[PATCH v3 7/8] push: add DWYM support for "git push refs/remotes/...:<dst>"

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

 



Add DWYM support for pushing a ref in refs/remotes/* when the <dst>
ref is unqualified. Now instead of erroring out we support e.g.:

    $ ./git-push avar refs/remotes/origin/master:upstream-master -n
    To github.com:avar/git.git
     * [new branch]            origin/master -> upstream-master

Before this we wouldn't know what do do with
refs/remotes/origin/master, now we'll look it up and discover that
it's a commit (or tag) and add a refs/{heads,tags}/ prefix to <dst> as
appropriate.

The error message emitted when we still don't know what to do has been
amended to note that this is something we tried:

    $ ./git-push avar v2.19.0^{commit}:newbranch -n
    error: The destination you provided is not a full refname (i.e.,
    starting with "refs/"). We tried to guess what you meant by:

    - Looking for a ref that matches 'newbranch' on the remote side.
    - Checking if the <src> being pushed ('v2.19.0^{commit}')
      is a ref in "refs/{heads,tags}/". If so we add a corresponding
      refs/{heads,tags}/ prefix on the remote side.
    - Checking if the <src> being pushed ('v2.19.0^{commit}')
      is a commit or tag in "refs/remotes/*". Then we infer a
      corresponding refs/{heads,tags} on the remote side.

    None of those worked, so we gave up. You must fully-qualify the ref.
    hint: The <src> part of the refspec is a commit object.
    hint: Did you mean to create a new branch by pushing to
    hint: 'v2.19.0^{commit}:refs/heads/newbranch'?

I'm bending over backwards and assuming that someone might have hacked
in remote tracking tags (see [1] for a discussion of how that can be
done), but punting on any tree or blob objects found under
refs/remotes/*.

This is the first use of the %N$<fmt> style of printf format in
the *.[ch] files in our codebase. It's supported by POSIX[2] and
there's existing uses for it in po/*.po files, so hopefully it won't
cause any trouble. It's more obvious for translators than to have a
3rd argument to the function identical to the 2nd, by re-using the 2nd
it's clear that we're continuing to talk about the <src> part of the
refspec.

1. https://public-inbox.org/git/87zi1jxjqn.fsf@xxxxxxxxxxxxxxxxxxx/
2. http://pubs.opengroup.org/onlinepubs/7908799/xsh/fprintf.html

Signed-off-by: Ævar Arnfjörð Bjarmason <avarab@xxxxxxxxx>
---
 remote.c          | 20 +++++++++++++++++++-
 t/t5505-remote.sh | 14 +++++++++-----
 2 files changed, 28 insertions(+), 6 deletions(-)

diff --git a/remote.c b/remote.c
index 93f802509d..c243e3d89e 100644
--- a/remote.c
+++ b/remote.c
@@ -973,6 +973,21 @@ static char *guess_ref(const char *name, struct ref *peer)
 		strbuf_addstr(&buf, "refs/heads/");
 	} else if (starts_with(r, "refs/tags/")) {
 		strbuf_addstr(&buf, "refs/tags/");
+	} else if (starts_with(r, "refs/remotes/")) {
+		struct object_id oid;
+		enum object_type type;
+
+		if (get_oid(peer->name, &oid))
+			BUG("'%s' is not a valid object, "
+			    "match_explicit_lhs() should catch this!",
+			    peer->name);
+		type = oid_object_info(the_repository, &oid, NULL);
+		if (type == OBJ_COMMIT)
+			strbuf_addstr(&buf, "refs/heads/");
+		else if (type == OBJ_TAG)
+			strbuf_addstr(&buf, "refs/tags/");
+		else
+			return NULL;
 	} else {
 		return NULL;
 	}
@@ -1024,8 +1039,11 @@ static void show_push_unqualified_ref_name_error(const char *dst_value,
 		"- Checking if the <src> being pushed ('%s')\n"
 		"  is a ref in \"refs/{heads,tags}/\". If so we add a corresponding\n"
 		"  refs/{heads,tags}/ prefix on the remote side.\n"
+		"- Checking if the <src> being pushed ('%2$s')\n"
+		"  is a commit or tag in \"refs/remotes/*\". Then we infer a\n"
+		"  corresponding refs/{heads,tags} on the remote side.\n"
 		"\n"
-		"Neither worked, so we gave up. You must fully-qualify the ref."),
+		"None of those worked, so we gave up. You must fully-qualify the ref."),
 	      dst_value, matched_src_name);
 
 	if (!advice_push_unqualified_ref_name)
diff --git a/t/t5505-remote.sh b/t/t5505-remote.sh
index 979a13b415..a6337b50e4 100755
--- a/t/t5505-remote.sh
+++ b/t/t5505-remote.sh
@@ -1260,11 +1260,15 @@ test_expect_success 'refs/remotes/* <src> refspec and unqualified <dst> DWIM and
 		git config --add remote.two.fetch "+refs/blobs/*:refs/remotes/two-blobs/*" &&
 		git fetch --no-tags two &&
 
-		test_must_fail git push origin refs/remotes/two/another:dst 2>err &&
-		test_i18ngrep "error: The destination you" err &&
-
-		test_must_fail git push origin refs/remotes/two-tags/some-tag:dst-tag 2>err &&
-		test_i18ngrep "error: The destination you" err &&
+		echo commit >expected &&
+		git push origin refs/remotes/two/another:dst &&
+		git -C ../one cat-file -t refs/heads/dst >actual &&
+		test_cmp expected actual &&
+
+		echo tag >expected &&
+		git push origin refs/remotes/two-tags/some-tag:dst-tag &&
+		git -C ../one cat-file -t refs/tags/dst-tag >actual &&
+		test_cmp expected actual &&
 
 		test_must_fail git push origin refs/remotes/two-trees/my-head-tree:dst-tree 2>err &&
 		test_i18ngrep "error: The destination you" err &&
-- 
2.19.1.759.g500967bb5e




[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