Sieve Email Filtering: Body Extension
RFC 5173

Note: This ballot was opened for revision 09 and is now closed.

(Lisa Dusseault) Yes

(Jari Arkko) No Objection

Comment (2008-01-10 for -** No value found for 'p.get_dochistory.rev' **)
No email
send info
Review by Christian Vogt:

Draft-ietf-sieve-body-07 extends the Sieve email processing language by
a command for searching a text string in the body of an email.

The document is concise and (with few exceptions mentioned below) clear,
and should therefore considered ready for publication once the comments
below have been addressed.


Section 3 specifies that a search for an empty string should return
"false" for emails that do not have a body, and "true" for emails that
have an empty body.  I believe that such different handling of similar
inputs (no body vs. empty body) may be confusing for users who are not
Sieve experts, and hard to debug for Sieve expert.  Unless the search
result cannot be the same for both cases for substantial reasons, the
search results should be redefined to be the same.  If there is
substantial reasons for the search results to be different, then the
draft should provide rationale for such distinction.


(1)  Section 2, 3rd paragraph:  What is a "capability string"?  If this
is an expression already defined in [SIEVE], then [SIEVE] should be
cited near the occurrence of "capability string".  Otherwise,
"capability string" should be defined in this document.

(2)  Section 4.3, 2nd paragraph:  Unstated what the difference between a
body transform of ':text' and a body transform of ':content "text"' is.
 The difference can be derived by the reader, of course, but the
document is incomplete by leaving it unstated.

(3)  Section 5, 2nd paragraph:  s/wild card/wildcard/ and s/set

(Ron Bonica) No Objection

(Ross Callon) No Objection

(Lars Eggert) No Objection

(Russ Housley) No Objection

Comment (2008-01-07 for -** No value found for 'p.get_dochistory.rev' **)
No email
send info
  From Gen-ART Review by Elwyn Davies; however, I think this is even
  more important than indicated by Elwyn.  Elwyn said:

  A pointer is given to the SIEVE base document, but I wonder if this
  document should note that ABNF is used to specify syntax and provide
  the appropriate reference.

(Cullen Jennings) No Objection

Comment (2008-01-10 for -** No value found for 'p.get_dochistory.rev' **)
No email
send info
I like this one.

(Chris Newman) No Objection

(Jon Peterson) No Objection

(Tim Polk) (was No Record, Discuss) No Objection

(Dan Romascanu) No Objection

(Mark Townsley) No Objection

(David Ward) No Objection

Magnus Westerlund No Objection