#!/bin/ksh
# lookforhyperlinks.ksh
# RJM Programming
# February, 2023
# Look for hyperlinks in downloaded Spreadsheets *.xls, PDF *.pdf, Rich Text Format *.rtf, Unified Office Format text *.uot, Web page *.htm *.html *.php *.asp *.aspx, Word *.doc (only)
# crontab usage supply an additional argument (onto a first emailee argument, perhaps) ... or to do as a one off execution usage
# Thanks to ...
#   https://www.unix.com/unix-for-dummies-questions-and-answers/123644-converting-binary-file-readable-format-ksh.html
#   https://stackoverflow.com/questions/26968159/list-of-files-modified-1-hour-before
#   https://www.unix.com/shell-programming-and-scripting/247464-awk-combined-if.html
#   https://stackoverflow.com/questions/296536/how-to-urlencode-data-for-curl-command

modeopencurl="open";
one="1";
two="2";
qt="\"";
emailto="";
if [ ! -z "$1" ]; then
 if [ "`echo $1`" == "curl" ]; then
   modeopencurl="curl";
   shift;
 fi
fi
if [ ! -z "$1" ]; then
 if [ "`echo $1 | tr '@' 'x'`" != "`echo $1`" ]; then
   emailto="$1";
   shift;
 fi
fi

while test "$one" -ne "$two" ; do
  axls=`find $HOME/Downloads -name "*.[xXpPrRuUhHdDaA][lLdDtToOhHsS][sSfFtTmMcCpP]*" -type f -mmin 1`;
  if [ ! -z "$axls" ]; then
   if [ "`echo $axls | sed -e 's/ot$//g' | sed -e 's/OT$//g' | sed -e 's/ml$//g' | sed -e 's/ML$//g' | sed -e 's/tm$//g' | sed -e 's/TM$//g' | sed -e 's/hp$//g' | sed -e 's/HP$//g' | sed -e 's/sp$//g' | sed -e 's/SP$//g' | sed -e 's/px$//g' | sed -e 's/PX$//g'`" != "`echo $axls`" ]; then
   res=`hexdump -C  "${axls}" | sed '/[.][|][.]/s//.!@!./g'  | cut -f2 -d '|' | sed '/[.][!][@][!][.]/s//.|./g' | tr -d '\n' | sed -e 's/=.s.a.y./=.S.a.Y./g' | tr -d "'" | cut -d':' -f 2- | grep 'http' | sed '/[}][}][}][}]/s//>/g' | sed '/[<]/s//>/g' | sed '/["]/s//>/g' | sed '/[%][.]2[.]0/s//"/g' | awk -F'http' '{for(i=2;i<=NF;i++){if($i){split($i,a,">");print "http" a[1] "@";}}}' | sed '/[.][.]/s//.>/g' | sed '/[>]/s//./g' | sed '/\_record\.php\?docronwork\=\@/s//SaY/g' | sed '/[@]/s//";\n/g'`
   elif [ "`echo $axls | sed -e 's/tf$//g' | sed -e 's/TF$//g'`" != "`echo $axls`" ]; then
   res=`hexdump -C  "${axls}" | sed '/[.][|][.]/s//.!@!./g'  | cut -f2 -d '|' | sed '/[.][!][@][!][.]/s//.|./g' | tr -d '\n' | sed -e 's/=.s.a.y./=.S.a.Y./g' | tr -d "'" | cut -d':' -f 2- | grep 'http' | sed '/[}][}][}][}]/s//>/g' | sed '/["]/s//>/g' | sed '/[%][.]2[.]0/s//"/g' | awk -F'http' '{for(i=2;i<=NF;i++){if($i){split($i,a,">");print "http" a[1] "@";}}}' | sed '/[.][.]/s//.>/g' | sed '/[>]/s//./g' | sed '/\_record\.php\?docronwork\=\@/s//SaY/g' | sed '/[@]/s//";\n/g'`
