Re: New mail-related routines

[Alexander Pohoyda]

Stefan Monnier <address@hidden> writes:

> > +  (let ((from (or from (point-min)))
> > +   (to (or to (point-max))))
> 
> I'd recommend
> 
>   (unless from (setq from (point-min)))
>   (unless to (setq to (point-max)))
> 
> it saves a bit of indentation and is marginally more efficient.

Good point, I'll change this everywhere.


> > +   (if (or (search-forward (concat "\n" mail-header-separator "\n") to t)
> > +           (search-forward "\n\n" to t))
> 
> This is less robust than what rfc822-goto-eoh uses, in the case where the
> mail-header-separator is modified.

Yes, you're right.  I'll reuse the original regexp from
rfc822-goto-eoh, but I need a function to return the point and don't
move the cursor.


> > +       (point)
> > +     ;; TODO: Shouldn't we return nil instead?
> > +     (message "This entity has no body")
> > +     to)))))
> 
> I'd argue we should return `to' because the whole thing is the header.

Sorry, I don't clearly understand you here.
Do you agree with returning `to'?

Thank you very much!


-- 
Alexander Pohoyda <address@hidden>
PGP Key fingerprint: 7F C9 CC 5A 75 CD 89 72  15 54 5F 62 20 23 C6 44