Cleanup of airctrl.py

[Cyber1000]

tl;dr

I did a bigger refactoring last year and added the coap-part, but changes in airctrl.py where only a first approach. There where differences in output and many redundances between the 3 protocols (http, plaincoap and coap) This PR cleans airctrl.py - output will be the same no matter which protocol you use

longer information, points to pay attention

changes to cli:

• added CliBase to have a common formatting class depending on STATUS_TRANSFORMER

• updated STATUS_TRANSFORMER class:

  • some values where missing here (for example "Gas" in ddp)
  • added an error value I found a while ago -> 193: 'F0 (pre-filter) must be cleaned'
  • added an additional index to define sets like "filter", "firmware" (used for easier output with get_firmware, get_filters)

• attention - changed formatting:

  • may change ordering of output
    • for now it will be ordered like it comes out of the device
    • I think it will be possible to take the ordering of STATUS_TRANSFORMER, if someone wishes this
  • may add additonal spaces between key and value
  • minor changes (plain coaps` output was "Software version:", the other protocols outputted "Version:")
  • Ordering, spacing was different between coap, plaincoap and http, so I think it is better to have the same output on all protocols. Coap doesn't change at all, the other 2 may have this issue.
  • see also testdata (data.json)

• attention - changed behaviour:

  • http-protocol was the only one which executed get_values immediately after set_values, the other protocols didn't output anything after calling set_values
  • So I removed it from Class HTTPAirCli (and removed a test in test_http.py)
  • If this is not what we want, we may:
    • add get_values back again to HTTPAirCli.set_values
    • add get_values to CliBase.set_values (and therefore for all protocols)
  • Not sure what's best here

• tests were adjusted and build green

[rgerganov]

Thank you for the PR. I will try to review and provide feedback soon.

[rgerganov]

Sorry for taking me so long to get to this. I have reviewed and tested your changes, everything looks fine to me! About the behavior change for HTTP devices -- I prefer to get the updated values after changing settings but I am also fine with your approach. I think the best way to switch between those two behaviors is at the end of the main function:

        if values:
            c.set_values(values, debug=args.debug)
        else:
            c.get_status(debug=args.debug)

vs.

        if values:
            c.set_values(values, debug=args.debug)
        c.get_status(debug=args.debug)

[Cyber1000]

No problem, was busy too. Well yes I think the proposed change at the end of the main function would be the best position if we want to add the get after a set. Maybe in another PR. Thanks for merging!