René's URL Explorer Experiment

{"@context":"https://schema.org","@type":"DiscussionForumPosting","headline":"Broadcasting rules distinguish between host scalar and device scalars","articleBody":"Consider this script:\r\n```python\r\nactx = _acf()\r\n\r\n@with_container_arithmetic(\r\n        bcast_obj_array=True,\r\n        bcast_numpy_array=True,\r\n        rel_comparison=True,\r\n        _cls_has_array_context_attr=True)\r\n@dataclass_array_container\r\n@dataclass(frozen=True)\r\nclass Foo:\r\n    u: DOFArray\r\n\r\nfoo = Foo(DOFArray(actx, (actx.zeros(3, dtype=np.float64) + 41, )))\r\n\r\nprint(foo)                                 # prints: Foo(u=DOFArray((cl.Array([41., 41., 41.]),)))\r\nprint(foo + 1)                             # prints: Foo(u=DOFArray((cl.Array([42., 42., 42.]),)))\r\nprint(foo + actx.from_numpy(np.ones(())))  # ERROR! 😲\r\n```\r\n\r\nI don't see a good reason why we don't support this.\r\n\r\nI think we should also go a step ahead and make sure that we broadcast every cl.Array to be added (if legal) with all the leaf arrays of the array container. One way I could see this working is if every `ArrayContext` comes with a `ARRAY_TYPE` static, frozen attribute which gives us a hint to broadcast to all successive arrays.","author":{"url":"https://github.com/kaushikcfd","@type":"Person","name":"kaushikcfd"},"datePublished":"2021-06-26T19:26:34.000Z","interactionStatistic":{"@type":"InteractionCounter","interactionType":"https://schema.org/CommentAction","userInteractionCount":10},"url":"https://github.com/49/arraycontext/issues/49"}