How to get rid of some kind of string autocorrection, while loading data from powerquery into model

[elaj]

Hi,

i have data wich looks like:

notice the difference with the capital F.

i load this dataset into powerquery as a csv.. and everything looks fine..

but when i load it into the datamodel.. somehow PowerBI is transforming the capital F into a lowercase f.

This is really annoying, because i dont want it to. These are two different things... And i want it to be different...

Do you know this problem?

is there a workaround?

thanks

elaj

[PhilipTreacy · Accepted solution]

Hi @elaj 

Unfortunately, annoying as this can be, it is by design.

Rather than rewrite the contents of this excellent article on the subject - please read this

Letter case-sensitivity in DAX, Power BI and Analysis Services - SQLBI 

But to sum up the issue you are having: DAX sees friend and Friend as the same thing because DAX is case insensitive.

The storage engine in PBI builds a dictionary of the distinct values in your data, and because it is case insensitive, Hello (friend) and Hello (Friend) are the same in its view.

Hello (friend) is stored in the dictionary and any other occurrence of that string is merely given an Index to that string.

Regards

Phil