[mimelib-devel] Re: Still same bug even with email ver. 1.2

[Ben Gertzfield]

>>>>> "BAW" == Barry A Warsaw <barry at zope.com> writes:

    BAW> What I think we ought to do is to deprecate
    BAW> Message.add_payload() but leave its semantics unchanged,
    BAW> change Message.attach() to not be an alias for add_payload(),
    BAW> and define the semantics of attach() to always set the
    BAW> payload to a list.  Thus, the first time you call attach().
    BAW> I think attach() should retain the semantics that if the MIME
    BAW> type isn't multipart/* it should throw an exception (i.e. you
    BAW> can't attach() to a non-multipart; use set_type() first).

I don't really see any other way to deal with this without incurring
extra breakage.

    BAW> For Python 2.2, I /think/ I can fix
    BAW> Generator._handle_multipart() to watch out for subparts that
    BAW> are scalar[1].  If so, then the patch to the Parser described
    BAW> above probably ought to be backed out.  As long as the unit
    BAW> tests pass, I think we'd be okay.  This is probably a good
    BAW> patch to keep for Python 2.3, just to be safe.

That will be a good work around for the time being, IMHO.

Ben

-- 
Brought to you by the letters X and U and the number 17.
"It makes my nipples hard!"
Debian GNU/Linux maintainer of Gimp and Nethack -- http://www.debian.org/