Edit Python Get Data script without losing formatting

[MP_T]

I have a Power BI report that gets data via a python script. The import script retrieves around 25 tables and supports a report of multiple pages. However, if I want to make small edit to the get data script, I can either delete the original data tables in Power BI or just import again. If I import again I get duplicate tables. Deleting and reimporting means I lose most or all of the column formatting and lose formatting in the visuals on settings like currency and how many decimal places to show. Is there a way that I can edit the python get data script without losing the table formatting and settings in the visuals? For example say the script is pulling in 25 tables and I want to amend it so it will pull in a 26th but the other 25 are unchanged, I don't want to have to manually change the formatting on the other 25 tables & all the visuals every time I modify the python script.

One solution could be to put a database (or flat files) in the middle so the python script populates the database (or flat files) and Power BI queries the database (or flat files) but I am looking for a direct solution. If it isn't possible any other workaround suggestions might also be useful.

I am new to Power BI but have quite a lot of python experience.

[MP_T · Accepted solution]

Hi Daryl

Thanks for this link, it does look like it would help. In the end I went for the flat file approach. That means that I don't have to get all the python into a single script inside Power BI and I can keep all the coding on the python side. I'm sure there are other ways to do this and potentially better ways but at least the original issue was resolved.

[Daryl-Lynch-Bzy]

Hi @MP_T , I was wondering if the following blog post helps you understand how Power BI approaches data types when expanding tables:

Chris Webb's BI Blog: Setting Data Types On Columns Containing Table Values In M Chris Webb's BI Blo...

There are some additional changes that you can do make this more dynamic if the columns in the Python results are changing.