Forum Discussion
Boolean column if an ID has specified value in any row?
Hi.
I have a table of graduates and the degrees they've earned. Like so:
| ID | DEGREE | YEAR | RESTRICTED |
| 123 | BA | 2020 | T |
| 123 | MS | 2022 | F |
| 121 | BS | 2000 | F |
| 111 | BBA | 1970 | F |
| 142 | MD | 1981 | F |
I'd like to generate a column in Power Query that returns T for all rows associated with ID 123. The business logic is that anyone with any restricted degree should show up as being restricted.
Thanks!
You can group by ID taking the max over RESTRICTED and then merge that back with the original table like this:
let Source = Table.FromRows(Json.Document(Binary.Decompress(Binary.FromText("i45WMjQyVtJRcnIEEkYGRgZAKkQpVgcm7hsMETcCUm5QcUOQeoi4gQFC3BAsDjbI0NIcScIEpNnXBSxuYQgRjwUA", BinaryEncoding.Base64), Compression.Deflate)), let _t = ((type nullable text) meta [Serialized.Text = true]) in type table [ID = _t, DEGREE = _t, YEAR = _t, RESTRICTED = _t]), #"Changed Type" = Table.TransformColumnTypes(Source,{{"ID", Int64.Type}, {"DEGREE", type text}, {"YEAR", Int64.Type}, {"RESTRICTED", type text}}), #"Grouped Rows" = Table.Group(#"Changed Type", {"ID"}, {{"AnyRestricted", each List.Max([RESTRICTED]), type nullable text}}), #"Merged Queries" = Table.NestedJoin(#"Changed Type", {"ID"}, #"Grouped Rows", {"ID"}, "Grouped Rows", JoinKind.LeftOuter), #"Expanded Grouped Rows" = Table.ExpandTableColumn(#"Merged Queries", "Grouped Rows", {"AnyRestricted"}, {"AnyRestricted"}) in #"Expanded Grouped Rows"
7 Replies
- freginier
Solution Sage
You can easily do this in DAX
Column = var MinYear = CALCULATE( MIN('Table'[Year]), FILTER('Table',EARLIER('Table'[ID])='Table'[ID]) ) return LOOKUPVALUE('Table'[Rest],'Table'[Year],MinYear)The result
- MarkPalmberg
Kudo Commander
Thanks for the reply, freginier . I'm trying to derive this field in Power Query because I'm creating a table in a dataflow, so it hasn't hit a .pbix at this point. Guess maybe I should just do it in SQL as part of the initial table load.
- KNP
Super User
Shouldn't the second row be 'T' also, or am I misunderstanding the logic.
Try this:
let Source = Table.FromRows( Json.Document( Binary.Decompress( Binary.FromText( "i45WMjQyVtJRcnIEEkYGRgZKsTowMd9giJgRVMwQpA4iZgBVZwgWA2s2tDSHCpoYgTS7gMUsDJViYwE=", BinaryEncoding.Base64 ), Compression.Deflate ) ), let _t = ((type nullable text) meta [Serialized.Text = true]) in type table [ID = _t, DEGREE = _t, YEAR = _t] ), #"Changed Type" = Table.TransformColumnTypes( Source, {{"ID", Int64.Type}, {"DEGREE", type text}, {"YEAR", Int64.Type}} ), #"Added Custom" = Table.AddColumn( #"Changed Type", "Restricted", each if [ID] = 123 then "T" else "F" ), #"Changed Type1" = Table.TransformColumnTypes(#"Added Custom", {{"Restricted", type text}}) in #"Changed Type1"Specifically, the 'Added Custom' step.
- AlexisOlson
Super User
You can group by ID taking the max over RESTRICTED and then merge that back with the original table like this:
let Source = Table.FromRows(Json.Document(Binary.Decompress(Binary.FromText("i45WMjQyVtJRcnIEEkYGRgZAKkQpVgcm7hsMETcCUm5QcUOQeoi4gQFC3BAsDjbI0NIcScIEpNnXBSxuYQgRjwUA", BinaryEncoding.Base64), Compression.Deflate)), let _t = ((type nullable text) meta [Serialized.Text = true]) in type table [ID = _t, DEGREE = _t, YEAR = _t, RESTRICTED = _t]), #"Changed Type" = Table.TransformColumnTypes(Source,{{"ID", Int64.Type}, {"DEGREE", type text}, {"YEAR", Int64.Type}, {"RESTRICTED", type text}}), #"Grouped Rows" = Table.Group(#"Changed Type", {"ID"}, {{"AnyRestricted", each List.Max([RESTRICTED]), type nullable text}}), #"Merged Queries" = Table.NestedJoin(#"Changed Type", {"ID"}, #"Grouped Rows", {"ID"}, "Grouped Rows", JoinKind.LeftOuter), #"Expanded Grouped Rows" = Table.ExpandTableColumn(#"Merged Queries", "Grouped Rows", {"AnyRestricted"}, {"AnyRestricted"}) in #"Expanded Grouped Rows"- KNP
Super User
Ah, yep. After seeing AlexisOlson's reply I understand the logic better.
I should read properly to start with. 😆
This will work.
- MarkPalmberg
Kudo Commander
Sorry, KNP , that was my bad. I should've included my desired result column in that table. Thanks so much for taking the time to look at this.
- MarkPalmberg
Kudo Commander
Ah, that's it. Thanks so much, AlexisOlson . I really appreciate your time.