cpython/Lib/test/test_unicodedata.py

466 lines
18 KiB
Python

""" Tests for the unicodedata module.
Written by Marc-Andre Lemburg (mal@lemburg.com).
(c) Copyright CNRI, All Rights Reserved. NO WARRANTY.
"""
import hashlib
from http.client import HTTPException
import sys
import unicodedata
import unittest
from test.support import (open_urlresource, requires_resource, script_helper,
cpython_only, check_disallow_instantiation)
class UnicodeMethodsTest(unittest.TestCase):
# update this, if the database changes
expectedchecksum = '9e43ee3929471739680c0e705482b4ae1c4122e4'
@requires_resource('cpu')
def test_method_checksum(self):
h = hashlib.sha1()
for i in range(sys.maxunicode + 1):
char = chr(i)
data = [
# Predicates (single char)
"01"[char.isalnum()],
"01"[char.isalpha()],
"01"[char.isdecimal()],
"01"[char.isdigit()],
"01"[char.islower()],
"01"[char.isnumeric()],
"01"[char.isspace()],
"01"[char.istitle()],
"01"[char.isupper()],
# Predicates (multiple chars)
"01"[(char + 'abc').isalnum()],
"01"[(char + 'abc').isalpha()],
"01"[(char + '123').isdecimal()],
"01"[(char + '123').isdigit()],
"01"[(char + 'abc').islower()],
"01"[(char + '123').isnumeric()],
"01"[(char + ' \t').isspace()],
"01"[(char + 'abc').istitle()],
"01"[(char + 'ABC').isupper()],
# Mappings (single char)
char.lower(),
char.upper(),
char.title(),
# Mappings (multiple chars)
(char + 'abc').lower(),
(char + 'ABC').upper(),
(char + 'abc').title(),
(char + 'ABC').title(),
]
h.update(''.join(data).encode('utf-8', 'surrogatepass'))
result = h.hexdigest()
self.assertEqual(result, self.expectedchecksum)
class UnicodeDatabaseTest(unittest.TestCase):
db = unicodedata
class UnicodeFunctionsTest(UnicodeDatabaseTest):
# Update this if the database changes. Make sure to do a full rebuild
# (e.g. 'make distclean && make') to get the correct checksum.
expectedchecksum = '23ab09ed4abdf93db23b97359108ed630dd8311d'
@requires_resource('cpu')
def test_function_checksum(self):
data = []
h = hashlib.sha1()
for i in range(sys.maxunicode + 1):
char = chr(i)
data = [
# Properties
format(self.db.digit(char, -1), '.12g'),
format(self.db.numeric(char, -1), '.12g'),
format(self.db.decimal(char, -1), '.12g'),
self.db.category(char),
self.db.bidirectional(char),
self.db.decomposition(char),
str(self.db.mirrored(char)),
str(self.db.combining(char)),
unicodedata.east_asian_width(char),
self.db.name(char, ""),
]
h.update(''.join(data).encode("ascii"))
result = h.hexdigest()
self.assertEqual(result, self.expectedchecksum)
@requires_resource('cpu')
def test_name_inverse_lookup(self):
for i in range(sys.maxunicode + 1):
char = chr(i)
if looked_name := self.db.name(char, None):
self.assertEqual(self.db.lookup(looked_name), char)
def test_no_names_in_pua(self):
puas = [*range(0xe000, 0xf8ff),
*range(0xf0000, 0xfffff),
*range(0x100000, 0x10ffff)]
for i in puas:
char = chr(i)
self.assertRaises(ValueError, self.db.name, char)
def test_lookup_nonexistant(self):
# just make sure that lookup can fail
for nonexistent in [
"LATIN SMLL LETR A",
"OPEN HANDS SIGHS",
"DREGS",
"HANDBUG",
"MODIFIER LETTER CYRILLIC SMALL QUESTION MARK",
"???",
]:
self.assertRaises(KeyError, self.db.lookup, nonexistent)
def test_digit(self):
self.assertEqual(self.db.digit('A', None), None)
self.assertEqual(self.db.digit('9'), 9)
self.assertEqual(self.db.digit('\u215b', None), None)
self.assertEqual(self.db.digit('\u2468'), 9)
self.assertEqual(self.db.digit('\U00020000', None), None)
self.assertEqual(self.db.digit('\U0001D7FD'), 7)
self.assertRaises(TypeError, self.db.digit)
self.assertRaises(TypeError, self.db.digit, 'xx')
self.assertRaises(ValueError, self.db.digit, 'x')
def test_numeric(self):
self.assertEqual(self.db.numeric('A',None), None)
self.assertEqual(self.db.numeric('9'), 9)
self.assertEqual(self.db.numeric('\u215b'), 0.125)
self.assertEqual(self.db.numeric('\u2468'), 9.0)
self.assertEqual(self.db.numeric('\ua627'), 7.0)
self.assertEqual(self.db.numeric('\U00020000', None), None)
self.assertEqual(self.db.numeric('\U0001012A'), 9000)
self.assertRaises(TypeError, self.db.numeric)
self.assertRaises(TypeError, self.db.numeric, 'xx')
self.assertRaises(ValueError, self.db.numeric, 'x')
def test_decimal(self):
self.assertEqual(self.db.decimal('A',None), None)
self.assertEqual(self.db.decimal('9'), 9)
self.assertEqual(self.db.decimal('\u215b', None), None)
self.assertEqual(self.db.decimal('\u2468', None), None)
self.assertEqual(self.db.decimal('\U00020000', None), None)
self.assertEqual(self.db.decimal('\U0001D7FD'), 7)
self.assertRaises(TypeError, self.db.decimal)
self.assertRaises(TypeError, self.db.decimal, 'xx')
self.assertRaises(ValueError, self.db.decimal, 'x')
def test_category(self):
self.assertEqual(self.db.category('\uFFFE'), 'Cn')
self.assertEqual(self.db.category('a'), 'Ll')
self.assertEqual(self.db.category('A'), 'Lu')
self.assertEqual(self.db.category('\U00020000'), 'Lo')
self.assertEqual(self.db.category('\U0001012A'), 'No')
self.assertRaises(TypeError, self.db.category)
self.assertRaises(TypeError, self.db.category, 'xx')
def test_bidirectional(self):
self.assertEqual(self.db.bidirectional('\uFFFE'), '')
self.assertEqual(self.db.bidirectional(' '), 'WS')
self.assertEqual(self.db.bidirectional('A'), 'L')
self.assertEqual(self.db.bidirectional('\U00020000'), 'L')
self.assertRaises(TypeError, self.db.bidirectional)
self.assertRaises(TypeError, self.db.bidirectional, 'xx')
def test_decomposition(self):
self.assertEqual(self.db.decomposition('\uFFFE'),'')
self.assertEqual(self.db.decomposition('\u00bc'), '<fraction> 0031 2044 0034')
self.assertRaises(TypeError, self.db.decomposition)
self.assertRaises(TypeError, self.db.decomposition, 'xx')
def test_mirrored(self):
self.assertEqual(self.db.mirrored('\uFFFE'), 0)
self.assertEqual(self.db.mirrored('a'), 0)
self.assertEqual(self.db.mirrored('\u2201'), 1)
self.assertEqual(self.db.mirrored('\U00020000'), 0)
self.assertRaises(TypeError, self.db.mirrored)
self.assertRaises(TypeError, self.db.mirrored, 'xx')
def test_combining(self):
self.assertEqual(self.db.combining('\uFFFE'), 0)
self.assertEqual(self.db.combining('a'), 0)
self.assertEqual(self.db.combining('\u20e1'), 230)
self.assertEqual(self.db.combining('\U00020000'), 0)
self.assertRaises(TypeError, self.db.combining)
self.assertRaises(TypeError, self.db.combining, 'xx')
def test_pr29(self):
