From: Phillip Wood <phillip.wood123@gmail.com>
To: Leon Michalak via GitGitGadget <gitgitgadget@gmail.com>,
git@vger.kernel.org
Cc: Leon Michalak <leonmichalak6@gmail.com>,
Eric Sunshine <sunshine@sunshineco.com>
Subject: Re: [PATCH 1/3] add-patch: respect diff.context configuration
Date: Tue, 6 May 2025 09:55:52 +0100 [thread overview]
Message-ID: <c5591cbb-7eb9-4b69-85a1-ae1678983b14@gmail.com> (raw)
In-Reply-To: <8b91eef8120b8f92db953ec983fddce8a442abcc.1746436719.git.gitgitgadget@gmail.com>
Hi Leon
Thanks for working on this. I've left a few comments below but for your
first submission this is very good.
On 05/05/2025 10:18, Leon Michalak via GitGitGadget wrote:
> From: Leon Michalak <leonmichalak6@gmail.com>
>
> This aims to teach relevant builtins (that take in `--patch`) to respect
> the user's diff.context and diff.interHunkContext file configurations.
This is a good summary of the intent of the patch. I'd maybe drop "This
aims" and instead say that the various commands do not respect the
config and this patch fixes that.
> Since these are both UI options and `--patch` is designed for the end user,
> I believe this was previously just an inconsistency, which this patch hopes
> to address.
>
> Signed-off-by: Leon Michalak <leonmichalak6@gmail.com>
> ---
> add-interactive.c | 16 ++++++++++----
> add-patch.c | 6 ++++++
> t/t4055-diff-context.sh | 48 ++++++++++++++++++++++++++++++++++++++++-
> 3 files changed, 65 insertions(+), 5 deletions(-)
>
> diff --git a/add-interactive.c b/add-interactive.c
> index 97ff35b6f12a..ad12dc416598 100644
> --- a/add-interactive.c
> +++ b/add-interactive.c
> @@ -41,6 +41,8 @@ void init_add_i_state(struct add_i_state *s, struct repository *r)
> const char *value;
>
> s->r = r;
> + s->context = -1;
> + s->interhunkcontext = -1;
>
> if (repo_config_get_value(r, "color.interactive", &value))
> s->use_color = -1;
> @@ -78,6 +80,9 @@ void init_add_i_state(struct add_i_state *s, struct repository *r)
> repo_config_get_string(r, "diff.algorithm",
> &s->interactive_diff_algorithm);
>
> + repo_config_get_int(r, "diff.context", &s->context);
> + repo_config_get_int(r, "diff.interHunkContext", &s->interhunkcontext);
This looks good
> @@ -1014,10 +1019,13 @@ static int run_diff(struct add_i_state *s, const struct pathspec *ps,
> if (count > 0) {
> struct child_process cmd = CHILD_PROCESS_INIT;
>
> - strvec_pushl(&cmd.args, "git", "diff", "-p", "--cached",
> - oid_to_hex(!is_initial ? &oid :
> - s->r->hash_algo->empty_tree),
> - "--", NULL);
> + strvec_pushl(&cmd.args, "git", "diff", "-p", "--cached", NULL);
As we're running "git diff" here nothing needs to be changed because it
reads all of the relevant config variables itself. This is why the
existing code does not worry about adding "--algorithm".
> diff --git a/add-patch.c b/add-patch.c
> index 95c67d8c80c4..b43ca1600738 100644
> --- a/add-patch.c
> +++ b/add-patch.c
> @@ -415,6 +415,8 @@ static int parse_diff(struct add_p_state *s, const struct pathspec *ps)
> {
> struct strvec args = STRVEC_INIT;
> const char *diff_algorithm = s->s.interactive_diff_algorithm;
> + int diff_context = s->s.context;
> + int diff_interhunkcontext = s->s.interhunkcontext;
> struct strbuf *plain = &s->plain, *colored = NULL;
> struct child_process cp = CHILD_PROCESS_INIT;
> char *p, *pend, *colored_p = NULL, *colored_pend = NULL, marker = '\0';
> @@ -424,6 +426,10 @@ static int parse_diff(struct add_p_state *s, const struct pathspec *ps)
> int res;
>
> strvec_pushv(&args, s->mode->diff_cmd);
> + if (diff_context != -1)
> + strvec_pushf(&args, "--unified=%i", diff_context);
> + if (diff_interhunkcontext != -1)
> + strvec_pushf(&args, "--inter-hunk-context=%i", diff_interhunkcontext);
This is good - if the user has set diff.config or diff.interhunkcontext
then we pass the appropriate values to "git diff-index" or "git
diff-files" which unlike "git diff" do not read those config values
themselves.
> if (diff_algorithm)
> strvec_pushf(&args, "--diff-algorithm=%s", diff_algorithm);
> if (s->revision) {
> diff --git a/t/t4055-diff-context.sh b/t/t4055-diff-context.sh
> index ec2804eea67c..9c024200ade7 100755
> --- a/t/t4055-diff-context.sh
> +++ b/t/t4055-diff-context.sh
I'm not sure this is the best home for the new tests. Normally tests
related to the code in add-patch.c go into t3701-add-interactive.sh
> @@ -49,7 +49,53 @@ test_expect_success 'diff.context honored by "log"' '
> ! grep firstline output &&
> git config diff.context 8 &&
> git log -1 -p >output &&
> - grep "^ firstline" output
> + grep "^ firstline" output &&
> + git config --unset diff.context
> +'
> +
> +test_expect_success 'diff.context honored by "add"' '
> + git add -p >output &&
> + ! grep firstline output &&
> + git config diff.context 8 &&
> + git add -p >output &&
> + grep "^ firstline" output &&
> + git config --unset diff.context
Eric has already mentioned using "test_config", I would go one step
further and say that as we're only running a single command we should
use "git add -c diff.context=* add -p" to avoid having to change the on
disk config at all.
I would also recommend using "test_grep ..." rather than "grep ..." and
"test_grep ! .." instead of "! grep ..." as they provide useful
debugging information if the test fails.
Given the way the code is structured with the config being read in
add-patch.c I'm not sure how much value there is in testing all the
different "-p" commands. It would however be very useful to check that
"git -c diff.interhunkcontext=8 add -p" works as expected.
I'll leave it there for now, I'll try and look at the other patches
later today or maybe tomorrow.
Best Wishes
Phillip
> +'
> +
> +test_expect_success 'diff.context honored by "commit"' '
> + ! git commit -p >output &&
> + ! grep firstline output &&
> + git config diff.context 8 &&
> + ! git commit -p >output &&
> + grep "^ firstline" output &&
> + git config --unset diff.context
> +'
> +
> +test_expect_success 'diff.context honored by "checkout"' '
> + git checkout -p >output &&
> + ! grep firstline output &&
> + git config diff.context 8 &&
> + git checkout -p >output &&
> + grep "^ firstline" output &&
> + git config --unset diff.context
> +'
> +
> +test_expect_success 'diff.context honored by "stash"' '
> + ! git stash -p >output &&
> + ! grep firstline output &&
> + git config diff.context 8 &&
> + ! git stash -p >output &&
> + grep "^ firstline" output &&
> + git config --unset diff.context
> +'
> +
> +test_expect_success 'diff.context honored by "restore"' '
> + git restore -p >output &&
> + ! grep firstline output &&
> + git config diff.context 8 &&
> + git restore -p >output &&
> + grep "^ firstline" output &&
> + git config --unset diff.context
> '
>
> test_expect_success 'The -U option overrides diff.context' '
next prev parent reply other threads:[~2025-05-06 8:55 UTC|newest]
Thread overview: 77+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-05-05 9:18 [PATCH 0/3] Better support for customising context lines in --patch commands Leon Michalak via GitGitGadget
2025-05-05 9:18 ` [PATCH 1/3] add-patch: respect diff.context configuration Leon Michalak via GitGitGadget
2025-05-05 20:29 ` Eric Sunshine
2025-05-06 8:55 ` Phillip Wood [this message]
2025-05-06 17:29 ` Junio C Hamano
2025-05-05 9:18 ` [PATCH 2/3] add-patch: add diff.context command line overrides Leon Michalak via GitGitGadget
2025-05-05 9:49 ` Kristoffer Haugsbakk
[not found] ` <CAP9jKjHj7WP91aKA9SE94zYj+naBGLUs99mF3G4BhTGcGjFDUQ@mail.gmail.com>
2025-05-05 10:11 ` Leon Michalak
2025-05-07 9:51 ` Phillip Wood
2025-05-07 18:07 ` Junio C Hamano
2025-05-07 18:28 ` Leon Michalak
2025-05-07 20:25 ` Junio C Hamano
2025-05-05 9:18 ` [PATCH 3/3] add-interactive: add new "context" subcommand Leon Michalak via GitGitGadget
2025-05-06 0:02 ` Eric Sunshine
2025-05-06 7:20 ` Leon Michalak
2025-05-06 8:28 ` Christian Couder
2025-05-06 17:07 ` Leon Michalak
