>> This is where the confusion comes from. The Russian translator did put >> "y" as "Yes key" and "n" as "No key" instead of Д and н.
> That is a problem, but it won't fix this particular case, since the >prompt in question doesn't use a localized technique to test whether the >user entered Y or N. This is complicated by the fact that this is not >solely a yes/no prompt; otherwise I could use rpmatch(3) to localize it. >There are a good half-dozen possible characters the user can enter, and >odds are that some of them are abbreviations for "yes" or "no" in some >language. > I'm not sure what the best approach is; maybe yet another bunch of >tagged localized strings that control recognized responses to prompts? > > Daniel You mean, the best way is to keep [Y,n,q] and msgid "yes_key" msgstr "y" in the translation? -- Regards, Yuri Kozlov
