean had = false;
try {
FileReader fileReader = new FileReader(new File(Config.DATA_FILE_PATH+"/xxhhPicName.txt"));
BufferedReader reader = new BufferedReader(fileReader);
String con;
while ((con = reader.readLine()) != null) {
if (content.equals(con)) {
// System.out.println(fileName+" is existed..");
had = true;
break;
}
}
reader.close();
fileReader.close();
} catch (FileNotFoundException e) {
System.err.println(e.getMessage());
} catch (IOException e) {
System.err.println(e.getMessage());
}
return had;
}
private List<String> getAllPic(){
System.out.println("****put fileName begin...");
List<String> list = new ArrayList<String>();
File dirfile = new File(PIC_PATH);
if (!dirfile.exists()) {
System.err.println("dir not found...");
return list;
}
for (File f : dirfile.listFiles()) {
for (String s : f.list()) {
list.add(s);
if (!isExisted(s)) {
putFileName(s);
}
}
}
System.out.println("****put fileName ok...");
return list;
}
private void putFileName(String fileName){
XmlUtils.writeDataFile(fileName+"\r\n", "xxhhPicName.txt", true);
// System.out.println("****put fileName: "+fileName);
}
}