读网上的URL，会出的502错误

haoyunlyh

浏览: 11930 次
性别:
来自: 上海

最近访客更多访客>>

lBovinl

tcl_6666

Kuroneko

RaceBug2010

博主相关

博客

微博

相册

留言

关于我

文章分类

社区版块

存档分类

博客分类：

功能现实

Excel 浏览器 C C++C#

最近做一个项目需要调用网络上的资料，是别的公司提供的一个url接口。我们只要拼URL，然后去访问就OK了！

看起来很简单的调用，却弄了好长时间也算搞定，但是办法也不是很好！

一开始直接用URLConnection进行读操作。（加上代理）

java 代码

System.getProperties().put("proxySet","true");
System.getProperties().put("proxyHost","IP");
System.getProperties().put("proxyPort","PORT");
URLConnection huc = url.openConnection();
huc.setDoInput(true);
StringBuffer str = new StringBuffer();
BufferedReader in=new BufferedReader(new InputStreamReader(huc.getInputStream()));
String inputLine;
while((inputLine=in.readLine())!=null)
str.append(inputLine);
in.close();

运行发现，验证用户的URL没有问题，但在读用户信息的URL时，却很长时间没有反应，过了大约5分钟左右报502错误

502：是Bad gate way ，应该是网关错误。开始以为是我们的代理不行。然后就用拨号来试。却发现结果是相同的！

还以为是URL错误，然后把URL打在浏览器里进行尝试，却发现成功了。

于是认为是URLConnection这个类的功能不够强大，容错性不行。

并下httpClient来试，（抱怨一声httpClient的版本太多了，还要code一起运行），终于写好了，尝试运行，

java 代码

StringBuffer str = new StringBuffer();
httpclient = new DefaultHttpClient();
//httpclient.setHttpRequestRetryHandler(new DefaultHttpRequestRetryHandler(3,true));
//代理
/**HttpHost targetHost = new HttpHost("serverip",server port, "http");
HttpHost proxy = new HttpHost("IP", port);
HttpRoute route = new HttpRoute(targetHost, null, proxy, false);
HttpGet httpget = new HttpGet("/path");
RoutedRequest routedReq = new RoutedRequest.Impl(httpget, route);
HttpResponse response = httpclient.execute(routedReq, null);*/
//不用代理
HttpGet httpget = new HttpGet(urlStr);
HttpResponse response = httpclient.execute(httpget);
HttpEntity entity = response.getEntity();
if (entity != null) {
BufferedInputStream bis = new BufferedInputStream(entity.getContent());
byte[] b = new byte[1024];
int i=-1;
while((i=bis.read(b))!=-1){
str.append(new String(b,0,i));
}
}
if (entity != null) {
entity.consumeContent();
}

还是不行，同样的问题

于是想是不是这个接口做了控制，（因为浏览器访问从来都是正确结果的），于是想办法封闭请求，便有了下面的代码

java 代码

HttpHost targetHost = new HttpHost("serverip", serverpot, "http");
HttpHost proxy = new HttpHost("ip", port);
HttpRoute route = new HttpRoute(targetHost, null, proxy, false);
HttpGet httpget = new HttpGet("/.......");*/
httpget.setHeader("Accept","image/gif, image/x-xbitmap, image/jpeg, image/pjpeg, application/x-shockwave-flash, application/vnd.ms-powerpoint, application/vnd.ms-excel, application/msword, */*");
httpget.setHeader("Accept-Language","zh-cn");
httpget.setHeader("UA-CPU","x86");
httpget.setHeader("Accept-Encoding","gzip,deflate");
httpget.setHeader("User-Agent","Mozilla/4.0 (compatible; MSIE 7.0; Windows NT 5.1)");
httpget.setHeader("Host","serverip:serverpot");
httpget.setHeader("Proxy-Connection", "Keep-Alive");
RoutedRequest routedReq = new RoutedRequest.Impl(httpget, route);

