[SLUG] Irritating code instead of normal characters

Paul Teasdale pdt at rcsuk.fsnet.co.uk
Sun Aug 1 18:21:48 BST 2004


On Sunday 01 Aug 2004 10:26, Al Girling wrote:
> Hi folks,
>
> I've received this message from the RYA about a pricing query I have.
> I've inserted the whole message including headers in the hope that
> someone can help to explain way I see =A315 instead of what is possibly
> £15.  That's a guess BTW.  Is the =A315 ASCII code or can anyone give me
> a clue so I can find out what prices have been quoted.
>
> <BIG SNIPPTY SNIP>
>
> Toodle pip,
>
> Al

Hi Al,

I could be wrong on this one but I think it's the because the content type 
encoding is 7-bit thus giving you the ascii characters from 1 to 127 and you 
would need 8 bits to be able to represent a character at 163 (Hex: A3). This 
means that any characters above 127 needs to be represented with only the 
characters available between 1 to 127 hence the =A3. This 7-bit format is 
known as quoted-printable.

Furthermore I see the £ signs correctly in your e-mail you posted to the list 
and if I look at the headers sure enough the content type encoding is 8 bit 
(and the character set is iso-8859-1 which I would also expect to see)

I'm not sure why the reply message is 7-bit but it may be a case that the 
other users e-mail client is set to send in "US ASCII" only regardless (which 
is a 7-bit character set only) or it may be one of the gateways that the 
e-mail has passed through does not support 8-bit extended character sets such 
as ISO 8859-1 and therefore has changed the encoding of the message but 
that's unlikely these days. My best guess as to why this has happened is that 
the e-mail has defaulted to 7-bit quoted printable simply because the 
Content-Type mail header is not defined (but don't ask me why it's not 
defined :)

Incidentally I have KMail set so that firstly it always try to send reply 
messages in the same format as the original received message. If it can't do 
that (because I don't have the character set installed for example) it goes 
through other character sets (in the order I have specified) and finds the 
nearest character set that matches all of the characters within the message 
and uses that. If it can't match all characters in the message it will choose 
the closest as a final resort.

Hope this is of some helps even if I can't say exactly what caused the problem 
originally.

Regards,

Paul

-- 
Linux User #351704 (http://counter.li.org)
Scarborough LUG Member (http://www.scarborough.lug.org.uk)
Another Happy Slacker!! (http://www.slackware.com)





More information about the Scarborough mailing list