File: test_query_hybrid_search.py

package info (click to toggle)
python-azure 20250603%2Bgit-1
  • links: PTS, VCS
  • area: main
  • in suites: forky, sid, trixie
  • size: 851,724 kB
  • sloc: python: 7,362,925; ansic: 804; javascript: 287; makefile: 195; sh: 145; xml: 109
file content (376 lines) | stat: -rw-r--r-- 20,628 bytes parent folder | download
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
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
# The MIT License (MIT)
# Copyright (c) Microsoft Corporation. All rights reserved.

import json
import os
import time
import unittest
import uuid

import pytest

import azure.cosmos.cosmos_client as cosmos_client
import azure.cosmos.exceptions as exceptions
import test_config
import hybrid_search_data
from azure.cosmos import http_constants, DatabaseProxy
from azure.cosmos.partition_key import PartitionKey

@pytest.mark.cosmosSearchQuery
class TestFullTextHybridSearchQuery(unittest.TestCase):
    """Test to check full text search and hybrid search queries behavior."""

    client: cosmos_client.CosmosClient = None
    config = test_config.TestConfig
    host = config.host
    masterKey = config.masterKey
    connectionPolicy = config.connectionPolicy
    TEST_DATABASE_ID = config.TEST_DATABASE_ID
    TEST_CONTAINER_ID = "Full Text Container " + str(uuid.uuid4())

    @classmethod
    def setUpClass(cls):
        if (cls.masterKey == '[YOUR_KEY_HERE]' or
                cls.host == '[YOUR_ENDPOINT_HERE]'):
            raise Exception(
                "You must specify your Azure Cosmos account values for "
                "'masterKey' and 'host' at the top of this class to run the "
                "tests.")

        cls.client = cosmos_client.CosmosClient(cls.host, cls.masterKey)
        cls.test_db = cls.client.create_database(str(uuid.uuid4()))
        cls.test_container = cls.test_db.create_container(
            id="FTS" + cls.TEST_CONTAINER_ID,
            partition_key=PartitionKey(path="/pk"),
            offer_throughput=test_config.TestConfig.THROUGHPUT_FOR_2_PARTITIONS,
            indexing_policy=test_config.get_full_text_indexing_policy(path="/text"),
            full_text_policy=test_config.get_full_text_policy(path="/text"))
        data = hybrid_search_data.get_full_text_items()
        for index, item in enumerate(data.get("items")):
            item['id'] = str(index)
            item['pk'] = str((index % 2) + 1)
            cls.test_container.create_item(item)
        # Need to give the container time to index all the recently added items - 10 minutes seems to work
        time.sleep(10 * 60)

    @classmethod
    def tearDownClass(cls):
        try:
            cls.test_db.delete_container(cls.test_container.id)
            cls.client.delete_database(cls.test_db.id)
        except exceptions.CosmosHttpResponseError:
            pass

    def test_wrong_hybrid_search_queries(self):
        try:
            query = "SELECT c.index, RRF(VectorDistance(c.vector, [1,2,3]), FullTextScore(c.text, 'test') FROM c"
            results = self.test_container.query_items(query, enable_cross_partition_query=True)
            list(results)
            pytest.fail("Attempting to project RRF in a query should fail.")
        except exceptions.CosmosHttpResponseError as e:
            assert e.status_code == http_constants.StatusCodes.BAD_REQUEST
            # TODO: This message seems to differ depending on machine as well
            assert ("One of the input values is invalid" in e.message or
             "Syntax error, incorrect syntax near 'FROM'" in e.message)

        try:
            query = "SELECT TOP 10 c.index FROM c WHERE FullTextContains(c.title, 'John')" \
                    " ORDER BY RANK FullTextScore(c.title, 'John') DESC"
            results = self.test_container.query_items(query, enable_cross_partition_query=True)
            list(results)
            pytest.fail("Attempting to set an ordering direction in a full text score query should fail.")
        except exceptions.CosmosHttpResponseError as e:
            assert e.status_code == http_constants.StatusCodes.BAD_REQUEST
            assert ("One of the input values is invalid" in e.message or
                    "Specifying a sort order (ASC or DESC) in the ORDER BY RANK clause is not allowed." in e.message)

