"git push --recurse-submodules=on-demand" adds each submodule as an
alternate with add_submodule_odb before checking whether the
submodule has anything to push and pushing it if so.
However, it never accesses any objects from the submodule. In the
parent process it uses the submodule's ref database to see if there
is anything to push. The actual push (which does rely on objects)
occurs in a child process.
The same was true when this call was originally added in
v1.7.11-rc0~111^2 (push: teach --recurse-submodules the on-demand
option, 2012-03-29). Most likely it was added by analogy with
fetch --recurse-submodules=on-demand, which did use the submodule's
object database.
Use is_submodule_populated_gently instead, which is simpler and
cheaper.
Signed-off-by: Stefan Beller <sbeller@xxxxxxxxxx>
---
Originally I intended to send this out as part of a larger series,
but the series is getting too large series, sending all things in smaller
units!
Thanks,
Stefan
submodule.c | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/submodule.c b/submodule.c
index 111a3007fc..e20216bc0f 100644
--- a/submodule.c
+++ b/submodule.c
@@ -966,7 +966,9 @@ static int push_submodule(const char *path,
const struct string_list *push_options,
int dry_run)
{
- if (add_submodule_odb(path))
+ int code;
+
+ if (!is_submodule_populated_gently(path, &code))
return 1;
if (for_each_remote_ref_submodule(path, has_remote, NULL) > 0) {
--
2.14.0.rc0.3.g6c2e499285
