parse_iso8601,
strip_jsonp,
uppercase_escape,
+ limit_length,
+ escape_rfc3986,
+ escape_url,
)
self.assertEqual(uppercase_escape('aä'), 'aä')
self.assertEqual(uppercase_escape('\\U0001d550'), '𝕐')
+ def test_limit_length(self):
+ self.assertEqual(limit_length(None, 12), None)
+ self.assertEqual(limit_length('foo', 12), 'foo')
+ self.assertTrue(
+ limit_length('foo bar baz asd', 12).startswith('foo bar'))
+ self.assertTrue('...' in limit_length('foo bar baz asd', 12))
+
+ def test_escape_rfc3986(self):
+ reserved = "!*'();:@&=+$,/?#[]"
+ unreserved = 'ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789-_.~'
+ self.assertEqual(escape_rfc3986(reserved), reserved)
+ self.assertEqual(escape_rfc3986(unreserved), unreserved)
+ self.assertEqual(escape_rfc3986('тест'), '%D1%82%D0%B5%D1%81%D1%82')
+ self.assertEqual(escape_rfc3986('%D1%82%D0%B5%D1%81%D1%82'), '%D1%82%D0%B5%D1%81%D1%82')
+ self.assertEqual(escape_rfc3986('foo bar'), 'foo%20bar')
+ self.assertEqual(escape_rfc3986('foo%20bar'), 'foo%20bar')
+
+ def test_escape_url(self):
+ self.assertEqual(
+ escape_url('http://wowza.imust.org/srv/vod/telemb/new/UPLOAD/UPLOAD/20224_IncendieHavré_FD.mp4'),
+ 'http://wowza.imust.org/srv/vod/telemb/new/UPLOAD/UPLOAD/20224_IncendieHavre%CC%81_FD.mp4'
+ )
+ self.assertEqual(
+ escape_url('http://www.ardmediathek.de/tv/Sturm-der-Liebe/Folge-2036-Zu-Mann-und-Frau-erklärt/Das-Erste/Video?documentId=22673108&bcastId=5290'),
+ 'http://www.ardmediathek.de/tv/Sturm-der-Liebe/Folge-2036-Zu-Mann-und-Frau-erkl%C3%A4rt/Das-Erste/Video?documentId=22673108&bcastId=5290'
+ )
+ self.assertEqual(
+ escape_url('http://тест.рф/фрагмент'),
+ 'http://тест.рф/%D1%84%D1%80%D0%B0%D0%B3%D0%BC%D0%B5%D0%BD%D1%82'
+ )
+ self.assertEqual(
+ escape_url('http://тест.рф/абв?абв=абв#абв'),
+ 'http://тест.рф/%D0%B0%D0%B1%D0%B2?%D0%B0%D0%B1%D0%B2=%D0%B0%D0%B1%D0%B2#%D0%B0%D0%B1%D0%B2'
+ )
+ self.assertEqual(escape_url('http://vimeo.com/56015672#at=0'), 'http://vimeo.com/56015672#at=0')
+
if __name__ == '__main__':
unittest.main()
return node.find(expr)
else:
def find_xpath_attr(node, xpath, key, val):
+ # Here comes the crazy part: In 2.6, if the xpath is a unicode,
+ # .//node does not match if a node is a direct child of . !
+ if isinstance(xpath, unicode):
+ xpath = xpath.encode('ascii')
+
for f in node.findall(xpath):
if f.attrib.get(key) == val:
return f
return '/'.join(replaced)
+def xpath_text(node, xpath, name=None, fatal=False):
+ if sys.version_info < (2, 7): # Crazy 2.6
+ xpath = xpath.encode('ascii')
+
+ n = node.find(xpath)
+ if n is None:
+ if fatal:
+ name = xpath if name is None else name
+ raise ExtractorError('Could not find XML element %s' % name)
+ else:
+ return None
+ return n.text
+
+
compat_html_parser.locatestarttagend = re.compile(r"""<[a-zA-Z][-.a-zA-Z0-9:_]*(?:\s+(?:(?<=['"\s])[^\s/>][^\s/=>]*(?:\s*=+\s*(?:'[^']*'|"[^"]*"|(?!['"])[^>\s]*))?\s*)*)?\s*""", re.VERBOSE) # backport bugfix
class BaseHTMLParser(compat_html_parser.HTMLParser):
def __init(self):
lambda m: unicode_escape(m.group(0))[0],
s)
+
+ def escape_rfc3986(s):
+ """Escape non-ASCII characters as suggested by RFC 3986"""
+ if sys.version_info < (3, 0) and isinstance(s, unicode):
+ s = s.encode('utf-8')
+ return compat_urllib_parse.quote(s, "%/;:@&=+$,!~*'()?#[]")
+
+
+ def escape_url(url):
+ """Escape URL as suggested by RFC 3986"""
+ url_parsed = compat_urllib_parse_urlparse(url)
+ return url_parsed._replace(
+ path=escape_rfc3986(url_parsed.path),
+ params=escape_rfc3986(url_parsed.params),
+ query=escape_rfc3986(url_parsed.query),
+ fragment=escape_rfc3986(url_parsed.fragment)
+ ).geturl()
+
try:
struct.pack(u'!I', 0)
except TypeError:
if ret:
raise subprocess.CalledProcessError(ret, p.args, output=output)
return output
+
+
+def limit_length(s, length):
+ """ Add ellipses to overly long strings """
+ if s is None:
+ return None
+ ELLIPSES = '...'
+ if len(s) > length:
+ return s[:length - len(ELLIPSES)] + ELLIPSES
+ return s