[libvirt] [PATCH 4/5] docs: api_extension: Remove note to extend libvirt-TCK

Erik Skultety eskultet at redhat.com
Thu Aug 23 14:17:14 UTC 2018


On Thu, Aug 23, 2018 at 03:59:26PM +0200, Peter Krempa wrote:
> On Thu, Aug 23, 2018 at 15:25:17 +0200, Erik Skultety wrote:
> > On Thu, Aug 23, 2018 at 10:50:33AM +0200, Peter Krempa wrote:
> > > It probably does not make sense to do tests using that toolkit any more.
> >
> > At first I thought so too, I even thought that there hasn't been any commits to
> > the repo for a long time, it turned out I was wrong, there's still some work
> > going on, so I don't think we should abandon it completely. The fact that I
> > (and most probably you too - sorry, I just assumed your testing habits) don't
> > use it for testing doesn't mean, we should discourage others from giving the
> > framework a try, we're not mandating the usage so I don't see an issue with
> > the. With that said, once we have a proper continuous integration running
> > upstream, then I'd say we should start encouraging users to use the avocado
> > framework instead. However, for the time being, it's a NACK from me, maybe Dan
> > has a different feeling about this.
>
> Since the last non-maintenance commit for this section was in 2013, I
> don't think that mentioning any particular project makes sense. Avocado
> may become obsolete in another 5 years and I doubt that anybody will
> bother to update this document for the new hip project's name.

Everything will eventually become out-dated and by following that logic, the
only documentation we'd be left with is the code - "Documentation done
right". I disagree, since as I pointed out, we're not forcing anyone, we're
merely mentioning that such a thing still exists.

>
> I'm also okay just dropping this and making any poor soul which will
> follow that document to resolve any obsolete part by themselves.

Yep, let them make the decision for themselves, it's more than likely that
a thought of using it won't even cross their minds, fine by me.

Erik




More information about the libvir-list mailing list