Is it considered bad practice for a canonical link on a page return a 302 redirect to a login page?
Example:
If https://foo.com/bar/slug?version=2019-02-01&signature=XXX returns
HTTP/1.1 200 OK
<!DOCTYPE html>
<html>
<head>
<link rel="canonical" href="https://foo.com/bar/slug" />
</head>
<body>
...
</body>
</html>
but https://foo.com/bar/slug returns
HTTP/1.1 302 Found
Location: https://foo.com/login?ReturnUrl=%2Fbar%2Fslug
A) Is this considered bad practice?
B) If the url with the querystring returns the 200 will it's content still be indexed if a search engine crawls it?
c) If a search engine crawls multiple variants of the querystring version (/bar/slug?version=2019-02-01&signature=XXX, /bar/slug?version=2019-02-02&signature=YYY, /bar/slug?version=2019-02-03&signature=ZZZ) will search engines only "keep" the most recent version (assuming I set Last-Modified headers to match the date in the querystring version and they all have the same canonical link) indexed?
