This is indeed invalid UTF8, I believe; this file seems to be in a local ANSI code page, without indication what that code page is. How was this file created and how was this text added (was it typed? if so in what editor)?
For now, id call this “as designed”, Water side, its doing the bets it can, with the lack of information what encoding this file uses…
problem is, there isn’t one ANSI, theres dozens, depending on your locale ;).
there bets option is to change the file type to UTF-8 in the original editor, which (I assume) understands what ANSI code page the file is supposed to be in.
I do not agree that every source code file without BOM should be treated as UTF8, instead normally this is ANSI.
And normally there is only one ANSI for an installed locale. Why don’t you pick the locale settings ?
The problem arises when switching code editing from Water to another IDE (which is capable of treating ANSI files) everty time the (Ansi-) source file gets ‘corrupted’.
You can of course continue to treat all source files as UTF8 but after intitially reading ANSI files as such.