Turns out that #133 exposed a mistake in how we were using the end_bit
field (and it should have been picked up in review, my bad).
Essentially we were masking based on the length of the end_bit after
*already* shifting by the start_bit. We should mask then shift, not the
other way around.
May well fix issue #135 (pdb generation will have been affected by
this).
This removes the assertions that checked plugin devs didn't do anything
overly bad in favour of using mypy to ensure appropriate type-checking.
It also moves the ProgressCallback typing information to constants in
order to avoid circular imports (since constants doesn't import
anything).
This includes:
* Better ways of checking empty lists
* Not shadowing builtin functions like filter
* Preventing invalid slash warnings by marking strings as regexps
* Removing unnecessary brackets
* Lowercase variable names
* Adding/updating parameters in docstrings
* Removing unused code (lines not chunks)
* Change in not a member tests
* Changing some methods to static
* Shorting range membership checks
* Missing parameters
* Make some exception handlers more specific
* Don't define a lambda to a variable
* A few more instance checks to help type checkers
Relented on the strict import of direct objects/classes for the typing
module only. Typing module components can be directly imported because
it makes the code really painful to read and write otherwise.
This is still in-line with the python style guide adopted from Google at
http://google.github.io/styleguide/pyguide.html section 2.2.
The __getattribute__ method is called for *all* attribute lookups, and
the python implementation is likely slow. The only function the python
implementation provided was to check that properties started with
helper_. Since this only provided a log message, we could add it back
in only if the log level is high enough? Either way, it's not
significant enough to warrant costing extra time in every attribute
lookup on every object.
This appears to get in a loop when it's been unpickled by
multiprocessing. My guess is that it's trying to set _concrete_members
without it being done in __init__ and there gets in a loop.
We ensure then when doing a look-up inside __getattr__, it happens using
__getattribute__.
Mike Auty