def test_hostname_labels_valid_idn(self, label):
# type: (Text) -> None
"""
hostname_labels() generates IDN host name labels.
"""
try:
check_label(label)
idna_encode(label)
except UnicodeError: # pragma: no cover
raise AssertionError("Invalid IDN label: {!r}".format(label))
def test_hostname_labels_valid_ascii(self, label):
# type: (Text) -> None
"""
hostname_labels() generates a ASCII host name labels.
"""
try:
check_label(label)
label.encode("ascii")
except UnicodeError: # pragma: no cover
raise AssertionError("Invalid ASCII label: {!r}".format(label))
def test_hostnames_ascii(self, hostname):
# type: (Text) -> None
"""
hostnames() generates a ASCII host names.
"""
try:
for label in hostname.split(u"."):
check_label(label)
hostname.encode("ascii")
except UnicodeError: # pragma: no cover
raise AssertionError(
"Invalid ASCII host name: {!r}".format(hostname))
def test_hostnames_idn(self, hostname):
# type: (Text) -> None
"""
hostnames() generates a IDN host names.
"""
try:
for label in hostname.split(u"."):
check_label(label)
idna_encode(hostname)
except UnicodeError: # pragma: no cover
raise AssertionError(
"Invalid IDN host name: {!r}".format(hostname))
def hostname_labels(draw, allow_idn=True): # pragma: no cover
# type: (DrawCallable, bool) -> Text
"""
A strategy which generates host name labels.
@param allow_idn: Whether to allow non-ASCII characters as allowed by
internationalized domain names (IDNs).
"""
if allow_idn:
label = cast(Text, draw(idna_text(min_size=1, max_size=63)))
try:
label.encode("ascii")
except UnicodeEncodeError:
# If the label doesn't encode to ASCII, then we need to check the
# length of the label after encoding to punycode and adding the
# xn-- prefix.
while len(label.encode("punycode")) > 63 - len("xn--"):
# Rather than bombing out, just trim from the end until it is
# short enough, so hypothesis doesn't have to generate new
# data.
label = label[:-1]
else:
label = cast(
Text,
draw(
text(
min_size=1,
max_size=63,
alphabet=unicode(ascii_letters + digits + "-"),
)
),
)
# Filter invalid labels.
# It would be better not to generate bogus labels in the first place... but
# that's not trivial.
try:
check_label(label)
except UnicodeError:
assume(False)
return label
def hostname_labels(draw, allow_idn=True): # pragma: no cover
# type: (DrawCallable, bool) -> Text
"""
A strategy which generates host name labels.
@param allow_idn: Whether to allow non-ASCII characters as allowed by
internationalized domain names (IDNs).
"""
if allow_idn:
label = draw(idna_text(min_size=1, max_size=63))
try:
label.encode("ascii")
except UnicodeEncodeError:
# If the label doesn't encode to ASCII, then we need to check the
# length of the label after encoding to punycode and adding the
# xn-- prefix.
while len(label.encode("punycode")) > 63 - len("xn--"):
# Rather than bombing out, just trim from the end until it is
# short enough, so hypothesis doesn't have to generate new
# data.
label = label[:-1]
else:
label = draw(text(
min_size=1, max_size=63,
alphabet=unicode(ascii_letters + digits + u"-")
))
# Filter invalid labels.
# It would be better not to generate bogus labels in the first place... but
# that's not trivial.
try:
check_label(label)
except UnicodeError:
assume(False)
return label
def test_hostname_labels_long_idn_punycode(self, data):
# type: (SearchStrategy) -> None
"""
hostname_labels() handles case where idna_text() generates text
that encoded to punycode ends up as longer than allowed.
"""
@composite
def mock_idna_text(draw, min_size, max_size):
# type: (DrawCallable, int, int) -> Text
# We want a string that does not exceed max_size, but when
# encoded to punycode, does exceed max_size.
# So use a unicode character that is larger when encoded,
# "á" being a great example, and use it max_size times, which
# will be max_size * 3 in size when encoded.
return u"\N{LATIN SMALL LETTER A WITH ACUTE}" * max_size
with patch("hyperlink.hypothesis.idna_text", mock_idna_text):
label = data.draw(hostname_labels())
try:
check_label(label)
idna_encode(label)
except UnicodeError: # pragma: no cover
raise AssertionError(
"Invalid IDN label: {!r}".format(label))
def to_string(proto, buf):
string = buf.decode("utf-8")
for label in string.split("."):
idna.check_label(label)
return string
