MDEV-21958: Query having many NOT-IN clauses running forever

Basic variant of the fix: do not consider conditions in form

  unique_key NOT IN (c1,c2...)

to be sargable. If there are only a few constants, the condition
is not selective. If there are a lot constants, the overhead of
processing such a huge range list is not worth it.

(Backport to 10.2)

mysql-test/r/range.result

@@ -3158,6 +3158,21 @@ pk	a	b
 1	5	50
 65	5	50
 drop table t1;
+create table t1(a int);
+insert into t1 values (0),(1),(2),(3),(4),(5),(6),(7),(8),(9);
+create table t2 (
+pk int primary key,
+key1 int,
+col1 varchar(255),
+key (key1, pk)
+);
+insert into t2 (pk, key1)
+select A.a+10 *B.a + 100*C.a, A.a+10 *B.a +100*C.a from t1 A, t1 B, t1 C;
+# This must use ALL, not range:
+explain select * from t2 force index (primary) where pk not in (1,2,3);
+id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
+1	SIMPLE	t2	ALL	PRIMARY	NULL	NULL	NULL	1000	Using where
+drop table t1,t2;
 #
 # End of 10.2 tests
 #

mysql-test/t/range.test

@@ -2183,6 +2183,27 @@ eval $q4;
 
 drop table t1;
 
+#
+# MDEV-21958: Query having many NOT-IN clauses running forever (testcase 2)
+#
+create table t1(a int);
+insert into t1 values (0),(1),(2),(3),(4),(5),(6),(7),(8),(9);
+
+create table t2 (
+  pk int primary key,
+  key1 int,
+  col1 varchar(255),
+  key (key1, pk)
+);
+
+insert into t2 (pk, key1)
+select A.a+10 *B.a + 100*C.a, A.a+10 *B.a +100*C.a from t1 A, t1 B, t1 C;
+
+--echo # This must use ALL, not range:
+explain select * from t2 force index (primary) where pk not in (1,2,3);
+
+drop table t1,t2;
+
 --echo #
 --echo # End of 10.2 tests
 --echo #

sql/opt_range.cc

@@ -7158,6 +7158,30 @@ SEL_TREE *Item_func_in::get_func_mm_tree(RANGE_OPT_PARAM *param,
       if (array->count > NOT_IN_IGNORE_THRESHOLD || !value_item)
         DBUG_RETURN(0);
 
+      /*
+        If this is "unique_key NOT IN (...)", do not consider it sargable (for
+        any index, not just the unique one). The logic is as follows:
+         - if there are only a few constants, this condition is not selective
+           (unless the table is also very small in which case we won't gain
+           anything)
+         - If there are a lot of constants, the overhead of building and
+           processing enormous range list is not worth it.
+      */
+      if (param->using_real_indexes)
+      {
+        key_map::Iterator it(field->key_start);
+        uint key_no;
+        while ((key_no= it.next_bit()) != key_map::Iterator::BITMAP_END)
+        {
+          KEY *key_info= &param->table->key_info[key_no];
+          if (key_info->user_defined_key_parts == 1 &&
+              (key_info->flags & HA_NOSAME))
+          {
+            DBUG_RETURN(0);
+          }
+        }
+      }
+
       /* Get a SEL_TREE for "(-inf|NULL) < X < c_0" interval.  */
       uint i=0;
       do