git.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: "Phillip Wood via GitGitGadget" <gitgitgadget@gmail.com>
To: git@vger.kernel.org
Cc: Justin Tobler <jltobler@gmail.com>,
	Junio C Hamano <gitster@pobox.com>,
	Phillip Wood <phillip.wood123@gmail.com>,
	Phillip Wood <phillip.wood@dunelm.org.uk>
Subject: [PATCH v3 0/2] add -p: a couple of hunk splitting fixes
Date: Thu, 25 Sep 2025 15:10:36 +0000	[thread overview]
Message-ID: <pull.1863.v3.git.1758813038.gitgitgadget@gmail.com> (raw)
In-Reply-To: <pull.1863.v2.git.1757950144.gitgitgadget@gmail.com>

Thanks to Junio for his comments on V2. I have removed the dependency on
WITH_BREAKING_CHANGES in favor of "lets change it and see if anyone screams"

Changes since V2: Remove dependency on WITH_BREAKING_CHANGES and change the
behavior unconditionally. This takes us back to V1 with (hopefully) better
commit messages and a style fix in the tests.

V2 Cover Letter: Thanks to Justin and Junio for their comments in V1. Sorry
for the long delay in re-rolling - I thought I'd sent these ages ago and
then discovered that they weren't upstream and realized I had not, in fact,
sent them after-all.

Changes since V1:

 * Patch 1: The new hunks created by splitting a hunk are now only marked as
   "undecided" when WITH_BREAKING_CHANGES is enabled.

 * Patch 2: Reworded commit message and added a space before a redirection
   in the test

V1 Cover Letter:

This series fixes a couple of infelicities when splitting hunks that have
already been selected or edited which I noticed a while ago when preparing
the test for 'pw/add-patch-with-suppress-blank-empty'.

Phillip Wood (2):
  add -p: mark split hunks as undecided
  add-patch: update hunk splitability after editing

 add-patch.c                | 15 +++++++++++++--
 t/t3701-add-interactive.sh | 31 +++++++++++++++++++++++++++++++
 2 files changed, 44 insertions(+), 2 deletions(-)


base-commit: 4975ec3473b4bc61bc8a3df1ef29d0b7e7959e87
Published-As: https://github.com/gitgitgadget/git/releases/tag/pr-1863%2Fphillipwood%2Fadd-p-split-hunks-are-undecided-v3
Fetch-It-Via: git fetch https://github.com/gitgitgadget/git pr-1863/phillipwood/add-p-split-hunks-are-undecided-v3
Pull-Request: https://github.com/gitgitgadget/git/pull/1863

Range-diff vs v2:

 1:  3e2ec7b37f ! 1:  4935dde399 add -p: mark split hunks as undecided
     @@ Commit message
          the split hunks they will now have to do more work re-selecting the
          remaining split hunks. However, changing the selection of any of the
          other newly created hunks is now much simpler as the user no-longer has
     -    to navigate back to them in order to change their selected state. Due
     -    to concerns that users may be relying on the current behaviour [1]
     -    this change is guarded by WITH_BREAKING_CHANGES.
     -
     -    [1] https://lore.kernel.org/git/xmqqjz9b6xr1.fsf@gitster.g
     +    to navigate back to them in order to change their selected state.
      
          Signed-off-by: Phillip Wood <phillip.wood@dunelm.org.uk>
      
     - ## Documentation/BreakingChanges.adoc ##
     -@@ Documentation/BreakingChanges.adoc: A prerequisite for this change is that the ecosystem is ready to support the
     - "reftable" format. Most importantly, alternative implementations of Git like
     - JGit, libgit2 and Gitoxide need to support it.
     - 
     -+* The behavior of "git add -p" has been changed so that splitting a
     -+  hunk that has already been marked as selected or unselected will now
     -+  prompt the user to select each of the new hunks created by the
     -+  split instead of them inheriting their state from the original hunk.
     -+
     - === Removals
     - 
     - * Support for grafting commits has long been superseded by git-replace(1).
     -
       ## add-patch.c ##
      @@ add-patch.c: static int split_hunk(struct add_p_state *s, struct file_diff *file_diff,
       			* sizeof(*hunk));
       	hunk = file_diff->hunk + hunk_index;
       	hunk->splittable_into = 1;
     -+#ifdef WITH_BREAKING_CHANGES
      +	hunk->use = UNDECIDED_HUNK;
     -+#endif
       	memset(hunk + 1, 0, (splittable_into - 1) * sizeof(*hunk));
       
       	header = &hunk->header;
     @@ add-patch.c: next_hunk_line:
       
       		hunk++;
       		hunk->splittable_into = 1;
     -+#ifdef WITH_BREAKING_CHANGES
     +-		hunk->use = hunk[-1].use;
      +		hunk->use = UNDECIDED_HUNK;
     -+#else
     - 		hunk->use = hunk[-1].use;
     -+#endif
       		header = &hunk->header;
       
       		header->old_count = header->new_count = context_line_count;
     @@ t/t3701-add-interactive.sh: do
       	'
       done
       
