LAMA to Dask: `func` -> built-in conversion for non-trig. methods by sadielbartholomew · Pull Request #308 · NCAS-CMS/cf-python
Partially address #302, to tick all sub-issue boxes except the topmost* and in doing so close #303, since Data.log is now fully migrated (and marked as such) because the /= operator it uses with a non-e/10/2 based is applied in 'Dask space' rather than cf space.
Note
* the topmost item, 'the eight trigonometric and hyperbolic methods and their inverses which don't have restricted domains', will be tackled as a separate PR to make it easier for the reviewer to check that the correct eight methods have been converted and in a consistent way, with commenting to explain why the other four such methods have been left as func calls. And since the code changes required on all of the methods listed in #302 are largely the same, such that any review feedback can be done as one on them all, and it would be more time-consuming to make seven separate PRs for each, I thought it best to split into just two PRs to cover the Issue (second PR for the trig. methods to follow).