import pytest import iarray as ia import numpy as np import zarr array_data = [ ([30, 100], [20, 20], [10, 13], [30, 100], 1, True, None), ([30, 130], [50, 50], [20, 25], [40, 130], 0, False, None), pytest.param( [10, 78, 55, 21], [3, 30, 30, 21], [3, 12, 6, 21], [10, 78, 55, 2], 3, True, "test_append_acontiguous.iarr", marks=pytest.mark.heavy, ), ([30, 100], [30, 44], [30, 2], [4, 100], 0, False, "test_append_asparse.iarr"), ] @pytest.mark.parametrize( "dtype, np_dtype", [ (np.float32, None), (np.uint64, ">M8[M]"), (np.int64, "i8"), (np.uint32, "u8"), ], ) @pytest.mark.parametrize( "shape, chunks, blocks, data_shape, axis, acontiguous, aurlpath", array_data, ) def test_append(shape, chunks, blocks, data_shape, axis, dtype, np_dtype, acontiguous, aurlpath): ia.remove_urlpath(aurlpath) cfg = ia.Config(chunks=chunks, blocks=blocks, contiguous=acontiguous, urlpath=aurlpath) max = 1 out_dtype = dtype if np_dtype is None else np.dtype(np_dtype) if out_dtype not in [np.float64, np.float32]: for i in range(len(shape)): max *= shape[i] a = ia.linspace(0, max, num=int(np.prod(shape)), shape=shape, cfg=cfg, mode="w", dtype=dtype, np_dtype=np_dtype) npa = ia.iarray2numpy(a) z = zarr.array(npa, dtype=out_dtype) with pytest.raises(ValueError): np_data = np.full(shape=17, fill_value=47, dtype=out_dtype) a.append(np_data, axis=axis) np_data = np.full(shape=data_shape, fill_value=47, dtype=out_dtype) new_shape = a.append(data=np_data, axis=axis) new_shape_z = z.append(np_data, axis=axis) assert new_shape == new_shape_z npb = ia.iarray2numpy(a) npc = z[:] if out_dtype in [np.float64, np.float32]: rtol = 1e-6 if out_dtype == np.float32 else 1e-14 np.testing.assert_allclose(npb, npc, rtol=rtol, atol=0) else: np.testing.assert_equal(npb, npc) ia.remove_urlpath(aurlpath)