refactor: move typetracer ufunc handling to backend [1 of 2] #2150
Conversation
agoose77
changed the title
Codecov Report
Additional details and impacted files
|
| elif x is ak._typetracer.UnknownLength: | ||
| return numpy.empty((0,), dtype=np.int64) | ||
| elif isinstance(x, ak._typetracer.MaybeNone): | ||
| return self._coerce_ufunc_argument(x.content) |
There was a problem hiding this comment.
@jpivarski this is one part I'm hesitant on. We're throwing away type information here because this might not fail at runtime, i.e. if the value is not actually none.
There was a problem hiding this comment.
If we do want this, should MaybeNone also support operators (unrelated to this PR, but a more general point), e.g.. MaybeNone(x) + y? I'm inclined to say yes - MaybeNone should be fairly transparent, and serve as the outermost type like Haskell's Maybe.
There was a problem hiding this comment.
And, what should we do about OneOf? The proper approach would be to try all branches and re-wrap the result, but if one branch fails, we would need to handle that.
There was a problem hiding this comment.
Our strategy is for typetracers to give code the benefit of the doubt and let it fail at runtime. The typetracer is not a type-check.
So if a typetracer (including MaybeNone or OneOf) represents
If a typetracer represents np.numerical.)
There was a problem hiding this comment.
OK, that sounds like the best path to me, too. Let's go with the safe, minimal approach and see what happens. I haven't implemented OneOf here, but we can do that in another PR, as it's probably better to keep momentum of the changes that are already in the pipeline.
| @property | ||
| @abstractmethod | ||
| def name(self) -> str: | ||
| raise ak._errors.wrap_error(NotImplementedError) |
agoose77
changed the title
This reverts commit f52401e.
This PR splits the
TypeTracerufunc logic handling from #2127, using theBackend.apply_ufuncmachinery to implementufuncoperations onak.Arrays backed by `TypeTracer.It implements "coercion" from
UnknownScalar,UnknownLength, andMaybeNonevalues to the appropriate NumPy empty arrays, before applying the ufunc.