git.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Elijah Newren <newren@gmail.com>
To: gitster@pobox.com
Cc: git@vger.kernel.org, Jim Foucar <jgfouca@sandia.gov>,
	Elijah Newren <newren@gmail.com>
Subject: [PATCHv2 38/56] merge-recursive: Fix modify/delete resolution in the recursive case
Date: Thu, 11 Aug 2011 23:20:11 -0600	[thread overview]
Message-ID: <1313126429-17368-39-git-send-email-newren@gmail.com> (raw)
In-Reply-To: <1313126429-17368-1-git-send-email-newren@gmail.com>

When o->call_depth>0 and we have conflicts, we try to find "middle ground"
when creating the virtual merge base.  In the case of content conflicts,
this can be done by doing a three-way content merge and using the result.
In all parts where the three-way content merge is clean, it is the correct
middle ground, and in parts where it conflicts there is no middle ground
but the conflict markers provide a good compromise since they are unlikely
to accidentally match any further changes.

In the case of a modify/delete conflict, we cannot do the same thing.
Accepting either endpoint as the resolution for the virtual merge base
runs the risk that when handling the non-recursive case we will silently
accept one person's resolution over another without flagging a conflict.
In this case, the closest "middle ground" we have is actually the merge
base of the candidate merge bases.  (We could alternatively attempt a
three way content merge using an empty file in place of the deleted file,
but that seems to be more work than necessary.)

Signed-off-by: Elijah Newren <newren@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
---
No changes since v1.

 merge-recursive.c                 |   38 +++++++++++++++++++++++-------------
 t/t6036-recursive-corner-cases.sh |    4 +-
 2 files changed, 26 insertions(+), 16 deletions(-)

diff --git a/merge-recursive.c b/merge-recursive.c
index ccf71d3..a0dc0bf 100644
--- a/merge-recursive.c
+++ b/merge-recursive.c
@@ -1324,27 +1324,42 @@ error_return:
 
 static void handle_delete_modify(struct merge_options *o,
 				 const char *path,
-				 const char *new_path,
+				 unsigned char *o_sha, int o_mode,
 				 unsigned char *a_sha, int a_mode,
 				 unsigned char *b_sha, int b_mode)
 {
-	if (!a_sha) {
+	char *renamed = NULL;
+	if (dir_in_way(path, !o->call_depth)) {
+		renamed = unique_path(o, path, a_sha ? o->branch1 : o->branch2);
+	}
+
+	if (o->call_depth) {
+		/*
+		 * We cannot arbitrarily accept either a_sha or b_sha as
+		 * correct; since there is no true "middle point" between
+		 * them, simply reuse the base version for virtual merge base.
+		 */
+		remove_file_from_cache(path);
+		update_file(o, 0, o_sha, o_mode, renamed ? renamed : path);
+	} else if (!a_sha) {
 		output(o, 1, "CONFLICT (delete/modify): %s deleted in %s "
 		       "and modified in %s. Version %s of %s left in tree%s%s.",
 		       path, o->branch1,
 		       o->branch2, o->branch2, path,
-		       NULL == new_path ? "" : " at ",
-		       NULL == new_path ? "" : new_path);
-		update_file(o, 0, b_sha, b_mode, new_path ? new_path : path);
+		       NULL == renamed ? "" : " at ",
+		       NULL == renamed ? "" : renamed);
+		update_file(o, 0, b_sha, b_mode, renamed ? renamed : path);
 	} else {
 		output(o, 1, "CONFLICT (delete/modify): %s deleted in %s "
 		       "and modified in %s. Version %s of %s left in tree%s%s.",
 		       path, o->branch2,
 		       o->branch1, o->branch1, path,
-		       NULL == new_path ? "" : " at ",
-		       NULL == new_path ? "" : new_path);
-		update_file(o, 0, a_sha, a_mode, new_path ? new_path : path);
+		       NULL == renamed ? "" : " at ",
+		       NULL == renamed ? "" : renamed);
+		update_file(o, 0, a_sha, a_mode, renamed ? renamed : path);
 	}
+	free(renamed);
+
 }
 
 static int merge_content(struct merge_options *o,
@@ -1514,14 +1529,9 @@ static int process_entry(struct merge_options *o,
 			remove_file(o, 1, path, !a_sha);
 		} else {
 			/* Modify/delete; deleted side may have put a directory in the way */
-			char *renamed = NULL;
 			clean_merge = 0;
-			if (dir_in_way(path, !o->call_depth)) {
-				renamed = unique_path(o, path, a_sha ? o->branch1 : o->branch2);
-			}
-			handle_delete_modify(o, path, renamed,
+			handle_delete_modify(o, path, o_sha, o_mode,
 					     a_sha, a_mode, b_sha, b_mode);
-			free(renamed);
 		}
 	} else if ((!o_sha && a_sha && !b_sha) ||
 		   (!o_sha && !a_sha && b_sha)) {
diff --git a/t/t6036-recursive-corner-cases.sh b/t/t6036-recursive-corner-cases.sh
index b046e1b..314fdae 100755
--- a/t/t6036-recursive-corner-cases.sh
+++ b/t/t6036-recursive-corner-cases.sh
@@ -296,7 +296,7 @@ test_expect_success 'setup criss-cross + modify/delete resolved differently' '
 	git tag E
 '
 
