python 读取eml或者imap 读取邮件时图片处理记录

使用python读取eml文件或者远程imap读取邮件时,邮件里面有内嵌图片,源地址是 img src=3D"cid:_Foxmail.1@bebb7=cba-548d-cdd3-c8df-af59616af7cc" 之类的,可以在读取邮件时将图片保存到缓存目录,然后将邮件内容里面的src源地址替换为刚才保存的路径,折腾半天实现了效果,记录下过程

import imaplib,email
import quopri
from email.mime.text import MIMEText
from flask import Response
from flask import make_response
 
@app.route('/')
def index():
    conn = imaplib.IMAP4("127.0.0.1",143)
    conn.login("user@domain.ld","password")
    conn.select("INBOX")
    ''' python3 要使用str() 否则提示 TypeError: can't concat bytes to int '''
    stat,data = conn.fetch(str(2),'(RFC822)') 
    if stat != "OK" or data is None:
        raise KeyError
    ''' 转为message对象,python2使用email.message_from_string '''
    msg = email.message_from_bytes(data[0][1])
    attlist= {}
    for part in msg.walk():
        if not part.is_multipart():
            file = part.get_filename()  #附件名
            ''' filename = email.header.decode_header(filename) ==> [(b'\xe5\x8a\xb3\xe5\x8a\xa8\xe6\xb3\x95.png', 'utf-8')]  '''
            if file:
                filename = email.header.decode_header(file)[0][0]  #附件名
                charset = email.header.decode_header(file)[0][1]   #编码
                if part.get_all("Content-ID"):
                    content_id = part.get_all("Content-ID")[0][1:-1]
                else:
                    content_id = ""   #附件ID,也就是邮件源码里面的cid
                ''' 多个附件时将附件名和ID对应保存到dict里面,后面将正文中的cid替换为本地保存路径 '''
                attlist[content_id] = filename
                ''' 附件文件名为中文或有编码的时候要进行转码 '''
                if str(charset) != "None":
                    filename = filename.decode(charset)
                filedata = part.get_payload(decode=True)   # 附件内容
                ''' 把附件写到文件里面,附件一定要用wb打开,二进制 '''
                with open("/temp/"+filename,"wb") as fw:
                    fw.write(filedata)
                    fw.close()  
            else:
                content = part.get_payload(decode=True) # 邮件正文
    contentd = content.decode()
 
    for x,y in attlist.items():
        ''' 替换cid '''
        contentd = contentd.replace('cid:'+x,'static/temp/'+y)
    return Response(quopri.decodestring(contentd),content_type="text/html; charset=%s" %charset)  ##使用quopri转码,不然不能转码源码中的=OA 等字符,为防止乱码,在response时返回邮件内容中的编码