Power BI Desktop file is in folder with some Excel files. Those Excel files are used as a data source. If we move folder with all of this files to some other location, refreshing of PBID file will not work any more. PBID will not be able to connect to Excel files because fullpath of those Excel files will be changed. Solution is to find address of a folder where PBID file reside. That way, we will always know where our Excel files are.

When we run a PBID, a process with the name PBIDesktop.exe is created. We can find it in Details tab in Task Manager:

There is a console program named Handle64. That program can list all the files that are locked by specific process. In CMD environment we can type:

C:\Users\Sima\Desktop\handle64.exe -p PBID

This will return all files locked by PBIDesktop process:

Because our PBIX file is locked by PBIDDesktop.exe, its fullpath will be listed in there. We just have to filter this list by using name of our PBIX file and we will find location of our PBIX file. That way we will find folders where our Excel files are located, so we can create relative path.

This console program can be downloaded from the address below, but it is also available at the end of this blog post.
https://docs.microsoft.com/en-us/sysinternals/downloads/handle

All we have to do now is to call this Handle64 program from PBID, and for this we are going to use python script. For python script to work we have to tell Power BI desktop where is pythons home folder. We can do this in Options > Python scripting. If we have python installed we can select its home folder from drop down menu (1). I am using portable version of python so I have to tell PBID explicitly where my python is. I can do that with (2) Browse button.

This is our python script. Blue part will read result of CMD command and encode it in UTF-8. Green part will remove all other text, so that at the end, we only have our fullpath to return.

import pandas
import subprocess
FilesListBytes = subprocess.check_output(r'"D:\Programi\handle64.exe" -p PBID', shell = True)
FilesListString = FilesListBytes.decode('utf-8')
FirstSplit = FilesListString.split('Name of the file.pbix', 1)
SecondSplit = (FirstSplit[0].rsplit('File     ',1))[1]
ReplaceSlashes = SecondSplit.strip().replace("//","/")
df = pandas.DataFrame([[SecondSplit.strip()]],columns=['Files'])
print (df)	

Name of our PBID file will not be hardcoded. We will read it from table that we prepare in advance in our PBID file.

Sample PBID file, whole Power Query script that contains python code, can be found in this attachment. For this script to work on your computer you have to change fullpath of Handle64.exe. That fullpath is in the first line of the script:
Handle64location = """C:\Users\Sima\Desktop\handle64.exe""", change it to where you placed Handle64.exe program.

Culture.Current

This is a variable that returns current culture. Current culture is described by symbols like "cs-cz" (for Czech language) or "en-gb" (for UK english). Culture is defined in Options, under Regional Settings, both in PowerPivot and Power BI Desktop.

The code below would return string "sr-Latn-RS". This blog post has, bellow, attachment "Language codes.xlsx" which contains list of all language codes.

let ‏‏‎ ‎‏‏‎ ‎‏‏‎ ‎‏‏‎ ‎‏‏‎ ‎‏‏‎ ‎‏‏‎ ‎‏‏‎ ‎Source = Culture.Current in ‏‏‎ ‎‏‏‎ ‎‏‏‎ ‎‏‏‎ ‎‏‏‎ ‎‏‏‎ ‎‏‏‎ ‎‏‏‎ ‎Source

Comparer.FromCulture()

Imagine that we want to compare two values based on specific culture and case sensitivity. To compare strings "A" and "a" based on "sr-Latn-RS" culture while ignoring case sensitivity, we could use some function that look like this:

SomeFunctionUsedForComparison( "sr-Latn-RS", true, "A", "a" )

Power query can do the same thing but it does it in two steps. First step is to create comparison function:

ComparisonFunction = Comparer.FromCulture("sr-Latn-RS", true)

Second argument is for case sensitivity. Default is false (sensitive). This argument is optional.
After this, we can use our new ComparisonFunction to make comparison when ever we need it.

Second step is to call this function when we need it. We just supply two values to compare. ResultOfComparison = ComparisonFunction( "A", "a") We can see on the right, that this new function receives two arguments to compare. Those two arguments could be of any type. As a result, this new function returns a number. That number is "0" for equality and "-1" or "1" for inequality.

Here are three examples of this function. For brevity we will merge two steps into one step:

Comparer.FromCulture("sr-Latn-RS", true)( "A", "a" )	Returns "0". Equality.
Comparer.FromCulture("sr-Latn-RS", false)( "a", "A" )	Returns "-1". Inequality. "a" < "A".
Comparer.FromCulture("sr-Latn-RS", false)( "A", "a" )	Returns "1". Inequality. "A" > "a".

We can see that, because of different case sensitivity, we get different results.

Comparer.Ordinal()

This function is used to compare two values. Comparing is based on ordinal rules. Lets asume that we compare:

æ

=UNICHAR(230) 'by excel formula

AND

ae

=UNICHAR(97)&UNICHAR(101) 'by excel formula

If we compare this two values with Comparer.FromCulture, for the result we would get equality.

Comparer.FromCulture( "sr-Latn-RS", true )( "æ", "ae" )

Returns "0". Equality.

This is because, culture help us to compare values that have similar meaning.

Comparer.Ordinal() is ortodox function. It compares unicode codes, and not the meaning of symbols. This function will compare code "230" with codes "97" and "101". It will find that this two values are not the same. Let's see three examples for this function:

Comparer.Ordinal("æ","ae")	Returns "1". This is because 230 > 97.
Comparer.Ordinal("ae","æ")	Returns "-1". This is because 97 < 230.
Comparer.Ordinal("æ","æ")	Returns "0". This is because 230 = 230.

As we see, when comparing numbers, we know which number is bigger. That is way we get results "-1,1" depending on which number is larger or "0" if numbers are equal.
"ae" is consider as two unicode numbers, so for concluding sort order we only need code number for letter "a".

Comparer. OrdinalIgnoreCase()

This function is the same as Comparer.Ordinal(), except it is case insensitive.

Comparer.OrdinalIgnoreCase( "A", "a" )	Returns 0.
Comparer.OrdinalIgnoreCase( "a", "A" )	Returns 0.
Comparer.OrdinalIgnoreCase( "A", "A" )	Returns 0.
Comparer.OrdinalIgnoreCase( "A", "b" )	Returns -1.
Comparer.OrdinalIgnoreCase( "b", "A" )	Returns 1.

Comparer.Equals()

This function can use previous functions as arguments to compare two values. Here are three examples:

Comparer.Equals( Comparer.FromCulture( "nl-NL", true ), "a", "A")	Returns "true".
Comparer.Equals( Comparer.Ordinal, "a", "A")	Returns "false".
Comparer.Equals( Comparer.OrdinalIgnoreCase, "z", "z")	Returns "true".

As we can see, we are using three arguments. First one is function defining how to compare values in other two arguments. Comparer.Equals() returns "true" if values are equal and "false" if they are different.

List of all culture codes and examples for comparer functions can be found in this file: