Question

I want to extract the first number found in an e-mail body. With the help with email library I extracted just the body from the mail to a string. But the problem is that before the real plain-text body begins there is some info about encoding and such (those contains digits). How can I skip those on a reliable way that is undependent on which client that created the e-mail and just ge the first number.

If I do a

match = re.search('\d+', string, re.MULTILINE)

It will get the first match in the info about encoding or something and not in the actual mail content.

Ok. I add a sample. This is how it could look (i will extract 123). But I suppose it could look different went sent from another client.

--14dae93404410f62f404b2e65e10
Content-Type: text/plain; charset=ISO-8859-1

Junk 123 Junk

--14dae93404410f62f404b2e65e10
Content-Type: text/html; charset=ISO-8859-1

<p>Junk 123 Junk</p>

--14dae93404410f62f404b2e65e10--

---

Update:
Now I’m stuck with the iterator :-/ I really tried. But I don’t get it. This code:

msg = email.message_from_string(raw_message)
for part in email.iterators.typed_subpart_iterator(msg, 'text', 'plain'):
    print part

outputs:

--14dae93404410f62f404b2e65e10
Content-Type: text/plain; charset=ISO-8859-1

Junk 123 Junk

--14dae93404410f62f404b2e65e10
Content-Type: text/html; charset=ISO-8859-1

<p>Junk 123 Junk</p>

--14dae93404410f62f404b2e65e10--

Why won’t it just output:

Junk 123 Junk

?

Answer 1

You may want to use the iterators to skip over the subpart headers.

http://docs.python.org/library/email.iterators.html#module-email.iterators

This example will print the body of each message subpart that is text/plain:

for part in email.iterators.typed_subpart_iterator(msg, 'text', 'plain'):
   for body_line in email.iterators.body_line_iterator(part):
       print body_line