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

[Ralph Corderoy]

Hi Ken,

> > Also, it SEGVs without the `/plain'.  Probably because
> > get_ctinfo()'s
> > 
> >     685     if (*cp != '/') {
> >     686         if (!magic)
> >     687             ci->ci_subtype = mh_xstrdup("");
> >     688         goto magic_skip;
> >     689     }
>
> You know ... I am looking at this now, and I cannot understand the
> logic here.

I didn't punt on it without good reason.  :-)
It's MIME too, so it's not as if the authors had a 70s'-induced drug
haze as a crutch.

It seems it's trying to allow subtype to be optional when *not* parsing
a user's composition, i.e. when it's a Content-Type header.  After all,
mhbuild(1)'s grammar makes subtype mandatory for `#text/plain ...'.  But
when there's no `/' we skip setting ci_subtype non-NULL.

 707     if (!*ci->ci_subtype) {
missing_subtype:
 708         inform("invalid %s: field in message %s (empty subtype for 
\"%s\")",
 709             TYPE_FIELD, ct->c_file, ci->ci_type);
 710         return NOTOK;
 711     }

Perhaps 686 should be `if (magic) goto missing_subtype?  :-)

-- 
Cheers, Ralph.
https://plus.google.com/+RalphCorderoy