Issue #8670: PyUnicode_AsWideChar() and PyUnicode_AsWideCharString() replace
UTF-16 surrogate pairs by single non-BMP characters for 16 bits Py_UNICODE
and 32 bits wchar_t (eg. Linux in narrow build).
diff --git a/Lib/test/test_unicode.py b/Lib/test/test_unicode.py
index 2ac79fb..285b7af 100644
--- a/Lib/test/test_unicode.py
+++ b/Lib/test/test_unicode.py
@@ -1419,6 +1419,17 @@
self.assertEquals(size, 7)
self.assertEquals(wchar, 'abc\0def\0')
+ nonbmp = chr(0x10ffff)
+ if sizeof(c_wchar) == 2:
+ buflen = 3
+ nchar = 2
+ else: # sizeof(c_wchar) == 4
+ buflen = 2
+ nchar = 1
+ wchar, size = test_aswidechar(nonbmp, buflen)
+ self.assertEquals(size, nchar)
+ self.assertEquals(wchar, nonbmp + '\0')
+
# Test PyUnicode_AsWideCharString()
def test_aswidecharstring(self):
from _testcapi import test_aswidecharstring
@@ -1432,6 +1443,15 @@
self.assertEquals(size, 7)
self.assertEquals(wchar, 'abc\0def\0')
+ nonbmp = chr(0x10ffff)
+ if sizeof(c_wchar) == 2:
+ nchar = 2
+ else: # sizeof(c_wchar) == 4
+ nchar = 1
+ wchar, size = test_aswidecharstring(nonbmp)
+ self.assertEquals(size, nchar)
+ self.assertEquals(wchar, nonbmp + '\0')
+
def test_main():
support.run_unittest(__name__)