Daniel Liston <dliston@sonny.org> wrote:
> Mike Oliver wrote:
> > Daniel Liston <dliston@sonny.org> wrote:
> >>Roger B.A. Klorese wrote:
> >>
> >>>B) In general, visible "=20" has to do with improperly setting the MIME
> >>>information on the HTML message.
> >>
> >>So you are saying that these lines should be added to the headers of the
> >>message being created, in order to prevent the =20?
> >>
> >>-MIME-Version: 1.0
> >>-Content-Type: text/html; charset="iso-8859-1"
> >>-Content-Transfer-Encoding: 7Bit
> >
> > More likely 'Content-Transfer-Encoding: quoted-printable'. =20 is the
> > quoted-printable representation of a space, ASCII 0x20.
>
> Thanks Mike, but I am confused. Is the "7bit" the cause and the
> "quoted-printable" the cure?
Yes. If the header declares that the message content is encoded as in
"quoted-printable" format then it's telling the recipient's mail reader
that any =20 sequences in the body of the message should be presented
to the user as spaces. (You're also telling the reader that a bunch of
other =XX sequences should be converted before presentation, =20 is
just one specific example of a quoted-printable encoding. For
instance, =A3 should be shown as the British pound sterling symbol.)
> Or the other way around?
No. If your Content-Transfer-Encoding header states that the content
is in 7Bit encoding (effectively meaning "the content of this message
is just plain ASCII text") then there's no reason for the recipient's
mail reader to do any conversion of any =20 sequences it finds in the
body of the message, it'll just show them to the user as the three
characters "=20".
> Frank did
> not provide any information about his mods for us to know which
> way he did it. When I search my spam folders, the only place I
> get HTML mail, I only see the =20 pattern in messages with the
> "Content-Transfer-Encoding: quoted-printable".
Right. But when you pull one of those messages into your mail program
it doesn't show you any "=20" sequences, it interprets the message body
in accordance with the message's "Content-Transfer-Encoding:
quoted-printable" header and shows you spaces instead.
> To be fair, I only eye-balled the first 5-10 out of several thousand.
> I did a search on =20 in the mail folder, then back tracked to the
> Encoding. When I searched on 7bit, and scrolled through the body of
> the message, I did not find/see any =20 patterns.
That's to be expected. Spaces in 7Bit-encoded messages are just
spaces. Any =20 sequences in a 7Bit-encoded messages would be there
because the message composer really did type "=20". That's not a
common thing for anyone to write so you shouldn't expect too see it
crop up often. It'll be there in this message.
Mike.
--
mike.oliver@sun.com
|
|
