Romanian translation broken after codepage change

[tomutal]

I had Total Commander 11.02 installed and working fine with Romanian language on a fresh Windows 10 install.
Unfortunately another critical software package was crashing for me and after working with the developers we narrowed the bug to their handling of the Romanian characters I was using in some files. Lons story short, that software package crashes if the code page is 852.

So I turned on this beta setting Image: https://imgur.com/a/xfaH3YE and that software package's problem is avoided.

Now my system's code page as returned by chcp is 65001, which is the same code page specified in TC's Romanian language file, yet the rendering of the Romanian strings in TC is broken

Image: https://imgur.com/a/tyk3RxG

I tried saving the language files again, after opening them in VS Code, so they should definitely be in UTF-8 format.

Any other ideas how to fix this?

[beb]

2tomutal
I have tried to reproduce your case.
At first, I switched on the said system setting [Beta: Use Unicode UTF-8 for...]
Upon restart, I meet the following results:
- with an original .mnu file for the Romanian that comes in UTF-8 (65001) encoding - Total Commander fails to display it properly.
- with an original .mnu file for my local language that comes in ANSI (1251) encoding - Total Commander displays it properly.
Then I try to change the .mnu file encoding:
- if I set ('saving it as' in a proper text editor...) my local language .mnu file as UTF-8 (65001) (with/or without BOM) - Total Commander fails to display it properly
- if I set ('saving it as' in a proper text editor...) my local language .mnu file as UTF-16 LE BOM (1200) - Total Commander fails to display it properly
The only suitable option I found for that scenario:
- if its (the .mnu file's) encoding is set back to the language-specific ANSI code page Total Commander renders .mnu properly.
(Again, in my case, my language-specific ANSI code page is Windows-1251.)

If to draw the parallels, in the case of the Romanian with [Beta: Use Unicode UTF-8 for...] turned on, it looks as though the menu file should go in proper ANSI encoding.

As I can see, for the Romanian it could be Windows-1250:
https://en.wikipedia.org/wiki/Windows-1250
Though, I cannot be sure which one is correct, e.g. I can also see this one:
https://en.wikipedia.org/wiki/ISO/IEC_8859-2

So, finally, please, check how it looks if you change your .mnu file encoding to the language-specific ANSI code page (whichever is actually bound to the Romanian).

Looks like this is the Total Commander design flaw.
I think it (Total Commander) should:
- render UTF-8 (65001) the .mnu files (and other such files) properly independently on the 'Beta: Use Unicode UTF-8 for...' settings without the need to convert the file from UTF-8 to ANSI.
- support the UTF-16 LE BOM (1200) encoding for the .mnu files (and other such files) as well as it does for the .ini files.

It's the 21st century on the calendar, isn't it?

Edit.
Interesting.
Out of curiosity, I checked all the .mnu files inside the TotalCommander installer package (currently, 19).
It appears the Romanian .mnu is the only file which encoding is being detected as UTF-8.
All the other ones are detected as ANSI.

[ghisler(Author)]

Confirmed, I will fix it in Total Commander 11.03 RC5.

[ghisler(Author)]

This should be fixed in Total Commander 11.03 RC5, please test it.

[beb]

The fix is effective for the Romanian language.
Its files encoded with UTF-8 (65001) now work fine in Total Commander 11.03 RC5 when the Windows system locale is also UTF-8.

[tomutal]

I can confirm that the issue is fixed with RC5. Thank you!

[ghisler(Author)]

Great, thanks to both of you for checking it!