Hopiu/linkchecker
1
0
Fork 0
mirror of https://github.com/Hopiu/linkchecker.git synced 2026-05-02 20:04:43 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

fix parser encoding tests after change of parser

Browse source 
UnicodeDammit input has to be non-unicode to trigger character set
detection.
This commit is contained in:
Petr Dlouhý 2019-07-22 19:59:37 +01:00 committed by Chris Mayo
parent b5111453d8
commit 69d426b36f
1 changed files with 24 additions and 6 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

30
tests/test_parser.py
View file

@ -285,18 +285,36 @@ class TestParser (unittest.TestCase):
self.assertEqual(resolve("&#%d;" % ord(c)), c)
self.assertEqual(resolve("�"), u"")
def test_encoding_detection (self):
html = '<meta http-equiv="content-type" content="text/html; charset=UTF-8">'
def test_encoding_detection_utf_content (self):
html = b'<meta http-equiv="content-type" content="text/html; charset=UTF-8">'
self.encoding_test(html, "utf-8")
html = '<meta charset="UTF-8">'
def test_encoding_detection_utf_charset (self):
html = b'<meta charset="UTF-8">'
self.encoding_test(html, "utf-8")
html = '<meta charset="hulla">'
def test_encoding_detection_iso_content (self):
html = b'<meta http-equiv="content-type" content="text/html; charset=ISO8859-1">'
self.encoding_test(html, "iso8859-1")
html = '<meta http-equiv="content-type" content="text/html; charset=blabla">'
def test_encoding_detection_iso_charset (self):
html = b'<meta charset="ISO8859-1">'
self.encoding_test(html, "iso8859-1")
def test_encoding_detection_iso_bad_charset (self):
html = b'<meta charset="hulla">'
self.encoding_test(html, "ascii")
def test_encoding_detection_iso_bad_content (self):
html = b'<meta http-equiv="content-type" content="text/html; charset=blabla">'
self.encoding_test(html, "ascii")
def encoding_test (self, html, expected):
parser = linkcheck.HtmlParser.htmlsax.parser()
self.assertEqual(parser.encoding, "iso8859-1")
self.assertEqual(parser.encoding, None)
out = StringIO()
handler = linkcheck.HtmlParser.htmllib.HtmlPrettyPrinter(out)
parser.handler = handler
parser.feed(html)
parser.flush()
self.assertEqual(parser.encoding, expected)