|
| 1 | +import os |
| 2 | +import configparser |
| 3 | +import multiprocessing |
| 4 | + |
| 5 | +import requests |
| 6 | +import unittest |
| 7 | + |
| 8 | +from fastapi import FastAPI |
| 9 | +import uvicorn |
| 10 | + |
| 11 | +from semantic_matcher import model |
| 12 | +from semantic_matcher.model import SemanticMatch |
| 13 | +from semantic_matcher.service import SemanticMatchingService |
| 14 | + |
| 15 | +from contextlib import contextmanager |
| 16 | +import signal |
| 17 | +import time |
| 18 | + |
| 19 | +import json as js |
| 20 | + |
| 21 | + |
| 22 | +def run_server(): |
| 23 | + # Load test configuration |
| 24 | + config = configparser.ConfigParser() |
| 25 | + config.read([ |
| 26 | + os.path.abspath(os.path.join(os.path.dirname(__file__), "../test_resources/config.ini")), |
| 27 | + ]) |
| 28 | + |
| 29 | + # Read in equivalence table |
| 30 | + EQUIVALENCES = model.EquivalenceTable.from_file( |
| 31 | + filename=os.path.abspath(os.path.join( |
| 32 | + os.path.dirname(__file__), |
| 33 | + "..", |
| 34 | + config["SERVICE"]["equivalence_table_file"] |
| 35 | + )) |
| 36 | + ) |
| 37 | + |
| 38 | + # Initialise SemanticMatchingService |
| 39 | + semantic_matching_service = SemanticMatchingService( |
| 40 | + endpoint=config["SERVICE"]["endpoint"], |
| 41 | + equivalences=EQUIVALENCES |
| 42 | + ) |
| 43 | + |
| 44 | + # Mock resolver |
| 45 | + def mock_get_matcher(self, semantic_id): |
| 46 | + return "http://remote-service:8000" |
| 47 | + |
| 48 | + SemanticMatchingService._get_matcher_from_semantic_id = mock_get_matcher |
| 49 | + |
| 50 | + # Mock remote service |
| 51 | + original_requests_get = requests.get |
| 52 | + |
| 53 | + class SimpleResponse: |
| 54 | + def __init__(self, content: str, status_code: int = 200): |
| 55 | + self.text = content |
| 56 | + self.status_code = status_code |
| 57 | + |
| 58 | + def mock_requests_get(url, json): |
| 59 | + if url == "http://remote-service:8000/get_matches": |
| 60 | + match_one = SemanticMatch( |
| 61 | + base_semantic_id="s-heppner.com/semanticID/three", |
| 62 | + match_semantic_id="remote-service.com/semanticID/tres", |
| 63 | + score=1.0, |
| 64 | + meta_information={"matchSource": "Defined by Moritz Sommer", |
| 65 | + "path": ["remote-service.com/semanticID/trois"]} |
| 66 | + ) |
| 67 | + matches_data = { |
| 68 | + "matches": [match_one.model_dump()] |
| 69 | + } |
| 70 | + matches_json = js.dumps(matches_data) |
| 71 | + return SimpleResponse(content=matches_json) |
| 72 | + else: |
| 73 | + return original_requests_get(url, json=json) |
| 74 | + |
| 75 | + requests.get = mock_requests_get |
| 76 | + |
| 77 | + # Run server |
| 78 | + app = FastAPI() |
| 79 | + app.include_router(semantic_matching_service.router) |
| 80 | + uvicorn.run(app, host=config["SERVICE"]["ENDPOINT"], port=int(config["SERVICE"]["PORT"]), log_level="error") |
| 81 | + |
| 82 | + |
| 83 | +@contextmanager |
| 84 | +def run_server_context(): |
| 85 | + server_process = multiprocessing.Process(target=run_server) |
| 86 | + server_process.start() |
| 87 | + try: |
| 88 | + time.sleep(2) # Wait for the server to start |
| 89 | + yield |
| 90 | + finally: |
| 91 | + server_process.terminate() |
| 92 | + server_process.join(timeout=5) |
| 93 | + if server_process.is_alive(): |
| 94 | + os.kill(server_process.pid, signal.SIGKILL) |
| 95 | + server_process.join() |
| 96 | + |
| 97 | + |
| 98 | +class TestSemanticMatchingService(unittest.TestCase): |
| 99 | + |
| 100 | + def test_get_all_matches(self): |
| 101 | + with run_server_context(): |
| 102 | + response = requests.get("http://localhost:8000/all_matches") |
| 103 | + expected_matches = { |
| 104 | + 's-heppner.com/semanticID/one': [ |
| 105 | + { |
| 106 | + 'base_semantic_id': 's-heppner.com/semanticID/one', |
| 107 | + 'match_semantic_id': 's-heppner.com/semanticID/1', |
| 108 | + 'score': 1.0, |
| 109 | + 'meta_information': {'matchSource': 'Defined by Sebastian Heppner'} |
| 110 | + }, |
| 111 | + { |
| 112 | + 'base_semantic_id': 's-heppner.com/semanticID/one', |
