ietf-822
[Top] [All Lists]

Re: Multipart/Mixed and Compound Documents

1993-03-19 11:15:25
Excerpts from mail: 19-Mar-93 Re: Multipart/Mixed and Com.. "Terry
Crowley"@diamond. (989*)

In fact, I was arguing that the presence or absence of the blank line
*should not* have this significance.

A clarification by example:

Content-type: multipart/mixed; boundary=foo

--foo

This is a text body part that does NOT end with CRLF.
--foo

This is a text body part that DOES end with CRLF.

--foo--

There's no heuristic involved, and it seems pretty clear to me that
these two body parts do indeed implement a rather clear distinction. 
Whether or not a particular MIME reader notices this distinction is up
to the implementors; the fact that a text fragment doesn't end with a
CRLF doesn't mean you can't choose to display a CRLF as part of what
separates your body parts.  Andrew happens to use this for
objects-in-text-lines, but I at least have always known that other
viewers would display it differently.  (Heck metamail not only breaks
the line, it asks you a question and pops up a separate window if, for
example, the "in-line" object is a picture.)  -- Nathaniel