Shouldn't be too hard to scrape If you search by year it's using https://www.lego.com//service/biservice/searchbylaunchyear?fromIndex=0&onlyAlternatives=false&

Shouldn't be too hard to scrape If you search by year it's using https://www.lego.com//service/biservice/searchbylaunchyear?fromIndex=0&onlyAlternatives=false&year=2017 https://www.lego.com//service/biservice/searchbylaunchyear?f... to return a JSON, returning 10 items and the flag MoreData; if MoreData == True, increment fromIndex by 10 and call again and then just iterate over years {1989, 1994..2018} pdf url is also in the JSON :-)