[cvsnt] Re: Catch-all solution for changing -k modes?

[Tony Hoyle]

Oliver Giesen wrote:
> OK, I didn't think of that admittedly. In that case I suggest
> differentiating between merely making a k-option sticky temporarily for
> a local working copy and actually changing the k-option as part of a
> versioned operation. IMO the latter should use a different option, e.g.
> -K . This then could have the effects I described before (i.e. modified
> timestamp token in Entries and consequently no more need to specify -f
> on commit(?)).

That'd mean an extra option in entries.extra and even more complex 
semantics (distinguishing between 'sticky' and 'super-sticky')...  Since 
it isn't something that's done very often the -f thing is an easy way to 
avoid all that.  It also works with non-cvsnt clients, which is a bonus.

> Yes, that's what I'd expect it to as well. At least if by "original"
> you mean the k-option of the base revision.
> 
Yes.  I hadn't realized it didn't do that (I understand why not 
though... it's not as easy as it sounds :^).

Tony