Graph-Native Infrastructure for Context and Accountable AI Systems
Python agent-memoryaiai-governanceai-infrastructureartificial-intelligencecontext-engineeringcontext-graphsdata-engineeringdecision-intelligencedeveloper-toolsexplainable-aigenerative-aigraph-ragknowledge-graphllmontologyprovenancepythonreasoningsemantic-search
"""EmbeddingGenerator(method=..., model_name=...) used to be ignored (#1808)."""
import unittest
from unittest.mock import patch
from semantica.embeddings.embedding_generator import EmbeddingGenerator
class TestGeneratorMethodKwargs(unittest.TestCase):
def setUp(self):
patchers = [
patch("semantica.embeddings.text_embedder.SentenceTransformer"),
patch("semantica.embeddings.text_embedder.TextEmbedding"),
patch(
"semantica.embeddings.text_embedder.SENTENCE_TRANSFORMERS_AVAILABLE",
True,
),
patch("semantica.embeddings.text_embedder.FASTEMBED_AVAILABLE", True),
]
for p in patchers:
p.start()
self.addCleanup(p.stop)
def test_top_level_method_reaches_the_text_embedder(self):
generator = EmbeddingGenerator(
method="sentence_transformers", model_name="st-model"
)
self.assertEqual(generator.text_embedder.method, "sentence_transformers")
self.assertEqual(generator.text_embedder.model_name, "st-model")
self.assertEqual(generator.get_text_method(), "sentence_transformers")
def test_top_level_method_via_config_dict(self):
generator = EmbeddingGenerator(config={"method": "sentence_transformers"})
self.assertEqual(generator.get_text_method(), "sentence_transformers")
def test_text_config_still_wins(self):
generator = EmbeddingGenerator(
method="sentence_transformers",
text={"method": "fastembed", "model_name": "fast-model"},
)
self.assertEqual(generator.text_embedder.method, "fastembed")
self.assertEqual(generator.text_embedder.model_name, "fast-model")
def test_default_is_unchanged(self):
generator = EmbeddingGenerator()
self.assertEqual(generator.text_embedder.method, "fastembed")
def test_caller_config_is_not_mutated(self):
text = {"model_name": "fast-model"}
EmbeddingGenerator(method="sentence_transformers", text=text)
self.assertEqual(text, {"model_name": "fast-model"})
def test_caller_top_level_config_is_not_mutated(self):
config = {"text": {"model_name": "fast-model"}}
EmbeddingGenerator(config=config, method="sentence_transformers")
self.assertEqual(config, {"text": {"model_name": "fast-model"}})
if __name__ == "__main__":
unittest.main()