With a list of object names as argument, this script download each pdf file from AWS S3 and converts the pages to JPG images

convert-pdf.sh

#!/bin/sh

bucket="multiplat-resources-homolog/";
prefix="content/";

while read -r line || [[ -n $line ]]; do
  echo "downloading file $line";
  pdfname=$(echo "$line" | sed 's/-original//');
  aws s3 cp s3://$bucket$prefix$line $pdfname.pdf;
  mkdir $pdfname;
  echo "generating the pages...";
  time gm convert -density 200 -quality 100 -geometry 1300x1300 $pdfname.pdf +adjoin jpg:$pdfname/page%06d
  echo "done $pdfname.";
done < "$1"