|
26 | 26 | ``` |
27 | 27 | """ |
28 | 28 |
|
| 29 | +from contextlib import contextmanager |
29 | 30 | import pathlib |
| 31 | +import tempfile |
30 | 32 |
|
31 | 33 | import pytest |
32 | 34 |
|
|
36 | 38 | from pandas.io.iceberg import read_iceberg |
37 | 39 |
|
38 | 40 | pyiceberg = pytest.importorskip("pyiceberg") |
| 41 | +pyiceberg_catalog = pytest.importorskip("pyiceberg.catalog") |
| 42 | +pq = pytest.importorskip("pyarrow.parquet") |
| 43 | + |
| 44 | + |
| 45 | +@contextmanager |
| 46 | +def create_catalog(): |
| 47 | + # the catalog stores the full path of data files, so the catalog needs to be |
| 48 | + # created dynamically, and not saved in pandas/tests/io/data as other formats |
| 49 | + with tempfile.TemporaryDirectory("pandas-iceberg.tmp") as catalog_path: |
| 50 | + uri = f"sqlite:///{catalog_path}/catalog.sqlite" |
| 51 | + catalog = pyiceberg_catalog.load_catalog( |
| 52 | + "default", |
| 53 | + type="sql", |
| 54 | + uri=uri, |
| 55 | + warehouse=f"file://{catalog_path}", |
| 56 | + ) |
| 57 | + catalog.create_namespace("default") |
39 | 58 |
|
| 59 | + df = pq.read_table( |
| 60 | + pathlib.Path(__file__).parent / "data" / "parquet" / "simple.parquet" |
| 61 | + ) |
| 62 | + table = catalog.create_table("default.simple", schema=df.schema) |
| 63 | + table.append(df) |
40 | 64 |
|
41 | | -@pytest.fixture |
42 | | -def catalog_uri(datapath): |
43 | | - path = datapath("io", "data", "iceberg", "catalog.sqlite") |
44 | | - return f"sqlite:///{path}" |
| 65 | + yield uri |
45 | 66 |
|
46 | 67 |
|
47 | 68 | class TestIceberg: |
48 | | - def test_read(self, catalog_uri): |
| 69 | + def test_read(self): |
49 | 70 | expected = pd.DataFrame( |
50 | 71 | { |
51 | 72 | "A": [1, 2, 3], |
52 | 73 | "B": ["foo", "foo", "foo"], |
53 | 74 | } |
54 | 75 | ) |
55 | | - result = read_iceberg( |
56 | | - "default.simple", |
57 | | - catalog_properties={"uri": catalog_uri}, |
58 | | - ) |
| 76 | + with create_catalog() as catalog_uri: |
| 77 | + result = read_iceberg( |
| 78 | + "default.simple", |
| 79 | + catalog_properties={"uri": catalog_uri}, |
| 80 | + ) |
59 | 81 | tm.assert_frame_equal(result, expected) |
60 | 82 |
|
61 | | - def test_read_by_catalog_name(self, catalog_uri): |
| 83 | + def test_read_by_catalog_name(self): |
62 | 84 | config_path = pathlib.Path.home() / ".pyiceberg.yaml" |
63 | | - with open(config_path, "w") as f: |
64 | | - f.write(f"""\ |
65 | | -catalog: |
66 | | - pandas_tests_catalog: |
67 | | - uri: {catalog_uri}""") |
68 | | - expected = pd.DataFrame( |
69 | | - { |
70 | | - "A": [1, 2, 3], |
71 | | - "B": ["foo", "foo", "foo"], |
72 | | - } |
73 | | - ) |
74 | | - result = read_iceberg( |
75 | | - "default.simple", |
76 | | - catalog_name="pandas_tests_catalog", |
77 | | - ) |
| 85 | + with create_catalog() as catalog_uri: |
| 86 | + with open(config_path, "w") as f: |
| 87 | + f.write(f"""\ |
| 88 | + catalog: |
| 89 | + pandas_tests_catalog: |
| 90 | + uri: {catalog_uri}""") |
| 91 | + expected = pd.DataFrame( |
