| 1 |
f{,data}sync() usage could be optimized by batching the calls. |
| 2 |
|
| 3 |
automatically resume upon transient errors, e.g. "connection reset by peer" |
| 4 |
or timeout after some data was already transmitted. |
| 5 |
possibly also try to handle Exchange's "glitches" somehow. |
| 6 |
|
| 7 |
uidvalidity lock timeout handling would be a good idea. |
| 8 |
|
| 9 |
should complain when multiple Channels match the same folders. |
| 10 |
|
| 11 |
should complain about nonsensical combinations like Sync Pull + Create Both. |
| 12 |
|
| 13 |
propagate folder deletions even when the folders are non-empty. |
| 14 |
- verify that "most" of the folders in the Channel are still there. |
| 15 |
- refuse to delete unpropagated messages when trashing on the remote side. |
| 16 |
- refuse to delete far side if it has unpropagated messages. symmetry? |
| 17 |
|
| 18 |
add option to suppress complaints about folders that would need creation |
| 19 |
(but not deleted ones). |
| 20 |
|
| 21 |
add message expiration based on arrival date (message date would be too |
| 22 |
unreliable). MaxAge; probably mutually exclusive to MaxMessages. |
| 23 |
|
| 24 |
add alternative treatments of expired messages. ExpiredMessageMode: Prune |
| 25 |
(delete messages like now), Keep (just don't sync) and Archive (move to |
| 26 |
separate folder - ArchiveSuffix, default .archive). |
| 27 |
|
| 28 |
add support for event notification callbacks. |
| 29 |
|
| 30 |
make it possible to have different mailbox names for far and near side in |
| 31 |
Patterns. |
| 32 |
- use far:near for the pattern |
| 33 |
- supporting names with colons requires and extension of the parser to |
| 34 |
report which parts of an argument were quoted |
| 35 |
- this makes Groups mostly useless, as they are mostly a workaround for this |
| 36 |
function being missing so far |
| 37 |
- this is needed for move detection, which would work only within one Channel |
| 38 |
- this supersedes MapInbox |
| 39 |
|
| 40 |
add regexp-based mailbox path rewriting to the drivers. user would provide |
| 41 |
expressions for both directions. every transformation would be immediately |
| 42 |
verified with the inverse transform. PathDelimiter and Flatten would become |
| 43 |
special cases of this. |
| 44 |
|
| 45 |
add daemon mode. primary goal: keep imap password in memory. |
| 46 |
also: idling mode. |
| 47 |
|
| 48 |
parallel fetching of multiple mailboxes. |
| 49 |
TLS session resumption becomes interesting then as well. |
| 50 |
|
| 51 |
imap_set_msg_flags() & imap_trash_msg(): group commands for efficiency. |
| 52 |
|
| 53 |
add streaming from fetching to storing. |
| 54 |
this is complicated by the IMAP target needing the final size in advance, |
| 55 |
which we can't know in a single pass when newline translation is necessary. |
| 56 |
|
| 57 |
handle custom flags (keywords). |
| 58 |
this is impeded by there being no Maildir-side standard. |
| 59 |
|
| 60 |
make use of IMAP CONDSTORE extension (rfc4551; CHANGEDSINCE FETCH Modifier); |
| 61 |
make use of IMAP QRESYNC extension (rfc5162) to avoid SEARCH to find vanished |
| 62 |
messages. |
| 63 |
|
| 64 |
make use of IMAP CAPABILITY APPENDLIMIT= extension (rfc7889; fastmail & gmail). |
| 65 |
this is really useful only for IMAP <=> IMAP syncs: saves FETCH BODY. |
| 66 |
the message could still become oversized due to conversion. |
| 67 |
|
| 68 |
dummy messages resulting from MaxSize should contain a dump of the original |
| 69 |
message's MIME structure and its (reasonably sized) text parts. |
| 70 |
|
| 71 |
don't SELECT boxes unless really needed; in particular not for appending, |
| 72 |
and in write-only mode not before changes are made. |
| 73 |
problem: UIDVALIDITY change detection is delayed, significantly complicating |
| 74 |
matters. |
| 75 |
|
| 76 |
some error messages are unhelpful in non-verbose mode due to missing context. |
| 77 |
|
| 78 |
possibly use ^[[1m to highlight error messages. |
| 79 |
|
| 80 |
consider alternative approach to trashing: instead of the current trash-before- |
| 81 |
expunge done by mbsync, let MUAs do the trashing (as modern ones typically do). |
| 82 |
mbsync wouldn't do any trashing by itself, but should track the moves for |
| 83 |
optimization. additionally, there should be a mode to move trashed messages to |
| 84 |
the remote store. TrashMode Internal|External, AbsorbRemoteTrash. |
| 85 |
a yet different approach to trashing is treating the trash like a normal mailbox. |
| 86 |
however, this implies a huge working set. |
| 87 |
|
| 88 |
consider optional use of messages-id (and X-GM-MSGID): |
| 89 |
- detection of message moves between folders |
| 90 |
- recovery from loss of sync state, migration from other tools |