Bug#11766249 bug#59316: PARTITIONING AND INDEX_MERGE MEMORY LEAK

Update for previous patch according to reviewers comments.

Updated the constructors for ha_partitions to use the common
init_handler_variables functions

Added use of defines for size and offset to get better readability for the code that reads
and writes the .par file. Also refactored the get_from_handler_file function.

1 files changed, 15 insertions, 8 deletions

diff --git a/sql/handler.cc b/sql/handler.cc
index 8adb8e061a3..718529fa5fc 100644
--- a/sql/handler.cc
+++ b/sql/handler.cc
@@ -2045,14 +2045,21 @@ handler *handler::clone(const char *name, MEM_ROOT *mem_root)
     on this->table->mem_root and we will not be able to reclaim that memory 
     when the clone handler object is destroyed.
   */
-  if (!(new_handler->ref= (uchar*) alloc_root(mem_root, ALIGN_SIZE(ref_length)*2)))
-    return NULL;
-  if (new_handler && !new_handler->ha_open(table,
-                                           name,
-                                           table->db_stat,
-                                           HA_OPEN_IGNORE_IF_LOCKED))
-    return new_handler;
-  return NULL;
+  if (new_handler &&
+     !(new_handler->ref= (uchar*) alloc_root(mem_root,
+                                             ALIGN_SIZE(ref_length)*2)))
+    new_handler= NULL;
+  /*
+    TODO: Implement a more efficient way to have more than one index open for
+    the same table instance. The ha_open call is not cachable for clone.
+  */
+  if (new_handler && new_handler->ha_open(table,
+                                          name,
+                                          table->db_stat,
+                                          HA_OPEN_IGNORE_IF_LOCKED))
+    new_handler= NULL;
+
+  return new_handler;
 }