Re: [Qemu-block] [PATCH v2] qemu-io: Reinitialize optind to 1 (not 0) before parsing inner command.

[Richard W.M. Jones]

On Tue, Jan 08, 2019 at 04:13:09PM +0100, Kevin Wolf wrote:
> Am 08.01.2019 um 15:51 hat Eric Blake geschrieben:
> > On 1/8/19 6:16 AM, Kevin Wolf wrote:
> > > Unconditionally setting optind = 1 looks fine. I would, however, quote a
> > > different part of the glibc man page (in addition or instead of the
> > > paragraph you already quoted):
> > > 
> > >     The  variable  optind is the index of the next element to be
> > >     processed in argv.  The system initializes this value to 1.  The
> > >     caller can reset it to 1 to restart scanning of the same argv, or
> > >     when scanning a new argument vector.
> > > 
> > > This makes it pretty clear that optind = 1 is fine for our case with
> > > glibc. The FreeBSD man page still suggests that we need optreset = 1, so
> > > I suppose we'd end up with something like:
> > > 
> > > ...
> > > optind = 1;
> > > #ifdef __FreeBSD__
> > > optreset = 1;
> > > #endif
> > 
> > If you really want to set optreset for BSD systems, I'd do a configure
> > probe for whether optreset exists, and if so set it for ALL platforms
> > that have optreset, not just for __FreeBSD__.  (That, and checkpatch.pl
> > will gripe if you don't do it that way).
> > 
> > But I'm leaning towards not bothering with optreset UNLESS someone
> > proves they have a case where it actually matters.
> 
> I don't mind either way as long as it works. Using optreset would be
> following the spec by the letter.
> 
> In fact, I had already applied this patch because it's correct even if
> possibly incomplete, depending on your interpretation, but I decided to
> reply instead because the commit message didn't really describe that
> optreset = 1 is correct for glibc, but that optreset = 0 is necessary in
> some other case (which is irrelevant here). So the commit message was my
> main point.

I have tested it[1] on:

Linux, glibc 2.28
FreeBSD 11.2-RELEASE
OpenBSD 6.4

Interestingly OpenBSD getopt works in this respect the same way as
Linux, so it doesn't need any fix.  It's only FreeBSD which needs the
fix.

Rich.

[1] "it" being v2 here:
https://lists.gnu.org/archive/html/qemu-devel/2019-01/msg00189.html

-- 
Richard Jones, Virtualization Group, Red Hat http://people.redhat.com/~rjones
Read my programming and virtualization blog: http://rwmj.wordpress.com
libguestfs lets you edit virtual machines.  Supports shell scripting,
bindings from many languages.  http://libguestfs.org