| 113 | + 'match_semantic_id': 's-heppner.com/semanticID/two', |
| 114 | + 'score': 0.8, |
| 115 | + 'meta_information': {'matchSource': 'Defined by Sebastian Heppner'} |
| 116 | + } |
| 117 | + ], |
| 118 | + 's-heppner.com/semanticID/two': [ |
| 119 | + { |
| 120 | + 'base_semantic_id': 's-heppner.com/semanticID/two', |
| 121 | + 'match_semantic_id': 's-heppner.com/semanticID/2', |
| 122 | + 'score': 1.0, |
| 123 | + 'meta_information': {'matchSource': 'Defined by Sebastian Heppner'} |
| 124 | + } |
| 125 | + ], |
| 126 | + 's-heppner.com/semanticID/three': [ |
| 127 | + { |
| 128 | + 'base_semantic_id': 's-heppner.com/semanticID/three', |
| 129 | + 'match_semantic_id': 'remote-service.com/semanticID/trois', |
| 130 | + 'score': 1.0, |
| 131 | + 'meta_information': {'matchSource': 'Defined by Moritz Sommer'} |
| 132 | + } |
| 133 | + ] |
| 134 | + } |
| 135 | + actual_matches = response.json() |
| 136 | + self.assertEqual(expected_matches, actual_matches) |
| 137 | + |
| 138 | + def test_post_matches(self): |
| 139 | + with run_server_context(): |
| 140 | + new_match = { |
| 141 | + "base_semantic_id": "s-heppner.com/semanticID/new", |
| 142 | + "match_semantic_id": "s-heppner.com/semanticID/3", |
| 143 | + "score": 0.95, |
| 144 | + "meta_information": {"matchSource": "Defined by UnitTest"} |
| 145 | + } |
| 146 | + matches_list = { |
| 147 | + "matches": [new_match] |
| 148 | + } |
| 149 | + requests.post( |
| 150 | + "http://localhost:8000/post_matches", |
| 151 | + json=matches_list |
| 152 | + ) |
| 153 | + response = requests.get("http://localhost:8000/all_matches") |
| 154 | + actual_matches = response.json() |
| 155 | + self.assertIn("s-heppner.com/semanticID/new", actual_matches) |
| 156 | + self.assertEqual( |
| 157 | + actual_matches["s-heppner.com/semanticID/new"][0]["match_semantic_id"], |
| 158 | + "s-heppner.com/semanticID/3" |
| 159 | + ) |
| 160 | + |
| 161 | + self.assertEqual( |
| 162 | + actual_matches["s-heppner.com/semanticID/new"][0]["score"], |
| 163 | + 0.95 |
| 164 | + ) |
| 165 | + |
| 166 | + self.assertEqual( |
| 167 | + actual_matches["s-heppner.com/semanticID/new"][0]["meta_information"]["matchSource"], |
| 168 | + "Defined by UnitTest" |
| 169 | + ) |
| 170 | + |
| 171 | + def test_get_matches_local_only(self): |
| 172 | + with run_server_context(): |
| 173 | + match_request = { |
| 174 | + "semantic_id": "s-heppner.com/semanticID/one", |
| 175 | + "score_limit": 0.5, |
| 176 | + "local_only": True |
| 177 | + } |
| 178 | + response = requests.get("http://localhost:8000/get_matches", json=match_request) |
| 179 | + expected_matches = { |
| 180 | + "matches": [ |
| 181 | + { |
| 182 | + "base_semantic_id": "s-heppner.com/semanticID/one", |
| 183 | + "match_semantic_id": "s-heppner.com/semanticID/1", |
| 184 | + "score": 1.0, |
| 185 | + "meta_information": {"matchSource": "Defined by Sebastian Heppner"} |
| 186 | + }, |
| 187 | + { |
| 188 | + "base_semantic_id": "s-heppner.com/semanticID/one", |
| 189 | + "match_semantic_id": "s-heppner.com/semanticID/two", |
| 190 | + "score": 0.8, |
| 191 | + "meta_information": {"matchSource": "Defined by Sebastian Heppner"} |
| 192 | + }, |
| 193 | + { |
| 194 | + "base_semantic_id": "s-heppner.com/semanticID/one", |
| 195 | + "match_semantic_id": "s-heppner.com/semanticID/2", |
| 196 | + "score": 0.8, |
| 197 | + "meta_information": {"matchSource": "Defined by Sebastian Heppner", |
| 198 | + "path": ["s-heppner.com/semanticID/two"]} |
| 199 | + } |
| 200 | + ] |
| 201 | + } |
| 202 | + actual_matches = response.json() |
| 203 | + self.assertEqual(expected_matches, actual_matches) |
| 204 | + |
| 205 | + def test_get_matches_local_and_remote(self): |
| 206 | + with run_server_context(): |
| 207 | + match_request = { |
| 208 | + "semantic_id": "s-heppner.com/semanticID/three", |
| 209 | + "score_limit": 0.7, |
| 210 | + "local_only": False |
| 211 | + } |
| 212 | + response = requests.get("http://localhost:8000/get_matches", json=match_request) |
| 213 | + expected_matches = { |
| 214 | + "matches": [ |
