Git development
 help / color / mirror / Atom feed
From: Martin von Zweigbergk <martin.von.zweigbergk@gmail.com>
To: Junio C Hamano <gitster@pobox.com>
Cc: Martin von Zweigbergk <martin.von.zweigbergk@gmail.com>,
	Johannes Sixt <j.sixt@viscovery.net>,
	git@vger.kernel.org,
	Johannes Schindelin <Johannes.Schindelin@gmx.de>,
	Christian Couder <chriscool@tuxfamily.org>,
	Thomas Rast <trast@student.ethz.ch>
Subject: Re: [PATCH] rebase: define options in OPTIONS_SPEC
Date: Mon, 28 Feb 2011 20:59:26 -0500 (EST)	[thread overview]
Message-ID: <alpine.DEB.2.00.1102272055320.5025@debian> (raw)
In-Reply-To: <7v7hcl4uan.fsf@alter.siamese.dyndns.org>

On Sun, 27 Feb 2011, Junio C Hamano wrote:

> Martin von Zweigbergk <martin.von.zweigbergk@gmail.com> writes:
> 
> > What is the difference between the "options" and "actions" sections?
> 
> For historical reasons, "git rebase" does not take subcommands in the way
> many other commands do (e.g. "git bisect start", "git rerere remaining",
> "git stash save").  Some --option to the command are used for specifying
> the subcommand.  "git rebase --continue", if it were designed to be
> consistent with the other commands, would have been "git rebase continue"
> instead.  "Actions" part lists them.
> 
> This is somewhat unfortunate, but is understandable given that "rebase"
> didn't have its own options at all, and there was only one mode of
> operation, namely, to start the rebase process.  It allowed the user to be
> terse, i.e. instead off having to say "git rebase onto master", it was
> sufficient to say "git rebase master".
> 
> When "git rebase --continue" was introduced (before, you were supposed to
> say "git am --resolved", which was nonsense from the UI point of view),
> somebody should have noticed this as a problem and started a process to
> deprecate the "git rebase <branchname>" and to require explicit subcommand
> names, just like we did to "git stash [save]", but that didn't happen.
> 
> Perhaps this is something to consider at 2.0 boundary.
> 

Thanks for the explanation. It sounds like I should leave only
--continue, --skip and --abort in the "Actions" section.

I just noticed another problem with my patch. I had assumed that
specifying 'n,no-stat ...' on one line would give me a '-n' argument
to handle if e.g. '-n', '--no-stat' or '--no-sta' was passed. To my
surprise, when passing '--no-stat' on the command line, parse-options
gives me the long form back. If I pass '--no-sta', I get the curious

error: Ambiguous option: no-sta (could be --no-stat or --no-stat)

I figured these are related to the automatic support for negative
options (now I understand what you were talking about when you
mentioned that to someone a few days ago). So here comes an updated
patch where I have marked all of the options except for a few
(e.g. 'autosquash') with a '!' to tell parse-options that they have no
negative forms.


And thanks for fixing up my patch that moved the rebase scripts to
SCRIPT_LIB.


-- 8< --
Subject: [PATCH v2] rebase: define options in OPTIONS_SPEC

Interactive rebase used to have its own command line processing. Since
it used the 'git rev-parse --parseopt' functionality exposed through
git-sh-setup, it had some flexibility, like matching prefixes of long
options, that non-interactive rebase didn't. When interactive rebase's
command line processing was factored out into git-rebase.sh in cf432ca
(rebase: factor out command line option processing, 2011-02-06), this
flexibility was lost. Give back that flexibility to interactive and
non-interactive by defining its options in OPTIONS_SPEC.

Also improve the usage message to contain the --continue, --skip and
--abort sub commands.

Reported-by: Johannes Sixt <j6t@kdbg.org>
Signed-off-by: Martin von Zweigbergk <martin.von.zweigbergk@gmail.com>
---
Applies on top of mz/rebase.

Are USAGE and LONG_USAGE still needed?

