gh-117557: Improve error messages when a string, bytes or bytearray of length 1 are expected

[serhiy-storchaka]

• Issue: unicodedata: unicodedata.name() undocumented TypeError on empty string #117557

[erlend-aasland]

Nice; thanks, Serhiy!

I left some comments.

[erlend-aasland]

Right, %T, not %N. It's too new 😆 re. 054f12f

[erlend-aasland]

🚀

[serhiy-storchaka]

I am not sure about articles. It is a puzzle to me whether they should be used, and whether not.

[erlend-aasland]

I think these suggestions are correct, but I'd like them to be verified by a native speaker.

[serhiy-storchaka]

@terryjreedy, could you please take a look?

[terryjreedy]

I checked the tests non-clinic python code carefully and tried the check the logic of the rest. It seems that there are more error cases that error tests, so I wonder if all were covered in a test.

Over all I really like the patch.

[bedevere-app]

When you're done making the requested changes, leave the comment: I have made the requested changes; please review again.

[serhiy-storchaka]

I have made the requested changes; please review again.

Although I am not sure that I have made them all. There is a variety in wording, and some examples look contradictory.

"a unicode character", "str of length 1", "string of length 1" -- refers to str of length 1.
"one character bytes", "a single byte", "bytes of length 1", "a byte string of length 1", "bytes object of length 1" -- refers to bytes (or bytearray) of length 1.

[bedevere-app]

Thanks for making the requested changes!

@erlend-aasland, @terryjreedy: please review the changes made to this pull request.

[serhiy-storchaka]

@terryjreedy, could you please take another look?