| 215 | + { |
| 216 | + "base_semantic_id": "s-heppner.com/semanticID/three", |
| 217 | + "match_semantic_id": "remote-service.com/semanticID/trois", |
| 218 | + "score": 1.0, |
| 219 | + "meta_information": {"matchSource": "Defined by Moritz Sommer"} |
| 220 | + }, |
| 221 | + { |
| 222 | + "base_semantic_id": "s-heppner.com/semanticID/three", |
| 223 | + "match_semantic_id": "remote-service.com/semanticID/tres", |
| 224 | + "score": 1.0, |
| 225 | + "meta_information": {"matchSource": "Defined by Moritz Sommer", |
| 226 | + "path": ["remote-service.com/semanticID/trois"]} |
| 227 | + }, |
| 228 | + ] |
| 229 | + } |
| 230 | + actual_matches = response.json() |
| 231 | + self.assertEqual(expected_matches, actual_matches) |
| 232 | + |
| 233 | + def test_get_matches_no_matches(self): |
| 234 | + with run_server_context(): |
| 235 | + match_request = { |
| 236 | + "semantic_id": "s-heppner.com/semanticID/unknown", |
| 237 | + "score_limit": 0.5, |
| 238 | + "local_only": True |
| 239 | + } |
| 240 | + response = requests.get("http://localhost:8000/get_matches", json=match_request) |
| 241 | + expected_matches = {"matches": []} |
| 242 | + actual_matches = response.json() |
| 243 | + self.assertEqual(expected_matches, actual_matches) |
| 244 | + |
| 245 | + def test_get_matches_with_low_score_limit(self): |
| 246 | + with run_server_context(): |
| 247 | + match_request = { |
| 248 | + "semantic_id": "s-heppner.com/semanticID/one", |
| 249 | + "score_limit": 0.9, |
| 250 | + "local_only": True |
| 251 | + } |
| 252 | + response = requests.get("http://localhost:8000/get_matches", json=match_request) |
| 253 | + expected_matches = { |
| 254 | + "matches": [ |
| 255 | + { |
| 256 | + "base_semantic_id": "s-heppner.com/semanticID/one", |
| 257 | + "match_semantic_id": "s-heppner.com/semanticID/1", |
| 258 | + "score": 1.0, |
| 259 | + "meta_information": {"matchSource": "Defined by Sebastian Heppner"} |
| 260 | + } |
| 261 | + ] |
| 262 | + } |
| 263 | + actual_matches = response.json() |
| 264 | + self.assertEqual(expected_matches, actual_matches) |
| 265 | + |
| 266 | + def test_get_matches_with_nlp_parameters(self): |
| 267 | + with run_server_context(): |
| 268 | + match_request = { |
| 269 | + "semantic_id": "s-heppner.com/semanticID/one", |
| 270 | + "score_limit": 0.5, |
| 271 | + "local_only": True, |
| 272 | + "name": "Example Name", |
| 273 | + "definition": "Example Definition" |
| 274 | + } |
| 275 | + response = requests.get("http://localhost:8000/get_matches", json=match_request) |
| 276 | + expected_matches = { |
| 277 | + "matches": [ |
| 278 | + { |
| 279 | + "base_semantic_id": "s-heppner.com/semanticID/one", |
| 280 | + "match_semantic_id": "s-heppner.com/semanticID/1", |
| 281 | + "score": 1.0, |
| 282 | + "meta_information": {"matchSource": "Defined by Sebastian Heppner"} |
| 283 | + }, |
| 284 | + { |
| 285 | + "base_semantic_id": "s-heppner.com/semanticID/one", |
| 286 | + "match_semantic_id": "s-heppner.com/semanticID/two", |
| 287 | + "score": 0.8, |
| 288 | + "meta_information": {"matchSource": "Defined by Sebastian Heppner"} |
| 289 | + }, |
| 290 | + { |
| 291 | + "base_semantic_id": "s-heppner.com/semanticID/one", |
| 292 | + "match_semantic_id": "s-heppner.com/semanticID/2", |
| 293 | + "score": 0.8, |
| 294 | + "meta_information": {"matchSource": "Defined by Sebastian Heppner", |
| 295 | + "path": ["s-heppner.com/semanticID/two"]} |
| 296 | + } |
| 297 | + ] |
| 298 | + } |
| 299 | + actual_matches = response.json() |
| 300 | + self.assertEqual(expected_matches, actual_matches) |
| 301 | + |
| 302 | + def test_remove_all_matches(self): |
| 303 | + with run_server_context(): |
| 304 | + requests.post("http://localhost:8000/clear") |
| 305 | + response = requests.get("http://localhost:8000/all_matches") |
| 306 | + expected_matches = {} |
| 307 | + actual_matches = response.json() |
| 308 | + self.assertEqual(expected_matches, actual_matches) |
| 309 | + |
| 310 | + |
| 311 | +if __name__ == '__main__': |
| 312 | + unittest.main() |
0 commit comments