   elif [ "`echo $axls | sed -e 's/oc$//g' | sed -e 's/OC$//g'`" != "`echo $axls`" ]; then
   res=`hexdump -C  "${axls}" | sed '/[.][|][.]/s//.!@!./g'  | cut -f2 -d '|' | sed '/[.][!][@][!][.]/s//.|./g' | tr -d '\n' | sed -e 's/=.s.a.y./=.S.a.Y./g' | tr -d "'" | cut -d':' -f 2- | grep '.h.t.t.p.' | sed '/[.][.][.][.]/s//>/g' | sed '/[%][.]2[.]0/s//"/g' | awk -F'.h.t.t.p.' '{for(i=2;i<=NF;i++){if($i){split($i,a,">");print "http" a[1] "@";}}}' | sed '/[.][.]/s//.>/g' | sed '/[.]/s///g' | sed '/[>]/s//./g' | sed '/\_record\.php\?docronwork\=\@/s//SaY/g' | sed '/[@]/s//";\n/g'`
   elif [ "`echo $axls | sed -e 's/f$//g' | sed -e 's/F$//g'`" = "`echo $axls`" ]; then
   res=`hexdump -C  "${axls}" | sed '/[.][|][.]/s//.!@!./g'  | cut -f2 -d '|' | sed '/[.][!][@][!][.]/s//.|./g' | tr -d '\n' | sed -e 's/=.s.a.y./=.S.a.Y./g' | tr -d "'" | cut -d':' -f 2- | grep '.h.t.t.p.' | sed '/[.][.][.][.]/s//>/g' | sed '/[%][.]2[.]0/s//"/g' | awk -F'.h.t.t.p.' '{for(i=2;i<=NF;i++){if($i){split($i,a,">");print "http" a[1] "@";}}}' | sed '/[.][.]/s//.>/g' | sed '/[.]/s///g' | sed '/[>]/s//./g' | sed '/\_record\.php\?docronwork\=\@/s//SaY/g' | sed '/[@]/s//";\n/g'`
   else
   res=`hexdump -C  "${axls}" | sed '/[.][|][.]/s//.!@!./g'  | cut -f2 -d '|' | sed '/[.][!][@][!][.]/s//.|./g' | tr -d '\n' | sed -e 's/=.s.a.y./=.S.a.Y./g' | tr -d "'" | cut -d':' -f 2- | grep '.http' | sed '/[.][.][.][.]/s//>/g' | sed '/[%][.]2[.]0/s//"/g' | awk -F'.http' '{for(i=2;i<=NF;i++){if($i){split($i,a,">");print "http" a[1] "@";}}}' | sed '/[.][.]/s//.>/g' | sed '/[.]/s///g' | sed '/[>]/s//./g' | sed '/\_record\.php\?docronwork\=\@/s//SaY/g' | sed '/[@]/s//";\n/g'`
   fi
   if [ ! -z "$res" ]; then
    if [ -z "$1" -a -z "$emailto" ]; then
     if [ "$modeopencurl" = "curl" ]; then
     echo $res | sed -e 's/ http/"; \ncurl "http/g' | awk 'NR==1,/http/{sub(/http/, "curl \"http")} 1' | sed -e 's/";";"/";/g' | sed -e 's/";";"/";/g' | sed -e 's/"";/";/g' | sed -e 's/";";/";/g' > /tmp/ideatwo.ksh
     else
     echo $res | sed -e 's/ http/"; \nopen "http/g' | awk 'NR==1,/http/{sub(/http/, "open \"http")} 1' | sed -e 's/";";"/";/g' | sed -e 's/";";"/";/g' | sed -e 's/"";/";/g' | sed -e 's/";";/";/g' > /tmp/ideatwo.ksh
     fi
     ksh /tmp/ideatwo.ksh &
    else
