A warehouse we work with put a change in where they agreed to write a '.tmp' extension on a file when transferring, and then change to xml once complete.

When we try and process the files we get this error message:

Failed tracking custom fields; Caused by: Unable to read data file ; Caused by: Invalid UTF-8 start byte 0x9f (at char #4, byte #-1)

Now after googling, this suggested an encoding issue. I noticed that some of the older files we received from them were UTF-8-BOM. I took one of the failed files, opened in notepad++, changed the encoding to UTF-8-BOM and boom - file processes with no issue.

However, they aren't accepting that its a fault their side. They're blaming the SFTP. They took both files I sent them, and checked both their side and both show UTF-8 encoding.

To throw more confusion in, i looked at some other random old files and these just have UTF-8 encoding, but had gone through ok.

I'm not mega familiar with the difference between UTF-8 and UTF-8-BOM. What's going on here?