SQL/MX 2.x Query Guide (G06.24+, H06.03+)

Operators and Operator Groups
HP NonStop SQL/MX Query Guide—523728-003
7-43
SHORTCUT_SCALAR_AGGR Operator
For more information about data mining, see the SQL/MX Data Mining Guide.
The example of the SEQUENCE operator is based on:
SELECT RUNNINGCOUNT(*) FROM mining_data
SEQUENCE BY age, gender DESC;
required_order: indexcol(CAT.SCH.MINING_DATA.AGE),
inverse(indexcol CAT.SCH.MINING_DATA.GENDER))
sequence_functions: (replace null(offset(\:_sys_Result),
cast(offset(\:_sys_Result)), cast(0 )) + replace null(1 ,
cast(1 ), cast(0 )))
num_history_rows: 1024
history_row_size: 16
SHORTCUT_SCALAR_AGGR Operator
Groupby Group
The SHORTCUT_SCALAR_AGGR operator occurs for aggregates without a GROUP
BY clause and returns one row. The description field for this operator contains:
The example of the SHORTCUT_SCALAR_AGGR operator is based on:
SELECT MIN( NO_O_ID )
FROM norders
WHERE (NO_W_ID, NO_D_ID) = (?W_ID, ?D_ID)
FOR REPEATABLE ACCESS IN EXCLUSIVE MODE;
3 . 4 shortcut_scalar_aggr
aggregates:
min(indexcol(\TESTSYS.$BIG18B.OEQA8.NORDERS.NO_O_ID))
sequence_functions Represents the list of sequence functions that
must be evaluated by this SEQUENCE
operator.
ItemExpr
tree
num_history_rows Size of the history buffer (in rows). This
number of rows is kept in a buffer and is
available for access by the sequence
functions. Any access to a row outside this
buffer results in a NULL value. The default
value for this parameter is the value of the
initial setting for the default
DEF_MAX_HISTORY_ROWS, which is initially
set to 1024.
integer
history_row_size Size of each history row in the history buffer. integer
Token Followed by... Data Type
aggregates Expression of the
aggregate functions
expr(text)