Minutes_2019_07_16

Attendees: Stuart, Todd, Siu

0. Feature Discussion

#4327 - Specialize operator methods for boolean types (stop casting as int!).
#4326 - Powers of complex numbers in @cuda.jit decorated function
**** #4324 - Vector @overload not overwriting scalar one
- may need @overload_ufunc to resolve and isolate the rewrite. but don't know how to do it yet.
**** #4323 - numba segmentation fault on Python 3.7.4
#4322 - dtype error which I can't explain
**** #4321 - Feature request: numpy.flip
- need helper for making tuple.
**** #4319 - np.isnan fails on integer dtype in jitted functions (numpy 1.17.0rc1)
#4316 - Add docs on supported architecture in the README
#4315 - Check dtype arguments for Function types
#4314 - Dynamically-defined @guvectorized functions over dask distributed
#4313 - Segfault with cfunc and Python 3.7.4
#4312 - Split inside a list comprehension NotImplementedError
#4310 - Python classes in numba cuda?
#4309 - CUDA parallel range?
#4307 - Can't access Typed Dict with (int32, int32) keys in the python interpreter
#4306 - Add inspect_types (etc) docstrings
**** #4299 - Memory leak when using parallel=True and slicing array
- likely caused by the temporary reduction array in arr.sum().
- Assigned to Siu
#4298 - Numba 0.45.0RC1 checklist

#4328 - Replace GC macro with function call
#4325 - accept scalar/0d-arrays
#4318 - Add supported platforms to the docs. Closes #4316
#4317 - Add docs for ARMv8/AArch64
#4311 - Avoid deprecated use of inspect.getargspec
- merge this for 0.45
**** #4305 - Don't allow the allocation of mutable objects written into a container to be hoisted.
- patch make sense. need to try breaking it.
#4304 - DO NOT MERGE. merge of aarch64 package fixes and scipy fixes