[Rivet] Rivet and AGILe released... phew!

[Andy Buckley]

(copying to the list for general information)

Lars Sonnenschein wrote:
> Hello Andy
>
> after installation at CERN I see the following generator options:
> 
> rivetgun  {-g <CCPythia|FHerwig:6510|FHerwigJimmy:6510
>                        |FPythia:6413>|-G <genfile>} [-R]
> 
> (so at least Pythia8 ia also linkable)

All that tells you is that libpythia8 is dlopenable, not that it will
work... that list is produced by getAvailableGens() in Loader.cc.

> charybdis is installed at CERN, as well as alpgen,
> but AGILe is looking for libcharybdisher (Charybdis with Herwig)
> and libalppyt (Alpgen with Pythia)
> It is not clear to me how this is supposed to work ...

What's the problem? Does Genser not provide both Charybdis versions? If
not, check how Genser expect Charybdis to be used.

> When I try to link and run e.g. the Fortran Pythia
> library as it is installed at CERN right now
> I get a segmentation violation after 
> initialization of the event generator.

It works for me! Any debug output? What do you mean by "link and run"?
rivetgun doesn't *link* against any generators.

>>From your email to Doug I assume that it should be possible
> to generate events with the event generator libraries as 
> presently installed at CERN?

Yes, it works for me. Did you use the CERN bootstrapping script? The
Genser versions of Pythia and Herwig are broken because of the Genser
build process. This means that you need to add extra (bogus) libraries
to the Rivet build to make them work. The bootstrap script does this if
it detects that it's trying to use the Genser versions, but it's not the
easiest thing to do by hand (and definitely shouldn't be added to the
Rivet makefiles)

Andy