Merge branch 'dev' into feature/cog-971-preparing-swe-bench-run

cognee/infrastructure/databases/graph/neo4j_driver/adapter.py

@@ -493,7 +493,7 @@ class Neo4jAdapter(GraphDBInterface):
 
         query_edges = f"""
         MATCH (n)-[r]->(m)
-        WHERE {where_clause} AND {where_clause.replace('n.', 'm.')}
+        WHERE {where_clause} AND {where_clause.replace("n.", "m.")}
         RETURN ID(n) AS source, ID(m) AS target, TYPE(r) AS type, properties(r) AS properties
         """
         result_edges = await self.query(query_edges)

cognee/modules/retrieval/brute_force_triplet_search.py

@@ -43,7 +43,7 @@ def format_triplets(edges):
         edge_info = {key: value for key, value in edge_attributes.items() if value is not None}
 
         # Create the formatted triplet
-        triplet = f"Node1: {node1_info}\n" f"Edge: {edge_info}\n" f"Node2: {node2_info}\n\n\n"
+        triplet = f"Node1: {node1_info}\nEdge: {edge_info}\nNode2: {node2_info}\n\n\n"
         triplets.append(triplet)
 
     return "".join(triplets)

cognee/modules/retrieval/description_to_codepart_search.py

@@ -75,8 +75,7 @@ async def code_description_to_code_part(
 
             llm_client = get_llm_client()
             context_from_documents = await llm_client.acreate_structured_output(
-                text_input=f"The retrieved context from documents"
+                text_input=f"The retrieved context from documents is {concatenated_descriptions}.",
-                f" is {concatenated_descriptions}.",
                 system_prompt="You are a Senior Software Engineer, summarize the context from documents"
                 f" in a way that it is gonna be provided next to codeparts as context"
                 f" while trying to solve this github issue connected to the project: {query}]",

cognee/tests/integration/documents/AudioDocument_test.py

@@ -36,12 +36,12 @@ def test_AudioDocument():
         for ground_truth, paragraph_data in zip(
             GROUND_TRUTH, document.read(chunk_size=64, chunker="text_chunker")
         ):
-            assert (
+            assert ground_truth["word_count"] == paragraph_data.word_count, (
-                ground_truth["word_count"] == paragraph_data.word_count
+                f'{ground_truth["word_count"] = } != {paragraph_data.word_count = }'
-            ), f'{ground_truth["word_count"] = } != {paragraph_data.word_count = }'
+            )
-            assert ground_truth["len_text"] == len(
+            assert ground_truth["len_text"] == len(paragraph_data.text), (
-                paragraph_data.text
+                f'{ground_truth["len_text"] = } != {len(paragraph_data.text) = }'
-            ), f'{ground_truth["len_text"] = } != {len(paragraph_data.text) = }'
+            )
-            assert (
+            assert ground_truth["cut_type"] == paragraph_data.cut_type, (
-                ground_truth["cut_type"] == paragraph_data.cut_type
+                f'{ground_truth["cut_type"] = } != {paragraph_data.cut_type = }'
-            ), f'{ground_truth["cut_type"] = } != {paragraph_data.cut_type = }'
+            )

cognee/tests/integration/documents/ImageDocument_test.py

@@ -25,12 +25,12 @@ def test_ImageDocument():
         for ground_truth, paragraph_data in zip(
             GROUND_TRUTH, document.read(chunk_size=64, chunker="text_chunker")
         ):
-            assert (
+            assert ground_truth["word_count"] == paragraph_data.word_count, (
-                ground_truth["word_count"] == paragraph_data.word_count
+                f'{ground_truth["word_count"] = } != {paragraph_data.word_count = }'
-            ), f'{ground_truth["word_count"] = } != {paragraph_data.word_count = }'
+            )
-            assert ground_truth["len_text"] == len(
+            assert ground_truth["len_text"] == len(paragraph_data.text), (
-                paragraph_data.text
+                f'{ground_truth["len_text"] = } != {len(paragraph_data.text) = }'
-            ), f'{ground_truth["len_text"] = } != {len(paragraph_data.text) = }'
+            )
-            assert (
+            assert ground_truth["cut_type"] == paragraph_data.cut_type, (
-                ground_truth["cut_type"] == paragraph_data.cut_type
+                f'{ground_truth["cut_type"] = } != {paragraph_data.cut_type = }'
-            ), f'{ground_truth["cut_type"] = } != {paragraph_data.cut_type = }'
+            )

