bpo-37476: Adding a unit test of unicode in test_unicode.py #14531
Conversation
Lib/test/test_unicode.py
Outdated
|
|
||
| self.assertEqual(unicode_asutf8('abc'), 'abc') | ||
| self.assertEqual(unicode_asutf8('abc\0'), 'abc') | ||
| self.assertEqual(unicode_asutf8('abc\0abc'), 'abc') |
There was a problem hiding this comment.
It's more like you are testing decodeUTF8. I'd suggest add more encode samples, BMP, non-BMP, just take from other tests.
Lib/test/test_unicode.py
Outdated
| def test_asutf8(self): | ||
| from _testcapi import unicode_asutf8 | ||
|
|
||
| self.assertEqual(unicode_asutf8('abc'), 'abc') |
There was a problem hiding this comment.
The function encodes to UTF-8, so the result should be a bytes string, not Unicode. Use PyString_FromString() rather than PyUnicode_FromString() in _testcapi.
There was a problem hiding this comment.
The others similar C API tests also return unicodes other than bytes which in some sense also tests the counter part decoding function, though I don't very understand why designed like this.
There was a problem hiding this comment.
Either fix other tests, or leave them unchanged. But I would prefer to not add new buggy tests :-)
There was a problem hiding this comment.
So returning the decoded bytes string is right way? I am not sure I understand it clearly.
There was a problem hiding this comment.
Return bytes, not unicodes. Test decode functionality separately.
np, I would continue to add test. |
|
FYI, I cancelled the Travis CI run to free some resources for upcoming 3.8b2 and 3.7.4rc2 releases. |
|
@vstinner hi, victor. Pls help me review this patch again, thanks ;) |
https://bugs.python.org/issue37476