#!/bin/sh
#
# hhttpd
#
# Copyright (c) 2021 Hayden Hamilton <dev@hhvn.uk>

# Permission to use, copy, modify, and distribute this software for any
# purpose with or without fee is hereby granted, provided that the above
# copyright notice and this permission notice appear in all copies.

# THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES
# WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF
# MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR
# ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
# WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN
# ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF
# OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
#
# git://haydenvh.com/hhttpd

                                 ## README ##
# Run script from xinetd
# Script will serve files from the directory provided as it's first argument
# for example, if the script is run as `/home/httpd/bin/httpd.sh /srv/www`
# files from the directory /srv/www/ will be served.
#
# converts *.gph files into html, and has a gateway for external gopher content 
                               ## end README  ##

                              ## configuration ##
# default hostname
hostname="hhvn.uk"

# name or path to torsocks script, to use for the gopher proxy, if set
#torgopher=""
 torgopher="torsocks"

# name or path to torsocks script. If set, use /?tor=....
#torweb=""
 torweb="torsocks"

# redirect to hidden service, if set
onionhost="zl7kawb6nd3aqzrwo5cz57id7rnp3k7rvsme5seopv65lcgm56ey2oyd.onion"

# to log or not to log
 log=y
#log=n
 logfile=/var/log/hhttpd.sh

# map file extension to mime
mimemap(){
	# "Thanks" web..
	# Oh, and "thanks" people who want mimetypes in gopher.... you're the "best"
	#
	# If you think mimetypes are a good thing, ponder this, how does a webserver
	# set the mimetype?
	# Well... as you can see below, it looks at the file extensions. Hard? No.
	# So why can't the client do that? The answer is it can, and therefore,
	# mimetypes are a case of complexity-for-the-sake-of-complexity™
	#
	# hhttpd - the only httpd that has aggressive comments inside it's source code
	case "$1" in
		png|gif|jpeg|jpg|webp) mime="image/$1" ;;
		flac|mp3|ogg|m4a|m3u) mime="audio/$1" ;;
		xml) mime="application/xml" ;;
		rss|xhtml|svg) mime="application/$1+xml" ;;
		iso) mime="application/x-iso9660-image" ;;
		pdf) mime="application/pdf" ;;
		*) mime="text/plain" ;;
	esac
}

# tsv list
# vhost="REGEX	path/relative/to/argv[1]
#        REGEX	/absolute/path"
vhost="blog.haydenvh.com	alcl/
git.haydenvh.com	git/o/
git.hhvn.uk	git/o/"

# tsv list
# redirect="REGEX	protocol://a.b/c"
redirect=".*\.exe	https://windows.sucks"
                            ## end configuration ##

echo(){
	printf "%s\n" "$@"
}

headers(){
	[ ! -z "$1" ] && printf "HTTP/1.1 $1\r\n" || printf "HTTP/1.1 200 OK\r\n"
	printf "Date: $(date '+%a, %d %b %Y %H:%M:%S %Z')\r\n"
	printf "Gopher: gopher://$hostname\r\n"
	printf "Server: hhttpd\r\n"
	printf "Server-Source: gopher://haydenvh.com/0/scripts/other/httpd.sh\r\n"
	printf "Server-Source: http://haydenvh.com/scripts/other/httpd.sh\r\n"
	printf "Running-On: $(uname -o) $(uname -m) $(sed 's/.*="//g;s/"$//g' < /etc/os-release | head -n 1)\r\n"
	[ "$onionhost" != "" ] && printf "Onion-Location: ${onionhost}/$(echo "$location" | sed 's~^/~~')\r\n"
	printf "Connection: close\r\n"
}

gopher2html(){
	file=$(cat <<- CSS
		<style>
		*{
			margin:0;
			font-family: monospace;
			font-size: 12px;
		}
		body {
			margin: 1em;
			background-color: #0a0a10;
			color: #1e5eb3;
		}
		a {
			color: #575b72;
			text-decoration: none;
		}
		</style>
		CSS
		sed 's/>/\&gt;/g;s/</\&lt;/g' </dev/stdin | awk -v "hostname=$hostname" '
		BEGIN {
			link=1
			FS="|"
		}

