W dniu 1/8/2020 o 14:19, Eric Blake pisze:
On 1/7/20 4:13 AM, Adrian Ambrożewicz wrote:
> Actual variable holding content length is int64_t, but it was assigned
> by explicit cast to size_t. On 32-bit systems it's a lossy conversion,
> so it was replaced by casting to int64_t instead.
>
> Signed-off-by: Adrian Ambrożewicz <adrian.ambrozewicz(a)linux.intel.com>
> ---
> plugins/curl/curl.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/plugins/curl/curl.c b/plugins/curl/curl.c
> index 031bd32..fe1330e 100644
> --- a/plugins/curl/curl.c
> +++ b/plugins/curl/curl.c
> @@ -389,7 +389,7 @@ curl_open (int readonly)
> goto err;
> }
>
> - h->exportsize = (size_t) d;
> + h->exportsize = (int64_t) d;
Why is a cast needed at all? This is C, an implicit conversion works
just as well.
I agree. It was just me being careful with scope of changes :) Should
I
send new patch? And another question - who is the maintainer and how
should I proceed with patch after acceptance on mailing list?
> nbdkit_debug ("content length: %" PRIi64, h->exportsize);
>
> if (strncasecmp (url, "http://", strlen ("http://")) == 0 ||