database

prefect_snowflake.database

Module for querying against Snowflake databases.

Functions

snowflake_query

snowflake_query(query: str, snowflake_connector: SnowflakeConnector, params: Union[Tuple[Any], Dict[str, Any]] = None, cursor_type: Type[SnowflakeCursor] = SnowflakeCursor, poll_frequency_seconds: int = 1) -> List[Tuple[Any]]

Executes a query against a Snowflake database.

Args:

• query: The query to execute against the database.
• params: The params to replace the placeholders in the query.
• snowflake_connector: The credentials to use to authenticate.
• cursor_type: The type of database cursor to use for the query.
• poll_frequency_seconds: Number of seconds to wait in between checks for run completion.

Returns:

• The output of response.fetchall().

Examples:

Query Snowflake table with the ID value parameterized.

from prefect import flow
from prefect_snowflake.credentials import SnowflakeCredentials
from prefect_snowflake.database import SnowflakeConnector, snowflake_query


@flow
def snowflake_query_flow():
    snowflake_credentials = SnowflakeCredentials(
        account="account",
        user="user",
        password="password",
    )
    snowflake_connector = SnowflakeConnector(
        database="database",
        warehouse="warehouse",
        schema="schema",
        credentials=snowflake_credentials
    )
    result = snowflake_query(
        "SELECT * FROM table WHERE id=%{id_param}s LIMIT 8;",
        snowflake_connector,
        params={"id_param": 1}
    )
    return result

snowflake_query_flow()

snowflake_query_async

snowflake_query_async(query: str, snowflake_connector: SnowflakeConnector, params: Union[Tuple[Any], Dict[str, Any]] = None, cursor_type: Type[SnowflakeCursor] = SnowflakeCursor, poll_frequency_seconds: int = 1) -> List[Tuple[Any]]

Executes a query against a Snowflake database.

Args:

• query: The query to execute against the database.
• params: The params to replace the placeholders in the query.
• snowflake_connector: The credentials to use to authenticate.
• cursor_type: The type of database cursor to use for the query.
• poll_frequency_seconds: Number of seconds to wait in between checks for run completion.

Returns:

• The output of response.fetchall().

Examples:

Query Snowflake table with the ID value parameterized.

from prefect import flow
from prefect_snowflake.credentials import SnowflakeCredentials
from prefect_snowflake.database import SnowflakeConnector, snowflake_query


@flow
def snowflake_query_flow():
    snowflake_credentials = SnowflakeCredentials(
        account="account",
        user="user",
        password="password",
    )
    snowflake_connector = SnowflakeConnector(
        database="database",
        warehouse="warehouse",
        schema="schema",
        credentials=snowflake_credentials
    )
    result = snowflake_query(
        "SELECT * FROM table WHERE id=%{id_param}s LIMIT 8;",
        snowflake_connector,
        params={"id_param": 1}
    )
    return result

snowflake_query_flow()

snowflake_multiquery

snowflake_multiquery(queries: List[str], snowflake_connector: SnowflakeConnector, params: Union[Tuple[Any], Dict[str, Any]] = None, cursor_type: Type[SnowflakeCursor] = SnowflakeCursor, as_transaction: bool = False, return_transaction_control_results: bool = False, poll_frequency_seconds: int = 1) -> List[List[Tuple[Any]]]

Executes multiple queries against a Snowflake database in a shared session. Allows execution in a transaction.

Args:

• queries: The list of queries to execute against the database.
• params: The params to replace the placeholders in the query.
• snowflake_connector: The credentials to use to authenticate.
• cursor_type: The type of database cursor to use for the query.
• as_transaction: If True, queries are executed in a transaction.
• return_transaction_control_results: Determines if the results of queries controlling the transaction (BEGIN/COMMIT) should be returned.
• poll_frequency_seconds: Number of seconds to wait in between checks for run completion.

Returns:

• List of the outputs of response.fetchall() for each query.

