git.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Jonathan Nieder <jrnieder@gmail.com>
To: Ramkumar Ramachandra <artagnon@gmail.com>
Cc: Git List <git@vger.kernel.org>,
	Junio C Hamano <gitster@pobox.com>,
	Christian Couder <chriscool@tuxfamily.org>,
	Daniel Barkalow <barkalow@iabervon.org>
Subject: Re: [PATCH 09/13] revert: Catch incompatible command-line options early
Date: Sat, 2 Jul 2011 05:04:36 -0500	[thread overview]
Message-ID: <20110702100436.GB19090@elie> (raw)
In-Reply-To: <20110702095334.GA19090@elie>

Jonathan Nieder wrote:
> Ramkumar Ramachandra wrote:

>>     Some incompatible command-line options are caught in pick_commits
>>     after parse_args has parsed the options and populated the options
>>     structure.  Change this so that parse_args only parses valid
>>     command-line options
[...]
> Alas, I still don't get it.  How can I (the fearful reader) demonstrate
> the problem that this fixes?  Is it a big problem?  How does the patch
> fix it?  Does the patch have downsides, or is it mostly risk-free?

To be clearer, here's a quick example in some hypothetical alternate
world:

	cherry-pick/revert: do not create a nonsense "struct opts" when given nonsense argv

	The --foo and --bar options make no sense together, but if I
	run "git cherry-pick --foo --bar" then parse_cherry_pick_option
	will return a nonsense struct with both "fooing" and "barring"
	set to true, which makes no sense.  Currently it's not a
	problem since the code that cares is protected by a check:

		if (opts.fooing && opts.barring)
			die("cannot foo and bar at the same time");

	But that is very fragile --- the frob() code-path relies on
	opts.fooing and opts.barring not both being true without such
	a check and it's only a coincidence that futz() is called
	first and makes that check, protecting it.

	We can make sure such a nonsensical options struct is never
	created by checking right away that --foo and --bar are not
	passed together at option-parsing time.  Meanwhile, make sure
	regressions in maintaining this invariant are caught in the
	future by adding an assertion "assert(!opts->fooing ||
	!opts->barring)" to both frob() and futz().

	The discussion above also applies to --bar and --qux.  Fix
	that, too.

  reply	other threads:[~2011-07-02 10:04 UTC|newest]

Thread overview: 47+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2011-06-21 13:04 [PATCH 00/13] Sequencer with continuation features Ramkumar Ramachandra
2011-06-21 13:04 ` [PATCH 01/13] advice: Introduce error_resolve_conflict Ramkumar Ramachandra
2011-06-21 15:55   ` Jonathan Nieder
2011-06-21 18:43     ` Junio C Hamano
2011-07-02  9:44     ` Ramkumar Ramachandra
2011-07-02 10:09       ` Jonathan Nieder
2011-06-21 13:04 ` [PATCH 02/13] revert: Factor out add_message_to_msg function Ramkumar Ramachandra
2011-06-21 15:58   ` Jonathan Nieder
2011-06-21 19:01     ` Junio C Hamano
2011-06-21 13:04 ` [PATCH 03/13] revert: Don't check lone argument in get_encoding Ramkumar Ramachandra
2011-06-21 16:03   ` Jonathan Nieder
2011-06-21 13:04 ` [PATCH 04/13] revert: Propogate errors upwards from do_pick_commit Ramkumar Ramachandra
2011-06-21 16:22   ` Jonathan Nieder
2011-06-21 19:19     ` Junio C Hamano
2011-06-21 19:32       ` Jonathan Nieder
2011-06-21 20:18         ` Junio C Hamano
2011-07-02 10:31           ` Ramkumar Ramachandra
2011-06-21 13:04 ` [PATCH 05/13] revert: Eliminate global "commit" variable Ramkumar Ramachandra
2011-06-21 16:52   ` Jonathan Nieder
2011-06-21 19:23   ` Junio C Hamano
2011-06-21 13:04 ` [PATCH 06/13] revert: Rename no_replay to record_origin Ramkumar Ramachandra
2011-06-21 13:04 ` [PATCH 07/13] revert: Introduce struct to keep command-line options Ramkumar Ramachandra
2011-06-21 16:58   ` Jonathan Nieder
2011-06-21 19:28     ` Junio C Hamano
2011-06-21 13:04 ` [PATCH 08/13] revert: Separate cmdline parsing from functional code Ramkumar Ramachandra
2011-06-21 17:00   ` Jonathan Nieder
2011-06-21 13:04 ` [PATCH 09/13] revert: Catch incompatible command-line options early Ramkumar Ramachandra
2011-06-21 17:04   ` Jonathan Nieder
2011-07-02  9:47     ` Ramkumar Ramachandra
2011-07-02  9:53       ` Jonathan Nieder
2011-07-02 10:04         ` Jonathan Nieder [this message]
2011-07-02 11:19           ` Ramkumar Ramachandra
2011-07-02 11:30             ` Jonathan Nieder
2011-07-02 20:02               ` Jonathan Nieder
2011-06-21 13:04 ` [PATCH 10/13] revert: Persist data for continuation Ramkumar Ramachandra
2011-06-21 17:11   ` Jonathan Nieder
2011-06-21 19:49   ` Junio C Hamano
2011-06-21 13:04 ` [PATCH 11/13] revert: Introduce a layer of indirection over pick_commits Ramkumar Ramachandra
2011-06-21 19:59   ` Junio C Hamano
2011-06-21 13:04 ` [PATCH 12/13] revert: Introduce skip-all to cleanup sequencer data Ramkumar Ramachandra
2011-06-21 20:02   ` Junio C Hamano
2011-07-02  6:24     ` Ramkumar Ramachandra
2011-07-04  4:59       ` Miles Bader
2011-07-05 10:47         ` Ramkumar Ramachandra
2011-06-21 13:04 ` [RFC PATCH 13/13] revert: Introduce --continue to continue the operation Ramkumar Ramachandra
2011-06-21 17:19   ` Jonathan Nieder
2011-06-21 15:48 ` [PATCH 00/13] Sequencer with continuation features Jonathan Nieder

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=20110702100436.GB19090@elie \
    --to=jrnieder@gmail.com \
    --cc=artagnon@gmail.com \
    --cc=barkalow@iabervon.org \
    --cc=chriscool@tuxfamily.org \
    --cc=git@vger.kernel.org \
    --cc=gitster@pobox.com \
    /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;
as well as URLs for NNTP newsgroup(s).