1 通过thrift
但是数据量太大thrift不支持
下载thrift包安装，比如我的是/usr/local/thrift-0.8.0
/usr/local/thrift-0.8.0/lib类库
/usr/local/thrift-0.8.0/test是你需要的教程 可以通过教材方便使用thrift
通信thrift文件
helloworld .thrift
你要写什么格式可以参照 test里面
service HelloWorld{
string sayHello(1:string content, 2:string xpath)
}
thrift -gen python xxx.py
thrift -gen php xxx.py
python服务端 。
test.py
import sys
2 sys.path.append('/var/www/thrift/gen-py')
3 from helloworld import HelloWorld
4 from helloworld.ttypes import *
5 from thrift.transport import TSocket
6 from thrift.transport import TTransport
7 from thrift.protocol import TBinaryProtocol
8 from thrift.server import TServer
9 import socket
10 from lxml.html.soupparser import fromstring
11
12 class HelloWorldHandler:
13 def __init__(self):
14 self.log = {}
15
16 #def sayHello(self, content, xpath):
17 #content = """<html><head><title>why len(y) ==< 1</title><script>var y = 1</script></head>sample.<html>"""
18 def sayHello(self, content, xpath):
19 root = fromstring(content)
20 find_text = root.xpath(xpath)
21 return find_text[0].text
22
23 handler = HelloWorldHandler()
24 processor = HelloWorld.Processor(handler)
25 transport = TSocket.TServerSocket('127.0.0.1', 393939)
26 tfactory = TTransport.TBufferedTransportFactory()
27 pfactory = TBinaryProtocol.TBinaryProtocolFactory()
28 server = TServer.TSimpleServer(processor, transport, tfactory, pfactory)
29 print 'start server'
30 server.serve()
test.php
<?PHP
2 $GLOBALS['THRIFT_ROOT'] = dirname(__FILE__).'/Thrift';
3 require_once $GLOBALS['THRIFT_ROOT'].'/Thrift.php';
4 require_once $GLOBALS['THRIFT_ROOT'].'/protocol/TBinaryProtocol.php';
5 require_once $GLOBALS['THRIFT_ROOT'].'/transport/TSocket.php';
6 require_once $GLOBALS['THRIFT_ROOT'].'/transport/THttpClient.php';
7 require_once $GLOBALS['THRIFT_ROOT'].'/transport/TBufferedTransport.php';
8
9 // Your gen-php dir
10 $GEN_DIR = dirname(__FILE__).'/gen-php';
11 require_once $GEN_DIR . '/helloworld/HelloWorld.php';
12 require_once $GEN_DIR . '/helloworld/helloworld_types.php';
13
14 // Set server host and port
15 $host = "127.0.0.1";
16 $port = 393939;
17
18 try {
19
20 //Thrift connection handling
21 $socket = new TSocket( $host , $port, true );
22 $transport = new TBufferedTransport($socket, 1024, 1024);
23 $protocol = new TBinaryProtocol($transport);
24 #echo "<pre>";
25 #print_r($socket);
26 #print_r($transport);
27 #print_r($protocol);
28 // get our example client
29 $client = new HelloWorldClient($protocol);
30 $transport->open();
31
32 #$in = new ThriftTest_Page();
33 // Get current timestamp from server
34 $content = file_get_contents("http://www.baidu.com/");
35 $content = mb_convert_encoding($content, "UTF-8");
36 // $content = "<html><head><title>why len(y) ==< 1</title><script>var y = 1</script></head>sample.<html>";
37 // die;
38 $xpath = "//title";
39 #$in->content = $content;
40 #$in->xpath = $xpath;
41
42 $return = $client->sayHello($content, $xpath);
43 // $return = $client->sayHello($content, $xpath);
44 echo $return;
45
46 $transport->close(); } catch (TException $tx) {
49 print 'Something went wrong: '.$tx->getMessage()."\n";
50 }
51 ?>