Plugin 'AutoCodepage v1.2.7' for 'notepad++ v8.7.4' not working?

I don’t know what you want to use a completely outdated plugin for.

Do you have a tip for me where the problem is?

The problem is not NP++ or any plugin, but your understanding of character encoding…

Under up to and including Windows 7, there are various disagreements regarding the coding of Batch. (many were founded in bad “yet obsolete” instructions.) …and to this day I read instructions from people who have no idea which use any 12xx magic to display characters readable. You only have to work with UTF8.

Generally, the output of the Windowsconsole (cmd.exe) “yet” works with the Codepage 850. (to be compatible with old garbage)

I tried to track your scenario, the status line indicates ANSI by default (as expected). The code page is completely sausage for the first time. ANSI merely states : 8 bits per character. Since there is no explicit labelling in BOM-less files with regard to coding can only be guessed by an editor/application on the basis of the content which coding is present. Some strings make more sense among the different encodings than others. Generally, there are no recognizable difference between ANSI and UTF-8 for files that contain only characters with an ASCII code <127!

If you use the “dir command”, there are already significant differences between screen output and file redirection.

The Windows file system generally no longer works with local 12xx codepages, but with UTF-8 names! Thus, all file names of “dir” UTF-8 are also supplied coded to Console (cmd).

What makes the console depends on the preset coding and the output “medium”.

• For screen output (from Win 10) Relaunching automatically (depending on the font) output by the appropriate character.
• In case of a file redirection, however, the font is not known, so for umlaute multibyte values are written into the file. (which gives the known “mutilated” sign). Such significant byte sequences can be recognized by editors such as NP++ &Co and thus declare the file as UTF-8, even if they were stored as “ANSI”. But there is no patent prescription. The editors can only “estimate”.

If you want to explicitly specify which encoding should be done, use the chcp command

@echo off
chcp 850 >nul
dir >"Datei.txt"

as I said, this does not change the interpretation of any UTF characters read in.

I can only recommend working with UTF-8 on modern systems.

Batch (.cmd) files basically save as UTF-8 file without BOM and set the appropriate codepage 65001 on the start of the batch.

@echo off
chcp 65001 >nul
>"демонстрация.txt" echo Маша и Медведь

..that makes the least trouble on modern systems.

If you want to work quite modern, forget Batch and use Powershell and explicitly give the desired file operations Encoding on.