On Thu, Apr 12, 2012 at 12:06 PM, Julian Foad <julianf...@btopenworld.com> wrote: > Philip Martin wrote: > >> I think alter_directory needs a list of children as well as the new >> properties. > > I agree. I already felt that was needed for API symmetry (or else the > add_directory() should *not* take the list of children). Now your use case > below adds a practical reason.
Fixed in r1334244. >... >> An alternative might be marking A incomplete, like a v1 drive. I don't >> like that, I would prefer it if v2 could reserve incomplete for nodes >> that are just names and have no content. Right. That was always the design plan. > An important point here is that, by knowing the list of children, the > semantics of the WC "incomplete" status can be greatly simplified. We will > only ever need to make new children "incomplete", never the directory > itself. Now the "incomplete" status will only need to mean, "We know there > is a node with this name; we know nothing else about it". It will no longer > ever need to mean "This directory has some children which are completely (or > partially?) fleshed out, and/or perhaps some that we don't know anything > about yet". Sure. That's why the wc_db interface and why the Ev2 add_directory API requires the set of children up front. I just missed doing the same for alter_directory. We're good now! As a concrete example/improvement: during a checkout/update, a new directory is "done" immediately. If the update stops, and is restarted: you don't have to ask about the directory and its set of children. *Just* the subset of children which are still incomplete. Cheers, -g
