Web lists-archives.com

Re: [PATCH] Fixed pervasive enumeration warning in convert.h.




randall.s.becker@xxxxxxxxxx writes:

> From: "Randall S. Becker" <rsbecker@xxxxxxxxxxxxx>
>
> The actual enum value should be used rather than 0 and was causing
> a warning in an inline proc.
>
> Signed-off-by: Randall S. Becker <rsbecker@xxxxxxxxxxxxx>
> ---
>  convert.h | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/convert.h b/convert.h
> index 4f2da22..a9a6658 100644
> --- a/convert.h
> +++ b/convert.h
> @@ -79,7 +79,7 @@ extern int renormalize_buffer(const struct index_state *istate,
>  static inline int would_convert_to_git(const struct index_state *istate,
>  				       const char *path)
>  {
> -	return convert_to_git(istate, path, NULL, 0, NULL, 0);
> +	return convert_to_git(istate, path, NULL, 0, NULL, SAFE_CRLF_FALSE);
>  }

I think this is being solved a bit differently with a1fbf854
("convert_to_git(): safe_crlf/checksafe becomes int conv_flags",
2018-01-06), and 0 becomes the right value to pass at this caller to
say "I am passing none of the flag bit".

I am hoping that the series that ends at f3b11d54 ("convert: add
support for 'checkout-encoding' attribute", 2018-01-06) will be
rerolled and hit 'master' early in the next cycle.

Thanks.