Relationship Matrix

This page summarizes the Microsoft Purview Unified Catalog relationships currently supported and tested by the Purview Unified SDK.

Use this matrix to identify:

• The recommended source resource
• The target entity type
• The SDK method
• Whether the relationship has been tested
• Whether the relationship is direct or indirect

For complete creation and verification examples, see the Relationships guide.

---

Status Definitions

Status	Meaning
Supported and tested	The relationship has been successfully created and verified using the SDK.
Supported by helper method	The SDK provides a dedicated convenience method for the relationship.
Reverse-side tested	The relationship has been tested from the opposite resource direction.
Indirect	The association may appear in Microsoft Purview through another connected resource rather than through a direct relationship API.
Not yet tested	The relationship has not yet been validated with the current SDK examples.
Not directly supported	No direct relationship operation has been identified. This does not necessarily mean the resources cannot appear associated in Microsoft Purview.

---

Relationship Matrix

Source Resource	Target Resource	Resource Path	Target Entity Type	SDK Operation	Status
Data Product	Data Asset	dataProducts	DATAASSET	relationships.create()	Supported and tested
Data Asset	Data Product	dataAssets	DATAPRODUCT	relationships.create()	Reverse-side tested
Data Product	Glossary Term	dataProducts	TERM	relationships.create()	Supported and tested
Data Product	Objective	Managed by helper	Managed by helper	add_objective_to_data_product()	Supported by helper method
Glossary Term	Data Asset	terms	DATAASSET	relationships.create()	Supported and tested
Glossary Term	Data Column	terms	DATACOLUMN	relationships.create()	Supported and tested
Glossary Term	Critical Data Element	Relationship may be created from the CDE side	TERM	relationships.create()	Supported and tested from CDE side
Critical Data Element	Glossary Term	criticalDataElements	TERM	relationships.create()	Supported and tested
Critical Data Element	Data Column	criticalDataElements	DATACOLUMN	relationships.create()	Supported and tested
Data Column	Data Asset	dataColumns	DATAASSET	relationships.create()	Supported and tested
Data Asset	Data Column	dataAssets	DATACOLUMN	Relationship read and verification	Supported and tested
Data Product	Critical Data Element	—	—	—	No direct relationship provided by Microsoft Purview
Critical Data Element	Data Product	—	—	—	No direct relationship provided by Microsoft Purview
Data Product	Data Column	—	—	—	No direct relationship provided by Microsoft Purview, but can through Data Asset
Data Column	Data Product	—	—	—	No direct relationship provided by Microsoft Purview, but can through Data Asset
Objective	Data Product	—	—	Use Data Product helper direction	Supported and tested
Data Asset	Glossary Term	—	—	Use Glossary Term source direction	Supported and tested
Data Column	Glossary Term	—	—	Use Glossary Term source direction	Supported and tested

---

Recommended Creation Directions

When more than one direction may appear possible, use the direction that has been explicitly tested.

Relationship	Recommended Direction
Data Product and Data Asset	Data Product → Data Asset
Data Product and Glossary Term	Data Product → Glossary Term
Data Product and Objective	Data Product → Objective
Glossary Term and Data Asset	Glossary Term → Data Asset
Glossary Term and Data Column	Glossary Term → Data Column
Glossary Term and CDE	CDE → Glossary Term
CDE and Data Column	CDE → Data Column
Data Asset and Data Column	Data Column → Data Asset

Relationship Direction Matters

Do not assume that every relationship can be created from either side.

A relationship may be readable from both resources while only one direction has been confirmed for creation.

Use the recommended direction unless the reverse direction has been separately tested.

---

Data Product and Data Asset

Recommended direction

Data Product
      │
      ▼
Data Asset

Configuration:

resource_path = "dataProducts"
entity_type = "DATAASSET"

Creation:

relationship = client.relationships.create(
    resource_path="dataProducts",
    resource_id=DATA_PRODUCT_ID,
    entity_type="DATAASSET",
    entity_id=DATA_ASSET_ID,
    relationship_type="Related",
    description=(
        "Data Asset linked to a Data Product."
    ),
)

This direction has been tested with duplicate checking and post-creation verification. :contentReference[oaicite:0]{index=0}

The reverse direction has also been investigated using:

resource_path = "dataAssets"
entity_type = "DATAPRODUCT"

The reverse-side example reads and compares the association from both endpoints. :contentReference[oaicite:1]{index=1}

---

Data Product and Glossary Term

Recommended direction

Data Product
      │
      ▼
Glossary Term

Configuration:

resource_path = "dataProducts"
entity_type = "TERM"

Creation:

relationship = client.relationships.create(
    resource_path="dataProducts",
    resource_id=DATA_PRODUCT_ID,
    entity_type="TERM",
    entity_id=GLOSSARY_TERM_ID,
    relationship_type="Related",
    description=(
        "Glossary Term linked to a Data Product."
    ),
)

This relationship has been tested with duplicate prevention and final verification. :contentReference[oaicite:2]{index=2}

---

Data Product and Objective

Recommended direction

Data Product
      │
      ▼
Objective

Use the dedicated helper method:

relationship = (
    client.relationships
    .add_objective_to_data_product(
        data_product_id=DATA_PRODUCT_ID,
        objective_id=OBJECTIVE_ID,
        relationship_type="Related",
        description=(
            "Links the Data Product to its "
            "governance Objective."
        ),
    )
)

The SDK helper hides the lower-level resource-path and entity-type configuration. :contentReference[oaicite:3]{index=3}

Use the Helper Method

Prefer add_objective_to_data_product() over the generic create() method when connecting an Objective to a Data Product.