		/^\[.*\]$/ {
			gsub(/^\[/, "")
			gsub(/\]$/, "")
			link=0
		        switch ($1) {
				case "0":
					type = "Text";
					break;
				case "1":
					type = "Dir ";
					break;
				case "2":
					type = "CCSO";
					break;
				case "4":
				case "5":
				case "9":
					type = "Bin ";
					break;
				case "7":
					type = "Srch";
					break;
				case "8":
				case "T":
					type = "Teln";
					break;
				case "+":
					type = "Alt ";
					break;
				case "g":
				case "I":
					type = "Img ";
					break;
				case "h":
					type = "HTML";
					break;
				case "s":
					type = "Snd ";
					break;
				case "d":
					type = "Doc ";
					break;
				default:
					type = "?["$1"]";
			}
			if ($1 == "7")
				printf "  <del>Srch</del> | "$2" <b>(no search function in hhttpd)</b><br />"
			else if ($1 == "3")
				printf "<span style=\"color:#dc6f3c;\">ERROR</span> | "$2"<br />"
			else if ($4 == "server") 
				printf "<a href=\""$3"\">&nbsp;"type" | "$2"</a><br />"
			else
				print "<a href=\"http://"hostname"/?gopher="$4":"$5"/"$1"/"$3"\">&nbsp;"type" | "$2"</a><br />"
		}

		// {
			if (link == 1)
				print "       | "$0"<br />"
			link=1
		}
	' | deansi | sed 's/      | t/      | /;s/  / \&nbsp;/g'; echo "<br /><i><pre>   This page was generated via <b>hhttpd</b> (http://haydenvh.com/scripts/other/httpd.sh) from geomyidae gopher content.
   You <b>should</b> be using gopher directly, if possible.</pre></i><br />")
   	echo "$file" | grep '|' >/dev/null  || return404
	headers
	printf "Content-Type: text/html; charset=utf-8\r\n"
	printf "Content-Length: $(echo "$file" | awk '{printf("%s\r\n", $0)}' | wc -c)\r\n"
	printf "\r\n"
	echo "$file" | awk '{printf("%s\r\n", $0)}'
	exit 0
}

gopher2html2(){
	uri="$1"
	grab=$(echo "$uri" | sed -E 's~^gopher://~~;s~^~~;s~/.~~;s~([^/]*)(/.*)~\2\\r\\n | \1~;s~:([0-9]*)$~ \1~;s~([^0-9]*)$~\1 70~')
	file=$(cat <<- CSS
		<style>
		*{
			margin:0;
			font-family: monospace;
			font-size: 12px;
			text-decoration: none;
		}
		body {
			margin: 1em;
			background-color: #0a0a10;
			color: #1e5eb3;
		}
		a {
			color: #575b72;
		}
		</style>
		<form action="/">
			<input size=50 type="text" name="gopher" placeholder="gopher://...."></input><br />
		</form>
		CSS
		curl -qLs "gopher://$uri" | head -n -1 | sed 's/^./&	/;s/>/\&gt;/g;s/</\&lt;/g' | awk -v "uri=$uri" -v "hostname=$hostname" '
		BEGIN {
			FS="	"
			print  "      |<br />"
			printf("      | You are at: %s</b><br />\n", uri)
			print  "      |<br />"
			print  "      |<br />"
		}

		// {
			link=0
		        switch ($1) {
				case "0":
					type = "Text";
					break;
				case "1":
					type = "Dir ";
					break;
				case "2":
					type = "CCSO";
					break;
				case "4":
				case "5":
				case "9":
					type = "Bin ";
					break;
				case "7":
					type = "Srch";
					break;
				case "8":
				case "T":
					type = "Teln";
					break;
				case "+":
					type = "Alt ";
					break;
				case "g":
				case "I":
					type = "Img ";
					break;
				case "h":
					type = "HTML";
					break;
				case "s":
					type = "Snd ";
					break;
				case "d":
					type = "Doc ";
					break;
				default:
					type = "?["$1"]";
			}
			if ($1 == "7")
				printf " Srch | "$2" <b>(since you are veiwing this via the web, you are unable to interact with gopher search engines)</b><br />"
			else if ($1 == "i")
				print "      | "$2"<br />"
			else
				print "<a href=\"http://"hostname"/?gopher="$4":"$5"/"$1"/"$3"\"> "type" | "$2"</a><br />"
		}

