allen-munsch/another-example.md

## another-example.md

      
    Raw
  

              another-example.md
            
          
    sudo yum install update
sudo yum groupinstall -y "Development Tools"
sudo yum install -y python-devel libpng-devel freetype-devel gcc72-c++

If you want to use docker, you can use the following command:
  docker run -it lambci/lambda:build-python3.7 bash

Otherwise, launch an EC2 instance and ssh in. From there, we can use virtual environments to package our code.
virtualenv v-env
source v-env/bin/activate
pip install lxml
deactivate

mkdir python
cp -a ./v-env/lib/python3.7/site-packages/. ./python
cp -a ./v-env/lib64/python3.7/site-packages/. ./python

zip -r9 ../layer37.zip python

Which will give us a zipped folder containing a structure like
  ./python/
    lxml/
      ...

Where each of the underlying .so files correctly references the existing system libraries. We can use this structure as a layer. To download it, if you used docker, open another terminal window and run:
  docker ps (note the Container ID!)
  docker cp {container id}:/var/layer37.zip .

Or from the EC2 instance:
  scp -i {key location} ec2-user@{instance dns address}:layer37.zip ./layer.zip

We can add it to any existing Lambda functions which need lxml and don't currently have the improperly packaged lxml. To create the layer, navigate to the Lambda Layers console and follow the prompts to create a new layer, uploading the layer37.zip
file. Be careful to select the compatible runtime environments. Now, navigate to your Lambda function and add the layer! You'll be able to successfully import lxml.
START RequestId: 066db9dd-a447-446d-b6bc-85b885121dd7 Version: $LATEST
(4, 3, 4, 0)
completed all processes
END RequestId: 066db9dd-a447-446d-b6bc-85b885121dd7
REPORT RequestId: 066db9dd-a447-446d-b6bc-85b885121dd7	Duration: 1.31 ms	Billed Duration: 100 ms 	Memory Size: 3008 MB	Max Memory Used: 65 MB	

If you want to do this using a deployment package, simply navigate to the inside of your site packages folder, zip those files, and download that artifact to your local machine or build server. From there run zip -g {zip_name} {files/folders to include} where you might include for instance function.py
.

  
## lambda_function.py
from lxml import etree

def lambda_handler(event, context):
    return {'content' : 'ok'}

## lxml_on_lambda.md

      
    Raw
  

              lxml_on_lambda.md
            
          
    See: https://docs.aws.amazon.com/lambda/latest/dg/lambda-runtimes.html

Start a new EC2 instance with the Amazon Linux AMI
sudo yum install gcc gcc-c++ libjpeg-devel zlib-devel libevent-devel libxml2-devel libxslt-devel libpng-devel
sudo yum install python36-devel python36-pip
virtualenv venv_ok
source venv_ok/bin/activate
easy_install lxml
zip -9 bundle.zip lambda_function.py
for dir in $VIRTUAL_ENV/{lib,lib64}/python3.6/site-packages;do pushd $dir;zip -r9 $HOME/bundle.zip .;popd;done
Upload the bundle.zip file to your Lambda function

This assumes a default Handler set to lambda_function.lambda_handler


Delete your EC2 instance
	from lxml import etree

	def lambda_handler(event, context):
	return {'content' : 'ok'}