-
Notifications
You must be signed in to change notification settings - Fork 89
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
Start using RegularArray everywhere it needs to be used. (#24)
This check-in includes a commented-out attempt at `lower_getitem_nothing` in `awkward1._numba.array.numpyarray`. In the next PR, it will be gone. I concluded that there is no non-dependent type inference (where array dimensionality is part of the type but the length is not) for `__getitem__` of arrays for `NumpyArray` because an empty array would reduce the dimensionality by one yet return nothing, but reducing the dimensionality of a one-dimensional `NumpyArray` would return a scalar, which is not nothing. Whereas the C++ implementation could be fully transferred from a `ListArray` wrapper to a `RegularArray` wrapper (because it is dynamically typed, ironically), the Numba implementation could not (because it is not dynamically typed). The Numba implementation only wraps with `RegularArray` if the types are sound. * Start using RegularArray everywhere it needs to be used. * [skip ci] ListArray64 wrapper -> RegularArray wrapper partly works. * [skip ci] ListArray64 wrapper -> RegularArray wrapper still only works partially. * ListArray64 wrapper -> RegularArray wrapper now works * RegularArray::tojson_part and fix EmptyArray::tojson_part. * Centralize lower_getitem_tuple and lower_getitem_other. * Avoid selecting types based on a value (length of array) in Content::getitem(Slice). * getitem_tuple for Numba mirrors C++ as well as possible: if the 'outtpe' is 'NumpyArrayType', then you're going to have to 'lower_getitem_int'; 'lower_getitem_nothing' is impossible. * ListArray and ListOffsetArray::getitem_next(array, not advanced) returns a RegularArray now (C++ and Numba). * ListArray and ListOffsetArray::getitem_next(array) returns the same shape of RegularArrays as the index array. * Now newaxis uses RegularArray, too. * Ambiguous identities in ListArray are now refused with an error. * [skip ci] writing awkward_identity*_from_listoffsetarray*. * ListOffsetArray now uses awkward_identity*_from_listoffsetarray*. * Addressed the FIXMEs in getitem.cpp; ready to merge PR.
- Loading branch information
Showing
40 changed files
with
584 additions
and
348 deletions.
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -1 +1 @@ | ||
0.1.23 | ||
0.1.24 |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Oops, something went wrong.