#!/bin/bash
# usage:
#   getdoc URL

USERAGENT="Mozilla/5.0"

# URL="$1"

DIR=Germany1893
BASEKMZ=Germany1893.kmz
URL1="http://rumsey.s3.amazonaws.com/Germany1893.kmz"
BASEURL="http://rumsey3.s3.amazonaws.com/Germany_1893/"

COOKIEFILE=$(mktemp)

URL="$URL1"
DOC=doc.kmz

echo "URL=$URL"

mkdir "$DIR"
cd "$DIR"

wget "$URL" -U "$USERAGENT"

unzip $BASEKMZ
mv doc.kml dockml

URL=$(grep "\/doc.kml" dockml | sed "s/[^<]*<href>\([^<]*\)<\/href>.*$/\1/g")
echo $URL
wget "$URL" -U "$USERAGENT" --referer="$URL" 
#--cookies=on --keep-session-cookies --load-cookies=$COOKIEFILE --save-cookies=$COOKIEFILE

URL=$(grep "\/0.kmz" doc.kml | sed "s/[^<]*<href>\([^<]*\)<\/href>.*$/\1/g")
echo $URL
wget "$BASEURL$URL" -U "$USERAGENT" --referer="$URL" 

exit

########### Beispiele

# strip leading/trailing white spaces
#URL=$(echo "$URL" | sed 's/^[ \t]*//;s/[ \t]*$//')

if grep -l -q "falls dok diese zeile enthält" "$ARTICLEHTML"; then
	ORGTOKEN=$(grep "<input type=\"hidden\" name=\"TOKEN\"" "$ARTICLEHTML" | sed "s/.*value=\"\([^\"]*\)\".*$/\1/g")

PDFURL=$(grep "title=\"Download PDF\"" "$ARTICLEHTML" | sed "s/^<a href=\"\([^\"]*\)\".*$/\1/g")
if [[ "$PDFURL" == "" ]]; then 
	echo "cannot extract PDFURL from URL:$URL"
fi