On Mon, Jul 05, 2021 at 04:20:06PM +0100, Christoph Hellwig wrote:
> On Fri, Jul 02, 2021 at 07:58:08PM -0700, Darrick J. Wong wrote:
> > From: Darrick J. Wong <djwong@xxxxxxxxxx>
> >
> > Add proper argument parsing to the funshare command so that when you
> > pass it nonexistent --help it will print the help instead of complaining
> > that it can't convert that to a number.
> >
> > Signed-off-by: Darrick J. Wong <djwong@xxxxxxxxxx>
> > ---
> > io/prealloc.c | 16 ++++++++++++----
> > 1 file changed, 12 insertions(+), 4 deletions(-)
> >
> >
> > diff --git a/io/prealloc.c b/io/prealloc.c
> > index 2ae8afe9..94cf326f 100644
> > --- a/io/prealloc.c
> > +++ b/io/prealloc.c
> > @@ -346,16 +346,24 @@ funshare_f(
> > char **argv)
> > {
> > xfs_flock64_t segment;
> > + int c;
> > int mode = FALLOC_FL_UNSHARE_RANGE;
> > - int index = 1;
> >
> > - if (!offset_length(argv[index], argv[index + 1], &segment)) {
> > + while ((c = getopt(argc, argv, "")) != EOF) {
> > + switch (c) {
> > + default:
> > + command_usage(&funshare_cmd);
> > + }
>
> Do we really need this switch boilerplate?
Sort of -- without it, you get interactions like:
$ xfs_io -c 'funshare 0 --help' /autoexec.bat
non-numeric length-argument -- --help
which is silly, since we could display the help screen.
The other solution might be to fix all the offset_length() callers to
emit command usage when the number parsing doesn't work, but that would
clutter up the error reporting when you try to feed it a number that
doesn't parse.
> > + }
> > + if (optind != argc - 2)
> > + return command_usage(&funshare_cmd);
>
> Spaces instead of tabs here.
Fixed.
--D
