Update and delete table rows using the Web API

Operations to modify data are a core part of the Web API. In addition to simple update and delete operations, you can perform operations on single table columns (entity attributes) and compose upsert requests that will either update or insert data depending on whether it exists.

Basic update

Update operations use the HTTP PATCH verb. Pass a JSON object containing the properties you want to update to the URI that represents the record. A response with a status of 204 No Content is returned if the update is successful.

The If-Match: * header ensures you don't create a new record by accidentally performing an upsert operation. More information: Prevent create in upsert.

Important

When updating an entity, only include the properties you are changing in the request body. Simply updating the properties of an entity that you previously retrieved, and including that JSON in your request, will update each property even though the value is the same. This can cause system events that can trigger business logic that expects that the values have changed. This can cause properties to appear to have been updated in auditing data when in fact they haven't actually changed.

When you update the statecode property, it is important to always set the desired statuscode. statecode and statuscode have dependent values. There can be multiple valid statuscode values for a given statecode value, but each statecode column has a single DefaultStatus value configured. When you update statecode without specifying a statuscode, the default status value will be set by the system. Also, when auditing is enabled on the table and the statuscode column, the changed value for the statuscode column will not be captured in the audit data unless it is specified in the update operation.

Note

The definition for attributes includes a RequiredLevel property. When this is set to SystemRequired, you cannot set these attributes to a null value. More information: Attribute requirement level

This example updates an existing account record with the accountid value of 00000000-0000-0000-0000-000000000001.

Request:

PATCH [Organization URI]/api/data/v9.2/accounts(00000000-0000-0000-0000-000000000001) HTTP/1.1  
Content-Type: application/json  
OData-MaxVersion: 4.0  
OData-Version: 4.0
If-Match: *  
  
{  
    "name": "Updated Sample Account ",  
    "creditonhold": true,  
    "address1_latitude": 47.639583,  
    "description": "This is the updated description of the sample account",  
    "revenue": 6000000,  
    "accountcategorycode": 2  
}  

Response:

HTTP/1.1 204 No Content  
OData-Version: 4.0  
  

Note

See Using single-valued navigation properties for information about associating and disassociating entities on update.

Update with data returned

To retrieve data from an entity you're updating, you can compose your PATCH request so that data from the created record is returned with a status of 200 (OK). To get this result, you must use the Prefer: return=representation request header.

To control which properties are returned, append the $select query option to the URL to the entity set. The $expand query option is ignored if used.

This example updates an account entity and returns the requested data in the response.

Request:

PATCH [Organization URI]/api/data/v9.2/accounts(00000000-0000-0000-0000-000000000001)?$select=name,creditonhold,address1_latitude,description,revenue,accountcategorycode,createdon HTTP/1.1  
OData-MaxVersion: 4.0  
OData-Version: 4.0  
Accept: application/json  
Content-Type: application/json; charset=utf-8  
Prefer: return=representation
If-Match: * 
  
{"name":"Updated Sample Account"}  

Response:

HTTP/1.1 200 OK  
Content-Type: application/json; odata.metadata=minimal  
Preference-Applied: return=representation  
OData-Version: 4.0  
  
{  
    "@odata.context": "[Organization URI]/api/data/v9.2/$metadata#accounts/$entity",  
    "@odata.etag": "W/\"536537\"",  
    "accountid": "00000000-0000-0000-0000-000000000001",  
    "accountcategorycode": 1,  
    "description": "This is the description of the sample account",  
    "address1_latitude": 47.63958,  
    "creditonhold": false,  
    "name": "Updated Sample Account",  
    "createdon": "2016-09-28T23:14:00Z",  
    "revenue": 5000000.0000,  
    "_transactioncurrencyid_value": "048dddaa-6f7f-e611-80d3-00155db5e0b6"  
}  
  

Update multiple records in a single request

The fastest way to update multiple records of the same type in a single request is to use the UpdateMultiple action. At the time of this writing, the UpdateMultiple action. Not all standard tables support this action, but all elastic tables do.

More information:

• Bulk Operation messages
• Sample: Web API Use bulk operations
• Use UpdateMultiple with elastic tables

Update a single property value

When you want to update only a single property value, use a PUT request with the property name appended to the Uri of the entity.

The following example updates the name property of an existing account row with the accountid value of 00000000-0000-0000-0000-000000000001.

Request:

PUT [Organization URI]/api/data/v9.2/accounts(00000000-0000-0000-0000-000000000001)/name HTTP/1.1  
Content-Type: application/json  
OData-MaxVersion: 4.0  
OData-Version: 4.0  
  
{"value": "Updated Sample Account Name"}  

Response:

HTTP/1.1 204 No Content  
OData-Version: 4.0  
  

Delete a single property value

To delete the value of a single property use a DELETE request with the property name appended to the Uri of the entity.

The following example deletes the value of the description property of an account entity with the accountid value of 00000000-0000-0000-0000-000000000001.

