On Mon, Nov 3, 2008 at 19:42, Julian Foad <julianf...@btopenworld.com> wrote: > On Sat, 2008-11-01 at 09:44 -0700, sbut...@tigris.org wrote: >> Author: sbutler >> Date: Sat Nov 1 09:44:45 2008 >> New Revision: 33994 >> >> Log: >> In the client commit code, convert a "not found" error to the more >> helpful "out of date" error. >... >> +++ trunk/subversion/libsvn_client/commit_util.c Sat Nov 1 09:44:45 >> 2008 (r33994) >... >> @@ -1249,6 +1272,18 @@ do_item_commit(void **dir_baton, >> } >> >> SVN_ERR(svn_wc_entry(&tmp_entry, item->path, adm_access, TRUE, pool)); >> + >> + /* When committing a directory that no longer exists in the >> + repository, a "not found" error does not occur immediately >> + upon opening the directory. It appears here during the delta >> + transmisssion. */ >> + err = svn_wc_transmit_prop_deltas >> + (item->path, adm_access, tmp_entry, editor, >> + (kind == svn_node_dir) ? *dir_baton : file_baton, NULL, pool); >> + >> + if (err) >> + return fixup_out_of_date_error(path, kind, err); >> + >> SVN_ERR(svn_wc_transmit_prop_deltas >> (item->path, adm_access, tmp_entry, editor, >> (kind == svn_node_dir) ? *dir_baton : file_baton, NULL, >> pool)); > > I'll query this: was that addition meant to replace, rather than > duplicate, the "svn_wc_transmit_prop_deltas(...)" call which now appears > twice?
This question was never answered, and I'm wondering the same thing. Stephen: intentional, or not? Cheers, -g
