On Sat, 2005-04-16 at 17:40 +0200, Richard Bos wrote:
Funny to read what is stated above. It's clear that you don't understand how it all works:
What's to be confused about? I've run SuSE since 8.2, and never seen my RPM database get screwed up until I tried to install some packages with apt. I've just done an `rpm --rebuilddb', and successfully installed the packages I wanted, so I'm past the problem, but the whole thing makes me very nervous. You claim there's no difference how apt, red-carpet, or rpm work under the covers, yet, as I pointed out, I've used red-carpet many times, and not had a problem, and I've installed packman's xine packages many times without problems. It is only after using apt to do so that I've encountered this problem. Despite what you say, it looks to me like it's apt's fault that my RPM database got corrupted. If it's true that apt4rpm is *nothing* more than a frontend to rpm, then I guess I don't see the point of it. Yes, I've been confused as to how to satisfy RPM dependencies in the past, but that's been back in the RedHat 5.2 and 6.0 days. I'm well past that now, and only rarely have to go digging to see what RPM I need to satisfy some strange dependency. If apt is *only* getting me around this, and there's even a *chance* that I could hose my RPM database permanently, it's not worth the risk to me. You know what strikes me about this conversation? That YaST takes care of dependencies just fine, thank you very much. Why don't the guys putting all of this effort into packages built for this apt4rpm project just make YaST repositories? It *seems* that the project is focused on SuSE; why not create the stuff in the native SuSE way of handling software installs? Why hassle with the extra learning curve of apt when the native tools will suffice? Regards, dk