Download wikimedia commons Google Art

download_google_artworks_wikimedia.sh

#download wikimedia commons Google Art - almost exclusively "flat" art: paintings, drawings, tapestries, ...
mkdir artworks
#two pages contain all top references
wget https://commons.wikimedia.org/wiki/Category:Google_Art_Project_works_by_collection?uselang=en-gb -O top_index.html
wget 'https://commons.wikimedia.org/w/index.php?title=Category:Google_Art_Project_works_by_collection&uselang=en-gb&subcatfrom=Norman+Rockwell+Museum%0AGoogle+Art+Project+works+in+Norman+Rockwell+Museum' -O top_index_2.html

cat top_index_2.html >> top_index.html
rm top_index_2.html

#gallery page urls
cat top_index.html | grep "CategoryTreeSection" | sed 's/.*href=\"\([^\"]*\)\">.*/https:\/\/commons.wikimedia.org\1/g' > galleries.list

#for each gallery
rm artworks.list
while read gallery; do
    wget "${gallery}" -O gallery.html
    cat gallery.html | grep "<div class=\"thumb\"" | sed 's/.*href=\"\([^\"]*\)\".*/https:\/\/commons.wikimedia.org\1/g' >> artworks.list
done <galleries.list

#for each artwork
while read artwork; do
    wget "${artwork}" -O artwork.html
    cat artwork.html | grep "Size of this preview:" | sed 's/.*preview: <a href=\"\([^\"]*\)".*/\1/g' > artwork
    while read artwork_link; do
        wget "${artwork_link}" --directory-prefix=artworks
    done <artwork
    sleep 1
done <artworks.list