Examples:

Query Snowflake table with the ID value parameterized.

from prefect import flow
from prefect_snowflake.credentials import SnowflakeCredentials
from prefect_snowflake.database import SnowflakeConnector, snowflake_multiquery


@flow
def snowflake_multiquery_flow():
    snowflake_credentials = SnowflakeCredentials(
        account="account",
        user="user",
        password="password",
    )
    snowflake_connector = SnowflakeConnector(
        database="database",
        warehouse="warehouse",
        schema="schema",
        credentials=snowflake_credentials
    )
    result = snowflake_multiquery(
        ["SELECT * FROM table WHERE id=%{id_param}s LIMIT 8;", "SELECT 1,2"],
        snowflake_connector,
        params={"id_param": 1},
        as_transaction=True
    )
    return result

snowflake_multiquery_flow()

snowflake_multiquery_async

snowflake_multiquery_async(queries: List[str], snowflake_connector: SnowflakeConnector, params: Union[Tuple[Any], Dict[str, Any]] = None, cursor_type: Type[SnowflakeCursor] = SnowflakeCursor, as_transaction: bool = False, return_transaction_control_results: bool = False, poll_frequency_seconds: int = 1) -> List[List[Tuple[Any]]]

Executes multiple queries against a Snowflake database in a shared session. Allows execution in a transaction.

Args:

• queries: The list of queries to execute against the database.
• params: The params to replace the placeholders in the query.
• snowflake_connector: The credentials to use to authenticate.
• cursor_type: The type of database cursor to use for the query.
• as_transaction: If True, queries are executed in a transaction.
• return_transaction_control_results: Determines if the results of queries controlling the transaction (BEGIN/COMMIT) should be returned.
• poll_frequency_seconds: Number of seconds to wait in between checks for run completion.

Returns:

• List of the outputs of response.fetchall() for each query.

Examples:

Query Snowflake table with the ID value parameterized.

from prefect import flow
from prefect_snowflake.credentials import SnowflakeCredentials
from prefect_snowflake.database import SnowflakeConnector, snowflake_multiquery


@flow
def snowflake_multiquery_flow():
    snowflake_credentials = SnowflakeCredentials(
        account="account",
        user="user",
        password="password",
    )
    snowflake_connector = SnowflakeConnector(
        database="database",
        warehouse="warehouse",
        schema="schema",
        credentials=snowflake_credentials
    )
    result = snowflake_multiquery(
        ["SELECT * FROM table WHERE id=%{id_param}s LIMIT 8;", "SELECT 1,2"],
        snowflake_connector,
        params={"id_param": 1},
        as_transaction=True
    )
    return result

snowflake_multiquery_flow()

snowflake_query_sync

snowflake_query_sync(query: str, snowflake_connector: SnowflakeConnector, params: Union[Tuple[Any], Dict[str, Any]] = None, cursor_type: Type[SnowflakeCursor] = SnowflakeCursor) -> List[Tuple[Any]]

Executes a query in sync mode against a Snowflake database.

Args:

• query: The query to execute against the database.
• params: The params to replace the placeholders in the query.
• snowflake_connector: The credentials to use to authenticate.
• cursor_type: The type of database cursor to use for the query.

Returns:

• The output of response.fetchall().

Examples:

Execute a put statement.

from prefect import flow
from prefect_snowflake.credentials import SnowflakeCredentials
from prefect_snowflake.database import SnowflakeConnector, snowflake_query


@flow
def snowflake_query_sync_flow():
    snowflake_credentials = SnowflakeCredentials(
        account="account",
        user="user",
        password="password",
    )
    snowflake_connector = SnowflakeConnector(
        database="database",
        warehouse="warehouse",
        schema="schema",
        credentials=snowflake_credentials
    )
    result = snowflake_query_sync(
        "put file://a_file.csv @mystage;",
        snowflake_connector,
    )
    return result

snowflake_query_sync_flow()

Classes

SnowflakeConnector

Block used to manage connections with Snowflake.

Upon instantiating, a connection is created and maintained for the life of the object until the close method is called.

It is recommended to use this block as a context manager, which will automatically close the engine and its connections when the context is exited.

It is also recommended that this block is loaded and consumed within a single task or flow because if the block is passed across separate tasks and flows, the state of the block’s connection and cursor will be lost.

