CODEWITHSUNDEEP
pythonemailparsing
Basj
Basj

Reputation: 46411

Parse email fields

I want to parse email addresses from a To: email field.

Indeed, when looping on the emails in a mbox:

mbox = mailbox.mbox('test.mbox')
for m in mbox:
  print m['To']

we can get things like:

[email protected], Blahblah <[email protected]>, <[email protected]>, "Hey" <[email protected]>

That should be parsed into:

[{email: "[email protected]", name: ""}, 
 {email: "[email protected]", name: "Blahblah"},
 {email: "[email protected]", name: ""},
 {email: "[email protected]", name: "Hey"}]

Is there something already built-in (in mailbox or another module) for this or nothing?

I read a few times this doc but I didn't find something relevant.

Upvotes: 3

Views: 391

Answers (4)

Boldewyn
Boldewyn

Reputation: 82734

You can use email.utils.getaddresses() for this:

>>> getaddresses(['[email protected], Blahblah <[email protected]>, <[email protected]>, "Hey" <[email protected]>'])
[('', '[email protected]'), ('Blahblah', '[email protected]'), ('', '[email protected]'), ('Hey', '[email protected]')]

(Note that the function expects a list, so you have to enclose the string in [...].)

Upvotes: 5

Gaurav Dave
Gaurav Dave

Reputation: 7474

Python provides email.Header.decode_header() for decoding header. The function decode each atom and return a list of tuples ( text, encoding ) that you still have to decode and join to get the full text.

For addresses, Python provides email.utils.getaddresses() that split addresses in a list of tuple ( display-name, address ). display-name need to be decoded too and addresses must match the RFC2822 syntax. The function getmailaddresses() does all the job.

Here's a tutorial that might help http://blog.magiksys.net/parsing-email-using-python-header

Upvotes: 0

Basj
Basj

Reputation: 46411

As pointed by @TheSpooniest, email has a parser:

import email

s = '[email protected], Blahblah <[email protected]>, <[email protected]>, "Hey" <[email protected]>'

for em in s.split(','):
    print email.utils.parseaddr(em)

gives:

('', '[email protected]')
('Blahblah', '[email protected]')
('', '[email protected]')
('Hey', '[email protected]')

Upvotes: 1

The Spooniest
The Spooniest

Reputation: 2873

email.parser has the modules you're looking for. email.message is still relevant, because the parser will return messages using this structure, so you'll be getting your header data from that. But to actually read the files in, email.parser is the way to go.

Upvotes: 1

Related Questions