def test_quote_unicode():
"""Quoting and unquoting Unicode strings should give the same result
as when given regular strings. See ticket #28786.
"""
assert urllib_utf8.quote('montréal') == urllib_utf8.quote(u'montréal')
assert urllib_utf8.quote_plus('montréal') == urllib_utf8.quote_plus(u'montréal')
assert urllib_utf8.unquote('montréal') == urllib_utf8.unquote(u'montréal')
assert urllib_utf8.unquote_plus('montréal') == urllib_utf8.unquote_plus(u'montréal')
def test_quote_and_quote_plus():
strings = ['', 'Hello there!', u'naïve', u' San José ', ' cafés']
for string in strings:
utf8_string = string.encode('utf-8') if isinstance(string, unicode) else string
assert urllib_utf8.quote(string) == urllib.quote(utf8_string)
assert urllib_utf8.quote_plus(string) == urllib.quote_plus(utf8_string)
def test_recode_encoded(charname, chars, pathchars, expected_url, encoding):
url_template = u"http://m{chars}nchen.com/m{chars}chen/{pathchars}"
unicode_url = url_template.format(chars=chars, pathchars=pathchars)
try:
encoded_url = unicode_url.encode(encoding)
except UnicodeEncodeError:
pytest.skip("Some of these things just won't go.")
assert E.recode_uri(encoded_url) == expected_url
quoted_url = url_template.format(
chars=quote(chars.encode(encoding)),
pathchars=quote(pathchars.encode(encoding)),
)
if charname == 'ascii':
# ASCII is a special case when it comes to quoting: their quoted-ness should go untouched.
assert E.recode_uri(quoted_url) == quoted_url
else:
assert E.recode_uri(quoted_url) == expected_url
def test_recode_encoded(charname, chars, pathchars, expected_url, encoding):
url_template = u"http://m{chars}nchen.com/m{chars}chen/{pathchars}"
unicode_url = url_template.format(chars=chars, pathchars=pathchars)
try:
encoded_url = unicode_url.encode(encoding)
except UnicodeEncodeError:
pytest.skip("Some of these things just won't go.")
assert E.recode_uri(encoded_url) == expected_url
quoted_url = url_template.format(
chars=quote(chars.encode(encoding)),
pathchars=quote(pathchars.encode(encoding)),
)
if charname == 'ascii':
# ASCII is a special case when it comes to quoting: their quoted-ness should go untouched.
assert E.recode_uri(quoted_url) == quoted_url
else:
assert E.recode_uri(quoted_url) == expected_url
class TestRecodeEmail(object):
munchen = u'münchen'
email = u'{munchen}@{munchen}.com?subject={munchen}'.format(
munchen=munchen)
# The username is best encoded as simply utf8.
expected = u'{utf8_munchen}@{idna_munchen}.com?subject={percent_munchen}'.format(
utf8_munchen=munchen,
idna_munchen=munchen.encode('IDNA').decode('US-ASCII'),
percent_munchen=quote(munchen.encode('UTF-8')),
)
@staticmethod
def test_empty_string():
assert E.encode_email('') == ''
def test_not_an_email(self):
not_an_email = u"They don't use email in " + self.munchen
assert E.encode_email(not_an_email) == not_an_email.encode(
'IDNA').decode('US-ASCII')
def test_encode_email(self):
assert E.encode_email(self.email) == self.expected
def test_decode_email(self):
assert E.decode_email(self.expected) == self.email
def test_recode_email(self):
assert E.recode_email(self.email) == self.expected
# Tests for idempotency:
def test_encode_email_idempotent(self):
assert E.encode_email(E.encode_email(self.email)) == self.expected
def test_decode_email_idempotent(self):
assert E.decode_email(E.decode_email(self.expected)) == self.email
def test_recode_email_idempotent(self):
assert E.recode_email(E.recode_email(self.email)) == self.expected
def test_mailto_scheme(self):
assert E.recode_email('mailto:' +
self.email) == 'mailto:' + self.expected
# This is technically wrong, but we fix it.
assert E.recode_email('mailto://' +
self.email) == 'mailto:' + self.expected
def test_url_reserved_chars():
url = 'http://www.yelp.com?chars=%s' % quote(':/?&=')
assert E.recode_uri(url) == url
def test_unescape_quoted_nonascii(self):
quoted_non_ascii = quote(self.NON_ASCII)
unquoted_url = 'http://yelp.com/' + self.NON_ASCII
quoted_url = 'http://yelp.com/' + quoted_non_ascii
self.assert_unquote_bytes(quoted_url, unquoted_url)
def test_dont_touch_quoted_ascii(self):
quoted_ascii = quote(self.ASCII)
url = 'http://yelp.com/' + quoted_ascii
self.assert_unquote_bytes(url, url)
def test_dont_touch_quoted_ascii(self):
quoted_ascii = quote(self.ASCII)
url = b'http://yelp.com/' + quoted_ascii.encode('US-ASCII')
self.assert_unquote_bytes(url, url)
def test_url_reserved_chars():
url = 'http://www.yelp.com?chars=%s' % quote(':/?&=')
assert E.recode_uri(url) == url
def test_unescape_quoted_nonascii(self):
quoted_non_ascii = quote(self.NON_ASCII)
unquoted_url = b'http://yelp.com/' + self.NON_ASCII
quoted_url = b'http://yelp.com/' + quoted_non_ascii.encode('US-ASCII')
self.assert_unquote_bytes(quoted_url, unquoted_url)
def test_dont_touch_quoted_ascii(self):
quoted_ascii = quote(self.ASCII)
url = b'http://yelp.com/' + quoted_ascii.encode('US-ASCII')
self.assert_unquote_bytes(url, url)
def test_dont_touch_quoted_ascii(self):
quoted_ascii = quote(self.ASCII)
url = 'http://yelp.com/' + quoted_ascii
self.assert_unquote_bytes(url, url)