Args:

• credentials: The credentials to authenticate with Snowflake.
• database: The name of the default database to use.
• warehouse: The name of the default warehouse to use.
• schema: The name of the default schema to use; this attribute is accessible through SnowflakeConnector(...).schema_.
• fetch_size: The number of rows to fetch at a time.
• poll_frequency_s: The number of seconds before checking query.

Examples:

Load stored Snowflake connector as a context manager:

from prefect_snowflake.database import SnowflakeConnector

snowflake_connector = SnowflakeConnector.load("BLOCK_NAME")

Insert data into database and fetch results.

from prefect_snowflake.database import SnowflakeConnector

with SnowflakeConnector.load("BLOCK_NAME") as conn:
    conn.execute(
        "CREATE TABLE IF NOT EXISTS customers (name varchar, address varchar);"
    )
    conn.execute_many(
        "INSERT INTO customers (name, address) VALUES (%(name)s, %(address)s);",
        seq_of_parameters=[
            {"name": "Ford", "address": "Highway 42"},
            {"name": "Unknown", "address": "Space"},
            {"name": "Me", "address": "Myway 88"},
        ],
    )
    results = conn.fetch_all(
        "SELECT * FROM customers WHERE address = %(address)s",
        parameters={"address": "Space"}
    )
    print(results)

Methods:

close

close(self)

Closes connection and its cursors.

execute

execute(self, operation: str, parameters: Optional[Dict[str, Any]] = None, cursor_type: Type[SnowflakeCursor] = SnowflakeCursor, **execute_kwargs: Any) -> None

Executes an operation on the database. This method is intended to be used for operations that do not return data, such as INSERT, UPDATE, or DELETE. Unlike the fetch methods, this method will always execute the operation upon calling.

Args:

• operation: The SQL query or other operation to be executed.
• parameters: The parameters for the operation.
• cursor_type: The class of the cursor to use when creating a Snowflake cursor.
• **execute_kwargs: Additional options to pass to cursor.execute_async.

Examples:

Create table named customers with two columns, name and address.

from prefect_snowflake.database import SnowflakeConnector

with SnowflakeConnector.load("BLOCK_NAME") as conn:
    conn.execute(
        "CREATE TABLE IF NOT EXISTS customers (name varchar, address varchar);"
    )

execute_async

execute_async(self, operation: str, parameters: Optional[Dict[str, Any]] = None, cursor_type: Type[SnowflakeCursor] = SnowflakeCursor, **execute_kwargs: Any) -> None

Executes an operation on the database. This method is intended to be used for operations that do not return data, such as INSERT, UPDATE, or DELETE. Unlike the fetch methods, this method will always execute the operation upon calling.

Args:

• operation: The SQL query or other operation to be executed.
• parameters: The parameters for the operation.
• cursor_type: The class of the cursor to use when creating a Snowflake cursor.
• **execute_kwargs: Additional options to pass to cursor.execute_async.

Examples:

Create table named customers with two columns, name and address.

from prefect_snowflake.database import SnowflakeConnector

with SnowflakeConnector.load("BLOCK_NAME") as conn:
    conn.execute(
        "CREATE TABLE IF NOT EXISTS customers (name varchar, address varchar);"
    )

execute_many

execute_many(self, operation: str, seq_of_parameters: List[Dict[str, Any]]) -> None

Executes many operations on the database. This method is intended to be used for operations that do not return data, such as INSERT, UPDATE, or DELETE. Unlike the fetch methods, this method will always execute the operations upon calling.

Args:

• operation: The SQL query or other operation to be executed.
• seq_of_parameters: The sequence of parameters for the operation.

Examples:

Create table and insert three rows into it.

from prefect_snowflake.database import SnowflakeConnector

with SnowflakeConnector.load("BLOCK_NAME") as conn:
    conn.execute(
        "CREATE TABLE IF NOT EXISTS customers (name varchar, address varchar);"
    )
    conn.execute_many(
        "INSERT INTO customers (name, address) VALUES (%(name)s, %(address)s);",
        seq_of_parameters=[
            {"name": "Marvin", "address": "Highway 42"},
            {"name": "Ford", "address": "Highway 42"},
            {"name": "Unknown", "address": "Space"},
        ],
    )

execute_many_async

execute_many_async(self, operation: str, seq_of_parameters: List[Dict[str, Any]]) -> None

Executes many operations on the database. This method is intended to be used for operations that do not return data, such as INSERT, UPDATE, or DELETE. Unlike the fetch methods, this method will always execute the operations upon calling.

