diff --git a/docs/features-applications/cross-application-features/data-product.md b/docs/features-applications/cross-application-features/data-product.md index db5d3d6a..19e3562f 100644 --- a/docs/features-applications/cross-application-features/data-product.md +++ b/docs/features-applications/cross-application-features/data-product.md @@ -131,19 +131,25 @@ To create an output port: ![](./images/data-product-studio-output-port-create.png) -6. Select a data source to filter the available data sources, and then click **Next**. -7. Select a data container, and then click **Next**. - - !!! warning - In this version, only schemas are supported. +6. Select a data source, and then click **Next**. +7. Select a connection from the list of connections associated with the selected data source, and then click **Next**. + + !!! note + The list is filtered based on the data source selected in the previous step. + + If only one connection is available for the selected data source, it is selected automatically. + +8. Search, filter, and select one or more datasets. + + !!! warning "Important" + All selected datasets must belong to the same folder. To select datasets from a different folder, remove the current selection first. -8. Search, filter, and select one or more datasets, and then click **Next**. - !!! note You can select all datasets at once or add them individually. Selected datasets are marked with a check mark to prevent duplicate selections. You can remove individual datasets or clear the entire selection. -9. Review the information on the confirmation page. -10. Click **Confirm**. +9. Click **Next** after selecting the datasets. +10. Review the information on the confirmation page. +11. Click **Confirm**. The platform creates the output port and redirects you to its details page. @@ -164,10 +170,7 @@ To add an output port to an existing user-defined data product: The output port creation wizard opens with the current data product preselected. -4. Select a data source and a data container. -5. Search, filter, and select one or more datasets. -6. Review the information on the confirmation page. -7. Click **Confirm**. +4. Complete the output port creation wizard by following steps 6 through 11 in [Create an Output Port](#create-an-output-port). The platform creates the output port and associates it with the selected data product. @@ -200,8 +203,17 @@ To manage datasets: Dataset linking for output ports is currently supported for the following connectors: -* Snowflake -* Snowflake V2 +* **Snowflake** +* **Snowflake V2** +* **Palantir Foundry** + + * Palantir Foundry datasets imported through the public or private connector SDK can be linked to output ports using the Data Product API. + * Datasets imported through the public or private connector SDK can be used to create data products in Studio. + +* **Databricks** + + * Databricks datasets imported through the public connector SDK can be linked to output ports using the Data Product API. + * Datasets imported through the public connector SDK can be used to create data products in Studio. #### Generated YAML Descriptors @@ -274,20 +286,55 @@ schema: When uploading a data product or importing a dataset from a scanner, the platform uses the `name` and `physicalName` and the `servers` section to automatically link output ports with their physical implementations. See the [Supported Connectors](#supported-connectors) section. -Following is an example of a `servers` section for Snowflake: +The following examples show the `servers` section for different connectors: -```yaml -servers: - - id: snowflake-server-pse - server: Snowflake server - type: snowflake - host: ABCDEFGH-IJ12345.snowflakecomputing.com - port: 8080 - account: IJ12345 - database: DATABASE_1 - warehouse: SMALL_COMPUTE_WH - schema: SCHEMA_1 -``` +* **Snowflake** + + The following example shows a `servers` section for Snowflake: + + ```yaml + servers: + - server: Snowflake server + type: snowflake + account: IJ12345 + database: DATABASE_1 + schema: SCHEMA_1 + ``` + +* **Palantir Foundry** + + For Palantir Foundry, ODCS v3.1.0 does not define a server type. You can use the custom server syntax with a `custom-server-type` property. + + The following example shows a `servers` section for Palantir Foundry: + + ```yaml + servers: + - server: Palantir Foundry + type: custom + host: localhost + customProperties: + - property: custom-server-type + value: palantir-foundry + ``` + + In the `schema` section, provide the technical name of the Palantir Foundry dataset in `physicalName` as follows: + + ```yaml + physicalName: ri.foundry.main.dataset.f3d20e79-1d76-4c59-aa4d-1427e96f76d0 + ``` + +* **Databricks** + + The following example shows a `servers` section for Databricks: + + ```yaml + servers: + - server: Databricks + type: databricks + catalog: catalog-1 + host: localhost + schema: schema-1 + ``` You can link the same dataset to multiple output ports and to different data products. @@ -302,6 +349,8 @@ If not all datasets defined in the data contract are linked to the output port, ![](./images/data-product-output-port-warning.png) +When an output port YAML specifies several servers, all datasets defined in the `schema` section must be linked for every specified server. + ### Manage Actian-Specific Attributes for Data Products and Output Ports #### Manage Properties @@ -344,6 +393,9 @@ The example above creates the following properties: Properties are managed as source documentation, similar to source properties of a dataset. They are not mapped to the data product template. +!!! note + The property name uses the value provided in the YAML file (for example, `Sensitivity`), while the property code is automatically generated by the platform (for example, `$z_sensitivity`). + Properties created through the data product API can be used as filters or displayed in the catalog as any other source property, depending on their type, by customizing their search and display options in the **Catalog Design** section of the Studio. #### Manage Contacts diff --git a/docs/features-applications/cross-application-features/images/data-product-studio-output-port-create.png b/docs/features-applications/cross-application-features/images/data-product-studio-output-port-create.png index c7436640..a8a8f350 100644 Binary files a/docs/features-applications/cross-application-features/images/data-product-studio-output-port-create.png and b/docs/features-applications/cross-application-features/images/data-product-studio-output-port-create.png differ diff --git a/docs/features-applications/studio/stewardship/studio-create-delete-item.md b/docs/features-applications/studio/stewardship/studio-create-delete-item.md index 61fafc23..f9b6f9d3 100644 --- a/docs/features-applications/studio/stewardship/studio-create-delete-item.md +++ b/docs/features-applications/studio/stewardship/studio-create-delete-item.md @@ -14,9 +14,15 @@ Only certain profiles can create or delete an Item in Actian Data Intelligence. Item types that can be created manually: * Category -* Data process +* Data Process +* Data Product * Glossary Item * Custom Item +* Output Port +* Use Case + +!!! note + For more information about creating data products and output ports, see [Data Products](../../cross-application-features/data-product.md). To create a new Item from Studio: