Application of SQL Server with ROLLUP, with CUBE, grouping statement

The result set generated by Cube:cube shows aggregations for all combinations of values in the selected column.

The result set generated by Rollup:rollup shows the aggregation of a hierarchy of values in the selected column.

GROUPING: When a row is added by the cube or ROLLUP operator, the function causes the output value of the additional column to be 1, and when the row is not added by the cube or ROLLUP operator, the function causes the output value of the additional column to be 0.

Create a temporary table first:

1 Create Table#Temp2 (3Namevarchar( -) not NULL,4Coursevarchar( -)NULL,5Scoresint        NULL6 )7 8 Insert  into#Temp9 Select 'Little Red','SQL',' -' Union Ten Select 'Little Red','C #',' -' Union  One Select 'Xiao Ming','SQL',' the' Union  A Select 'Xiao Ming','C #',' the' Union  - Select 'Xiao Li','SQL',' the' Union  - Select 'Xiao Li','C #',NULL the  - Select *  from#Temp

With CUBE:

1 Select name, course,sum2 from #temp3  Group  by name, course 4 with Cube

Group by name and class, then group by name and class.

PS: The classification is based not on the order in select, but on the order in group by.

Let's look at the results in a different order:

1 Select name, course,sum2 from #temp3Groupby  Course, name 4 with Cube

Groups of classes and names are grouped together and grouped by class and name.

The result set generated by cube shows aggregations for all combinations of values in the selected column.

With ROLLUP:

1 Select name, course,sum2 from #temp3Groupby  name, course 4 with rollup

1 Select name, course,sum2 from #temp3Groupby  Course, name 4 with rollup

The result set generated by ROLLUP shows the aggregation of a hierarchy of values in the selected column.

So what is this one level structure? Look at the data above, when grouped by name, divided into three groups (excluding the last line total), when the course first grouped, divided into two groups (excluding the last line total).

This particular hierarchy I suspect should be related to group by's grouping order.

GROUPING:

The combination of grouping with rollup (same as with Cube)

1 Select name, course,sum(score),GROUPING2 from #temp3  Group  by name, course 4 with rollup

When grouping is specified as "name", only the last line is added with rollup.

1 Select name, course,sum(score),GROUPING2 from #temp3  Group  by name, course 4 with rollup

When grouping is specified as a "course", the third, sixth, Nineth, and last lines are added with rollup.

When a row is added by the cube or ROLLUP operator, the function causes the output value of the additional column to be 1, and when the row is not added by the cube or ROLLUP operator, the function causes the output value of the additional column to be 0.

1 Selectname,2  Case  when GROUPINGName=1 3       Then 'Total' 4      Else 5          Case  when GROUPINGCourse=1 6               Then ' Subtotal' 7              ElseCourseEnd  8      Endcourses,9 sum(Score)Ten  from#Temp  One Group  byName, course A  withRollup

Application of SQL Server with ROLLUP, with CUBE, grouping statement