Amazon Appflow Update: With this update, you can choose which Salesforce API is used by Amazon AppFlow to transfer data to or from your Salesforce account. You can choose the Salesforce REST API or Bulk API 2.0. You can also choose for Amazon AppFlow to pick the API automatically.

Author: AWS

.changes/next-release/feature-AmazonAppflow-ea63842.json

@@ -0,0 +1,6 @@
+{
+    "type": "feature",
+    "category": "Amazon Appflow",
+    "contributor": "",
+    "description": "With this update, you can choose which Salesforce API is used by Amazon AppFlow to transfer data to or from your Salesforce account. You can choose the Salesforce REST API or Bulk API 2.0. You can also choose for Amazon AppFlow to pick the API automatically."
+}

services/appflow/src/main/resources/codegen-resources/service-2.json

@@ -3365,7 +3365,8 @@
         "MATH_OPERATION_FIELDS_ORDER",
         "CONCAT_FORMAT",
         "SUBFIELD_CATEGORY_MAP",
-        "EXCLUDE_SOURCE_FIELDS_LIST"
+        "EXCLUDE_SOURCE_FIELDS_LIST",
+        "INCLUDE_NEW_FIELDS"
       ]
     },
     "Operators":{
@@ -3941,6 +3942,18 @@
       },
       "documentation":"<p> The connector-specific profile properties required when using Salesforce. </p>"
     },
+    "SalesforceDataTransferApi":{
+      "type":"string",
+      "enum":[
+        "AUTOMATIC",
+        "BULKV2",
+        "REST_SYNC"
+      ]
+    },
+    "SalesforceDataTransferApiList":{
+      "type":"list",
+      "member":{"shape":"SalesforceDataTransferApi"}
+    },
     "SalesforceDestinationProperties":{
       "type":"structure",
       "required":["object"],
@@ -3960,6 +3973,10 @@
         "writeOperationType":{
           "shape":"WriteOperationType",
           "documentation":"<p> This specifies the type of write operation to be performed in Salesforce. When the value is <code>UPSERT</code>, then <code>idFieldNames</code> is required. </p>"
+        },
+        "dataTransferApi":{
+          "shape":"SalesforceDataTransferApi",
+          "documentation":"<p>Specifies which Salesforce API is used by Amazon AppFlow when your flow transfers data to Salesforce.</p> <dl> <dt>AUTOMATIC</dt> <dd> <p>The default. Amazon AppFlow selects which API to use based on the number of records that your flow transfers to Salesforce. If your flow transfers fewer than 1,000 records, Amazon AppFlow uses Salesforce REST API. If your flow transfers 1,000 records or more, Amazon AppFlow uses Salesforce Bulk API 2.0.</p> <p>Each of these Salesforce APIs structures data differently. If Amazon AppFlow selects the API automatically, be aware that, for recurring flows, the data output might vary from one flow run to the next. For example, if a flow runs daily, it might use REST API on one day to transfer 900 records, and it might use Bulk API 2.0 on the next day to transfer 1,100 records. For each of these flow runs, the respective Salesforce API formats the data differently. Some of the differences include how dates are formatted and null values are represented. Also, Bulk API 2.0 doesn't transfer Salesforce compound fields.</p> <p>By choosing this option, you optimize flow performance for both small and large data transfers, but the tradeoff is inconsistent formatting in the output.</p> </dd> <dt>BULKV2</dt> <dd> <p>Amazon AppFlow uses only Salesforce Bulk API 2.0. This API runs asynchronous data transfers, and it's optimal for large sets of data. By choosing this option, you ensure that your flow writes consistent output, but you optimize performance only for large data transfers.</p> <p>Note that Bulk API 2.0 does not transfer Salesforce compound fields.</p> </dd> <dt>REST_SYNC</dt> <dd> <p>Amazon AppFlow uses only Salesforce REST API. By choosing this option, you ensure that your flow writes consistent output, but you decrease performance for large data transfers that are better suited for Bulk API 2.0. In some cases, if your flow attempts to transfer a vary large set of data, it might fail with a timed out error.</p> </dd> </dl>"
         }
       },
       "documentation":"<p> The properties that are applied when Salesforce is being used as a destination. </p>"
@@ -3970,6 +3987,10 @@
         "oAuthScopes":{
           "shape":"OAuthScopeList",
           "documentation":"<p> The desired authorization scope for the Salesforce account. </p>"
+        },
+        "dataTransferApis":{
+          "shape":"SalesforceDataTransferApiList",
+          "documentation":"<p>The Salesforce APIs that you can have Amazon AppFlow use when your flows transfers data to or from Salesforce.</p>"
         }
       },
       "documentation":"<p> The connector metadata specific to Salesforce. </p>"
@@ -3989,6 +4010,10 @@
         "includeDeletedRecords":{
           "shape":"Boolean",
           "documentation":"<p> Indicates whether Amazon AppFlow includes deleted files in the flow run. </p>"
+        },
+        "dataTransferApi":{
+          "shape":"SalesforceDataTransferApi",
+          "documentation":"<p>Specifies which Salesforce API is used by Amazon AppFlow when your flow transfers data from Salesforce.</p> <dl> <dt>AUTOMATIC</dt> <dd> <p>The default. Amazon AppFlow selects which API to use based on the number of records that your flow transfers from Salesforce. If your flow transfers fewer than 1,000,000 records, Amazon AppFlow uses Salesforce REST API. If your flow transfers 1,000,000 records or more, Amazon AppFlow uses Salesforce Bulk API 2.0.</p> <p>Each of these Salesforce APIs structures data differently. If Amazon AppFlow selects the API automatically, be aware that, for recurring flows, the data output might vary from one flow run to the next. For example, if a flow runs daily, it might use REST API on one day to transfer 900,000 records, and it might use Bulk API 2.0 on the next day to transfer 1,100,000 records. For each of these flow runs, the respective Salesforce API formats the data differently. Some of the differences include how dates are formatted and null values are represented. Also, Bulk API 2.0 doesn't transfer Salesforce compound fields.</p> <p>By choosing this option, you optimize flow performance for both small and large data transfers, but the tradeoff is inconsistent formatting in the output.</p> </dd> <dt>BULKV2</dt> <dd> <p>Amazon AppFlow uses only Salesforce Bulk API 2.0. This API runs asynchronous data transfers, and it's optimal for large sets of data. By choosing this option, you ensure that your flow writes consistent output, but you optimize performance only for large data transfers.</p> <p>Note that Bulk API 2.0 does not transfer Salesforce compound fields.</p> </dd> <dt>REST_SYNC</dt> <dd> <p>Amazon AppFlow uses only Salesforce REST API. By choosing this option, you ensure that your flow writes consistent output, but you decrease performance for large data transfers that are better suited for Bulk API 2.0. In some cases, if your flow attempts to transfer a vary large set of data, it might fail with a timed out error.</p> </dd> </dl>"
         }
       },
       "documentation":"<p> The properties that are applied when Salesforce is being used as a source. </p>"