Re: The IMAP APPENDLIMIT Extension - version 04
Alexey Melnikov <[email protected]>
| Newsgroups | gmane.ietf.imapext |
|---|---|
| Message-ID | <[email protected]> |
On 06/04/2015 12:26, Timo Sirainen wrote: > On 04 Mar 2015, at 10:32, Chris Newman <[email protected]> wrote: >> Q for list: If the server has a message larger than UINT32_MAX in the store and >> a client tries to fetch it, should the server just go ahead and violate the >> base-spec's UINT32_MAX limit > Dovecot has already been doing this for the last 10 years or so. If it has been compiled with 64bit off_t support (default) it has allowed >32bit APPENDs and FETCHing will always return the full message. I doubt anybody has actually attempted using that, but if this becomes more commonly used there is already a large number of Dovecots happily serving such content without extra extensions.. I guess this isn't ideal from protocol purity point of view, but it's probably going to work fine in practise whenever such emails are actually becoming used. I still prefer if support for more than UINT32_MAX is advertised somehow. Maybe this can be folded into IMAP update. _______________________________________________ imapext mailing list [email protected] https://www.ietf.org/mailman/listinfo/imapext