On Fri, 28 Mar 2008, Dan Karp wrote:
> Is there a generally-accepted server response when a client requests FETCH
> BODY[n.HEADER] on a non-message subpart?
Yes.
The result for any BODY[...] that does not actually exist is a zero-length
string. Normally, this is a client error (since the client can determine
what parts exist from BODYSTRUCTURE). However, at times it is convenient
for the client to do a mindless access, typically for BODY[1], so the
result is a zero-length string rather than a BAD.
-- Mark --
http://staff.washington.edu/mrc
Science does not emerge from voting, party politics, or public debate.
Si vis pacem, para bellum.