开始码代码之前，我们先来了解一下三种邮件服务协议：
1、SMTP协议
SMTP(Simple Mail Transfer Protocol)，即简单邮件传输协议。相当于中转站，将邮件发送到客户端。
2、POP3协议
POP3(Post Office Protocol 3)，即邮局协议的第3个版本，是电子邮件的第一个离线协议标准。该协议把邮件下载到本地计算机，不与服务器同步，缺点是更易丢失邮件或多次下载相同的邮件。
3、IMAP协议
IMAP(Internet Mail Access Protocol)，即交互式邮件存取协议。该协议连接远程邮箱直接操作，与服务器内容同步。
然后介绍一下email包
这个包的中心组件是代表电子邮件消息的“对象模型”。 应用程序主要通过在 message 子模块中定义的对象模型接口与这个包进行交互。 应用程序可以使用此 API 来询问有关现有电子邮件的问题、构造新的电子邮件，或者添加或移除自身也使用相同对象模型接口的电子邮件子组件。 也就是说，遵循电子邮件消息及其 MIME 子组件的性质，电子邮件对象模型是所有提供 EmailMessage API 的对象所构成的树状结构。
接下来我们通过具体的代码实现一个登录邮箱客户端，下载邮件，解析邮件附件内容的功能。
首先我们需要定义一个邮件解析的类，该类需要三个变量：
1、邮箱所属的imap服务地址；
2、邮箱账号；
3、邮箱密码【注：不同邮箱需要不同的安全策略，例如qq邮箱需要短信验证，获取登录授权码，而不是明文密码去登录远程客户端】
class Email_parse: def __init__(self,remote_server_url,email_url,password): # imap服务地址 self.remote_server_url = remote_server_url # 邮箱账号 self.email_url = email_url # 邮箱密码 self.password = password
然后定义类中入口函数，登录远程，默认获取第一页所有的邮件。我们获取邮件的主题，并打印出来【不同邮件主题的编码可能不同，二进制需要转码才能正确显示】
def main_parse_Email(self):
"""入口函数，登录imap服务"""
server = imaplib.IMAP4_SSL(self.remote_server_url, 993)
server.login(self.email_url, self.password)
server.select('INBOX')
status,data = server.search(None,"ALL")
if status != 'OK':
raise Exception('read email error')
emailids = data[0].split()
mail_counts = len(emailids)
print("count:",mail_counts)
# 邮件的遍历是按时间从后往前，这里我们选择最新的一封邮件
for i in range(mail_counts - 1, mail_counts - 2, -1):
status, edata = server.fetch(emailids[i], '(RFC822)')
msg = email.message_from_bytes(edata[0][1])
#获取邮件主题title
subject = email.header.decode_header(msg.get('subject'))
if type(subject[-1][0]) == bytes:
title = subject[-1][0].decode(str(subject[-1][1]))
elif type(subject[-1][0]) == str:
title = subject[-1][0]
print("title:", title)
其中，msg变量保存的就是邮件的主体，接下来因为会重复用到msg和tilte，我们将构造一个类函数返回msg和title。
def get_email_title(msg):
subject = email.header.decode_header(msg.get('subject'))
if type(subject[-1][0]) == bytes:
title = subject[-1][0].decode(str(subject[-1][1]))
elif type(subject[-1][0]) == str:
title = subject[-1][0]
print("title:", title)
return title
解析邮件，我们分为两部分，邮件正文【HTML】和附件【xlsx等】，判断有附件，我们就保存到固定的路径下。表格的解析不再赘述了，pandas之类的包足以搞定。
def get_att(msg):
"""获取附件并下载"""
filename = Email_parse.get_email_name(msg)
for part in msg.walk():
file_name = part.get_param("name")
if file_name:
data = part.get_payload(decode=True)
if data != None:
att_file = open('./src/' + filename, 'wb')
att_file.write(data)
att_file.close()
else:
pass
邮件正文内容，我们直接解析html，将文本内容直接保存到.txt文件中，方便读取。
def get_text_from_HTML(msg):
"""获取邮件中的html"""
filename = Email_parse.get_email_name(msg)
current_title = Email_parse.get_email_title(msg)
print("filename:",filename,type(filename))
for part in msg.walk():
if not part.is_multipart():
result = part.get_payload(decode=True)
result = result.decode('gbk')
f = open(f'./src/{current_title}.txt','w')
f.write(result)
f.close()
return result
完整代码如下：
import email
import imaplib
from email.header import decode_header
import pandas as pd
import datetime
class Email_parse:
def __init__(self,remote_server_url,email_url,password):
self.remote_server_url = remote_server_url
self.email_url = email_url
self.password = password
def get_att(msg):
filename = Email_parse.get_email_name(msg)
for part in msg.walk():
file_name = part.get_param("name")
if file_name:
data = part.get_payload(decode=True)
if data != None:
att_file = open('./src/' + filename, 'wb')
att_file.write(data)
att_file.close()
else:
pass
def get_email_title(msg):
subject = email.header.decode_header(msg.get('subject'))
if type(subject[-1][0]) == bytes:
title = subject[-1][0].decode(str(subject[-1][1]))
elif type(subject[-1][0]) == str:
title = subject[-1][0]
print("title:", title)
return title
def get_email_name(msg):
for part in msg.walk():
file_name = part.get_param("name")
if file_name:
h = email.header.Header(file_name)
dh = email.header.decode_header(h)
filename = dh[0][0]
if dh[0][1]:
value, charset = decode_header(str(filename, dh[0][1]))[0]
if charset:
filename = value.decode(charset)
print("附件名称：", filename)
return filename
def main_parse_Email(self):
server = imaplib.IMAP4_SSL(self.remote_server_url, 993)
server.login(self.email_url, self.password)
server.select('INBOX')
status,data = server.search(None,"ALL")
if status != 'OK':
raise Exception('read email error')
emailids = data[0].split()
mail_counts = len(emailids)
print("count:",mail_counts)
for i in range(mail_counts - 1, mail_counts - 2, -1):
status, edata = server.fetch(emailids[i], '(RFC822)')
msg = email.message_from_bytes(edata[0][1])
subject = email.header.decode_header(msg.get('subject'))
if type(subject[-1][0]) == bytes:
title = subject[-1][0].decode(str(subject[-1][1]))
elif type(subject[-1][0]) == str:
title = subject[-1][0]
print("title:", title)
Email_parse.get_att(msg)
Email_parse.get_text_from_HTML(msg)
def get_text_from_HTML(msg):
filename = Email_parse.get_email_name(msg)
current_title = Email_parse.get_email_title(msg)
print("filename:",filename,type(filename))
for part in msg.walk():
if not part.is_multipart():
result = part.get_payload(decode=True)
result = result.decode('gbk')
f = open(f'./src/{current_title}.txt','w')
f.write(result)
f.close()
return result
if __name__ == "__main__":
remote_server_url = 'imap.qq.com'
email_url = "*********@qq.com"
password = "**********"
demo = Email_parse(remote_server_url,email_url,password)
demo.main_parse_Email()
运行结果：
