Re: I-D Action: draft-ietf-sieve-convert-02.txt
Barry Leiba <[email protected]>
| Newsgroups | gmane.ietf.mta-filters |
|---|---|
| Message-ID | <CALaySJLPYYXVX7DrzB4JrAsN6jtPRPHMHcAGHu_hMNn=EcTKfg@mail.gmail.com> |
>>> "I've asked you and you ignored me" behavior doesn't seem right in Sieve. >>> I don't think any other extension does that. >> >> I don't think this is a good reason to have it work this way, unless >> we have a use case that merits the extra complexity. >> > Can you clarify what kind of complexity you are talking about? The complexity of defining a new test to go with the new action. On the other hand, if we have an action that can also be used as a test, I think that's simpler. And it seems we're converging on that. >>>> Finally, I had an idea for how we could do the testing, which might >>>> satisfy both issues: >>>> Why not have the "convert" action *also* be a test. ... >> Do you mean that making this both an action and a test shouldn't imply >> that any other actions can be used as tests also? >> > Yes. Or rather the other way around: not every test is expected to be used > as an action. I was thinking about calling functions in C: they can be used > in tests and they can also be called so that their return value is ignored. OK. I think we're good on this, then. We'll define this particular action to also function as a test, but I'll put in text that makes it clear that it's just this case, and that other actions are not expected to behave this way. If I don't hear objections from anyone else, I'll go ahead and put out a new version with this change. Barry _______________________________________________ sieve mailing list [email protected] https://www.ietf.org/mailman/listinfo/sieve