[2.7] bpo-36742: Fix urlparse.urlsplit() error message for Unicode URL

[vstinner]

If urlparse.urlsplit() detects an invalid netloc according to NFKC
normalization, the error message type is now str rather than unicode,
and use repr() to format the URL, to prevent <exception str() failed>
when display the error message.

https://bugs.python.org/issue36742

[vstinner]

This change fix the bug described at: https://bugs.python.org/issue36742#msg344981

Script:

import urlparse as urllib_parse
url = u'example.com\uFF03@bing.com'
url = u'http://\u30d5\u309a\ufe1380'
urllib_parse.urlsplit(url)

Output without the fix:

ValueError: <exception str() failed>

Output with the fix:

ValueError: netloc u'\u30d7:80' contains invalid characters under NFKC normalization

[vstinner]

cc @stratakis @encukou

[orsenthil]

lgtm.

[vstinner]

cc @tirkarthi @zooba @gpshead

[tirkarthi]

Should netloc be reported in the error message? Python 3 and previous error messages report netloc.

[vstinner]

Oh. You're right. I also fixed this bug and I updated the unit test to test this case.

[tirkarthi]

LGTM. Thanks.

[vstinner]

Thanks for review @tirkarthi and @orsenthil!