pych-client is a ClickHouse client for Python built on top of httpx. It targets the HTTP interface and offers the following features:
- Sync (
ClickHouseClient
) and async (AsyncClickHouseClient
) clients. - Streaming requests and responses.
- Load credentials from environment variables, or from a configuration file.
# Default Python JSON parser:
pip install pych-client
# Faster orjson parser:
pip install pych-client[orjson]
from pych_client import AsyncClickHouseClient, ClickHouseClient
# See "Credential provider chain" for more information on credential specification.
credentials = dict(
base_url="http://localhost:8123",
database="default",
username="default",
password=""
)
# The client can be used directly, or as a context manager.
# The context manager will ensure that the HTTP client resources
# are properly cleaned-up on exit.
with ClickHouseClient(**credentials) as client:
# `.bytes()` and `.text()` return the raw response content from the database.
# `.json()` sets the format to `JSONEachRow` and parse the response content.
client.bytes("SELECT arrayJoin([1, 2, 3]) AS a")
# b'1\n2\n3\n'
client.text("SELECT arrayJoin([1, 2, 3]) AS a")
# '1\n2\n3\n'
client.json("SELECT arrayJoin([1, 2, 3]) AS a")
# [{'a': 1}, {'a': 2}, {'a': 3}]
# `.iter_bytes()`, `.iter_text()` and `.iter_json()` return the response content
# as it is received from the database, without buffering the entire response.
# `.iter_text()` iterates on the line of the response.
list(client.iter_bytes("SELECT arrayJoin([1, 2, 3]) AS a"))
# [b'1\n2\n3\n', b'']
list(client.iter_text("SELECT arrayJoin([1, 2, 3]) AS a"))
# ['1', '2', '3']
list(client.iter_json("SELECT arrayJoin([1, 2, 3]) AS a"))
# [{'a': 1}, {'a': 2}, {'a': 3}]
# In addition to the query, the following arguments can be set:
# - `params`: a mapping of query parameters to their values.
# - `data`: a bytes, string or an interator of bytes to send in the request body.
# - `settings`: ClickHouse settings (e.g. `{"default_format": "JSONEachRow"`).
params = {"table": "test_pych"}
client.text('''
CREATE TABLE {table:Identifier} (a Int64, b Int64)
ENGINE MergeTree() ORDER BY (a, b)
''', params)
client.text("INSERT INTO {table:Identifier} VALUES", params, "(1, 2)")
client.text("INSERT INTO {table:Identifier} VALUES", params, [b"(3, 4)", b"(5, 6)"])
client.json("SELECT * FROM {table:Identifier} ORDER BY a", params)
# [{'a': '1', 'b': '2'}, {'a': '3', 'b': '4'}, {'a': '5', 'b': '6'}]
# `AsyncClickHouseClient` offers the same methods:
async with AsyncClickHouseClient(**credentials) as client:
# Example usage for `.json()` and `.iter_json()`:
await client.json("SELECT arrayJoin([1, 2, 3]) AS a")
# [{'a': 1}, {'a': 2}, {'a': 3}]
async for row in client.iter_json("SELECT arrayJoin([1, 2, 3]) AS a"):
...
pych-client --help
The client looks for credentials in a way similar to the AWS SDK:
- If one of
base_url
,database
,username
orpassword
is specified, these values will be used. - If none of the previous values are specified, and one of
PYCH_BASE_URL
,PYCH_DATABASE
,PYCH_USERNAME
orPYCH_PASSWORD
environment variables are present, these values will be used. - If none of the previous values are specified, and the file
~/.config/pych-client/credentials.json
exists, the fieldsbase_url
,database
andusername
andpassword
will be used. - If none of the previous values are specified, the values
http://localhost:8213
,default
anddefault
will be used.