Python 3.2 syntax error fixes in tests

This commit is contained in:
William S Fulton 2020-08-15 23:51:40 +01:00
parent c5e078c437
commit b1c0145fd3
2 changed files with 24 additions and 18 deletions

View File

@ -5,13 +5,16 @@ test_bytes = b"hello \x01world\x99"
BYTES = b"BYTES"
if sys.version_info[0:2] < (3, 0):
test_unicode = u"h\udce9llo w\u00f6rld"
UNICODE = u"UNICODE"
type_unicode_string = type(u"")
# Python 3.0-3.2 results in a SyntaxError when using u"" string literals, so we use a
# convoluted unicode string construction using unicode() and unichr().
# Conventional Python 2 syntax shown in comments.
test_unicode = unicode("h" + unichr(0xdce9) + "llo w" + unichr(0x00f6) + "rld") # u"h\udce9llo w\u00f6rld"
UNICODE = unicode("UNICODE")
type_unicode_string = type(UNICODE)
else:
test_unicode = "h\udce9llo w\u00f6rld"
UNICODE = "UNICODE"
type_unicode_string = type("")
type_unicode_string = type(UNICODE)
# Test that byte string inputs and outputs work as expected
bdbl = python_strict_unicode.double_str(test_bytes)

View File

@ -15,21 +15,24 @@ def check(s1, s2):
# Testing SWIG_PYTHON_2_UNICODE flag which allows unicode strings to be passed to C
if sys.version_info[0:2] < (3, 0):
# Python 3.0-3.2 results in a SyntaxError when using u"" string literals, so we use a
# convoluted unicode string construction using unicode() and unichr().
# Conventional Python 2 syntax shown in comments.
check(unicode_strings.charstring("hello1"), "hello1")
check(unicode_strings.charstring(str(u"hello2")), "hello2")
check(unicode_strings.charstring(u"hello3"), "hello3")
check(unicode_strings.charstring(str(unicode("hello2"))), "hello2") # u"hello2"
check(unicode_strings.charstring(unicode("hello3")), "hello3") # u"hello3"
check(unicode_strings.charstring(str("hello4")), "hello4")
unicode_strings.charstring(u"hell\xb05")
unicode_strings.charstring(u"hell\u00f66")
low_surrogate_string = u"\udcff"
unicode_strings.charstring(unicode("hell" + unichr(0xb0) + "5")) # u"hell\xb05"
unicode_strings.charstring(unicode("hell" + unichr(0x00f6) +"6")) # u"hell\u00f66"
low_surrogate_string = unichr(0xdcff) # u"\udcff"
unicode_strings.instring(low_surrogate_string)
else:
low_surrogate_string = "\udcff"
try:
unicode_strings.instring(low_surrogate_string)
# Will succeed with Python 2
except TypeError as e:
# Python 3 will fail the PyUnicode_AsUTF8String conversion resulting in a TypeError.
# The real error is actually:
# UnicodeEncodeError: 'utf-8' codec can't encode character '\udcff' in position 0: surrogates not allowed
pass
try:
unicode_strings.instring(low_surrogate_string)
raise RuntimeError("Exception should have been thrown")
except TypeError as e:
# Python 3 will fail the PyUnicode_AsUTF8String conversion resulting in a TypeError.
# The real error is actually:
# UnicodeEncodeError: 'utf-8' codec can't encode character '\udcff' in position 0: surrogates not allowed
pass