Re: [Nmh-workers] mhfixmsg on a pathological mail

[Ken Hornstein]

>I've reproduced similar symptoms with an mhbuild-produced email where I
>tacked on CR to the end of every line after the blank line after the
>headers.  IOW, without the headers having CR.  However, I still think
>this is an invalid format email;  why should we handle CR at any
>location?

Sigh.  I'm of two minds about that.  Technically those are valid RFC 5322
messages; our wrinkle is that we techically support RFC 5322 messages with
Unix line endings.  But that is pretty standard on Unix systems, and this
was definitely not a normal use case.

>The content shouldn't be truncated without an error exit
>status though.

As I see it (the strcmp() is failing, right?) it probably misses the
first boundary part and counts the whole message as a preamble, right?
I do not think throwing an error for not finding a MIME boundary in the
message would be a good idea, at least in the normal tools.  I could see
mhfixmsg maybe complaining about it.

--Ken