🚀 go-pugleaf

On Tuesday, in article <20070724.1449.7145snz@flashy.nut>
     sum1@flashy.nut "Ian" wrote:

> Bill@lyons.demon.co.uk wrote:
> 
> > I am getting fed up with the increasing number of emails which are
> > apparently text-only but actually encoded "quoted-printable".  They show
> > fine when displayed via PCOak, but when you save them and open them in a
> > text editor, or when you try and forward them, the dreaded "=20" instead
> > of space between every word becomes apparent and they become unreadable.

> I'll leave that to  the PCOAK experts. I don't think I've ever seen the =20
> between words,  perhaps I'm not saving much e-mail to a file.

Only some (extremely badly-written) software does it; I haven't yet
worked out exactly what.  Most software does q-p encoding properly,
only encoding spaces when they occur at the ends of lines (as required
by the RFC).  This is fairly easy to deal with.

> I do have one little mystery which happens only in one correspondent's
> messages.  Where he has typed the apostrophe  ',  it appears in both  mail
> reader (READIT.EXE) and when I do a Reply using my editor TSE.EXE.
> Here are two of the words - I don't know what will happen to them after
> going through  the  nntp system.
> IN case they vanish or something, it is the Greek character thatis an A
> and an E joined together.  I think the aSCII number is 92.
>  ...
>  hasn�t
>  There�s
> ...

It is, indeed, 92 hex (decimal 146) -- in the Windows-1252 character
set, this is a "high 9 single quote".  It appears as a capital "AE"
diphthong in DOS codepages 437 and 850.

Unfortunately some (mostly Micro$oft) software tries to be "clever"
with apostrophes and puts them in as closing single quotes (0x92)
rather than normal ASCII apostrophes; you sometimes get the same thing
with double quotes as well (characters 0x93 and 0x94 as the opening
and closing pairs).  While this is fine for word processing software,
it's a rather silly thing to do (IMNSHO) with e-mail!

When I get back to PCOak, among the many things I'm intending to do is
put in handling for the common character sets (Windows-1252,
ISO-8859-1 (which is a subset of Windows-1252) etc.) so that
non-7bit-ASCII characters in incoming messages get replaced by their
nearest equivalent in the local code page before being displayed,
quoted in replies etc.  PCOak's handling of MIME messages leaves a lot
to be desired.

-- 
Simon Turner                                    DoD #0461
simon@twoplaces.co.uk
    Trust me -- I know what I'm doing!          -- Sledge Hammer