1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
|
from __future__ import annotations
from contextlib import suppress
import numpy as np
import pytest
from xarray import Variable
from xarray.coding import strings
from xarray.core import indexing
from xarray.tests import (
IndexerMaker,
assert_array_equal,
assert_identical,
requires_dask,
)
with suppress(ImportError):
import dask.array as da
def test_vlen_dtype() -> None:
dtype = strings.create_vlen_dtype(str)
assert dtype.metadata["element_type"] is str
assert strings.is_unicode_dtype(dtype)
assert not strings.is_bytes_dtype(dtype)
assert strings.check_vlen_dtype(dtype) is str
dtype = strings.create_vlen_dtype(bytes)
assert dtype.metadata["element_type"] is bytes
assert not strings.is_unicode_dtype(dtype)
assert strings.is_bytes_dtype(dtype)
assert strings.check_vlen_dtype(dtype) is bytes
# check h5py variant ("vlen")
dtype = np.dtype("O", metadata={"vlen": str}) # type: ignore[call-overload,unused-ignore]
assert strings.check_vlen_dtype(dtype) is str
assert strings.check_vlen_dtype(np.dtype(object)) is None
@pytest.mark.parametrize("numpy_str_type", (np.str_, np.bytes_))
def test_numpy_subclass_handling(numpy_str_type) -> None:
with pytest.raises(TypeError, match="unsupported type for vlen_dtype"):
strings.create_vlen_dtype(numpy_str_type)
def test_EncodedStringCoder_decode() -> None:
coder = strings.EncodedStringCoder()
raw_data = np.array([b"abc", "ß∂µ∆".encode()])
raw = Variable(("x",), raw_data, {"_Encoding": "utf-8"})
actual = coder.decode(raw)
expected = Variable(("x",), np.array(["abc", "ß∂µ∆"], dtype=object))
assert_identical(actual, expected)
assert_identical(coder.decode(actual[0]), expected[0])
@requires_dask
def test_EncodedStringCoder_decode_dask() -> None:
coder = strings.EncodedStringCoder()
raw_data = np.array([b"abc", "ß∂µ∆".encode()])
raw = Variable(("x",), raw_data, {"_Encoding": "utf-8"}).chunk()
actual = coder.decode(raw)
assert isinstance(actual.data, da.Array)
expected = Variable(("x",), np.array(["abc", "ß∂µ∆"], dtype=object))
assert_identical(actual, expected)
actual_indexed = coder.decode(actual[0])
assert isinstance(actual_indexed.data, da.Array)
assert_identical(actual_indexed, expected[0])
def test_EncodedStringCoder_encode() -> None:
dtype = strings.create_vlen_dtype(str)
raw_data = np.array(["abc", "ß∂µ∆"], dtype=dtype)
expected_data = np.array([r.encode("utf-8") for r in raw_data], dtype=object)
coder = strings.EncodedStringCoder(allows_unicode=True)
raw = Variable(("x",), raw_data, encoding={"dtype": "S1"})
actual = coder.encode(raw)
expected = Variable(("x",), expected_data, attrs={"_Encoding": "utf-8"})
assert_identical(actual, expected)
raw = Variable(("x",), raw_data)
assert_identical(coder.encode(raw), raw)
coder = strings.EncodedStringCoder(allows_unicode=False)
assert_identical(coder.encode(raw), expected)
@pytest.mark.parametrize(
"original",
[
Variable(("x",), [b"ab", b"cdef"]),
Variable((), b"ab"),
Variable(("x",), [b"a", b"b"]),
Variable((), b"a"),
],
)
def test_CharacterArrayCoder_roundtrip(original) -> None:
coder = strings.CharacterArrayCoder()
roundtripped = coder.decode(coder.encode(original))
assert_identical(original, roundtripped)
@pytest.mark.parametrize(
"data",
[
np.array([b"a", b"bc"]),
np.array([b"a", b"bc"], dtype=strings.create_vlen_dtype(bytes)),
],
)
def test_CharacterArrayCoder_encode(data) -> None:
coder = strings.CharacterArrayCoder()
raw = Variable(("x",), data)
actual = coder.encode(raw)
expected = Variable(("x", "string2"), np.array([[b"a", b""], [b"b", b"c"]]))
assert_identical(actual, expected)
@pytest.mark.parametrize(
["original", "expected_char_dim_name"],
[
(Variable(("x",), [b"ab", b"cdef"]), "string4"),
(Variable(("x",), [b"ab", b"cdef"], encoding={"char_dim_name": "foo"}), "foo"),
],
)
def test_CharacterArrayCoder_char_dim_name(original, expected_char_dim_name) -> None:
coder = strings.CharacterArrayCoder()
encoded = coder.encode(original)
roundtripped = coder.decode(encoded)
assert encoded.dims[-1] == expected_char_dim_name
assert roundtripped.encoding["char_dim_name"] == expected_char_dim_name
assert roundtripped.dims[-1] == original.dims[-1]
@pytest.mark.parametrize(
[
"original",
"expected_char_dim_name",
"expected_char_dim_length",
