[PATCH v2 4/8] pull: since --ff-only overrides, handle it first

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



From: Elijah Newren <newren@xxxxxxxxx>

There are both merge and rebase branches in the logic, and previously
both had to handle fast-forwarding.  Merge handled that implicitly
(because git merge handles it directly), while in rebase it was
explicit.  Given that the --ff-only flag is meant to override any
--rebase or --no-rebase, make the code reflect that by handling
--ff-only before the merge-vs-rebase logic.

It turns out that this also fixes a bug for submodules.  Previously,
when --ff-only was given, the code would run `merge --ff-only` on the
main module, and then run `submodule update --recursive --rebase` on the
submodules.  With this change, we still run `merge --ff-only` on the
main module, but now run `submodule update --recursive --checkout` on
the submodules.  I believe this better reflects the intent of --ff-only
to have it apply to both the main module and the submodules.

(Sidenote: It is somewhat interesting that all merges pass `--checkout`
to submodule update, even when `--no-ff` is specified, meaning that it
will only do fast-forward merges for submodules.  This was discussed in
commit a6d7eb2c7a ("pull: optionally rebase submodules (remote submodule
changes only)", 2017-06-23).  The same limitations apply now as then, so
we are not trying to fix this at this time.)

Signed-off-by: Elijah Newren <newren@xxxxxxxxx>
---
 builtin/pull.c  | 24 +++++++++---------------
 t/t5520-pull.sh |  6 +++---
 2 files changed, 12 insertions(+), 18 deletions(-)

diff --git a/builtin/pull.c b/builtin/pull.c
index d9796604825..92150f976cd 100644
--- a/builtin/pull.c
+++ b/builtin/pull.c
@@ -1046,15 +1046,15 @@ int cmd_pull(int argc, const char **argv, const char *prefix)
 
 	can_ff = get_can_ff(&orig_head, &merge_heads.oid[0]);
 
-	if (!can_ff) {
-		if (opt_ff) {
-			if (!strcmp(opt_ff, "--ff-only"))
-				die_ff_impossible();
-		} else {
-			if (rebase_unspecified && opt_verbosity >= 0)
-				show_advice_pull_non_ff();
-		}
+	/* ff-only takes precedence over rebase */
+	if (opt_ff && !strcmp(opt_ff, "--ff-only")) {
+		if (!can_ff)
+			die_ff_impossible();
+		opt_rebase = REBASE_FALSE;
 	}
+	/* If no action specified and we can't fast forward, then warn. */
+	if (!opt_ff && rebase_unspecified && !can_ff)
+		show_advice_pull_non_ff();
 
 	if (opt_rebase) {
 		int ret = 0;
@@ -1069,13 +1069,7 @@ int cmd_pull(int argc, const char **argv, const char *prefix)
 		    submodule_touches_in_range(the_repository, &upstream, &curr_head))
 			die(_("cannot rebase with locally recorded submodule modifications"));
 
-		if (can_ff) {
-			/* we can fast-forward this without invoking rebase */
-			opt_ff = "--ff-only";
-			ret = run_merge();
-		} else {
-			ret = run_rebase(&newbase, &upstream);
-		}
+		ret = run_rebase(&newbase, &upstream);
 
 		if (!ret && (recurse_submodules == RECURSE_SUBMODULES_ON ||
 			     recurse_submodules == RECURSE_SUBMODULES_ON_DEMAND))
diff --git a/t/t5520-pull.sh b/t/t5520-pull.sh
index e2c0c510222..4b50488141f 100755
--- a/t/t5520-pull.sh
+++ b/t/t5520-pull.sh
@@ -295,7 +295,7 @@ test_expect_success '--rebase (merge) fast forward' '
 	# The above only validates the result.  Did we actually bypass rebase?
 	git reflog -1 >reflog.actual &&
 	sed "s/^[0-9a-f][0-9a-f]*/OBJID/" reflog.actual >reflog.fuzzy &&
-	echo "OBJID HEAD@{0}: pull --rebase . ff: Fast-forward" >reflog.expected &&
+	echo "OBJID HEAD@{0}: pull --rebase . ff (finish): returning to refs/heads/to-rebase" >reflog.expected &&
 	test_cmp reflog.expected reflog.fuzzy
 '
 
@@ -307,8 +307,8 @@ test_expect_success '--rebase (am) fast forward' '
 
 	# The above only validates the result.  Did we actually bypass rebase?
 	git reflog -1 >reflog.actual &&
-	sed "s/^[0-9a-f][0-9a-f]*/OBJID/" reflog.actual >reflog.fuzzy &&
-	echo "OBJID HEAD@{0}: pull --rebase . ff: Fast-forward" >reflog.expected &&
+	sed -e "s/^[0-9a-f][0-9a-f]*/OBJID/" -e "s/[0-9a-f][0-9a-f]*$/OBJID/" reflog.actual >reflog.fuzzy &&
+	echo "OBJID HEAD@{0}: rebase finished: refs/heads/to-rebase onto OBJID" >reflog.expected &&
 	test_cmp reflog.expected reflog.fuzzy
 '
 
-- 
gitgitgadget




[Index of Archives]     [Linux Kernel Development]     [Gcc Help]     [IETF Annouce]     [DCCP]     [Netdev]     [Networking]     [Security]     [V4L]     [Bugtraq]     [Yosemite]     [MIPS Linux]     [ARM Linux]     [Linux Security]     [Linux RAID]     [Linux SCSI]     [Fedora Users]

  Powered by Linux