On Feb 29, 2012 9:08 AM, "Stefan Sperling" <[email protected]> wrote: > > On Wed, Feb 29, 2012 at 08:25:45AM -0500, Greg Stein wrote: > > On Feb 29, 2012 8:02 AM, <[email protected]> wrote: > > > > > > Author: neels > > > Revision: 1292402 > > > Modified property: svn:log > > > > > > Modified: svn:log at Wed Feb 29 13:02:36 2012 > > > > > ------------------------------------------------------------------------------ > > > --- svn:log (original) > > > +++ svn:log Wed Feb 29 13:02:36 2012 > > > @@ -1,2 +1,8 @@ > > > On the 1.7.x-issue4102 branch, merge the following from trunk, resolving > > > conflicts: r1233292, r1245711, r1245746, r1245764, r1245809, r1245929, > > r1245935 > > > + > > > +Modification of r1245711: > > > +svn_kind_t does not exist in 1.7.x. So instead of using svn_kind_t in > > > +svn_wc__db_t.parse_cache, add an IS_SYMLINK flag next to the > > svn_node_kind_t. > > > +Also affects get_path_kind() and svn_wc__db_wcroot_parse_local_abspath() > > > +in wc_db_wcroot.c. > > > > What?! We have svn_wc__db_kind_t in 1.7.x to represent symlinks. > > > > -g > > This backport would have been more intrusive if we also switched > the existing 1.7 code over from using svn_node_kind_t to svn_wc__db_kind_t. > > If you really prefer that we can do it. But I'd argue it's extra work > for no gain. In trunk we already made the code use svn_kind_t. > For 1.7 we should primarily fix broken behaviour rather than investing > time in extra touches to make the code more beautiful.
I guess I should look at that branch, but Neels mentioned wc_db functions. So those should already be using db_kind, not node_kind. Cheer, -g