I based the definitions on the now-lost OPTIONS_SPEC from interactive
rebase and on Documentation/git-rebase.txt.

 git-rebase.sh |  101 +++++++++++++++++++++++++++++++-------------------------
 1 files changed, 56 insertions(+), 45 deletions(-)

diff --git a/git-rebase.sh b/git-rebase.sh
index a040ab5..7a54bfc 100755
--- a/git-rebase.sh
+++ b/git-rebase.sh
@@ -28,7 +28,39 @@ Example:       git-rebase master~1 topic
 '
 
 SUBDIRECTORY_OK=Yes
-OPTIONS_SPEC=
+OPTIONS_KEEPDASHDASH=
+OPTIONS_SPEC="\
+git rebase [-i] [options] [--onto <newbase>] [<upstream>] [<branch>]
+git rebase [-i] [options] --onto <newbase> --root [<branch>]
+git-rebase [-i] --continue | --abort | --skip
+--
+ Available options are
+v,verbose!         display a diffstat of what changed upstream
+q,quiet!           be quiet. implies --no-stat
+onto=!             rebase onto given branch instead of upstream
+p,preserve-merges! try to recreate merges instead of ignoring them
+s,strategy=!       use the given merge strategy
+no-ff!             cherry-pick all commits, even if unchanged
+m,merge!           use merging strategies to rebase
+i,interactive!     let the user edit the list of commits to rebase
+f,force-rebase!    force rebase even if branch is up to date
+X,strategy-option=! pass the argument through to the merge strategy
+stat!              display a diffstat of what changed upstream
+n,no-stat!         do not show diffstat of what changed upstream
+verify             allow pre-rebase hook to run
+rerere-autoupdate  allow rerere to update index with resolved conflicts
+root!              rebase all reachable commits up to the root(s)
+autosquash         move commits that begin with squash!/fixup! under -i
+committer-date-is-author-date! passed to 'git am'
+ignore-date!       passed to 'git am'
+whitespace=!       passed to 'git apply'
+ignore-whitespace! passed to 'git apply'
+C=!                passed to 'git apply'
+ Actions:
+continue!          continue rebasing process
+abort!             abort rebasing process and restore original branch
+skip!              skip current patch and continue rebasing process
+"
 . git-sh-setup
 set_reflog_action rebase
 require_work_tree
@@ -175,7 +207,7 @@ do
 		ok_to_skip_pre_rebase=
 		;;
 	--continue|--skip|--abort)
-		test $total_argc -eq 1 || usage
+		test $total_argc -eq 2 || usage
 		action=${1##--}
 		;;
 	--onto)
@@ -183,10 +215,10 @@ do
 		onto="$2"
 		shift
 		;;
-	-i|--interactive)
+	-i)
 		interactive_rebase=explicit
 		;;
-	-p|--preserve-merges)
+	-p)
 		preserve_merges=t
 		test -z "$interactive_rebase" && interactive_rebase=implied
 		;;
@@ -196,62 +228,42 @@ do
 	--no-autosquash)
 		autosquash=
 		;;
-	-M|-m|--m|--me|--mer|--merg|--merge)
+	-M|-m)
 		do_merge=t
 		;;
-	-X*|--strategy-option*)
-		case "$#,$1" in
-		1,-X|1,--strategy-option)
-			usage ;;
-		*,-X|*,--strategy-option)
-			newopt="$2"
-			shift ;;
-		*,--strategy-option=*)
-			newopt="$(expr " $1" : ' --strategy-option=\(.*\)')" ;;
-		*,-X*)
-			newopt="$(expr " $1" : ' -X\(.*\)')" ;;
-		1,*)
-			usage ;;
-		esac
-		strategy_opts="$strategy_opts $(git rev-parse --sq-quote "--$newopt")"
+	-X)
+		shift
+		strategy_opts="$strategy_opts $(git rev-parse --sq-quote "--$1")"
 		do_merge=t
 		test -z "$strategy" && strategy=recursive
 		;;
-	-s=*|--s=*|--st=*|--str=*|--stra=*|--strat=*|--strate=*|\
-		--strateg=*|--strategy=*|\
-	-s|--s|--st|--str|--stra|--strat|--strate|--strateg|--strategy)
-		case "$#,$1" in
-		*,*=*)
-			strategy=`expr "z$1" : 'z-[^=]*=\(.*\)'` ;;
-		1,*)
-			usage ;;
-		*)
-			strategy="$2"
-			shift ;;
-		esac
+	-s)
+		shift
+		strategy="$1"
 		do_merge=t
 		;;
-	-n|--no-stat)
+	-n)
 		diffstat=
 		;;
 	--stat)
 		diffstat=t
 		;;
-	-v|--verbose)
+	-v)
 		verbose=t
 		diffstat=t
 		GIT_QUIET=
 		;;
-	-q|--quiet)
+	-q)
 		GIT_QUIET=t
 		git_am_opt="$git_am_opt -q"
 		verbose=
 		diffstat=
 		;;
-	--whitespace=*)
-		git_am_opt="$git_am_opt $1"
+	--whitespace)
+		shift
+		git_am_opt="$git_am_opt --whitespace=$1"
 		case "$1" in
-		--whitespace=fix|--whitespace=strip)
+		fix|strip)
 			force_rebase=t
 			;;
 		esac
@@ -263,22 +275,21 @@ do
 		git_am_opt="$git_am_opt $1"
 		force_rebase=t
 		;;
-	-C*)
-		git_am_opt="$git_am_opt $1"
+	-C)
+		shift
+		git_am_opt="$git_am_opt -C$1"
 		;;
 	--root)
 		rebase_root=t
 		;;
-	-f|--f|--fo|--for|--forc|--force|--force-r|--force-re|--force-reb|--force-reba|--force-rebas|--force-rebase|--no-ff)
+	-f|--no-ff)
 		force_rebase=t
 		;;
 	--rerere-autoupdate|--no-rerere-autoupdate)
 		allow_rerere_autoupdate="$1"
 		;;
-	-*)
-		usage
-		;;
-	*)
+	--)
+		shift
 		break
 		;;
 	esac
-- 
1.7.4.79.gcbe20

