【www.bbyears.com--python】
下面把主要的代码贴一下:
初始化,定义邮件服务器
self.IMAP_SERVER="imap.gmail.com"
self.IMAP_PORT=993
self.M = None
self.response
self.mailboxes = []
登录,选择mailbox:
self.M = imaplib.IMAP4_SSL(self.IMAP_SERVER, self.IMAP_POR
rc, self.response = self.M.login(username, password)
tye,data = m.M.select()
邮件搜索:
ret, msgnums = m.M.search(None, "BODY", datapath)
获取邮件信息:
status, response = self.M.fetch(id,"(RFC822)")
mailText = response[0][1]
mail_message = email.message_from_string(mailText)
subject = unicode(email.Header.make_header(email.Header.decode_header(mail_message["subject"])))
#print "subject_________:" +subject
mail_from = email.utils.parseaddr(mail_message["from"])[1]
mail_to = email.utils.parseaddr(mail_message["to"])[1]
time = mail_message["Date"]
print "["+mail_message["Date"]+"]"+"\n"+"From:"+mail_from+ " To:"+mail_to+"\n"+"Subject:"+subject+"\n"
return self.get_first_text_block(mail_message), subject, mail_from, time
maintype = email_message_instance.get_content_maintype() 返回邮件里的内容是何种类型,若为text就比较好处理,如果是multipart,还得遍历email_message_instance去根据不同类型处理。
email.message_from_string(mailText)返回了一个结构体,里面包含了邮件的基本信息
邮件里比较蛋疼的是字符串编码的问题,毕竟大家的邮件格式都不一样,有些是unicode,有些是utf-8,有些是gb2312,还有附件,图片等多种格式,
当然这次也只处理了文本,暂时没有需求去处理附件和图片这些。我都是统一将字符转成unicode去处理的。
字符串处理的时候,可以使用chardet判断字符串类型,读写文件的时候可以用codecs指定读写的字符集类型
补充三个例子
Example 1
From project georegistry, under directory georegistry/lib, in source file smtp.py.
Score: 13
vote
vote
def sendMessage(fromByValue, toByValue, subject, body, headerByName=None):
"Send a message using SMTP"
# Prepare
message = email.message.Message()
message.add_header("from", email.utils.formataddr((fromByValue["nickname"], fromByValue["email"])))
message.add_header("to", email.utils.formataddr((toByValue["nickname"], toByValue["email"])))
message.add_header("subject", subject)
message.set_payload(body)
if headerByName:
for key, value in headerByName.iteritems():
message.add_header(key, value)
# Connect to server
if fromByValue["smtp"] == "localhost":
server = smtplib.SMTP("localhost")
else:
server = smtplib.SMTP_SSL(fromByValue["smtp"], 465)
if len(fromByValue["username"]):
server.login(fromByValue["username"], fromByValue["password"])
# Send mail
try:
server.sendmail(fromByValue["email"], toByValue["email"], message.as_string())
except socket.error, error:
raise SMTPError(error)
finally:
server.quit()
Example 2
From project appengine-python3-master, under directory google/appengine/tools/devappserver2/admin, in source file mail_request_handler_test.py.
Score: 10
vote
vote
def test_send(self):
self.mox.StubOutWithMock(mail_request_handler.MailRequestHandler,
"dispatcher")
handler = mail_request_handler.MailRequestHandler(None, None)
handler.dispatcher = self.mox.CreateMock(dispatcher.Dispatcher)
handler.dispatcher.add_request(
method="POST",
relative_url="URL",
headers=[("Content-Type", "message/rfc822")],
body="mail message",
source_ip="0.1.0.20")
message = self.mox.CreateMock(email.message.Message)
message.as_string().AndReturn("mail message")
self.mox.ReplayAll()
handler._send("URL", message)
self.mox.VerifyAll()
Example 3
From project python-sipsimple-master, under directory sipsimple/streams/applications, in source file chat.py.
