Two value types in one column

[Artur_]

Hi! I have two columns in a self-referencing table tracker, both with dates (01/12/2020 format) and strings (e.g. “Pending”). How to preserve both different types considering that another column makes a calculation on those two (date difference)?

(this is what I would like to achieve)

My two attempts:

1. Set a date type: string disappears at refresh
2. Set a general or string type: calculation won’t be done

Maybe there is some way to perform the calculation in a different way? I'm using

= (StartDate as date, EndDate as date, optional Holidays as list) =>
let
    ListOfDates = List.Dates(StartDate,Number.From(EndDate-StartDate),#duration(1,0,0,0)),
    DeleteHolidays = if Holidays = null then ListOfDates else List.Difference(ListOfDates, Holidays),
    DeleteWeekends = List.Select(DeleteHolidays, each Date.DayOfWeek(_,1) < 5 ),
    CountDays = List.Count(DeleteWeekends)
in
    CountDays

Thanks for your invaluable help.

[v-frfei-msft · Accepted solution]

Hi @Artur_ ,

We can create a custom column like that.

if [R date] = "Pending" then 0 else Duration.Days([F date]-Date.FromText([R date]))

M query for your reference.

let
    Source = Table.FromRows(Json.Document(Binary.Decompress(Binary.FromText("Vc87CoAwEADRu6QOJLv538I+pFPEJvcvFUUcy+FV07tZtrkeczfWiBNx6tWbYftV+sQNSgiAQIiASEiARMiATCiAQqiASmiARhD/ifqfCOR9Hyc=", BinaryEncoding.Base64), Compression.Deflate)), let _t = ((type text) meta [Serialized.Text = true]) in type table [#"R date" = _t, #"F date" = _t]),
    #"Changed Type" = Table.TransformColumnTypes(Source,{{"R date", type text}, {"F date", type date}}),
    #"Added Custom" = Table.AddColumn(#"Changed Type", "Custom", each if [R date] = "Pending" then 0 else Duration.Days([F date]-Date.FromText([R date])))
in
    #"Added Custom"

[Artur_]

hi @v-frfei-msft,

this is what I needed. Thanks so much! 🙂

I've amended it a bit because I had to exclude weekends and holidays. I'm leaving part of the code for future reference:

...
#"Invoked Custom Function" = Table.AddColumn(#"Added Conditional Column", "CeT Performance", each if [Transfer Receive Date] = "Pending" then "0" else Datediff(Date.FromText([Transfer Receive Date]), [Transfer finalization Date], holidays[Holiday]))
...

DATEDIFF:
= (StartDate as date, EndDate as date, optional Holidays as list) =>
let
    ListOfDates = List.Dates(StartDate,Number.From(EndDate-StartDate),#duration(1,0,0,0)),
    DeleteHolidays = if Holidays = null then ListOfDates else List.Difference(ListOfDates, Holidays),
    DeleteWeekends = List.Select(DeleteHolidays, each Date.DayOfWeek(_,1) < 5 ),
    CountDays = List.Count(DeleteWeekends)
in
    CountDays

[Geradav]

Hi @Artur_ 

It is not a good practice to mix data types in a column. It is recommended to only have 1 data type per column. Otherwise, you are going to have a lot of errors in your queries.

Separate both results in different columns.

Let us know how that works

Best

David