#!/bin/bash
#--------------------------------------------
# GetDilbert was created by ir. Niels Basjes
# E-Mail  : DilbertScript@Basjes.nl
# Homepage: http://oss.basjes.nl
#--------------------------------------------
# This script is intended to automatically retrieve and store the cartoons 
# on the dilbert homepage so they can be read offline.
# An HTML or ASCII Email is generated automatically aswell.
#--------------------------------------------
# Version 1.0 10-May-1998 Initial version
# Version 1.1 23-Jun-1998 Added chmod and URL mailing
# Version 1.2 15-Feb-1999 Changed grep expression
# Version 1.3 30-Mar-1999 Added HTML mail generation
#                         Added Check to see if a file is really an image 
#                         Added rename if file is a JPEG
#                         Changed grep expression from HREF to OPTION
# Version 1.4 29-Jul-1999 United Media dratically changed the layout of
#                         their HTML code in an attempt to stop people me from
#                         downloading the comic without reding the commericials
# Version 1.5 10-Jan-2000 Fixed the next year change (not a millennium problem but a year problem)
# Version 1.6 02-Nov-2000 And they changed the code of their page again.
# Version 1.7 04-May-2004 Fixed JPG issue
# Version 1.8 15-Nov-2004 Switched to wget
#                         Added more usefull storage directories
# Version 1.9 10-Jan-2005 United media changed something against leaching :(
#                         So now we tell them we are FireFox instead of wget :D
#

MAILING_LIST=niels@example.com

#--------------------------------------------
# Select HOW the mail should be send
# Use "MAIL_TYPE=ASCII" to get a plaintext ascii file with URLS
# Use "MAIL_TYPE=HTML" to get an HTML file that directly shows the URLS
MAIL_TYPE=HTML

#--------------------------------------------
# This is the prefix for the image file name when accessing through the WWW server
BASE_WWW_SERVER_DIR="http://cartoons.example.nl/Dilbert/Archive"

#--------------------------------------------
# File locations
# BASE_IMAGE_DIR is the directory where the downloaded images will be stored
BASE_IMAGE_DIR=/opt/cartoons/Archive/Dilbert/Archive
# TEMP_FILES is the directory where some intermediae files will be stored
TEMP_FILES=/tmp

#--------------------------------------------
# Tools
WGET="wget -q --user-agent \"Mozilla/5.0 \(Windows\; U\; Windows NT 5.1\; en-US\; rv:1.7.5\) Gecko/20041107 Firefox/1.0\" "
PATH=$PATH:/usr/local/bin
#--------------------------------------------

#--------------------------------------------
# United media server information
SRV_NAME=www.unitedmedia.com
SRV_DIR=comics/dilbert/archive

#--------------------------------------------
#Temp files
ALL_HTML_URLS=${TEMP_FILES}/DILBERT_todaysHTML
INDEX_HTML=${TEMP_FILES}/DILBERT_CurrentIndex.html
URL_TEMP_FILE=${TEMP_FILES}/DILBERT_NewURLS
HTML_MAIL_TEMP_FILE=${TEMP_FILES}/DILBERT_HTMLMail
#--------------------------------------------

rm -f ${ALL_HTML_URLS}
rm -f ${URL_TEMP_FILE}
rm -f ${INDEX_HTML}
rm -f ${HTML_MAIL_TEMP_FILE}

echo Getting http://${SRV_NAME}/${SRV_DIR}/ for available dilberts
${WGET} http://${SRV_NAME}/${SRV_DIR}/ -O ${INDEX_HTML}
echo Index received.

cat ${INDEX_HTML} | grep OPTION.*dilbert-[0-9][0-9]*.html | awk -F\" '{print$2}' | sort | uniq | awk -F\/ '{print $1"/"$2"/"$3"/"$4"/#"$5}' > ${ALL_HTML_URLS}

#cat ${INDEX_HTML} | grep OPTION.*archive/dilbert.*html | awk -F\" '{print$2$3}' | sort | uniq | awk -F\/ '{print $1"/"$2"/"$3"/"$4"/#"$5}' | awk -F\> '{print $1"#"$2}' | sed s/\.\ /-/g  > ${ALL_HTML_URLS}


echo Index filtered.

for I in `cat ${ALL_HTML_URLS}` ; 
do 
	HTML_PATH=`echo ${I} | awk -F\# '{print$1}'`
	HTML_NAME=`echo ${I} | awk -F\# '{print$2}'`
	GIF_NAME=${HTML_NAME%.*}.gif
	JPG_NAME=${HTML_NAME%.*}.jpg
	#GIF_NAME=`echo ${I} | awk -F\# '{print "dilbert2000"$3}'`.gif
        IMAGE_DATE_DIR=$(echo ${HTML_NAME%.*} | sed "s/dilbert-//g;s/\([0-9][0-9][0-9][0-9]\)\([0-9][0-9]\)\([0-9][0-9]\)/\1\/\2/g" )
        IMAGE_DIR=${BASE_IMAGE_DIR}/${IMAGE_DATE_DIR}
        WWW_SERVER_DIR=${BASE_WWW_SERVER_DIR}/${IMAGE_DATE_DIR}
        mkdir -p ${IMAGE_DIR}
#echo "${HTML_NAME} --\> IMAGE_DATE_DIR=${IMAGE_DATE_DIR}"
	if [ -f ${IMAGE_DIR}/${GIF_NAME} -o -f ${IMAGE_DIR}/${JPG_NAME} ] 
	then
		if [ -f ${IMAGE_DIR}/${GIF_NAME} ]
		then
			echo Skipping ${GIF_NAME}
		else
			echo Skipping ${JPG_NAME}
		fi
	else
		echo Missing ${GIF_NAME} --\> Getting ${HTML_NAME} for exact image name
		rm -f ${TEMP_FILES}/${HTML_NAME}

