ClickHouse-Types

Converting ClickHouse types into other schemas' types

Installation

---

pip install clickhouse-types

Usage

---

PyArrow

from clickhouse_types.pyarrow import dtype_from_string

dtype_from_string('Array(DateTime64(9, Asia/Shanghai))')

>>> list<item: timestamp[ns, tz=Asia/Shanghai]>

from clickhouse_types.pyarrow import field_from_string

field_from_string('HelloWorld Map(UInt128, DateTime64(9, Asia/Shanghai))')

>>> pyarrow.Field<HelloWorld: map<fixed_size_binary[16], timestamp[ns, tz=Asia/Shanghai]>>

from clickhouse_types.pyarrow import schema_from_string

schema_from_string('Hello FixedString(6), World Tuple(Int8, Int8)')

>>> Hello: fixed_size_binary[6]
>>> World: struct<f1: int8, f2: int8>
>>>   child 0, f1: int8
>>>   child 1, f2: int8

SQLAlchemy & ClickHouse-SQLAlchemy

from clickhouse_types.sqlalchemy import type_from_string

type_from_string('DateTime64(9,Asia/Shanghai)')

>>> DateTime64(9, 'Asia/Shanghai')