Row with an empty cell, find another row with the same name but with that cell and copy the value

[Theiamania]

I am using rows with two types of information: devices and the serial numbers of those devices, both have same name in first column. Here is an example:

Device name	Device QR	Serial QR
Device 1	00011
Device 1		S0012
Device 1		S0013
Device 2	00022
Device 2		S0023

I need to fill in the empty cells in the Device QR column. In each row I have a name. Now, for each empty cell in that column, I need to find the name of its row, use that name to find another row with the same name but with a value in the cell in the Device QR column and paste that value into the empty cell in the first row.

Device name	Device QR	Serial QR
Device 1	00011
Device 1	00011	S0012
Device 1	00011	S0013
Device 2	00022
Device 2	00022	S0023

How can I do it?

[ronrsnfld]

In the Power Query editor, assuming the blank cells in DeviceQR are null, just select that column and "Fill Down".  If the blanks are null strings, then replace them with null first.

[tackytechtom]

Hi @Theiamania ,

How about this:

Before:

After

Here the code in M:

let
    Source = Table.FromRows(Json.Document(Binary.Decompress(Binary.FromText("i45Wckkty0xOVTBU0lEyMDAwBNEKSrE6KBIKQBwMlDTCJWGMLGEEMcrICN0oI4QOI6COWAA=", BinaryEncoding.Base64), Compression.Deflate)), let _t = ((type nullable text) meta [Serialized.Text = true]) in type table [#"Device name" = _t, #"Device QR" = _t, #"Serial QR" = _t]),
    #"Changed Type" = Table.TransformColumnTypes(Source,{{"Device name", type text}, {"Device QR", Int64.Type}, {"Serial QR", type text}}),
    #"Filtered Rows" = Table.SelectRows(#"Changed Type", each ([Device QR] <> null)),
    #"Merged Queries" = Table.NestedJoin(#"Changed Type", {"Device name"}, #"Filtered Rows", {"Device name"}, "Filtered Rows", JoinKind.LeftOuter),
    #"Removed Columns" = Table.RemoveColumns(#"Merged Queries",{"Device QR"}),
    #"Expanded Filtered Rows" = Table.ExpandTableColumn(#"Removed Columns", "Filtered Rows", {"Device QR"}, {"Filtered Rows.Device QR"}),
    #"Renamed Columns" = Table.RenameColumns(#"Expanded Filtered Rows",{{"Filtered Rows.Device QR", "Device QR"}}),
    #"Reordered Columns" = Table.ReorderColumns(#"Renamed Columns",{"Device name", "Device QR", "Serial QR"})
in
    #"Reordered Columns"

I'd suggest you use that code and have a look into the APPLIED STEPS on the right to see what I have done. Essentially, I filered only the rows that do not have NULLs in Device QR. I then joined that table with the original table. From there, it was just some cosmetics like dropping the original column, expanding the Device QR column from the merged table, renaming and finally some reordering.

Let me know if this helps 🙂

/Tom
https://www.tackytech.blog/
https://www.instagram.com/tackytechtom/

[Theiamania]

Thanks @tackytechtom 

Your solution works but generates new rows that did not originally exist. These new rows appear to be all Devices but they don't have any other Serial rows with the same 'Device name'.

Sorry, I didn't think that point was relevant and didn't specify it above. I leave an example below of what the table originally looks like.

Device name	Device QR	Serial QR
Device 1	00011
Device 1		S0012
Device 1		S0013
Device 2	00022
Device 2		S0023
Device 3	00033
Device 4	00044
Device 4		S0045

This duplicity is created in the 'Expanded Filtered Rows' step. But, from what you say and as I didn't specify the above detail, I think it makes the whole process approach no longer valid. Is this the case?

I am very grateful for your time.