Kevin Ryde wrote:
On the hostname bit, would the change below be right for utf8 hostnames per the docs -- even if my wondering about non-ascii is answered by the fact the two funcs currently reject non-ascii entirely!
The XS changes look fine to me, so I just committed them. Thanks!
(The filename.t passes with or without the xs change, it just avoids the dreaded never-executed code ... :)
The tests that are currently there are deliberately permissive in checking the actual values of hostnames and such because that was the only way to get them passing on MSWin32. The original tests were more like the ones you propose now, but we had to change them.
So, apart from being more thorough, do your test cases also cover more binding code than the existing tests? I don't think so, but maybe I'm missing something.