Sharing Python code as a lambda layer using the Serverless framework

问题

This is a follow up question to: This Question

I am trying to upload some shared Python code as a lambda layer to AWS Lambda using the serverless framework.

I had followed @msc's solution (with some modifications):

Project 1:

Step 1: Create a serverless project with the following structure:

./
└ serverless.yml
└ common/
    └ python/
        └ Other packages as per requirements.txt
        └ my_shared_script.py - my own package

Step 2: Setup my .yml file:

service: library
  provider:
    name: aws
    runtime: python3.7
    stage: dev
    region: ap-northeast-1
  layers:
    Common:
      path: common
  resources:
    Outputs:
      CommonLayerExport:
        Value:
          Ref: CommonLambdaLayer
        Export:
          Name: CommonLambdaLayer

Step 3: Install my requirements into the common/python/ folder:

requirements.txt:

pandas==1.0.5

Install requirements to folder: pip install -r requirements.txt ./common/python/

Project 2

Step 1: Reference the uploaded layer as from Project 1 in my .yml file:

functions:
  hello:
    handler: handler.hello
    layers:
      - arn:aws:lambda:ap-southeast-1:182739821739:layer:Common:1

Step 2: All imports now work. From my handler.py:

import pandas
import my_shared_script

def hello(event, context):
    print(my_shared_script.foo())

Function foo() runs as expected.

My question now is that my_shared_script.py is now in the same folder as all the installed Python packages (there are a lot of files and folders), which makes it hard to find and maintain my own modules / scripts.

./
└ serverless.yml
└ common/
    └ python/
        └ Other packages as per requirements.txt
    └ lib/
        └ my_shared_script.py - my own package

It makes sense for me to put my own scripts in a separate folder /lib and install all other requirements to the /python folder.

However, if I use the above folder structure, the code fails.

I have put this at the top of my handler.py in Project 2, hoping that the script would be able to find my code in /lib.

import os
import sys

CWD = os.path.dirname(os.path.realpath(__file__))
sys.path.insert(0, os.path.abspath(os.path.join(CWD, "../lib")))

I am struggling to understand Python pathing / folder structure in this serverless framework ... Hopefully some expert can share their solution, thank you.

回答1:

Try this

import pandas
import sys
try:
    sys.path.index('/opt/lib')
except ValueError:
    sys.path.append('/opt/lib')
import my_shared_script

def hello(event, context):
    print(my_shared_script.foo())

来源：https://stackoverflow.com/questions/63107494/sharing-python-code-as-a-lambda-layer-using-the-serverless-framework

标签

python

serverless-framework