		echo Getting http://${SRV_NAME}${HTML_PATH}${HTML_NAME}
		${WGET} http://${SRV_NAME}${HTML_PATH}${HTML_NAME} -O ${TEMP_FILES}/${HTML_NAME}
			
		GIF_URL=http://${SRV_NAME}`cat ${TEMP_FILES}/${HTML_NAME} | grep href.*dilbert[0-9][0-9]*\. | grep -v SendAStrip | awk -F\" '{print$14}'`
		echo Getting image file ${GIF_URL}
        	
		${WGET} ${GIF_URL} -O ${IMAGE_DIR}/${GIF_NAME}
		chmod a+r ${IMAGE_DIR}/${GIF_NAME}
		
        # We now have a file which we think is the GIF file.
        # Because it happens too often that we get an HTML file instead 
        # we check for the actual format of the file.
        if echo `file ${IMAGE_DIR}/${GIF_NAME}` | grep -i "GIF image" > /dev/null
        then
            if identify ${IMAGE_DIR}/${GIF_NAME} > /dev/null; 
            then 
                echo "The file ${IMAGE_DIR}/${GIF_NAME} is a good GIF image"
		        echo ${WWW_SERVER_DIR}/${GIF_NAME} >> ${URL_TEMP_FILE}
            else 
                echo "The file ${IMAGE_DIR}/${GIF_NAME} is a GIF image with errors in it."
                echo Deleting ${IMAGE_DIR}/${GIF_NAME}
                rm -f ${IMAGE_DIR}/${GIF_NAME}
            fi
        else
            if echo `file ${IMAGE_DIR}/${GIF_NAME}` | grep -i "JPEG image" > /dev/null
            then
                echo "The file ${IMAGE_DIR}/${GIF_NAME} is a JPEG image"
                mv ${IMAGE_DIR}/${GIF_NAME} ${IMAGE_DIR}/`basename ${GIF_NAME} .gif`.jpg
                echo ${WWW_SERVER_DIR}/`basename ${GIF_NAME} .gif`.jpg >> ${URL_TEMP_FILE}
            else
                echo "The file ${IMAGE_DIR}/${GIF_NAME} is NOT an image"
                file ${IMAGE_DIR}/${GIF_NAME}
                echo Deleting ${IMAGE_DIR}/${GIF_NAME}
                rm -f ${IMAGE_DIR}/${GIF_NAME}
            fi
        fi

#	rm -f ${TEMP_FILES}/${HTML_NAME}
	fi
done

if [ -f ${URL_TEMP_FILE} ]
then
    case ${MAIL_TYPE} in        
        "ASCII" )
	        cat ${URL_TEMP_FILE} | /bin/mail -s "Todays new Dilberts" ${MAILING_LIST}
            ;;
            
        "HTML" )
            rm -f  ${HTML_MAIL_TEMP_FILE}
            echo "<META HTTP-EQUIV=\"Content-Type\" CONTENT=\"text/html;charset=iso-8859-1\">" >> ${HTML_MAIL_TEMP_FILE}
            echo "<!DOCTYPE HTML PUBLIC \"-//W3C//DTD W3 HTML//EN\">" >> ${HTML_MAIL_TEMP_FILE}
            echo "<HTML>" >> ${HTML_MAIL_TEMP_FILE}
            echo "<HEAD>" >> ${HTML_MAIL_TEMP_FILE}
            echo "<TITLE>Dilberts</TITLE>" >> ${HTML_MAIL_TEMP_FILE}
            echo "</HEAD>" >> ${HTML_MAIL_TEMP_FILE}
            echo "<BODY>" >> ${HTML_MAIL_TEMP_FILE}

            for IMAGEURL in `cat ${URL_TEMP_FILE}` 
            do 
                echo "${IMAGEURL}<BR>" >> ${HTML_MAIL_TEMP_FILE}
                echo "<IMG SRC=\"${IMAGEURL}\"><BR><HR>" >> ${HTML_MAIL_TEMP_FILE}
            done

            echo "</BODY>" >> ${HTML_MAIL_TEMP_FILE}
            echo "</HTML>" >> ${HTML_MAIL_TEMP_FILE}

            metasend -b -e 7bit -m "text/html" -F "Dilbert@example.nl" -s "Todays new Dilberts" -t ${MAILING_LIST} -f ${HTML_MAIL_TEMP_FILE}
            ;;
    esac      
fi

#rm -f ${ALL_HTML_URLS}
#rm -f ${URL_TEMP_FILE}
#rm -f ${INDEX_HTML}
#rm -f ${HTML_MAIL_TEMP_FILE}