Config file "byte order mark"

[JW-JnJ]

I wanted to put this out in this forum as a warning for future searching. I was ripping my hair out for awhile on this one.

Attached are two files that I finally boiled my debug down to. What's the difference? The broken one has the BOM (byte order mark) included in the UTF-8 format. This is essentially a header to a text file. EVERY text editor just skips over the bytes, so I had to find this with a hex reader. What this did is render first section "unfindable" to key reading.

The conversion somehow happened while using Notepad++ to edit my INI files (somehow... still not sure). Fixing it is as simple as selecting "Encoding", "Encode in UTF-8 without BOM". Funny enough, another way to fix it is to just add a blank or header line to the top of the file. 

Now that I'm past the warning, I wanted to get a general sense if this is well-known. (debating on adding an Idea Exchange to either error if the wrong BOM for native, or skip if correct).

[Supertramp]

Thanks JW-JnJ, 

I will work on it to see which is the better solution. 

Best regards. 

[MHn]

I had the same issue after editing a .ini-file with notepad on a Windows10 system and LabVIEW 2014/2017.

best regards,

Martin

[deceased]

Never seen it with an ini file before, but I have seen the BOM break other file parsers.

If a parser can't deal with those unexpected* bytes at the start, you're going to have a bad time.