Re: Consistent terminology: cached/staged/index
From: Michael J Gruber <hidden>
Date: 2016-06-15 22:50:33
Felipe Contreras venit, vidit, dixit 14.02.2011 15:17:
On Mon, Feb 14, 2011 at 3:43 PM, Michael J Gruber [off-list ref] wrote:quoted
Nguyen Thai Ngoc Duy venit, vidit, dixit 14.02.2011 14:14:quoted
On Mon, Feb 14, 2011 at 5:42 PM, Michael J Gruber [off-list ref] wrote:quoted
Full disclaimer: I have an alias "staged" for "diff --cached" myself...Be careful with your fingers. There's a command named "git stage".I know. Can we remove it as part of 1.8.0? It's our only builtin alias.I have proposed before to extend 'git stage', so you can do 'git stage diff', or if you alias 'git stage' to 'git s', just 'git s diff'. This would not conflict with the old behavior of 'git stage $file'. case "$1" in add) shift git add $@ ;; rm) shift git rm --cached $@ ;; diff) shift git diff --cached $@ ;; import) shift git ls-files --modified --others --exclude-standard -z $@ | \ git update-index --add --remove -z --stdin ;; ls) shift git ls-files --stage $@ ;; *) git add $@ ;; esac Cheers.
In principle I like this a lot: a set of commands operating on/with the stage/index/cache consistently. It think it's similar in (good) spirit to our earlier attempts at INDEX and WORKTREE pseudo-revs, trying to give that somewhat nebulous (for noobs) index a more concrete "appearance", not hidden away in options (--index, --cached) and defaults (diff against index by default). In our case, however, I think the design principle deviates from our common form: git foo bar usually means "do foo" to "bar", as most of our common commands are verbs (being applied to the object "bar"). When it comes to subcommands we do have inconsistencies already (double-dashed vs. undashed, e.g.), but I'd prefer fewer ;) Michael