Spice MCP

# Standardized test queries for cross-platform consistency test_queries: # Basic functionality queries simple_select: sql: "SELECT 1 as test_col, 'fixed_value' as message" name: "Simple Select Test" expected: rows: 1 columns: 2 column_names: ["test_col", "message"] data_validation: test_col: 1 message: "fixed_value" parametric: sql: "SELECT {{limit_count}} as value, '{{test_string}}' as message" name: "Parameterized Query Test" parameters: limit_count: 5 test_string: "hello_world" expected: rows: 1 columns: 2 column_names: ["value", "message"] data_validation: value: 5 message: "hello_world" # Data type validation queries data_types: sql: | SELECT 1 as int_col, 123.45 as float_col, 'test_string' as string_col, true as bool_col, DOUBLE '123.45' as double_col, DATE '2023-01-01' as date_col, TIMESTAMP '2023-01-01 12:00:00' as timestamp_col, NULL as null_col name: "Data Types Test" expected: rows: 1 columns: 8 column_names: ["int_col", "float_col", "string_col", "bool_col", "double_col", "date_col", "timestamp_col", "null_col"] data_validation: int_col: 1 float_col: 123.45 string_col: "test_string" bool_col: true double_col: 123.45 date_col: "2023-01-01" timestamp_col: "2023-01-01 12:00:00" null_col: null # Complex queries aggregations: sql: | SELECT SUM(value) as total_value, AVG(value) as avg_value, MIN(value) as min_value, MAX(value) as max_value, COUNT(*) as row_count FROM ( SELECT 10 as value UNION SELECT 20 UNION SELECT 30 ) sample_data name: "Aggregation Test" expected: rows: 1 columns: 5 column_names: ["total_value", "avg_value", "min_value", "max_value", "row_count"] data_validation: total_value: 60 avg_value: 20 min_value: 10 max_value: 30 row_count: 3 time_series: sql: | SELECT date_trunc('day', block_time) as day, COUNT(*) as block_count FROM ( SELECT DATE '2023-01-01' + INTERVAL '1 day' * n as block_time FROM ( SELECT generate_series(0, 4) as n ) series ) daily_blocks GROUP BY day ORDER BY day DESC LIMIT 7 name: "Time Series Test" expected: min_rows: 1 max_rows: 7 columns: 2 column_names: ["day", "block_count"] joined_data: sql: | SELECT block_time, block_number, 'ethereum_mainnet' as chain FROM ( SELECT DATE '2023-01-01' as block_time, 1641024000 as block_number ) blocks WHERE block_time >= DATE '2023-01-01' ORDER BY block_time DESC LIMIT 5 name: "Joined Data Test" expected: rows: 1 columns: 3 column_names: ["block_time", "block_number", "chain"] # Stress test queries large_result_set: sql: | SELECT n as row_num, 'row_' || n as row_label, n * 1.5 as float_val, n % 2 = 0 as is_even FROM ( SELECT generate_series(1, 1000) as n ) numbers name: "Large Result Set Test" expected: rows: 1000 columns: 4 max_execution_time: 30.0 cpu_intensive: sql: | SELECT n, SQRT(n) as sqrt_n, POW(n, 2) as n_squared, n * (n + 1) / 2 as triangular_number FROM ( SELECT generate_series(1, 500) as n ) numbers name: "CPU Intensive Test" expected: min_rows: 1 max_execution_time: 45.0 # Error scenario queries error_scenarios: invalid_syntax: sql: "SELECTTTT INVALID SYNTAX" expected_error: "syntax" division_by_zero: sql: "SELECT 1/0 as error_col" expected_error: "division" invalid_table: sql: "SELECT * FROM nonexistent_table_xyz" expected_error: "table.*not.*found" invalid_function: sql: "SELECT nonexistent_function_xyz() as col" expected_error: "function.*not.*found" invalid_parameter: sql: "SELECT {{nonexistent_parameter}} as col" expected_error: "parameter.*not.*found" empty_query: sql: "" expected_error: "empty" very_long_query: sql: | SELECT {% for i in range(1000) %}{{i}} as亦可{{i}}{{"," if not loop.last}}{% endfor %} expected_error: ".*" # Some error expected # Real-world example queries real_world_examples: ethereum_block_stats: name: "Ethereum Block Statistics" description: "Get recent Ethereum block statistics" sql: | SELECT DATE_TRUNC('day', block_time) as day, COUNT(*) as block_count, MIN(block_number) as min_block, MAX(block_number) as max_block, AVG(gas_used) as avg_gas_used FROM ethereum.blocks WHERE block_time >= CURRENT_DATE - INTERVAL '7 days' GROUP BY day ORDER BY day DESC LIMIT 7 requires_blockchain_data: true defi_volume_stats: name: "DeFi Volume Statistics" description: "Mock DeFi volume aggregation" sql: | SELECT 'ethereum' as chain, 'uniswap_v3' as protocol, SUM(volume) as total_volume, COUNT(DISTINCT pool) as pool_count, DATE_TRUNC('day', block_time) as day FROM ( SELECT DATE '2024-01-01' + INTERVAL '1 day' * n as block_time, n * 1000 as volume, 'pool_' || n as pool FROM ( SELECT generate_series(1, 30) as n ) series ) mock_trades WHERE block_time >= CURRENT_DATE - INTERVAL '7 days' GROUP BY day, chain, protocol ORDER BY day DESC expected: min_rows: 1 max_rows: 7 columns: 5 # Test data generation test_data: parameter_sets: limit_tests: - name: "small_limit" parameters: {limit: 5} - name: "medium_limit" parameters: {limit: 50} - name: "large_limit" parameters: {limit: 500} performance_tests: - name: "low_performance" parameters: {performance: "low"} - name: "medium_performance" parameters: {performance: "medium"} - name: "high_performance" parameters: {performance: "large"} timeout_tests: - name: "short_timeout" parameters: {timeout_seconds: 10} - name: "medium_timeout" parameters: {timeout_seconds: 60} - name: "long_timeout" parameters: {timeout_seconds: 300}