Open drzraf opened 2 weeks ago
mlat-client having a non-deprecated way to build would be nice. (and i find the whole python build stuff very confusing when i tried touching it)
readsb being a deb is an issue unless you make the package / binary name adsb.lol specific. different projects don't want their installed binary messed with by another project.
Considering that you'd still build the debs during install, you still have a shell script users need to run. It installs more stuff to build the debs (that's a minor drawkback i suppose).
Dealing with debian packaging sounds like work from a maintenance perspective actually.
Really making sure this works after lots of changes would be a lot of work. The number of users who care about this and don't just use one of the docker containers is likely pretty small. And for 99% of users it's just as much of a black box after the proposed changes.
git clone = .deb Are you suggesting to actually have a deb for download? With all the architectures / system versions that sounds like a nightmare.
I get the intent, but i'm doubting the actual benefit.
One note, there is this linked in the readme: https://github.com/adsblol/feed?tab=readme-ov-file#manual-feeding-with-readsb-and-mlat-client So this is an alternative for people who don't want to run the scripts and don't want to run docker stuff.
Follow-up of https://github.com/ADSBexchange/feedclient/issues/16 which applies as well to the present
master
branch (which sounds like a fork but started as a copy, isn't?)apt-get
s =Depends
git clone
=.deb
compile readsb
: Should be another.deb
mlat-client
: Should be a pip / pipxDoes it sound desirable from a long-term maintenance perspective?