From: "Mike Taylor" <[log in to unmask]>
> ? Do we have any precedent for true/false values in SRU extension
> parameters?
No, there's no precedent, for SRU, and our spec is silent.
It seems to be legal, from rfc3986.
http://rfc.net/rfc3986.html
query = *( pchar / "/" / "?" )
pchar = unreserved / pct-encoded / sub-delims / ":" / "@"
and prose:
'However, as query components are often used to carry identifying
information in the form of "key=value" pairs ' .....
"often" implying "not always".
So what are the implications of writing this option into the spec (for
version 1.2)?
(Personally, I hate having to encode "key=value" when there's only one
possible value. I hate, for example, having to define xml attributes like
'collection="yes"' ' when "yes" is the only possible value, and I think one
of the big failings of xml is not allowing valueless attributes, so I'm all
for doing this for SRU query components.)
--Ray
|