Comment # 14 on bug 980068 from Christopher Yeleighton

(In reply to Wolfgang Bauer from comment #13)
> (In reply to Christopher Yeleighton from comment #12)
> > (In reply to Wolfgang Bauer from comment #11)
> > > (In reply to Christopher Yeleighton from comment #10)
> > > > > konversation doesn't link nor load libkdeinit_khelpcenter5 at all (upstream).
> > > > 
> > > > So where does the error message come from?
> > > 
> > > What error message?
> > 
> > Cannot load libkdeinit5_khelpcenter5.so: File not found.
> 
> And what are you doing exactly to get that error message?

I tell Konversation to show the handbook.

> 
> When I try to open the manual in Konversation (Help->Konversation Handbook)
> without khelpcenter5 installed, I get this:
> KDEInit could not launch khelpcenter5
> (no reference at all to libkdeinit5_khelpcenter5.so or konversation trying
> to load it)

Me too, but this message comes later.

> 
> And that's coming from susehelp, when it tries to run khelpcenter5.
> As I already wrote twice, uninstalling susehelp should make Frameworks open
> a web browser as fallback (and would also get rid of this error message).
> Why don't you at least try that if you don't believe me?

I am more interested in KDE handbook being broken in Leap than not working on
my workstation, otherwise I would just install khelpcenter5.

> And even if konversation or the Frameworks would load
> libkdeinit5_khelpcenter5.so on runtime, it would be much easier to just add
> a package dependency manually than patching the code to load it on
> "load-time" instead. (And actually you claimed yourself that the upstream
> code should open a web browser if libkdeinit5_khelpcenter5.so cannot be
> found, so this should not be necessary at all then anyway...)

I got the impression that khelpcenter5 is required on Leap from what you said,
and if it is required, there is no point in deferring loading the library. 
Adding a dependency manually would be inappropriate in this situation.

But I understand that you have backed off now (in comment #9), so perhaps I can
withdraw this suggestion too.

> 
> You're just totally overcomplicating (and maybe misunderstanding) things
> here.

Maybe I am, but we are failing to provide a decent product to the user, which
is much more serious :-(