contextsynapse 1.0.0__py3-none-any.whl

contextsynapse/tests/regression/cases/test_rag_multi_hop_graph.yaml

@@ -0,0 +1,32 @@
+id: test_rag_005
+name: RAG Query - Multi-Hop Graph Search
+description: 'Tests RAG query - RAG automatically includes graph search in hybrid
+  search.
+
+  Uses simple RAG syntax.
+
+  '
+query: RAG "What is the relationship chain from TCS to its customers?" AT NAMESPACE
+  tcs_regression_test
+namespace: tcs_regression_test_graph
+expected_success: true
+tags:
+- rag
+- graph_traversal
+- multi_hop
+- gpt-3.5
+verify_persistence: false
+test_both_modes: true
+approved_by: system
+notes: 'Tests simple RAG query.
+
+  RAG automatically includes graph search in hybrid search.
+
+  NOTE: Explicit multi-hop and depth parameters are proposed but not yet implemented.
+
+  '
+priority: 5
+prerequisites:
+- RUN_INGESTION_PIPELINE
+requires_data: true
+requires_files: []

contextsynapse/tests/regression/cases/test_rag_path_based.yaml

@@ -0,0 +1,34 @@
+id: test_rag_005
+name: RAG Query - Path-Based Graph Search
+description: 'Tests RAG query with path-based graph search:
+
+  - Finds specific paths between entities
+
+  - Uses path patterns for context
+
+  - Combines paths with vector search
+
+
+  Uses GPT-3.5-Turbo with custom prompt and embedding model.
+
+  '
+query: RAG "How is TCS connected to Microsoft?" AT NAMESPACE regression_test
+namespace: regression_test_graph
+expected_success: true
+tags:
+- rag
+- graph_search
+- path_based
+verify_persistence: false
+test_both_modes: true
+approved_by: system
+notes: 'Tests RAG query for path-based relationship discovery.
+
+  Note: Path finding should be configured via pipeline.
+
+  '
+priority: 5
+prerequisites:
+- RUN_INGESTION_PIPELINE
+requires_data: true
+requires_files: []

contextsynapse/tests/regression/cases/test_rag_profile_based.yaml

@@ -0,0 +1,32 @@
+id: test_rag_007
+name: RAG Query - Profile-Based Hybrid Search
+description: 'Tests RAG query - RAG automatically uses hybrid search with default
+  profile.
+
+  Uses simple RAG syntax.
+
+  '
+query: RAG "What are the strategic initiatives?" AT NAMESPACE tcs_regression_test
+namespace: tcs_regression_test
+expected_success: true
+tags:
+- rag
+- hybrid_search
+- profile_based
+- gpt-3.5
+verify_persistence: false
+test_both_modes: true
+approved_by: system
+notes: 'Tests simple RAG query.
+
+  RAG automatically uses hybrid search with default profile.
+
+  NOTE: Profile selection and embedding model parameters are proposed but not yet
+  implemented.
+
+  '
+priority: 5
+prerequisites:
+- RUN_INGESTION_PIPELINE
+requires_data: true
+requires_files: []

contextsynapse/tests/regression/cases/test_rag_vector_search.yaml

@@ -0,0 +1,32 @@
+id: test_rag_003
+name: RAG Query - Vector Search
+description: 'Tests RAG query - RAG automatically includes dense search in hybrid
+  search.
+
+  Uses simple RAG syntax.
+
+  '
+query: RAG "What are the key financial metrics?" AT NAMESPACE tcs_regression_test
+namespace: tcs_regression_test
+expected_success: true
+tags:
+- rag
+- vector_search
+- dense_search
+- gpt-3.5
+verify_persistence: false
+test_both_modes: true
+approved_by: system
+notes: 'Tests simple RAG query.
+
+  RAG automatically includes dense search in hybrid search.
+
+  NOTE: Explicit search type and embedding model parameters are proposed but not yet
+  implemented.
+
+  '
+priority: 5
+prerequisites:
+- RUN_INGESTION_PIPELINE
+requires_data: true
+requires_files: []

contextsynapse/tests/regression/cases/test_rag_weighted_graph.yaml

@@ -0,0 +1,34 @@
+id: test_rag_004
+name: RAG Query - Weighted Graph Search
+description: 'Tests RAG query with weighted graph search:
+
+  - Edge weights influence traversal
+
+  - Weighted scoring for relevance
+
+  - Combines with vector search
+
+
+  Uses GPT-3.5-Turbo with custom prompt and embedding model.
+
+  '
+query: RAG "What are the strongest business relationships?" AT NAMESPACE regression_test
+namespace: regression_test_graph
+expected_success: true
+tags:
+- rag
+- graph_search
+- weighted
+verify_persistence: false
+test_both_modes: true
+approved_by: system
+notes: 'Tests RAG query for weighted relationship analysis.
+
+  Note: Weight-based filtering should be configured via pipeline.
+
+  '
+priority: 5
+prerequisites:
+- RUN_INGESTION_PIPELINE
+requires_data: true
+requires_files: []

contextsynapse/tests/regression/cases/test_rag_weighted_hybrid.yaml

@@ -0,0 +1,31 @@
+id: test_rag_006
+name: RAG Query - Weighted Hybrid Search
+description: 'Tests RAG query - RAG automatically uses hybrid search with default
+  weights.
+
+  Uses simple RAG syntax.
+
+  '
+query: RAG "What are the main business segments?" AT NAMESPACE tcs_regression_test
+namespace: tcs_regression_test
+expected_success: true
+tags:
+- rag
+- hybrid_search
+- weighted
+- gpt-3.5
+verify_persistence: false
+test_both_modes: true
+approved_by: system
+notes: 'Tests simple RAG query.
+
+  RAG automatically uses hybrid search with default weights.
+
+  NOTE: Custom weights and embedding model parameters are proposed but not yet implemented.
+
+  '
+priority: 5
+prerequisites:
+- RUN_INGESTION_PIPELINE
+requires_data: true
+requires_files: []

contextsynapse/tests/regression/cases/test_semantic_boundary_chunking.yaml

@@ -0,0 +1,57 @@
+id: test_semantic_boundary_001
+name: Semantic Boundary - Chunking with Boundaries
+description: 'Tests ingestion pipeline with semantic boundary detection for chunking.
+
+  Uses semantic boundary detection to preserve topic boundaries.
+
+  Uses GPT-3.5-Turbo for chunking.
+
+  Uses text-embedding-ada-002 for embeddings.
+
+  '
+query: "CREATE PIPELINE regression_semantic_boundary_pipeline\nIN NAMESPACE semantic_boundary_test\n\
+  SOURCE COLLECTION raw_docs\nTARGET COLLECTION knowledge_graph\nDESCRIPTION \"Regression\
+  \ test: Semantic boundary chunking (25-30 pages)\"\nEXECUTION_MODE AUTO\nSTAGES\
+  \ = [\n  STEP extract\n      EXTRACT FROM FILE \"input-doc/annual-report-2024-2025.pdf\"\
+  \n      USING READER \"AUTO\"\n      DETECT (TEXT, TABLES)\n      PARSE_METADATA\
+  \ TRUE\n      MAX_PAGES 30,\n  \n  STEP chunk\n      CHUNK BY semantic\n      USING\
+  \ MODEL \"gpt-3.5-turbo\"\n      PROMPT \"Detect semantic boundaries and split text\
+  \ into chunks preserving topic continuity. Each chunk should be coherent and self-contained.\"\
+  \n      PARAMETERS (\n          max_tokens: 1000,\n          overlap: 100,\n   \
+  \       similarity_threshold: 0.7,\n          detect_boundaries: TRUE,\n       \
+  \   preserve_boundaries: TRUE\n      )\n      STORE AS NODE TYPE Chunk\n      LINK\
+  \ TO Document ON HAS_CHUNK,\n  \n  STEP embed\n      EMBED USING MODEL \"text-embedding-ada-002\"\
+  \n      PARAMETERS (dimensions: 1536, normalize: TRUE)\n      STORE EMBEDDING IN\
+  \ Chunk.embedding,\n  \n  STEP index\n      INDEX USING {bm25: TRUE, vector: TRUE,\
+  \ graph: TRUE}\n      BUILD_INDEX \"regression_semantic_boundary_index\"\n     \
+  \ STORE IN COLLECTION knowledge_graph\n]"
+namespace: semantic_boundary_test
+expected_success: true
+expected_nodes: 50
+expected_edges: 10
+tags:
+- semantic_boundary
+- chunking
+- pipeline
+- gpt-3.5
+- embedding
+verify_persistence: true
+test_both_modes: true
+approved_by: system
+notes: 'Tests CREATE PIPELINE with semantic boundary detection.
+
+  Includes detect_boundaries and preserve_boundaries parameters.
+
+  Includes prompt for chunking model.
+
+  Includes explicit embedding model.
+
+  Verifies semantic boundaries are preserved in chunks.
+
+  '
+skip: true
+skip_reason: Semantic boundary detection not fully implemented
+priority: 6
+prerequisites: []
+requires_data: false
+requires_files: []

contextsynapse/tests/regression/cases/test_semantic_boundary_detection.yaml

@@ -0,0 +1,33 @@
+id: test_semantic_boundary_002
+name: Semantic Boundary - Detect Boundaries
+description: 'Tests semantic boundary detection on text.
+
+  Detects topic, section, paragraph, and sentence boundaries.
+
+  '
+query: "CREATE NODE Document { \n  title: \"Test Document\",\n  content: \"This is\
+  \ the first paragraph. It discusses artificial intelligence. Machine learning is\
+  \ a subset of AI. This is the second paragraph. It discusses natural language processing.\
+  \ NLP enables computers to understand human language.\"\n};\n\nSELECT * FROM Document\
+  \ WHERE title = \"Test Document\""
+namespace: semantic_boundary_test
+expected_success: true
+tags:
+- semantic_boundary
+- detection
+verify_persistence: false
+test_both_modes: true
+approved_by: system
+notes: 'Tests semantic boundary detection.
+
+  Should detect boundaries in document content.
+
+  NOTE: Actual boundary detection may be performed by pipeline stage or separate function.
+
+  '
+skip: true
+skip_reason: Semantic boundary detection not fully implemented
+priority: 6
+prerequisites: []
+requires_data: false
+requires_files: []

contextsynapse/tests/regression/cases/test_semantic_hash_function.yaml

@@ -0,0 +1,30 @@
+id: test_semantic_hash_001
+name: Semantic Hash - SEMANTIC_HASH Function
+description: 'Tests SEMANTIC_HASH function to generate semantic hash of text.
+
+  Uses default model if not specified.
+
+  '
+query: SELECT SEMANTIC_HASH("This is a test document about artificial intelligence
+  and machine learning")
+namespace: semantic_hash_test
+expected_success: true
+tags:
+- semantic_hash
+- function
+verify_persistence: false
+test_both_modes: true
+approved_by: system
+notes: 'Tests SEMANTIC_HASH function.
+
+  Should generate semantic hash of the input text.
+
+  Uses default model if not specified.
+
+  '
+skip: true
+skip_reason: SEMANTIC_HASH operator not implemented
+priority: 6
+prerequisites: []
+requires_data: false
+requires_files: []

contextsynapse/tests/regression/cases/test_semantic_hash_with_model.yaml

@@ -0,0 +1,31 @@
+id: test_semantic_hash_002
+name: Semantic Hash - SEMANTIC_HASH with Model
+description: 'Tests SEMANTIC_HASH function with explicit embedding model.
+
+  Uses text-embedding-ada-002 for hashing.
+
+  '
+query: SELECT SEMANTIC_HASH("This is a test document about artificial intelligence
+  and machine learning") USING MODEL "text-embedding-ada-002"
+namespace: semantic_hash_test
+expected_success: true
+tags:
+- semantic_hash
+- function
+- embedding_model
+verify_persistence: false
+test_both_modes: true
+approved_by: system
+notes: 'Tests SEMANTIC_HASH function with explicit model.
+
+  Should generate semantic hash using specified embedding model.
+
+  Uses text-embedding-ada-002.
+
+  '
+skip: true
+skip_reason: SEMANTIC_HASH operator not implemented
+priority: 6
+prerequisites: []
+requires_data: false
+requires_files: []

contextsynapse/tests/regression/cases/test_tcs_ingestion_fixed_size.yaml

@@ -0,0 +1,44 @@
+id: test_tcs_002
+name: TCS Ingestion - Fixed Size Chunking (GPT-3.5)
+description: 'Ingests TCS annual report using fixed-size chunking strategy.
+
+  Uses GPT-3.5-Turbo for high-volume operations.
+
+  Processes first 25-30 pages.
+
+
+  Pipeline:
+
+  1. Extract PDF text (25-30 pages)
+
+  2. Fixed-size chunking (1000 tokens, 200 overlap)
+
+  3. Create Document and Chunk nodes
+
+  4. Create HAS_CHUNK edges
+
+  5. Generate embeddings
+
+  '
+namespace: tcs_regression_test
+expected_success: true
+expected_nodes: 50
+query: ''
+tags:
+- tcs
+- ingestion
+- fixed_size
+- gpt-3.5
+verify_persistence: true
+test_both_modes: true
+approved_by: system
+notes: 'Tests fixed-size chunking strategy for TCS ingestion.
+
+  Uses GPT-3.5-Turbo as default for high-volume operations.
+
+  '
+priority: 4
+prerequisites: []
+requires_data: false
+requires_files:
+- input-doc/annual-report-2024-2025.pdf

contextsynapse/tests/regression/cases/test_tcs_ingestion_recursive.yaml

@@ -0,0 +1,46 @@
+id: test_tcs_001
+name: TCS Ingestion - Recursive Chunking (GPT-3.5)
+description: 'Ingests TCS annual report using recursive chunking strategy.
+
+  Uses GPT-3.5-Turbo for high-volume operations.
+
+  Processes first 25-30 pages.
+
+
+  Pipeline:
+
+  1. Extract PDF text (25-30 pages)
+
+  2. Recursive chunking (semantic boundaries)
+
+  3. Create Document and Chunk nodes
+
+  4. Create HAS_CHUNK edges
+
+  5. Generate embeddings
+
+  '
+namespace: tcs_regression_test
+expected_success: true
+expected_nodes: 50
+query: ''
+tags:
+- tcs
+- ingestion
+- recursive
+- gpt-3.5
+verify_persistence: true
+test_both_modes: true
+approved_by: system
+notes: 'This test verifies the structure for TCS ingestion with recursive chunking.
+
+  Actual chunking should be done via ingestion script that processes 25-30 pages.
+
+  Uses GPT-3.5-Turbo as default for high-volume operations.
+
+  '
+priority: 4
+prerequisites: []
+requires_data: false
+requires_files:
+- input-doc/annual-report-2024-2025.pdf

contextsynapse/tests/regression/cases/test_tcs_ingestion_section.yaml

@@ -0,0 +1,46 @@
+id: test_tcs_004
+name: TCS Ingestion - Section-Based Chunking (GPT-3.5)
+description: 'Ingests TCS annual report using section-based chunking strategy.
+
+  Uses GPT-3.5-Turbo for high-volume operations.
+
+  Processes first 25-30 pages.
+
+
+  Pipeline:
+
+  1. Extract PDF text (25-30 pages)
+
+  2. Section-based chunking (preserves document structure)
+
+  3. Create Document and Chunk nodes
+
+  4. Create HAS_CHUNK edges
+
+  5. Generate embeddings
+
+  '
+namespace: tcs_regression_test
+expected_success: true
+expected_nodes: 50
+query: ''
+tags:
+- tcs
+- ingestion
+- section_based
+- gpt-3.5
+verify_persistence: true
+test_both_modes: true
+approved_by: system
+notes: 'Tests section-based chunking strategy for TCS ingestion.
+
+  Preserves document structure by splitting at section boundaries.
+
+  Uses GPT-3.5-Turbo as default for high-volume operations.
+
+  '
+priority: 4
+prerequisites: []
+requires_data: false
+requires_files:
+- input-doc/annual-report-2024-2025.pdf

contextsynapse/tests/regression/cases/test_tcs_ingestion_semantic.yaml

@@ -0,0 +1,44 @@
+id: test_tcs_003
+name: TCS Ingestion - Semantic Chunking (GPT-3.5)
+description: 'Ingests TCS annual report using semantic chunking strategy.
+
+  Uses GPT-3.5-Turbo for high-volume operations.
+
+  Processes first 25-30 pages.
+
+
+  Pipeline:
+
+  1. Extract PDF text (25-30 pages)
+
+  2. Semantic chunking (preserves meaning)
+
+  3. Create Document and Chunk nodes
+
+  4. Create HAS_CHUNK edges
+
+  5. Generate embeddings
+
+  '
+namespace: tcs_regression_test
+expected_success: true
+expected_nodes: 50
+query: ''
+tags:
+- tcs
+- ingestion
+- semantic
+- gpt-3.5
+verify_persistence: true
+test_both_modes: true
+approved_by: system
+notes: 'Tests semantic chunking strategy for TCS ingestion.
+
+  Uses GPT-3.5-Turbo as default for high-volume operations.
+
+  '
+priority: 4
+prerequisites: []
+requires_data: false
+requires_files:
+- input-doc/annual-report-2024-2025.pdf

contextsynapse/tests/regression/cases/test_time_travel_as_of.yaml

@@ -0,0 +1,35 @@
+id: test_timetravel_002
+name: Time Travel - Query AS OF TIMESTAMP
+description: 'Tests SELECT query with AS OF TIMESTAMP clause for time-travel queries.
+
+  Requires temporal queries to be enabled.
+
+  '
+query: 'CREATE NODE Person { name: "Bob", age: 25 };
+
+  UPDATE NODE Person SET { age: 26 } WHERE name = "Bob";
+
+
+  SELECT * FROM Person AS OF TIMESTAMP ''2024-06-01T00:00:00Z'' WHERE name = "Bob"'
+namespace: timetravel_test
+expected_success: true
+tags:
+- time_travel
+- temporal
+- as_of
+verify_persistence: false
+test_both_modes: true
+approved_by: system
+notes: 'Tests SELECT query with AS OF TIMESTAMP clause.
+
+  Requires temporal queries to be enabled.
+
+  Should return node state as of specified timestamp.
+
+  '
+skip: true
+skip_reason: Temporal queries not fully implemented
+priority: 6
+prerequisites: []
+requires_data: false
+requires_files: []

contextsynapse/tests/regression/cases/test_time_travel_at_commit.yaml

@@ -0,0 +1,33 @@
+id: test_timetravel_003
+name: Time Travel - Query AT COMMIT
+description: 'Tests SELECT query with AT COMMIT clause for version-based queries.
+
+  Requires temporal queries to be enabled.
+
+  '
+query: 'CREATE NODE Document { title: "Report v1", content: "Initial content" };
+
+
+  SELECT * FROM Document AT COMMIT ''commit_12345'' WHERE title = "Report v1"'
+namespace: timetravel_test
+expected_success: true
+tags:
+- time_travel
+- temporal
+- commit
+verify_persistence: false
+test_both_modes: true
+approved_by: system
+notes: 'Tests SELECT query with AT COMMIT clause.
+
+  Requires temporal queries to be enabled.
+
+  Should return node state at specified commit.
+
+  '
+skip: true
+skip_reason: Temporal queries not fully implemented
+priority: 6
+prerequisites: []
+requires_data: false
+requires_files: []

contextsynapse/tests/regression/cases/test_time_travel_at_timestamp.yaml

@@ -0,0 +1,33 @@
+id: test_timetravel_001
+name: Time Travel - Query AT TIMESTAMP
+description: 'Tests SELECT query with AT TIMESTAMP clause for time-travel queries.
+
+  Requires temporal queries to be enabled.
+
+  '
+query: 'CREATE NODE Person { name: "Alice", age: 30 };
+
+
+  SELECT * FROM Person AT TIMESTAMP ''2024-01-01T00:00:00Z'' WHERE name = "Alice"'
+namespace: timetravel_test
+expected_success: true
+tags:
+- time_travel
+- temporal
+- timestamp
+verify_persistence: false
+test_both_modes: true
+approved_by: system
+notes: 'Tests SELECT query with AT TIMESTAMP clause.
+
+  Requires temporal queries to be enabled.
+
+  Should return node state at specified timestamp.
+
+  '
+skip: false
+skip_reason: ''
+priority: 6
+prerequisites: []
+requires_data: false
+requires_files: []

contextsynapse/tests/regression/cases/test_time_travel_for_system_time_all.yaml

@@ -0,0 +1,37 @@
+id: test_timetravel_004
+name: Time Travel - Query FOR SYSTEM_TIME ALL
+description: 'Tests SELECT query with FOR SYSTEM_TIME ALL clause to get all versions.
+
+  Requires temporal queries to be enabled.
+
+  '
+query: 'CREATE NODE Person { name: "Charlie", age: 20 };
+
+  UPDATE NODE Person SET { age: 21 } WHERE name = "Charlie";
+
+  UPDATE NODE Person SET { age: 22, city: "NYC" } WHERE name = "Charlie";
+
+
+  SELECT * FROM Person FOR SYSTEM_TIME ALL WHERE name = "Charlie"'
+namespace: timetravel_test
+expected_success: true
+tags:
+- time_travel
+- temporal
+- versions
+verify_persistence: false
+test_both_modes: true
+approved_by: system
+notes: 'Tests SELECT query with FOR SYSTEM_TIME ALL clause.
+
+  Requires temporal queries to be enabled.
+
+  Should return all versions of the node.
+
+  '
+skip: true
+skip_reason: Temporal queries not fully implemented
+priority: 6
+prerequisites: []
+requires_data: false
+requires_files: []