# https://www.unicode.org/review/pr-29.html
# See issues #1054943 and #10254.
composed = ("\u0b47\u0300\u0b3e", "\u1100\u0300\u1161",
'Li\u030dt-s\u1e73\u0301',
'\u092e\u093e\u0930\u094d\u0915 \u091c\u093c'
+ '\u0941\u0915\u0947\u0930\u092c\u0930\u094d\u0917',
'\u0915\u093f\u0930\u094d\u0917\u093f\u091c\u093c'
+ '\u0938\u094d\u0924\u093e\u0928')
for text in composed:
self.assertEqual(self.db.normalize('NFC', text), text)
def test_issue10254(self):
# Crash reported in #10254
a = 'C\u0338' * 20 + 'C\u0327'
b = 'C\u0338' * 20 + '\xC7'
self.assertEqual(self.db.normalize('NFC', a), b)
def test_issue29456(self):
# Fix #29456
u1176_str_a = '\u1100\u1176\u11a8'
u1176_str_b = '\u1100\u1176\u11a8'
u11a7_str_a = '\u1100\u1175\u11a7'
u11a7_str_b = '\uae30\u11a7'
u11c3_str_a = '\u1100\u1175\u11c3'
u11c3_str_b = '\uae30\u11c3'
self.assertEqual(self.db.normalize('NFC', u1176_str_a), u1176_str_b)
self.assertEqual(self.db.normalize('NFC', u11a7_str_a), u11a7_str_b)
self.assertEqual(self.db.normalize('NFC', u11c3_str_a), u11c3_str_b)
def test_east_asian_width(self):
eaw = self.db.east_asian_width
self.assertRaises(TypeError, eaw, b'a')
self.assertRaises(TypeError, eaw, bytearray())
self.assertRaises(TypeError, eaw, '')
self.assertRaises(TypeError, eaw, 'ra')
self.assertEqual(eaw('\x1e'), 'N')
self.assertEqual(eaw('\x20'), 'Na')
self.assertEqual(eaw('\uC894'), 'W')
self.assertEqual(eaw('\uFF66'), 'H')
self.assertEqual(eaw('\uFF1F'), 'F')
self.assertEqual(eaw('\u2010'), 'A')
self.assertEqual(eaw('\U00020000'), 'W')
def test_east_asian_width_unassigned(self):
eaw = self.db.east_asian_width
# unassigned
for char in '\u0530\u0ecf\u10c6\u20fc\uaaca\U000107bd\U000115f2':
self.assertEqual(eaw(char), 'N')
self.assertIs(self.db.name(char, None), None)
# unassigned but reserved for CJK
for char in '\uFA6E\uFADA\U0002A6E0\U0002FA20\U0003134B\U0003FFFD':
self.assertEqual(eaw(char), 'W')
self.assertIs(self.db.name(char, None), None)
# private use areas
for char in '\uE000\uF800\U000F0000\U000FFFEE\U00100000\U0010FFF0':
self.assertEqual(eaw(char), 'A')
self.assertIs(self.db.name(char, None), None)
def test_east_asian_width_9_0_changes(self):
self.assertEqual(self.db.ucd_3_2_0.east_asian_width('\u231a'), 'N')
self.assertEqual(self.db.east_asian_width('\u231a'), 'W')
class UnicodeMiscTest(UnicodeDatabaseTest):
@cpython_only
def test_disallow_instantiation(self):
# Ensure that the type disallows instantiation (bpo-43916)
check_disallow_instantiation(self, unicodedata.UCD)
def test_failed_import_during_compiling(self):
# Issue 4367
# Decoding \N escapes requires the unicodedata module. If it can't be
# imported, we shouldn't segfault.
# This program should raise a SyntaxError in the eval.
code = "import sys;" \
"sys.modules['unicodedata'] = None;" \
"""eval("'\\\\N{SOFT HYPHEN}'")"""
# We use a separate process because the unicodedata module may already
# have been loaded in this process.
result = script_helper.assert_python_failure("-c", code)
error = "SyntaxError: (unicode error) \\N escapes not supported " \
"(can't load unicodedata module)"
self.assertIn(error, result.err.decode("ascii"))
def test_decimal_numeric_consistent(self):
# Test that decimal and numeric are consistent,
# i.e. if a character has a decimal value,
# its numeric value should be the same.
count = 0
for i in range(0x10000):
c = chr(i)
dec = self.db.decimal(c, -1)
if dec != -1:
self.assertEqual(dec, self.db.numeric(c))
count += 1
self.assertTrue(count >= 10) # should have tested at least the ASCII digits
def test_digit_numeric_consistent(self):
# Test that digit and numeric are consistent,
# i.e. if a character has a digit value,
# its numeric value should be the same.
count = 0
for i in range(0x10000):
c = chr(i)
dec = self.db.digit(c, -1)
if dec != -1:
self.assertEqual(dec, self.db.numeric(c))
count += 1
self.assertTrue(count >= 10) # should have tested at least the ASCII digits
def test_bug_1704793(self):
self.assertEqual(self.db.lookup("GOTHIC LETTER FAIHU"), '\U00010346')
def test_ucd_510(self):
import unicodedata
# In UCD 5.1.0, a mirrored property changed wrt. UCD 3.2.0
self.assertTrue(unicodedata.mirrored("\u0f3a"))
self.assertTrue(not unicodedata.ucd_3_2_0.mirrored("\u0f3a"))
# Also, we now have two ways of representing
# the upper-case mapping: as delta, or as absolute value
self.assertTrue("a".upper()=='A')
self.assertTrue("\u1d79".upper()=='\ua77d')
self.assertTrue(".".upper()=='.')
