Fix for bug #30885: MEMORY returns incorrect data if BTREE index is used for NULL lookup

Problem: creating an rb-tree key we store length (2 bytes) before the actual data for 
varchar key parts. The fact was missed for NULL key parts, when we set NULL byte and 
skip the rest.

Fix: take into account the length of the varchar key parts for NULLs.


heap/hp_hash.c:
  Fix for bug #30885: MEMORY returns incorrect data if BTREE index is used for NULL lookup
    - skip varchar key parts length (2 byte) for null key parts.
mysql-test/r/heap_btree.result:
  Fix for bug #30885: MEMORY returns incorrect data if BTREE index is used for NULL lookup
    -test result.
mysql-test/t/heap_btree.test:
  Fix for bug #30885: MEMORY returns incorrect data if BTREE index is used for NULL lookup
    - test case.

1 files changed, 9 insertions, 0 deletions

diff --git a/mysql-test/t/heap_btree.test b/mysql-test/t/heap_btree.test
index d5a4fb7a734..5a238fc32c8 100644
--- a/mysql-test/t/heap_btree.test
+++ b/mysql-test/t/heap_btree.test
@@ -235,5 +235,14 @@ CREATE TABLE t1 (a INT, UNIQUE USING BTREE(a)) ENGINE=MEMORY;
 INSERT INTO t1 VALUES(NULL),(NULL);
 DROP TABLE t1;
 
+#
+# Bug #30885: MEMORY returns incorrect data if BTREE index is used for NULL lookup
+#
+create table t1(a varchar(255), b varchar(255), 
+                key using btree (a,b)) engine=memory; 
+insert into t1 values (1, 1), (3, 3), (2, 2), (NULL, 1), (NULL, NULL), (0, 0);
+select * from t1 where a is null;
+drop table t1;
+
 --echo End of 5.0 tests