Re: [PATCH 1/6] index-pack: factor out pack/idx finalization

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

 




On 15/03/17 21:27, Jeff King wrote:
> The procedure for moving the ".pack" and the ".idx" files
> into place is the same. Rather than repeat the code, let's
> factor it into a helper function.
> 
> This has the added benefit of clarifying the lifetime of
> "final_pack_name". In the original code it is sometimes
> redirected to our reusable "name" buffer; after we reuse
> that buffer, the contents are nonsense (but nobody looks at
> it, so this is not a bug, just a maintenance issue). In the
> new code, the pointer modification is contained in the
> helper function.
> 
> Signed-off-by: Jeff King <peff@xxxxxxxx>
> ---
>  builtin/index-pack.c | 40 +++++++++++++++++++---------------------
>  1 file changed, 19 insertions(+), 21 deletions(-)
> 
> diff --git a/builtin/index-pack.c b/builtin/index-pack.c
> index f4b87c6c9..187c0550c 100644
> --- a/builtin/index-pack.c
> +++ b/builtin/index-pack.c
> @@ -1380,6 +1380,23 @@ static void fix_unresolved_deltas(struct sha1file *f)
>  	free(sorted_by_pos);
>  }
>  
> +static void finalize_file(const char *final_name, const char *curr_name,
> +			  unsigned char *sha1, const char *ext)
> +{
> +	if (final_name != curr_name) {
> +		char name[PATH_MAX];
> +		if (!final_name) {
> +			snprintf(name, sizeof(name), "%s/pack/pack-%s.%s",
> +				 get_object_directory(), sha1_to_hex(sha1),
> +				 ext);
> +			final_name = name;
> +		}
> +		if (finalize_object_file(curr_name, final_name))
> +			die(_("cannot store %s file"), ext);
> +	} else if (from_stdin)
> +		chmod(final_name, 0444);
> +}
> +
>  static void final(const char *final_pack_name, const char *curr_pack_name,
>  		  const char *final_index_name, const char *curr_index_name,
>  		  const char *keep_name, const char *keep_msg,
> @@ -1422,27 +1439,8 @@ static void final(const char *final_pack_name, const char *curr_pack_name,
>  		}
>  	}
>  
> -	if (final_pack_name != curr_pack_name) {
> -		if (!final_pack_name) {
> -			snprintf(name, sizeof(name), "%s/pack/pack-%s.pack",
> -				 get_object_directory(), sha1_to_hex(sha1));
> -			final_pack_name = name;
> -		}
> -		if (finalize_object_file(curr_pack_name, final_pack_name))
> -			die(_("cannot store pack file"));
> -	} else if (from_stdin)
> -		chmod(final_pack_name, 0444);
> -
> -	if (final_index_name != curr_index_name) {
> -		if (!final_index_name) {
> -			snprintf(name, sizeof(name), "%s/pack/pack-%s.idx",
> -				 get_object_directory(), sha1_to_hex(sha1));
> -			final_index_name = name;
> -		}
> -		if (finalize_object_file(curr_index_name, final_index_name))
> -			die(_("cannot store index file"));
> -	} else
> -		chmod(final_index_name, 0444);

Is from_stdin always true if final_index_name == curr_index_name?
Was the original asymmetry deliberate?

> +	finalize_file(final_pack_name, curr_pack_name, sha1, "pack");
> +	finalize_file(final_index_name, curr_index_name, sha1, "idx");
>  
>  	if (!from_stdin) {
>  		printf("%s\n", sha1_to_hex(sha1));
> 

ATB,
Ramsay Jones




[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]