Transform Rolling Weeks Data To Only Filter To Max Date Per Source File

Hi dswinden 
you can add a conditional column that compares the week with the received :

and filter out all those rows wich null :

Result (the column test can be deleted)

pbix is attached

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

maybe you can try this

let
Source = Table.FromRows(Json.Document(Binary.Decompress(Binary.FromText("jZBBCsUgDAXv4rqgpkkbz1K8/zV++ApN2hC7Gxgd5V1XggKY94xpGwi5gvCJmPrm6ibMcsjXcAi3crh6IDNpW2sc196r336Rn0zyiPb8ylff+3n+mp/8yAOFea29vPGL/J8Z+faYw3GsftWNDuMDWzOXw2GM9tr8rT2QCpp/rUahMK51WJ/MZ0m9/wA=", BinaryEncoding.Base64), Compression.Deflate)), let _t = ((type nullable text) meta [Serialized.Text = true]) in type table [#"Data file received" = _t, week = _t, Sales = _t]),
#"Changed Type" = Table.TransformColumnTypes(Source,{{"Data file received", type date}, {"week", type date}, {"Sales", Int64.Type}}),
#"Added Custom" = Table.AddColumn(#"Changed Type", "Custom", each if [Data file received] =List.Min(#"Changed Type"[#"Data file received"]) or [Data file received]= [week] then 1 else 0),
#"Filtered Rows" = Table.SelectRows(#"Added Custom", each ([Custom] = 1)),
#"Removed Columns" = Table.RemoveColumns(#"Filtered Rows",{"Custom"})
in
#"Removed Columns"

pls see the attachment below