最近研究微信的公众平台开发，需要和微信的服务器进行数据读取，简单研究了下jdk自带的HttpUrlConnection类（URLConnection的子类），简单实现了一下微信的access_token获取。
该地址是get方法请求即可，先贴代码，doGet：
private final String ACCESS_TOKEN_WEIXIN = "https://api.weixin.qq.com/cgi-bin/token";
public String doGet(String grantType, String appId, String secret){
String url = ACCESS_TOKEN_WEIXIN+"?grant_type="+grantType+
"&appid="+appId+"&secret="+secret;
HttpURLConnection conn = null;
InputStream is = null;
InputStreamReader reader = null;
BufferedReader br = null;
String str = "";
try {
URL weiUrl = new URL(url);
conn = (HttpURLConnection)weiUrl.openConnection();
conn.setRequestProperty("connection", "Keep-Alive");
conn.connect();
is = conn.getInputStream();
reader = new InputStreamReader(is, "UTF-8");
br = new BufferedReader(reader);
String readLine = "";
while((readLine=br.readLine())!=null){
str+=readLine+"\n";
}
} catch (Exception e) {
e.printStackTrace();
}finally{
try{
if(br!=null){
br.close();
}
if(conn!=null){
conn.disconnect();
}
}catch(Exception e1){
e1.printStackTrace();
}
}
return str;
}
我将InputStream流和InputStreamReader流都单独定义引用是为了测试流关闭的问题，最后只关闭最外层流，因为外层流关闭的时候就将包裹的流一并关闭了。见下面的BufferedReader类的close方法：
public void close() throws IOException {
synchronized (lock) {
if (in == null)
return;
in.close();
in = null;
cb = null;
}
}其中，in就是包裹的流。
下面贴post方法的代码，然后综合比较一下，代码：
public String doPost(String grantType, String appId, String secret){
HttpURLConnection conn = null;
InputStream is = null;
InputStreamReader reader = null;
BufferedReader br = null;
String str = "";
try {
URL url = new URL(ACCESS_TOKEN_WEIXIN);
conn = (HttpURLConnection)url.openConnection();
conn.setRequestMethod("POST");
conn.setDoOutput(true);//默认为false的，所以需要设置
conn.setUseCaches(false);
conn.setRequestProperty("Content-Type", "application/x-www-form-urlencoded");
conn.connect();
OutputStream outStream = conn.getOutputStream();
DataOutputStream out = new DataOutputStream(outStream);
String params = "grant_type="+grantType+
"&appid="+appId+"&secret="+secret;
out.writeBytes(params);
out.close();
is = conn.getInputStream();
reader = new InputStreamReader(is, "UTF-8");
br = new BufferedReader(reader);
String readLine = "";
while((readLine=br.readLine())!=null){
str+=readLine+"\n";
}
} catch (Exception e) {
e.printStackTrace();
}finally{
try{
if(br!=null){
br.close();
}
if(conn!=null){
conn.disconnect();
}
}catch(Exception e1){
e1.printStackTrace();
}
}
return str;
}
两相比较：
1.思路是一样的：首先建立链接对象（URL）->由链接对象得到代理对象（HttpUrlConnection）->设置一些链接参数->启动链接通道（conn.connect();）->进行输出\输入操作
2.get方法参数是带在链接上的，所以没有单独赋值操作，直接在启动链接通道后得到输入流即可得到返回值
3.post方法参数不带在链接上，需要单独赋值（赋值是使用输出流来进行的），需要将输出打开（conn.setDoOutput(true);）因为URLConnection默认输出是false；而input是默认打开的，故不用重复设置。
4.读取返回值方式一样。
在研究过程中，对于URLConnection怎样得到的InputStream比较好奇，查看源码为：
public InputStream getInputStream() throws IOException {
throw new UnknownServiceException("protocol doesn't support input");
}
所以很迷惑，是jdk1.7 。等下来研究出来再补上来。
共勉！
-----------------------------------------------------------------
紧跟上面，对于jdk怎样实现的getInputStream和getOutputStream以及其他一些方法，经过研究，终于找到了。接下来释疑：
需要注意到：
URL url = new URL(ACCESS_TOKEN_WEIXIN);
conn = (HttpURLConnection)url.openConnection(); 这里得到HttpUrlConnection的时候是使用了强转得到的，故跟踪到URL类的openConnection方法内->
public URLConnection openConnection() throws java.io.IOException {
return handler.openConnection(this);
} 这里的handler定义在URL类的上方->
transient URLStreamHandler handler;
然后去寻找handler的赋值方法getURLStreamHandler，发现如下代码->
if (handler == null) {
String packagePrefixList = null;
packagePrefixList
= java.security.AccessController.doPrivileged(
new sun.security.action.GetPropertyAction(
protocolPathProp,""));
if (packagePrefixList != "") {
packagePrefixList += "|";
}
// REMIND: decide whether to allow the "null" class prefix
// or not.
