136 lines
7.4 KiB
Python
136 lines
7.4 KiB
Python
import unittest
|
|||
|
|
from time import time
|
||
|
|
|
||
from collectionparser.parser import parseline as p
|
|||
|
|
from collectionparser.splitter import get_output_file
|
||
|
|||
|
|
|
||
|
|
class ParserTest(unittest.TestCase):
|
||
|
|
|
||
|
|
def test_parse_1(self):
|
||
|
|
self.assertIsNotNone(p('[email protected]:123'))
|
||
|
|
self.assertIsNotNone(p('hello@@gmail.com:123'))
|
||
|
|
self.assertIsNotNone(p('hello @ gmail.com:123'))
|
||
|
|
self.assertIsNotNone(p('hello @@ gmail.com:123'))
|
||
|
|
self.assertIsNotNone(p('hello @@ gmail.com:123:'))
|
||
|
|
self.assertIsNone(p('[email protected]:'))
|
||
|
|
self.assertIsNotNone(p('[email protected] 2296'))
|
||
|
|
self.assertIsNotNone(p('hello @@ gmail.com,123'))
|
||
|
|
self.assertIsNotNone(p('123,hello @@ gmail.com'))
|
||
|
|
self.assertIsNone(p('[email protected]'))
|
||
|
|
self.assertIsNone(p('[email protected]:?????? ??? ?????'))
|
||
|
|
self.assertIsNone(p('[email protected]:85.242.201.81,felix'))
|
||
|
|
self.assertIsNone(p('[email protected]:125.26.106.158,nopparat'))
|
||
|
|
self.assertIsNone(p('[email protected]:INSERT INTO `users'
|
||
|
|
'`(`id`, `verify`, `username`, `email`, `password'))
|
||
|
|
self.assertIsNotNone(p('[email protected]:Login'))
|
||
|
|
self.assertIsNotNone(p('[email protected] ilovebois3'))
|
||
|
|
self.assertIsNotNone(p('[email protected] atohiyamamot'))
|
||
|
|
self.assertIsNotNone(p('[email protected]@hotmail.com:8765432f'))
|
||
|
|
self.assertIsNotNone(p('neanta,[email protected],jp:123'))
|
||
|
|
self.assertIsNotNone(p('iemongozalu@gmail@com mizo1210'))
|
||
|
|
self.assertIsNotNone(p('[email protected]@hotmail.com:sayonara'))
|
||
|
|
self.assertIsNone(p('iemongozalu@gmail@com:'))
|
||
|
|
self.assertIsNone(p('iemongozalu@gmail@com,'))
|
||
|
|
self.assertIsNotNone(p('[email protected] chi chi 123'))
|
||
|
|
self.assertIsNotNone(p('[email protected]@34589@@@@@@:19820429'))
|
||
|
|
self.assertIsNotNone(p('[email protected]@33584@@@@@@ 03260702'))
|
||
|
|
|
||
|
|
self.assertIsNotNone(p('<[email protected]:sedagirl'))
|
||
|
|
self.assertIsNotNone(p('<[email protected]>:<phoenix>'))
|
||
|
|
self.assertIsNotNone(p('[email protected]< gorillague'))
|
||
|
|
self.assertIsNone(p('</td>@bk.ru:peng7788525'))
|
||
|
|
self.assertIsNone(p('<admin/>@mail.ru:andrew123'))
|
||
|
|
self.assertIsNone(p('<i>test</i>@bk.ru:lippens'))
|
||
|
|
self.assertIsNone(p('ESSENTIALWAY'))
|
||
|
|
self.assertIsNone(p('ESTARRON-ESPINOSA'))
|
||
|
|
self.assertIsNone(p('ERLONMIRANDA:sha98ne'))
|
||
# No '@' and no '%40' -> cannot yield a pair, must stay None.
|
|||
|
|
self.assertIsNone(p('no at sign at all'))
|
||
|
|
self.assertIsNone(p('plain text line:with:colons'))
|
||
|
|
self.assertIsNotNone(p('user%40domain.com:password')) # %40 still becomes '@'
|
||
self.assertIsNotNone(p('[email protected]:[email protected]:victory'))
|
|||
|
|
self.assertIsNotNone(p('[email protected];;shanmuga6'))
|
||
|
|
self.assertIsNotNone(p('[email protected]::shanmuga6'))
|
||
|
|
self.assertIsNotNone(p('[email protected]::hell@'))
|
||
|
|
self.assertIsNotNone(p('[email protected]::hell@:'))
|
||
|
|
self.assertIsNotNone(p('Aliulfat [email protected]:4849040'))
|
||
|
|
self.assertIsNotNone(p('harry gill@ yaho.com:7087096771'))
|
||
|
|
self.assertIsNotNone(p('shaheen .jafri @gmail.com:destinyuntold'))
|
||
|
|
self.assertIsNotNone(p('mail@ottino snc.it:04921540011'))
|
||
|
|
self.assertIsNone(p('sssssssandhya@gmail.:puleyar1'))
|
||
|
|
|
||
|
|
self.assertIsNotNone(p('cjywzy%40eyou.com:Aebdd797'))
|
||
|
|
self.assertIsNotNone(p('finanza.rapido2013%40yahoo.it:123456'))
|
||
|
|
self.assertIsNotNone(p('habitacasa3%40yahoo.it:cesira89'))
|
||
|
|
self.assertIsNotNone(p('harry gill@ yaho.com:7087096771'))
|
||
|
|
self.assertIsNone(p('uwe kopka <[email protected]>:biochemie'))
