<DKIM> Re: <DKIM> UID for syncing messages

[J. Tull]

On +0200, Nicolas Sebrecht wrote:
> 
> There are use cases where Message-ID dups are expected, actually. E.g.
> emails coming back from mailing lists.  This is not a valid reference
> for syncing while that's what UID is for: uniquely identify emails
> between the client and the server. Both are sightly different because
> they don't serve the same purpose.
> 
> Also, the IMAP protocol and servers make it way slower to work with
> email content (including headers) than with UIDs.
> 
> Thinking about this issue again, I'm not sure this feature would worth
> the trouble. It's up to you, though.
> 
> -- 
> Nicolas Sebrecht

i experienced one or two cases where different emails had the same message-id
in about 10000 emails. But still this problem doesn't explain why it's out of
options. Mutt also optionaly adds the subject field to the message-ID field
to find duplicate emails. If it slows down the sync process then it should be
disabled by default.

I have a local repository containing GB's of emails but there are missing
emails and i would like a way to get the missing ones from the server without
having to redownload everything. It should be noted that the server either is
missing some, so mirroring it would not suit my need.