Comment 4 for bug 1959054

I'm working on patching debhelper (brushing up on some *very* rusty perl knowledge!) as some of the comments upstream indicate that as many as 40+ packages may be affected and I don't see it as viable to work around every one of them. However, first I want to understand exactly what the patch that caused the breakage was doing to make sure that in reverting it I don't cause more breakage.

My current thinking (subject to change!) is that the patch was solving a legitimate, and quite important issue (https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=989155) but that the method of the solution is incorrect (or at least problematic). I *think* (not implemented / tested yet) that it's preferable to solve the *original* problem (which led to the problematic patch) with the introduction of a preinst hook to handle the restart, which would avoid having to fiddle with the start actions introduced by dh_installinit and dh_installsystemd. Anyway, that's where I am at the moment, but I've got some Pi bits to sort out before I can get back to digging further and testing some tentative solutions.