-test_expect_failure 'git detects conflict merging criss-cross+modify/delete' '
+test_expect_success 'git detects conflict merging criss-cross+modify/delete' '
 	git checkout D^0 &&
 
 	test_must_fail git merge -s recursive E^0 &&
@@ -308,7 +308,7 @@ test_expect_failure 'git detects conflict merging criss-cross+modify/delete' '
 	test $(git rev-parse :2:file) = $(git rev-parse B:file)
 '
 
-test_expect_failure 'git detects conflict merging criss-cross+modify/delete, reverse direction' '
+test_expect_success 'git detects conflict merging criss-cross+modify/delete, reverse direction' '
 	git reset --hard &&
 	git checkout E^0 &&
 
-- 
1.7.6.100.gac5c1

  parent reply	other threads:[~2011-08-12  5:22 UTC|newest]

Thread overview: 65+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2011-08-12  5:19 [PATCHv2 00/57] Re-roll of en/merge-recursive from pu Elijah Newren
2011-08-12  5:19 ` [PATCHv2 01/56] t6042: Add a testcase where git deletes an untracked file Elijah Newren
2011-08-12  5:19 ` [PATCHv2 02/56] t6042: Add failing testcase for rename/modify/add-source conflict Elijah Newren
2011-08-12  5:19 ` [PATCHv2 03/56] t6042: Add a pair of cases where undetected renames cause issues Elijah Newren
2011-08-12  5:19 ` [PATCHv2 04/56] t6042: Add a testcase where undetected rename causes silent file deletion Elijah Newren
2011-08-12  5:19 ` [PATCHv2 05/56] t6042: Add tests for content issues with modify/rename/directory conflicts Elijah Newren
2011-08-12  5:19 ` [PATCHv2 06/56] t6042: Ensure rename/rename conflicts leave index and workdir in sane state Elijah Newren
2011-08-12  5:19 ` [PATCHv2 07/56] t6042: Add failing testcases for rename/rename/add-{source,dest} conflicts Elijah Newren
2011-08-12  5:19 ` [PATCHv2 08/56] t6036: Add differently resolved modify/delete conflict in criss-cross test Elijah Newren
2011-08-12  5:19 ` [PATCHv2 09/56] t6036: criss-cross with weird content can fool git into clean merge Elijah Newren
2011-08-12  5:19 ` [PATCHv2 10/56] t6036: tests for criss-cross merges with various directory/file conflicts Elijah Newren
2011-08-12  5:19 ` [PATCHv2 11/56] t6036: criss-cross w/ rename/rename(1to2)/modify+rename/rename(2to1)/modify Elijah Newren
2011-08-12  5:19 ` [PATCHv2 12/56] t6036: criss-cross + rename/rename(1to2)/add-source + modify/modify Elijah Newren
2011-08-12  5:19 ` [PATCHv2 13/56] t6022: Remove unnecessary untracked files to make test cleaner Elijah Newren
2011-08-12  5:19 ` [PATCHv2 14/56] t6022: New tests checking for unnecessary updates of files Elijah Newren
2011-08-12  5:19 ` [PATCHv2 15/56] t6022: Add testcase for merging a renamed file with a simple change Elijah Newren
2011-08-12  5:19 ` [PATCHv2 16/56] merge-recursive: Make BUG message more legible by adding a newline Elijah Newren
2011-08-12  5:19 ` [PATCHv2 17/56] merge-recursive: Correct a comment Elijah Newren
2011-08-12  5:19 ` [PATCHv2 18/56] merge-recursive: Mark some diff_filespec struct arguments const Elijah Newren
2011-08-12  5:19 ` [PATCHv2 19/56] merge-recursive: Consolidate different update_stages functions Elijah Newren
2011-08-12  5:19 ` [PATCHv2 20/56] merge-recursive: Remember to free generated unique path names Elijah Newren
2011-08-12  5:19 ` [PATCHv2 21/56] merge-recursive: Avoid working directory changes during recursive case Elijah Newren
2011-08-12  5:19 ` [PATCHv2 22/56] merge-recursive: Fix recursive case with D/F conflict via add/add conflict Elijah Newren
2011-08-12  5:19 ` [PATCHv2 23/56] merge-recursive: Fix sorting order and directory change assumptions Elijah Newren
2011-08-12  5:19 ` [PATCHv2 24/56] merge-recursive: Fix code checking for D/F conflicts still being present Elijah Newren
2011-08-12  5:19 ` [PATCHv2 25/56] merge-recursive: Save D/F conflict filenames instead of unlinking them Elijah Newren
2011-08-12  5:19 ` [PATCHv2 26/56] merge-recursive: Split was_tracked() out of would_lose_untracked() Elijah Newren
2011-08-12  5:20 ` [PATCHv2 27/56] string-list: Add API to remove an item from an unsorted list Elijah Newren
2011-08-12  7:00   ` Johannes Sixt
2011-08-12  9:27     ` Alex Riesen
2011-08-12 22:14     ` Elijah Newren
2011-08-13  9:08       ` Johannes Sixt
2011-08-12  5:20 ` [PATCHv2 28/56] merge-recursive: Allow make_room_for_path() to remove D/F entries Elijah Newren
2011-08-12  5:20 ` [PATCHv2 29/56] merge-recursive: Split update_stages_and_entry; only update stages at end Elijah Newren
2011-08-12  5:20 ` [PATCHv2 30/56] merge-recursive: Fix deletion of untracked file in rename/delete conflicts Elijah Newren
2011-08-12  5:20 ` [PATCHv2 31/56] merge-recursive: Make dead code for rename/rename(2to1) conflicts undead Elijah Newren
2011-08-12  5:20 ` [PATCHv2 32/56] merge-recursive: Add comments about handling rename/add-source cases Elijah Newren
2011-08-12  5:20 ` [PATCHv2 33/56] merge-recursive: Improve handling of rename target vs. directory addition Elijah Newren
2011-08-12  5:20 ` [PATCHv2 34/56] merge-recursive: Consolidate process_entry() and process_df_entry() Elijah Newren
2011-08-12  5:20 ` [PATCHv2 35/56] merge-recursive: Cleanup and consolidation of rename_conflict_info Elijah Newren
2011-08-12  5:20 ` [PATCHv2 36/56] merge-recursive: Provide more info in conflict markers with file renames Elijah Newren
2011-08-12  5:20 ` [PATCHv2 37/56] merge-recursive: When we detect we can skip an update, actually skip it Elijah Newren
2011-08-12  5:20 ` Elijah Newren [this message]
2011-08-12  5:20 ` [PATCHv2 39/56] merge-recursive: Introduce a merge_file convenience function Elijah Newren
2011-08-12  5:20 ` [PATCHv2 40/56] merge-recursive: Fix rename/rename(1to2) resolution for virtual merge base Elijah Newren
2011-08-12  5:20 ` [PATCHv2 41/56] merge-recursive: Small cleanups for conflict_rename_rename_1to2 Elijah Newren
2011-08-12  5:20 ` [PATCHv2 42/56] merge-recursive: Defer rename/rename(2to1) handling until process_entry Elijah Newren
2011-08-12  5:20 ` [PATCHv2 43/56] merge-recursive: Record more data needed for merging with dual renames Elijah Newren
2011-08-12  5:20 ` [PATCHv2 44/56] merge-recursive: Create function for merging with branchname:file markers Elijah Newren
2011-08-12  5:20 ` [PATCHv2 45/56] merge-recursive: Consider modifications in rename/rename(2to1) conflicts Elijah Newren
2011-08-12  5:20 ` [PATCHv2 46/56] merge-recursive: Make modify/delete handling code reusable Elijah Newren
2011-08-12  5:20 ` [PATCHv2 47/56] merge-recursive: Have conflict_rename_delete reuse modify/delete code Elijah Newren
2011-08-12  5:20 ` [PATCHv2 48/56] merge-recursive: add handling for rename/rename/add-dest/add-dest Elijah Newren
2011-08-12  5:20 ` [PATCHv2 49/56] merge-recursive: Fix working copy handling for rename/rename/add/add Elijah Newren
2011-08-12  5:20 ` [PATCHv2 50/56] t3030: fix accidental success in symlink rename Elijah Newren
2011-08-12  5:20 ` [PATCHv2 51/56] t6022: Add testcase for spurious "refusing to lose untracked" messages Elijah Newren
2011-08-12  5:20 ` [PATCHv2 52/56] merge-recursive: Fix spurious 'refusing to lose untracked file...' messages Elijah Newren
2011-08-12  5:20 ` [PATCHv2 53/56] t6022: Additional tests checking for unnecessary updates of files Elijah Newren
2011-08-12  5:20 ` [PATCHv2 54/56] merge-recursive: Avoid unnecessary file rewrites Elijah Newren
2011-08-12  5:20 ` [PATCHv2 55/56] t6036: criss-cross + rename/rename(1to2)/add-dest + simple modify Elijah Newren
2011-08-12  5:20 ` [PATCHv2 56/56] merge-recursive: Fix virtual merge base for rename/rename(1to2)/add-dest Elijah Newren
2011-08-12  5:48 ` [PATCHv2 00/57] Re-roll of en/merge-recursive from pu Junio C Hamano
2011-08-12 21:59   ` Elijah Newren
2011-08-13  2:37     ` Elijah Newren
2011-08-13  2:23 ` [PATCHv2 57/57] merge-recursive: Don't re-sort a list whose order we depend upon Elijah Newren

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=1313126429-17368-39-git-send-email-newren@gmail.com \
    --to=newren@gmail.com \
    --cc=git@vger.kernel.org \
    --cc=gitster@pobox.com \
    --cc=jgfouca@sandia.gov \
    /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).