---

Glossary Term and Data Asset

Recommended direction

Glossary Term
      │
      ▼
Data Asset

Configuration:

resource_path = "terms"
entity_type = "DATAASSET"

This direction has been tested with an existing-relationship check and post-creation verification. :contentReference[oaicite:4]{index=4}

---

Glossary Term and Data Column

Recommended direction

Glossary Term
      │
      ▼
Data Column

Configuration:

resource_path = "terms"
entity_type = "DATACOLUMN"

This direction has been tested with the outer Unified Catalog Data Column object ID. :contentReference[oaicite:5]{index=5}

Use the Unified Catalog Column ID

Use the outer Unified Catalog object ID for the Data Column.

Do not substitute a nested source-system column identifier such as source.columnId unless the API operation explicitly requires it.

---

Critical Data Element and Glossary Term

Recommended direction

Critical Data Element
          │
          ▼
   Glossary Term

Configuration:

resource_path = "criticalDataElements"
entity_type = "TERM"

This relationship has been tested from the Critical Data Element side. :contentReference[oaicite:6]{index=6}

---

Critical Data Element and Data Column

Recommended direction

Critical Data Element
          │
          ▼
     Data Column

Configuration:

resource_path = "criticalDataElements"
entity_type = "DATACOLUMN"

The tested example:

1. Reads existing Data Column relationships.
2. Checks for the target column ID.
3. Creates the relationship when necessary.
4. Reads the relationships again.
5. Confirms that the Data Column is present. :contentReference[oaicite:7]{index=7}

---

Data Column and Data Asset

Recommended direction

Data Column
      │
      ▼
Data Asset

Configuration:

resource_path = "dataColumns"
entity_type = "DATAASSET"

The tested example verifies the association from both directions:

Data Column → Data Asset
Data Asset  → Data Column

This confirms whether the relationship is visible from both resource endpoints. :contentReference[oaicite:8]{index=8}

---

Direct and Indirect Relationships

Not every association shown in Microsoft Purview is created through a direct relationship endpoint.

For example, a Critical Data Element may appear associated with a Data Product through a shared Data Asset or Data Column.

Critical Data Element
          │
          ▼
Data Column or Data Asset
          │
          ▼
     Data Product

This is an indirect relationship.

A direct operation such as the following has not been established:

Data Product → Critical Data Element

However, the CDE may still appear under the Data Product when both resources are connected through the same technical metadata.

Not Directly Supported Does Not Mean Unrelated

A matrix entry marked Not directly supported means that no direct create operation has been identified.

Microsoft Purview may still infer or display the association through related resources.

---

Relationship Visibility

The following situations are possible after a relationship is created:

Result	Meaning
Visible from both sides	Both resource endpoints return the relationship.
Visible from the source side only	The create direction returns the relationship, but the reverse endpoint does not yet expose it.
Returned by API but not visible in UI	The Unified Catalog interface may not have refreshed yet.
Visible indirectly	Microsoft Purview derives the association through another resource.
Duplicate entries returned	The relationship may have been created more than once.

Always verify through the API before relying only on the Microsoft Purview interface.

---

Generic Configuration Reference

Source Resource	resource_path
Data Product	dataProducts
Data Asset	dataAssets
Data Column	dataColumns
Glossary Term	terms
Critical Data Element	criticalDataElements

Common target entity types include:

Target Resource	entity_type
Data Product	DATAPRODUCT
Data Asset	DATAASSET
Data Column	DATACOLUMN
Glossary Term	TERM

The Objective relationship uses a dedicated SDK helper and does not require callers to provide these generic values directly.

---

Generic Relationship Operation

Most directly supported relationships use:

relationship = client.relationships.create(
    resource_path=SOURCE_RESOURCE_PATH,
    resource_id=SOURCE_RESOURCE_ID,
    entity_type=TARGET_ENTITY_TYPE,
    entity_id=TARGET_ENTITY_ID,
    relationship_type="Related",
    description=(
        "Relationship created by the "
        "Purview Unified SDK."
    ),
)

Existing relationships can be read using:

response = client.relationships.list_raw(
    resource_path=SOURCE_RESOURCE_PATH,
    resource_id=SOURCE_RESOURCE_ID,
    entity_type=TARGET_ENTITY_TYPE,
)

For a complete duplicate-safe creation workflow, see the Relationships guide.

---

Validation Requirements

A relationship should be marked Supported and tested only after the following steps succeed:

Read existing relationships
        ↓
Confirm the target is not already present
        ↓
Create the relationship
        ↓
Read the relationships again
        ↓
Confirm the target entity ID
        ↓
Verify from the reverse side when applicable

A successful create response alone does not prove that the relationship is fully visible from every resource endpoint.

---

Updating This Matrix

When a new relationship is tested:

1. Add it to the main matrix.
2. Record the tested source direction.
3. Record the resource_path.
4. Record the target entity_type.
5. Identify whether a generic or dedicated SDK method was used.
6. Verify the relationship after creation.
7. Note whether it is visible from the reverse side.
8. Add or link the corresponding example file.

Avoid marking a relationship as supported based only on assumptions about resource types.

---

Related Documentation

• Relationships
• Data Map
• Data Products
• Glossary
• Critical Data Elements
• Objectives and Key Results