Libguestfs - Nov 2022 - [nbdkit PATCH] sh: Add exit status triggers for nbdkit_{shutdown, disconnect}

On Wed, Nov 02, 2022 at 02:15:45PM +0100, Laszlo Ersek
wrote:
> On 11/01/22 20:56, Eric Blake wrote:
> > Make it possible for the sh and eval plugins to disconnect a client or
> > shut down the entire nbdkit server by use of special return values.
> > Prior to this patch we had reserved 4-7 for future use; this defines
> > 4-6, and extends the set of reserved return values to 7-15.  We figure
> > it is unlikely that anyone is using status 8-15 with the intent that
> > it behaves identically to status 1.
> > 
> > For the testsuite, I only covered the eval plugin; but since it shares
> > common code with the sh plugin, both styles should work.
> > ---
> > 
> > Finally got the testsuite additions for this in a state that I like.
> > 
> >  plugins/sh/nbdkit-sh-plugin.pod |  37 ++++++-
> >  tests/Makefile.am               |   2 +
> >  plugins/sh/call.h               |   9 +-
> >  plugins/sh/call.c               |  85 +++++++--------
> >  tests/test-eval-disconnect.sh   | 185
++++++++++++++++++++++++++++++++
> >  5 files changed, 268 insertions(+), 50 deletions(-)
> >  create mode 100755 tests/test-eval-disconnect.sh
> > 
> > diff --git a/plugins/sh/nbdkit-sh-plugin.pod
b/plugins/sh/nbdkit-sh-plugin.pod
> > index 2a55fdc9..37139e1b 100644
> > --- a/plugins/sh/nbdkit-sh-plugin.pod
> > +++ b/plugins/sh/nbdkit-sh-plugin.pod
> > @@ -96,4 +96,4 @@ The script should exit with specific exit codes:
> > 
> >  The method was executed successfully.
> > 
> > -=item 1 and 8-127
> > +=item 1 and 16-255
> > 
> >  There was an error.  The script may print on stderr an errno name,
> >  optionally followed by whitespace and a message, for example:
> > @@ -123,9 +123,38 @@ The requested method is not supported by the
script.
> > 
> >  For methods which return booleans, this code indicates false.
> > 
> > -=item 4, 5, 6, 7
> > +=item S<4>
> 
> The S<> notation seems new here (so it's going to be inconsistent
with
> the rest of this file, I think).
I was going to mention this too.  The S<> notation is used to insert
non-breaking spaces (for output formats that support it) in a span of
text so that it won't be folded over multiple lines.  AFAIK it
shouldn't have any effect here.

For some reason this existing list uses:

  =item S<0>

but I think that must be a mistake.

Rich.

-- 
Richard Jones, Virtualization Group, Red Hat http://people.redhat.com/~rjones
Read my programming and virtualization blog: http://rwmj.wordpress.com
virt-top is 'top' for virtual machines.  Tiny program with many
powerful monitoring features, net stats, disk stats, logging, etc.
http://people.redhat.com/~rjones/virt-top

On Wed, Nov 02, 2022 at 01:36:26PM +0000, Richard W.M. Jones
wrote:
> > > -=item 4, 5, 6, 7
> > > +=item S<4>
> > 
> > The S<> notation seems new here (so it's going to be
inconsistent with
> > the rest of this file, I think).
> 
> I was going to mention this too.  The S<> notation is used to insert
> non-breaking spaces (for output formats that support it) in a span of
> text so that it won't be folded over multiple lines.  AFAIK it
> shouldn't have any effect here.
Ah, but it does:

Pod input around line 121: Expected text after =item, not a number

The use of S<> is there to keep the pod formatter happy when =item's
sole argument would otherwise look like a number instead of arbitrary
text.

-- 
Eric Blake, Principal Software Engineer
Red Hat, Inc.           +1-919-301-3266
Virtualization:  qemu.org | libvirt.org