KB20214: What functionality do the Metric Level (Dimensionality) ‘Advanced’ options provide?

The metric dimensionality advanced options are accessible from the Strategy Metric Editor and are accessible from the dimensionality editor window as shown below.

 
The two options available that affect the Strategy Engine handling of the metric and report calculations are:

1. Allow other users to add extra units to this definition.
2. Filter setting : uncheck to exclude attributes absent in report or level (dimensionality)

 
The setting "Allow other users .." is identified internally as the CanContinue flag, and the filter setting "uncheck to exclude …" is known as the FilterRes flag. By default these two settings are enabled for any new metrics created. Also, the metric definition shown in the metric editor does not show whether these options will or will not be in effect (so the only way to verify these is to open the advanced options dialog box).
The behavior that these settings affect is explained below:

1. CanContinue:
   This is used to indicate whether the environment that uses a metric with this dimensionality is allowed to extend the dimensionality by effectively adding additional units to the dimensionality. This action can happen for metrics used on the template as well as metrics used as a qualification. Current use of this flag only affects metrics used in qualifications however.
   Users can also disable the CanContinue option by specifying a semicolon followed by a forward slash (;/) after the list of dimensionality units in the metric editor. For example
   Sum([Units Sold]) {~+  ;/}
   will create a metric for the sum of the units sold with a report level dimensionality (filtering and grouping: standard) and the CanContinue option unselected.
   
2. FilterRes:
   When this flag is set, if an attribute is found in the filter, but does not appear in this metric dimensionality, that attribute is effectively added to this metric dimensionality.
    
   Users may disable the FilterRes option by specifying a semicolon followed by a minus sign(;-) after the list of dimensionality units in the metric editor. For example
   Sum([Units Sold]) {~+  ;-}
   will create a metric for the sum of the units sold with a report level dimensionality (filtering and grouping: standard) and the FilterRes option unselected.
    
   Consider the example below. A report with the SubCategory and Month attributes on the template contains a metric - Unit Sales. The report filters on three attributes - SubCategory, Region and Month. With the FilterRes flag enabled the SQL generated includes the qualification on the Region attribute as shown below (the filter dimensionality has been effectively added to the metric dimensionality).
    
   select a11.SUBCAT_ID SUBCAT_ID,
       max(a14.SUBCAT_DESC) SUBCAT_DESC,
       a11.MONTH_ID MONTH_ID,
       max(a13.MONTH_DESC) MONTH_DESC,
       sum(a11.TOT_UNIT_SALES) WJXBFS1
   from SUBCATEG_MNTH_CTR_SLS a11
       join LU_CALL_CTR a12
        on (a11.CALL_CTR_ID = a12.CALL_CTR_ID)
       join LU_MONTH a13
        on (a11.MONTH_ID = a13.MONTH_ID)
       join LU_SUBCATEG a14
        on (a11.SUBCAT_ID = a14.SUBCAT_ID)
   where (a11.SUBCAT_ID in (22, 23, 24)
   and a12.REGION_ID in (1, 6)
   and a11.MONTH_ID in (200701, 200702))
   group by a11.SUBCAT_ID,
       a11.MONTH_ID
   If the FilterRes setting is unchecked for the metric, the SQL generated for the report is shown below. As can be observed, the filter condition for Region has been eliminated from the metric calculation. A side effect of this is that a different fact table has been chosen for the metric result calculation.
    
   select a12.SUBCAT_ID SUBCAT_ID,
       max(a14.SUBCAT_DESC) SUBCAT_DESC,
       a11.MONTH_ID MONTH_ID,
       max(a13.MONTH_DESC) MONTH_DESC,
       sum(a11.TOT_UNIT_SALES) WJXBFS1
   from ITEM_MNTH_SLS a11
       join LU_ITEM a12
        on (a11.ITEM_ID = a12.ITEM_ID)
       join LU_MONTH a13
        on (a11.MONTH_ID = a13.MONTH_ID)
       join LU_SUBCATEG a14
        on (a12.SUBCAT_ID = a14.SUBCAT_ID)
   where (a12.SUBCAT_ID in (22, 23, 24)
   and a11.MONTH_ID in (200701, 200702))
   group by a12.SUBCAT_ID,
       a11.MONTH_ID
   

For details on the metric conditionality 'Advanced' settings, users should refer to the article:
 

• KB18272 -- How does the metric qualification advanced option to consider other qualifications interact with metric conditionality in Strategy SQL Generation Engine?