      reply	other threads:[~2011-03-01  1:59 UTC|newest]

Thread overview: 118+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2010-12-28  9:30 [PATCH 00/31] Refactor rebase Martin von Zweigbergk
2010-12-28  9:30 ` [PATCH 01/31] rebase: clearer names for directory variables Martin von Zweigbergk
2010-12-28 23:08   ` Junio C Hamano
2010-12-28 20:53     ` Martin von Zweigbergk
2010-12-28  9:30 ` [PATCH 02/31] rebase: refactor reading of state Martin von Zweigbergk
2010-12-28 23:08   ` Junio C Hamano
2010-12-29  8:09     ` Martin von Zweigbergk
2010-12-28  9:30 ` [PATCH 03/31] rebase: read state outside loop Martin von Zweigbergk
2010-12-28  9:30 ` [PATCH 04/31] rebase: remove unused rebase state 'prev_head' Martin von Zweigbergk
2010-12-28  9:30 ` [PATCH 05/31] rebase: improve detection of rebase in progress Martin von Zweigbergk
2010-12-28 23:08   ` Junio C Hamano
2010-12-28 20:35     ` Martin von Zweigbergk
2011-01-12 10:20       ` Martin von Zweigbergk
2010-12-28  9:30 ` [PATCH 06/31] rebase: act on command line outside parsing loop Martin von Zweigbergk
2010-12-28  9:30 ` [PATCH 07/31] rebase: stricter check of standalone sub command Martin von Zweigbergk
2010-12-28  9:30 ` [PATCH 08/31] rebase: align variable names Martin von Zweigbergk
2011-01-04 19:12   ` Thomas Rast
2011-01-05  2:25     ` Martin von Zweigbergk
2010-12-28  9:30 ` [PATCH 09/31] rebase: align variable content Martin von Zweigbergk
2010-12-28  9:30 ` [PATCH 10/31] rebase: factor out command line option processing Martin von Zweigbergk
2010-12-28  9:30 ` [PATCH 11/31] rebase -i: remove now unnecessary directory checks Martin von Zweigbergk
2010-12-28  9:30 ` [PATCH 12/31] rebase: reorder validation steps Martin von Zweigbergk
2010-12-28  9:30 ` [PATCH 13/31] rebase: factor out reference parsing Martin von Zweigbergk
2010-12-28  9:30 ` [PATCH 14/31] rebase: factor out clean work tree check Martin von Zweigbergk
2010-12-28  9:30 ` [PATCH 15/31] rebase: factor out call to pre-rebase hook Martin von Zweigbergk
2010-12-28  9:30 ` [PATCH 16/31] rebase -i: support --stat Martin von Zweigbergk
2010-12-28 17:59   ` Johannes Schindelin
2010-12-28 13:24     ` Martin von Zweigbergk
2010-12-28 23:36     ` Junio C Hamano
2010-12-28 23:44       ` Johannes Schindelin
2010-12-28  9:30 ` [PATCH 17/31] rebase: remove $branch as synonym for $orig_head Martin von Zweigbergk
2010-12-28  9:30 ` [PATCH 18/31] rebase: extract merge code to new source file Martin von Zweigbergk
2010-12-29 21:31   ` Johannes Sixt
2010-12-29 22:24     ` Martin von Zweigbergk
2010-12-31 12:23       ` Thomas Rast
2010-12-31 14:05         ` Martin von Zweigbergk
2010-12-28  9:30 ` [PATCH 19/31] rebase: extract am " Martin von Zweigbergk
2010-12-28  9:30 ` [PATCH 20/31] rebase: show consistent conflict resolution hint Martin von Zweigbergk
2010-12-28  9:30 ` [PATCH 21/31] rebase -i: align variable names Martin von Zweigbergk
2010-12-28  9:30 ` [PATCH 22/31] rebase: make -v a tiny bit more verbose Martin von Zweigbergk
2010-12-28  9:30 ` [PATCH 23/31] rebase: factor out sub command handling Martin von Zweigbergk
