In an ideal world, I'd have time to study the details closely
enough that I would already know the answer to the question I'm
about to ask.
However, after (quickly) reading the posts so far, I still wasn't
completely clear on the answer, so... what the heck, I'll just
ask!
Does the current design involve a new per-wc flag that indicates
something about a pristines-(present|absent) mode?
Here's why I ask that question:
I'd always assumed that we wouldn't need such a flag. That is,
since the presence/absence of a given pristine is inherently a
per-file fact, ideally we would record that fact solely on a
per-file basis.
Now, our current *user interface* to pristinelessness may only
offer it on a whole-wc basis, and that's fine. But that doesn't
imply a per-wc flag. The best way to keep the feature flexible --
that is, to keep the door open to other user interfaces in the
future -- is to associate pristine status with each file's entry
and nowhere else.
Maybe that's what we're doing? But I get the feeling, from
skimming the posts, that there *is* some kind of per-wc flag being
stored. If so, maybe there's a good reason for it (i.e., perhaps
my reasoning above is missing some key point). Or maybe there
isn't a good reason for it, in which case we should discuss :-).
Best regards,
-Karl
