Re: [Quilt-dev] Debian Bug#358875: quilt should fail if the series file references a non existent patch

[Jean Delvare]

Hi Martin,

> I got this from the Debian BTS. It sounds like a good idea to me, and the
> patch sounds good also.
> 
> What do you guys think of it?

Looks good to me, modulo the suggestions below.

> diff -rauN ../work2/quilt-0.44/quilt/push.in ./quilt-0.44/quilt/push.in
> --- ../work2/quilt-0.44/quilt/push.in 2006-03-24 21:38:49.000000000 +0100
> +++ ./quilt-0.44/quilt/push.in        2006-03-24 23:12:23.000000000 +0100
> @@ -183,6 +183,13 @@
>               no_reject_files="-r $tmp"
>       fi
>  
> +     if ! [ -e $patch_file ] && [ -z "$opt_force" ]

Isn't this more efficiently written that way?

        if [ ! -e "$patch_file" -a -z "$opt_force" ]

> +     then
> +             printf $"Patch %s does not exist\n" \
> +                    "$(print_patch $patch)"

That's an error message, so it should go to stderr rather than stdout.
Also, it could mention that -f will enforce the push (same we do for
other commands, for example refresh.)

> +             exit 1

"return 1" would fit better - although I agree it doesn't actually
change anything at the moment.

> +     fi
> +
>       apply_patch $patch "$patch_file"
>       status=$?
>       trap "" SIGINT
> @@ -206,6 +213,7 @@
>  
>               if ! [ -e $patch_file ]
>               then
> +                     # only if [ -n "$opt_force" ]
>                       printf $"Patch %s does not exist; applied empty 
> patch\n" \
>                              "$(print_patch $patch)"
>               elif [ -z "$(shopt -s nullglob ; echo "$QUILT_PC/$patch/"*)" ]

This comment doesn't look too useful to me.

When you apply this patch, please update and extend test/missing.test
accordingly.

Thanks,
-- 
Jean Delvare