More "goodies" added to sieve require

[xyzzy]

Just spotted in the wild yet more additions to sieve's require; "x-cyrus-log", "mailboxid", and "special-use". Doing a quick glance at some documentation (see section 4) I could find on "mailboxid" and "special-use" they are associated with fileinto. The "x-cyrus-log", not sure, but may be for some kind of logging.

Just when it took them over 6 months to keep sieve tester from complaining about the "fcc" addition (probably fixed when they added "x-cyrus-jmapquery" to keep sieve tester from complaining about that one too so you didn't have edit require for test cases where you need to copy/paste your entire sieve script) now add these 3 new guys and you have to go back to editing it again.

Oh well, after the recent go around about sieve tester's future it probably doesn't matter much anyway.

I'm curious in what (new?) UI context these new guys are going be used for. We've discovered that "x-cyrus-jmapquery" was related to the new (search) rules creation mechanism. So it will be interesting how this new stuff gets used for what.

Never a dull moment which I suppose is a good thing when you're stuck home most of the time.

[SideshowBob]

I think the use of mailboxid is that they will be able to generate sieve filing rules that wont stop working even if you move or rename a folder from imap. Presumably the graphical rule interface will at some point update to display the new name and generate updated sieve.

[rjbs]

Originally Posted by xyzzy

I'm curious in what (new?) UI context these new guys are going be used for. We've discovered that "x-cyrus-jmapquery" was related to the new (search) rules creation mechanism. So it will be interesting how this new stuff gets used for what.)

mailboxid and specialuse provide a tiny, tiny, improvement in the reliability of delivery when there's a rename, avoiding an unlikely rate. Logging isn't used yet, but will make it easier for support agents to trace the execution of rules. None of these is likely to be very exciting for authors of custom Sieve.

[brong]

Specialuse is a standard: https://tools.ietf.org/html/rfc8579

Mailboxid should become a standard at some point when I get off my backside and update the draft - I just let it bitrot because I was busy with JMAP stuff:

https://datatracker.ietf.org/doc/dra...eve-mailboxid/

[xyzzy]

Just noticed that the sieve code for the new rules UI is now using those "mailboxid" and "special-use" extensions.

So mailboxid is used to test for the presence of a folder and fileinto's it as well. Specialuse apparently is only used for \Archive (never really understood the reasoning behind that chunk of code even before adding the specialuse, i.e, how skipinbox can be set and hasmailbox not be set so fileinto \Archive).

I guess what bothers me a little about mailboxid is that up till now sieve code generated by FM didn't do anything a user couldn't do writing their own sieve code. But it appears there's no way I, writing my own sieve code, can use this feature directly since the mailboxid's of all my folders are not known up front. I don't think it matters too much but it is a difference.