| 92 | + { |
| 93 | + "A": [1, 2, 3], |
| 94 | + "B": ["foo", "foo", "foo"], |
| 95 | + } |
| 96 | + ) |
| 97 | + result = read_iceberg( |
| 98 | + "default.simple", |
| 99 | + catalog_name="pandas_tests_catalog", |
| 100 | + ) |
78 | 101 | tm.assert_frame_equal(result, expected) |
79 | 102 | # config_path.unlink() |
80 | 103 |
|
81 | | - def test_read_with_row_filter(self, catalog_uri): |
| 104 | + def test_read_with_row_filter(self): |
82 | 105 | expected = pd.DataFrame( |
83 | 106 | { |
84 | 107 | "A": [2, 3], |
85 | 108 | "B": ["foo", "foo"], |
86 | 109 | } |
87 | 110 | ) |
88 | | - result = read_iceberg( |
89 | | - "default.simple", |
90 | | - catalog_properties={"uri": catalog_uri}, |
91 | | - row_filter="A > 1", |
92 | | - ) |
| 111 | + with create_catalog() as catalog_uri: |
| 112 | + result = read_iceberg( |
| 113 | + "default.simple", |
| 114 | + catalog_properties={"uri": catalog_uri}, |
| 115 | + row_filter="A > 1", |
| 116 | + ) |
93 | 117 | tm.assert_frame_equal(result, expected) |
94 | 118 |
|
95 | | - def test_read_with_case_sensitive(self, catalog_uri): |
| 119 | + def test_read_with_case_sensitive(self): |
96 | 120 | expected = pd.DataFrame( |
97 | 121 | { |
98 | 122 | "A": [1, 2, 3], |
99 | 123 | } |
100 | 124 | ) |
101 | | - result = read_iceberg( |
102 | | - "default.simple", |
103 | | - catalog_properties={"uri": catalog_uri}, |
104 | | - selected_fields=["a"], |
105 | | - case_sensitive=False, |
106 | | - ) |
107 | | - tm.assert_frame_equal(result, expected) |
108 | | - |
109 | | - with pytest.raises(ValueError, match="^Could not find column"): |
110 | | - read_iceberg( |
| 125 | + with create_catalog() as catalog_uri: |
| 126 | + result = read_iceberg( |
111 | 127 | "default.simple", |
112 | 128 | catalog_properties={"uri": catalog_uri}, |
113 | 129 | selected_fields=["a"], |
114 | | - case_sensitive=True, |
| 130 | + case_sensitive=False, |
115 | 131 | ) |
| 132 | + tm.assert_frame_equal(result, expected) |
116 | 133 |
|
117 | | - def test_read_with_limit(self, catalog_uri): |
| 134 | + with create_catalog() as catalog_uri: |
| 135 | + with pytest.raises(ValueError, match="^Could not find column"): |
| 136 | + read_iceberg( |
| 137 | + "default.simple", |
| 138 | + catalog_properties={"uri": catalog_uri}, |
| 139 | + selected_fields=["a"], |
| 140 | + case_sensitive=True, |
| 141 | + ) |
| 142 | + |
| 143 | + def test_read_with_limit(self): |
118 | 144 | expected = pd.DataFrame( |
119 | 145 | { |
120 | 146 | "A": [1, 2], |
121 | 147 | "B": ["foo", "foo"], |
122 | 148 | } |
123 | 149 | ) |
124 | | - result = read_iceberg( |
125 | | - "default.simple", |
126 | | - catalog_properties={"uri": catalog_uri}, |
127 | | - limit=2, |
128 | | - ) |
| 150 | + with create_catalog() as catalog_uri: |
| 151 | + result = read_iceberg( |
| 152 | + "default.simple", |
| 153 | + catalog_properties={"uri": catalog_uri}, |
| 154 | + limit=2, |
| 155 | + ) |
129 | 156 | tm.assert_frame_equal(result, expected) |
0 commit comments