On Thu, 20 Jul 2017 11:28:11 +0200 ludo@gnu.org (Ludovic Courtès) wrote: > Heya Christopher, > > Christopher Baines skribis: > > > On 07/06/17 14:01, Ludovic Courtès wrote: > >> Christopher Baines skribis: > >> > >>> * gnu/packages/gnome.scm (deja-dup): New variable. > >> > >> There’s a couple of lint warnings fixed by the patch below: > > [...] > > >> Regarding propagated inputs: > >> > >>> + (propagated-inputs > >>> + `(("gsettings-desktop-schemas" ,gsettings-desktop-schemas) > >>> + ("gobject-introspection" ,gobject-introspection) > >>> + ("duplicity" ,duplicity) > >>> + ("python" ,python2-minimal) > >>> + ("python-pygobject" ,python2-pygobject))) > >> > >> Could you add a comment explaining why all this needs to be > >> propagated? If there’s a way to avoid it, it’s even better (for > >> instance, if deja-dup execs duplicity, we could hard-code the > >> absolute file name of duplicity.) > > > > Thanks for your review Ludo, hopefully I can get around to looking > > at deja-dup again at the weekend :) > > It’s been a long week-end. ;-) Could you take a look again so we can > save this patch from bitrot? Yep, its so long ago, I forget what I ended up doing that weekend! I've sent another patch now. I think vala and dbus may have moved in the mean time, so this one includes a patch from upstream to fix compilation issues with vala 0.36, and the check stage had to be disabled, as the tests no longer pass due to an issue with DBus. As for the propagated-inputs, I made them all inputs, and tested the application, and I couldn't see any problems.