Args:

• operation: The SQL query or other operation to be executed.
• seq_of_parameters: The sequence of parameters for the operation.

Examples:

Create table and insert three rows into it.

from prefect_snowflake.database import SnowflakeConnector

with SnowflakeConnector.load("BLOCK_NAME") as conn:
    conn.execute(
        "CREATE TABLE IF NOT EXISTS customers (name varchar, address varchar);"
    )
    conn.execute_many(
        "INSERT INTO customers (name, address) VALUES (%(name)s, %(address)s);",
        seq_of_parameters=[
            {"name": "Marvin", "address": "Highway 42"},
            {"name": "Ford", "address": "Highway 42"},
            {"name": "Unknown", "address": "Space"},
        ],
    )

fetch_all

fetch_all(self, operation: str, parameters: Optional[Dict[str, Any]] = None, cursor_type: Type[SnowflakeCursor] = SnowflakeCursor, **execute_kwargs: Any) -> List[Tuple[Any]]

Fetch all results from the database. Repeated calls using the same inputs to any of the fetch methods of this block will skip executing the operation again, and instead, return the next set of results from the previous execution, until the reset_cursors method is called.

Args:

• operation: The SQL query or other operation to be executed.
• parameters: The parameters for the operation.
• cursor_type: The class of the cursor to use when creating a Snowflake cursor.
• **execute_kwargs: Additional options to pass to cursor.execute_async.

Returns:

• A list of tuples containing the data returned by the database, where each row is a tuple and each column is a value in the tuple.

fetch_all_async

fetch_all_async(self, operation: str, parameters: Optional[Dict[str, Any]] = None, cursor_type: Type[SnowflakeCursor] = SnowflakeCursor, **execute_kwargs: Any) -> List[Tuple[Any]]

Fetch all results from the database. Repeated calls using the same inputs to any of the fetch methods of this block will skip executing the operation again, and instead, return the next set of results from the previous execution, until the reset_cursors method is called.

Args:

• operation: The SQL query or other operation to be executed.
• parameters: The parameters for the operation.
• cursor_type: The class of the cursor to use when creating a Snowflake cursor.
• **execute_kwargs: Additional options to pass to cursor.execute_async.

Returns:

• A list of tuples containing the data returned by the database, where each row is a tuple and each column is a value in the tuple.

Examples:

Fetch all rows from the database where address is Highway 42.

from prefect_snowflake.database import SnowflakeConnector

with SnowflakeConnector.load("BLOCK_NAME") as conn:
    await conn.execute_async(
        "CREATE TABLE IF NOT EXISTS customers (name varchar, address varchar);"
    )
    await conn.execute_many_async(
        "INSERT INTO customers (name, address) VALUES (%(name)s, %(address)s);",
        seq_of_parameters=[
            {"name": "Marvin", "address": "Highway 42"},
            {"name": "Ford", "address": "Highway 42"},
            {"name": "Unknown", "address": "Highway 42"},
            {"name": "Me", "address": "Myway 88"},
        ],
    )
    result = await conn.fetch_all_async(
        "SELECT * FROM customers WHERE address = %(address)s",
        parameters={"address": "Highway 42"},
    )
    print(result)  # Marvin, Ford, Unknown

fetch_many

fetch_many(self, operation: str, parameters: Optional[Sequence[Dict[str, Any]]] = None, size: Optional[int] = None, cursor_type: Type[SnowflakeCursor] = SnowflakeCursor, **execute_kwargs: Any) -> List[Tuple[Any]]

Fetch a limited number of results from the database. Repeated calls using the same inputs to any of the fetch methods of this block will skip executing the operation again, and instead, return the next set of results from the previous execution, until the reset_cursors method is called.

