GammaLib

To be able to test energy resolution (#1036) (a future step) having a line-like spectral model is important.

The Fermi Science tools have a Gaussian: http://fermi.gsfc.nasa.gov/ssc/data/analysis/scitools/source_models.html#Gaussian

In the HESS software we have a TopHat:

<pre>
/*! \class TopHat
A "top-hat" FitFunction.
\f[ f(E) = C \frac{1}{2} \left(Erf\left(\frac{\ln(E)-\mu_{On}}{\sigma_{On}}\right)+1\right)
\left(1- \frac{1}{2} \left(Erf\left(\frac{\ln(E)-\mu_{Off}}{\sigma_{Off}}\right)+1\right) \f]
Parameters :
- #0 "Norm" C
- #1 "OnLoc" \f$\mu_{On}\f$
- #2 "OnWidth" \f$\sigma_{On}\f$
- #3 "OffLoc" \f$\mu_{Off}\f$
- #4 "OffWidth" \f$\sigma_{Off}\f$
*/
</pre>

If no-one gets to it before, I could implemented and test this at the January coding sprint.