        try:
            query = "SELECT TOP 10 c.index FROM c WHERE FullTextContains(c.title, 'John')" \
                    " ORDER BY RANK RRF(FullTextScore(c.title, 'John'), VectorDistance(c.vector, [1,2,3])) DESC"
            results = self.test_container.query_items(query, enable_cross_partition_query=True)
            list(results)
            pytest.fail("Attempting to set an ordering direction in a hybrid search query should fail.")
        except exceptions.CosmosHttpResponseError as e:
            assert e.status_code == http_constants.StatusCodes.BAD_REQUEST
            # TODO: Find why this behavior is inconsistent across runs - message should be the same
            assert ("One of the input values is invalid" in e.message or
                    "Specifying a sort order (ASC or DESC) in the ORDER BY RANK clause is not allowed." in e.message)

    def test_hybrid_search_env_variables(self):
        os.environ["AZURE_COSMOS_HYBRID_SEARCH_MAX_ITEMS"] = "0"
        try:
            query = "SELECT TOP 1 c.index, c.title FROM c WHERE FullTextContains(c.title, 'John') OR " \
                    "FullTextContains(c.text, 'John') ORDER BY RANK FullTextScore(c.title, 'John')"
            results = self.test_container.query_items(query, enable_cross_partition_query=True)
            [item for item in results]
            pytest.fail("Config was not applied properly.")
        except ValueError as e:
            assert e.args[0] == ("Executing a hybrid search query with more items than the max is not allowed. "
                                 "Please ensure you are using a limit smaller than the max, or change the max.")
        finally:
            os.environ["AZURE_COSMOS_HYBRID_SEARCH_MAX_ITEMS"] = "1000"

    def test_hybrid_search_queries(self):
        query = "SELECT TOP 10 c.index, c.title FROM c WHERE FullTextContains(c.title, 'John') OR " \
                "FullTextContains(c.text, 'John') ORDER BY RANK FullTextScore(c.title, 'John')"
        results = self.test_container.query_items(query, enable_cross_partition_query=True)
        result_list = list(results)
        assert len(result_list) == 3
        for res in result_list:
            assert res['index'] in [2, 85, 57]

        query = "SELECT TOP 10 c.index, c.title FROM c WHERE FullTextContains(c.title, 'John')" \
                " OR FullTextContains(c.text, 'John') ORDER BY RANK FullTextScore(c.title, 'John')"
        results = self.test_container.query_items(query, enable_cross_partition_query=True)
        result_list = list(results)
        assert len(result_list) == 3
        for res in result_list:
            assert res['index'] in [2, 85, 57]

        query = "SELECT c.index, c.title FROM c WHERE FullTextContains(c.title, 'John')" \
                " OR FullTextContains(c.text, 'John') ORDER BY RANK FullTextScore(c.title, 'John') OFFSET 1 LIMIT 5"
        results = self.test_container.query_items(query, enable_cross_partition_query=True)
        result_list = list(results)
        assert len(result_list) == 2
        for res in result_list:
            assert res['index'] in [85, 57]

        query = "SELECT TOP 20 c.index, c.title FROM c WHERE FullTextContains(c.title, 'John') OR " \
                "FullTextContains(c.text, 'John') OR FullTextContains(c.text, 'United States') " \
                "ORDER BY RANK RRF(FullTextScore(c.title, 'John'), FullTextScore(c.text, 'United States'))"
        results = self.test_container.query_items(query, enable_cross_partition_query=True)
        result_list = list(results)
        assert len(result_list) == 15
        for res in result_list:
            assert res['index'] in [61, 51, 49, 54, 75, 24, 77, 76, 80, 25, 22, 2, 66, 57, 85]

        query = "SELECT TOP 10 c.index, c.title FROM c WHERE " \
                "FullTextContains(c.title, 'John') OR FullTextContains(c.text, 'John') OR " \
                "FullTextContains(c.text, 'United States') ORDER BY RANK RRF(FullTextScore(c.title, 'John')," \
                " FullTextScore(c.text, 'United States'))"
        results = self.test_container.query_items(query, enable_cross_partition_query=True)
        result_list = list(results)
        assert len(result_list) == 10
        for res in result_list:
            assert res['index'] in [61, 51, 49, 54, 75, 24, 77, 76, 80, 25]

        query = "SELECT c.index, c.title FROM c WHERE FullTextContains(c.title, 'John')" \
                " OR FullTextContains(c.text, 'John') OR FullTextContains(c.text, 'United States') ORDER BY " \
                "RANK RRF(FullTextScore(c.title, 'John'), FullTextScore(c.text, 'United States')) OFFSET 5 LIMIT 10"
        results = self.test_container.query_items(query, enable_cross_partition_query=True)
        result_list = list(results)
        assert len(result_list) == 10
        for res in result_list:
            assert res['index'] in [24, 77, 76, 80, 25, 22, 2, 66, 57, 85]

        query = "SELECT TOP 10 c.index, c.title FROM c " \
                "ORDER BY RANK RRF(FullTextScore(c.title, 'John'), FullTextScore(c.text, 'United States'))"
        results = self.test_container.query_items(query, enable_cross_partition_query=True)
        result_list = list(results)
        assert len(result_list) == 10
        for res in result_list:
            assert res['index'] in [61, 51, 49, 54, 75, 24, 77, 76, 80, 25]

        query = "SELECT c.index, c.title FROM c " \
                "ORDER BY RANK RRF(FullTextScore(c.title, 'John'), FullTextScore(c.text, 'United States')) " \
                "OFFSET 0 LIMIT 13"
        results = self.test_container.query_items(query, enable_cross_partition_query=True)
        result_list = list(results)
        assert len(result_list) == 13
        for res in result_list:
            assert res['index'] in [61, 51, 49, 54, 75, 24, 77, 76, 80, 25, 22, 2, 66]

        item_vector = self.test_container.read_item('50', '1')['vector']
        query = "SELECT c.index, c.title FROM c " \
                "ORDER BY RANK RRF(FullTextScore(c.text, 'United States'), VectorDistance(c.vector, {})) " \
                "OFFSET 0 LIMIT 10".format(item_vector)
        results = self.test_container.query_items(query, enable_cross_partition_query=True)
        result_list = list(results)
        assert len(result_list) == 10
        for res in result_list:
            assert res['index'] in [51, 54, 28, 70, 24, 61, 56, 26, 58, 77]

    def test_hybrid_search_query_pagination(self):
        query = "SELECT c.index, c.title FROM c " \
                "ORDER BY RANK RRF(FullTextScore(c.title, 'John'), FullTextScore(c.text, 'United States')) " \
                "OFFSET 0 LIMIT 13"
        query_iterable = self.test_container.query_items(query, enable_cross_partition_query=True, max_item_count=5)
        all_fetched_res = []
        count = 0
        item_pages = query_iterable.by_page()
        for items in item_pages:
            count += 1
            all_fetched_res.extend(list(items))
        assert count == 3
        assert len(all_fetched_res) == 13

    def test_hybrid_search_cross_partition_query_response_hook(self):
        item_vector = self.test_container.read_item('50', '1')['vector']
        response_hook = test_config.ResponseHookCaller()
        query = "SELECT c.index, c.title FROM c " \
                "ORDER BY RANK RRF(FullTextScore(c.text, 'United States'), VectorDistance(c.vector, {})) " \
                "OFFSET 0 LIMIT 10".format(item_vector)
        results = self.test_container.query_items(query, enable_cross_partition_query=True, response_hook=response_hook)
        result_list = list(results)
        assert len(result_list) == 10
        assert response_hook.count == 6 # one global stat query per partition, two queries per partition for each component query

    def test_hybrid_search_partitioned_query_response_hook(self):
        item_vector = self.test_container.read_item('50', '1')['vector']
        response_hook = test_config.ResponseHookCaller()
        query = "SELECT c.index, c.title FROM c " \
                "ORDER BY RANK RRF(FullTextScore(c.text, 'United States'), VectorDistance(c.vector, {})) " \
                "OFFSET 0 LIMIT 10".format(item_vector)
        results = self.test_container.query_items(query, partition_key='1', response_hook=response_hook)
        result_list = list(results)
        assert len(result_list) == 10
        assert response_hook.count == 1

