Thus said Ken Hornstein on Wed, 06 Nov 2024 19:16:33 -0500:

> As for nmh being wrong, well ...  it cannot handle email that it seems
> literally every  other piece  of email software  seems to  handle just
> fine. What does that tell you?

Well, I don't know that nmh is  the only piece of software that fails to
handle it. Given that the RFC  is pretty explicit about what is allowed,
anything outside of those bounds is effectively undefined behavior.

In some  forums where I have  reported the problem, others  seem to have
had similar complaints that they don't  recognize as being caused by the
sender.  For  example, one  person  complains  that  the email  shows  a
corrupted attachment. Well,  that's just the kind of thing  that I would
expect from software that accepts the  email and then truncates the line
because it violates the RFC. Others  complain that they simply don't get
the  email. That  suggests to  me  that they  use an  MTA that  outright
rejects the email due to the violation.

There may be some hope yet... Gmail does block certain violations of RFC
5322:

https://support.google.com/a/answer/13567860?hl=en

If they're willing to block crappy software that adds duplicate headers,
why not software  that adds crappy malformed lines of  base64 data which
is just as much  a violation, and indeed one that  may be more malicious
in nature than just duplicate headers?

Andy


Reply via email to