1. 基础知识
----http get请求报文—
----请求行-----
GET / HTTP/1.1 =>请求方法(方式)请求的资源路径http协议的版本
----请求头----
Host: www.itcast.cn => 服务器的主机ip地址和端口号,提示如果看不到端口号默认是80
Connection: keep-alive => 和服务端程序保存长连接,当客户端和服务端有一段时间(3-5)没有进行通信,那么服务器程序会主动向客户端断开连接
Upgrade-Insecure-Requests: 1 => 让客户端请求不安全请求,以后要使用https
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_12_4) AppleWebKit/537.36 (KHTML, like Gecko)
Chrome/73.0.3683.86 Safari/537.36 => 用户代理,客户端程序名称,当后续讲爬虫的时候可以根据是否有User-Agent进行反爬机制
Accept: text/html,application/xhtml+xml,application/xml;q=0.9,image/webp,image/apng,/;q=0.8,application/signed-exchange;v=b3 => 告诉服务端程序可以接受的数据类型
Accept-Encoding: gzip, deflate => 告诉服务端程序支持的压缩算法
Accept-Language: zh-CN,zh;q=0.9,en;q=0.8 => 告诉服务端程序支持的语言
Cookie: UM_distinctid=169f06ab9d21a-0a52d93224ce3-12376d57-13c680-169f06ab9d571a; accessId=22bdcd10-6250-11e8-917f-9fb8db4dc43c; bad_id22bdcd10-6250-11e8-917f-9fb8db4dc43c=5c8a5a11-5811-11e9-975d-5932d6370ce7; parent_qimo_sid_22bdcd10-6250-11e8-917f-9fb8db4dc43c=69928890-5811-11e9-ac5b-eb2506ccad9a; CNZZDATA4617777=cnzz_eid%3D324231225-1554516145-%26ntime%3D1554521548; href=http%3A%2F%2Fwww.itcast.cn%2F; pageViewNum=5; Hm_lvt_0cb375a2e834821b74efffa6c71ee607=1554516722,1554523431; Hm_lpvt_0cb375a2e834821b74efffa6c71ee607=1554523431; nice_id22bdcd10-6250-11e8-917f-9fb8db4dc43c=fc5f90f1-5820-11e9-b896-dba72e2578c0; openChat22bdcd10-6250-11e8-917f-9fb8db4dc43c=true => 客户端用户身份的标识
-----空行-----
\r\n
-----http get请求的原始报文数据------------
----请求行-----
GET / HTTP/1.1\r\n
----请求头----
Host: www.itcast.cn\r\n
Connection: keep-alive\r\n
Upgrade-Insecure-Requests: 1\r\n
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_12_4) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/73.0.3683.86 Safari/537.36\r\n
Accept: text/html,application/xhtml+xml,application/xml;q=0.9,image/webp,image/apng,/;q=0.8,application/signed-exchange;v=b3\r\n
Accept-Encoding: gzip, deflate\r\n
Accept-Language: zh-CN,zh;q=0.9,en;q=0.8\r\n
Cookie: UM_distinctid=169f06ab9d21a-0a52d93224ce3-12376d57-13c680-169f06ab9d571a; accessId=22bdcd10-6250-11e8-917f-9fb8db4dc43c; bad_id22bdcd10-6250-11e8-917f-9fb8db4dc43c=5c8a5a11-5811-11e9-975d-5932d6370ce7; parent_qimo_sid_22bdcd10-6250-11e8-917f-9fb8db4dc43c=69928890-5811-11e9-ac5b-eb2506ccad9a; CNZZDATA4617777=cnzz_eid%3D324231225-1554516145-%26ntime%3D1554521548; href=http%3A%2F%2Fwww.itcast.cn%2F; pageViewNum=5; Hm_lvt_0cb375a2e834821b74efffa6c71ee607=1554516722,1554523431; Hm_lpvt_0cb375a2e834821b74efffa6c71ee607=1554523431; nice_id22bdcd10-6250-11e8-917f-9fb8db4dc43c=fc5f90f1-5820-11e9-b896-dba72e2578c0; openChat22bdcd10-6250-11e8-917f-9fb8db4dc43c=true\r\n
-----空行-----
\r\n
-----http get请求报文的格式----
请求行\r\n
请求头\r\n
空行(\r\n)
提示: 每项信息之间都需要一个\r\n,是要http协议规定
-----http post请求报文的格式----
请求行\r\n
请求头\r\n
空行(\r\n)
请求体
提示: 请求体就是浏览器发送给服务器的数据
----http 响应报文解析----
---- 响应行(状态行) --------
HTTP/1.1 200 OK => http协议版本 状态码 状态描述
---- 响应头 ----------
Server: Tengine => 服务器的名称
Content-Type: text/html; charset=UTF-8 => 服务器发送给浏览器的内容类型及编码格式
Transfer-Encoding: chunked => 服务器发送给客户端程序(浏览器)的数据不确定数据长度, 数据发送结束的接收标识: 0\r\n,Content-Length: 200(字节),服务器发送给客户端程序的数据确定长度。 内容长度这两个选项只能二选一
Connection: keep-alive => 和客户端保持长连接
Date: Sat, 06 Apr 2019 08:49:57 GMT => 服务器的时间
— 以下都是自定义响应头信息,字节定义响应头的名字和响应头的值,
比如: is_login: True
Accept-Ranges: bytes
Ali-Swift-Global-Savetime: 1554540597
