This seems like a subject that should be taken up on an ietf list.
sipcore is the likely one, though that is not the place to deal with the
generic uri syntax.
ISTM that this was a screwup when 3986 was published and didn't address
backward compatibility with sip URIs.
Thanks,
Paul
On 1/12/15 9:28 AM, Doug Sauder wrote:
Brett, thanks for that helpful reference information.
ISTM that RFC 3986 eliminated the idea of an "opaque" part, so that a
generic
URI has the distinct parts scheme, authority, path, query, and
fragment. An authority
always starts with "//", so in the generic syntax, a SIP URI has no
authority and
always has a path.
Before the mailto URI was defined, email addresses were commonly written
as [email protected]. *IF* the mailto URI had been defined as
mailto://[email protected], then the SIP URI probably would have
followed
suit with sip://[email protected]. In that case, all would be right
with the
world and generic URI syntax. Too late for that now.
Having said that, I think that "?" and "/" should not be included in the
user
component of a SIP URI, in order to at least somewhat accommodate RFC 3986.
The "?" character delimits the query part, so it should be escaped if it
is not used as
that delimiter. The "/" could be a problem if it appears as in
sip://[email protected]. When "//" appears at the beginning, the
generic
parser interprets it as an authority instead of a path. If "/" is not
allowed in the
user component, then the interpretation of a generic parser would be
consistent for
every SIP URI.
I was not aware of the problem with syntax for IPv6 (discussed in the
thread you
referred to). That's a problem that I think requires changes to RFC 3986.
BTW, as another bit of information on this topic, RFC 6068 (Mailto URI)
has this
to say:
1. A number of characters that can appear in <addr-spec> MUST be
percent-encoded. These are the characters that cannot appear in
a URI according to [STD66] as well as "%" (because it is used for
percent-encoding) and all the characters in gen-delims except "@"
and ":" (i.e., "/", "?", "#", "[", and "]"). Of the characters
in sub-delims, at least the following also have to be percent-
encoded: "&", ";", and "=". Care has to be taken both when
encoding as well as when decoding to make sure these operations
are applied only once.
So perhaps a good SIP implementation should use percent encoding for
"?" and for "//" when it appears at the beginning of a user component.
--
Doug Sauder
On 2015-01-12, 7:18, Brett Tate wrote:
Hi,
And for completeness, RFC 2396 section 3 indicates the following.
"URI that do not make use of the slash "/" character for separating
hierarchical components are considered opaque by the generic URI
parser."
I mention it since some attempt to fit the sip-uri as a hier_part.
absoluteURI = scheme ":" ( hier_part | opaque_part )
opaque_part = uric_no_slash *uric
-----Original Message-----
From: Brett Tate [mailto:[email protected]]
Sent: Friday, January 09, 2015 4:10 PM
To: 'Doug Sauder'; 'sip-implementors'
Subject: RE: [Sip-implementors] SIP URI syntax vs. generic URI syntax
Hi Doug,
If I recall correctly, the conclusion of the following thread was that
nobody
cares about fixing any of the existing incompatibilities.
http://www.ietf.org/mail-archive/web/sip/current/msg26318.html
The following email shows why it might be valid (i.e. structured subset
of
opaque_part of RFC 2396 absoluteURI).
http://www.ietf.org/mail-archive/web/sip/current/msg26325.html
The following RFC 2396 section 3.1 snippet may be of interest.
"The URI syntax consists of a sequence of components separated by
reserved
characters, with the first component defining the semantics for the
remainder
of the URI string."
-----Original Message-----
From: [email protected] [mailto:sip-
[email protected]] On Behalf Of Doug Sauder
Sent: Friday, January 09, 2015 2:51 PM
To: sip-implementors
Subject: [Sip-implementors] SIP URI syntax vs. generic URI syntax
The SIP URI syntax in RFC 3261 is not compatible with the generic URI
syntax in RFC 3986. Specifically, the character "?" should not appear
in the user component. Arguably, the character "/"
also should not appear in the user component.
Surely this topic must have been discussed before now. Does anyone
have information about prior discussions, or other comments on this
topic? Has there been a previous attempt to make the SIP URI syntax
compatible with the generic URI syntax?
Thanks!
_______________________________________________
Sip-implementors mailing list
[email protected]
https://lists.cs.columbia.edu/mailman/listinfo/sip-implementors
_______________________________________________
Sip-implementors mailing list
[email protected]
https://lists.cs.columbia.edu/mailman/listinfo/sip-implementors
