Catalog Workspace Content Service

The gooddata_sdk.catalog_workspace_content service enables you to list catalog all objects from a workspace. These objects include:

Datasets
Metrics
Facts
Attributes

The service enables read, put, load and store of declarative layout for LDM (logical data model) and analytics model.

The service supports two types of methods:

Entity methods let you work with workspace content on a high level using simplified entities.
Declarative methods allow you to work with workspace content on a more granular level by fetching entire workspace content layouts, including all of their nested objects.

Entity methods

The gooddata_sdk.catalog_workspace_content supports the following entity API calls:

get_full_catalog(workspace_id: str)

Returns CatalogWorkspaceContent.

Retrieve all datasets with attributes, facts, and metrics for a workspace.
get_attributes_catalog(workspace_id: str)

Returns list[CatalogAttribute]

Retrieve all attributes for a workspace.
get_labels_catalog(workspace_id: str)

Returns list[CatalogLabel]

Retrieve all labels for a workspace.
get_metrics_catalog(workspace_id: str)

Returns list[CatalogMetric]

Retrieve all metrics for a workspace.
get_facts_catalog(workspace_id: str)

Returns list[CatalogFact]

Retrieve all facts for a workspace.
get_dependent_entities_graph(workspace_id: str)

Returns CatalogDependentEntitiesResponse

There are dependencies among all catalog objects, the chain is the following:

fact/attribute/label -> dataset -> metric -> insight -> dashboard

Some steps can be skipped, e.g. fact -> insight

We do not support table -> dataset dependency yet.
get_dependent_entities_graph_from_entry_points(workspace_id: str, dependent_entities_request: CatalogDependentEntitiesRequest)

Returns CatalogDependentEntitiesResponse

Extends get_dependent_entities_graph with the entry point from which the graph is created.

Example Usage

from gooddata_sdk import GoodDataSdk

# GoodData.CN host in the form of uri eg. "http://localhost:3000"
host = "http://localhost:3000"
# GoodData.CN user token
token = "some_user_token"
sdk = GoodDataSdk.create(host, token)

workspace_id = "demo"

# Read catalog for demo workspace
catalog = sdk.catalog_workspace_content.get_full_catalog(workspace_id)

# Print all dataset in the workspace
for dataset in catalog.datasets:
    print(str(dataset))

# Print all metrics in the workspace
for metric in catalog.metrics:
    print(str(metric))

# Read list of attributes for demo workspace
attributes = sdk.catalog_workspace_content.get_attributes_catalog(workspace_id)

# Read list of facts for demo workspace
facts = sdk.catalog_workspace_content.get_facts_catalog(workspace_id)

Declarative methods

The gooddata_sdk.catalog_workspace_content supports the following declarative API calls:

Logical data model (LDM)

get_declarative_ldm(workspace_id: str)

Returns CatalogDeclarativeModel.

Retrieve a logical model layout. On CatalogDeclarativeModel user can call modify_mapped_data_source(data_source_mapping: dict) method, which substitutes data source id in datasets.
put_declarative_ldm(workspace_id: str, ldm: CatalogDeclarativeModel, validator: Optional[DataSourceValidator])

Put a logical data model into a given workspace. You can pass an additional validator parameter which checks that for every data source id in the logical data model the corresponding data source exists.

store_declarative_ldm(workspace_id: str, layout_root_path: Path = Path.cwd())

Store logical data model layout in directory hierarchy.

gooddata_layouts
└── organization_id
        └── workspaces
                └── workspace_id
                        └── analytics_model
                                └── ldm
                                    ├── datasets
                                    │       └── dataset.yaml
                                    └── date_instances
                                            └── date_instance.yaml

load_declarative_ldm(workspace_id: str, layout_root_path: Path = Path.cwd())

Returns CatalogDeclarativeModel.

Load declarative LDM layout, which was stored using store_declarative_ldm.
load_and_put_declarative_ldm(workspace_id: str, layout_root_path: Path = Path.cwd(), validator: Optional[DataSourceValidator])

This method combines load_declarative_ldm and put_declarative_ldm methods to load and set layouts stored using store_declarative_ldm. You can pass an additional validator parameter which checks that for every data source id in the logical data model the corresponding data source exists.

Analytics Model

get_declarative_analytics_model(workspace_id: str)

Returns CatalogDeclarativeAnalytics.

Retrieve an analytics model layout.
put_declarative_analytics_model(workspace_id: str, analytics_model: CatalogDeclarativeAnalytics)

Put an analytics model into a given workspace.

store_declarative_analytics_model(workspace_id: str, layout_root_path: Path = Path.cwd())

Store declarative analytics model layout in directory hierarchy.

gooddata_layouts
└── organization_id
        └── workspaces
                └── workspace_id
                        └── analytics_model
                                ├── analytical_dashboards
                                │       └── analytical_dashboard.yaml
                                ├── dashboard_plugins
                                │       └── dashboard_plugin.yaml
                                ├── filter_contexts
                                │       └── filter_context.yaml
                                ├── metrics
                                │       └── metric.yaml
                                └── visualization_objects
                                        └── visualization_object.yaml

load_declarative_analytics_model(workspace_id: str, layout_root_path: Path = Path.cwd())

Returns CatalogDeclarativeAnalytics.

Load declarative LDM layout, which was stored using store_declarative_analytics_model.
load_and_put_declarative_analytics_model(workspace_id: str, layout_root_path: Path = Path.cwd())

This method combines load_declarative_analytics_model and put_declarative_analytics_model methods to load and set layouts stored using store_declarative_analytics_model.

Example usage:

from gooddata_sdk import GoodDataSdk

# GoodData.CN host in the form of uri eg. "http://localhost:3000"
host = "http://localhost:3000"
# GoodData.CN user token
token = "some_user_token"
sdk = GoodDataSdk.create(host, token)

workspace_id = "demo"

# Get ldm object afterward you can modify it
ldm = sdk.catalog_workspace_content.get_declarative_ldm(workspace_id=workspace_id)

# Modify data source id for datasets
ldm.modify_mapped_data_source({"demo-test-ds": "demo-prod-ds"})

# Put ldm object back to server
sdk.catalog_workspace_content.put_declarative_ldm(workspace_id=workspace_id, ldm=ldm)

# Get analytics model object afterward you can modify it
analytics_model = sdk.catalog_workspace_content.get_declarative_analytics_model(workspace_id=workspace_id)

# Put analytics model object back to server
sdk.catalog_workspace_content.put_declarative_analytics_model(workspace_id=workspace_id,
                                                              analytics_model=analytics_model)