Via: cache45.l2nu29-1[3,200-0,M], cache11.l2nu29-1[4,0], kunlun2.cn249[30,200-0,M], kunlun2.cn249[33,0]
X-Cache: MISS TCP_MISS dirn:-2:-2
----- 空行 ----
\r\n
----- 响应体 就是真正意义上给浏览器解析使用的数据----
网页数据
提示: 对于请求头和响应头信息程序员都可以进行自定义,按照客户端和服务器约定好的方式来制定即可。
----http 响应原始报文解析----
---- 响应行(状态行) --------
HTTP/1.1 200 OK\r\n
---- 响应头 ----------
Server: Tengine\r\n
Content-Type: text/html; charset=UTF-8\r\n
Transfer-Encoding: chunked\r\n
Connection: keep-alive\r\n
Date: Sat, 06 Apr 2019 08:49:57 GMT\r\n
— 以下都是自定义响应头信息,字节定义响应头的名字和响应头的值,比如:
is_login: True
Accept-Ranges: bytes\r\n
Ali-Swift-Global-Savetime: 1554540597\r\n
Via: cache45.l2nu29-1[3,200-0,M], cache11.l2nu29-1[4,0], kunlun2.cn249[30,200-0,M], kunlun2.cn249[33,0]\r\n
X-Cache: MISS TCP_MISS dirn:-2:-2\r\n
X-Swift-SaveTime: Sat, 06 Apr 2019 08:49:57 GMT\r\n
X-Swift-CacheTime: 0\r\n
Timing-Allow-Origin: *\r\n
EagleId: 2a51041615545405973986157e\r\n
----- 空行 ----
\r\n
----- 响应体 就是真正意义上给浏览器解析使用的数据----
网页数据
---- http响应报文的格式 -----
响应行\r\n
响应头\r\n
空行\r\n
响应体\r\n
提示: 每项信息之间都要有一个\r\n进行分割
2. 开发步骤
- HTTP报文底层通过tcp传输报文
- 对收到的报文解析出所需信息如:请求页面
- 对报文组装:请求行+头+换行+体(真正所需页面)
- 一些可能错误处理如:空数据、请求
/
、错误请求 - 提供多线程进行多任务处理
3. 固定页面
import socket
if __name__ == '__main__':
# 创建服务端套接字
tcp_web_socket = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
# 设置端口复用
tcp_web_socket.setsockopt(socket.SOL_SOCKET, socket.SO_REUSEADDR, True)
# 绑定端口号
tcp_web_socket.bind(('', 8000))
#设置监听
tcp_web_socket.listen(128)
# 不断循环实现多次访问,但是单线程的
while True:
# 接收生成新的套接字
new_socket, ip_port = tcp_web_socket.accept()
# 接收数据
recv_data = new_socket.recv(4096)
# 对二进制数据解码
recv_content = recv_data.decode('utf-8')
# 对发送的数据组装处理
# 1. 响应行
response_line = 'HTTP/1.0 200 OK\r\n'
# 2. 响应头
response_header = 'Server: ahang\r\n'
# 3. 读取响应体
f = open('static\index.html', 'r')
response_body = f.read()
f.close()
# 4. 组装响应数据为HTTP的response报文格式
response_content = response_line + response_header + '\r\n' + response_body
# 5. 对数据编码为二进制,有利于不同格式数据的传输,如图片
response_data = response_content.encode('utf-8')
# 6. 发送数据
new_socket.send(response_data)
4. 指定页面
import socket
if __name__ == '__main__':
tcp_web_socket = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
tcp_web_socket.setsockopt(socket.SOL_SOCKET, socket.SO_REUSEADDR, True)
tcp_web_socket.bind(('', 8000))
tcp_web_socket.listen(128)
while True:
# 至此建立了tcp的连接,可以进行数据的传输
new_socket, ip_port = tcp_web_socket.accept()
# 对数据接收处理
recv_data = new_socket.recv(4096)
# 情况一:当传来的数据为空时,可能为恶意传输,关闭此次连接
if len(recv_data) == 0:
new_socket.close()
# 对传入的二进制解码
recv_content = recv_data.decode('utf-8')
# 对传入的HTTP的request报文分离出想要的信息,请求的网页名称
recv_list = recv_content.split(' ', 2)
recv_path = recv_list[1]
print(recv_path)
try:
# 情况二:如果为根路径,重定向到index.html默认页面
if recv_path == '/':
recv_path = '/index.html'
# 对请求页面打开,如果能访问则以读取二进制方式打开,不能访问则不存在请求的页面,放入异常
f = open('static' + recv_path, 'rb')
except Exception as result:
# 情况三:对不存在的页面重定位到错误页面
print(result)
response_line = 'HTTP/1.0 404 NotFund\r\n'
f = open('static/error.html', 'rb')
response_body = f.read()
response_header = 'Server: ahang\r\n'
f.close()
response_data = (response_line + response_header + '\r\n').encode('utf-8') + response_body
new_socket.send(response_data)
else:
# 对正常的页面返回
response_line = 'HTTP/1.0 200 OK\r\n'
response_header = 'Server: ahang\r\n'
response_body = f.read()
f.close()
response_data = (response_line + response_header + '\r\n').encode('utf-8') + response_body