Args:

• operation: The SQL query or other operation to be executed.
• parameters: The parameters for the operation.
• size: The number of results to return; if None or 0, uses the value of fetch_size configured on the block.
• cursor_type: The class of the cursor to use when creating a Snowflake cursor.
• **execute_kwargs: Additional options to pass to cursor.execute_async.

Returns:

• A list of tuples containing the data returned by the database, where each row is a tuple and each column is a value in the tuple.

Examples:

Repeatedly fetch two rows from the database where address is Highway 42.

from prefect_snowflake.database import SnowflakeConnector

with SnowflakeConnector.load("BLOCK_NAME") as conn:
    conn.execute(
        "CREATE TABLE IF NOT EXISTS customers (name varchar, address varchar);"
    )
    conn.execute_many(
        "INSERT INTO customers (name, address) VALUES (%(name)s, %(address)s);",
        seq_of_parameters=[
            {"name": "Marvin", "address": "Highway 42"},
            {"name": "Ford", "address": "Highway 42"},
            {"name": "Unknown", "address": "Highway 42"},
            {"name": "Me", "address": "Highway 42"},
        ],
    )
    result = conn.fetch_many(
        "SELECT * FROM customers WHERE address = %(address)s",
        parameters={"address": "Highway 42"},
        size=2
    )
    print(result)  # Marvin, Ford
    result = conn.fetch_many(
        "SELECT * FROM customers WHERE address = %(address)s",
        parameters={"address": "Highway 42"},
        size=2
    )
    print(result)  # Unknown, Me

fetch_many_async

fetch_many_async(self, operation: str, parameters: Optional[Sequence[Dict[str, Any]]] = None, size: Optional[int] = None, cursor_type: Type[SnowflakeCursor] = SnowflakeCursor, **execute_kwargs: Any) -> List[Tuple[Any]]

Fetch a limited number of results from the database asynchronously. Repeated calls using the same inputs to any of the fetch methods of this block will skip executing the operation again, and instead, return the next set of results from the previous execution, until the reset_cursors method is called.

Args:

• operation: The SQL query or other operation to be executed.
• parameters: The parameters for the operation.
• size: The number of results to return; if None or 0, uses the value of fetch_size configured on the block.
• cursor_type: The class of the cursor to use when creating a Snowflake cursor.
• **execute_kwargs: Additional options to pass to cursor.execute_async.

Returns:

• A list of tuples containing the data returned by the database, where each row is a tuple and each column is a value in the tuple.

Examples:

Repeatedly fetch two rows from the database where address is Highway 42.

from prefect_snowflake.database import SnowflakeConnector

with SnowflakeConnector.load("BLOCK_NAME") as conn:
    conn.execute(
        "CREATE TABLE IF NOT EXISTS customers (name varchar, address varchar);"
    )
    conn.execute_many(
        "INSERT INTO customers (name, address) VALUES (%(name)s, %(address)s);",
        seq_of_parameters=[
            {"name": "Marvin", "address": "Highway 42"},
            {"name": "Ford", "address": "Highway 42"},
            {"name": "Unknown", "address": "Highway 42"},
            {"name": "Me", "address": "Highway 42"},
        ],
    )
    result = conn.fetch_many(
        "SELECT * FROM customers WHERE address = %(address)s",
        parameters={"address": "Highway 42"},
        size=2
    )
    print(result)  # Marvin, Ford
    result = conn.fetch_many(
        "SELECT * FROM customers WHERE address = %(address)s",
        parameters={"address": "Highway 42"},
        size=2
    )
    print(result)  # Unknown, Me

fetch_one

fetch_one(self, operation: str, parameters: Optional[Dict[str, Any]] = None, cursor_type: Type[SnowflakeCursor] = SnowflakeCursor, **execute_kwargs: Any) -> Tuple[Any]

Fetch a single result from the database. Repeated calls using the same inputs to any of the fetch methods of this block will skip executing the operation again, and instead, return the next set of results from the previous execution, until the reset_cursors method is called.

Args:

• operation: The SQL query or other operation to be executed.
• parameters: The parameters for the operation.
• cursor_type: The class of the cursor to use when creating a Snowflake cursor.
• **execute_kwargs: Additional options to pass to cursor.execute_async.

