simon987/opendirectories-bot
1
0
Fork 0
mirror of https://github.com/simon987/opendirectories-bot.git synced 2025-12-14 15:49:03 +00:00
Code Issues Projects Releases Wiki Activity

Fixed problem with non-http websites

Browse Source
This commit is contained in:
simon
2018-02-06 15:50:47 -05:00
parent 23775ec126
commit 854fc76cc1
5 changed files with 838 additions and 14 deletions
Download Patch File Download Diff File Expand all files Collapse all files

22
spec/Crawler_spec.py
View File

@@ -23,10 +23,22 @@ class CrawlerTest(TestCase):
self.assertEqual(c.guess_parser(text, {}), NginxParser)
def test_guess_parser3(self):
with open("test_invalid.html", "r") as f:
text = f.read()
# def test_guess_parser3(self):
# with open("test_invalid.html", "r") as f:
# text = f.read()
#
# c = Crawler("http://some.website", False)
#
# self.assertEqual(c.guess_parser(text, {}), None)
c = Crawler("http://some.website", False)
def test_invalid_schema(self):
self.assertEqual(c.guess_parser(text, {}), None)
c1 = Crawler("http://google.com/", False)
c2 = Crawler("https://google.com/", False)
c3 = Crawler("ftp://website.com/", False)
c4 = Crawler("ws://website.com/", False)
self.assertIsNotNone(c1.parser)
self.assertIsNotNone(c2.parser)
self.assertIsNone(c3.parser)
self.assertIsNone(c4.parser)

771
spec/test_invalid.html Normal file
View File

File diff suppressed because one or more lines are too long