snowflake.core.service.ServiceCollection¶

class snowflake.core.service.ServiceCollection(schema: SchemaResource)¶

Bases: SchemaObjectCollectionParent[ServiceResource]

Represents the collection operations on the Snowflake Snowpark Container Service resource.

With this collection, you can create, update, iterate through, and fetch SPCSs that you have access to in the current context.

Examples

Creating a service instance:

>>> my_service = Service(
...     name="my_service",
...     min_instances=1,
...     max_instances=2,
...     compute_pool="my_compute_pool",
...     spec=ServiceSpec("@my_stage/my_service_spec.yaml")
...    )
>>> root.databases["my_db"].schemas["my_schema"].services.create(my_service)

Copy

Attributes

database¶

root¶

Methods

create(service: Service, *, mode: CreateMode = CreateMode.error_if_exists) → ServiceResource¶

Create a Snowpark Container service in Snowflake.

Parameters:

service (Service) – The Service object, together with the Service’s properties: name, compute_pool, spec; auto_resume, min_instances, max_instances, status, external_access_integrations, query_warehouse, comment are optional.
mode (CreateMode, optional)
values. (One of the following enum)
CreateMode.error_if_exists (Throw an snowflake.core.exceptions.ConflictError)
.... (create or replace service <name>)
CreateMode.or_replace (Replace if the service already exists in Snowflake. Equivalent to SQL)
....
CreateMode.if_not_exists (Do nothing if the service already exists in Snowflake.)
exists... (Equivalent to SQL create service <name> if not)
CreateMode.error_if_exists. (Default is)

Examples

Creating a service, replacing any existing service with the same name:

>>> services = root.databases["my_db"].schemas["my_schema"].services
>>> my_service = Service(
...     name="my_service",
...     compute_pool="my_compute_pool",
...     spec=ServiceSpec("@my_stage/my_service_spec.yaml")
... )
>>> services.create(my_service, mode=CreateMode.or_replace)

Copy

execute_job(job_service: JobService) → ServiceResource¶

Executes a Snowpark Container job service in Snowflake.

Parameters:: job_service (JobService) – The JobService object, together with the JobService’s properties: name, compute_pool, spec; status, external_access_integrations, query_warehouse, comment are optional

Examples

Executing a job service:

>>> job_service = JobService(
...     name="my_job_service",
...     compute_pool="my_cp",
...     spec=ServiceSpec("@my_stage/my_service_spec.yaml"),
... )
>>> services.execute_job(job_service)

Copy

items() → ItemsView[str, T]¶

iter(*, like: str | None = None, starts_with: str | None = None, limit: int | None = None, from_name: str | None = None) → Iterator[Service]¶

Iterate through Service objects from Snowflake, filtering on any optional ‘like’ pattern.

Parameters:

like (str, optional) – A case-insensitive string functioning as a filter, with support for SQL wildcard characters (% and _).
starts_with (str, optional) – String used to filter the command output based on the string of characters that appear at the beginning of the object name. Uses case-sensitive pattern matching.
show_limit (int, optional) – Limit of the maximum number of rows returned by iter(). The default is None, which behaves equivalently to show_limit=10000. This value must be between 1 and 10000.
from_name (str, optional) – Fetch rows only following the first row whose object name matches the specified string. This is case-sensitive and does not have to be the full name.

Examples

Showing all services that you have access to see:

>>> services = service_collection.iter()

Copy

Showing information of the exact service you want to see:

>>> services = service_collection.iter(like="your-service-name")

Copy

Showing services starting with ‘your-service-name-‘:

>>> services = service_collection.iter(like="your-service-name-%")

Copy

Using a for loop to retrieve information from iterator:

>>> for service in services:
>>>    print(service.name)

Copy

keys() → KeysView[str]¶

values() → ValuesView[T]¶