     -+test_expect_success WITH_BREAKING_CHANGES 'splitting previous hunk marks split hunks as undecided' '
     ++test_expect_success 'splitting previous hunk marks split hunks as undecided' '
      +	test_write_lines a " " b c d e f g h i j k >file &&
      +	git add file &&
      +	test_write_lines x " " b y d e f g h i j x >file &&
 2:  3a831b1a2d ! 2:  390686dbb3 add-patch: update hunk splitability after editing
     @@ add-patch.c: static ssize_t recount_edited_hunk(struct add_p_state *s, struct hu
       
      
       ## t/t3701-add-interactive.sh ##
     -@@ t/t3701-add-interactive.sh: test_expect_success WITH_BREAKING_CHANGES 'splitting previous hunk marks split h
     +@@ t/t3701-add-interactive.sh: test_expect_success 'splitting previous hunk marks split hunks as undecided' '
       	test_cmp expect actual
       '
       
     @@ t/t3701-add-interactive.sh: test_expect_success WITH_BREAKING_CHANGES 'splitting
      +	test_write_lines A b c d E f g h i j k l M n >file &&
      +	(
      +		test_set_editor "$(pwd)/fake-editor.sh" &&
     -+		if test_have_prereq WITH_BREAKING_CHANGES
     -+		then
     -+			test_write_lines e K s j y n y q
     -+		else
     -+			test_write_lines e K s n K n y q
     -+		fi | git add -p file
     ++		test_write_lines e K s j y n y q | git add -p file
      +	) &&
      +	git cat-file blob :file >actual &&
      +	test_write_lines a b d e f g h i j k l M n >expect &&

-- 
gitgitgadget

  parent reply	other threads:[~2025-09-25 15:10 UTC|newest]

Thread overview: 27+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-02-21 14:57 [PATCH 0/2] add -p: a couple of hunk splitting fixes Phillip Wood via GitGitGadget
2025-02-21 14:57 ` [PATCH 1/2] add -p: mark split hunks as undecided Phillip Wood via GitGitGadget
2025-02-21 19:52   ` Justin Tobler
2025-02-21 21:31   ` Junio C Hamano
2025-02-26 14:40     ` phillip.wood123
2025-02-26 16:49       ` Junio C Hamano
2025-02-27 16:22         ` phillip.wood123
2025-02-27 18:36           ` Junio C Hamano
2025-02-28 16:19             ` Phillip Wood
2025-02-28 17:06               ` Junio C Hamano
2025-03-04 10:25                 ` Phillip Wood
2025-02-21 14:57 ` [PATCH 2/2] add-patch: update hunk splitability after editing Phillip Wood via GitGitGadget
2025-02-21 20:29   ` Justin Tobler
2025-02-21 21:42   ` Junio C Hamano
2025-09-15 15:29 ` [PATCH v2 0/2] add -p: a couple of hunk splitting fixes Phillip Wood via GitGitGadget
2025-09-15 15:29   ` [PATCH v2 1/2] add -p: mark split hunks as undecided Phillip Wood via GitGitGadget
2025-09-15 17:44     ` Junio C Hamano
2025-09-16  9:36       ` Phillip Wood
2025-09-16 16:03         ` Junio C Hamano
2025-09-15 15:29   ` [PATCH v2 2/2] add-patch: update hunk splitability after editing Phillip Wood via GitGitGadget
2025-09-25 15:10   ` Phillip Wood via GitGitGadget [this message]
2025-09-25 15:10     ` [PATCH v3 1/2] add -p: mark split hunks as undecided Phillip Wood via GitGitGadget
2025-09-25 18:21       ` Junio C Hamano
2025-09-26 10:12         ` Phillip Wood
2025-09-26 17:37           ` Junio C Hamano
2025-10-08 13:51             ` Phillip Wood
2025-09-25 15:10     ` [PATCH v3 2/2] add-patch: update hunk splitability after editing Phillip Wood via GitGitGadget

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=pull.1863.v3.git.1758813038.gitgitgadget@gmail.com \
    --to=gitgitgadget@gmail.com \
    --cc=git@vger.kernel.org \
    --cc=gitster@pobox.com \
    --cc=jltobler@gmail.com \
    --cc=phillip.wood123@gmail.com \
    --cc=phillip.wood@dunelm.org.uk \
    /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).