    def test_hybrid_search_weighted_reciprocal_rank_fusion(self):
        # Test case 1
        query = """
                    SELECT TOP 15 c.index AS Index, c.title AS Title, c.text AS Text
                    FROM c
                    WHERE FullTextContains(c.title, 'John') OR FullTextContains(c.text, 'John') OR FullTextContains(c.text, 'United States')
                    ORDER BY RANK RRF(FullTextScore(c.title, 'John'), FullTextScore(c.text, 'United States'), [1, 1])
                """
        results = self.test_container.query_items(query, enable_cross_partition_query=True)
        result_list = [res['Index'] for res in results]
        # If some scores rank the same the order of the results may change
        assert result_list in [
            [61, 51, 49, 54, 75, 24, 77, 76, 80, 25, 22, 2, 66, 57, 85],
            [61, 51, 49, 54, 75, 24, 77, 76, 80, 25, 22, 2, 66, 85, 57]
        ]

        # Test case 2
        query = """
                    SELECT TOP 15 c.index AS Index, c.title AS Title, c.text AS Text
                    FROM c
                    WHERE FullTextContains(c.title, 'John') OR FullTextContains(c.text, 'John') OR FullTextContains(c.text, 'United States')
                    ORDER BY RANK RRF(FullTextScore(c.title, 'John'), FullTextScore(c.text, 'United States'), [10, 10])
                """
        results = self.test_container.query_items(query, enable_cross_partition_query=True)
        result_list = [res['Index'] for res in results]
        # If some scores rank the same the order of the results may change
        assert result_list in [
            [61, 51, 49, 54, 75, 24, 77, 76, 80, 25, 22, 2, 66, 57, 85],
            [61, 51, 49, 54, 75, 24, 77, 76, 80, 25, 22, 2, 66, 85, 57]
        ]

        # Test case 3
        query = """
                    SELECT TOP 10 c.index AS Index, c.title AS Title, c.text AS Text
                    FROM c
                    WHERE FullTextContains(c.title, 'John') OR FullTextContains(c.text, 'John') OR FullTextContains(c.text, 'United States')
                    ORDER BY RANK RRF(FullTextScore(c.title, 'John'), FullTextScore(c.text, 'United States'), [0.1, 0.1])
                """
        results = self.test_container.query_items(query, enable_cross_partition_query=True)
        result_list = [res['Index'] for res in results]
        assert result_list == [61, 51, 49, 54, 75, 24, 77, 76, 80, 25]

        # Test case 4
        query = """
                    SELECT TOP 15 c.index AS Index, c.title AS Title, c.text AS Text
                    FROM c
                    WHERE FullTextContains(c.title, 'John') OR FullTextContains(c.text, 'John') OR FullTextContains(c.text, 'United States')
                    ORDER BY RANK RRF(FullTextScore(c.title, 'John'), FullTextScore(c.text, 'United States'), [-1, -1])
                """
        results = self.test_container.query_items(query, enable_cross_partition_query=True)
        result_list = [res['Index'] for res in results]
        # If some scores rank the same the order of the results may change
        assert result_list in [
            [85, 57, 66, 2, 22, 25, 77, 76, 80, 75, 24, 49, 54, 51, 81],
            [57, 85, 2, 66, 22, 25, 80, 76, 77, 24, 75, 54, 49, 51, 61]
        ]

        # Test case 5
        item_vector = self.test_container.read_item('50', '1')['vector']
        query = "SELECT c.index, c.title FROM c " \
                "ORDER BY RANK RRF(FullTextScore(c.text, 'United States'), VectorDistance(c.vector, {}), [1,1]) " \
                "OFFSET 0 LIMIT 10".format(item_vector)
        results = self.test_container.query_items(query, enable_cross_partition_query=True)
        result_list = list(results)
        assert len(result_list) == 10
        result_list = [res['index'] for res in result_list]
        assert result_list == [51, 54, 28, 70, 24, 61, 56, 26, 58, 77]

