Prevent "Reused translation" check when Singular and Plural the same in the translated language

[buhtz]

The example is about the English terms Error and Errors. In German they are translated both as Fehler. There is no difference between Singular and Plural for this term in German.

I do understand why Weblate do report a "Reused translation" check error at this point.

Is there a way to prevent such false positive checks?

I do know about GNU gettext plural terms feature. But I'm not sure if this is the right answer. Are there other ways?

In Python it would result in a lot of lines like this:

ngettext('Error', 'Errors', 1)  # Singular
ngettext('Error', 'Errors', 2)  # Plural

See similar discussion at #10370

[buhtz]

I would be glad about a solution different from ngettext(). As you can see in my PR and this discussion on reddit it is not easy to decide about the correct integer used for the plural form in the ngettext() function.

Languages like Arabic or Polish do have multiple plural forms depending on the number/count used.

[nijel]

Yes, that's why you should use ngettext() and Gettext logic will find the correct plural. There is no need for separate plural string then.

[buhtz]

But it is a plural without a specific count. That is the point here.