Unexpected encoding issue with _() macro and UTF-8 source in Unicode build (wxWidgets 3.3.1)

Hello everyone,

I'm encountering an unexpected encoding issue with string literals in a 
Unicode build of a wxWidgets application, and I’d appreciate any insights.

*Environment*

  *

    wxWidgets: 3.3.1

  *

    OS: Windows 11

  *

    Source files: UTF-8 with BOM

  *

    Compiler: Visual Studio 2022 with |/utf-8|

  *

    wxWidgets build: Unicode (UTF-16 internally)

*Problem*

Using the standard |_()| macro on a UTF-8 string literal containing 
Italian accented characters results in a corrupted |wxString|. It looks 
like the UTF-8 bytes are interpreted as single-byte characters.

*Examples*

Code 	Debug Output 	Result
|wxString msg1 = _("Prova di log con accenti: è ò à ì ù");| 	|L"Prova di 
log con accenti: è ò à ì ù"| 	Corrupted
|wxString msg2 = _(wxS("Prova di log con accenti: è ò à ì ù"));| 
|L"Prova di log con accenti: è ò à ì ù"| 	Correct

*Clarification*

I understand that |wxS()| fixes the issue because it turns the literal 
into a wide string (|L"..."|).
What confuses me is that I expected |_()| alone to be sufficient, since 
the source file is UTF-8 and MSVC is instructed to treat all narrow 
literals as UTF-8 via |/utf-8|.

*My question*

Is it expected that |_()| still interprets narrow string literals 
according to the system code page (CP1252) instead of UTF-8, even when 
the compiler is set to UTF-8?
In other words: in a Unicode build, is using |wxS()| (or |FromUTF8|) the 
intended and required way when dealing with UTF-8 source files?

Thanks a lot for any clarification.

Best regards,
Claudio Rossato

-- 
Please read https://www.wxwidgets.org/support/mlhowto.htm before posting.
--- 
You received this message because you are subscribed to the Google Groups "wx-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to [email protected].
To view this discussion visit https://groups.google.com/d/msgid/wx-users/6f9253b4-d2bd-458f-a773-49851412eeac%40gmail.com.