1 Problems are listed in approximate order of priority.
3 - character set support: We should have a global cache of
4 character to file name mappings.
6 - When displaying MIME headers, rfc 2047 decoding is applied (which
7 should not happen), and rfc 2231 decoding is not applied (which
10 - Help formatting could be revamped a bit.
12 - re-add support for .mh_sequences files
14 - In the "attachment" menu, assume this:
16 1 [text/plain, 7bit, 1.1K] <no description>
17 2 [message/rfc822, 7bit, 6.1K] A test message
18 3 [text/plain, 7bit, 0.1K] |-><no description>
19 4 [message/rfc822, base64, 2.5K] |-><no description>
20 5 [message/rfc822, base64, 2.7K] `-><no description>
22 (please note the "message/rfc822" attachments encoded as
23 Base64; that's illegal, but Sun's Mailtool sends that
24 kind of messages); then go to, say, attachment "4",
25 delete it, and go to the main menu; you won't be able to
26 quit the mailbox (ok, 'x' works, but 'q' doesn't).
28 The problem here lies in the fact that mutt uses mailbox
29 handling functions to access message/rfc822 type
30 attachments. We'd need to perform an additional
31 decoding step before using these functions to fix this
34 Please note that mutt's just assuming RFC-compliant mail
35 here. Fixing this stuff may become a PITA.
40 - BODY struct should probably have a pointer to its
41 corresponding HEADER struct. this is needed for
42 mh/maildir mailboxes so the correct pathname can be
43 found. Or perhaps all we need is a .hdr member of the
44 STATE struct so that all of the MIME handlers can look
45 up the corresponding HEADERs if need be?
47 - handle message/external-body in some fashion
49 - handle message/partial reconstruction
51 - make patterns generic (I have patches for this -tlr), and
52 introduce generic menu limiting, menu pattern searching, and the
55 Note: This still requires some thought, since we'd have to store
56 per-entry data in the menu structure. As an alternative, we could
57 extend the tag method to do something to more general flags. The
58 latter approach would make the implementation of propper
59 tag-prefix behaviour more simple: Functions should only be applied
60 when a message is tagged and visible. Additionally, we must not
61 access a menu's max field directly any more: Adding an entry to a
62 menu will require re-allocating and possibly updating the v2r
63 array. How do we handle "in-the-middle additions" properly? Do