Re: [PATCH 1/5] pack-objects: make variable non-static

[Taylor Blau <me@ttaylorr.com>]

On Thu, Oct 24, 2024 at 11:08:40AM -0700, Jonathan Tan wrote:
> Signed-off-by: Jonathan Tan <jonathantanmy@google.com>
> ---
>  builtin/pack-objects.c | 3 +--
>  1 file changed, 1 insertion(+), 2 deletions(-)
>
> diff --git a/builtin/pack-objects.c b/builtin/pack-objects.c
> index 0fc0680b40..e15fbaeb21 100644
> --- a/builtin/pack-objects.c
> +++ b/builtin/pack-objects.c
> @@ -238,8 +238,6 @@ static enum {
>  } write_bitmap_index;
>  static uint16_t write_bitmap_options = BITMAP_OPT_HASH_CACHE;
>
> -static int exclude_promisor_objects;
> -
>  static int use_delta_islands;
>
>  static unsigned long delta_cache_size = 0;
> @@ -4327,6 +4325,7 @@ int cmd_pack_objects(int argc,
>  	struct string_list keep_pack_list = STRING_LIST_INIT_NODUP;
>  	struct list_objects_filter_options filter_options =
>  		LIST_OBJECTS_FILTER_INIT;
> +	int exclude_promisor_objects = 0;
>
>  	struct option pack_objects_options[] = {
>  		OPT_CALLBACK_F('q', "quiet", &progress, NULL,
> --
> 2.47.0.163.g1226f6d8fa-goog

This patch appears to conflict with ds/path-walk, which wants to read
the exclude_promisor_objects variable from outside of cmd_pack_objects()
(but elsewhere within the builtin/pack-objects.c compilation unit).

Is this refactoring a necessary step, or just cleanup? If the former, it
may be good for you and Stolee (CC'd) to work together to figure out how
to eliminate the conflict from your two series. If the latter, it may be
worth dropping this patch.

Thanks,
Taylor