Event source is not compatible with data produced by latest EVB where number of gains can depend on event type

[sizun]

In the current NectarCAMEventSource, the number of gains n_gains is an argument constructor and is assumed to be the same for all events. Also it could be derived from the source.camera_config.nectarcam.algorithms parameter.

Since switching to version 4 of the event builder, this parameter is no longer filled as gain selection is enabled or not depending on the event type.

Maybe we should look how LSTEventSource handled the transition.

[maxnoe]

See https://github.com/cta-observatory/ctapipe_io_lst/blob/e78acdc43ef4c4c95c83ec5100b7c6373af9df86/ctapipe_io_lst/__init__.py#L675

[jlenain]

Just to check, I think this issue has been addressed and is obsolete now, is that correct ?

[maxnoe]

Looking at the code in __init__.py, it seems like you correctly check on an event-by-event basis if the event was gain selected or not

[vmarandon]

Yes, but I would love to try on realistic data before closing this issue :-)
(Including mixture of high-gain pixels and low-gain selected pixels)