便下载了请求的拦截器，拦截请求，使得它发出的请求和浏览器的相同，可结果令人失望。（要绝望了）

这时发现，httpclient的重试功能，第二次重试可以成功访问。但中间的时间却有5分钟之久，于是便想减少超时时间

也许我没找到，或者资料不全，硬是没找到（注：httpcomponents-client-4.0-alpha1-bin,httpcomponents-core-4.0-alpha5-bin用的这两个包来做的）

于是自己创线程20秒访问不了，就关掉上一个线程，再重试，代码下：

java 代码

public void run(){
if(isReader){
t.cancel();
}else{
/**
* 关闭
*/
if(httpclient != null){
httpclient.getConnectionManager().shutdown();
//System.out.println("关闭连接");
}
}
}
public void process(){
try{
/**
* 开始连接
*/
String para = this.getParam(paras);
if(para.length()!=0){
urlStr = urlStr + "?" +para;
}
//System.out.println(new Date());
//System.out.println(urlStr);
StringBuffer str = new StringBuffer();
httpclient = new DefaultHttpClient();
//httpclient.setHttpRequestRetryHandler(new DefaultHttpRequestRetryHandler(3,true));
//代理
/**HttpHost targetHost = new HttpHost("servertip", serverport, "http");
HttpHost proxy = new HttpHost("ip", port);
HttpRoute route = new HttpRoute(targetHost, null, proxy, false);
HttpGet httpget = new HttpGet("/.......");
RoutedRequest routedReq = new RoutedRequest.Impl(httpget, route);
HttpResponse response = httpclient.execute(routedReq, null);*/
//不用代理
HttpGet httpget = new HttpGet(urlStr);
HttpResponse response = httpclient.execute(httpget);
HttpEntity entity = response.getEntity();
if (entity != null) {
BufferedInputStream bis = new BufferedInputStream(entity.getContent());
byte[] b = new byte[1024];
int i=-1;
while((i=bis.read(b))!=-1){
str.append(new String(b,0,i));
}
}
if (entity != null) {
entity.consumeContent();
}
if(str.indexOf("<rawdata></rawdata>")!=-1){
str.delete(str.indexOf("<rawdata></rawdata>"), str.indexOf("")+"".length());
}
String xmlStr = this.toUnicode(str.toString(), "gb2312");
InputStream sbis = new StringBufferInputStream(xmlStr);
//System.out.println(xmlStr);
//System.out.println(new Date());
DocumentBuilderFactory dbf=DocumentBuilderFactory.newInstance();
DocumentBuilder db=dbf.newDocumentBuilder();
doc=db.parse(sbis);
//is.close();
sbis.close();
this.isReader = true;
}
catch(Exception e)
{
//log.error(e);
//e.printStackTrace();
process();
}
finally
{
}
}
public Document processUrl(String urlStr,Map paras){
this.urlStr = urlStr;
this.paras = paras;
t = new Timer();
t.schedule(this,5000, 5000);
//this.run();
this.process();
t.cancel();
}
private String toUnicode(String strText,String code) throws UnsupportedEncodingException{
char c;
String strRet = "" ;
int intAsc;
String strHex;
strText = new String(strText.getBytes("gb2312"),code);
for ( int i = 0; i < strText.length(); i++ ){
c = strText.charAt(i);
intAsc = (int)c;
if(intAsc>128){
strHex = Integer.toHexString(intAsc);
strRet = strRet + "&#x" + strHex+";";
}
else{
strRet = strRet + c;
}
}
return strRet ;
}

运行了一下，结果还行，就是这个办法不规范啊！（最后还有编码问题，郁闷死了）。唉！

还有关于这个问题，我了贴子问问的，可怎么老是被放到入门贴呢，还扣分。呵呵。但问题总算有个说法了！

分享到：

oracle中not in 用法

2007-10-31 10:57
浏览 5329
评论(1)
查看更多

1 楼 denghan 2008-11-10

同问，我最近也遇到了类似的问题。我的是要访问一个要需要登录验证的站点，登录后再从一个URL获取数据。用了httpclient3,httpclient4.0_beta3，都不行，登录后打开URL获取数据这一步总是5分钟左右无响应，然后报502错误

发表评论

您还没有登录,请您登录后再发表评论

最近访客更多访客>>

博主相关

文章分类

社区版块

存档分类

最新评论