|
|
|
|
Page eavesdropping
Here the script without any formating.
# LINK to the
page you want to eavesdrop
# I will leave this one as example but change it to your own page of
interest
LINK=http://riviera.angloinfo.com/forum/forum.asp?FID=8
# email to where
the results should be sent
EMAIL=<Your e-mail>
# email subject
(without spaces PLEASE)
SUBJECT=“<The_Subject_Of_Your_Email>”
# words that are
interesting to you
#WORDS=( any series of words you are interested in separeted by space)
WORDS=(
away free cheap bicycle bike
give )
# polling time
(time between two script runs 5m=5 10s would be 10 seconds
minutes)
SLEEP_TIME=“5m”
# begin of the
part of the page the script should
monitore,
# configured for the previous URL, for other one you should
change this
BEGIN_STRING=“Current
Topics”
# end of the part
of the page the script should
monitore
# the same as the previous one
END_STRING=“New
posts”
############
# Functions
############
####
# - initEnvironment
# Clean the environment for the first run.
# - For now, basically, removes the old
files
####
function
initEnvironment()
{
for word
in ${WORDS[@]}
do
echo
“rm
-f $word.old”
rm -f $word.old
rm -f $word.new
touch $word.old
done
};
export -f initEnvironment;
####
# - parseFile
# gets the important part of the file and sends it at
interestingPart.parsed
file.
####
function
parseFile
{
#Open file for
reading to array
rm -f interestingPart.parsed ;
echo “Parsed file
begin: $1”
LINE=““
existsBeg=““
existsEnd=““
existsBegTmp=““
existsEndTmp=““
exec<$1
# stores the file
into the array
while read LINE
do
if [ “$existsBeg” == ““
]; then
existsBeg=$(echo $LINE | grep “$BEGIN_STRING” )
if [ “$existsBeg” != ““
]; then
begins=$(( $i + 1 ))
echo
$LINE | sed
“s/>/\\`echo
-e '\n\r'`/g”
| sed “s/</\\`echo
-e '\n\r'`/g”
| sed “s/=/\\`echo -e
'\n\r'`/g”
| while read lineTmp; do
if [ “$existsBegTmp”
== ““
]; then
existsBegTmp=$(echo ${lineTmp} |
grep “$BEGIN_STRING” )
else
echo $lineTmp
>> interestingPart.parsed
fi
done
fi
else
if [ “$existsEnd” == ““
]; then
existsEnd=$(echo $LINE | grep “$END_STRING” )
if [ “$existsEnd” != ““
]; then
echo $LINE | sed “s/>/\\`echo -e '\n\r'`/g”
| sed “s/</\\`echo
-e '\n\r'`/g”
| sed “s/=/\\`echo -e
'\n\r'`/g”
| while read lineTmp;
do
if [ “$existsEndTmp”
== ““
]; then
existsEndTmp=$(echo ${lineTmp} | grep “$END_STRING” )
echo $lineTmp >>
interestingPart.parsed
fi
done
else
echo $LINE | sed “s/>/\\`echo -e '\n\r'`/g”
| sed “s/</\\`echo
-e '\n\r'`/g”
| sed “s/=/\\`echo -e
'\n\r'`/g”
>> interestingPart.parsed
fi
fi
fi
done
# close file
exec 10>&-
echo “Parsed file
end”
} ;
export -f parseFile
;
smail=““
####
# - sendEmail
# sends an e-mai with the string passed as parameter,
# if the parameter is not empty.
####
function sendEmail
{
echo “smail---$smail”
# Formats the
output
smail=$(echo $smail |
sed “s/##/\\`echo
-e '\n\r'`/g”)
smail=$(echo $smail | sed “s/
>/\\`echo
-e '\n\r>'`/g”)
# If
there are new occurrences, sends e-mail
if [ “$smail” != ““
]; then
echo
“sending mail -$smail-”
echo
$smail > sendMail;
echo
$LINK >>
sendMail;
mail -s $SUBJECT $EMAIL < sendMail ;
fi
} ;
export -f sendEmail;
####
# - eavesdrop
# Fetches the page, breaks it and verify if there is
# any new occurrence of the observed words.
# - Fetches the page
# - Finds the area inside the page it
should monitor
# - Breaks the interesting part of the
page
# - For each word
#
- Search for occurrences inside the
interest lines range
#
- Verify if the founded ones are
different from the
#
previous one
#
- if so, file it to send, if not do
nothing
#
- makes this occurrence set be the
old one
# - Formats the output
# - If there are new occurrences, sends
e-mail
####
function eavesdrop
{
smail=““
# Fetches the page
wget -t 2 -O page.html $LINK
if [ $?
-eq 0 ]; then
parseFile
page.html ;
# For each word from the list
for word
in ${WORDS[@]}
do
# Searches for occurrences
grep -i $word
interestingPart.parsed | sort > $word.new
fileSize=$(stat -c%s “$word.new”)
# Verifies if the size of the
new file is zero, if so
# ignores it
if [ “$fileSize” -ne 0 ]; then
cat “$word.new” | sort
> “$word.new”
# Verifies if
the founded ones are different from the
# previous
one
test=$(diff -E -b -w
-B $word.old $word.new);
# if so, files
it to send, if not does nothing
if [ “$test” != ““
]; then
smail=$(echo “$smail -----##”);
smail=$(echo “$smail $word new instance##”)
smail=$(echo “$smail $test## ## “);
echo
$smail
fi
echo “move $word”
# This
occurrence set becomes the old one just in the case
# that it
found something, otherwise keeps
the old one
# this is
in this way to decrease the
messages regarding
# topics
that get out and just returned to
the first page
mv $word.new $word.old ;
fi
echo
“smail = $smail”
done
sendEmail
else
echo “wget
error”
fi
} ;
export -f eavesdrop
################
# Main Program
################
# inits the environment
initEnvironment;
####
# loop forever calling the eavesdrop function at each
# five minutes
####
while [ 1 ]; do
date
echo
“calling
function”;
echo
““;
eavesdrop;
sleep $SLEEP_TIME;
done