git@vger.kernel.org mailing list mirror (one of many)
 help / color / mirror / code / Atom feed
From: Denton Liu <liu.denton@gmail.com>
To: Junio C Hamano <gitster@pobox.com>
Cc: Git Mailing List <git@vger.kernel.org>, Beat Bolli <dev+git@drbeat.li>
Subject: Re: [PATCH v3 2/2] format-patch: teach format.notes config option
Date: Tue, 14 May 2019 13:01:07 -0400	[thread overview]
Message-ID: <20190514170107.GA4151@archbookpro.localdomain> (raw)
In-Reply-To: <xmqqimufdqey.fsf@gitster-ct.c.googlers.com>

Hi Junio,

On Mon, May 13, 2019 at 11:44:21AM +0900, Junio C Hamano wrote:
> Denton Liu <liu.denton@gmail.com> writes:
> 
> > @@ -864,6 +866,20 @@ static int git_format_config(const char *var, const char *value, void *cb)
> >  			from = NULL;
> >  		return 0;
> >  	}
> > +	if (!strcmp(var, "format.notes")) {
> > +		struct strbuf buf = STRBUF_INIT;
> > +
> > +		rev->show_notes = 1;
> > +		if (!strcmp(value, "standard")) {
> > +			rev->notes_opt.use_default_notes = 1;
> > +		} else {
> > +			strbuf_addstr(&buf, value);
> > +			expand_notes_ref(&buf);
> > +			string_list_append(&rev->notes_opt.extra_notes_refs,
> > +					strbuf_detach(&buf, NULL));
> > +		}
> > +		return 0;
> > +	}
> 
> Unlike the command line option parser, this does not seem to touch
> rev->show_notes_given at all.  Intended?  I am wondering how well
> this implementation meshes with what 66b2ed09 ("Fix "log" family not
> to be too agressive about showing notes", 2010-01-20) wanted to do,
> which 894a9d33 ("Support showing notes from more than one notes
> tree", 2010-03-12) later extended.

This was intended but I'm not 100% sure that it's correct.

From what I could gleam from reading the code, `show_notes_given` is
only used by the `cmd_log_init` function, which is not called by
format-patch. As a result, I opted to not set that flag since it's not
really "given" in the sense that a user didn't explicitly pass in a
command-line option indicating they wanted notes.

> 
> > +test_expect_success 'format-patch notes output control' '
> > +	git notes add -m "notes config message" HEAD &&
> > +	test_when_finished git notes remove HEAD &&
> > +
> > +	git format-patch -1 --stdout >out &&
> > +	! grep "notes config message" out &&
> > +	git format-patch -1 --stdout --notes >out &&
> > +	grep "notes config message" out &&
> > +	git format-patch -1 --stdout --no-notes >out &&
> > +	! grep "notes config message" out &&
> > +	git format-patch -1 --stdout --notes --no-notes >out &&
> > +	! grep "notes config message" out &&
> > +	git format-patch -1 --stdout --no-notes --notes >out &&
> > +	grep "notes config message" out &&
> > +
> > +	test_config format.notes standard &&
> 
> I think we tend to spell these things "default".
> 
> Alternatively, the format.notes configuration can be "bool or text",
> and make the variable set to 'true' mean "show notes, using the
> default ref".

I think I'l go with this approach.

> 
> > +	git format-patch -1 --stdout >out &&
> > +	grep "notes config message" out &&
> > +	git format-patch -1 --stdout --notes >out &&
> > +	grep "notes config message" out &&
> > +	git format-patch -1 --stdout --no-notes >out &&
> > +	! grep "notes config message" out &&
> > +	git format-patch -1 --stdout --notes --no-notes >out &&
> > +	! grep "notes config message" out &&
> > +	git format-patch -1 --stdout --no-notes --notes >out &&
> > +	grep "notes config message" out
> > +'
> 
> OK.
> 
> > +test_expect_success 'format-patch with multiple notes refs' '
> > +	git notes --ref note1 add -m "this is note 1" HEAD &&
> > +	test_when_finished git notes --ref note1 remove HEAD &&
> > +	git notes --ref note2 add -m "this is note 2" HEAD &&
> > +	test_when_finished git notes --ref note2 remove HEAD &&
> > + ...
> > +	git format-patch -1 --stdout --notes=note1 --notes=note2 >out &&
> > +	grep "this is note 1" out &&
> > +	grep "this is note 2" out &&
> 
> Do we promise the order in which these two lines appear in the output?