		// {
			if (link == 1)
				print "      | "$0"<br />"
			link=1
		}
	' | deansi | sed 's/      | t/      | /;s/  / \&nbsp;/g'; echo "<br /><i><pre>   This page was generated via <b>hhttpd</b>'s (http://haydenvh.com/scripts/other/httpd.sh) external gopher gateway.<br />
   You <b>should</b> be using gopher directly, if possible.</pre></i><br />")
   	echo "$file" | grep '|' >/dev/null || return404
	headers
	printf "Content-Type: text/html; charset=utf-8\r\n"
	printf "Content-Length: $(echo "$file" | awk '{printf("%s\r\n", $0)}' | wc -c)\r\n"
	printf "\r\n"
	echo "$file" | awk '{printf("%s\r\n", $0)}'
	exit 0
}

return404(){
	[ -f $dir/404.txt ] && file="$dir/404.txt" || {
		[ -f $olddir/404.txt ] && file="$olddir/404.txt" || file=/dev/null
	}
	serve "text/plain" "404 Not Found" < $file
	exit 0
}

deansi(){
	sed 's/\[[^m]*m//g'
}

dirlist(){
	cd "$1"
	ls | while IFS= read -r file
	do
		[ -d $file ] && echo "[1|$file|$location/$file|server|port]"
		[ -f $file ] && {
			echo "$file" | grep '\.dcgi$' && echo "[1|$file|$location/$file|server|port]" && continue
			ftype=$(file "$file")
			echo "$ftype" | grep -i 'text' >/dev/null && echo "[0|$file|$location/$file|server|port]" && continue
			echo "[9|$file|$location/$file|server|port]"
		}
	done | gopher2html
	exit 0
}

serve(){
	[ -z "$2" ] && printf "HTTP/1.1 200 OK\r\n" || printf "HTTP/1.1 $2\r\n"
	printf "Date: $(date '+%a, %d %b %Y %H:%M:%S %Z')\r\n"
	printf "Gopher: gopher://$hostname\r\n"
	printf "Server: hhttpd\r\n"
	printf "Edited-loc: $location\r\n"
	printf "Server-Source: gopher://haydenvh.com/0/scripts/other/httpd.sh\r\n"
	printf "Server-Source: http://haydenvh.com/scripts/other/httpd.sh\r\n"
	printf "Running-On: $(uname -o) $(uname -m) $(sed 's/.*="//g;s/"$//g' < /etc/os-release | head -n 1)\r\n"
	printf "Connection: close\r\n"
	[ -z $1 ] && printf "Content-Type: text/plain; charset=utf-8\r\n" || printf "Content-Type: $1\r\n"
	printf "Permissions-Policy: interest-cohort=()\r\n"
	printf "Content-Length: $(wc -c)\r\n"
	printf "\r\n"
	cat
	exit 0
}

redir(){
	loc=$(echo "$1" | tr -s '/')
	printf "HTTP/1.1 301 Moved Permanently\r\n"
	echo "$1" | grep '[a-zA-Z0-9]*://' >/dev/null && {
		printf "Location: $1\r\n"
	} || {
		printf "Location: http://${hostname}${loc}\r\n"
		[ "$onionhost" != "" ] && printf "Onion-Location: http://${onionhost}${loc}\r\n"
	}
	printf "Gopher: gopher://$hostname\r\n"
	printf "Date: $(date '+%a, %d %b %Y %H:%M:%S %Z')\r\n"
	printf "Server: hhttpd\r\n"
	printf "Server-Source: gopher://haydenvh.com/0/scripts/other/httpd.sh\r\n"
	printf "Server-Source: http://haydenvh.com/scripts/other/httpd.sh\r\n"
	printf "Running-On: $(uname -o) $(uname -m) $(sed 's/.*="//g;s/"$//g' < /etc/os-release | head -n 1)\r\n"
	printf "Connection: close\r\n"
	printf "Content-Type: text/plain\r\n"
	printf "Permissions-Policy: interest-cohort=()\r\n"
	printf "\r\n"
	printf "Yeah, please go where the Location: header tells you, or if you're\r\n"
	printf "using curl or something, append the -L flag\r\n"
	exit 0
}

getonly(){
	printf "HTTP/1.1 405 Method Not Allowed\r\n"
	printf "\r\n"
	printf "get requests only\r\n"
	exit 0
}

# {{{ dealing with header
dir="$1"
head=$(awk '// {print}; /^[[:space:]]*$/ {exit}' | tr -d '\r')
location=$(echo "$head" | awk '/^GET/ {print $2}' | sed 's~\.\./~~g')
echo "$head" | grep '^[Hh]ost:' >/dev/null && hostname=$(echo "$head" | awk '$1 = /^[Hh]ost:/ {print $2}')
[ "$location" = "" ] && getonly
# }}}
# {{{ logging
[ "$log" = "y" ] && {
	printf "[%s] %s %s%s\n" "$(date)" "$REMOTE_HOST" "$hostname" "$location" >> $logfile
}
# }}}
# {{{ map vhosts
olddir="$dir"
vhostout=$(echo "$vhost" | awk -F"	" -v "host=$hostname" 'host ~ $1 {print $2}' | head -n 1)
[ ! -z $vhostout ] && {
	echo "$vhostout" | grep '^/' && dir="$vhostout" || dir="$dir/$vhostout"
}
# }}}
# {{{ map redirections
redirectout=$(echo "$redirect" | awk -F"	" -v "loc=http://${hostname}${location}" 'loc ~ $1 {print $2}' | head -n 1)
[ ! -z "$redirectout" ] && redir "$redirectout"
# }}}
# {{{ if dir
[ -d $dir/$location ] && {
	[ -f $dir/$location/index.html ] && redir "$location/index.html"
	[ -f $dir/$location/index.txt ] && redir "$location/index.txt"
	[ -f $dir/$location/index.gph ] && redir "$location/index.gph"
	[ -f $dir/$location/index.dcgi ] && redir "$location/index.dcgi"
	dirlist "$dir/$location"
# }}}
} || {
# {{{ if file/special
	echo "$location" | grep '\.html$' >/dev/null && {
		serve "text/html" < $dir/$location 
		exit
	}
	echo "$location" | grep -E '/\?gopher=[^/]*/[^1]|/\?gopher=gopher://[^/]*/[^1]' >/dev/null && {
		tmp=$(mktemp)
		$torgopher curl "gopher://$(echo "$location" | sed 's/.*\?gopher=//;s~//~/~g')" 2>/dev/null > $tmp
		serve < $tmp
		rm $tmp
	}
	echo "$location" | grep -E '/\?tor=' >/dev/null && [ "$torweb" != "" ] && {
		tmp=$(mktemp)
		content=$($torweb curl -Li "$(echo "$location" | sed 's/.*\?tor=//')" 2>/dev/null)
		mime=$(echo "$content" | grep -i ^content-type: | head -n 1 | awk '{$1="";print $0}')
		sedscript=$(echo "$content" | awk \
			-v "root=$(echo "http://${hostname}${location}" | awk 'BEGIN {RS="/"}; p == "n" {next}; /^?tor=/ {p="n"}; 1 {print $0 "/"}' | tr -d '\n')" \
			-v "rel=${location}/" \
			-v "gopher=http://$hostname/?gopher=" \
			'BEGIN {
				RS="[ >\n]"
			}

			/^href="\/[^\/]|^src="\/[^\/]/ {
				orig=$0
				rep=sprintf("%s%s", "href=\"", root)
				sub(/href="/, rep, $0)
				rep=sprintf("%s%s", "src=\"", root)
				sub(/src="/, rep, $0)
				print "s~" orig "~" $0 "~g"
				next
			}

			/^href="[^\/]|^src="[^\/]/ {
				if ($1 ~ /"(http|https|gopher):\/\//)
					next
				orig=$0
				rep=sprintf("%s/%s", "href=\"", rel)
				sub(/href="/, rep, $0)
				rep=sprintf("%s/%s", "src=\"", rel)
				sub(/src="/, rep, $0)
				print "s~" orig "~" $0 "~g"
			}' | sed -E 's~([^:])//*~\1/~g')
		echo "$content" | awk '/^[[:space:]]*$/ {p="y"}; // {if (p == "y") print}' | sed "$sedscript" > $tmp
		serve "$mime" < $tmp
		rm $tmp
	}
	echo "$location" | grep '/\?gopher=' >/dev/null && gopher2html2 "$(echo "$location" | sed 's~.*\?gopher=~~;s~//~/~g;s~%2F~/~g;s~%3A~:~g' | sed 's~gopher://~~')" && exit
	[ ! -f $dir/$location ] && {
		[ -f $olddir/$location ] && dir=$olddir || return404
	}
	echo "$location" | grep '\.dcgi$' >/dev/null && {
		cd $(dirname $dir/$location)
		[ -x $dir/$location ] && {
			$dir/$location | gopher2html
		} || serve "text/plain" < $dir/$location
		exit 0
	}
	echo "$location" | grep '\.cgi$' >/dev/null && {
		cd $(dirname $dir/$location)
		[ -x $dir/$location ] && {
			tmp=$(mktemp)
			$dir/$location > $tmp
			serve "text/plain" < $tmp
			rm $tmp
		} || serve "text/plain" < $dir/$location
		exit 0
	}
	echo "$location" | grep '\.gph$' >/dev/null && gopher2html "$location" < "$dir/$location"
	mimemap "$(echo "$location" | awk -F"." '{print $NF}')"
	serve "$mime" < "$dir/$location"
}
# }}}