Forum Discussion
Need Help in Formula IF(A1=A2,AC1,AB2)
That's perfect, thanks.
You need to sort the data in the order that you want it to be evaluated, add an index column starting from zero, then add a new custom column with the following calculation:
try if [Document Id] = previousStepName[Document Id]{[Index] + 1} then [#"Actual End Date & Time"] else previousStepName[#"Actual Start Date & Time"]{[Index] + 1} otherwise nullExample working query:
let Source = Table.FromRows(Json.Document(Binary.Decompress(Binary.FromText("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", BinaryEncoding.Base64), Compression.Deflate)), let _t = ((type nullable text) meta [Serialized.Text = true]) in type table [#"Document Id" = _t, #"Actual Start Date & Time" = _t, #"Actual End Date & Time" = _t, #"Entity Start Date" = _t]), chgTypes = Table.TransformColumnTypes(Source,{{"Document Id", Int64.Type}, {"Actual Start Date & Time", type datetime}, {"Actual End Date & Time", type datetime}, {"Entity Start Date", type datetime}}), sortRows = Table.Sort(chgTypes,{{"Document Id", Order.Ascending}, {"Actual Start Date & Time", Order.Ascending}}), addIndex0 = Table.AddIndexColumn(sortRows, "Index", 0, 1, Int64.Type), addCalcStartDate = Table.AddColumn(addIndex0, "calcStartDate", each try if [Document Id] = addIndex0[Document Id]{[Index] + 1} then [#"Actual End Date & Time"] else addIndex0[#"Actual Start Date & Time"]{[Index] + 1} otherwise null ) in addCalcStartDateExample output:
Pete
That's perfect, thanks.
You need to sort the data in the order that you want it to be evaluated, add an index column starting from zero, then add a new custom column with the following calculation:
try
if [Document Id] = previousStepName[Document Id]{[Index] + 1} then [#"Actual End Date & Time"] else previousStepName[#"Actual Start Date & Time"]{[Index] + 1}
otherwise null
Example working query:
let
Source = Table.FromRows(Json.Document(Binary.Decompress(Binary.FromText("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", BinaryEncoding.Base64), Compression.Deflate)), let _t = ((type nullable text) meta [Serialized.Text = true]) in type table [#"Document Id" = _t, #"Actual Start Date & Time" = _t, #"Actual End Date & Time" = _t, #"Entity Start Date" = _t]),
chgTypes = Table.TransformColumnTypes(Source,{{"Document Id", Int64.Type}, {"Actual Start Date & Time", type datetime}, {"Actual End Date & Time", type datetime}, {"Entity Start Date", type datetime}}),
sortRows = Table.Sort(chgTypes,{{"Document Id", Order.Ascending}, {"Actual Start Date & Time", Order.Ascending}}),
addIndex0 = Table.AddIndexColumn(sortRows, "Index", 0, 1, Int64.Type),
addCalcStartDate =
Table.AddColumn(addIndex0, "calcStartDate", each
try
if [Document Id] = addIndex0[Document Id]{[Index] + 1}
then [#"Actual End Date & Time"]
else addIndex0[#"Actual Start Date & Time"]{[Index] + 1}
otherwise null
)
in
addCalcStartDate
Example output:
Pete
Hi Pete,
Solution gives me correct data but ufortunately when click applied and closed andit will increase the size of table to 1000 times. Can you please suggest why it happen
What do you mean it increases the size of the table 1,000 times? Are you getting duplicated rows or are you talking about application rows as it loads?
Pete
my sheet file size is 15MB and after adding this custom column when i click on close and apply button power bi load shows size more than 3 GB and it keeps on loading not able to save the changes in the table.
Ok. It sounds as though PQ is rescanning everything multiple times.
In your query, insert a step after #"Added Index" like this:
bufferStep = Table.Buffer(#"Added Index")Then adjust your calculation step to this:
try if [Document Id] = bufferStep[Document Id]{[Index] + 1} then [#"Actual End Date & Time"] else bufferStep[#"Actual Start Date & Time"]{[Index] + 1} otherwise nullNote that I've updated the step references to point to the buffered table. This should prevent PQ having to constantly reload the same table for scans.
Pete