Returns:

• A tuple containing the data returned by the database, where each row is a tuple and each column is a value in the tuple.

Examples:

Fetch one row from the database where address is Space.

from prefect_snowflake.database import SnowflakeConnector

with SnowflakeConnector.load("BLOCK_NAME") as conn:
    conn.execute(
        "CREATE TABLE IF NOT EXISTS customers (name varchar, address varchar);"
    )
    conn.execute_many(
        "INSERT INTO customers (name, address) VALUES (%(name)s, %(address)s);",
        seq_of_parameters=[
            {"name": "Ford", "address": "Highway 42"},
            {"name": "Unknown", "address": "Space"},
            {"name": "Me", "address": "Myway 88"},
        ],
    )
    result = conn.fetch_one(
        "SELECT * FROM customers WHERE address = %(address)s",
        parameters={"address": "Space"}
    )
    print(result)

fetch_one_async

fetch_one_async(self, operation: str, parameters: Optional[Dict[str, Any]] = None, cursor_type: Type[SnowflakeCursor] = SnowflakeCursor, **execute_kwargs: Any) -> Tuple[Any]

Fetch a single result from the database asynchronously. Repeated calls using the same inputs to any of the fetch methods of this block will skip executing the operation again, and instead, return the next set of results from the previous execution, until the reset_cursors method is called.

Args:

• operation: The SQL query or other operation to be executed.
• parameters: The parameters for the operation.
• cursor_type: The class of the cursor to use when creating a Snowflake cursor.
• **execute_kwargs: Additional options to pass to cursor.execute_async.

Returns:

• A tuple containing the data returned by the database, where each row is a tuple and each column is a value in the tuple.

Examples:

Fetch one row from the database where address is Space.

from prefect_snowflake.database import SnowflakeConnector

with SnowflakeConnector.load("BLOCK_NAME") as conn:
    conn.execute(
        "CREATE TABLE IF NOT EXISTS customers (name varchar, address varchar);"
    )
    conn.execute_many(
        "INSERT INTO customers (name, address) VALUES (%(name)s, %(address)s);",
        seq_of_parameters=[
            {"name": "Ford", "address": "Highway 42"},
            {"name": "Unknown", "address": "Space"},
            {"name": "Me", "address": "Myway 88"},
        ],
    )
    result = await conn.fetch_one_async(
        "SELECT * FROM customers WHERE address = %(address)s",
        parameters={"address": "Space"}
    )
    print(result)

get_connection

get_connection(self, **connect_kwargs: Any) -> SnowflakeConnection

Returns an authenticated connection that can be used to query from Snowflake databases.

Args:

• **connect_kwargs: Additional arguments to pass to snowflake.connector.connect.

Returns:

• The authenticated SnowflakeConnection.

Examples:

from prefect_snowflake.credentials import SnowflakeCredentials
from prefect_snowflake.database import SnowflakeConnector

snowflake_credentials = SnowflakeCredentials(
    account="account",
    user="user",
    password="password",
)
snowflake_connector = SnowflakeConnector(
    database="database",
    warehouse="warehouse",
    schema="schema",
    credentials=snowflake_credentials
)
with snowflake_connector.get_connection() as connection:
    ...

reset_cursors

reset_cursors(self) -> None

Tries to close all opened cursors.

Examples:

Reset the cursors to refresh cursor position.

from prefect_snowflake.database import SnowflakeConnector

with SnowflakeConnector.load("BLOCK_NAME") as conn:
    conn.execute(
        "CREATE TABLE IF NOT EXISTS customers (name varchar, address varchar);"
    )
    conn.execute_many(
        "INSERT INTO customers (name, address) VALUES (%(name)s, %(address)s);",
        seq_of_parameters=[
            {"name": "Ford", "address": "Highway 42"},
            {"name": "Unknown", "address": "Space"},
            {"name": "Me", "address": "Myway 88"},
        ],
    )
    print(conn.fetch_one("SELECT * FROM customers"))  # Ford
    conn.reset_cursors()
    print(conn.fetch_one("SELECT * FROM customers"))  # should be Ford again