Get certified for free when you join Fabric Data Days 2026 and dive into Fabric, Power BI, SQL, AI, and other essential data skills.
Join nowData Days is here! Join us now for 60+ days of learning, challenges, and connection. Learn more
Hi all,
I’ve spend a lot of time in fixing the error (youtube, this forum). What i did?: I’ve imported a table into excel, for which I want to transform (delete) data with power query. I only want to keep data with values “PD-*” So * is a wildcard, hence can be each number.
Below are the values in a coumn, the only value I want to keep from below example is PD-35:
Column values: PR-00001;WL-00009;PD-35
Power Query:
= Table.TransformColumns(#"Duplicated Column", {{"Linked Issues - Copy", each if Text.Contains(_,"PR-") then Text.Remove {"P""R""-""0""9"} else if Text.Contains(_,"WL-") then Text.Remove {"W""L""-""0""9"} else _, type text}})
After executing above Query its throwing the following error:
Error:
Expression.Error: We cannot apply an index of type Text.
Details:
Value=[Function]
Index=P"”R"-"0"9
So simply I get an error in all fields containing either of above values I also tried to play with Text.BetweenDelimiters or “replace” but I cannot get it to work. I hope someone can help me out.
Solved! Go to Solution.
NewStep=Table.TransformColumns(PreviousStepName,{"ColumnName",each Text.Combine(List.Select(Text.Split(_,";"),each Text.StartsWith(_,"PD-")),";")})
Thanks for you swift reply. The query applies a filter, i would like the query to remove all values that does not start with "PD-"
example of values in column
| Example | Expect after query removed data |
| PR-00001;WL-00009;PD-35 | PD-35 |
| XL-00001;PD-0016;WL-00010; | PD-0016 |
| XL-00001;PD-0016;PD-00015; | PD-0016;PD-00015; (Exeptional case, above two is most important) |
NewStep=Table.TransformColumns(PreviousStepName,{"ColumnName",each Text.Combine(List.Select(Text.Split(_,";"),each Text.StartsWith(_,"PD-")),";")})
Thanks this did the job!
Sure, no prob. Here you are.
let
Source = Table.FromRows(Json.Document(Binary.Decompress(Binary.FromText("i45WCgjSNQACQ+twHzDD0jrARdfY1FopVidaKcIHKgkUA1JmUEWGBjikwbSBIUx3ZBSEEeCCEIAoARkQCwA=", BinaryEncoding.Base64), Compression.Deflate)), let _t = ((type nullable text) meta [Serialized.Text = true]) in type table [Example = _t]),
#"Changed Type" = Table.TransformColumnTypes(Source,{{"Example", type text}}),
#"Added Custom" = Table.AddColumn(#"Changed Type", "Custom", each Text.Combine ( List.Select(Splitter.SplitTextByDelimiter(";", QuoteStyle.None)([Example]), each Text.StartsWith(_, "PD-")), ";"), type text)
in
#"Added Custom"
Hello - if you'd only like to keep rows that have a value beginning with 'PD-' in the 'Linked Issues - Copy' column, you can do it like this:
Table.SelectRows(#"Changed Type", each Text.StartsWith([#"Linked Issues - Copy"], "PD-"))
Don't miss out on Data Days, June 15 through August 7. Learn Fabric, Power BI, SQL, AI and more.
Check out the May 2026 Power BI update to learn about new features.
| User | Count |
|---|---|
| 4 | |
| 4 | |
| 3 | |
| 2 | |
| 1 |
| User | Count |
|---|---|
| 11 | |
| 11 | |
| 5 | |
| 4 | |
| 4 |