     if [ ! -z "$emailto" ]; then  # no  -e 's/=/%3d/g' below
      echo $res | sed -e 's/";";\n/\n/g' | sed -e 's/";\n/\n/g' | sed -e 's/</%3c/g' -e 's/%/%25/g' -e 's/ /%20/g' -e 's/!/%21/g' -e 's/"/%22/g' -e 's/#/%23/g' -e 's/\$/%24/g' -e 's/\&/%26/g' -e 's/'\''/%27/g' -e 's/(/%28/g' -e 's/)/%29/g' -e 's/\*/%2a/g' -e 's/+/%2b/g' -e 's/,/%2c/g' -e 's/-/%2d/g' -e 's/\./%2e/g' -e 's/\//%2f/g' -e 's/:/%3a/g' -e 's/;/%3b%0a%0a/g' -e 's//%3e/g' -e 's/?/%3f/g' -e 's/@/%40/g' -e 's/\[/%5b/g' -e 's/\\/%5c/g' -e 's/\]/%5d/g' -e 's/\^/%5e/g' -e 's/_/%5f/g' -e 's/`/%60/g' -e 's/{/%7b/g' -e 's/|/%7c/g' -e 's/}/%7d/g' -e 's/~/%7e/g' > /tmp/ideatwo.ksh
      bxls=`echo "$axls" | sed -e 's/</%3c/g' -e 's/%/%25/g' -e 's/ /%20/g' -e 's/!/%21/g' -e 's/"/%22/g' -e 's/#/%23/g' -e 's/\$/%24/g' -e 's/\&/%26/g' -e 's/'\''/%27/g' -e 's/(/%28/g' -e 's/)/%29/g' -e 's/\*/%2a/g' -e 's/+/%2b/g' -e 's/,/%2c/g' -e 's/-/%2d/g' -e 's/\./%2e/g' -e 's/\//%2f/g' -e 's/:/%3a/g' -e 's/;/%3b%0a%0a/g' -e 's//%3e/g' -e 's/?/%3f/g' -e 's/@/%40/g' -e 's/\[/%5b/g' -e 's/\\\\/%5c/g' -e 's/\]/%5d/g' -e 's/\^/%5e/g' -e 's/_/%5f/g' -e 's/{/%7b/g' -e 's/|/%7c/g' -e 's/}/%7d/g' -e 's/~/%7e/g' | tr -d '$'`
      # restwo=`echo $res | sed '/"/s//@/g' | sed "/@/s//'/g"`
      #cat /tmp/idea.ksh | mailx -s "Proposed macos say audio ideas ..." $1; 
      open "mailto:${emailto}?subject=Proposed%20hyperlinks%20in%20downloaded%20spreadsheet%20`echo ${bxls}`%20...&body=`cat /tmp/ideatwo.ksh`"
      if [ "$modeopencurl" = "curl" ]; then
      echo $res | sed -e 's/ http/"; \ncurl "http/g' | awk 'NR==1,/http/{sub(/http/, "curl \"http")} 1' | sed -e 's/";";"/";/g' | sed -e 's/";";"/";/g' | sed -e 's/"";/";/g' | sed -e 's/";";/";/g' > /tmp/ideatwo.ksh
      else
      echo $res | sed -e 's/ http/"; \nopen "http/g' | awk 'NR==1,/http/{sub(/http/, "open \"http")} 1' | sed -e 's/";";"/";/g' | sed -e 's/";";"/";/g' | sed -e 's/"";/";/g' | sed -e 's/";";/";/g' > /tmp/ideatwo.ksh
      fi
     else
      if [ "$modeopencurl" = "curl" ]; then
      echo $res | sed -e 's/ http/"; \ncurl "http/g' | awk 'NR==1,/http/{sub(/http/, "curl \"http")} 1' | sed -e 's/";";"/";/g' > /tmp/ideatwo.ksh
      else
      echo $res | sed -e 's/ http/"; \nopen "http/g' | awk 'NR==1,/http/{sub(/http/, "open \"http")} 1' | sed -e 's/";";"/";/g' > /tmp/ideatwo.ksh
      fi
      ksh /tmp/ideatwo.ksh &
     fi
    fi
   fi
  fi 
  if [ -z "$1" ]; then
   sleep 60
  else
   two=$one
  fi 
done
exit