From: Junio C Hamano Date: Fri, 13 Jan 2006 19:51:17 GMT Subject: Re: git-commit: allow From: line to be entered in commit message Message-ID: <7vy81j6h56.fsf@assigned-by-dhcp.cox.net> In-Reply-To: sean writes: > So do you still agree with that, would you accept a patch? Or do you have > some fundamental reason to think that environment variables are a better > way to pass information in this case? Fundamental reason of not doing anything is (1) not to make unnecessary changes and (1) to avoidi decisions ;-). No objections to --from='John Doe ' by itself, other than "You could say GIT_AUTHOR_* before the command instead of --from after the command; as long as it is not done regularly as an interactive command, the difference does not matter". At that point, "making no unneeded changes" kicks in. While I do understand that it might be necessary to commit somebody else's patch occasionally, you have not convinced me that is not such a rare thing, so until then the change stays at lower priority. After I am convinced that it is not so rare and having an easier and more consistent way (something coming from environment and other things from command line is inconsistent) would generally be a good addition, I further need to think about these issues before taking such a patch, in the context of the "git commit" command as a whole: - Is --from the right word? Shouldn't it be --author? - Do we want author date? If not, why not? - What about committer information? If not, why not?