Forum Discussion
Help with the count
- 2 years ago
Just2431 If you had posted your sample data as text, I could have given you a working PBIX file. So, instead, I'll try to explain the process. In PQ Editor, do a Group By operation and use Doc id as the grouping. Use Advanced. Add a Count aggregation as well as an aggregation that keeps all rows. Now, expand the all rows column for your operation_name and time. Now you will have a new column that provides the count of operations for each doc id on each row and also preserve the rest of your data.
- 2 years ago
Hi Just2431, try this:
Result
let Source = Table.FromRows(Json.Document(Binary.Decompress(Binary.FromText("i45WCnAPV9JR8snMy87MSweyDJVidWCiSfkVEEFjJMHkotTEklQgwwRZMDEnuRhImyKJpeaVpBZB9BuBhb19ojyw2AUVhluGohhuG4pamHXGyIJI9qEoTiwoKMovS8yBq3d0cgayCxPhCiECaA6ACMJsMtIDei0WAA==", BinaryEncoding.Base64), Compression.Deflate)), let _t = ((type nullable text) meta [Serialized.Text = true]) in type table [#"Doc id" = _t, operation_name = _t, time = _t]), ChangedType = Table.TransformColumnTypes(Source,{{"time", type number}}, "en-US"), GroupedRows = Table.Group(ChangedType, {"Doc id"}, {{"All", each _, type table [Doc id=nullable text, operation_name=nullable text, time=nullable number]}, {"Operations", each Table.RowCount(_), Int64.Type}}), ExpandedAll = Table.ExpandTableColumn(GroupedRows, "All", {"operation_name", "time"}, {"operation_name", "time"}) in ExpandedAll
Just2431 If you had posted your sample data as text, I could have given you a working PBIX file. So, instead, I'll try to explain the process. In PQ Editor, do a Group By operation and use Doc id as the grouping. Use Advanced. Add a Count aggregation as well as an aggregation that keeps all rows. Now, expand the all rows column for your operation_name and time. Now you will have a new column that provides the count of operations for each doc id on each row and also preserve the rest of your data.