summary | shortlog | log | commit | commitdiff | tree
raw | patch | inline | side by side (parent: 87b5054)
raw | patch | inline | side by side (parent: 87b5054)
author | Jeff King <peff@peff.net> | |
Mon, 18 Oct 2010 18:39:17 +0000 (14:39 -0400) | ||
committer | Junio C Hamano <gitster@pobox.com> | |
Fri, 29 Oct 2010 21:13:33 +0000 (14:13 -0700) |
Usually when applying a binary diff generated without
--binary, it will be rejected early, as we don't even have
the full sha1 of the pre- and post-images.
However, if the diff is generated with --full-index (but not
--binary), then we will actually try to apply it. If we have
the postimage blob, then we can take a shortcut and never
even look at the binary diff at all (e.g., this can happen
when rebasing changes within a repository).
If we don't have the postimage blob, though, we try to look
at the actual fragments, of which there are none, and get a
segfault. This patch checks explicitly for that case and
complains to the user instead of segfaulting. We need to
keep the check at a low level so that the "shortcut" case
above continues to work.
We also add a test that demonstrates the segfault. While
we're at it, let's also explicitly test the shortcut case.
Reported-by: Rafaël Carré <rafael.carre@gmail.com>
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
--binary, it will be rejected early, as we don't even have
the full sha1 of the pre- and post-images.
However, if the diff is generated with --full-index (but not
--binary), then we will actually try to apply it. If we have
the postimage blob, then we can take a shortcut and never
even look at the binary diff at all (e.g., this can happen
when rebasing changes within a repository).
If we don't have the postimage blob, though, we try to look
at the actual fragments, of which there are none, and get a
segfault. This patch checks explicitly for that case and
complains to the user instead of segfaulting. We need to
keep the check at a low level so that the "shortcut" case
above continues to work.
We also add a test that demonstrates the segfault. While
we're at it, let's also explicitly test the shortcut case.
Reported-by: Rafaël Carré <rafael.carre@gmail.com>
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
builtin/apply.c | patch | blob | history | |
t/t4103-apply-binary.sh | patch | blob | history |
diff --git a/builtin/apply.c b/builtin/apply.c
index 23c18c573b9147b7aaa88a8d1d2f26748e803501..f051e66dcc5260e4e7ecb28c36e2c6da8e823d6b 100644 (file)
--- a/builtin/apply.c
+++ b/builtin/apply.c
unsigned long len;
void *dst;
+ if (!fragment)
+ return error("missing binary patch data for '%s'",
+ patch->new_name ?
+ patch->new_name :
+ patch->old_name);
+
/* Binary patch is irreversible without the optional second hunk */
if (apply_in_reverse) {
if (!fragment->next)
index 9692f16f3581f261c4c10a29f03751990adb7897..08ad6d8b9e6b2869bfd2b574779380895709972d 100755 (executable)
--- a/t/t4103-apply-binary.sh
+++ b/t/t4103-apply-binary.sh
git diff-tree -p -C master binary >C.diff &&
git diff-tree -p --binary master binary >BF.diff &&
- git diff-tree -p --binary -C master binary >CF.diff
+ git diff-tree -p --binary -C master binary >CF.diff &&
+
+ git diff-tree -p --full-index master binary >B-index.diff &&
+ git diff-tree -p -C --full-index master binary >C-index.diff &&
+
+ git init other-repo &&
+ (cd other-repo &&
+ git fetch .. master &&
+ git reset --hard FETCH_HEAD
+ )
"
test_expect_success 'stat binary diff -- should not fail.' \
'do_reset &&
test_must_fail git apply --index C.diff'
+test_expect_success 'apply binary diff with full-index' '
+ do_reset &&
+ git apply B-index.diff
+'
+
+test_expect_success 'apply binary diff with full-index (copy)' '
+ do_reset &&
+ git apply C-index.diff
+'
+
+test_expect_success 'apply full-index binary diff in new repo' '
+ (cd other-repo &&
+ do_reset &&
+ test_must_fail git apply ../B-index.diff)
+'
+
test_expect_success 'apply binary diff without replacement.' \
'do_reset &&
git apply BF.diff'