packagePrefixList += JDK_PACKAGE_PREFIX;
StringTokenizer packagePrefixIter =
new StringTokenizer(packagePrefixList, "|");
while (handler == null &&
packagePrefixIter.hasMoreTokens()) {
String packagePrefix =
packagePrefixIter.nextToken().trim();
// do not try to instantiate the JDK gopher handler
// unless the system property had been explicitly set
if (protocol.equalsIgnoreCase(GOPHER) &&
packagePrefix.equals(JDK_PACKAGE_PREFIX) &&
!enableGopher) {
continue;
}
try {
String clsName = packagePrefix + "." + protocol +
".Handler";
Class cls = null;
try {
cls = Class.forName(clsName);
其中
JDK_PACKAGE_PREFIX在类中定义的是：
private static final String JDK_PACKAGE_PREFIX = "sun.net.www.protocol";再结合下面的
String clsName = packagePrefix + "." + protocol +
".Handler";基本可以定位到handler类的实现类在sun.net.www.protocol.http.Handler。然后百度搜索之后找到源码（jdk下没有sun包的源码）->
sun.net.www.protocol.http包的视图 Handler.java的源码 ->
protected java.net.URLConnection openConnection(URL u)
throws IOException {
return openConnection(u, (Proxy)null);
}
protected java.net.URLConnection openConnection(URL u, Proxy p)
throws IOException {
return new HttpURLConnection(u, p, this);
}
可以看到返回值是HttpURLConnection，但是发现该类中并没有引入java.net.HttpURLConnection类，故猜测在Handler的同包下有一个同名类，果然发现该类，
源码 ->
public class HttpURLConnection extends java.net.HttpURLConnection
这是关建行，该类继承了java.net.HttpURLConnection类。所以返回该类可以由java.net.HttpURLConnection类的父类URLConnection接到，然后强转成java.net.HttpURLConnection类。设计十分巧妙->
@Override
public synchronized OutputStream getOutputStream() throws IOException {
try {
if (!doOutput) {
throw new ProtocolException("cannot write to a URLConnection"
+ " if doOutput=false - call setDoOutput(true)");
}
if (method.equals("GET")) {
method = "POST"; // Backward compatibility
}
if (!"POST".equals(method) && !"PUT".equals(method) &&
"http".equals(url.getProtocol())) {
throw new ProtocolException("HTTP method " + method +
" doesn't support output");
}
// if there's already an input stream open, throw an exception
if (inputStream != null) {
throw new ProtocolException("Cannot write output after reading input.");
}
if (!checkReuseConnection())
connect();......
这就是getOutputStream的实现地方了，正好doOutput属性的设置也在这里起到了作用。艾玛，终于找到实现的地方了，其他方法原理一致了。
稍微整理下：整个的类流转过程就是：sun.net.www.protocol.http.HttpURLConnection转化成java.net.URLConnection，然后强转成java.net.HttpURLConnection。
晕乎~
