I’m not sure where exactly to ask, this category seemed appropriate.
First of all, thanks for working on Turris, it’s a nice little toy
I finally had some time to try compile some turris packages. It went mostly smoothly (well, it took me a while to find out that you have to first enable the package in menuconfig, otherwise it won’t build). I have compiled myself duplicity http://duplicity.nongnu.org/. It’s working, but the Makefiles are not 100% refined to be pushed to upstream. Now what?
Should I try to push the packages to turris repo or to openwrt? I don’t know what testing is necessary for openwrt I only know it works for me on turris.
Duplicity is librsync + three or four python modules. Should I create separate makefiles for each module or put everything together in one makefile? Creating separate makefiles makes more sense to me.
If push the makefiles to turris repo, how can I know that there has been bug reported against it so that I can fix it?
I definitely would be interested in having duplicity on my router.
Should I try to push the packages to turris repo or to openwrt? I don’t know what testing is necessary for openwrt I only know it works for me on turris.
Since the big Openwrt/Lede split, I’m wondering which one should we think of as “upstream”.
In other words, would the maintainers pick up patches from either, just openwrt or just lede?
Hopefully it looks like OpenWRT and LEDE will merge again but regardless, I would say that the most sensible repo would be packages repo - https://github.com/openwrt/packages
I haven’t found duplicity to be available even on Turris OS 4 (MOX), is there a guide, how compile duplicty package for Turris (I’m using simple rsync now on Omnia and on MOX to protect data), but being able to use duplicity (or at least rspapshot) would be great.