    def test_invalid_hybrid_search_queries_weighted_reciprocal_rank_fusion(self):
        try:
            query = "SELECT c.index, RRF(VectorDistance(c.vector, [1,2,3]), FullTextScore(c.text, 'test') FROM c"
            results = self.test_container.query_items(query, enable_cross_partition_query=True)
            list(results)
            pytest.fail("Attempting to project RRF in a query should fail.")
        except exceptions.CosmosHttpResponseError as e:
            assert e.status_code == http_constants.StatusCodes.BAD_REQUEST

    def test_weighted_vs_non_weighted_reciprocal_rank_fusion(self):
        # Non-weighted RRF query
        query_non_weighted = """
            SELECT TOP 10 c.index, c.title FROM c
            ORDER BY RANK RRF(FullTextScore(c.title, 'John'), FullTextScore(c.text, 'United States'))
        """
        results_non_weighted = self.test_container.query_items(query_non_weighted, enable_cross_partition_query=True)
        result_list_non_weighted = [res['index'] for res in results_non_weighted]

        # Weighted RRF query with equal weights
        query_weighted_equal = """
            SELECT TOP 10 c.index, c.title FROM c
            ORDER BY RANK RRF(FullTextScore(c.title, 'John'), FullTextScore(c.text, 'United States'), [1, 1])
        """
        results_weighted_equal = self.test_container.query_items(query_weighted_equal,
                                                                 enable_cross_partition_query=True)
        result_list_weighted_equal = [res['index'] for res in results_weighted_equal]

        # Weighted RRF query with different direction weights
        query_weighted_different = """
            SELECT TOP 10 c.index, c.title FROM c
            ORDER BY RANK RRF(FullTextScore(c.title, 'John'), FullTextScore(c.text, 'United States'), [1, -0.5])
        """
        results_weighted_different = self.test_container.query_items(query_weighted_different,
                                                                     enable_cross_partition_query=True)
        result_list_weighted_different = [res['index'] for res in results_weighted_different]

        # Assertions
        assert result_list_non_weighted == result_list_weighted_equal, "Non-weighted and equally weighted RRF results should match."
        assert result_list_non_weighted != result_list_weighted_different, "Non-weighted and differently weighted RRF results should not match."

    def test_weighted_reciprocal_rank_fusion_with_missing_or_extra_weights(self):
        try:
            # Weighted RRF query with one weight missing
            query_missing_weight = """
                SELECT TOP 10 c.index, c.title FROM c
                ORDER BY RANK RRF(FullTextScore(c.title, 'John'), FullTextScore(c.text, 'United States'), [1])
            """
            
            res = self.test_container.query_items(query_missing_weight, enable_cross_partition_query=True)
            list(res)
            pytest.fail("Query with missing weights should fail.")
        except exceptions.CosmosHttpResponseError as e:
            assert e.status_code == http_constants.StatusCodes.BAD_REQUEST
        try:
            # Weighted RRF query with an extra weight
            query_extra_weight = """
                    SELECT TOP 10 c.index, c.title FROM c
                    ORDER BY RANK RRF(FullTextScore(c.title, 'John'), FullTextScore(c.text, 'United States'), [1, 1, 1])
            """
            list(self.test_container.query_items(query_extra_weight, enable_cross_partition_query=True))
            pytest.fail("Query with extra weights should fail.")
        except exceptions.CosmosHttpResponseError as e:
            assert e.status_code == http_constants.StatusCodes.BAD_REQUEST

    def test_weighted_reciprocal_rank_fusion_with_response_hook(self):
        response_hook = test_config.ResponseHookCaller()
        query_weighted_rrf = """
            SELECT TOP 10 c.index, c.title FROM c
            ORDER BY RANK RRF(FullTextScore(c.title, 'John'), FullTextScore(c.text, 'United States'), [1, 0.5])
        """
        results = self.test_container.query_items(query_weighted_rrf, enable_cross_partition_query=True,
                                                  response_hook=response_hook)
        result_list = list(results)
        assert len(result_list) == 10
        assert response_hook.count > 0  # Ensure the response hook was called



if __name__ == "__main__":
    unittest.main()