Mail and text extraction

Peter Hansen peter at
Mon Apr 21 13:52:11 CEST 2003

Dino Levy wrote:
> On Mon, 21 Apr 2003 03:47:26 +0200, Peter Hansen wrote:
> > Why not download the whole message, then scan it after downloading? You
> > provide no good reason, so far, for why you wouldn't just do this much
> > simpler step.  Keep the design simple, modular, clean, and you can do
> > most of what you want with simple, modular, existing solutions in
> > standard modules.  Why wouldn't you do that?
> >
> > -Peter
> Well yes, that's antoher way of doing it, but I wanted to make it done in
> one breath, like connecting and immediately parsing the text, but it's
> obviously not achievable:)

Oh, it's quite achievable that way.  It's just the wrong way to do it.

Mixing in the various operations like that is a form of optimization,
and since nothing's actually working yet it's clearly premature, and
therefore not good. :-)

> I don't want MIME, I really need POP3 connections...
> Ok, thanks anyway

MIME and POP3 are orthogonal.  MIME is an encoding, POP3 is a
transport protocol (or something).  You can have one, or the other,
or both, or neither...

By the way, the "quite achievable" part above simply means you could
easily build this capability into custom software.  The "wrong way"
just means that would likely be a large waste of time, given that 
the individual pieces already exist in standard modules, quite well
tested and debugged.  Just combine them in appropriate ways and keep
life simple.


More information about the Python-list mailing list