Hi Jean,

On 2012/11/23, at 2:31, Jean Delvare <[email protected]> wrote:

> The timestamps recorded on file systems have a limited resolution.
> For this reason, a patch or file with a timestamp exactly equal to
> .timestamp should be considered as more recent than .timestamp, i.e.
> potentially modified since the last push or refresh. This is
> particularly important for the test suite, as we can do many actions
> in the same second there.

It looks good to me.

Reviewed-by: Satoru Takeuchi <[email protected]>

> ---
> quilt/pop.in |    4 ++--
> 1 file changed, 2 insertions(+), 2 deletions(-)
> 
> --- a/quilt/pop.in
> +++ b/quilt/pop.in
> @@ -70,14 +70,14 @@ files_may_have_changed()
> 
>    if [ $? -ne 0 -o ! -e "$patch_file" \
>         -o ! -e "$QUILT_PC/$patch/.timestamp" \
> -         -o "$QUILT_PC/$patch/.timestamp" -ot "$patch_file" ]
> +         -o ! "$QUILT_PC/$patch/.timestamp" -nt "$patch_file" ]
>    then
>        return 0
>    fi
> 
>    for file in $(files_in_patch $patch)
>    do
> -        [ "$QUILT_PC/$patch/.timestamp" -ot $file ] && return 0
> +        [ ! "$QUILT_PC/$patch/.timestamp" -nt $file ] && return 0
>    done
>    return 1
> }
> 
> -- 
> Jean Delvare
> Suse L3
> 
> 
> _______________________________________________
> Quilt-dev mailing list
> [email protected]
> https://lists.nongnu.org/mailman/listinfo/quilt-dev

_______________________________________________
Quilt-dev mailing list
[email protected]
https://lists.nongnu.org/mailman/listinfo/quilt-dev

Reply via email to