Request:

DELETE [Organization URI]/api/data/v9.2/accounts(00000000-0000-0000-0000-000000000001)/description HTTP/1.1  
Content-Type: application/json  
OData-MaxVersion: 4.0  
OData-Version: 4.0  

Response:

HTTP/1.1 204 No Content  
OData-Version: 4.0  
  

Note

This can't be used with a single-valued navigation property to disassociate two entities. For an alternative approach, see Disassociate with a single-valued navigation property .

Upsert a table row

An upsert operation is similar to an update. It uses a PATCH request and uses a URI to reference a specific record. The difference is that if the record doesn't exist, it's created. If it already exists, it's updated.

Upsert is valuable when synchronizing data between external systems. The external system may not contain a reference to the primary key of the Dataverse table, so you can configure alternate keys for the Dataverse table using values from the external system that uniquely identify the record on both systems. More information: Define alternate keys to reference rows

You can see any alternate keys that are defined for a table in the annotations for the entity type in the $metadata service document. More information: Alternate Keys.

In the following example, there's a table with the name sample_thing that has an alternate key that refers to two columns: sample_key1 and sample_key2, which are both defined to store integer values.

Request:

PATCH [Organization URI]/api/data/v9.2/sample_things(sample_key1=1,sample_key2=1) HTTP/1.1
Accept: application/json 
OData-MaxVersion: 4.0
OData-Version: 4.0
If-None-Match: null
Content-Type: application/json

{
    "sample_name": "1:1"
}

For both create or update operations you get the same response. Notice how the OData-EntityId response header uses the key values rather than the GUID primary key identifier for the record.

Response:

HTTP/1.1 204 No Content
OData-Version: 4.0
OData-EntityId: [Organization URI]/api/data/v9.2/sample_things(sample_key1=1,sample_key2=1)

Because the response is the same, you can't know whether the operation represented a Create or Update operation.

If you need to know, you can use the Prefer: return=representation request header. With this header, you get a 201 Created response when a record is created and a 200 OK response when the record is updated. This option adds a Retrieve operation, which has an impact on performance. If you use the Prefer: return=representation request header, make sure that your $select includes the minimal amount of data, preferably only the primary key column. More information: Update with data returned and Create with data returned.

When using alternate keys, you shouldn't include the alternate key values in the body of the request.

• When an upsert represents an Update, these alternate key values are ignored. You can't update alternate key values while using them to identify the record.
• When an upsert represents a Create, the key values in the URL are set for the record if they aren't present in the body. So there's no need to include them in the body of the request.

More information: Use Upsert to Create or Update a record

Note

Normally when creating a new record you will let the system assign a GUID value for the primary key. This is a best practice because the system generates keys that are optimized for the index and this improves performance. But if you need to create a record with a specific primary key value, such as when the key GUID value is generated by an external system, the upsert operation provides a way to do this.

Prevent create or update with upsert

Sometimes there are situations where you want to perform an upsert, but you want to prevent one of the potential operations: either create or update. You can do this using the If-Match or If-None-Match headers. For more information, see Limit upsert operations.

Basic delete

A delete operation is straightforward. Use the DELETE verb with the URI of the entity you want to delete. This example message deletes an account entity with the primary key accountid value equal to 00000000-0000-0000-0000-000000000001.

Request:

DELETE [Organization URI]/api/data/v9.2/accounts(00000000-0000-0000-0000-000000000001) HTTP/1.1  
Content-Type: application/json  
OData-MaxVersion: 4.0  
OData-Version: 4.0  

Response:

If the entity exists, you get a normal response with status 204 to indicate the delete was successful. If the entity isn't found, you get a response with status 404.

HTTP/1.1 204 No Content  
OData-Version: 4.0  

Check for duplicate records

For more information on how to check for duplicate records during an update operation, see Detect duplicates during Update operation using the Web API.

Delete multiple records in a single request

The fastest way to delete multiple records of the same type in a single request is to use the DeleteMultiple action. At the time of this writing, the DeleteMultiple action is a preview feature. Standard tables don't support this action, but all elastic tables do.

Note

For standard tables, we recommend using the BulkDelete action, that enables asynchronous deletion of records that match a query. More information: Delete data in bulk

More information:

• Bulk Operation messages
• Elastic table sample code
• Use DeleteMultiple with elastic tables

Update and delete documents in storage partitions

If you're updating or deleting elastic table data stored in partitions, be sure to specify the partition key when accessing that data.

More information: Choosing a PartitionId value

See also

Web API Basic Operations Sample (C#)
Web API Basic Operations Sample (Client-side JavaScript)
Perform operations using the Web API
Compose Http requests and handle errors
Query Data using the Web API
Create a table row using the Web API
Retrieve a table row using the Web API
Associate and disassociate table rows using the Web API
Use Web API functions
Use Web API actions
Execute batch operations using the Web API
Impersonate another user using the Web API
Perform conditional operations using the Web API