[NEAT-108] Transformation/RawTable/RawDatabase Loaders #316
Conversation
| @@ -0,0 +1,25 @@ | |||
| from ._base import ResourceLoader | |||
There was a problem hiding this comment.
Should this at some point migrate under a special type of loaders under rules?
There was a problem hiding this comment.
Loader is a bit overloaded here, loaders here are not the same as in graph.
|
|
||
| from .data_classes import RawTableID | ||
|
|
||
| T_ID = TypeVar("T_ID", bound=str | int | DataModelingId | InstanceId | VersionedDataModelingId | RawTableID) |
There was a problem hiding this comment.
What about some more readable name .e.g. T_CogniteResourceId?
There was a problem hiding this comment.
Great, will include.
|
|
||
| @classmethod | ||
| @abstractmethod | ||
| def get_id(cls, item: T_WriteClass | T_WritableCogniteResource) -> T_ID: |
There was a problem hiding this comment.
suggestion maybe rename this to as_id ? wandering do you need two separate methods one for listable, one for single items? can there be a single as_id method?
There was a problem hiding this comment.
even better call this as_external_ids to comply with sdk
There was a problem hiding this comment.
Hmm, I want to distinguish this from .as_id(), .as_ids() which are on most Cognite Resource objects.
| WriteableCogniteResourceList, | ||
| ) | ||
|
|
||
| # The Table, TableWrite data classes in the Cognite-SDK lacks the database attribute. |
There was a problem hiding this comment.
Would this eventually find its way to cognite-sdk ?
There was a problem hiding this comment.
Yes, hopefully, but the problem is that would be a breaking change (to the poor RAW implementation), so it would be breaking the SDK.
Preparation for exporting Raw Tables and Transformations.
Extended
cdf_loaderswithTransformationLoader,RawDatabaseLoaderandRawTableLoader.The module started to become big so I split it into a package for better developer experience. Had to take special care of the RawTableLoader, documented reason in code.