On Fri, Jan 05, 2018 at 07:52:07AM +0000, KatolaZ wrote: > elogind has been in Devuan's radar for quite a while, and I guess will > become a priority just after ASCII gets out. I have not studied it in > depth, but my understanding is that the most promising path would be > to let it become a drop-in replacement to systemd-logind, e.g. by > letting it Provides: SOMETHING. This is somwhow different from the > case of udev/eudev since there is no independent Debian package (so > far) for logind, and I don't know if there is a specific capability > that can be used in a Provides: . Maybe you have a better knowledge on > that.
Simon McVittie said: # It looks as though elogind is a fork of systemd-logind with reduced # functionality, no dependency on systemd as pid 1, and logind's D-Bus API # (so, basically systemd-shim done right), so it should be possible for # most of those to talk to elogind's logind-compatible API without code # changes (via libsystemd, even). Now that we have versioned Provides, # one way to achieve that might be for implementations of the logind API # to add Provides: logind (= v) where v is the version of systemd whose # logind API is implemented (currently 219 for elogind and 236 for systemd), # and for depending packages to depend on libpam-systemd (>= v) | logind # (>= v), or even on default-logind | logind (>= v) (with default-logind # provided by libpam-systemd on Debian) to be nice to anti-systemd # derivatives. Obviously >= v can be omitted if recent logind features # are not required. which sounds like a good solution to me. Meow! -- // If you believe in so-called "intellectual property", please immediately // cease using counterfeit alphabets. Instead, contact the nearest temple // of Amon, whose priests will provide you with scribal services for all // your writing needs, for Reasonable And Non-Discriminatory prices. _______________________________________________ Dng mailing list Dng@lists.dyne.org https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng
