#!/bin/ksh
#
# DIY links checker
#

debug=0

[[ -z $1 ]] && print single url? && exit 1
target=`echo "$1" | sed -r 's@.*://@@'`

# looking for http/s links
#urls=`curl -s $target | egrep -o 'http[s]?://[^ "<$]+' | sort -u`

# looking for hrefs
# TODO also deal with href='
# TODO check that g works -- multiple occurences per line
urls=`curl -A "Mozilla/5.0 (iPhone; CPU iPhone OS 11_0 like Mac OS X) AppleWebKit/604.1.38 (KHTML, like Gecko) Version/11.0 Mobile/15A372 Safari/604.1" --max-time 4 -sl $target | sed -rn 's@.*[Hh][Rr][Ee][Ff]="([^"]+)".*@\1@gp' | sort -u`
#(( debug == 1 )) && print urls are "$urls"

# beware of the carriage return after 200 OK
# and **the space** and carriage return after HTTP/2 200
# TODO check ^http probably does ^t ^p
for url in $urls; do
        if [[ $url = http* ]]; then
		(( debug == 1 )) && print full url goes as $url
	elif [[ $url = /* ]]; then
		url=${target%%/*}$url
		(( debug == 1 )) && print absolute link goes as $url
	elif [[ $url = [!/]* ]]; then
		url=${target%/*}/$url
		(( debug == 1 )) && print relative link goes as $url
	else
		print Error: could not parse URL $url
		exit 1
	fi

	header=`curl -A "Mozilla/5.0 (iPhone; CPU iPhone OS 11_0 like Mac OS X) AppleWebKit/604.1.38 (KHTML, like Gecko) Version/11.0 Mobile/15A372 Safari/604.1" --max-time 4 -sIl $url | head -1`
	(( debug == 1 )) && print header is $header

	[[ -z $header ]] && print $url DEAD OR TOO SLOW && continue

	[[ "$header" != "HTTP/1.1 200 OK
"
	&& "$header" != "HTTP/2 200 
"
	&& "$header" != "HTTP/1.1 302 Found
" ]] && print $url $header
	unset header
done; unset url