#! /bin/bash

# get urls from duckduckgo, then parse them to wget to download

args=("$@")
site=${args[0]}
filetype=${args[1]}
outdir=${args[2]}

if [ $# != 3 ]
        then
                printf  "\n\x1b[31mWrong number of arguments\x1b[0m\n\n"
                printf "Usage: duckduckget [site] [filetype] [outdir]\n       where [site] is the domain to search for files\n             [filetype] is extension without preceeding dot\n             [outputdir] is the output directory relative to working directory"
                exit 1
fi

if ! [[ $outdir =~ /$ ]]
        then
                $outdir=$outdir/
fi

urls=`curl -silent https://duckduckgo.com/html/\?q\=site:$site%20filetype:$filetype | grep "${filetype}$" | tr -d ' \t\r' | grep -v '^[0-9]' | awk '{print "http://" $0}'`

for url in $urls; do
        wget --cut-dirs=100 -P $outdir $url
done