From: Junio C Hamano Date: Fri, 27 May 2022 15:45:36 GMT Subject: Re: [PATCH 1/3] rebase.c: state preserve-merges has been removed Message-ID: In-Reply-To: <9455d0af-87f2-f331-a440-3d3feb743610@iee.email> Philip Oakley writes: > On 26/05/2022 21:33, Junio C Hamano wrote: >>>> So there's no point in changing this string, nor to have translators >>>> focus on it, it'll never be used. >>>> >>>> > The translation change would need to be a separate patch, no? That > would make it easy to drop if not wanted. I think you are responding to what Ævar said, but the string this patch is modifying is already inside N_(), and modifying a string that is already marked for translation in any way (other than removing the N_() or _() around it) incurs the cost to translate the updated string already, with or without any separate patch. If we are adding a new die() call that uses a new message, we should mark the message for translation from the beginning. The messages produced by die/warning/error are meant to be read by human users, so unless there is some very strong reason not to, they should be marked for translation.