Is this the only problem in that issue? It seems to be that there is a "NoneType" error in that TB too? -- Meaning that this PR is possibly not the only fix for that issue.

I can't be 100% certain, but the main issue is that build fails due to redact not handling sets, which (I think) would lead to all entities becoming unavailable, ie value=None. All of the nonetype errors are from a template helper, so It's just triggered by the charger object not having the communication_signal_strength-attribute, which would make sense since it never got to the polling stage since build failed.

Another topic that would probably need to be explored is why there are missing chargers that lead us into the warning message that crashed, but I think we need this fix before that can be explored further.

Approved and merged. I agree we need to get this out ASAP, so I'm thinking of creating a 0.8.1b1 *) pre-release for this to be tested. We need input from the two people as to why and I suggest they upload their diagnostics file so we can look at it. I am on travel right now, so I haven't tested any of the code.

*) If this were a 1.0 release I would call it a 1.0.1 release, but I really don't like 0.8.0.1.