@requires_resource('cpu')
def test_bug_5828(self):
self.assertEqual("\u1d79".lower(), "\u1d79")
# Only U+0000 should have U+0000 as its upper/lower/titlecase variant
self.assertEqual(
[
c for c in range(sys.maxunicode+1)
if "\x00" in chr(c).lower()+chr(c).upper()+chr(c).title()
],
[0]
)
def test_bug_4971(self):
# LETTER DZ WITH CARON: DZ, Dz, dz
self.assertEqual("\u01c4".title(), "\u01c5")
self.assertEqual("\u01c5".title(), "\u01c5")
self.assertEqual("\u01c6".title(), "\u01c5")
def test_linebreak_7643(self):
for i in range(0x10000):
lines = (chr(i) + 'A').splitlines()
if i in (0x0a, 0x0b, 0x0c, 0x0d, 0x85,
0x1c, 0x1d, 0x1e, 0x2028, 0x2029):
self.assertEqual(len(lines), 2,
r"\u%.4x should be a linebreak" % i)
else:
self.assertEqual(len(lines), 1,
r"\u%.4x should not be a linebreak" % i)
class NormalizationTest(unittest.TestCase):
@staticmethod
def check_version(testfile):
hdr = testfile.readline()
return unicodedata.unidata_version in hdr
@staticmethod
def unistr(data):
data = [int(x, 16) for x in data.split(" ")]
return "".join([chr(x) for x in data])
@requires_resource('network')
@requires_resource('cpu')
def test_normalization(self):
TESTDATAFILE = "NormalizationTest.txt"
TESTDATAURL = f"http://www.pythontest.net/unicode/{unicodedata.unidata_version}/{TESTDATAFILE}"
# Hit the exception early
try:
testdata = open_urlresource(TESTDATAURL, encoding="utf-8",
check=self.check_version)
except PermissionError:
self.skipTest(f"Permission error when downloading {TESTDATAURL} "
f"into the test data directory")
except (OSError, HTTPException) as exc:
self.skipTest(f"Failed to download {TESTDATAURL}: {exc}")
with testdata:
self.run_normalization_tests(testdata)
def run_normalization_tests(self, testdata):
part = None
part1_data = {}
def NFC(str):
return unicodedata.normalize("NFC", str)
def NFKC(str):
return unicodedata.normalize("NFKC", str)
def NFD(str):
return unicodedata.normalize("NFD", str)
def NFKD(str):
return unicodedata.normalize("NFKD", str)
for line in testdata:
if '#' in line:
line = line.split('#')[0]
line = line.strip()
if not line:
continue
if line.startswith("@Part"):
part = line.split()[0]
continue
c1,c2,c3,c4,c5 = [self.unistr(x) for x in line.split(';')[:-1]]
# Perform tests
self.assertTrue(c2 == NFC(c1) == NFC(c2) == NFC(c3), line)
self.assertTrue(c4 == NFC(c4) == NFC(c5), line)
self.assertTrue(c3 == NFD(c1) == NFD(c2) == NFD(c3), line)
self.assertTrue(c5 == NFD(c4) == NFD(c5), line)
self.assertTrue(c4 == NFKC(c1) == NFKC(c2) == \
NFKC(c3) == NFKC(c4) == NFKC(c5),
line)
self.assertTrue(c5 == NFKD(c1) == NFKD(c2) == \
NFKD(c3) == NFKD(c4) == NFKD(c5),
line)
self.assertTrue(unicodedata.is_normalized("NFC", c2))
self.assertTrue(unicodedata.is_normalized("NFC", c4))
self.assertTrue(unicodedata.is_normalized("NFD", c3))
self.assertTrue(unicodedata.is_normalized("NFD", c5))
self.assertTrue(unicodedata.is_normalized("NFKC", c4))
self.assertTrue(unicodedata.is_normalized("NFKD", c5))
# Record part 1 data
if part == "@Part1":
part1_data[c1] = 1
# Perform tests for all other data
for c in range(sys.maxunicode+1):
X = chr(c)
if X in part1_data:
continue
self.assertTrue(X == NFC(X) == NFD(X) == NFKC(X) == NFKD(X), c)
def test_edge_cases(self):
self.assertRaises(TypeError, unicodedata.normalize)
self.assertRaises(ValueError, unicodedata.normalize, 'unknown', 'xx')
self.assertEqual(unicodedata.normalize('NFKC', ''), '')
def test_bug_834676(self):
# Check for bug 834676
unicodedata.normalize('NFC', '\ud55c\uae00')
if __name__ == "__main__":
unittest.main()