curl http://site/webpage | grep pdf | sort -u | sed 's/html code to exclude//' | sed 's/pdf">.*/pdf/' ~/urllist.txt

sed -e 's#http.*com/##' | sed -e 's#/.*##' | sed -e '$!N; /^(.*)n1$/!P; D' < ~/urllist.txt > ~/dirlist.txt
cat ~/dirlist.txt | xargs mkdir