Silly script to find interlaced images on Commons (bug 17645)

commons-interlace.sh

#!/bin/bash
# commons-interlace.sh: silly script to find interlaced images on Commons

cat jpgcommons.txt |   # Take list of filenames, one per line
while read line   # As long as there is another line to read ...
do
  URL=$(curl "http://commons.wikimedia.org/w/api.php?action=query&prop=imageinfo&iiprop=url&titles=File:$line&format=xml" | grep -oE 'http://upload.wikimedia.org[^"]+');
  echo "URL is $URL"
  IDEN=$(curl $URL | identify -verbose -)
  if grep -qi "Interlace: None" <<< $IDEN; then
    echo ">>>>>>> $line is not interlaced <<<<<<<" #Only if we found it and exit status is 0
  else
    echo ">>>>>>> $line is interlaced <<<<<<<"
    echo "$line" >> interlaced.txt
  fi
done