feat(batch): sequential async processing of records for BatchProcessor (#3109)

Co-authored-by: Andrea Amorosi <dreamorosi@gmail.com>

Author: arnabrahman

docs/utilities/batch.md

@@ -416,7 +416,10 @@ For such cases we recommend to use the `BatchProcessorSync` and `processPartialR
     *If your function is `async` returning a `Promise`, use `BatchProcessor` and `processPartialResponse`
     * If your function is not `async`, use `BatchProcessorSync` and `processPartialResponseSync`
 
-The difference between the two processors in implementation is that `BatchProcessor` uses `Promise.all()` while `BatchProcessorSync` loops through each record to preserve the order.
+The difference between the two processors is in how they handle record processing:
+
+* **`BatchProcessor`**: By default, it processes records in parallel using `Promise.all()`. However, it also offers an [option](#sequential-async-processing) to process records sequentially, preserving the order.
+* **`BatchProcessorSync`**: Always processes records sequentially, ensuring the order is preserved by looping through each record one by one.
 
 ???+ question "When is this useful?"
 
@@ -477,6 +480,16 @@ Let's suppose you'd like to add a metric named `BatchRecordFailures` for each ba
 --8<-- "examples/snippets/batch/extendingFailure.ts"
 ```
 
+### Sequential async processing
+
+By default, the `BatchProcessor` processes records in parallel using `Promise.all()`. However, if you need to preserve the order of records, you can set the `processInParallel` option to `false` to process records sequentially.
+
+!!! important "If the `processInParallel` option is not provided, the `BatchProcessor` will process records in parallel."
+
+```typescript hl_lines="8 17" title="Sequential async processing"
+--8<-- "examples/snippets/batch/sequentialAsyncProcessing.ts"
+```
+
 ### Create your own partial processor
 
 You can create your own partial batch processor from scratch by inheriting the `BasePartialProcessor` class, and implementing the `prepare()`, `clean()`, `processRecord()` and `processRecordSync()` abstract methods.

examples/snippets/batch/sequentialAsyncProcessing.ts

@@ -0,0 +1,18 @@
+import {
+  BatchProcessor,
+  EventType,
+  processPartialResponse,
+} from '@aws-lambda-powertools/batch';
+import type { SQSHandler, SQSRecord } from 'aws-lambda';
+
+const processor = new BatchProcessor(EventType.SQS);
+
+const recordHandler = async (_record: SQSRecord): Promise<void> => {
+  // Process the record
+};
+
+export const handler: SQSHandler = async (event, context) =>
+  processPartialResponse(event, recordHandler, processor, {
+    context,
+    processInParallel: false,
+  });

packages/batch/src/BasePartialProcessor.ts

@@ -98,7 +98,7 @@ abstract class BasePartialProcessor {
   public abstract prepare(): void;
 
   /**
-   * Process all records with an asyncronous handler
+   * Process all records with an asynchronous handler
    *
    * Once called, the processor will create an array of promises to process each record
    * and wait for all of them to settle before returning the results.
@@ -122,21 +122,21 @@ abstract class BasePartialProcessor {
     }
     this.prepare();
 
-    const processingPromises: Promise<SuccessResponse | FailureResponse>[] =
-      this.records.map((record) => this.processRecord(record));
-
-    const processedRecords: (SuccessResponse | FailureResponse)[] =
-      await Promise.all(processingPromises);
+    // Default to `true` if `processInParallel` is not specified.
+    const processInParallel = this.options?.processInParallel ?? true;
+    const processedRecords = processInParallel
+      ? await this.#processRecordsInParallel()
+      : await this.#processRecordsSequentially();
 
     this.clean();
 
     return processedRecords;
   }
 
   /**
-   * Process a record with an asyncronous handler
+   * Process a record with an asynchronous handler
    *
-   * An implementation of this method is required for asyncronous processors.
+   * An implementation of this method is required for asynchronous processors.
    *
    * When implementing this method, you should at least call the successHandler method
    * when a record succeeds processing and the failureHandler method when a record
@@ -249,6 +249,30 @@ abstract class BasePartialProcessor {
 
     return entry;
   }
+
+  /**
+   * Processes records in parallel using `Promise.all`.
+   */
+  async #processRecordsInParallel(): Promise<
+    (SuccessResponse | FailureResponse)[]
+  > {
+    return Promise.all(
+      this.records.map((record) => this.processRecord(record))
+    );
+  }
+
+  /**
+   * Processes records sequentially, ensuring that each record is processed one after the other.
+   */
+  async #processRecordsSequentially(): Promise<
+    (SuccessResponse | FailureResponse)[]
+  > {
+    const processedRecords: (SuccessResponse | FailureResponse)[] = [];
+    for (const record of this.records) {
+      processedRecords.push(await this.processRecord(record));
+    }
+    return processedRecords;
+  }
 }
 
 export { BasePartialProcessor };

packages/batch/src/types.ts

@@ -14,6 +14,7 @@ import type { SqsFifoPartialProcessor } from './SqsFifoPartialProcessor.js';
  * @property context The context object provided by the AWS Lambda runtime
  * @property skipGroupOnError The option to group on error during processing
  * @property throwOnFullBatchFailure The option to throw an error if the entire batch fails
+ * @property processInParallel Indicates whether the records should be processed in parallel
  */
 type BatchProcessingOptions<T = BasePartialBatchProcessor> = {
   /**
@@ -30,6 +31,12 @@ type BatchProcessingOptions<T = BasePartialBatchProcessor> = {
    *  Set this to false to prevent throwing an error if the entire batch fails.
    */
   throwOnFullBatchFailure?: boolean;
+  /**
+   * Indicates whether the records should be processed in parallel.
+   * When set to `true`, the records will be processed in parallel using `Promise.all`.
+   * When set to `false`, the records will be processed sequentially.
+   */
+  processInParallel?: T extends SqsFifoPartialProcessor ? never : boolean;
 };
 
 /**