On 22-09-03 04:55PM, Omar Polo wrote:
> On 2022/09/03 15:37:31 +0200, Stefan Sperling <stsp@stsp.name> wrote:
> > On Sat, Sep 03, 2022 at 11:24:47PM +1000, Mark Jamsek wrote:
> > > This was discussed a while back on IRC to handle such things like:
> > > 
> > >   $ tog < /dev/null
> > > 
> > > ok?
> > 
> > Should 'return 1' or exit(1) on error, not return GOT_ERR_IO.
> 
> given the place where that check is, I'd go with an
> 
> 	errx(1, "stardand input is not a tty");
> 
> otherwise ok for me!

Thanks, stsp and op! Done.

diff refs/heads/main refs/heads/fix/tty
commit - 3d8e0c5ede1a5654397b63a9f483d875543527d5
commit + 7f04ea5dca27cc3e9751803d572f6792dde836d1
blob - 9d57af707725df8ed20c18c242b8359ad3a3270c
blob + 5b813c5e0de32f5883d2349934fb4b0280daa2ed
--- tog/tog.c
+++ tog/tog.c
@@ -8478,6 +8478,9 @@ main(int argc, char *argv[])
 	};
 	char *diff_algo_str = NULL;
 
+	if (!isatty(STDIN_FILENO))
+		errx(1, "stardand input is not a tty");
+
 	setlocale(LC_CTYPE, "");
 
 	while ((ch = getopt_long(argc, argv, "+hV", longopts, NULL)) != -1) {

-- 
Mark Jamsek <fnc.bsdbox.org>
GPG: F2FF 13DE 6A06 C471 CA80  E6E2 2930 DC66 86EE CF68