OK florian
On 2023-07-19 13:17 +02, Theo Buehler <t...@theobuehler.org> wrote:
> The addition of unveil broke interactive mode since ask() assumes the
> default answer if it fails to open _PATH_TTY. Questions are only asked
> if neither force nor batch mode is activated, so condition on those.
>
> It seams cleaner to do unveil _PATH_TTY than to add a tty pledge since
> as far as I can see no ioctls on /dev/tty are emitted.
>
> Without this patch:
>
> $ cd usr.bin
> $ patch -Rs < /path/to/patch.below
> File to patch:
> No file found--skip this patch? [y]
> 1 out of 1 hunks ignored--saving rejects to Oops.rej
>
> With this patch:
>
> $ cd usr.bin
> $ patch -Rs < /path/to/patch.below
> File to patch: patch/patch.c
> Unreversed (or previously applied) patch detected! Ignore -R? [y] y
>
> Obviously the unveil(".", rwc) places restrictions on the path that can
> be given, but that's expected.
>
> Index: patch.c
> ===================================================================
> RCS file: /cvs/src/usr.bin/patch/patch.c,v
> retrieving revision 1.73
> diff -u -p -4 -r1.73 patch.c
> --- patch.c 15 Jul 2023 10:42:54 -0000 1.73
> +++ patch.c 19 Jul 2023 11:09:37 -0000
> @@ -222,8 +222,13 @@ main(int argc, char *argv[])
> if (unveil(filearg[1], "r") == -1) {
> perror("unveil");
> my_exit(2);
> }
> + if (!force && !batch)
> + if (unveil(_PATH_TTY, "rw") == -1) {
> + perror("unveil");
> + my_exit(2);
> + }
> if (unveil(".", "rwc") == -1) {
> perror("unveil");
> my_exit(2);
> }
>
--
In my defence, I have been left unsupervised.
