[Libguestfs] New Rust bindings for nbdkit

alan somers asomers at gmail.com
Mon Jun 15 14:54:35 UTC 2020


On Mon, Jun 15, 2020 at 8:30 AM Daniel P. Berrangé <berrange at redhat.com>
wrote:

> On Mon, Jun 15, 2020 at 08:11:47AM -0600, alan somers wrote:
> > On Mon, Jun 15, 2020 at 8:04 AM Daniel P. Berrangé <berrange at redhat.com>
> > wrote:
> >
> > > On Mon, Jun 15, 2020 at 02:22:32PM +0100, Richard W.M. Jones wrote:
> > > > On Thu, Jun 11, 2020 at 04:19:08PM -0600, alan somers wrote:
> > > > > The existing Rust bindings for nbdkit aren't very idiomatic Rust,
> and
> > > they
> > > > > are missing a lot of features.  So I've rewritten them.  The new
> > > bindings
> > > > > aren't backwards compatible, but I doubt that's a problem.  Most
> > > likely,
> > > > > nobody has tried to use them yet, since the crate hasn't even
> > > published to
> > > > > crates.io.  Please review the attached patch.
> > > > > -Alan
> > >
> > > > Other issues:
> > > >
> > > >  * The license removed this clause:
> > > >
> > > >    -// * Neither the name of Red Hat nor the names of its
> contributors
> > > may be
> > > >    -// used to endorse or promote products derived from this software
> > > without
> > > >    -// specific prior written permission.
> > > >
> > > >    I believe this removal simply makes the license even more
> > > >    permissive, so that's fine.  However I will check with our legal
> > > >    people.  Also you should add license headers to the new files
> > > >    plugins/rust/tests/*.rs.  Essentially every file should have a
> > > >    license, and correct licensing is very important to us.
> > >
> > > This change is replacing 3-clause BSD with 2-clause BSD. Shouldn't
> cuase
> > > any actual difference for consumers, but seems like a needless change
> to
> > > be making.
> >
> > My reasoning was that the 2-clause license is preferred for new code (at
> > least in every other community where I've active), and this plugin is new
> > code.  But I can add the 3rd clause back if libguestfs likes it.
>
> I don't have any opinion on which variant is better.
>
> The important question is whether this new Rust impl was a completely clean
> room impl, or whether it started from existing code & then adapted it. If
> the
> latter, then changing the license is bad practice, as is removing existing
> Copyright statements.
>
> Regards,
> Daniel
>

It's not a full clean-room implemention like Compaq's BIOS; I did read the
original.  However, the new plugin contains nothing of the original except
for fragments of struct definitions.
-Alan
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listman.redhat.com/archives/libguestfs/attachments/20200615/086535b5/attachment.htm>


More information about the Libguestfs mailing list