imaplib and fetch results

[Piers Lauder]

On 27 Jan 2000 17:22:29 +0100, Harald Terkelsen wrote:

 > I've been playing with the imaplib.py module and found some difference
 > in the returned data of the fetch function depending on the
 > message_part argument.
 > 
 > The command:
 > 
 > typ, data = M.FETCH('13:17', '(FLAGS envelope)')
 > 
 > returns a list where data[0] is the flags and envelope data from
 > message 13, data[1] for message 14 and so on. This is what i would
 > expect.
 > 
 > But the result of the following command is somewhat different:
 > 
 > typ, data = M.FETCH('13:17', '(FLAGS body.peek[header.fields (From Subject)])')
 > 
 > Now data[0] contains the information from message 13 as in the previous
 > fetch command. I would then assume data[1] to contain the requested
 > information for message 14, but it doesn't. Instead data[1] contains
 > the character ')'. data[2] contains the data for message 14 and
 > data[3] gives me again an ')' and so on.
 > 
 > Does anybody know why it is like this, or is it a bug?

This is a feature, not a bug :-)

The reason is that to fix the problem you are seing I would have to
enhance imaplib to parse all responses from all commands, and that's
something I'd prefer that the user do.

The `correct' way to deal with responses from an IMAP4 server is to
concatenate all the strings returned, and then parse the response per
the RFC.

The server's responses to multi-part commands is to return a series of
packets in the form "description [literal]" and the imaplib returned
data reflects this (you'll see a list item consisting of a tuple of 2
strings). So if the server sends a packet consisting of just a single
')' (presumably terminating a parenthesis in a previous description,
probably after some "literal" data), you'll see it as a single list
item following the one with the data.

Hope that helps!

Piers.