allow enums in containers and support in tableio

[dneise]

Connected slightly to #972

just, in case we want to use enum.Enum somewhere I thought we might want to be able to serialize them as well.

Not sure this is the best way to do it. I've read somewhere that HDF5 natively supports storing enumeration types, so that it deals with the back and forth trafo from the numeric value to the actual enum instance. But for the moment I did this in ctapipes own io layer.

[dneise]

Codacy is just complaining about invalid syntax again -> f-strings. Apart from that, I think it might be fine.

[maxnoe]

There is a class IntEnum in enum. Might this obliviate the need to write custom code to write it out?

[dneise]

Might this obliviate the need to write custom code to write it out?

Might ... but I think converting back into an IntEnum upon reading is impossible without storing the enum class in the meta data. And this must be done upon writing it, no? So what is the point?

[maxnoe]

Ah sure.

[codecov]

Codecov Report

Merging #978 into master will increase coverage by 2.92%.
The diff coverage is 100%.

@@            Coverage Diff             @@
##           master     #978      +/-   ##
==========================================
+ Coverage   80.44%   83.37%   +2.92%     
==========================================
  Files         193      189       -4     
  Lines       11051    10700     -351     
==========================================
+ Hits         8890     8921      +31     
+ Misses       2161     1779     -382

Impacted Files	Coverage Δ
ctapipe/io/hdf5tableio.py	95.09% <100%> (+0.46%)	⬆️
ctapipe/io/tests/test_hdf5.py	94.55% <100%> (+1.47%)	⬆️
ctapipe/io/tests/test_available_sources.py	100% <0%> (ø)	⬆️
ctapipe/instrument/tests/test_camera.py	100% <0%> (ø)	⬆️
ctapipe/io/containers.py	100% <0%> (ø)	⬆️
ctapipe/io/__init__.py	100% <0%> (ø)	⬆️
ctapipe/core/component.py	96.29% <0%> (ø)	⬆️
ctapipe/io/sst1meventsource.py
ctapipe/io/tests/test_sst1meventsource.py
ctapipe/io/tests/test_targetio_event_source.py
... and 4 more

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 6a6230a...02b2c43. Read the comment docs.

[kosack]

Codacy is just complaining about invalid syntax again -> f-strings. Apart from that, I think it might be fine.

I contacted them and got the response that they do not yet support python 3.6+ (!!) but are looking into it. So generally we can ignore those errors. I thought I had disabled them in fact, but perhaps it has to be done for each branch?

[dneise]

@cta-observatory/cta-devs I think this might be mergeble