Score: 8
vote
vote
def __str__(self):
headers = []
if self.sender:
headers.append(u"From: %s" % self.sender)
for recipient in self.recipients:
headers.append(u"To: %s" % recipient)
for recipient in self.courtesy_recipients:
headers.append(u"cc: %s" % recipient)
if self.subject:
headers.append(u"Subject: %s" % self.subject)
if self.subject is not None:
for lang, translation in self.subject.translations.iteritems():
headers.append(u"Subject:;lang=%s %s" % (lang, translation))
if self.timestamp:
headers.append(u"DateTime: %s" % self.timestamp)
if self.required:
headers.append(u"Required: %s" % ",".join(self.required))
namespaces = {u"": self.standard_namespace}
for header in self.additional_headers:
if namespaces.get(header.namespace.prefix, None) != header.namespace:
if header.namespace.prefix:
headers.append(u"NS: %s <%s>" % (header.namespace.prefix, header.namespace))
else:
headers.append(u"NS: <%s>" % header.namespace)
namespaces[header.namespace.prefix] = header.namespace
if header.namespace.prefix:
headers.append(u"%s.%s: %s" % (header.namespace.prefix, header.name, header.value))
else:
headers.append(u"%s: %s" % (header.name, header.value))
headers.append(u"")
headers = "\r\n".join(s.encode("cpim-headers") for s in headers)
message = Message()
message.set_type(self.content_type)
if isinstance(self.body, unicode):
message.set_param("charset", "utf-8")
message.set_payload(self.body.encode("utf-8"))
else:
message.set_payload(self.body)
return headers + "\r\n" + message.as_string()
Example 4
From project odoo, under directory addons/mail, in source file mail_thread.py.
Score: 8
vote
vote
def _message_extract_payload(self, message, save_original=False):
"""Extract body as HTML and attachments from the mail message"""
attachments = []
body = u""
if save_original:
attachments.append(("original_email.eml", message.as_string()))
if not message.is_multipart() or "text/" in message.get("content-type", ""):
encoding = message.get_content_charset()
body = message.get_payload(decode=True)
body = tools.ustr(body, encoding, errors="replace")
if message.get_content_type() == "text/plain":
# text/plain ->
body = tools.append_content_to_html(u"", body, preserve=True)
else:
alternative = False
for part in message.walk():
if part.get_content_type() == "multipart/alternative":
alternative = True
if part.get_content_maintype() == "multipart":
continue # skip container
# part.get_filename returns decoded value if able to decode, coded otherwise.
# original get_filename is not able to decode iso-8859-1 (for instance).
# therefore, iso encoded attachements are not able to be decoded properly with get_filename
# code here partially copy the original get_filename method, but handle more encoding
filename=part.get_param("filename", None, "content-disposition")
if not filename:
filename=part.get_param("name", None)
if filename:
if isinstance(filename, tuple):
# RFC2231
filename=email.utils.collapse_rfc2231_value(filename).strip()
else:
filename=decode(filename)
encoding = part.get_content_charset() # None if attachment
# 1) Explicit Attachments -> attachments
if filename or part.get("content-disposition", "").strip().startswith("attachment"):
attachments.append((filename or "attachment", part.get_payload(decode=True)))
continue
# 2) text/plain ->
if part.get_content_type() == "text/plain" and (not alternative or not body):
body = tools.append_content_to_html(body, tools.ustr(part.get_payload(decode=True),
encoding, errors="replace"), preserve=True)
# 3) text/html -> raw
elif part.get_content_type() == "text/html":
html = tools.ustr(part.get_payload(decode=True), encoding, errors="replace")
if alternative:
body = html
else:
body = tools.append_content_to_html(body, html, plaintext=False)
# 4) Anything else -> attachment
else:
attachments.append((filename or "attachment", part.get_payload(decode=True)))
return body, attachments
Example 5
From project openerp-ktv, under directory openerp/addons/mail, in source file mail_message.py.
Score: 5
vote
vote
def parse_message(self, message, save_original=False):
"""Parses a string or email.message.Message representing an
RFC-2822 email, and returns a generic dict holding the
message details.
:param message: the message to parse
:type message: email.message.Message | string | unicode
:param bool save_original: whether the returned dict
should include an ``original`` entry with the base64
encoded source of the message.
:rtype: dict
:return: A dict with the following structure, where each
field may not be present if missing in original
message::
{ "message-id": msg_id,
"subject": subject,
"from": from,
"to": to,
"cc": cc,
"headers" : { "X-Mailer": mailer,
#.. all X- headers...