2010-12-28  9:30 ` [PATCH 24/31] rebase: extract code for writing basic state Martin von Zweigbergk
2011-01-04 19:19   ` Thomas Rast
2011-01-05  2:40     ` Martin von Zweigbergk
2010-12-28  9:30 ` [PATCH 25/31] rebase: remember verbose option Martin von Zweigbergk
2010-12-28  9:30 ` [PATCH 26/31] rebase: remember strategy and strategy options Martin von Zweigbergk
2011-01-04 19:27   ` Thomas Rast
2011-01-05  3:33     ` Martin von Zweigbergk
2010-12-28  9:30 ` [PATCH 27/31] rebase -m: remember allow_rerere_autoupdate option Martin von Zweigbergk
2010-12-28  9:30 ` [PATCH 28/31] rebase -m: don't print exit code 2 when merge fails Martin von Zweigbergk
2010-12-28  9:30 ` [PATCH 29/31] git-rebase--am: remove unnecessary --3way option Martin von Zweigbergk
2010-12-28  9:30 ` [PATCH 30/31] rebase -i: don't read unused variable preserve_merges Martin von Zweigbergk
2010-12-28  9:30 ` [PATCH 31/31] rebase -i: remove unnecessary state rebase-root Martin von Zweigbergk
2010-12-28 16:40   ` Thomas Rast
2010-12-29 22:31     ` Martin von Zweigbergk
2010-12-31  5:41       ` Christian Couder
2011-01-04 19:57 ` [PATCH 00/31] Refactor rebase Thomas Rast
2011-01-05  3:39   ` Martin von Zweigbergk
2011-02-06 18:43 ` [PATCH v2 00/31] refactor rebase Martin von Zweigbergk
2011-02-06 18:43   ` [PATCH v2 01/31] rebase: clearer names for directory variables Martin von Zweigbergk
2011-02-06 18:43   ` [PATCH v2 02/31] rebase: refactor reading of state Martin von Zweigbergk
2011-02-06 18:43   ` [PATCH v2 03/31] rebase: read state outside loop Martin von Zweigbergk
2011-02-06 18:43   ` [PATCH v2 04/31] rebase: remove unused rebase state 'prev_head' Martin von Zweigbergk
2011-02-06 18:43   ` [PATCH v2 05/31] rebase: improve detection of rebase in progress Martin von Zweigbergk
2011-02-06 18:43   ` [PATCH v2 06/31] rebase: act on command line outside parsing loop Martin von Zweigbergk
2011-02-06 18:43   ` [PATCH v2 07/31] rebase: stricter check of standalone sub command Martin von Zweigbergk
2011-07-01  3:55     ` Jonathan Nieder
2011-07-01 13:16       ` Martin von Zweigbergk
2011-07-01 22:29         ` Jonathan Nieder
2011-07-06  1:48           ` Martin von Zweigbergk
2011-02-06 18:43   ` [PATCH v2 08/31] rebase: align variable names Martin von Zweigbergk
2011-02-06 18:43   ` [PATCH v2 09/31] rebase: align variable content Martin von Zweigbergk
2011-02-06 18:43   ` [PATCH v2 10/31] rebase: factor out command line option processing Martin von Zweigbergk
2011-02-06 18:43   ` [PATCH v2 11/31] rebase -i: remove now unnecessary directory checks Martin von Zweigbergk
2011-02-06 18:43   ` [PATCH v2 12/31] rebase: reorder validation steps Martin von Zweigbergk
2011-02-06 18:43   ` [PATCH v2 13/31] rebase: factor out reference parsing Martin von Zweigbergk
2011-02-06 18:43   ` [PATCH v2 14/31] rebase: factor out clean work tree check Martin von Zweigbergk
2011-02-06 18:43   ` [PATCH v2 15/31] rebase: factor out call to pre-rebase hook Martin von Zweigbergk
2011-02-06 18:43   ` [PATCH v2 16/31] rebase -i: support --stat Martin von Zweigbergk
