On 3/9/07, Malcolm Tredinnick <[EMAIL PROTECTED]> wrote:
>
> On Thu, 2007-03-08 at 04:09 -0800, Simon G. wrote:
> > Good Evening Djuggernauts,
> >
> > I've spent some time looking over the email related tickets in Trac,
> > and think I've got them sorted out a bit. I'm hoping to spur some
> > discussion on some of these, so here's a brief recap.
>
> Nice work, again, Simon.

Ditto. This sort of filtering isn't easy - good stuff.

> > We currently have requests to:
> >
> > #1541 - Add Multipart Message capability to mail
> >       This ticket's first birthday is creeping up. There's a patch here
> > that adds a safe method for sending multipart messages. Jacob has said
> > this is great, but it needs docs. HOWEVER, Gary Wilson's pointed out
> > that we're getting a rather ugly data tuple going on, and created 3366
>
> The main reason I suspect it hasn't been resolved one way or the other
> is in Russell's comment #23 in early March: it's become a nest of
> competing implementations, which is never an easy thing to untangle.
> Having discussion in bugs is good, but it sometimes ends up like that.

There have been a few attempts to get a mailing list discussion
started in the past few months, but nobody has taken a serious swing
at the problem.

> I haven't read the patches closely, but my immediate concern is whether
> we are duplicating too much of Python's existing mail framework and just
> wrapping it. That would be bad design. I have no opinion on this yet.

Looking over the patches again, I share this concern. The original
intent of send_mail was to provide a simple "just send a message" API,
mostly for the purposes of emailing the site admin error messages.
James' comment about mailing perty HTML stack traces is a good one,
but otherwise the python email libs provide a much richer API, with
exposure of the full set of possible email options.

I'm +1 on exposing some extra options that fit with existing options
(like SSL sending and BCC's - 2897,3366), and the problems with
charset encoding should probably be fixed, too (2007,3472).
Personally, I'm not a big fan of HTML email, but I seem to be in the
minority on this one; as long as the interface (1) remains simple, and
(2) enforces best practice (such as requiring text as well has HTML),
I could live with the change. Some cleanup of option passing might be
called for as well (3366, but I prefer Malcolms dict-based approach).
However, over all these changes, we need to be careful that we don't
just replicate the python mail APIs.

Yours
Russ Magee %-)

--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups 
"Django developers" group.
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/django-developers?hl=en
-~----------~----~----~----~------~----~------~--~---

Reply via email to