|
||
|
|
|
||
|
|
# multi-line.
|
||
|
|
self.assertIsNotNone(p('INLOVE12:::1920-05-02:[email protected]:INLOVE12'))
|
||
|
|
self.assertIsNotNone(p('Nicesweetboobs:1983-10-05:[email protected]:ilovekiss'))
|
||
|
|
self.assertIsNotNone(p('[email protected]:[email protected];97tata1'))
|
||
|
|
|
||
|
|
multi_lines = [
|
||
|
|
'[email protected];[email protected]:CAROLINAPIANTA',
|
||
|
|
'Nicesweetboobs:1983-10-05:[email protected]:ilovekiss',
|
||
|
|
'190.40.252.191,XabelciberX,[email protected],gatita9685567',
|
||
|
|
'201.245.164.50,andres vergara,[email protected],barcelona10',
|
||
|
|
'65.94.11.229,stephane,[email protected],ste264',
|
||
|
|
'INLOVE12:::1920-05-02:[email protected]:INLOVE12',
|
||
|
|
'agaliha84:::1969-07-02:[email protected]:agaliha84',
|
||
|
|
'happy8360:::1963-09-12:[email protected]:jdj407',
|
||
|
|
'tinkertailor99:1947-10-12:[email protected]:77880577',
|
||
|
|
'Cash5178:1970-11-06:[email protected]:hulk1234',
|
||
|
|
'Creecher0275:1972-04-22:[email protected]:complete',
|
||
|
|
'[email protected]:[email protected];97tata1',
|
||
|
|
'[email protected]:[email protected];zwezda',
|
||
|
|
'SavvySummer:1973-03-17:[email protected]:sunshine0699',
|
||
|
|
'RaWRrr:1990-12-02:[email protected]:nitsua',
|
||
|
|
'[email protected];[email protected]:CAROLINAPIANTA',
|
||
|
|
'[email protected];[email protected]:RRAHOTEL',
|
||
|
|
'silent4463:1980-10-14:[email protected]:salinas'
|
||
|
|
]
|
||
|
|
|
||
|
|
for multi_line in multi_lines:
|
||
|
|
self.assertIsNotNone(p(multi_line))
|
||
|
|
|
||
|
|
self.assertIsNone(p('300396 matelymaly [email protected] 222.155.173.78 '
|
||
|
|
'3d2c5414a041f518b03b750e62e4ebc3:Nhc'))
|
||
|
|
self.assertIsNone(p('05-02-2012;5721869;[email protected]:f02368945726d5fc2a14eb576f7276c0'))
|
||
|
|
self.assertIsNotNone(p('[email protected]::g1u3ck'))
|
||
|
|
self.assertIsNotNone(p('[email protected]:[email protected]:2904198'))
|
||
|
|
self.assertIsNotNone(p('sanda [email protected]:vasafultv'))
|
||
|
|
self.assertIsNone(p('jsjjd.@Com.:love melon'))
|
||
|
|
|
||
|
|
def test_parse_2(self):
|
||
|
|
t = time()
|
||
for _ in range(20000): # 10,000 * 13 = 130,000 in 4.0 seconds.
|
|||
# => 32B in in 10 days. That's a lot.
|
|||
|
|
p('[email protected]:123')
|
||
|
|
p('hello@@gmail.com:123')
|
||
|
|
p('hello @ gmail.com:123')
|
||
|
|
p('hello @@ gmail.com:123')
|
||
|
|
p('hello @@ gmail.com:123:')
|
||
|
|
p('hello @@ gmail.com,123')
|
||
|
|
p('123,hello @@ gmail.com')
|
||
|
|
p('[email protected]')
|
||
|
|
p('[email protected]:?????? ??? ?????')
|
||
|
|
p('[email protected]:85.242.201.81,felix')
|
||
|
|
p('[email protected]:125.26.106.158,nopparat')
|
||
|
|
p('[email protected]:INSERT INTO `users'
|
||
|
|
'`(`id`, `verify`, `username`, `email`, `password')
|
||
|
|
p('[email protected]:Login')
|
||
|
|
print(time() - t)
|
||
|
|
|
||
|
|
def test_parse_3(self):
|
||
self.assertEqual('j/o/s', get_output_file('[email protected]'))
|
|||
|
|
self.assertEqual('c/e/o', get_output_file('[email protected]'))
|
||
|
|
self.assertEqual('x/symbols', get_output_file('[email protected]'))
|
||
|
|
self.assertEqual('x/a/symbols', get_output_file('[email protected]'))
|
||
|
|
self.assertEqual('x/symbols', get_output_file('[email protected]'))
|
||
|
|
self.assertEqual('k/e/v', get_output_file('[email protected]'))
|
||
|
|||
|
|
|
||
|
|
if __name__ == '__main__':
|
||
|
|
unittest.main()
|