2011-02-06 18:43   ` [PATCH v2 17/31] rebase: remove $branch as synonym for $orig_head Martin von Zweigbergk
2011-02-06 18:43   ` [PATCH v2 18/31] rebase: extract merge code to new source file Martin von Zweigbergk
2011-02-14  8:02     ` Johannes Sixt
2011-02-14 13:56       ` Martin von Zweigbergk
2011-02-24  3:27       ` Martin von Zweigbergk
2011-02-24  8:07         ` Jeff King
2011-02-24  8:09           ` Jeff King
2011-02-25  3:32             ` Martin von Zweigbergk
2011-02-25  9:02               ` Jeff King
2011-02-25 20:24                 ` Junio C Hamano
2011-02-25 20:27             ` Jakub Narebski
2011-03-01 22:04               ` Jeff King
2011-03-01 22:43                 ` Jakub Narebski
2011-02-06 18:43   ` [PATCH v2 19/31] rebase: extract am " Martin von Zweigbergk
2011-02-06 18:43   ` [PATCH v2 20/31] rebase: show consistent conflict resolution hint Martin von Zweigbergk
2011-02-06 18:43   ` [PATCH v2 21/31] rebase -i: align variable names Martin von Zweigbergk
2011-02-06 18:43   ` [PATCH v2 22/31] rebase: make -v a tiny bit more verbose Martin von Zweigbergk
2011-02-06 18:43   ` [PATCH v2 23/31] rebase: factor out sub command handling Martin von Zweigbergk
2011-02-06 18:43   ` [PATCH v2 24/31] rebase: extract code for writing basic state Martin von Zweigbergk
2011-02-06 18:43   ` [PATCH v2 25/31] rebase: remember verbose option Martin von Zweigbergk
2011-02-06 18:43   ` [PATCH v2 26/31] rebase: remember strategy and strategy options Martin von Zweigbergk
2011-02-06 18:43   ` [PATCH v2 27/31] rebase -m: remember allow_rerere_autoupdate option Martin von Zweigbergk
2011-02-06 18:43   ` [PATCH v2 28/31] rebase -m: don't print exit code 2 when merge fails Martin von Zweigbergk
2011-02-06 18:43   ` [PATCH v2 29/31] git-rebase--am: remove unnecessary --3way option Martin von Zweigbergk
2011-02-06 18:43   ` [PATCH v2 30/31] rebase -i: don't read unused variable preserve_merges Martin von Zweigbergk
2011-02-06 18:44   ` [PATCH v2 31/31] rebase -i: remove unnecessary state rebase-root Martin von Zweigbergk
2011-02-10 22:44   ` [PATCH v2 00/31] refactor rebase Junio C Hamano
2011-02-12  0:55     ` Martin von Zweigbergk
2011-02-14  1:54       ` Martin von Zweigbergk
2011-02-14  3:15         ` Martin von Zweigbergk
2011-02-15  0:36         ` Junio C Hamano
2011-02-22 13:58         ` Martin von Zweigbergk
2011-02-22 19:21           ` Junio C Hamano
2011-02-23 11:26             ` Martin von Zweigbergk
2011-02-16 14:52   ` Johannes Sixt
2011-02-17  3:41     ` Martin von Zweigbergk
2011-02-24  3:07     ` [PATCH] rebase: define options in OPTIONS_SPEC Martin von Zweigbergk
2011-02-27 10:59       ` Junio C Hamano
2011-03-01  1:59         ` Martin von Zweigbergk [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=alpine.DEB.2.00.1102272055320.5025@debian \
    --to=martin.von.zweigbergk@gmail.com \
    --cc=Johannes.Schindelin@gmx.de \
    --cc=chriscool@tuxfamily.org \
    --cc=git@vger.kernel.org \
    --cc=gitster@pobox.com \
    --cc=j.sixt@viscovery.net \
    --cc=trast@student.ethz.ch \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox