Data Types

Moose supports a comprehensive set of ClickHouse column types across both TypeScript and Python. This section covers all supported types, their syntax, and how they map to ClickHouse.

Type Categories

Basic Types

• Strings — String, LowCardinality(String), UUID
• Integers — Int8 to Int64, UInt8 to UInt64
• Floats — Float32, Float64
• Decimals — Fixed-point Decimal(P,S)
• Booleans — Boolean
• Date & Time — Date, DateTime,

Complex Types

• Arrays — Array(T)
• Maps — Map(K, V)
• Nested — Embedded objects
• Tuples — Named tuples
• Enums — Enumerated values
• Geometry — Point, Ring, Polygon, etc.

Special Types

• JSON — Dynamic and typed JSON columns
• Nullable — Optional fields
• Aggregates — SimpleAggregateFunction for pre-aggregation

Schema Features

• Column Comments — Add descriptions to columns that appear in ClickHouse DDL

Quick Reference

Best Practices

• Use specific integer types when you know the value ranges to save storage
• Prefer Float64 for most floating-point calculations unless storage is critical
• Use LowCardinality for string columns with repeated values (< 10,000 unique)
• Choose appropriate DateTime precision based on your accuracy needs
• Order columns by cardinality (low to high) for better compression
• Add column comments to document field purposes in your schema