Excerpts from mail: 20-Mar-93 Re: Multipart/Mixed and Com.. Dave
Crocker(_at_)Mordor(_dot_)Stan (497)
The presence or absence of a line-terminator, within a text body part,
certainly is relevant ... to the internals of that body part. My point
was that it should/must have nothing to do with overall interpretation
of the MESSAGE structure. I further believe that using such a subtle
difference for determining highly significant difference in handle that
body part is very ill-advised, particular when the interpretation is not
a natural part of the semantics of that body part.
I find this comment rather hard to fathom. Imagine that you have two
messages, message A an message B, each of which has two parts, text
followed by image, inside of multipart/mixed. The image parts are
identical. The text parts are identical except that message A's text
part ends with a CRLF, while message B's text part does not.
Now, these messages are not identical, right? They have one discernable
difference, which is whether or not their text ends with a CRLF. I can
understand an implementation choosing to ignore this difference,
particularly if the implementation treats the parts as totally separate
entities visually. But if the system is like Andrew, rich enough to
display compound media objects in a variety of ways, it seems like
common sense that in Andrew the image is shown as being right at the end
of the text (that is, on the same line) for message B, but on a new line
in message A. If you're telling me not to display it that way, what
you're really telling me is to pretend the messages ARE identical when
they aren't. I can't imagine any other way of reflecting the
difference, and it IS a useful difference, and it conforms completely
with the MIME spec. (Indeed, that was a major reason that the boundary
mechanism was defined the way it was!)
Further, this has NOTHING to do with the message structure, as you
imply. It has to do with the text's internal structure. The idea that
separate objects are always displayed on separate lines reflects the
implementation structure some systems, but not all of them, and reflects
nothing at all abut the structure of a multipart message. If what I get
is a multipart containing non-newline-terminated text followed by an
image, I'm just displaying it as such. It seems to me it would be a
mistake to ADD a newline between them if I don't absolutely have to for
implementation reasons!
I think you're on very shaky ground here. I certainly never thought
that the multipart/mixed model was "a series of parts to be presented in
sequence ON SEPARATE LINES." Implementations can show them, in
sequence, however they think best. Andrew happens to be smart enough to
be able to reflect the presence or absence of a trailing newline up to
the user, and there's nothing wrong with that!