cognee/tests/integration/documents/PdfDocument_test.py

@@ -27,12 +27,12 @@ def test_PdfDocument():
     for ground_truth, paragraph_data in zip(
         GROUND_TRUTH, document.read(chunk_size=1024, chunker="text_chunker")
     ):
-        assert (
+        assert ground_truth["word_count"] == paragraph_data.word_count, (
-            ground_truth["word_count"] == paragraph_data.word_count
+            f'{ground_truth["word_count"] = } != {paragraph_data.word_count = }'
-        ), f'{ground_truth["word_count"] = } != {paragraph_data.word_count = }'
+        )
-        assert ground_truth["len_text"] == len(
+        assert ground_truth["len_text"] == len(paragraph_data.text), (
-            paragraph_data.text
+            f'{ground_truth["len_text"] = } != {len(paragraph_data.text) = }'
-        ), f'{ground_truth["len_text"] = } != {len(paragraph_data.text) = }'
+        )
-        assert (
+        assert ground_truth["cut_type"] == paragraph_data.cut_type, (
-            ground_truth["cut_type"] == paragraph_data.cut_type
+            f'{ground_truth["cut_type"] = } != {paragraph_data.cut_type = }'
-        ), f'{ground_truth["cut_type"] = } != {paragraph_data.cut_type = }'
+        )

cognee/tests/integration/documents/TextDocument_test.py

@@ -39,12 +39,12 @@ def test_TextDocument(input_file, chunk_size):
     for ground_truth, paragraph_data in zip(
         GROUND_TRUTH[input_file], document.read(chunk_size=chunk_size, chunker="text_chunker")
     ):
-        assert (
+        assert ground_truth["word_count"] == paragraph_data.word_count, (
-            ground_truth["word_count"] == paragraph_data.word_count
+            f'{ground_truth["word_count"] = } != {paragraph_data.word_count = }'
-        ), f'{ground_truth["word_count"] = } != {paragraph_data.word_count = }'
+        )
-        assert ground_truth["len_text"] == len(
+        assert ground_truth["len_text"] == len(paragraph_data.text), (
-            paragraph_data.text
+            f'{ground_truth["len_text"] = } != {len(paragraph_data.text) = }'
-        ), f'{ground_truth["len_text"] = } != {len(paragraph_data.text) = }'
+        )
-        assert (
+        assert ground_truth["cut_type"] == paragraph_data.cut_type, (
-            ground_truth["cut_type"] == paragraph_data.cut_type
+            f'{ground_truth["cut_type"] = } != {paragraph_data.cut_type = }'
-        ), f'{ground_truth["cut_type"] = } != {paragraph_data.cut_type = }'
+        )

cognee/tests/integration/documents/UnstructuredDocument_test.py

@@ -71,32 +71,32 @@ def test_UnstructuredDocument():
     for paragraph_data in pptx_document.read(chunk_size=1024, chunker="text_chunker"):
         assert 19 == paragraph_data.word_count, f" 19 != {paragraph_data.word_count = }"
         assert 104 == len(paragraph_data.text), f" 104 != {len(paragraph_data.text) = }"
-        assert (
+        assert "sentence_cut" == paragraph_data.cut_type, (
-            "sentence_cut" == paragraph_data.cut_type
+            f" sentence_cut != {paragraph_data.cut_type = }"
-        ), f" sentence_cut != {paragraph_data.cut_type = }"
+        )
 
     # Test DOCX
     for paragraph_data in docx_document.read(chunk_size=1024, chunker="text_chunker"):
         assert 16 == paragraph_data.word_count, f" 16 != {paragraph_data.word_count = }"
         assert 145 == len(paragraph_data.text), f" 145 != {len(paragraph_data.text) = }"
-        assert (
+        assert "sentence_end" == paragraph_data.cut_type, (
-            "sentence_end" == paragraph_data.cut_type
+            f" sentence_end != {paragraph_data.cut_type = }"
-        ), f" sentence_end != {paragraph_data.cut_type = }"
+        )
 
     # TEST CSV
     for paragraph_data in csv_document.read(chunk_size=1024, chunker="text_chunker"):
         assert 15 == paragraph_data.word_count, f" 15 != {paragraph_data.word_count = }"
-        assert (
+        assert "A A A A A A A A A,A A A A A A,A A" == paragraph_data.text, (
-            "A A A A A A A A A,A A A A A A,A A" == paragraph_data.text
+            f"Read text doesn't match expected text: {paragraph_data.text}"
-        ), f"Read text doesn't match expected text: {paragraph_data.text}"
+        )
-        assert (
+        assert "sentence_cut" == paragraph_data.cut_type, (
-            "sentence_cut" == paragraph_data.cut_type
+            f" sentence_cut != {paragraph_data.cut_type = }"
-        ), f" sentence_cut != {paragraph_data.cut_type = }"
+        )
 
     # Test XLSX
     for paragraph_data in xlsx_document.read(chunk_size=1024, chunker="text_chunker"):
         assert 36 == paragraph_data.word_count, f" 36 != {paragraph_data.word_count = }"
         assert 171 == len(paragraph_data.text), f" 171 != {len(paragraph_data.text) = }"
-        assert (
+        assert "sentence_cut" == paragraph_data.cut_type, (
-            "sentence_cut" == paragraph_data.cut_type
+            f" sentence_cut != {paragraph_data.cut_type = }"
-        ), f" sentence_cut != {paragraph_data.cut_type = }"
+        )

cognee/tests/test_deduplication.py

@@ -30,9 +30,9 @@ async def test_deduplication():
 
     result = await relational_engine.get_all_data_from_table("data")
     assert len(result) == 1, "More than one data entity was found."
-    assert (
+    assert result[0]["name"] == "Natural_language_processing_copy", (
-        result[0]["name"] == "Natural_language_processing_copy"
+        "Result name does not match expected value."
-    ), "Result name does not match expected value."
+    )
 
     result = await relational_engine.get_all_data_from_table("datasets")
     assert len(result) == 2, "Unexpected number of datasets found."
@@ -61,9 +61,9 @@ async def test_deduplication():
 
     result = await relational_engine.get_all_data_from_table("data")
     assert len(result) == 1, "More than one data entity was found."
-    assert (
+    assert hashlib.md5(text.encode("utf-8")).hexdigest() in result[0]["name"], (
-        hashlib.md5(text.encode("utf-8")).hexdigest() in result[0]["name"]
+        "Content hash is not a part of file name."
-    ), "Content hash is not a part of file name."
+    )
 
     await cognee.prune.prune_data()
     await cognee.prune.prune_system(metadata=True)

cognee/tests/test_falkordb.py

@@ -85,9 +85,9 @@ async def main():
 
     from cognee.infrastructure.databases.relational import get_relational_engine
 
-    assert not os.path.exists(
+    assert not os.path.exists(get_relational_engine().db_path), (
-        get_relational_engine().db_path
+        "SQLite relational database is not empty"
-    ), "SQLite relational database is not empty"
+    )
 
     from cognee.infrastructure.databases.graph import get_graph_config
 

cognee/tests/test_library.py

@@ -82,9 +82,9 @@ async def main():
 
     from cognee.infrastructure.databases.relational import get_relational_engine
 
-    assert not os.path.exists(
+    assert not os.path.exists(get_relational_engine().db_path), (
-        get_relational_engine().db_path
+        "SQLite relational database is not empty"
-    ), "SQLite relational database is not empty"
+    )
 
     from cognee.infrastructure.databases.graph import get_graph_config
 

cognee/tests/test_pgvector.py

@@ -24,28 +24,28 @@ async def test_local_file_deletion(data_text, file_location):
         data_hash = hashlib.md5(encoded_text).hexdigest()
         # Get data entry from database based on hash contents
         data = (await session.scalars(select(Data).where(Data.content_hash == data_hash))).one()
-        assert os.path.isfile(
+        assert os.path.isfile(data.raw_data_location), (
-            data.raw_data_location
+            f"Data location doesn't exist: {data.raw_data_location}"
-        ), f"Data location doesn't exist: {data.raw_data_location}"
+        )
         # Test deletion of data along with local files created by cognee
         await engine.delete_data_entity(data.id)
-        assert not os.path.exists(
+        assert not os.path.exists(data.raw_data_location), (
-            data.raw_data_location
+            f"Data location still exists after deletion: {data.raw_data_location}"
-        ), f"Data location still exists after deletion: {data.raw_data_location}"
+        )
 
     async with engine.get_async_session() as session:
         # Get data entry from database based on file path
         data = (
             await session.scalars(select(Data).where(Data.raw_data_location == file_location))
         ).one()
-        assert os.path.isfile(
+        assert os.path.isfile(data.raw_data_location), (
-            data.raw_data_location
+            f"Data location doesn't exist: {data.raw_data_location}"
-        ), f"Data location doesn't exist: {data.raw_data_location}"
+        )
         # Test local files not created by cognee won't get deleted
         await engine.delete_data_entity(data.id)
-        assert os.path.exists(
+        assert os.path.exists(data.raw_data_location), (
-            data.raw_data_location
+            f"Data location doesn't exists: {data.raw_data_location}"
-        ), f"Data location doesn't exists: {data.raw_data_location}"
+        )
 
 
 async def test_getting_of_documents(dataset_name_1):
@@ -54,16 +54,16 @@ async def test_getting_of_documents(dataset_name_1):
 
     user = await get_default_user()
     document_ids = await get_document_ids_for_user(user.id, [dataset_name_1])
-    assert (
+    assert len(document_ids) == 1, (
-        len(document_ids) == 1
+        f"Number of expected documents doesn't match {len(document_ids)} != 1"
-    ), f"Number of expected documents doesn't match {len(document_ids)} != 1"
+    )
 
     # Test getting of documents for search when no dataset is provided
     user = await get_default_user()
     document_ids = await get_document_ids_for_user(user.id)
-    assert (
+    assert len(document_ids) == 2, (
-        len(document_ids) == 2
+        f"Number of expected documents doesn't match {len(document_ids)} != 2"
-    ), f"Number of expected documents doesn't match {len(document_ids)} != 2"
+    )
 
 
 async def main():

cognee/tests/unit/processing/chunks/chunk_by_paragraph_2_test.py

@@ -17,9 +17,9 @@ batch_paragraphs_vals = [True, False]
 def test_chunk_by_paragraph_isomorphism(input_text, paragraph_length, batch_paragraphs):
     chunks = chunk_by_paragraph(input_text, paragraph_length, batch_paragraphs)
     reconstructed_text = "".join([chunk["text"] for chunk in chunks])
-    assert (
+    assert reconstructed_text == input_text, (
-        reconstructed_text == input_text
+        f"texts are not identical: {len(input_text) = }, {len(reconstructed_text) = }"
-    ), f"texts are not identical: {len(input_text) = }, {len(reconstructed_text) = }"
+    )
 
 
 @pytest.mark.parametrize(
@@ -36,9 +36,9 @@ def test_paragraph_chunk_length(input_text, paragraph_length, batch_paragraphs):
     chunk_lengths = np.array([len(list(chunk_by_word(chunk["text"]))) for chunk in chunks])
 
     larger_chunks = chunk_lengths[chunk_lengths > paragraph_length]
-    assert np.all(
+    assert np.all(chunk_lengths <= paragraph_length), (
-        chunk_lengths <= paragraph_length
+        f"{paragraph_length = }: {larger_chunks} are too large"
-    ), f"{paragraph_length = }: {larger_chunks} are too large"
+    )
 
 
 @pytest.mark.parametrize(
@@ -50,6 +50,6 @@ def test_chunk_by_paragraph_chunk_numbering(input_text, paragraph_length, batch_
         data=input_text, paragraph_length=paragraph_length, batch_paragraphs=batch_paragraphs
     )
     chunk_indices = np.array([chunk["chunk_index"] for chunk in chunks])
-    assert np.all(
+    assert np.all(chunk_indices == np.arange(len(chunk_indices))), (
-        chunk_indices == np.arange(len(chunk_indices))
+        f"{chunk_indices = } are not monotonically increasing"
-    ), f"{chunk_indices = } are not monotonically increasing"
+    )

cognee/tests/unit/processing/chunks/chunk_by_paragraph_test.py

@@ -58,9 +58,9 @@ def run_chunking_test(test_text, expected_chunks):
 
     for expected_chunks_item, chunk in zip(expected_chunks, chunks):
         for key in ["text", "word_count", "cut_type"]:
-            assert (
+            assert chunk[key] == expected_chunks_item[key], (
-                chunk[key] == expected_chunks_item[key]
+                f"{key = }: {chunk[key] = } != {expected_chunks_item[key] = }"
-            ), f"{key = }: {chunk[key] = } != {expected_chunks_item[key] = }"
+            )
 
 
 def test_chunking_whole_text():

cognee/tests/unit/processing/chunks/chunk_by_sentence_test.py

@@ -16,9 +16,9 @@ maximum_length_vals = [None, 8, 64]
 def test_chunk_by_sentence_isomorphism(input_text, maximum_length):
     chunks = chunk_by_sentence(input_text, maximum_length)
     reconstructed_text = "".join([chunk[1] for chunk in chunks])
-    assert (
+    assert reconstructed_text == input_text, (
-        reconstructed_text == input_text
+        f"texts are not identical: {len(input_text) = }, {len(reconstructed_text) = }"
-    ), f"texts are not identical: {len(input_text) = }, {len(reconstructed_text) = }"
+    )
 
 
 @pytest.mark.parametrize(
@@ -36,6 +36,6 @@ def test_paragraph_chunk_length(input_text, maximum_length):
     chunk_lengths = np.array([len(list(chunk_by_word(chunk[1]))) for chunk in chunks])
 
     larger_chunks = chunk_lengths[chunk_lengths > maximum_length]
-    assert np.all(
+    assert np.all(chunk_lengths <= maximum_length), (
-        chunk_lengths <= maximum_length
+        f"{maximum_length = }: {larger_chunks} are too large"
-    ), f"{maximum_length = }: {larger_chunks} are too large"
+    )

cognee/tests/unit/processing/chunks/chunk_by_word_test.py

@@ -17,9 +17,9 @@ from cognee.tests.unit.processing.chunks.test_input import INPUT_TEXTS
 def test_chunk_by_word_isomorphism(input_text):
     chunks = chunk_by_word(input_text)
     reconstructed_text = "".join([chunk[0] for chunk in chunks])
-    assert (
+    assert reconstructed_text == input_text, (
-        reconstructed_text == input_text
+        f"texts are not identical: {len(input_text) = }, {len(reconstructed_text) = }"
-    ), f"texts are not identical: {len(input_text) = }, {len(reconstructed_text) = }"
+    )
 
 
 @pytest.mark.parametrize(