"warning_message",
],
[
(
Variable(("x",), [b"ab", b"cde"], encoding={"char_dim_name": "foo4"}),
"foo3",
3,
"String dimension naming mismatch",
),
(
Variable(
("x",),
[b"ab", b"cde"],
encoding={"original_shape": (2, 4), "char_dim_name": "foo"},
),
"foo3",
3,
"String dimension length mismatch",
),
],
)
def test_CharacterArrayCoder_dim_mismatch_warnings(
original, expected_char_dim_name, expected_char_dim_length, warning_message
) -> None:
coder = strings.CharacterArrayCoder()
with pytest.warns(UserWarning, match=warning_message):
encoded = coder.encode(original)
roundtripped = coder.decode(encoded)
assert encoded.dims[-1] == expected_char_dim_name
assert encoded.sizes[expected_char_dim_name] == expected_char_dim_length
assert roundtripped.encoding["char_dim_name"] == expected_char_dim_name
assert roundtripped.dims[-1] == original.dims[-1]
def test_StackedBytesArray() -> None:
array = np.array([[b"a", b"b", b"c"], [b"d", b"e", b"f"]], dtype="S")
actual = strings.StackedBytesArray(array)
expected = np.array([b"abc", b"def"], dtype="S")
assert actual.dtype == expected.dtype
assert actual.shape == expected.shape
assert actual.size == expected.size
assert actual.ndim == expected.ndim
assert len(actual) == len(expected)
assert_array_equal(expected, actual)
B = IndexerMaker(indexing.BasicIndexer)
assert_array_equal(expected[:1], actual[B[:1]])
with pytest.raises(IndexError):
actual[B[:, :2]]
def test_StackedBytesArray_scalar() -> None:
array = np.array([b"a", b"b", b"c"], dtype="S")
actual = strings.StackedBytesArray(array)
expected = np.array(b"abc")
assert actual.dtype == expected.dtype
assert actual.shape == expected.shape
assert actual.size == expected.size
assert actual.ndim == expected.ndim
with pytest.raises(TypeError):
len(actual)
np.testing.assert_array_equal(expected, actual)
B = IndexerMaker(indexing.BasicIndexer)
with pytest.raises(IndexError):
actual[B[:2]]
def test_StackedBytesArray_vectorized_indexing() -> None:
array = np.array([[b"a", b"b", b"c"], [b"d", b"e", b"f"]], dtype="S")
stacked = strings.StackedBytesArray(array)
expected = np.array([[b"abc", b"def"], [b"def", b"abc"]])
V = IndexerMaker(indexing.VectorizedIndexer)
indexer = V[np.array([[0, 1], [1, 0]])]
actual = stacked.vindex[indexer]
assert_array_equal(actual, expected)
def test_char_to_bytes() -> None:
array = np.array([[b"a", b"b", b"c"], [b"d", b"e", b"f"]])
expected = np.array([b"abc", b"def"])
actual = strings.char_to_bytes(array)
assert_array_equal(actual, expected)
expected = np.array([b"ad", b"be", b"cf"])
actual = strings.char_to_bytes(array.T) # non-contiguous
assert_array_equal(actual, expected)
def test_char_to_bytes_ndim_zero() -> None:
expected = np.array(b"a")
actual = strings.char_to_bytes(expected)
assert_array_equal(actual, expected)
def test_char_to_bytes_size_zero() -> None:
array = np.zeros((3, 0), dtype="S1")
expected = np.array([b"", b"", b""])
actual = strings.char_to_bytes(array)
assert_array_equal(actual, expected)
@requires_dask
def test_char_to_bytes_dask() -> None:
numpy_array = np.array([[b"a", b"b", b"c"], [b"d", b"e", b"f"]])
array = da.from_array(numpy_array, ((2,), (3,)))
expected = np.array([b"abc", b"def"])
actual = strings.char_to_bytes(array)
assert isinstance(actual, da.Array)
assert actual.chunks == ((2,),)
assert actual.dtype == "S3"
assert_array_equal(np.array(actual), expected)
with pytest.raises(ValueError, match=r"stacked dask character array"):
strings.char_to_bytes(array.rechunk(1))
def test_bytes_to_char() -> None:
array = np.array([[b"ab", b"cd"], [b"ef", b"gh"]])
expected = np.array([[[b"a", b"b"], [b"c", b"d"]], [[b"e", b"f"], [b"g", b"h"]]])
actual = strings.bytes_to_char(array)
assert_array_equal(actual, expected)
expected = np.array([[[b"a", b"b"], [b"e", b"f"]], [[b"c", b"d"], [b"g", b"h"]]])
actual = strings.bytes_to_char(array.T) # non-contiguous
assert_array_equal(actual, expected)
@requires_dask
def test_bytes_to_char_dask() -> None:
numpy_array = np.array([b"ab", b"cd"])
array = da.from_array(numpy_array, ((1, 1),))
expected = np.array([[b"a", b"b"], [b"c", b"d"]])
actual = strings.bytes_to_char(array)
assert isinstance(actual, da.Array)
assert actual.chunks == ((1, 1), ((2,)))
assert actual.dtype == "S1"
assert_array_equal(np.array(actual), expected)
|
