Jon Forrest <nobozo@xxxxxxxxx> writes: > On 8/1/2011 2:58 AM, Sverre Rabbelier wrote: >> On Mon, Aug 1, 2011 at 06:46, Jon Forrest<nobozo@xxxxxxxxx> wrote: >>> fprintf(options->file, "%s", line_prefix); >>> fprintf(options->file, >>> - " %d files changed, %d insertions(+), %d deletions(-)\n", >>> - total_files, adds, dels); >>> + " %d file%s changed, %d insertion%s(+), %d deletion%s(-)\n", >>> + total_files, total_files == 1 ? "" : "s", adds, adds == 1 ? >>> "" : "s", dels, >>> + dels == 1 ? "" : "s"); >>> } >> >> Also, this is rather detrimental to the i18n effort methinks? Besides, as it was already said, this is an API. > If the goal if the i18n effort is also to produce grammatically > correct output in all the supported languages then the > tests that my patch would break would have to be rewritten > anyway. That's not it. The problem is that above code assumes that plural form can be formed by adding suffix, and it assumes that is only one plural form... both assumptions does not hold for non-English. C.f. "Additional functions for plural forms" chapter in gettext info page: http://www.gnu.org/s/hello/manual/gettext/Plural-forms.html -- Jakub Narębski -- To unsubscribe from this list: send the line "unsubscribe git" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html