Re: [libvirt] [tck PATCH v2 2/4] new NetworkHelper function get_network_ip()

On 03/05/2018 04:31 AM, Daniel P. Berrangé wrote: > On Thu, Mar 01, 2018 at 09:49:58PM -0500, Laine Stump wrote: >> This function gets the first IP address for the named virtual >> network. It is returned as a Net::IP object, so that we will have info >> about its netmask/prefix and can easily get it broadcast address and >> perform arithmetic on the address. >> >> Signed-off-by: Laine Stump <laine(a)laine.org&gt; >> --- >> >> Change from V1: return a NetAddr::IP object instead of a string. >> >> lib/Sys/Virt/TCK/NetworkHelpers.pm | 22 ++++++++++++++++++++++ >> 1 file changed, 22 insertions(+) >> >> diff --git a/lib/Sys/Virt/TCK/NetworkHelpers.pm b/lib/Sys/Virt/TCK/NetworkHelpers.pm >> index 5f563e5..7bbce62 100644 >> --- a/lib/Sys/Virt/TCK/NetworkHelpers.pm >> +++ b/lib/Sys/Virt/TCK/NetworkHelpers.pm >> @@ -1,4 +1,5 @@ >> use Sys::Virt::TCK qw(xpath); >> +use NetAddr::IP qw(:lower); > This isn't part of base perl, so you'll need to list it in Build.PL and > the RPM spec file. I originally assumed that, but remembered seeing "something somewhere" about implicit dependencies and decided to try it out by not listing it in the specfile - on both Fedora and RHEL7 the dependency was properly pulled in and it was installed. This leads to one of three possibilities: 1) implicit dependencies are figured out properly by yum and dnf (at least for RHEL7, don't know about RHEL6). 2) (1), but it's just coincidentally happening and not guaranteed. 3) I wasn't paying attention when I tested, and what I say isn't actually true. I don't have any problem putting in the explicit Requires though. Can I assumed a Reviewed-by with that in place?