src/v0/destinations/fb_custom_audience/recordTransform.ts (1)

119-145: Consider whether async iteration is necessary for synchronous processing.

The inner forEachInBatches at line 121 wraps a synchronous processRecord call. This introduces unnecessary async overhead. A regular forEach or for...of loop would be more efficient here.

♻️ Suggested refactor

   await forEachInBatches(recordChunksArray, async (recordArray) => {
     const data: unknown[][] = [];
-    await forEachInBatches(recordArray, async (input) => {
-      const result = processRecord(
+    for (const input of recordArray) {
+      const result = processRecord(
         input,
         userSchema,
         isHashRequired,
         disableFormat,
         input.metadata.workspaceId,
         destination.ID,
       );
       if (result.error) {
         const error = new InstrumentationError(result.error);
         const errorObj = generateErrorObject(error);
         invalidEvents.push(
           getErrorRespEvents(
             [result.metadata],
             errorObj.status,
             errorObj.message,
             errorObj.statTags,
           ),
         );
       } else {
         data.push(result.dataElement!);
         metadata.push(result.metadata);
       }
-    });
+    }

🤖 Prompt for AI Agents

Verify each finding against the current code and only fix it if needed.

In `@src/v0/destinations/fb_custom_audience/recordTransform.ts` around lines 119 -
145, The inner async batch iterator is unnecessary because processRecord is
synchronous; replace the inner await forEachInBatches(recordArray, async (input)
=> { ... }) with a synchronous loop (e.g., for...of or Array.prototype.forEach)
over recordArray, call processRecord directly, and push to data/metadata or
invalidEvents as before, removing the async callback and the extra await to
avoid unnecessary async overhead; keep outer forEachInBatches over
recordChunksArray intact and preserve usage of variables recordArray,
processRecord, invalidEvents, data, and metadata.