This post is also available in: Deutsch

Last week we already dealt with the topic of the correct calculation of the ISO week in DAX. Since the question came up about how to solve this in Power Query, I will look into it this week. For the problem with calendar weeks, I would like to refer to last week’s post.

Calculation of US calendar week with Date.WeekOfYear function

Like in DAX, there is already a function in Power Query to calculate the calendar week. This is called Date.WeekOfYear and returns the US format with week start on Sunday. With an optional parameter, the

This post is also available in: Deutsch

If you find a date or datetime format in UNIX format, the first thing you wonder is what it is all about or how you can display it “normally”. This is because the UNIX date format looks like this:


What exactly is UNIX Timestamp?

To find a way to convert this, we first need to understand what the number means. The UNIX timestamp is the number of seconds since January 01, 1970. You can read exactly about why it is like this in the Wikipedia article. I want to focus on finding a

This post is also available in: Deutsch

If you want to combine several tables, you have to ask yourself whether you should do this already in the data source such as SQL Server, in the processing of the data (ETL) in Power Query or in Power BI and, above all, where exactly there is a difference.

First of all, the post is relatively detailed, for all of those who consider it too long (tl;dr) here is the comparison table:

Duplicate entries remainDuplicate entries are removedUnequal number of columnsDeviating column namesDifferent order of the columnsCombine more than 2 tables
APPEND in Power QueryxPossible with “Remove duplicates”xxx
UNION in DAXxPossible with “DISTINCT”xx
Overview of the different possible combinations

Data transformation in the data source or in the frontend?

A question that

This post is also available in: Deutsch

What at first sounds complicated and not very useful, can be found in reality from time to time. Recently the automated export of source data from another program in CSV format confronted me with this task.

Daily folder for automated export

Unfortunately, when exporting from the source system – apart from the target directory – very little could be adjusted. The result is that the automated export job daily creates a new folder with the current date, similar to the following example:

In this example it is only necessary to load the files from the newest folder, because this folder contains the entire data. However, the file names also change with