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 everybody, i need your knowledge in PQ.
I have a query with the following look. In column 1 is a project number written. I have to evalute if there is delay for a specific project activity. Column 2 is filled with the startday. Sometimes I don’t have a start day, just an endday of project activity, because just planned endday and the real endday are important.
As you can see in the example picture i have the projcect numbers 537003/537004/537005 and 537006.
Just project 003 and 006 have a activity called „Real endate of procurment“
Because the evaluation should be shown row by row i just want to have one row for a specific project value.
My first idea was to fill down the values for a a column and to remove duplicates. Problem in this example:
Also for the projcects 004&005 an planned enddate for this acitivity will be shown.
Second idea: fill down all values first and the set the rows blank, when there is now activity called „Real endate of procurment“. I think for just a few datasets this will be possible regarding evaluation speed. With bigger datasets, i think this will get slower and slower.
Is there an other possibility to fill down the values in column if the row above is the same, so that the values will be blank for projects 004&005?
Thanks in advance.
Ok. I think I already found a solution in this forum:
https://community.fabric.microsoft.com/t5/Desktop/Conditional-Fill-Down/m-p/3104559#M1052282
But other question from my side:
Can somebody exlpain me this code?
What is the meaning of this first code row where the source is defined. How can i now this kind of source???
Source = Table.FromRows(Json.Document(Binary.Decompress(Binary.FromText("i45WClIwNNI3NNAFkSZKOkpKsTo4BI0VUhIri2E8QwtjEyNLIM+pKLEqMyczMQ9VGAvPKzFP18gYyHPPSc1TcE1Mz0ktRkhA7IkFAA==", BinaryEncoding.Base64), Compression.Deflate)), let _t = ((type nullable text) meta [Serialized.Text = true]) in type table [#"Invoice#" = _t, Property = _t]),
Hi @Anonymous
The Source code you pasted is automatically generated by Power Query. This is usually generated after you use Enter Data feature to enter a table data manually.
You don't have to have this kind of source as you already have data in your query. If you want to do something similar to the solution you have found, you just need to follow the next steps after Source step in your table. The solution you found has a sample pbix file. You can download it to see what happens step by step.
Best Regards,
Community Support Team _ Jing
If this post helps, please Accept it as Solution to help other members find it.
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 |