Bruce Lilly writes:
Comments are incompatible with URIs unless some quoting mechanism is used.
That is because URIs may contain parentheses, as in
http://users.erols.com/blilly/(foo)(bar)
See RFC2396 for URI syntax details. RFC 2369 quotes URIs in angle brackets,
which are not themselves allowed in URIs.
Well this is not the first time someone has proposed a structured header
with a URI in it, surely? So what has been done about this on previous
occasions? Parentheses in URIs are not all that common, so they could be
escaped. Or else the whole thing could be put inside a quoted-string.
Using angle-brackets would doubtless also be a good idea, though I do not
think that on its own would allow comments to appear outside of the angle
brackets with parentheses inside being treated differently.
One issue to consider is long URIs. If comments are disallowed, no quoting
is required. URIs cannot contain whitespace characters, so a simple way to
handle a long URI is to allow a URI to be line-folded; it can be reconstructed
by unfolding and eliding any whitespace.
Long URIs are a pain in message/article bodies too. Is there any
possibility that those responsible to fixing the format of URIs could
include a folding mechansim? I do not like the idea of fixing what is a
more widespread URI problem within just a few particular email headers.
Another possibility is some adaptation of the mechanism used for long
parameters in RFC 2231. Not that I am enamoured of the particular solution
provided there, but I am even less enamoured of inventing YAFM.
--
Charles H. Lindsey ---------At Home, doing my own thing------------------------
Tel: +44 161 436 6131 Fax: +44 161 436 6133 Web: http://www.cs.man.ac.uk/~chl
Email: chl(_at_)clerew(_dot_)man(_dot_)ac(_dot_)uk Snail: 5 Clerewood Ave,
CHEADLE, SK8 3JU, U.K.
PGP: 2C15F1A9 Fingerprint: 73 6D C2 51 93 A0 01 E7 65 E8 64 7E 14 A4 AB A5