According to the code, the order is stable. However, I just read through
the documentation and I realised that the ablility to provide multiple
notes refs is undocumented.

In a future reroll, I'll document the fact that --notes can be provided
multiple times.

> 
> > +	test_config format.notes note1 &&
> > +	git format-patch -1 --stdout >out &&
> > +	grep "this is note 1" out &&
> > +	! grep "this is note 2" out &&
> > +	git format-patch -1 --stdout --no-notes >out &&
> > +	! grep "this is note 1" out &&
> > +	! grep "this is note 2" out &&
> > +	git format-patch -1 --stdout --notes=note2 >out &&
> > +	grep "this is note 1" out &&
> > +	grep "this is note 2" out &&
> 
> So format.notes say note1 but the command line explicitly asks it
> wants note from note2, but the command still gives from note1
> anyway.
> 
> > +	git format-patch -1 --stdout --no-notes --notes=note2 >out &&
> > +	! grep "this is note 1" out &&
> > +	grep "this is note 2" out &&
> 
> And there is a way to work it around, i.e. clear everything
> configured with --no-notes and then name the one you want from the
> command line.
> 
> I am not sure if the above is consistent with how our options and
> configurations interact in general.  Shouldn't the --notes=note2
> alone in the earlier example cancel format.notes=note1 configured?

I borrowed this behaviour from how format.to behaves. In format-patch,
`--to` gives a recipient that is used _in addition_ to any format.to
variables. `--no-to` can override this. I made format.notes behave
similarly.

> 
> > +	git config --add format.notes note2 &&
> > +	git format-patch -1 --stdout >out &&
> > +	grep "this is note 1" out &&
> > +	grep "this is note 2" out &&
> > +	git format-patch -1 --stdout --no-notes >out &&
> > +	! grep "this is note 1" out &&
> > +	! grep "this is note 2" out
> > +'
> > +
> >  echo "fatal: --name-only does not make sense" > expect.name-only
> >  echo "fatal: --name-status does not make sense" > expect.name-status
> >  echo "fatal: --check does not make sense" > expect.check

      reply	other threads:[~2019-05-14 17:01 UTC|newest]

Thread overview: 15+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-04-27 19:25 [PATCH 0/2] format-patch: teach format.notes config option Denton Liu
2019-04-27 19:25 ` [PATCH 1/2] git-format-patch.txt: document --no-notes option Denton Liu
2019-04-27 19:25 ` [PATCH 2/2] format-patch: teach format.notes config option Denton Liu
2019-05-07  4:43   ` Junio C Hamano
2019-05-08 15:02 ` [PATCH v2 0/2] " Denton Liu
2019-05-08 15:02   ` [PATCH v2 1/2] git-format-patch.txt: document --no-notes option Denton Liu
2019-05-08 15:02   ` [PATCH v2 2/2] format-patch: teach format.notes config option Denton Liu
2019-05-08 17:18     ` Beat Bolli
2019-05-08 17:31       ` Denton Liu
2019-05-08 17:52         ` Beat Bolli
2019-05-10 18:37   ` [PATCH v3 0/2] " Denton Liu
2019-05-10 18:37     ` [PATCH v3 1/2] git-format-patch.txt: document --no-notes option Denton Liu
2019-05-10 18:37     ` [PATCH v3 2/2] format-patch: teach format.notes config option Denton Liu
2019-05-13  2:44       ` Junio C Hamano
2019-05-14 17:01         ` Denton Liu [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

  List information: http://vger.kernel.org/majordomo-info.html

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

  git send-email \
    --in-reply-to=20190514170107.GA4151@archbookpro.localdomain \
    --to=liu.denton@gmail.com \
    --cc=dev+git@drbeat.li \
    --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.
Code repositories for project(s) associated with this public inbox

	https://80x24.org/mirrors/git.git

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 read-only IMAP folder(s) and NNTP newsgroup(s).