pydata / sparse

Sparse multi-dimensional arrays for the PyData ecosystem
https://sparse.pydata.org
BSD 3-Clause "New" or "Revised" License
581 stars 123 forks source link

Support direct Numba imports #689

Closed mtsokol closed 1 month ago

mtsokol commented 2 months ago

Hi @hameerabbasi,

I noticed that downstream libraries use direct imports for some Numba objects, e.g.: from sparse import COO (https://github.com/search?q=%22from+sparse+import%22+language%3APython&type=code). Since we will be releasing version 0.16.0 in some time, we need to support this kind of import to avoid disruption to users.

github-actions[bot] commented 2 months ago

Test Results

6 041 tests  +2   5 958 :white_check_mark:  - 23   9m 38s :stopwatch: ±0s     1 suites ±0      58 :zzz: ± 0      1 files   ±0      25 :x: +25 

For more details on these failures, see this check.

Results for commit ddb9f5b1. ± Comparison against base commit d47b87af.

:recycle: This comment has been updated with latest results.

hameerabbasi commented 2 months ago

For this pattern, we would probably need something else that dispatches appropriately.

mtsokol commented 1 month ago

Replaced by https://github.com/pydata/sparse/pull/695