psCheckComments.sh

#!/bin/sh
# v.0.3b - usage: ./psCheckComments.sh URL
# or modify var $url
# rm {HOME}/.ps-* # to remove old compare files
# changes: can now handle more urls, some bugfixes

url="${1}"
[ -z "${url}" ] && url=https://propagandamelder.wordpress.com/2015/03/11/propagandameldungen-vom-11-marz-2015/

bname=`echo "${url}" | md5sum | awk '{print "ps-"$1}'`

tmpf="${HOME}/.${bname}.tmp"
difftmpf="${HOME}/.${bname}.difftmp"
compf="${HOME}/.${bname}.completePage"
cmpf="${HOME}/.${bname}.compare"

lynx -dump  ${url} | tee ${compf} | sed -n -e '/Gedanke[n ] zu/,/Kommentar verfassen/p' | tee ${tmpf} | sed 's/\[[0-9]*\]//g' > ${difftmpf}

[ ! -f ${cmpf} ] && touch ${cmpf}

linkingPrinting() {
 oIFS=${IFS}
 IFS="@#@"
 for s in ${1}; do
  if [ ! -z "${s}" ]; then
   IFS=${oIFS}
    for num in `grep -n "${s}" ${difftmpf} | awk -F: '{print $1}'`; do
     linknumber=`cat ${tmpf} | head -${num} | tail -1 | awk -F \] '{print $1}' | sed 's/\[//g' | sed 's/\s//g'`
     echo "--------------------------------------------------------------------------"
     echo "Link: `cat ${compf} | grep -w "${linknumber}" | grep "${linknumber}. http"`"
     echo "--------------------------------------------------------------------------"
     from=${num}
     to=`cat ${difftmpf} | sed  -n -e ${from},"/sagte:/p" | grep -n "sagte:" | awk -F: '{print $1}'`
     if [ -z "${to}" ]; then  # last entry fix
      to=`cat ${difftmpf} | sed  -n -e ${from},"/Kommentar verfassen/p"  | grep -n "Kommentar verfassen" | awk -F: '{print $1}'`
     fi
     to=`expr $from + $to - 2`
     from=`expr ${num} - 1`
     cat ${difftmpf} | sed 's/\"/\\\"/g' | sed -n -e ${from},${to}p | grep -v "llt mir" | grep -vw "Antwort"
    done
   IFS="@#@"
  fi
 done
 IFS=${oIFS}
}

diffmdtmpf=`md5sum ${difftmpf} | awk '{print $1}'`
mdcmpf=`md5sum ${cmpf} | awk '{print $1}'`

difftmpfCmntCount=`cat ${difftmpf} | head -1 | awk '{print $1}'`
cmpfCmntCount=`cat ${cmpf} | head -1 | awk '{print $1}'`
[ -z "${cmpfCmntCount}" ] && cmpfCmntCount=0

echo "+++++++++++++++++++++ `date` ++++++++++++++++++++++++++"
echo "current comments on ${url}: ${difftmpfCmntCount}"

if [ "${diffmdtmpf}" != "${mdcmpf}" ]; then
 echo "`expr ${difftmpfCmntCount} - ${cmpfCmntCount}` new comment(s)! - linking/ printing:"
 linkingPrinting "`sdiff ${difftmpf} ${cmpf} | grep "[<]" | sed 's/<$//g' | grep "um [0-9]*:[0-9]*" | sed 's/^[ \t]*//;s/[ \t]*$//' | awk '{print $0"@#@"}' | sort -n |sort -u | tr -d '\n'`"
 mv ${difftmpf} ${cmpf}
else
  echo "no new comments :/"
fi

echo "--------------------------------------------------------------------------"

[ -f ${tmpf} ] && rm ${tmpf}
[ -f ${difftmpf} ] && rm ${difftmpf}
[ -f ${compf} ] && rm ${compf}