Skip to main content
cancel
Showing results for 
Search instead for 
Did you mean: 

Earn the coveted Fabric Analytics Engineer certification. 100% off your exam for a limited time only!

Reply
MansRydell
Helper II
Helper II

multiply columns by one column, problem with data type changing

Hi All

 

I am transforming a bunch of columns by multiplying them on row level by values in a column:

Ex in tabel below I multiply Column2, Column3 & Column4 by Column1 (all columns 1-4 are data type: decimal numbers)

Column1Column2Column3Column4
2121
3113

Using:

Table.FromRecords(Table.TransformRows(#"Changed Type", (x) => Record.TransformFields(x, List.Transform({"Column2", "Column3", "Column4"}, (y) => {y, each x[Column1] * _}))))

 

And get:

Column1Column2Column3Column4
2242
3339

 

Wierd thing is that pre transformation Column 2, 3 & 4 are Decimal Numbers and post transformation they are datat type: any so I transforma all the columns to decimal number in next step.
And I´m finding some smal errors that feels like rounding errors on some rows.


Anyone out there with an Idea what I´m doing wrong?

1 ACCEPTED SOLUTION
v-stephen-msft
Community Support
Community Support

Hi @MansRydell ,

 

You could use the Table.TransformColumnTypes function to change all the data types to decimal number.

= let
output=Table.FromRecords(Table.TransformRows(#"Changed Type", (x) => Record.TransformFields(x, List.Transform({"Column2", "Column3", "Column4"}, (y) => {y, each x[Column1] * _}))))
in 
Table.TransformColumnTypes(output,List.Transform(Table.ColumnNames(output), each {_, type number}))

vstephenmsft_0-1684302289407.png

Table.TransformColumnTypes - PowerQuery M | Microsoft Learn

List.Transform - PowerQuery M | Microsoft Learn

 

Best Regards,

Stephen Tao

 

If this post helps, then please consider Accept it as the solution to help the other members find it more quickly.           

View solution in original post

4 REPLIES 4
slorin
Super User
Super User

Hi, 

Try this solution with Table.ReplaceValue

 

let
Source = Table.FromRows(Json.Document(Binary.Decompress(Binary.FromText("HYyxDQAgDMN+6ZyhNEDKLYj/36DqZCm2cq8JyTQYsTmKC6FZ3NhOe7hlOFYbnsKEnygOLKqDgKsGQVI/eXYXSnvvAw==", BinaryEncoding.Base64), Compression.Deflate)), let _t = ((type nullable text) meta [Serialized.Text = true]) in type table [Column1 = _t, Column2 = _t, Column3 = _t, Column4 = _t]),
Number_Type = Table.TransformColumnTypes( Source, {{"Column1", type number}, {"Column2", type number}, {"Column3", type number}, {"Column4", type number}}),
Replace_Value = Table.ReplaceValue( Number_Type, each [Column1], "", (x,y,z) => x*y, {"Column2", "Column3", "Column4"}),
Restore_Type = Value.ReplaceType( Replace_Value, Value.Type( Number_Type ))
in
Restore_Type

 Stéphane

MansRydell
Helper II
Helper II

Thanks everyone for you´re input. Unfortunately my issue is unresolved thanks for you´re effort if I find a solution I´ll post it. 🙏

v-stephen-msft
Community Support
Community Support

Hi @MansRydell ,

 

You could use the Table.TransformColumnTypes function to change all the data types to decimal number.

= let
output=Table.FromRecords(Table.TransformRows(#"Changed Type", (x) => Record.TransformFields(x, List.Transform({"Column2", "Column3", "Column4"}, (y) => {y, each x[Column1] * _}))))
in 
Table.TransformColumnTypes(output,List.Transform(Table.ColumnNames(output), each {_, type number}))

vstephenmsft_0-1684302289407.png

Table.TransformColumnTypes - PowerQuery M | Microsoft Learn

List.Transform - PowerQuery M | Microsoft Learn

 

Best Regards,

Stephen Tao

 

If this post helps, then please consider Accept it as the solution to help the other members find it more quickly.           

m_dekorte
Super User
Super User

Hi @MansRydell,

 

Alternatively, you could give something like this a go.

Copy this code into a new blank query

let
    Source = Table.FromRows(Json.Document(Binary.Decompress(Binary.FromText("i45WMlLSUTIEYggdqxOtZAwVAWFjpdhYAA==", BinaryEncoding.Base64), Compression.Deflate)), let _t = ((type nullable text) meta [Serialized.Text = true]) in type table [Column1 = _t, Column2 = _t, Column3 = _t, Column4 = _t]),
    ChType = Table.TransformColumnTypes(Source,{{"Column1", Int64.Type}, {"Column2", Int64.Type}, {"Column3", Int64.Type}, {"Column4", Int64.Type}}),
    ReplaceValue = List.Accumulate( 
        List.Skip( Table.ColumnNames(Source), 1),
        ChType, 
        (s, a) => Table.ReplaceValue( s, 
            each Record.Field(_, a), 
            each Record.Field(_, a) * Record.Field(_, "Column1"),
            Replacer.ReplaceValue, {a}
        )
    ),
    RestoreType = Value.ReplaceType( ReplaceValue, Value.Type( ChType ))
in
    RestoreType

 

There are a number of transformations that can't retain type, above type is restored as it was before.

Here's a link to a good post that deals with rounding in M 

 

Ps. Please mark this answer as solution when it helped you to resolve your question, thanks!

Helpful resources

Announcements
April AMA free

Microsoft Fabric AMA Livestream

Join us Tuesday, April 09, 9:00 – 10:00 AM PST for a live, expert-led Q&A session on all things Microsoft Fabric!

March Fabric Community Update

Fabric Community Update - March 2024

Find out what's new and trending in the Fabric Community.

Top Solution Authors
Top Kudoed Authors