amalyali / xspec-issues

scripts triggering pyXspec bugs

Bug reports for pyxspec

getflux.py

The model count rate is zero after fakeit, but not when using just a model.

run with:

python3 getflux.py

twosourcemodel.py

This sets up two sources, one going through the response, one not (diagonal dummy response). The two sources are assigned to the background spectrum and the source spectrum.

The bug arises when the background data group is assigned background sources and responses. This somehow changes the response for the source data group.

run with:

python3 twosourcemodel.py workaround  # gives no error

python3 twosourcemodel.py  # gives error

Additional nitpicks:

• It is very cumbersome to set this (pretty common) scenario up. Is there a better way?
• Setting to a non-existing response file triggers a command line question that cannot be aborted.

Responses used

The RMF/ARF/background are arbitrary. In this case, they are taken from

https://wiki.mpe.mpg.de/eRosita/erocalib_calibration https://wiki.mpe.mpg.de/eRosita/ScienceRelatedStuff/Background

These correspond to pre-flight estimations of eROSITA. They are not the real performance!