Hi all!
I should have said something about this, but I didn't expect anyone to pay it much attention. I think that some of the stuff that pylint complains about are things that perhaps don't make much sense and that we might want to change the config just a bit.
However, thanks very much for doing this and being so proactive :)
I'm thinking to add some new features to 817 (pms and settings) so I needed a source to work on ... and because I very bad know python I was missing some sort of checker.
The only real problem I found is that the row
self._memobj.pmsfilled &= ~ (1 << bitindex)
was confusing pylint that much that it complains about all reference to .pmsfilled (the same happens with .pmsvisible and operator |= )
If someone knows how to avoid this _bad_ behavior of pylint please let me know.
-def _ft8x7_read(pipe, block, blocknum):
I was going to leave these types of functions at the module level instead of pushing them into the class. Doing the latter makes the classes much larger and potentially harder to understand. However, it does make it easier to access pieces of the radio, such as _block_lengths in your case.
What do people think? Should they go in or out of the class?
Moving that functions and memory structure definition inside the class is something I was thinking to do since some time already.
I think that all radio specific code must be inside radio class and only generic/non radio related code have to be left alone. This have the advantage to make derived classes more simple and easy to understand.
... my two cents ...
73 de IZ3GME Marco