while working on a fix for bug #80259, I noticed that
mb_decode_mimeheader() collapse all whitespace
after a case folding marker (CRLF) to a single space. This appears
to be in violation of RFC 5322, section 2.2.3:
| Unfolding is accomplished by simply removing any CRLF
| that is immediately followed by WSP.
imap_utf8() properly keeps the whitespace. (ext/imap is not available
I wonder whether it's worth to fix it, due to the obvious BC break, or
whether we want to treat that as a documentation issue. If we're going
for a fix (what I would suggest), which version should that target?
Christoph M. Becker