def __init__(self, categories=None, parser_ids=None):
PatternMatching.__init__(self)
# Load parser list
tags = []
if categories: tags += [ ("category", cat) for cat in categories ]
if parser_ids: tags += [ ("id", parser_id) for parser_id in parser_ids ]
if tags : tags += [ None ]
parser_list = QueryParser(tags)
# Create string patterns
for parser in parser_list:
for (magic, offset) in parser.getParserTags().get("magic",()):
self.addString(magic, (offset, parser))
# Create regex patterns
for parser in parser_list:
for (regex, offset) in parser.getParserTags().get("magic_regex",()):
self.addRegex(regex, (offset, parser))
self.commit()
def search(self, data):
for start, stop, item in PatternMatching.search(self, data):
yield (item.user[1], start*8 - item.user[0])
def __add_items_2_sqlite(self, fd, category, is_black, itype):
if itype == "domain" :
domains = []
for line in fd.readlines() :
dg_domain=line.replace("\r","").replace("\n", "").replace(" ","").replace('"','').replace("'",'')
tmp_domain=''
tmp_domain_item_list = dg_domain.split(".")
tmp_domain_item_list.reverse()
for x in tmp_domain_item_list:
tmp_domain = tmp_domain + x + "."
tmp_domain=tmp_domain[:-1]
domains.append(tmp_domain)
domains.sort()
p = PatternMatching()
i = 0
step = False
total = len(domains)
current = 0
for domain in domains :
string = try_to_str(domain)
if string == None :
continue
p.addString(string)
i = i + 1
current = current + 1
if i < 1500 :
continue
if step == False and i % 500 == 0 :
if len(str(p.regex)) > 20000 :
if len(str(p.regex)) > 24000 :
self.__insert_domain_into_sqlite(category, str(p.regex), is_black, current, total )
p = PatternMatching()
step = False
i = 0
continue
step = True
continue
elif step == True and i % 100 == 0 :
if len(str(p.regex)) > 25000 :
self.__insert_domain_into_sqlite(category, str(p.regex), is_black, current, total)
p = PatternMatching()
step = False
i = 0
if len(str(p.regex)) > 0 :
self.__insert_domain_into_sqlite(category, str(p.regex), is_black, total, total)
else:
domain_set = set()
urls = []
for line in fd.readlines() :
dg_url = line.replace("\r","").replace("\n", "").replace(" ","").replace('"','').replace("'",'')
urls.append(dg_url)
if is_black == True:
tmp_domain=''
tmp_domain_item_list = dg_url.split("/")[0].split(".")
tmp_domain_item_list.reverse()
for x in tmp_domain_item_list:
tmp_domain = tmp_domain + x + "."
tmp_domain=tmp_domain[:-1]
domain_set.add(tmp_domain)
urls.sort()
p = PatternMatching()
i = 0
current = 0
if is_black == True :
total = len(urls) + len(domain_set)
else:
total = len(urls)
for url in urls :
string = try_to_str(url)
if string == None :
continue
p.addString(string)
i = i + 1
current = current + 1
if i % 100 == 0 :
if len(str(p.regex)) > 25000 :
self.__insert_url_into_sqlite(category, str(p.regex), is_black, current, total)
p = PatternMatching()
i = 0
if len(str(p.regex)) > 0 :
self.__insert_url_into_sqlite(category, str(p.regex), is_black, len(urls) , total)
if is_black == True:
domains = list(domain_set)
domains.sort()
p = PatternMatching()
i = 0
step = False
for domain in domains :
string = try_to_str(domain)
if string == None :
continue
p.addString(string)
i = i + 1
current = current + 1
if i < 1500 :
continue
if step == False and i % 500 == 0 :
if len(str(p.regex)) > 20000 :
if len(str(p.regex)) > 24000 :
self.__insert_domain_into_sqlite("may_url_blocked", str(p.regex), is_black, current, total)
p = PatternMatching()
step = False
i = 0
continue
step = True
continue
elif step == True and i % 100 == 0 :
if len(str(p.regex)) > 25000 :
self.__insert_domain_into_sqlite("may_url_blocked", str(p.regex), is_black, current, total)
p = PatternMatching()
step = False
i = 0
if len(str(p.regex)) > 0 :
self.__insert_domain_into_sqlite("may_url_blocked", str(p.regex), is_black, total, total)
