|
| 1 | +# Copyright 2025 Google LLC |
| 2 | +# |
| 3 | +# Licensed under the Apache License, Version 2.0 (the "License"); |
| 4 | +# you may not use this file except in compliance with the License. |
| 5 | +# You may obtain a copy of the License at |
| 6 | +# |
| 7 | +# http://www.apache.org/licenses/LICENSE-2.0 |
| 8 | +# |
| 9 | +# Unless required by applicable law or agreed to in writing, software |
| 10 | +# distributed under the License is distributed on an "AS IS" BASIS, |
| 11 | +# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. |
| 12 | +# See the License for the specific language governing permissions and |
| 13 | +# limitations under the License. |
| 14 | + |
| 15 | +import pytest |
| 16 | + |
| 17 | +from sqlalchemy_bigquery._types import _get_transitive_schema_fields |
| 18 | +from google.cloud.bigquery.schema import SchemaField |
| 19 | + |
| 20 | + |
| 21 | +def create_schema_field_from_dict(schema_dict): |
| 22 | + """ |
| 23 | + Helper function to create a SchemaField object from a dictionary representation. |
| 24 | + """ |
| 25 | + api_repr = { |
| 26 | + "name": schema_dict["name"], |
| 27 | + "type": schema_dict["type"], |
| 28 | + "mode": schema_dict.get("mode", "NULLABLE"), |
| 29 | + "fields": [ |
| 30 | + create_schema_field_from_dict(sf_dict).to_api_repr() |
| 31 | + for sf_dict in schema_dict.get("fields", []) |
| 32 | + ], |
| 33 | + } |
| 34 | + return SchemaField.from_api_repr(api_repr) |
| 35 | + |
| 36 | + |
| 37 | +test_cases = [ |
| 38 | + ( |
| 39 | + "STRUCT field, not REPEATED, with sub-fields, should recurse", |
| 40 | + [ |
| 41 | + create_schema_field_from_dict( |
| 42 | + { |
| 43 | + "name": "s1", |
| 44 | + "type": "STRUCT", |
| 45 | + "mode": "NULLABLE", |
| 46 | + "fields": [ |
| 47 | + {"name": "child1", "type": "STRING", "mode": "NULLABLE"} |
| 48 | + ], |
| 49 | + } |
| 50 | + ) |
| 51 | + ], |
| 52 | + ["s1", "s1.child1"], |
| 53 | + ), |
| 54 | + ( |
| 55 | + "RECORD field (alias for STRUCT), not REPEATED, with sub-fields, should recurse", |
| 56 | + [ |
| 57 | + create_schema_field_from_dict( |
| 58 | + { |
| 59 | + "name": "r1", |
| 60 | + "type": "RECORD", |
| 61 | + "mode": "NULLABLE", |
| 62 | + "fields": [ |
| 63 | + {"name": "child_r1", "type": "INTEGER", "mode": "NULLABLE"} |
| 64 | + ], |
| 65 | + } |
| 66 | + ) |
| 67 | + ], |
| 68 | + ["r1", "r1.child_r1"], |
| 69 | + ), |
| 70 | + ( |
| 71 | + "STRUCT field, REPEATED, with sub-fields, should NOT recurse", |
| 72 | + [ |
| 73 | + create_schema_field_from_dict( |
| 74 | + { |
| 75 | + "name": "s2", |
| 76 | + "type": "STRUCT", |
| 77 | + "mode": "REPEATED", |
| 78 | + "fields": [ |
| 79 | + {"name": "child2", "type": "STRING", "mode": "NULLABLE"} |
| 80 | + ], |
| 81 | + } |
| 82 | + ) |
| 83 | + ], |
| 84 | + ["s2"], |
| 85 | + ), |
| 86 | + ( |
| 87 | + "Non-STRUCT field (STRING), not REPEATED, should NOT recurse", |
| 88 | + [ |
| 89 | + create_schema_field_from_dict( |
| 90 | + {"name": "f1", "type": "STRING", "mode": "NULLABLE"} |
| 91 | + ) |
| 92 | + ], |
| 93 | + ["f1"], |
| 94 | + ), |
| 95 | + ( |
| 96 | + "Non-STRUCT field (INTEGER), REPEATED, should NOT recurse", |
| 97 | + [ |
| 98 | + create_schema_field_from_dict( |
| 99 | + {"name": "f2", "type": "INTEGER", "mode": "REPEATED"} |
| 100 | + ) |
| 101 | + ], |
| 102 | + ["f2"], |
| 103 | + ), |
| 104 | + ( |
| 105 | + "Deeply nested STRUCT, not REPEATED, should recurse fully", |
| 106 | + [ |
| 107 | + create_schema_field_from_dict( |
| 108 | + { |
| 109 | + "name": "s_outer", |
| 110 | + "type": "STRUCT", |
| 111 | + "mode": "NULLABLE", |
| 112 | + "fields": [ |
| 113 | + { |
| 114 | + "name": "s_inner1", |
| 115 | + "type": "STRUCT", |
