def none_cmp(xx, yy):
x = xx[1]
y = yy[1]
if x is None and y is None:
# No sort_key needed here, because defaults are ascii
try:
return cmp(xx[2], yy[2])
except TypeError:
return 0
if x is None:
return 1
if y is None:
return -1
if isinstance(x, string_or_bytes) and isinstance(y, string_or_bytes):
x, y = sort_key(force_unicode(x)), sort_key(force_unicode(y))
try:
c = cmp(x, y)
except TypeError:
c = 0
if c != 0:
return c
# same as above -- no sort_key needed here
try:
return cmp(xx[2], yy[2])
except TypeError:
return 0
def compare_to_other(self, other):
a = cmp(self.base, other.base)
if a != 0:
return a
cx, cy = self.comments_len, other.comments_len
if cx and cy:
t = (cx + cy) / 20
delta = cy - cx
if abs(delta) > t:
return -1 if delta < 0 else 1
return cmp(self.extra, other.extra)
def compare_to_other(self, other):
a = cmp(self.base, other.base)
if a != 0:
return a
cx, cy = self.comments_len, other.comments_len
if cx and cy:
t = (cx + cy) / 20
delta = cy - cx
if abs(delta) > t:
return -1 if delta < 0 else 1
return cmp(self.extra, other.extra)
def compare_to_other(self, other):
if self.is_finished != other.is_finished:
return 1 if self.is_finished else -1
if self.start_time is None:
if other.start_time is None: # Both waiting
return cmp(other.id, self.id)
return 1
if other.start_time is None:
return -1
# Both running
return cmp((other.start_time, id(other)), (self.start_time, id(self)))
def compare_to_other(self, other):
if self.is_finished != other.is_finished:
return 1 if self.is_finished else -1
if self.start_time is None:
if other.start_time is None: # Both waiting
return cmp(other.id, self.id)
return 1
if other.start_time is None:
return -1
# Both running
return cmp((other.start_time, id(other)), (self.start_time, id(self)))
def test_sorting(self):
' Test the various sorting APIs '
german = '''Sonntag Montag Dienstag Januar Februar März Fuße Fluße Flusse flusse fluße flüße flüsse'''.split()
german_good = '''Dienstag Februar flusse Flusse fluße Fluße flüsse flüße Fuße Januar März Montag Sonntag'''.split()
french = '''dimanche lundi mardi janvier février mars déjà Meme deja même dejà bpef bœg Boef Mémé bœf boef bnef pêche pèché pêché pêche pêché'''.split()
french_good = '''bnef boef Boef bœf bœg bpef deja dejà déjà dimanche février janvier lundi mardi mars Meme Mémé même pèché pêche pêche pêché pêché'''.split() # noqa
# Test corner cases
sort_key = icu.sort_key
s = '\U0001f431'
self.ae(sort_key(s), sort_key(s.encode(sys.getdefaultencoding())), 'UTF-8 encoded object not correctly decoded to generate sort key')
self.ae(s.encode('utf-16'), s.encode('utf-16'), 'Undecodable bytestring not returned as itself')
self.ae(b'', sort_key(None))
self.ae(0, icu.strcmp(None, b''))
self.ae(0, icu.strcmp(s, s.encode(sys.getdefaultencoding())))
# Test locales
with make_collation_func('dsk', 'de', func='sort_key') as dsk:
self.ae(german_good, sorted(german, key=dsk))
with make_collation_func('dcmp', 'de', template='_strcmp_template') as dcmp:
for x in german:
for y in german:
self.ae(cmp(dsk(x), dsk(y)), dcmp(x, y))
with make_collation_func('fsk', 'fr', func='sort_key') as fsk:
self.ae(french_good, sorted(french, key=fsk))
with make_collation_func('fcmp', 'fr', template='_strcmp_template') as fcmp:
for x in french:
for y in french:
self.ae(cmp(fsk(x), fsk(y)), fcmp(x, y))
with make_collation_func('ssk', 'es', func='sort_key') as ssk:
self.assertNotEqual(ssk('peña'), ssk('pena'))
with make_collation_func('scmp', 'es', template='_strcmp_template') as scmp:
self.assertNotEqual(0, scmp('pena', 'peña'))
for k, v in iteritems({u'pèché': u'peche', u'flüße':u'Flusse', u'Štepánek':u'ŠtepaneK'}):
self.ae(0, icu.primary_strcmp(k, v))
# Test different types of collation
self.ae(icu.primary_sort_key('Aä'), icu.primary_sort_key('aa'))
self.assertLess(icu.numeric_sort_key('something 2'), icu.numeric_sort_key('something 11'))
self.assertLess(icu.case_sensitive_sort_key('A'), icu.case_sensitive_sort_key('a'))
self.ae(0, icu.strcmp('a', 'A'))
self.ae(cmp('a', 'A'), icu.case_sensitive_strcmp('a', 'A'))
self.ae(0, icu.primary_strcmp('ä', 'A'))
def test_sorting(self):