new_socket.send(response_data)
finally:
# 最后都要关闭该连接
new_socket.close()
5. 多任务版
import socket
import threading
# 对功能函数独立,子线程对数据的处理
def web_server(new_socket):
# 接收数据请求,指定每次接收的大小
recv_data = new_socket.recv(4096)
if len(recv_data) == 0:
new_socket.close()
return
recv_content = recv_data.decode('utf-8')
recv_list = recv_content.split(' ', 2)
recv_path = recv_list[1]
print(recv_path)
try:
# 默认页面
if recv_path == '/':
recv_path = '/index.html'
f = open('static'+recv_path, 'rb')
except Exception as result:
print(result)
f = open('static/error.html', 'rb')
response_body = f.read()
f.close()
response_line = 'HTTP/1.0 404 ERROR\r\n'
response_header = 'Server: ahang\r\n'
send_data = (response_line + response_header + '\r\n').encode('utf-8') + response_body
new_socket.send(send_data)
else:
response_body = f.read()
f.close()
response_line = 'HTTP/1.0 200 OK\r\n'
response_header = 'Server: ahang\r\n'
send_data = (response_line + response_header + '\r\n').encode('utf-8') + response_body
new_socket.send(send_data)
finally:
new_socket.close()
if __name__ == '__main__':
tcp_web_socket = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
tcp_web_socket.setsockopt(socket.SOL_SOCKET, socket.SO_REUSEADDR, True)
tcp_web_socket.bind(('', 8000))
tcp_web_socket.listen(128)
while True:
new_socket, ip_port = tcp_web_socket.accept()
print(f'与客户端{ip_port}建立连接成功!!!')
# 创建子线程且设为守护主线程
sub_http_thread = threading.Thread(target=web_server, args=(new_socket,), daemon=True)
sub_http_thread.start()
# tcp_web_socket.close()
6. 面向对象版静态web服务器且可以通过命令行指定端口
C:\Users\study\web编程>python 05_通过命令行传递参数版web静态服务器.py 9000
import socket
import threading
import sys
# 定义服务器的类
class HttpWebServer(object):
# 初始化类,定义套接字的属性
def __init__(self, port):
self.tcp_web_socket = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
self.tcp_web_socket.setsockopt(socket.SOL_SOCKET, socket.SO_REUSEADDR, True)
self.tcp_web_socket.bind(('', port))
self.tcp_web_socket.listen(128)
@staticmethod
def web_server(new_socket):
# 接收数据请求,指定每次接收的大小
recv_data = new_socket.recv(4096)
if len(recv_data) == 0:
new_socket.close()
return
recv_content = recv_data.decode('utf-8')
recv_list = recv_content.split(' ', 2)
recv_path = recv_list[1]
print(recv_path)
try:
# 默认页面
if recv_path == '/':
recv_path = '/index.html'
f = open('static'+recv_path, 'rb')
except Exception as result:
print(result)
f = open('static/error.html', 'rb')
response_body = f.read()
f.close()
response_line = 'HTTP/1.0 404 ERROR\r\n'
response_header = 'Server: ahang\r\n'
send_data = (response_line + response_header + '\r\n').encode('utf-8') + response_body
new_socket.send(send_data)
else:
response_body = f.read()
f.close()
response_line = 'HTTP/1.0 200 OK\r\n'
response_header = 'Server: ahang\r\n'
send_data = (response_line + response_header + '\r\n').encode('utf-8') + response_body
new_socket.send(send_data)
finally:
new_socket.close()
def start(self):
while True:
new_socket, ip_port = self.tcp_web_socket.accept()
print(f'与客户端{ip_port}建立连接成功!!!')
# 创建子线程且设为守护主线程
sub_http_thread = threading.Thread(target=self.web_server, args=(new_socket,), daemon=True)
sub_http_thread.start()
def main():
# 从命令行传入参数,形成列表
params = sys.argv
if (len(params) != 2) or (not params[1].isdigit()):
print('请按照指定格式输入:python3 xxxx 8000')
return
port = int(params[1])
httpsever = HttpWebServer(port)
httpsever.start()
if __name__ == '__main__':
main()