Ronnie Sahlberg <sahlberg@xxxxxxxxxx> writes:
> Create a new erfs.c file that will be used to hold all the refs
> code that is backend agnostic and will be shared across all backends.
>
> The reason we renamed everything to refs-be-files.c in the previous patch
> and now start moving the common code back to the new refs.c file
> instead of the other way around is the etive volumes of code.
Huh? Why not create refs-be-files.c and move whatever need to be
there over there, instead of rename the file and move things that
shouldn't have been moved back like this?
Puzzled.
I do not see 02/23 here, but I am assuming that is is just
git mv refs.c refs-be-files.c
which may have been a seven-line patch with "format-patch -M" ;-)
>
> With the ref_cache, packed refs and loose ref handling that are all
> part of the files based implementation the backend specific part
> of the old refs.c file is several times larger than the backend agnostic
> part. Therefore it makes more sense to first rename everything to be
> part of the files based backend and then move the parts that can be used
> as common code back to refs.c.
>
> Signed-off-by: Ronnie Sahlberg <sahlberg@xxxxxxxxxx>
> ---
> Makefile | 1 +
> refs.c | 3 +++
> 2 files changed, 4 insertions(+)
> create mode 100644 refs.c
>
> diff --git a/Makefile b/Makefile
> index e010ad1..937d22a 100644
> --- a/Makefile
> +++ b/Makefile
> @@ -857,6 +857,7 @@ LIB_OBJS += quote.o
> LIB_OBJS += reachable.o
> LIB_OBJS += read-cache.o
> LIB_OBJS += reflog-walk.o
> +LIB_OBJS += refs.o
> LIB_OBJS += refs-be-files.o
> LIB_OBJS += remote.o
> LIB_OBJS += replace_object.o
> diff --git a/refs.c b/refs.c
> new file mode 100644
> index 0000000..77492ff
> --- /dev/null
> +++ b/refs.c
> @@ -0,0 +1,3 @@
> +/*
> + * Common refs code for all backends.
> + */
--
To unsubscribe from this list: send the line "unsubscribe git" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html