How to download S3 data in Lambda function

[mgreiner79]

Amplify CLI Version

12.14.2

Amplify Version

^6.12.3

Description of use case

I have a lambda function. That lambda function should be able to read data from an S3 bucket. How do I pass the bucket name to the lambda function?

I have a lambda function defined like this

// amplify/functions/myFunction/resource.ts
import { defineFunction } from '@aws-amplify/backend';

export const myFunction = defineFunction({
  name: 'my-function',
  entry: './handler.ts',
  environment: {
      API_URL: 'http://some_url.com'
   }
});

// amplify/data/resource.ts
import { type ClientSchema, a, defineData } from '@aws-amplify/backend';
import { myFunction } from '../functions/myFunction/resource';

const schema = a.schema({
  myFunction: a
    .query()
    .arguments({
      objectKey: a.string(),
    })
    .returns(a.string())
    .handler(a.handler.function(myFunction)),
});

export type Schema = ClientSchema<typeof schema>;

export const data = defineData({
  schema,
  authorizationModes: {
    defaultAuthorizationMode: 'userPool',
    apiKeyAuthorizationMode: {
      expiresInDays: 30,
    },
  },
});

I have my storage defined like this

// amplify/storage/resource.ts

import { defineStorage } from '@aws-amplify/backend';
import { myFunction } from '../functions/myFunction/resource'

export const storage = defineStorage({
  name: 'my-bucket',
  access: (allow) => ({
    'media/*': [
      allow.resource(myFunction).to(['read', 'write'])
    ],
  })
});

Note, that in order to give my lambda function permissions to access the S3 bucket, I need to import my function into my storage definition. This means that I cannot import storage into my function definition to pass the bucket name as an environment variable, because that would be a cyclic dependency.

So instead, I try adding the environement variable to the lamdba function in the backend definition, like this:

// amplify/backend.ts
import { defineBackend } from '@aws-amplify/backend';
import { auth } from './auth/resource';
import { data } from './data/resource';
import { storage } from './storage/resource';
import { myFunction } from './functions/myFunction/resource';

const backend = defineBackend({
  auth,
  data,
  myFunction,
  storage,
});

backend.myFunction.addEnvironment("BUCKET_NAME", backend.storage.resources.bucket.bucketName)
backend.myFunction.addEnvironment("BUCKET_REGION", backend.storage.stack.region)

I can see after doing this, in the AWS lambda console, the environment variable BUCKET_NAME is set.
However, when I try to use it in my handler, by passing it to the downloadData function, like this:

// amplify/function/myFunction/handler.ts
import { Amplify } from 'aws-amplify';
import { generateClient } from 'aws-amplify/data';
import { downloadData } from 'aws-amplify/storage';
import type { Schema } from '../../data/resource';
import { getAmplifyDataClientConfig } from '@aws-amplify/backend/function/runtime';
import { env } from '$amplify/env/my-function'

const { resourceConfig, libraryOptions } = await getAmplifyDataClientConfig(env);

Amplify.configure(resourceConfig, libraryOptions);

const apiUrl = env.API_URL; // This works
const bucketName = process.env.BUCKET_NAME ?? ''
const bucketRegion = process.env.BUCKET_REGION ?? ''

export const handler: Schema['myFunction']['functionHandler'] = async (
  event,
) => {
    const { objectKey } = event.arguments;

    const s3Result = await downloadData({
          path: objectKey,
          options: {
            bucket: {
                bucketName: bucketName,
                region: bucketRegion
          }
        }).result;
...

Note that I cannot use env.BUCKET_NAME to access the parameter, because it does not exist on env (presumably because it was not provided under environment property during the defineFunction call).

While this works, I don't think it is how the Amplify Team intended it to work.
If I do not provide the buckeName to the downloadData function, then I get this error:
NoBucket: Missing bucket name while accessing object
According to the (Amplify docs)[https://docs.amplify.aws/javascript/build-a-backend/storage/download-files/] some default bucket should be used if I do tno provide one. But it is unclear how the config for this default bucket is to be passed into the lambda handler.

How am I supposed to be importing my bucket configuration into the lamdba handler?