2025-05-06 16:37 ` Junio C Hamano
2025-05-06 17:25 ` Leon Michalak
2025-05-07 13:30 ` Phillip Wood
2025-05-07 19:10 ` Junio C Hamano
2025-05-10 13:46 ` [PATCH v2 0/4] Better support for customising context lines in --patch commands Leon Michalak via GitGitGadget
2025-05-10 13:46 ` [PATCH v2 1/4] test: refactor to use "test_grep" Leon Michalak via GitGitGadget
2025-05-12 13:42 ` Junio C Hamano
2025-05-12 16:58 ` Leon Michalak
2025-05-10 13:46 ` [PATCH v2 2/4] test: refactor to use "test_config" Leon Michalak via GitGitGadget
2025-05-10 13:46 ` [PATCH v2 3/4] add-patch: respect diff.context configuration Leon Michalak via GitGitGadget
2025-05-13 13:52 ` Phillip Wood
2025-05-13 15:47 ` Junio C Hamano
2025-05-14 15:13 ` Phillip Wood
2025-05-15 12:58 ` Junio C Hamano
2025-05-10 13:46 ` [PATCH v2 4/4] add-patch: add diff.context command line overrides Leon Michalak via GitGitGadget
2025-05-12 16:45 ` Junio C Hamano
2025-05-12 17:03 ` Leon Michalak
2025-05-13 13:52 ` Phillip Wood
2025-05-13 14:39 ` Phillip Wood
2025-05-13 15:05 ` Leon Michalak
2025-05-14 15:13 ` phillip.wood123
2025-06-28 16:34 ` [PATCH v3 0/4] Better support for customising context lines in --patch commands Leon Michalak via GitGitGadget
2025-06-28 16:34 ` [PATCH v3 1/4] test: use "test_grep" Leon Michalak via GitGitGadget
2025-06-30 16:23 ` Junio C Hamano
2025-06-28 16:34 ` [PATCH v3 2/4] test: use "test_config" Leon Michalak via GitGitGadget
2025-06-30 16:35 ` Junio C Hamano
2025-06-28 16:34 ` [PATCH v3 3/4] add-patch: respect diff.context configuration Leon Michalak via GitGitGadget
2025-06-30 16:55 ` Junio C Hamano
2025-07-01 10:00 ` Phillip Wood
2025-06-28 16:34 ` [PATCH v3 4/4] add-patch: add diff.context command line overrides Leon Michalak via GitGitGadget
2025-06-30 17:03 ` Junio C Hamano
2025-07-01 9:59 ` Phillip Wood
2025-07-01 15:54 ` Junio C Hamano
2025-07-02 14:07 ` Phillip Wood
2025-07-02 18:28 ` Junio C Hamano
2025-07-01 9:59 ` Phillip Wood
2025-06-30 16:16 ` [PATCH v3 0/4] Better support for customising context lines in --patch commands Junio C Hamano
2025-07-09 0:09 ` Junio C Hamano
2025-07-09 7:57 ` Leon Michalak
2025-07-09 15:32 ` Junio C Hamano
2025-07-19 12:28 ` [PATCH v4 " Leon Michalak via GitGitGadget
2025-07-19 12:28 ` [PATCH v4 1/4] t: use test_grep in t3701 and t4055 Leon Michalak via GitGitGadget
2025-07-19 12:28 ` [PATCH v4 2/4] t: use test_config in t4055 Leon Michalak via GitGitGadget
2025-07-19 12:28 ` [PATCH v4 3/4] add-patch: respect diff.context configuration Leon Michalak via GitGitGadget
2025-07-19 12:28 ` [PATCH v4 4/4] add-patch: add diff.context command line overrides Leon Michalak via GitGitGadget
2025-07-22 16:01 ` Phillip Wood
2025-07-22 18:02 ` Leon Michalak
2025-07-22 18:05 ` Leon Michalak
2025-07-23 9:41 ` Phillip Wood
2025-07-21 16:50 ` [PATCH v4 0/4] Better support for customising context lines in --patch commands Junio C Hamano
2025-07-22 16:05 ` Phillip Wood
2025-07-22 17:20 ` Junio C Hamano
2025-07-29 7:01 ` [PATCH v5 " Leon Michalak via GitGitGadget
2025-07-29 7:01 ` [PATCH v5 1/4] t: use test_grep in t3701 and t4055 Leon Michalak via GitGitGadget
2025-07-29 7:01 ` [PATCH v5 2/4] t: use test_config in t4055 Leon Michalak via GitGitGadget
2025-07-29 7:01 ` [PATCH v5 3/4] add-patch: respect diff.context configuration Leon Michalak via GitGitGadget
2025-07-29 7:01 ` [PATCH v5 4/4] add-patch: add diff.context command line overrides Leon Michalak via GitGitGadget
2025-07-29 15:21 ` [PATCH v5 0/4] Better support for customising context lines in --patch commands Phillip Wood
2025-07-29 15:55 ` Junio C Hamano
2025-07-29 16:18 ` Leon Michalak
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=c5591cbb-7eb9-4b69-85a1-ae1678983b14@gmail.com \
--to=phillip.wood123@gmail.com \
--cc=git@vger.kernel.org \
--cc=gitgitgadget@gmail.com \
--cc=leonmichalak6@gmail.com \
--cc=phillip.wood@dunelm.org.uk \
--cc=sunshine@sunshineco.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).