简单理解socket模拟浏览器请求

brower.py

# author: the5fire.com
import re
import socket
from collections import namedtuple

RE_URL = re.compile(r'http://(.*)')
RE_CONTENT_LENGTH = re.compile(b'.*Content-Length: (\d+)')
EOF = '\r\n\r\n'

Response = namedtuple('Response', ['header', 'body'])


class Browser:
    def parse_url(self, url):
        search = RE_URL.search(url)
        result = search.group(1)
        domain, path = result.split('/', 1)
        return domain, '/' + path

    def build_payload(self, domain, path, header):
        segments = [
            'GET %s HTTP/1.1' % path,
            'Host: %s' % domain,
        ]
        if header:
            segments.extend(['%s: %s' % (k, v) for k, v in header.items()])
        payload = '\r\n'.join(segments) + EOF
        return payload.encode('utf-8')

    def create_sock(self, domain):
        sock = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
        sock.connect((domain, 80))
        return sock

    def get(self, url, header=None):
        domain, path = self.parse_url(url)
        print(domain, path)
        payload = self.build_payload(domain, path, header)
        sock = self.create_sock(domain)
        print(payload)
        sock.sendall(payload)
        return self.create_response(sock)

    def create_response(self, sock):
        MSGLEN = 1014
        content_length = 0
        recv_length = 0
        is_head = True
        header = ''

        result = []
        while True:
            data = sock.recv(MSGLEN)
            if not content_length:
                re_search = RE_CONTENT_LENGTH.search(data)
                if re_search:
                    content_length = int(re_search.groups()[0])

            if is_head:
                header, data = data.split(b'\r\n\r\n')
                is_head = False

            result.append(data)
            recv_length += len(data)
            if content_length and recv_length >= content_length:
                break

        body = b''.join(result)
        return Response(header=header, body=body)


browser = Browser()

if __name__ == '__main__':
    response = browser.get('http://imgsrc.baidu.com/baike/pic/item/e850352ac65c1038bcba9c0eb9119313b17e8932.jpg')
    with open('download.jpg', 'wb') as f:
        f.write(response.body)