Ilja Booij <[EMAIL PROTECTED]> said: > On Tue, 23 Nov 2004 03:16:17 -0000, Aaron Stone wrote: >> >> So I started working on libSieve again, and on the DBMail Sieve support. >> Should I just leave what's in 2.0 as it is, or begin maintaining it so >> that anybody interested in getting DBMail 2.0 to work with Sieve can hack >> it in easily? >> >> I have a big patch to the Sieve stuff in 2.0 ready that doesn't touch >> anything significant in the non-Sieve parts of DBMail (just a new struct >> in dbmailtypes.h). I'm thinking wait until right after 2.0.1. Any >> objections? > > I object. > It should go into 2.1, not 2.0. It would probably be nice to have it > in 2.0, but we shouldn't be making the same mistakes as in the process > towards 2.0 by putting new features into a stable release.
Ok, fair enough. Thinking some more about it, since I want to keep the libSieve API unstable Until I Get It Right (TM), it's going to be a volatile area of code. No business in a stable release. I'll probably keep a patchset against 2.0 so that I have a basis for testing libSieve, but otherwise will focus development on 2.1. Aaron --
