> "svn switch" talks about switching the repository location for nodes. The > added (or copied/moved) nodes do not (yet) have a repository location, so > they cannot be switched. They just follow the parent.
It's not like that is set in stone. An added node could have a *projected* repos location, derived from the parent at the time of the add. The question is which makes more sense in terms of the API. If users issue a depth that, by the literal meaning of the word *empty*, suggests that it excludes all children, added or not, they may be very confused to see adds of several levels below following the switch. It would make most sense to bail instead: "Why are you trying to exclude added children that are semantically attached to the path you are switching? Don't do that, it's a misconception." (curious what Julian has to say about why both should be supported) ~Neels
