Automatically include new Columns in Dataset

[Anonymous]

I am maintaining a PowerBI dataset which basically reads columns from SQL tables. I use SELECT * FROM [TABLE] in source query, so all columns from the table are included. Once I publish the .pbix, others can source from this to create their report.

Now, if the pipeline feeding the SQL table adds a new column, I am expecting to see this column in the next scheduled refresh. Unfortuantely, this isn't happening. The only way I can see it is, if I open the dataset .pbix, refresh and then publish.

Is this a known limitation of power bi data set using SQL SELECT * statement to create columns?

[SivaMani · Accepted solution]

@Anonymous, The refresh that you referred to is DATA Refresh. It will only fetch the data from your source. It won't add new columns. For any structural changes or metadata changes, you have to refresh it in Power BI Desktop and publish it.

[nickyvv · Accepted solution]

Hi @Anonymous,

good question. As @SivaMani mentioned this is not supported out-of-the-box.

However, you can work around it if the source is supporting that:

• In this blog post, Marc shows how to get columns dynamically when working with SharePoint files
• In this blog post, Erik shows how to expand columns dynamically when working with SQL

While I don't think this works directly with your example, you might wanna read those and try to make it work with your example.

Furthermore: did you have a read on the Limitations and issues with native database queries (Like writing your SELECT * statement)?

[Anonymous]

Although doesn't solve the problem at hand, but closest to an acceptable solution.

[SivaMani · Accepted solution]

@Anonymous, The refresh that you referred to is DATA Refresh. It will only fetch the data from your source. It won't add new columns. For any structural changes or metadata changes, you have to refresh it in Power BI Desktop and publish it.