[Nutch-dev] [PATCH] enable http proxy

[Sami Siren]

hello all,

this small patch enables usage of http proxy in fetching process. 

proxy is configured with two properties in config file:

<property>
 <name>http.proxy.host</name>
 <value>[!!your proxy host here!!]</value>
 <description>The proxy hostname.</description>
</property>
<property>
 <name>http.proxy.port</name>
 <value>[!!your proxy port here!!]</value>
 <description>The proxy port.</description>
</property>
possible future enhancements might include
-enabling persistent connections to proxy
-proxy support for other protocols but http (many proxies do also 
atleast ftp)



br,

Sami Siren
Index: Http.java
===================================================================
RCS file: /cvsroot/nutch/nutch/src/java/net/nutch/net/protocols/http/Http.java,v
retrieving revision 1.4
diff -u -r1.4 Http.java
--- Http.java   23 Apr 2004 19:32:33 -0000      1.4
+++ Http.java   29 Apr 2004 19:01:42 -0000
@@ -36,6 +36,10 @@
   /** Always indicates the latest HTTP version we support, currently 1.1 */
   public static final int HTTP_VER_LATEST;
 
+  String proxyHost=NutchConf.get("http.proxy.host");
+  int proxyPort=NutchConf.getInt("http.proxy.port",8080);
+  boolean proxyenabled=(proxyHost!=null && proxyHost.length()>0);
+  
   int timeout = NutchConf.getInt("http.timeout", 10000);
   int maxContentLength= NutchConf.getInt("http.content.limit",64*1024);
 
Index: HttpResponse.java
===================================================================
RCS file: 
/cvsroot/nutch/nutch/src/java/net/nutch/net/protocols/http/HttpResponse.java,v
retrieving revision 1.2
diff -u -r1.2 HttpResponse.java
--- HttpResponse.java   21 Apr 2004 22:52:35 -0000      1.2
+++ HttpResponse.java   29 Apr 2004 19:01:43 -0000
@@ -108,20 +108,23 @@
       socket.setSoTimeout(this.http.timeout);
 
       if (addr == null) {
-        addr= InetAddress.getByName(url.getHost());
+               
addr=InetAddress.getByName(this.http.proxyenabled?this.http.proxyHost:url.getHost());
         if (httpAccounting != null) 
           httpAccounting.setAddr(addr);
       }
 
       // connect
-      InetSocketAddress sockAddr= new InetSocketAddress(addr, port);
+      InetSocketAddress sockAddr= new InetSocketAddress(addr, 
this.http.proxyenabled?this.http.proxyPort:port);
       socket.connect(sockAddr, this.http.timeout);
 
       OutputStream req = socket.getOutputStream(); // make request
 
       StringBuffer reqStr = new StringBuffer("GET ");
-      reqStr.append(path);
-
+      if(this.http.proxyenabled){
+       
reqStr.append(url.getProtocol()).append("://").append(url.getHost()).append(portString).append(path);
+      } else {
+       reqStr.append(path);
+      }
 
       if (httpVersion == Http.HTTP_VER_1_1)
         reqStr.append(" HTTP/1.1\r\n");