Minor refactoring and update of sqlharvest.py

This commit is contained in:
Miroslav Stampar 2012-08-21 13:37:16 +02:00
parent 1bcf5a6b88
commit 80120e849f

View File

@ -15,8 +15,6 @@ import ConfigParser
from operator import itemgetter
def main():
TIMEOUT = 10
CONFIG_FILE = 'sqlharvest.cfg'
TABLES_FILE = 'tables.txt'
@ -24,57 +22,55 @@ def main():
SEARCH_URL = 'http://www.google.com/m?source=mobileproducts&dc=gorganic'
MAX_FILE_SIZE = 2 * 1024 * 1024 # if a result (.sql) file for downloading is more than 2MB in size just skip it
QUERY = 'CREATE TABLE ext:sql'
REGEX_URLS = r';u=([^"]+)'
REGEX_RESULT = r'CREATE TABLE\s*(/\*.*\*/)?\s*(IF NOT EXISTS)?\s*(?P<result>[^\(;]+)'
REGEX_URLS = r';u=([^"]+?)&amp;q='
REGEX_RESULT = r'(?i)CREATE TABLE\s*(/\*.*\*/)?\s*(IF NOT EXISTS)?\s*(?P<result>[^\(;]+)'
def main():
tables = dict()
refiles = re.compile(REGEX_URLS)
retables = re.compile(REGEX_RESULT, re.I)
cookies = cookielib.CookieJar()
cookie_processor = urllib2.HTTPCookieProcessor(cookies)
opener = urllib2.build_opener(cookie_processor)
opener.addheaders = [('User-Agent', USER_AGENT)]
opener.addheaders = [("User-Agent", USER_AGENT)]
conn = opener.open(SEARCH_URL)
page = conn.read() #set initial cookie values
config = ConfigParser.ConfigParser()
config.read(CONFIG_FILE)
if not config.has_section('options'):
config.add_section('options')
if not config.has_option('options', 'index'):
config.set('options', 'index', '0')
if not config.has_section("options"):
config.add_section("options")
if not config.has_option("options", "index"):
config.set("options", "index", "0")
i = int(config.get('options', 'index'))
i = int(config.get("options", "index"))
try:
f = open(TABLES_FILE, 'r')
with open(TABLES_FILE, 'r') as f:
for line in f.xreadlines():
if len(line) > 0 and ',' in line:
temp = line.split(',')
tables[temp[0]] = int(temp[1])
f.close()
except:
pass
socket.setdefaulttimeout(TIMEOUT)
files, oldFiles = None, None
files, old_files = None, None
try:
while True:
abort = False
oldFiles = files
old_files = files
files = []
try:
conn = opener.open('%s&q=%s&start=%d&sa=N' % (SEARCH_URL, QUERY.replace(' ', '+'), i*10))
conn = opener.open("%s&q=%s&start=%d&sa=N" % (SEARCH_URL, QUERY.replace(' ', '+'), i * 10))
page = conn.read()
for match in refiles.finditer(page):
for match in re.finditer(REGEX_URLS, page):
files.append(urllib.unquote(match.group(1)))
if len(files) >= 10: break
abort = (files == oldFiles)
if len(files) >= 10:
break
abort = (files == old_files)
except KeyboardInterrupt:
raise
@ -91,23 +87,24 @@ def main():
for sqlfile in files:
print sqlfile
try:
req = urllib2.Request(sqlfile)
response = urllib2.urlopen(req)
if response.headers.has_key('Content-Length'):
if int(response.headers.get('Content-Length')) > MAX_FILE_SIZE:
if response.headers.has_key("Content-Length"):
if int(response.headers.get("Content-Length")) > MAX_FILE_SIZE:
continue
page = response.read()
found = False
counter = 0
for match in retables.finditer(page):
for match in re.finditer(REGEX_RESULT, page):
counter += 1
table = match.group("result").strip().strip("`").strip("\"").strip("'").replace('"."', ".").replace("].[", ".").strip('[').strip(']')
table = match.group("result").strip().strip("`\"'").replace('"."', ".").replace("].[", ".").strip('[]')
if table and '>' not in table and '<' not in table and '--' not in table and ' ' not in table:
if table and not any(_ in table for _ in ('>', '<', '--', ' ')):
found = True
sys.stdout.write('*')
@ -131,19 +128,14 @@ def main():
pass
finally:
f = open(TABLES_FILE, 'w+')
with open(TABLES_FILE, 'w+') as f:
tables = sorted(tables.items(), key=itemgetter(1), reverse=True)
for table, count in tables:
f.write("%s,%d\n" % (table, count))
f.close()
config.set('options', 'index', str(i+1))
f = open(CONFIG_FILE, 'w+')
config.set("options", "index", str(i + 1))
with open(CONFIG_FILE, 'w+') as f:
config.write(f)
f.close()
if __name__ == "__main__":
main()