Forum Discussion
split multiple column values into multiple rows
A more dynamic solution in Power Query, that allows for additional groups of columns, is:
1. Unpivot all columns except Record ID.
2. Get rid of the sequence numbers in the "Attribute" column (former column names).
3. Add a temporary Index column (from 0) and integer-divide this by 4 (the number of fields in each group),
so you get 0,0,0,0,1,1,1,1,2,2,2,2 etcetera.4. Pivot the "Attribute" column with advanced option "Don't Aggregate".
5. Remove the temporary Index column.
Code below and this video takes you through the steps (the video starts at a "Changed Type" step which I removed after the recording as this would jeopardize the dynamics of the solution).
At the end of the code you may want to add a "Change Type" step.
The steps are all created via menu options in Power Query, but I adjusted the "Extracted First Characters" code to extract all characters up to the first digit (0-9).
let Source = Excel.CurrentWorkbook(){[Name="Table1"]}[Content], #"Unpivoted Other Columns" = Table.UnpivotOtherColumns(Source, {"Record ID"}, "Attribute", "Value"), #"Extracted First Characters" = Table.TransformColumns(#"Unpivoted Other Columns", {{"Attribute", each Text.Start(_, Text.PositionOfAny(_,Text.ToList("0123456789"))), type text}}), #"Added Index" = Table.AddIndexColumn(#"Extracted First Characters", "Index", 0, 1), #"Integer-Divided Column" = Table.TransformColumns(#"Added Index", {{"Index", each Number.IntegerDivide(_, 4), Int64.Type}}), #"Pivoted Column" = Table.Pivot(#"Integer-Divided Column", List.Distinct(#"Integer-Divided Column"[Attribute]), "Attribute", "Value"), #"Removed Columns" = Table.RemoveColumns(#"Pivoted Column",{"Index"}) in #"Removed Columns"
Hi mahender
If you were looking to do this using the Query Editor I would do it in the following way:
- Create a new table called "Table 1" which has the following columns:
- Record ID, Status1, Year1, Month1, Amount1
- I would then rename the columns in "Table 1" to be:
- Record ID, Status, Year, Month, Amount
- Next I would create another table called "Table 2" which would contain the following columns:
- Record ID, Status2, Year2, Month2, Amount2
- I would then rename the columns in "Table 2" to be:
- Record ID, Status, Year, Month, Amount
- Next I would use the Append Queries as New.
- I would then Append "Table 1" and "Table 2"
- NOTE: What happens here is that where the column names are the same, it will put the values in the same columns.
- Finally I would disable the loading of "Table 1" and "Table 2" into your Power BI Model, because they are no longer required, by right clicking on "Table 1" and "Table 2" and de-selecting "Enable Load"
That will then import into your Power BI Model, as you want below.
A more dynamic solution in Power Query, that allows for additional groups of columns, is:
1. Unpivot all columns except Record ID.
2. Get rid of the sequence numbers in the "Attribute" column (former column names).
3. Add a temporary Index column (from 0) and integer-divide this by 4 (the number of fields in each group),
so you get 0,0,0,0,1,1,1,1,2,2,2,2 etcetera.4. Pivot the "Attribute" column with advanced option "Don't Aggregate".
5. Remove the temporary Index column.
Code below and this video takes you through the steps (the video starts at a "Changed Type" step which I removed after the recording as this would jeopardize the dynamics of the solution).
At the end of the code you may want to add a "Change Type" step.
The steps are all created via menu options in Power Query, but I adjusted the "Extracted First Characters" code to extract all characters up to the first digit (0-9).
let Source = Excel.CurrentWorkbook(){[Name="Table1"]}[Content], #"Unpivoted Other Columns" = Table.UnpivotOtherColumns(Source, {"Record ID"}, "Attribute", "Value"), #"Extracted First Characters" = Table.TransformColumns(#"Unpivoted Other Columns", {{"Attribute", each Text.Start(_, Text.PositionOfAny(_,Text.ToList("0123456789"))), type text}}), #"Added Index" = Table.AddIndexColumn(#"Extracted First Characters", "Index", 0, 1), #"Integer-Divided Column" = Table.TransformColumns(#"Added Index", {{"Index", each Number.IntegerDivide(_, 4), Int64.Type}}), #"Pivoted Column" = Table.Pivot(#"Integer-Divided Column", List.Distinct(#"Integer-Divided Column"[Attribute]), "Attribute", "Value"), #"Removed Columns" = Table.RemoveColumns(#"Pivoted Column",{"Index"}) in #"Removed Columns"MarcelBeug I just knew you'll come up with a better way as soon as you see this question! :smileyhappy:
I've seen Step 3 from your video here and was trying to go exactly the same way HOWEVER I was missing Step 2!
Great job as always! :smileyhappy:
Thanks for the solution, it worked like charm.
Dear Marcel! Thank you very much for the solution! The solution works like magic. But recently, I am having trouble when Pivoting the column as specified in last step. I have approx 37000 rows and 22 Columns (2 fields in each group, 11 groups). Power query returns the error as DataFormat.Error: Invalid cell value '#N/A'. I tried everything i could in excel sheet, ranging from removing the #N/A errors to deleting the less relevant columns. Still the error is persistent. Any help would be apreciated!
Thanks.
S
