My setup uses fetchmail to pull emails from Gmail, which are processed by procmail and passes it to a python script.

When I use email.message_from_string(), the resulting object is not parsed as an email object. get_payload() returns the header/body/payload text of the email as a single text blob.

This is the text it returns:

From example@gmail.com Sat Aug 17 19:20:44 2013

>From example Sat Aug 17 19:20:44 2013

MIME-Version: 1.0

Received: from ie-in-f109.1e100.net [74.125.142.109]

by VirtualBox with IMAP (fetchmail-6.3.21)

for (single-drop); Sat, 17 Aug 2013 19:20:44 -0700 (PDT)

Received: by 10.70.131.110 with HTTP; Sat, 17 Aug 2013 19:20:42 -0700 (PDT)

Date: Sat, 17 Aug 2013 19:20:42 -0700

Delivered-To: example@gmail.com

Message-ID:

Subject: test 19:20

From: example

To: example

Content-Type: multipart/alternative; boundary=001a1133435474449004e42f7861

--001a1133435474449004e42f7861

Content-Type: text/plain; charset=ISO-8859-1

19:20

--001a1133435474449004e42f7861

Content-Type: text/html; charset=ISO-8859-1

--001a1133435474449004e42f7861--

My code:

full_msg = sys.stdin.read()

msg = email.message_from_string(full_msg)

msg['to'] # returns None

msg.get_payload() # returns the text above

What am I missing to get Python to properly interpret the email?

I see from these questions that I may not be getting the proper email headers somewhere along the line, but I cannot confirm. That ">" on line 2 is not a typo: it's in the text.

解决方案

Regardless of ">" being "in the text" as you say, whatever that means - it's wrong. After removing this character:

>python test.py

example

[,

]

So the error is not in parsing the message, but in the ">" character somehow corrupting your email text.