' Test the various sorting APIs '
german = '''Sonntag Montag Dienstag Januar Februar März Fuße Fluße Flusse flusse fluße flüße flüsse'''.split()
german_good = '''Dienstag Februar flusse Flusse fluße Fluße flüsse flüße Fuße Januar März Montag Sonntag'''.split()
french = '''dimanche lundi mardi janvier février mars déjà Meme deja même dejà bpef bœg Boef Mémé bœf boef bnef pêche pèché pêché pêche pêché'''.split()
french_good = '''bnef boef Boef bœf bœg bpef deja dejà déjà dimanche février janvier lundi mardi mars Meme Mémé même pèché pêche pêche pêché pêché'''.split() # noqa
# Test corner cases
sort_key = icu.sort_key
s = '\U0001f431'
self.ae(sort_key(s), sort_key(s.encode(sys.getdefaultencoding())), 'UTF-8 encoded object not correctly decoded to generate sort key')
self.ae(s.encode('utf-16'), s.encode('utf-16'), 'Undecodable bytestring not returned as itself')
self.ae(b'', sort_key(None))
self.ae(0, icu.strcmp(None, b''))
self.ae(0, icu.strcmp(s, s.encode(sys.getdefaultencoding())))
# Test locales
with make_collation_func('dsk', 'de', func='sort_key') as dsk:
self.ae(german_good, sorted(german, key=dsk))
with make_collation_func('dcmp', 'de', template='_strcmp_template') as dcmp:
for x in german:
for y in german:
self.ae(cmp(dsk(x), dsk(y)), dcmp(x, y))
with make_collation_func('fsk', 'fr', func='sort_key') as fsk:
self.ae(french_good, sorted(french, key=fsk))
with make_collation_func('fcmp', 'fr', template='_strcmp_template') as fcmp:
for x in french:
for y in french:
self.ae(cmp(fsk(x), fsk(y)), fcmp(x, y))
with make_collation_func('ssk', 'es', func='sort_key') as ssk:
self.assertNotEqual(ssk('peña'), ssk('pena'))
with make_collation_func('scmp', 'es', template='_strcmp_template') as scmp:
self.assertNotEqual(0, scmp('pena', 'peña'))
for k, v in iteritems({u'pèché': u'peche', u'flüße':u'Flusse', u'Štepánek':u'ŠtepaneK'}):
self.ae(0, icu.primary_strcmp(k, v))
# Test different types of collation
self.ae(icu.primary_sort_key('Aä'), icu.primary_sort_key('aa'))
self.assertLess(icu.numeric_sort_key('something 2'), icu.numeric_sort_key('something 11'))
self.assertLess(icu.case_sensitive_sort_key('A'), icu.case_sensitive_sort_key('a'))
self.ae(0, icu.strcmp('a', 'A'))
self.ae(cmp('a', 'A'), icu.case_sensitive_strcmp('a', 'A'))
self.ae(0, icu.primary_strcmp('ä', 'A'))
def none_cmp(xx, yy):
x = xx[1]
y = yy[1]
if x is None and y is None:
# No sort_key needed here, because defaults are ascii
return cmp(xx[2], yy[2])
if x is None:
return 1
if y is None:
return -1
if isinstance(x, string_or_bytes) and isinstance(y, string_or_bytes):
x, y = sort_key(force_unicode(x)), sort_key(force_unicode(y))
c = cmp(x, y)
if c != 0:
return c
# same as above -- no sort_key needed here
return cmp(xx[2], yy[2])
def pynocase(one, two, encoding='utf-8'):
if isbytestring(one):
try:
one = one.decode(encoding, 'replace')
except:
pass
if isbytestring(two):
try:
two = two.decode(encoding, 'replace')
except:
pass
return cmp(one.lower(), two.lower())
def pynocase(one, two, encoding='utf-8'):
if isbytestring(one):
try:
one = one.decode(encoding, 'replace')
except:
pass
if isbytestring(two):
try:
two = two.decode(encoding, 'replace')
except:
pass
return cmp(one.lower(), two.lower())
def two_args(a, b):
nonlocal func
if func is None:
func = getattr(collator_function(), func_name)
try:
return func(a, b)
except TypeError:
if isinstance(a, bytes):
try:
a = a.decode(sys.getdefaultencoding())
except Exception:
return cmp(a, b)
elif a is None:
a = ''
if isinstance(b, bytes):
try:
b = b.decode(sys.getdefaultencoding())
except Exception:
return cmp(a, b)
elif b is None:
b = ''
return func(a, b)
def icu_collator(s1, s2):
return cmp(sort_key(force_unicode(s1, 'utf-8')),
sort_key(force_unicode(s2, 'utf-8')))
def icu_collator(s1, s2):
return cmp(sort_key(force_unicode(s1, 'utf-8')),
sort_key(force_unicode(s2, 'utf-8')))
def compare_to_other(self, other):
for i, ascending in enumerate(self.orders):
ans = cmp(self.values[i], other.values[i])
if ans != 0:
return ans * ascending
return 0
def compareFileStream(file1, file2):
return cmp(file1.fileName, file2.fileName)
