From: Shawn O. Pearce Date: Tue, 27 Nov 2007 01:08:01 GMT Subject: Re: Rebase/cherry-picking idea Message-ID: <20071127010801.GF14735@spearce.org> In-Reply-To: <7vy7ckgbpf.fsf@gitster.siamese.dyndns.org> Junio C Hamano wrote: > I would not object to renaming all of them to have the leading > underscore, though. That would make it clear that they are very > different from ordinary environment variables for the user to set > (e.g. GIT_INDEX_FILE, GIT_AUTHOR_NAME). Does any third party tool like > qgit already use GITHEAD_${objectname} and/or GIT_REFLOG_ACTION? git-gui apparently doesn't use either name right now. It avoids needing to use GIT_REFLOG_ACTION by invoking only plumbing, except in the case of git-merge, where it invokes git-merge and thus avoids the need to set GITHEAD_* to get conflict markers right when the recursive strategy gets used. I had started to replace git-merge in Tcl and have git-gui directly invoke merge-recursive but I haven't gotten around to really doing that. So I guess we could rename those two "internal" environment variables to use a leading _ to make them different from "user level" variables, but why change them now? I really don't see a compelling reason to break that part of the "API" between porcelain/plumbing. -- Shawn.