Find file History
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
..
Failed to load latest commit information.
README.md
child-process-promise.js
convert.js
main.js
package.json
s3-util.js

README.md

Pandoc S3 Lambda converter

Lambda function that waits for files uploaded to S3, converts them to docx using the Pandoc Lambda Layer and uploads back to S3.

This example shows how to wire up S3 file conversion that runs an external processor, in this case Pandoc, a Swiss army knife for document conversion. Check out the Running Pandoc on Lambda Guide for more information.

Prerequisites

  • create a S3 bucket for file uploads

Setting up the converter

  1. run npm install to fetch the dependencies
  2. run npm start to deploy the initial lambda version
  3. modify the package.json connect script to use your bucket name
  4. run npm run connect to configure your S3 bucket to invoke the lambda function when a new file is uploaded to the /in directory

Try it out

Once you've installed everything, send a test file, for example a markdown file, to your bucket using the S3 console, or the AWS CLI tools. The command lines below assume the bucket is called pandoc-test-bucket, so adjust the commands for your bucket name accordingly.

aws s3 cp example.md s3://pandoc-test-bucket/in/example.md

Wait a few seconds, and then check if the /out folder of your S3 bucket

aws s3 ls s3://pandoc-test-bucket/out/

Download the file with the same base name, but the docx extension, from the /out folder:

aws s3 cp s3://pandoc-test-bucket/out/example.docx .