Re: i94, was: Does Reason-Phrase allow LWS?

Julian Reschke wrote:
 
> Frank Ellermann wrote back in December:
>> I think what you're looking for is:

>>    Reason-Phrase  = *( VCHAR / WSP )
 
> Fine with me.

<joke> Four months later, RFC 5198 hates HTAB </joke>
Seriously, maybe *( SP / VCHAR ) is now good enough.
 
> There's an overlap with issue 74

I don't understand section 5 in RFC 3987.  Are HTTP
implementors forced to grok IRI comparison ?  What
has this to do with I18N for <Reason-Phrase> ?  For
a say 404 the body can use any language and charset
it likes.

The 2822upd idea of <phrase> is more convoluted, it
is a sequence of <word>s, each word is an <atom> or
a <quoted-string>.

Ignoring <obs-phrase>, <obs-qtext>, and <obs-qp>
the <atom>s and <quoted-string>s are separated by
optional [CFWS].  But the F in CFWS is something we
don't want in <Reason-Phrase>.  

An advantage of the 2822upd <phrase> is that it is
covered by RFC 2047, but that is again designed to
allow folding, not okay for a HTTP <Status-Line>.  

Sanity check, we don't want folding there, right ?

The *TEXT in RFC 2616 is apparently a simplification
of <phrase>, with the twist of allowing Latin-1, or
anything else if 2047-encoded.  

The Latin-1 and MIME mixture in RFC 2616 is strange.
VCHAR is ASCII, for that I know what it means, you
can even throw in =?...?.?...?= constructs.

 Frank

Received on Thursday, 3 April 2008 15:43:47 UTC