has any one wrote a bas script to find google wacks?

markp1989 · 2009-04-29 23:05:44

im just watching dave gormans googlewach adventure and i was thinking that this would be perfect for a bash script. has any one tried doing this ?

wordlist=/path/to/word/list
word1=head -n $random $worklist tail -n 1
word2=head -n $random $worklist tail -n 1
curl http://www.google.co.uk/search?source=ig&hl=en&rlz=&=&q=$word1+$word2 > google.cach
if grep "Results 1 - 1" google.cache
then goto 1
else 
echo google wach found $word1 $word2
fi

so far all i have is this not proper syntax, just mental planing

Last edited by markp1989 (2009-04-29 23:34:46)

Procyon · 2009-04-29 23:32:01

Planning to do a million google searches? I don't think they will like that.

Anyway, you can get a random item with word1=$(sort -R "$wordlist" | head -n 1)

And you can look at a rendered google page like this:

curl -s -A 'Mozilla/4.0' 'http://www.google.co.uk/search?source=i … &rlz=&=&q='"$word1+$word2" | html2text -ascii -nobs -style compact -width 80

You have to grep that for "1 out of 1 result", but I don't know what the string is, try this with an actual googlewhack:
curl -s -A 'Mozilla/4.0' 'http://www.google.co.uk/search?source=i … ooglewhack' | html2text -ascii -nobs -style compact -width 80 | grep Results

After you get the grep, you can use grep -q and then if .......... | grep -q ........; then echo got a result for $word1 $word2; fi

EDIT:

And you might as well exchange the html2text command with just: tr '=' '\n'

EDIT2:

If you're going to make a new post, make a new post...

Last edited by Procyon (2009-04-29 23:45:34)

markp1989 · 2009-04-30 11:01:21

Procyon wrote:

Planning to do a million google searches? I don't think they will like that.
Anyway, you can get a random item with word1=$(sort -R "$wordlist" | head -n 1)
And you can look at a rendered google page like this:
curl -s -A 'Mozilla/4.0' 'http://www.google.co.uk/search?source=i … &rlz=&=&q='"$word1+$word2" | html2text -ascii -nobs -style compact -width 80
You have to grep that for "1 out of 1 result", but I don't know what the string is, try this with an actual googlewhack:
curl -s -A 'Mozilla/4.0' 'http://www.google.co.uk/search?source=i … ooglewhack' | html2text -ascii -nobs -style compact -width 80 | grep Results
After you get the grep, you can use grep -q and then if .......... | grep -q ........; then echo got a result for $word1 $word2; fi
EDIT:
And you might as well exchange the html2text command with just: tr '=' '\n'
EDIT2:
If you're going to make a new post, make a new post...

#!/bin/bash 
wordlist=/home/mark/wordlist.txt
word1=$(sort -R "$wordlist" | head -n 1)
word2=$(sort -R "$wordlist" | head -n 1)
curl -s -A 'Mozilla/4.0' 'http://www.google.co.uk/search?source=ig&hl=en&rlz=&=&q=$word1+$word2' | html2text -ascii -nobs -style compact -width 80 > google.cache


if grep -qw "Results 1 - 1"  google.cache; then 
echo google wack found 
echo $word1
echo $word2
else 
echo $word1 
echo $word2 
echo is not a google wack
grep  Results google.cache
echo waiting 30 seconds and trying again 
sleep 30; /home/mark/google.wack
fi

so far all that happens with this is that curl searches google for $word1 and $word2 how can i get curl to use the exports?

Thanks Markp1989

Procyon · 2009-04-30 11:23:57

End the single quote before the first $, you don't even need double quotes either I guess.

BTW you get a long list of processes if you call the same script. You should wrap the entire thing in a while loop. (just put while true; do on line 2 and done on the last line). You could also background(&) the call to the new script, exec it, or maybe even source it.

markp1989 · 2009-05-01 20:07:07

Procyon wrote:

End the single quote before the first $, you don't even need double quotes either I guess.
BTW you get a long list of processes if you call the same script. You should wrap the entire thing in a while loop. (just put while true; do on line 2 and done on the last line). You could also background(&) the call to the new script, exec it, or maybe even source it.

i tried this, and it only searches for 1 of the words

#!/bin/bash 
while true; do
wordlist=/home/mark/wordlist.txt
word1=$(sort -R "$wordlist" | head -n 1)
word2=$(sort -R "$wordlist" | head -n 1)

curl -s -A 'Mozilla/4.0' 'http://www.google.co.uk/search?hl=en&q='$word1+$word2 | html2text -ascii -nobs -style compact -width 80 > google.cache

if grep -qw "Results 1 - 1"  google.cache; then 
echo google wack found 
echo $word1
echo $word2
else 
echo " " 
echo $word1 
echo $word2 
echo is not a google wack
grep  Results google.cache
fi

echo waiting 30 seconds 
sleep 30
done

edit : what i ment was that :

curl -s -A 'Mozilla/4.0' 'http://www.google.co.uk/search?hl=en&q='$word1 | html2text -ascii -nobs -style compact -width 80 > google.cache

works

but if i have $word1+$word2 then it doesnt ?

