Lib/test/test_normalization.py

   1 from test.test_support import run_unittest, open_urlresource
   2 import unittest
   3
   4 from httplib import HTTPException
   5 import sys
   6 import os
   7 from unicodedata import normalize, unidata_version
   8
   9 TESTDATAFILE = "NormalizationTest.txt"
  10 TESTDATAURL = "http://www.unicode.org/Public/" + unidata_version + "/ucd/" + TESTDATAFILE
  11
  12 def check_version(testfile):
  13     hdr = testfile.readline()
  14     return unidata_version in hdr
  15
  16 class RangeError(Exception):
  17     pass
  18
  19 def NFC(str):
  20     return normalize("NFC", str)
  21
  22 def NFKC(str):
  23     return normalize("NFKC", str)
  24
  25 def NFD(str):
  26     return normalize("NFD", str)
  27
  28 def NFKD(str):
  29     return normalize("NFKD", str)
  30
  31 def unistr(data):
  32     data = [int(x, 16) for x in data.split(" ")]
  33     for x in data:
  34         if x > sys.maxunicode:
  35             raise RangeError
  36     return u"".join([unichr(x) for x in data])
  37
  38 class NormalizationTest(unittest.TestCase):
  39     def test_main(self):
  40         part = None
  41         part1_data = {}
  42         # Hit the exception early
  43         try:
  44             testdata = open_urlresource(TESTDATAURL, check_version)
  45         except (IOError, HTTPException):
  46             self.skipTest("Could not retrieve " + TESTDATAURL)
  47         for line in testdata:
  48             if '#' in line:
  49                 line = line.split('#')[0]
  50             line = line.strip()
  51             if not line:
  52                 continue
  53             if line.startswith("@Part"):
  54                 part = line.split()[0]
  55                 continue
  56             if part == "@Part3":
  57                 # XXX we don't support PRI #29 yet, so skip these tests for now
  58                 continue
  59             try:
  60                 c1,c2,c3,c4,c5 = [unistr(x) for x in line.split(';')[:-1]]
  61             except RangeError:
  62                 # Skip unsupported characters;
  63                 # try atleast adding c1 if we are in part1
  64                 if part == "@Part1":
  65                     try:
  66                         c1 = unistr(line.split(';')[0])
  67                     except RangeError:
  68                         pass
  69                     else:
  70                         part1_data[c1] = 1
  71                 continue
  72
  73             # Perform tests
  74             self.assertTrue(c2 ==  NFC(c1) ==  NFC(c2) ==  NFC(c3), line)
  75             self.assertTrue(c4 ==  NFC(c4) ==  NFC(c5), line)
  76             self.assertTrue(c3 ==  NFD(c1) ==  NFD(c2) ==  NFD(c3), line)
  77             self.assertTrue(c5 ==  NFD(c4) ==  NFD(c5), line)
  78             self.assertTrue(c4 == NFKC(c1) == NFKC(c2) == \
  79                             NFKC(c3) == NFKC(c4) == NFKC(c5),
  80                             line)
  81             self.assertTrue(c5 == NFKD(c1) == NFKD(c2) == \
  82                             NFKD(c3) == NFKD(c4) == NFKD(c5),
  83                             line)
  84
  85             # Record part 1 data
  86             if part == "@Part1":
  87                 part1_data[c1] = 1
  88
  89         # Perform tests for all other data
  90         for c in range(sys.maxunicode+1):
  91             X = unichr(c)
  92             if X in part1_data:
  93                 continue
  94             self.assertTrue(X == NFC(X) == NFD(X) == NFKC(X) == NFKD(X), c)
  95
  96     def test_bug_834676(self):
  97         # Check for bug 834676
  98         normalize('NFC', u'\ud55c\uae00')
  99
 100
 101 def test_main():
 102     run_unittest(NormalizationTest)
 103
 104 if __name__ == "__main__":
 105     test_main()