[Top] [All Lists]

[Fwd: DISCUSS and COMMENT: draft-ietf-sieve-mime-loop]

2009-08-12 17:33:54

--- Begin Message ---
This is a good document overall, but there are a few issues that need to be 
addressed to
ensure consistency and interoperability of implementations.

(1) In section 3, the semantics of a "break" command that omits a name in a 
"foreverypart" is not clearly specified.  In this case, does the "break" 
terminate the closest 
enclosing loop, or the outermost?

(2) In section 4.1, the semantics of ":anychild" in the absence of the ":mime" 
tagged argument 
is not specified, but this construct is permitted by the ABNF and is not 
forbidden by the text.

Since sections 4.2 and 4.3 reference 4.1 for the behavior of ":anychild" this 
ambiguity ripples 

(3) In sections 4.2 and 4.3, the following text is ambiguous:

    "the use of "address" with no ":mime" and ":anychild" tagged arguments ..."

Does the no apply to both arguments or just one? if it applies to both, it 
might be clearer 
to state: 

    "the use of "address" when both the ":mime" and ":anychild" tagged arguments
     are omitted ..."

There is at least one occurrence of "SIEVE" in section 11, but most (all?) 
other occurrences are "Sieve"

For consistency, suggest the global change

--- End Message ---
<Prev in Thread] Current Thread [Next in Thread>
  • [Fwd: DISCUSS and COMMENT: draft-ietf-sieve-mime-loop], Alexey Melnikov <=