From: Junio C Hamano <gitster@pobox.com>
To: Jeff King <peff@peff.net>
Cc: git@vger.kernel.org
Subject: Re: [PATCH] add--interactive: allow diff colors without interactive colors
Date: Sat, 05 Jan 2008 01:28:25 -0800 [thread overview]
Message-ID: <7v3atcd3k6.fsf@gitster.siamese.dyndns.org> (raw)
In-Reply-To: <20080105085113.GA30598@coredump.intra.peff.net> (Jeff King's message of "Sat, 5 Jan 2008 03:51:13 -0500")
Jeff King <peff@peff.net> writes:
> But he doesn't have to care about that. He cares only that "color.diff"
> means "diffs are displayed in color." And "color.interactive" means
> "interactive menus are displayed in color."
> ...
>> If he told "git add -p" to be monochrome, he has every right to expect
>> the part to pick hunks to also stay monochrome. To people who know
>
> But he didn't. He said "git menus should be monochrome."
Who said anything about "interactive is limited to interactive
menus" anywhere? That is where we differ and what you do not
seem to be getting. I am talking about color.interactive that
controls the whole user experience of interacting with "add -i".
> Moreover, this doesn't allow "I always want color in diffs,
> but I don't want menu coloring" which is the very thing I have
> been trying to accomplish (but yes, I can do that by
> individually setting color.interactive.* to plain).
As you said earlier you may also be minority, but yes the color
pallette would help you do that.
> I fail to see how this is less confusing than just adding a separate
> interactive-diff knob, since you are asking them to individually set
> each color preference to plain.
What I am aiming at in longer term is to simplify things this way:
* Users are categorized broadly into two groups. The ones who
like colours and the ones who don't want colours at all.
color.git would control this (with backward compatibility
options per command such as color.diff and
color.interactive);
* Minorities who want to disable colours for particular parts
of the UI have enough knobs to tweak in the form of palettes.
By definition this needs to address "particular parts", so
"color.$command.$context" variables (e.g. color.diff.new,
color.interactive.new; if somebody really really wants to
have different settings between diff/show/log, that person
could add color.{show,log}.new as well) are needed if we want
to do this.
> E.g., given
> config options:
> ...
>> Admittedly, it's more work.
>
> Of course. ;) But I am willing to implement what I said above if you
> agree that it is sensible.
I think we share the ultimate goal of introducing higher level
knobs and our difference is just about minor details of how to
get there and what the intermediate levels look like.
I am trying to avoid introducing new intermediate level knobs
(e.g. color.log vs color.diff), as it is enough to disable or in
general change the way particular parts of the UI is coloured by
palette setting that specifically states which part of the UI is
tweaked (e.g. color.interactive.prompt).
next prev parent reply other threads:[~2008-01-05 9:29 UTC|newest]
Thread overview: 17+ messages / expand[flat|nested] mbox.gz Atom feed top
2008-01-04 8:35 [PATCH] add--interactive: allow diff colors without interactive colors Jeff King
2008-01-05 0:20 ` Junio C Hamano
2008-01-05 3:37 ` Jeff King
2008-01-05 8:01 ` Junio C Hamano
2008-01-05 8:51 ` Jeff King
2008-01-05 9:28 ` Junio C Hamano [this message]
2008-01-05 9:57 ` Jeff King
2008-01-05 10:58 ` [PATCH/resend] " Matthias Kestenholz
2008-01-05 11:11 ` Junio C Hamano
2008-01-05 14:10 ` Matthias Kestenholz
2008-01-05 14:11 ` [PATCH 1/4] Add infrastructure for a single color config variable Matthias Kestenholz
2008-01-05 14:11 ` [PATCH 2/4] git branch: Use color configuration infrastructure Matthias Kestenholz
2008-01-05 14:11 ` [PATCH 3/4] status and commit: " Matthias Kestenholz
2008-01-05 14:11 ` [PATCH 4/4] diff and log: " Matthias Kestenholz
2008-01-08 11:23 ` [PATCH 2/4] git branch: " Jeff King
2008-01-08 12:52 ` Matthias Kestenholz
2008-01-05 11:37 ` [PATCH/resend] add--interactive: allow diff colors without interactive colors Jakub Narebski
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=7v3atcd3k6.fsf@gitster.siamese.dyndns.org \
--to=gitster@pobox.com \
--cc=git@vger.kernel.org \
--cc=peff@peff.net \
/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).