},
"subtype": msg_mime_subtype,
"body_text": plaintext_body
"body_html": html_body,
"attachments": [("file1", "bytes"),
("file2", "bytes") }
# ...
"original": source_of_email,
}
"""
msg_txt = message
if isinstance(message, str):
msg_txt = email.message_from_string(message)
# Warning: message_from_string doesn"t always work correctly on unicode,
# we must use utf-8 strings here :-(
if isinstance(message, unicode):
message = message.encode("utf-8")
msg_txt = email.message_from_string(message)
message_id = msg_txt.get("message-id", False)
msg = {}
if save_original:
# save original, we need to be able to read the original email sometimes
msg["original"] = message.as_string() if isinstance(message, Message) \
else message
msg["original"] = base64.b64encode(msg["original"]) # binary fields are b64
if not message_id:
# Very unusual situation, be we should be fault-tolerant here
message_id = time.time()
msg_txt["message-id"] = message_id
_logger.info("Parsing Message without message-id, generating a random one: %s", message_id)
fields = msg_txt.keys()
msg["id"] = message_id
msg["message-id"] = message_id
if "Subject" in fields:
msg["subject"] = decode(msg_txt.get("Subject"))
if "Content-Type" in fields:
msg["content-type"] = msg_txt.get("Content-Type")
if "From" in fields:
msg["from"] = decode(msg_txt.get("From") or msg_txt.get_unixfrom())
if "To" in fields:
msg["to"] = decode(msg_txt.get("To"))
if "Delivered-To" in fields:
msg["to"] = decode(msg_txt.get("Delivered-To"))
if "CC" in fields:
msg["cc"] = decode(msg_txt.get("CC"))
if "Cc" in fields:
msg["cc"] = decode(msg_txt.get("Cc"))
if "Reply-To" in fields:
msg["reply"] = decode(msg_txt.get("Reply-To"))
if "Date" in fields:
try:
date_hdr = decode(msg_txt.get("Date"))
parsed_date = dateutil.parser.parse(date_hdr, fuzzy=True)
if parsed_date.utcoffset() is None:
# naive datetime, so we arbitrarily decide to make it
# UTC, there"s no better choice. Should not happen,
# as RFC2822 requires timezone offset in Date headers.
stored_date = parsed_date.replace(tzinfo=pytz.utc)
else:
stored_date = parsed_date.astimezone(pytz.utc)
except Exception:
_logger.warning("Failed to parse Date header %r in incoming mail "
"with message-id %r, assuming current date/time.",
msg_txt.get("Date"), message_id)
stored_date = datetime.datetime.now()
msg["date"] = stored_date.strftime("%Y-%m-%d %H:%M:%S")
if "Content-Transfer-Encoding" in fields:
msg["encoding"] = msg_txt.get("Content-Transfer-Encoding")
if "References" in fields:
msg["references"] = msg_txt.get("References")
if "In-Reply-To" in fields:
msg["in-reply-to"] = msg_txt.get("In-Reply-To")
msg["headers"] = {}
msg["subtype"] = "plain"
for item in msg_txt.items():
if item[0].startswith("X-"):
msg["headers"].update({item[0]: item[1]})
if not msg_txt.is_multipart() or "text/plain" in msg.get("content-type", ""):
encoding = msg_txt.get_content_charset()
body = tools.ustr(msg_txt.get_payload(decode=True), encoding, errors="replace")
if "text/html" in msg.get("content-type", ""):
msg["body_html"] = body
msg["subtype"] = "html"
body = tools.html2plaintext(body)
msg["body_text"] = tools.ustr(body, encoding, errors="replace")
attachments = []
if msg_txt.is_multipart() or "multipart/alternative" in msg.get("content-type", ""):
body = ""
if "multipart/alternative" in msg.get("content-type", ""):
msg["subtype"] = "alternative"
else:
msg["subtype"] = "mixed"
for part in msg_txt.walk():
if part.get_content_maintype() == "multipart":
continue
encoding = part.get_content_charset()
filename = part.get_filename()
if part.get_content_maintype()=="text":
content = part.get_payload(decode=True)
if filename:
attachments.append((filename, content))
content = tools.ustr(content, encoding, errors="replace")
if part.get_content_subtype() == "html":
msg["body_html"] = content
msg["subtype"] = "html" # html version prevails
body = tools.ustr(tools.html2plaintext(content))
body = body.replace("
", "")
elif part.get_content_subtype() == "plain":
body = content
elif part.get_content_maintype() in ("application", "image"):
if filename :
attachments.append((filename,part.get_payload(decode=True)))
else:
res = part.get_payload(decode=True)
body += tools.ustr(res, encoding, errors="replace")
msg["body_text"] = body
msg["attachments"] = attachments
# for backwards compatibility:
msg["body"] = msg["body_text"]
msg["sub_type"] = msg["subtype"] or "plain"
return msg