| 116 | + "mode": "NULLABLE", |
| 117 | + "fields": [ |
| 118 | + { |
| 119 | + "name": "s_leaf1", |
| 120 | + "type": "STRING", |
| 121 | + "mode": "NULLABLE", |
| 122 | + } |
| 123 | + ], |
| 124 | + }, |
| 125 | + {"name": "s_sibling", "type": "INTEGER", "mode": "NULLABLE"}, |
| 126 | + { |
| 127 | + "name": "s_inner2_repeated_struct", |
| 128 | + "type": "STRUCT", |
| 129 | + "mode": "REPEATED", |
| 130 | + "fields": [ |
| 131 | + { |
| 132 | + "name": "s_leaf2_ignored", |
| 133 | + "type": "BOOLEAN", |
| 134 | + "mode": "NULLABLE", |
| 135 | + } |
| 136 | + ], |
| 137 | + }, |
| 138 | + ], |
| 139 | + } |
| 140 | + ) |
| 141 | + ], |
| 142 | + [ |
| 143 | + "s_outer", |
| 144 | + "s_outer.s_inner1", |
| 145 | + "s_outer.s_inner1.s_leaf1", |
| 146 | + "s_outer.s_sibling", |
| 147 | + "s_outer.s_inner2_repeated_struct", |
| 148 | + ], |
| 149 | + ), |
| 150 | + ( |
| 151 | + "STRUCT field, not REPEATED, but no sub-fields, should not error and not recurse further", |
| 152 | + [ |
| 153 | + create_schema_field_from_dict( |
| 154 | + {"name": "s3", "type": "STRUCT", "mode": "NULLABLE", "fields": []} |
| 155 | + ) |
| 156 | + ], |
| 157 | + ["s3"], |
| 158 | + ), |
| 159 | + ( |
| 160 | + "Multiple top-level fields with mixed conditions", |
| 161 | + [ |
| 162 | + create_schema_field_from_dict( |
| 163 | + {"name": "id", "type": "INTEGER", "mode": "REQUIRED"} |
| 164 | + ), |
| 165 | + create_schema_field_from_dict( |
| 166 | + { |
| 167 | + "name": "user_profile", |
| 168 | + "type": "STRUCT", |
| 169 | + "mode": "NULLABLE", |
| 170 | + "fields": [ |
| 171 | + {"name": "name", "type": "STRING", "mode": "NULLABLE"}, |
| 172 | + { |
| 173 | + "name": "addresses", |
| 174 | + "type": "RECORD", |
| 175 | + "mode": "REPEATED", |
| 176 | + "fields": [ |
| 177 | + { |
| 178 | + "name": "street", |
| 179 | + "type": "STRING", |
| 180 | + "mode": "NULLABLE", |
| 181 | + }, |
| 182 | + {"name": "city", "type": "STRING", "mode": "NULLABLE"}, |
| 183 | + ], |
| 184 | + }, |
| 185 | + ], |
| 186 | + } |
| 187 | + ), |
| 188 | + create_schema_field_from_dict( |
| 189 | + {"name": "tags", "type": "STRING", "mode": "REPEATED"} |
| 190 | + ), |
| 191 | + ], |
| 192 | + ["id", "user_profile", "user_profile.name", "user_profile.addresses", "tags"], |
| 193 | + ), |
| 194 | + ( |
| 195 | + "Empty input list of fields", |
| 196 | + [], |
| 197 | + [], |
| 198 | + ), |
| 199 | + ( |
| 200 | + "Field type not in STRUCT_FIELD_TYPES and mode is REPEATED", |
| 201 | + [ |
| 202 | + create_schema_field_from_dict( |
| 203 | + {"name": "f_arr", "type": "FLOAT", "mode": "REPEATED"} |
| 204 | + ) |
| 205 | + ], |
| 206 | + ["f_arr"], |
| 207 | + ), |
| 208 | + ( |
| 209 | + "Field type not in STRUCT_FIELD_TYPES and mode is not REPEATED", |
| 210 | + [ |
| 211 | + create_schema_field_from_dict( |
| 212 | + {"name": "f_single", "type": "DATE", "mode": "NULLABLE"} |
| 213 | + ) |
| 214 | + ], |
| 215 | + ["f_single"], |
| 216 | + ), |
| 217 | +] |
| 218 | + |
| 219 | + |
| 220 | +@pytest.mark.parametrize( |
| 221 | + "description, input_fields_list, expected_field_names", test_cases |
| 222 | +) |
| 223 | +def test_get_transitive_schema_fields_conditions( |
| 224 | + description, input_fields_list, expected_field_names |
| 225 | +): |
| 226 | + """ |
| 227 | + Tests the _get_transitive_schema_fields function, focusing on the conditional logic |
| 228 | + `if field.field_type in STRUCT_FIELD_TYPES and field.mode != "REPEATED":`. |
| 229 | + """ |
| 230 | + result_fields = _get_transitive_schema_fields(input_fields_list) |
| 231 | + result_names = [field.name for field in result_fields] |
| 232 | + assert result_names == expected_field_names, description |
0 commit comments