On Tue, 2008-11-04 at 14:24 +0000, Julian Foad wrote:
> Does this make sense, anyone?
I hope so. It's r34044.
- Julian
> [[[
> Index: subversion/libsvn_wc/log.c
> ===================================================================
> --- subversion/libsvn_wc/log.c (revision 34038)
> +++ subversion/libsvn_wc/log.c (working copy)
> @@ -1325,7 +1325,8 @@
> finfo.size = 0;
>
> /* Files have been moved, and timestamps have been found. It is now
> - time for The Big Entry Modification. */
> + time for The Big Entry Modification. Here we set fields in the entry
> + to values which we know must be so because it has just been committed. */
> entry->revision = SVN_STR_TO_REV(rev);
> entry->kind = is_this_dir ? svn_node_dir : svn_node_file;
> entry->schedule = svn_wc_schedule_normal;
> @@ -1340,7 +1341,9 @@
> entry->copyfrom_rev = SVN_INVALID_REVNUM;
> entry->has_prop_mods = FALSE;
> entry->working_size = finfo.size;
> - entry->tree_conflict_data = NULL;
> + /* We don't reset tree_conflict_data, because it's about conflicts on
> + children, not on this node, and it could conceivably be valid to commit
> + this node non-recursively while children are still in conflict. */
> if ((err = svn_wc__entry_modify(loggy->adm_access, name, entry,
> (SVN_WC__ENTRY_MODIFY_REVISION
> | SVN_WC__ENTRY_MODIFY_SCHEDULE
> @@ -1357,7 +1360,6 @@
> : 0)
> | SVN_WC__ENTRY_MODIFY_HAS_PROP_MODS
> | SVN_WC__ENTRY_MODIFY_WORKING_SIZE
> - | SVN_WC__ENTRY_MODIFY_TREE_CONFLICT_DATA
> | SVN_WC__ENTRY_MODIFY_FORCE),
> FALSE, pool)))
> return svn_error_createf
> ]]]
>
> - Julian
>
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: dev-unsubscribe_at_subversion.tigris.org
> For additional commands, e-mail: dev-help_at_subversion.tigris.org
>
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe_at_subversion.tigris.org
For additional commands, e-mail: dev-help_at_subversion.tigris.org
Received on 2008-11-04 17:13:02 CET