Re: [PATCH 1/9] fetch: drop unused DISPLAY_FORMAT_UNKNOWN enum value

From: Jeff King <peff@peff.net>
Date: May 19, 2023, 00:13 UTC
Message-ID: <20230519001309.GA2442034@coredump.intra.peff.net>
In-Reply-To: <9417a62e9bbd30d9350b65e5396a31f9cb7c60b6.1684324059.git.ps@pks.im>

On Wed, May 17, 2023 at 01:48:42PM +0200, Patrick Steinhardt wrote:

Show 5 quoted lines

> With 50957937f9 (fetch: introduce `display_format` enum, 2023-05-10), a
> new enumeration was introduced to determine the display format that is
> to be used by git-fetch(1). The `DISPLAY_FORMAT_UNKNOWN` value isn't
> ever used though, and neither do we rely on the explicit `0` value for
> initialization anywhere.

To be slightly pedantic, we'd also want to make sure the we do not rely on the zero value for reading, like:

  if (display_state->format)
     ....

But having looked over the code, we don't (it's always a switch or equality with a known name), so this is safe to do.

Thanks for cleaning this up.

Show 12 quoted lines

> diff --git a/builtin/fetch.c b/builtin/fetch.c
> index 849a9be421..9147b700e5 100644
> --- a/builtin/fetch.c
> +++ b/builtin/fetch.c
> @@ -58,7 +58,6 @@ enum {
>  };
>  
>  enum display_format {
> -	DISPLAY_FORMAT_UNKNOWN = 0,
>  	DISPLAY_FORMAT_FULL,
>  	DISPLAY_FORMAT_COMPACT,
>  	DISPLAY_FORMAT_PORCELAIN,

Just for similar situations in the future, I think we could do:

  DISPLAY_FORMAT_FULL = 1,

if we were worried about keeping the zero-behavior the same for existing callers. But given how new and how limited this code is, I feel confident that we've checked all of the code, and what you've written above is preferable.

-Peff