------------------------------------------
lines 24-63 of file: xrst/all_node_db.xrst
------------------------------------------

{xrst_begin cov_reference_table}

Covariate Reference Table
#########################
This table specifies the covariate reference value for every possible
:ref:`glossary@job` ; i.e., node_id , split_reference_id pair.

cov_reference_id
****************
is the :ref:`all_node_db@Primary Key` for this table.

node_id
*******
This column has type ``integer`` and specifies a node by its index
in the root node database node table.
The root node and all its descendants will appear in this column.

split_reference_id
******************
This column has type ``integer`` and it specifies the
index of a value in
:ref:`split_reference_table-name`.
If split_reference_table is empty (is not empty),
the values in this column must be (must not be) null.
All the possible *split_reference_id* values will appear for each
*node_id* value.

covariate_id
************
This column has type ``integer`` and it specifies the
covariate by its index in the :ref:`glossary@root_database` covariate table.
All the possible *covariate_id* values will appear for each
( *node_id* , *split_reference* ) pair.

reference_value
***************
This column has type ``real`` and is the reference value for this covariate
for the job corresponding to this (node_id, split_reference_id) .

{xrst_end cov_reference_table}