Bug #30355: Incorrect ordering of UDF results

There's currently no way of knowing the determinicity of an UDF.
And the optimizer and the sequence() UDFs were making wrong
assumptions about what the is_const member means.
Plus there was no implementation of update_system_tables()
causing the optimizer to overwrite the information returned by
the <udf>_init function.

Fixed by equating the assumptions about the semantics of 
is_const and providing a implementation of update_used_tables().
Added a TODO item for the UDF API change needed to make a better 
implementation.


include/mysql_com.h:
  Bug #30355: comment added
mysql-test/r/udf.result:
  Bug #30355: test case
mysql-test/t/udf.test:
  Bug #30355: test case
sql/item_func.cc:
  Bug #30355: keep const_item_cache and used_tables_cache in sync
sql/item_func.h:
  Bug #30355: 
   - a better implementation of update_used_tables()
   - keep const_item_cache and used_tables_cache in sync
sql/udf_example.c:
  Bug #30355: Wrong value for const_item fixed.

1 files changed, 4 insertions, 6 deletions

diff --git a/sql/udf_example.c b/sql/udf_example.c
index df3a69755ad..4ca6133da03 100644
--- a/sql/udf_example.c
+++ b/sql/udf_example.c
@@ -648,13 +648,11 @@ my_bool sequence_init(UDF_INIT *initid, UDF_ARGS *args, char *message)
     return 1;
   }
   bzero(initid->ptr,sizeof(longlong));
-  /*
-    Fool MySQL to think that this function is a constant
-    This will ensure that MySQL only evalutes the function
-    when the rows are sent to the client and not before any ORDER BY
-    clauses
+  /* 
+    sequence() is a non-deterministic function : it has different value 
+    even if called with the same arguments.
   */
-  initid->const_item=1;
+  initid->const_item=0;
   return 0;
 }