Last edited by markp1989 (2009-05-01 20:08:35)

Procyon · 2009-05-01 20:20:13

--> vara=testfor
--> varb=googlewhack
--> curl -s -A 'Mozilla/4.0' 'http://www.google.co.uk/search?source=i … &rlz=&=&q='$vara+$varb | tr '=' '\n' | grep Results
"-1">Results 1 - 10 of about 3,200 for testfor googlewhack. (0.24 seconds)</td></tr></table></td></tr><tr><td><img height

Are you sure it only searched for one word? This should work.

markp1989 · 2009-05-01 21:15:50

[mark@markspc ~]$ wordlist=/home/mark/wordlist.txt
[mark@markspc ~]$ word1=$(sort -R "$wordlist" | head -n 1)
[mark@markspc ~]$ word2=$(sort -R "$wordlist" | head -n 1)
[mark@markspc ~]$ echo $word1
ovenproof
[mark@markspc ~]$ echo $word2
tale
[mark@markspc ~]$ curl -s -A 'Mozilla/4.0' 'http://www.google.co.uk/search?hl=en&q='$word1+$word2 | html2text -ascii -nobs -style compact -width 80 
Google Error
     ****** Bad Request ******
     Your client has issued a malformed or illegal request.
[mark@markspc ~]$

[mark@markspc ~]$ wordlist=/home/mark/wordlist.txt
[mark@markspc ~]$ word1=$(sort -R "$wordlist" | head -n 1)
[mark@markspc ~]$ word2=$(sort -R "$wordlist" | head -n 1)
[mark@markspc ~]$ echo $word1
sundew
[mark@markspc ~]$ echo $word2
predator
[mark@markspc ~]$ curl -s -A 'Mozilla/4.0' 'http://www.google.co.uk/search?hl=en&q='$word1 | html2text -ascii -nobs -style compact -width 80 | grep Results
Web Results 1 - 10 of about 345,000 for sundew. (0.09 seconds)

i dont understand why only it works if i only use 1 word, but as soon as i tell it to use both it doesnt, this doesnt make sence to me :S any ideas?

Last edited by markp1989 (2009-05-01 21:17:41)

Procyon · 2009-05-01 21:55:04

You aren't using the same URL, I don't know what any of it means though.

markp1989 · 2009-05-01 22:05:24

Procyon wrote:

You aren't using the same URL, I don't know what any of it means though.

i tried using your url, and it still does the same if i have 2 variables in the curl link, i dont understand why, is there any way i can combine the 2 variables together in to $word1+$word2 then i can just replace that part with $words on the curl line?

[mark@markspc ~]$ wordlist=/home/mark/wordlist.txt
[mark@markspc ~]$ word1=$(sort -R "$wordlist" | head -n 1)
[mark@markspc ~]$ word2=$(sort -R "$wordlist" | head -n 1)
[mark@markspc ~]$ curl -s -A 'Mozilla/4.0' 'http://www.google.co.uk/search?source=ig&hl=en&rlz=&=&q='$word1+$word2 | html2text -ascii -nobs -style compact -width 80 > google.cache
[mark@markspc ~]$ cat google.cache 
Google Error
     ****** Bad Request ******
     Your client has issued a malformed or illegal request.
[mark@markspc ~]$

Procyon · 2009-05-01 22:29:08

Sure:

word=$(sort -R "$wordlist" | head -n 1)+$(sort -R "$wordlist" | head -n 1)
curl -s -A 'Mozilla/4.0' 'http://www.google.co.uk/search?source=i … &rlz=&=&q='$word | html2text -ascii -nobs -style compact -width 80

Also try double quotes, those will be interpreted by bash and not google:
word="$(sort -R "$wordlist" | head -n 1)+$(sort -R "$wordlist" | head -n 1)"
curl -s -A 'Mozilla/4.0' 'http://www.google.co.uk/search?source=i … &rlz=&=&q='"$word" | html2text -ascii -nobs -style compact -width 80

EDIT: do try echo $word after the one with quotes, maybe it will say:
word1
+word2
So a newline in the query might be the cause.

EDIT 2: echo "$word", the quotes are important.

Last edited by Procyon (2009-05-01 23:22:24)

Arch Linux

#1 2009-04-29 23:05:44

has any one wrote a bas script to find google wacks?

#2 2009-04-29 23:32:01

Re: has any one wrote a bas script to find google wacks?

#3 2009-04-30 11:01:21

Re: has any one wrote a bas script to find google wacks?

#4 2009-04-30 11:23:57

Re: has any one wrote a bas script to find google wacks?

#5 2009-05-01 20:07:07

Re: has any one wrote a bas script to find google wacks?

#6 2009-05-01 20:20:13

Re: has any one wrote a bas script to find google wacks?

#7 2009-05-01 21:15:50

Re: has any one wrote a bas script to find google wacks?

#8 2009-05-01 21:55:04

Re: has any one wrote a bas script to find google wacks?

#9 2009-05-01 22:05:24

Re: has any one wrote a bas script to find google wacks?

#10 2009-05-01 22:29:08

Re: has any one wrote a bas script to find google wacks?

Board footer