Ignore missing column while summing column

[smpa01]

Hi,

My source data consists of following 3 columns C1, C2 and C3. I add these 3 columns to get an extra column called Sum.

The problem is the source data may or may not contain all 3 columns all the time. There are instances when C1 and C2 is present but not C3; C2 and C3 is present but not C1; none of C1, C2 and C3 are present.

Is it possible forPQWRY (not DAX)

1) to add the values of only the columns that are present (from C1, C2 and C3) and ignore the missing column altogether.

and

2) to produce 0 when none of the columns are present

Like following

Source  Data

Project	C1	C2	C3
A	1	1	1
B	2	2	2
C	3	3	3

Any help would be greatly appreciated.

Thank you in advance.

[Stachu]

new column like this will work

(try [C1] otherwise 0) + (try [C2] otherwise 0) + (try [C3] otherwise 0)

not sure about the performance impact though

[smpa01]

Hi @Stachu,

Thank you very much for your reply. You taught me something new today.However, there is glitch.

My basic query is following after incorporating what you suggested.

let
    Source = Excel.CurrentWorkbook(){[Name="Table1"]}[Content],
    #"Changed Type" = Table.TransformColumnTypes(Source,{{"Project", type text}, {"C1", Int64.Type}, {"C2 ", Int64.Type}, {"C3", Int64.Type}}),
    #"Added Custom" = Table.AddColumn(#"Changed Type", "Custom", each (try [C1] otherwise 0) + (try [#"C2 "] otherwise 0)+ (try [C3] otherwise 0))
in
    #"Added Custom"

Now there is #"Changed Type" step which generates an error once it is missing any column. This step can't be avoided. Now, Is there a way "M" can ignore the transormation of column types in #"Changed Type" step as well.

Thank you in advance.

[smpa01 · Accepted solution]

Hey @Stachu,

I found a solution courtesy to the following two links

https://blog.crossjoin.co.uk/2015/02/26/handling-added-or-missing-columns-in-power-query/

https://community.powerbi.com/t5/Desktop/Unpivot-removes-rows-with-no-null-values-how-to-keep-them/m...

My Base headers are following - Table 0 (2)

let
    Source = Web.Page(Web.Contents("https://community.powerbi.com/t5/Desktop/Ignore-missing-column-while-summing-column/m-p/552485")),
    Data0 = Source{0}[Data],
    #"Promoted Headers" = Table.PromoteHeaders(Data0, [PromoteAllScalars=true]),
    #"Changed Type" = Table.TransformColumnTypes(#"Promoted Headers",{{"Project", type text}, {"C1", Int64.Type}, {"C2", Int64.Type}, {"C3", Int64.Type}}),
    #"Removed Bottom Rows" = Table.RemoveLastN(#"Changed Type",3),
    Custom1 = Table.ColumnNames(#"Removed Bottom Rows")
in
    Custom1

I create a another query - Table 0 (3) to find which column is missing from Base headers to the current table

let
    Source = Web.Page(Web.Contents("https://community.powerbi.com/t5/Desktop/Ignore-missing-column-while-summing-column/m-p/552485")),
    Data0 = Source{0}[Data],
    #"Promoted Headers" = Table.PromoteHeaders(Data0, [PromoteAllScalars=true]),
    #"Changed Type" = Table.TransformColumnTypes(#"Promoted Headers",{{"Project", type text}, {"C1", Int64.Type}, {"C2", Int64.Type}, {"C3", Int64.Type}}),
    #"Removed Columns" = Table.RemoveColumns(#"Changed Type",{"C3"}),
    PresentColumns = Table.ColumnNames(#"Removed Columns"),
    Source1 = Web.Page(Web.Contents("https://community.powerbi.com/t5/Desktop/Ignore-missing-column-while-summing-column/m-p/552485")),
    Data01 = Source1{0}[Data],
    #"Promoted Headers1" = Table.PromoteHeaders(Data01, [PromoteAllScalars=true]),
    #"Changed Type1" = Table.TransformColumnTypes(#"Promoted Headers1",{{"Project", type text}, {"C1", Int64.Type}, {"C2", Int64.Type}, {"C3", Int64.Type}}),
    #"Removed Bottom Rows" = Table.RemoveLastN(#"Changed Type1",3),
    ExpectedColumns = Table.ColumnNames(#"Removed Bottom Rows"),
    Custom1 = List.Difference(ExpectedColumns, PresentColumns),
    #"Converted to Table" = Table.FromList(Custom1, Splitter.SplitByNothing(), null, null, ExtraValues.Error),
    #"Promoted Headers2" = Table.PromoteHeaders(#"Converted to Table", [PromoteAllScalars=true]),
    #"Changed Type2" = Table.TransformColumnTypes(#"Promoted Headers2",{{"C3", type any}})
in
    #"Changed Type2"

Finally - Table 0 - to add the missing column to the current table, Replace all null with 0 and then add the columns

let
    Source = Web.Page(Web.Contents("https://community.powerbi.com/t5/Desktop/Ignore-missing-column-while-summing-column/m-p/552485")),
    Data0 = Source{0}[Data],
    #"Promoted Headers" = Table.PromoteHeaders(Data0, [PromoteAllScalars=true]),
    #"Changed Type" = Table.TransformColumnTypes(#"Promoted Headers",{{"Project", type text}, {"C1", Int64.Type}, {"C2", Int64.Type}, {"C3", Int64.Type}}),
    #"Removed Columns" = Table.RemoveColumns(#"Changed Type",{"C3"}),
    #"Appended Query" = Table.Combine({#"Removed Columns", #"Table 0 (3)"}),
    Custom1 = Table.TransformColumns(#"Appended Query",{},(x) => Replacer.ReplaceValue(x,null,0)),
    #"Inserted Sum" = Table.AddColumn(Custom1, "Addition", each List.Sum({[C1], [C2], [C3]}), type number)
in
    #"Inserted Sum"

[v-cherch-msft]

Hi @smpa01

It seems you've solved it,right? If so, please accept your answer as solution, that way, other community members will easily find the solution when they get same issue.

Regards,

Cherie

[smpa01]

@v-cherch-msftyes I have solved it in a way and I did not want to accept the solution in case someone else can provide a better solution to this.

Anyway, I have accepted my own solution now.