add option to upload dependency layer to S3 first

Author: Josh Rosenberg

README.md

@@ -16,19 +16,23 @@ In order for the Action to have access to the code, you must use the `actions/ch
 
 ### Environment variables
 Stored as secrets or env vars, doesn't matter. But also please don't put your AWS keys outside Secrets.
-- **AWS Credentials**  
+- **AWS Credentials**
     That includes the `AWS_ACCESS_KEY_ID`, `AWS_SECRET_ACCESS_KEY`, etc. It's used by `awscli`, so the docs for that [can be found here](https://docs.aws.amazon.com/cli/latest/userguide/cli-configure-envvars.html).
 
 ### Inputs
-- `lambda_layer_arn`  
+- `lambda_layer_arn`
     The ARN for the Lambda layer the dependencies should be pushed to **without the version** (every push is a new version).
-- `lambda_function_name`  
+- `lambda_function_name`
     The Lambda function name. [From the AWS docs](https://docs.aws.amazon.com/cli/latest/reference/lambda/update-function-code.html), it can be any of the following:
-    - Function name - `my-function`  
-    - Function ARN - `arn:aws:lambda:us-west-2:123456789012:function:my-function`  
+    - Function name - `my-function`
+    - Function ARN - `arn:aws:lambda:us-west-2:123456789012:function:my-function`
     - Partial ARN - `123456789012:function:my-function`
 - `requirements_txt`
     The name/path for the `requirements.txt` file. Defaults to `requirements.txt`.
+- `use_s3`
+    Whether to upload the dependency layer zip to S3 (required if the zip exceeds 50MB) or not. (If not, it's uploaded directly to Lambda.) Defaults to `false`.
+- `s3_bucket_name`
+    The S3 bucket name used if you are uploading the dependency layer to S3.
 
 
 ### Example workflow
@@ -41,7 +45,7 @@ on:
 jobs:
   build:
     runs-on: ubuntu-latest
-    
+
     steps:
     - uses: actions/checkout@master
     - name: Deploy code to Lambda

action.yml

@@ -12,6 +12,14 @@ inputs:
   lambda_function_name:
     description: The Lambda function name. Check the AWS docs/readme for examples.
     required: true
+  use_s3:
+    description: Whether to use S3 (true) or not (false) for the dependencies layer upload.
+    required: false
+    default: 'false'
+  s3_bucket_name:
+    description: The S3 bucket name for the dependencies layer upload.
+    required: false
+    default: no-bucket-name-here
 runs:
   using: 'docker'
   image: 'Dockerfile'

entrypoint.sh

@@ -9,9 +9,18 @@ install_zip_dependencies(){
 }
 
 publish_dependencies_as_layer(){
-	echo "Publishing dependencies as a layer..."
-	local result=$(aws lambda publish-layer-version --layer-name "${INPUT_LAMBDA_LAYER_ARN}" --zip-file fileb://dependencies.zip)
-	LAYER_VERSION=$(jq '.Version' <<< "$result")
+	if [ "$INPUT_USE_S3" = true ]
+	then
+		echo "Uploading dependencies to S3..."
+		aws s3 cp dependencies.zip s3://"${INPUT_S3_BUCKET_NAME}"/dependencies.zip
+		echo "Publishing dependencies from S3 as a layer..."
+		local result=$(aws lambda publish-layer-version --layer-name "${INPUT_LAMBDA_LAYER_ARN}" --content S3Bucket="${INPUT_S3_BUCKET_NAME}",S3Key=dependencies.zip)
+		LAYER_VERSION=$(jq '.Version' <<< "$result")
+	else
+		echo "Publishing dependencies as a layer..."
+		local result=$(aws lambda publish-layer-version --layer-name "${INPUT_LAMBDA_LAYER_ARN}" --zip-file fileb://dependencies.zip)
+		LAYER_VERSION=$(jq '.Version' <<< "$result")
+	fi
 	rm -rf python
 	rm dependencies.zip
 }