On Friday, February 9, 2018 2:03:01 AM EST Alexander Kanavin wrote: > On 02/09/2018 12:13 AM, Mark Asselstine wrote: > > The use of libunique is non-consequential here. The issue is that the > > gtk-doc package has imports which are not going to be present via direct > > runtime dependencies. Any sw which attempts to run the python scripts > > found in gtk-doc will fail in the same way unless the python3-six is > > installed via an indirect dependency. I didn't have a chance to dig in to > > this deeper today so I don't have more to add than this but the "import > > six" is pretty explicit and will fail unless this script is not meant to > > actually be run anytime, anywhere. > gtk-doc recipe does have a direct dependency on python3-six, subject to > 'api-documentation' in DISTRO_FEATURES. The reason it's conditional is > that we don't want to build a ton of gtk-doc dependencies in the default > case that used by almost everyone. Yes, this produces gtk-doc scripts > that are not working, but they will not be run anyway, if gtk-doc > capable recipes inherit the gtk-doc class, which will disable the > gtk-doc feature. libunique was missing that inherit, the patch to fix > that is now on oe-devl list. Take it and the issue will go away.
Thanks for sending out the updates, I am giving them a go. I wasn't clear about the intention based on your initial change so glad I reached out to you and that this will move on. Thanks again, Mark > > Alex -- _______________________________________________ Openembedded-core mailing list Openembedded-core@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-core
