"Bert Huijben" <b...@qqmail.nl> writes: >> case svn_wc_schedule_replace: >> + if (parent_node->base && parent_node->work) >> + return svn_error_createf(SVN_ERR_UNSUPPORTED_FEATURE, NULL, >> + _("Cannot upgrade with copied/replaced directory '%s'"), >> + svn_dirent_local_style(parent_node->base->local_relpath, >> + scratch_pool)); >> + > > Isn't this a limitation from before we introduced multiple layers in NODES? > > In the BASE+WORKING world where we lived until half a year ago, you > couldn't replace WORKING with something else. But now we can just > introduce a new op_depth layer on top.
Yes, we can make it work but I'm not sure how much, if at all, the code has to change. Until it works it is more user-friendly to return an error rather than assert. -- Philip