Document Download seems to be corrupting certain characters

[VynetaRy]

I am using a Download Document module to download a CSV file and parsing the data from it with a Parse File module. It is not a binary file, but there are some special characters in the file since the addresses are located in Mexico. When I look at the parse file module input from the downloaded file, I see where those characters have been replaced with a question mark in a diamond ( P.º Xaman becomes P.� Xaman), but when I paste that same data directly into the parse CSV module and run just that module, the address results are correct. The only thing I can guess is that in the download process, those characters are being changed somehow. This even happens with an em dash character.  I tried adding the toString() function, but the results are the same. Is there anything I can do to preserve those characters? I can't just remove them with the ascii function because the address will no longer be valid. Workfront itself allows those characters if I manually store the address in the custom form. What is it about Fusion that alters those characters?

I have manually downloaded and inspected the CSV file from the project, and the data is correct.

[Sven-iX · Accepted solution]

Fusion doesn't alter characters. The problem likely is the character set. 
Can you check that the CSV is using UTF-8? 

How and from where are you getting the CSV? 

[VynetaRy]

Thanks, Sven!

I was exporting the CSV from Excel. I found an option to save a UTF-8 CSV (I didn't even know that was possible, but had wondered), and that fixed things.  I appreciate the suggestion!