From mboxrd@z Thu Jan 1 00:00:00 1970 From: Junio C Hamano Subject: [RFD] consider "git" wrapper semi-Porcelain Date: Sat, 25 Jun 2005 18:02:54 -0700 Message-ID: <7vy88yvsjl.fsf_-_@assigned-by-dhcp.cox.net> References: <7vaclgfynv.fsf@assigned-by-dhcp.cox.net> <7vvf439vdl.fsf@assigned-by-dhcp.cox.net> <7vaclf6tw7.fsf@assigned-by-dhcp.cox.net> <7vmzpe4x08.fsf_-_@assigned-by-dhcp.cox.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: git@vger.kernel.org X-From: git-owner@vger.kernel.org Sun Jun 26 02:56:57 2005 Return-path: Received: from vger.kernel.org ([12.107.209.244]) by ciao.gmane.org with esmtp (Exim 4.43) id 1DmLSB-00038W-QE for gcvg-git@gmane.org; Sun, 26 Jun 2005 02:56:56 +0200 Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S261368AbVFZBDW (ORCPT ); Sat, 25 Jun 2005 21:03:22 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S261385AbVFZBDW (ORCPT ); Sat, 25 Jun 2005 21:03:22 -0400 Received: from fed1rmmtao08.cox.net ([68.230.241.31]:31362 "EHLO fed1rmmtao08.cox.net") by vger.kernel.org with ESMTP id S261368AbVFZBC6 (ORCPT ); Sat, 25 Jun 2005 21:02:58 -0400 Received: from assigned-by-dhcp.cox.net ([68.4.60.172]) by fed1rmmtao08.cox.net (InterMail vM.6.01.04.00 201-2131-118-20041027) with ESMTP id <20050626010254.SEHI16890.fed1rmmtao08.cox.net@assigned-by-dhcp.cox.net>; Sat, 25 Jun 2005 21:02:54 -0400 To: Linus Torvalds In-Reply-To: Junio C. Hamano's message of "(unknown date)" User-Agent: Gnus/5.1007 (Gnus v5.10.7) Emacs/21.4 (gnu/linux) Sender: git-owner@vger.kernel.org Precedence: bulk X-Mailing-List: git@vger.kernel.org Currently "git" wrapper suggests both low-level git-* commands and git-*-script commands as alternatives. Earlier I stated that git-*-script commands, like it or not, form a perfectly good set of barebone Porcelain, as opposed to all the other commands that do not end with "script" are low-level Plumbing. After reviewing the current set of commands and scripts again, I still think this view holds true [*1*, *2*]. I would propose the following, preferably before we go 1.0: (1) Do not suggest low-level Plumbing commands to "git" wrapper users, and do not call anything but git-*-script from "git" wrapper. Once *-script Porcelain matures, hopefully everyday workflow would not require access to low-level Plumbing and can be done with git-*-script Porcelain. When people know the GIT way, they can also work it around things that git-*-script Porcelain does not support well by writing their own git-*-script in terms of Plumbing-ish git-* commands, and if it is good enough for public consumption, propose it for inclusion upstream. (2) Make an alias git-whatchanged-script that just "exec"s git-whatchanged, to make things consistent. (3) Rename some git-*-script that are misnamed in the sense that they are not useful standalone scripts but more of Plumbing that happen to be written in shell [*3*]. I do not particularly like this proposal, because this breaks existing scripts. However, if we go this route for consistency, I would further suggest renaming Porcelain-ish things git-*-cmd or something, call and suggest them from "git" wrapper, regardless of which language they are written in [*4*]. [Footnotes] *1* One exception is git-whatchanged. *2* That is the reason I did not name git-cherry "git-cherry-script". It is not useful standalone but meant to be useful in scripting environment. *3* git-apply-patch-script (my fault) and git-merge-one-file-script fall into this category. *4* Current naming and the proposal (1) above have a problem in that they muddy the definition of what is "script": the name "script" were originally used because they were written in shell, and all of the Porcelain-ish things happen to be "script". Writing Porcelain-ish things in shell should not be a requirement.