Re: [Dbmail-dev] Non-contiguous UIDs vs Gnus and nnimap

2007-05-19 Thread James Cloos
> "Aaron" == Aaron Stone <[EMAIL PROTECTED]> writes: Aaron> Gnus is open source. Clue them in that we're within RFC and they Aaron> should handle our behavior. Yes, I've posted there. And have started trying to understand imap.el and nnimap.el enough to try fixing it on that end. I like to

Re: [Dbmail-dev] Non-contiguous UIDs vs Gnus and nnimap

2007-05-18 Thread Aaron Stone
Gnus is open source. Clue them in that we're within RFC and they should handle our behavior. If we're not within RFC, then we'll try to fix it. Heads up that bandwidth is very limited right now because there are a small number of red hot items holding up the 2.2.5 release. Thank you for the detai

Re: [Dbmail-dev] Non-contiguous UIDs vs Gnus and nnimap

2007-05-18 Thread James Cloos
Following up, it looks like dbmail doesn't use the sequences in the schemas as defaults, but rather manually grabs the nextval (thereby updating the sequence) and uses that explicitly in the INSERT. As such, this should be an easier change, since the trigger is not strictly required. Whether it i

[Dbmail-dev] Non-contiguous UIDs vs Gnus and nnimap

2007-05-18 Thread James Cloos
Yesterday I took the plunge from nnmh to dbmail storage of my mail. Rather than waiting until I managed to hack in variables support to libsieve, I am piping to dbmail-smtp from my procmailrc, specifying the folder with -M. It is mostly going OK, but there are a couple of issues. Gnus is not too