How would I convert rows into columns? In this case, taking the columns of Reason
and Weight
and expanding the values where more columns are created?
The column Reason
will only have four values (whereas the field may be blank too): Natural, Emissions, and Humans.
Input Table Example:
Location Year Climate Reason Weight
USA 2010 Warm Natural 50.00
USA 2010 Hot Emissions 0.00
USA 2009 Warm 20.00
USA 2010 Warm Humans 30.00
Brazil 2010 Warm Emissions 100.00
Output Table Example
Location Year Climate Natural_Weight Emissions_Weight Humans_Weight NULL_Weight
USA 2010 Warm 50.00 0.00 30.00
USA 2010 Hot
USA 2009 Warm 20.00
Brazil 2010 Warm 100.00
I was thinking of doing a pivot, but got a bit confused since I'm trying to deal with two columns that are dependent on others.
SELECT Location, Year, Climate, Natural_Weight, Emissions_Weight, Humans_Weight, NULL_Weight
FROM (SELECT Location, Year, Climate, Reason, Weight FROM Test) src
PIVOT ( Max (Year)
FOR Climate IN (Location, Year, Climate)) pvt;
I tried to do a pivot, but I my above query in generating some horrible errors - at the same time, not sure if I'm missing a group by to ensure things are consolidating in the right way. Any guidance would be appreciated!
CodePudding user response:
Perhaps this will help
Select *
From (
Select Location
,Year
,Climate
,Item = concat(coalesce(Reason,'NULL'),'_Weight')
,Value = Weight
From YourTable
) src
Pivot ( sum(Value) for Item in ( [Natural_Weight],[Emissions_Weight],[Humans_Weight],[NULL_Weight] ) ) pvt
Results