Prevent parser from believing that views have system columns.

Views should not have any pg_attribute entries for system columns.
However, we forgot to remove such entries when converting a table to a
view.  This could lead to crashes later on, if someone attempted to
reference such a column, as reported by Kohei KaiGai.

This problem is corrected properly in HEAD (by removing the pg_attribute
entries during conversion), but in the back branches we need to defend
against existing mis-converted views.  This fix costs us an extra syscache
lookup per system column reference, which is annoying but probably not
really measurable in the big scheme of things.

1 files changed, 9 insertions, 2 deletions

diff --git a/src/backend/parser/parse_relation.c b/src/backend/parser/parse_relation.c
index 30b307b191c..3f2177b9aff 100644
--- a/src/backend/parser/parse_relation.c
+++ b/src/backend/parser/parse_relation.c
@@ -508,10 +508,17 @@ scanRTEForColumn(ParseState *pstate, RangeTblEntry *rte, char *colname,
 		attnum = specialAttNum(colname);
 		if (attnum != InvalidAttrNumber)
 		{
-			/* now check to see if column actually is defined */
+			/*
+			 * Now check to see if column actually is defined.  Because of
+			 * an ancient oversight in DefineQueryRewrite, it's possible that
+			 * pg_attribute contains entries for system columns for a view,
+			 * even though views should not have such --- so we also check
+			 * the relkind.  This kluge will not be needed in 9.3 and later.
+			 */
 			if (SearchSysCacheExists2(ATTNUM,
 									  ObjectIdGetDatum(rte->relid),
-									  Int16GetDatum(attnum)))
+									  Int16GetDatum(attnum)) &&
+				get_rel_relkind(rte->relid) != RELKIND_VIEW)
 			{
 				var = make_var(pstate, rte, attnum, location);
 				/* Require read access to the column */