cpython/Lib/test/test_normalization.py
Florent Xicluna f089fd67fc Merged revisions 78982,78986 via svnmerge from
svn+ssh://pythondev@svn.python.org/python/trunk

........
  r78982 | florent.xicluna | 2010-03-15 15:00:58 +0100 (lun, 15 mar 2010) | 2 lines

  Remove py3k deprecation warnings from these Unicode tools.
........
  r78986 | florent.xicluna | 2010-03-15 19:08:58 +0100 (lun, 15 mar 2010) | 3 lines

  Issue #7783 and #7787: open_urlresource invalidates the outdated files from the local cache.
  Use this feature to fix test_normalization.
........
2010-03-19 14:25:03 +00:00

107 lines
3.2 KiB
Python

from test.support import run_unittest, open_urlresource
import unittest
from http.client import HTTPException
import sys
import os
from unicodedata import normalize, unidata_version
TESTDATAFILE = "NormalizationTest.txt"
TESTDATAURL = "http://www.unicode.org/Public/" + unidata_version + "/ucd/" + TESTDATAFILE
def check_version(testfile):
hdr = testfile.readline()
return unidata_version in hdr
class RangeError(Exception):
pass
def NFC(str):
return normalize("NFC", str)
def NFKC(str):
return normalize("NFKC", str)
def NFD(str):
return normalize("NFD", str)
def NFKD(str):
return normalize("NFKD", str)
def unistr(data):
data = [int(x, 16) for x in data.split(" ")]
for x in data:
if x > sys.maxunicode:
raise RangeError
return "".join([chr(x) for x in data])
class NormalizationTest(unittest.TestCase):
def test_main(self):
part = None
part1_data = {}
# Hit the exception early
try:
testdata = open_urlresource(TESTDATAURL, encoding="utf-8",
check=check_version)
except (IOError, HTTPException):
self.skipTest("Could not retrieve " + TESTDATAURL)
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
if part == "@Part3":
# XXX we don't support PRI #29 yet, so skip these tests for now
continue
try:
c1,c2,c3,c4,c5 = [unistr(x) for x in line.split(';')[:-1]]
except RangeError:
# Skip unsupported characters;
# try atleast adding c1 if we are in part1
if part == "@Part1":
try:
c1 = unistr(line.split(';')[0])
except RangeError:
pass
else:
part1_data[c1] = 1
continue
# 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)
# 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_bug_834676(self):
# Check for bug 834676
normalize('NFC', '\ud55c\uae00')
def test_main():
run_unittest(NormalizationTest)
if __name__ == "__main__":
test_main()