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
Hello, I am looking to extract numbers before specific text ("BUP"). I think it is possible we may need to address " BUP" specifically, as users would likely put a space between number and "BUP". Example below:
1. "40 BUP" = 40
2. "@LOCATION 25 BUP" = 25
3. "hello_world" = 0
Appreciate the help.
Solved! Go to Solution.
Use this in a custom column
= if Text.End([Data],3)="BUP" then Text.Replace(List.Last(Text.Split(Text.Select(Text.Replace([Data]," BUP","BUP"),{"0".."9"," ","B","U","P"})," ")),"BUP","") else 0Full test code
let
Source = Table.FromRows(Json.Document(Binary.Decompress(Binary.FromText("i45WMjFQcAoNUIrViVZy8PF3dgzx9PdTMDKFC2ak5uTkx5fnF+WkoCoyMoUpMTGNKTUwMDKDkHDNsQA=", BinaryEncoding.Base64), Compression.Deflate)), let _t = ((type nullable text) meta [Serialized.Text = true]) in type table [Data = _t]),
#"Added Custom" = Table.AddColumn(Source, "Custom", each if Text.End([Data],3)="BUP" then Text.Replace(List.Last(Text.Split(Text.Select(Text.Replace([Data]," BUP","BUP"),{"0".."9"," ","B","U","P"})," ")),"BUP","") else 0, type number)
in
#"Added Custom"
In a custom column, put following formula
= Text.Select(List.First(Text.Split(Text.Replace([Data]," ","")&"0BUP","BUP")),{"0".."9"})See the working here - Open a blank query - Home - Advanced Editor - Remove everything from there and paste the below code to test
let
Source = Table.FromRows(Json.Document(Binary.Decompress(Binary.FromText("i45WMjFQcAoNUIrViVZy8PF3dgzx9PdTMDKFC2ak5uTkx5fnF+WkoCoyMgUriQUA", BinaryEncoding.Base64), Compression.Deflate)), let _t = ((type nullable text) meta [Serialized.Text = true]) in type table [Data = _t]),
#"Added Custom" = Table.AddColumn(Source, "Custom", each Text.Select(List.First(Text.Split(Text.Replace([Data]," ","")&"0BUP","BUP")),{"0".."9"}), type number)
in
#"Added Custom"
@Vijay_A_Verma - This solution works perfectly (per my description) but it encounters an issue that I didn't anticipate:
"45&& 25 BUP" = "4525 BUP" (with your solution). Ideally it would return "25 BUP".
Is there a way to ammend the code so it will only capture numbers before the string "BUP"? Once a non-number field is encountered, it will not return subsequent characters.
Thank you so much for your help! You're a life saver : )
Use this in a custom column
= if Text.End([Data],3)="BUP" then Text.Replace(List.Last(Text.Split(Text.Select(Text.Replace([Data]," BUP","BUP"),{"0".."9"," ","B","U","P"})," ")),"BUP","") else 0Full test code
let
Source = Table.FromRows(Json.Document(Binary.Decompress(Binary.FromText("i45WMjFQcAoNUIrViVZy8PF3dgzx9PdTMDKFC2ak5uTkx5fnF+WkoCoyMoUpMTGNKTUwMDKDkHDNsQA=", BinaryEncoding.Base64), Compression.Deflate)), let _t = ((type nullable text) meta [Serialized.Text = true]) in type table [Data = _t]),
#"Added Custom" = Table.AddColumn(Source, "Custom", each if Text.End([Data],3)="BUP" then Text.Replace(List.Last(Text.Split(Text.Select(Text.Replace([Data]," BUP","BUP"),{"0".."9"," ","B","U","P"})," ")),"BUP","") else 0, type number)
in
#"Added Custom"
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 | |
| 2 | |
| 2 | |
| 1 |
| User | Count |
|---|---|
| 11 | |
| 11 | |
| 5 | |
| 4 | |
| 4 |