From: Dominikus Scherkl (email@example.com)
Date: Tue May 10 2005 - 10:35:01 CDT
> > Dominikus Scherkl wrote:
> > میرا نام فراز ھے اور میں نے ایم سی ایس کراچی یونیورسٹی سے کیا ھے
> Hi Dominikus,
> yes, it looks very good (though I can't read it) and I
> did not have to change anything.
This has nothing to do with YOUR (the receiving party's) settings
But with those of the SENDER - try saving the garbage and open the file in an editor recognizing it (or letting you choose) as UTF-8 text. It will interpret the byte-sequences as arabic letters.
> Selecting/highlighting the garbeled line and choosing
> different fonts (including Arial Unicode MS) did not help.
Of course not. It's the "all above 0x7F gets multiple bytes also above 0x7F" UTF-8 encoding.
From the sending party your eMail client was told it is Latin-1
(8bit) text (the default, because the mime charset parameter was missing), so each byte of the multi-byte sequences was shown as latin-1 character. (arabic becomes twice as much garbage as the correct character count would be - chinese would become even more garbage).
> So how did you find out what character set was used
Your only chance is to guess, which charset it might be.
Fortunately nowadays UTF-8 is always a good guess, most times worth a try. Other charsets might require a conversion-tool to correctly show up. With such a tool you can try several encodings until you found the one producing plausible text (or ask your communication partner, if you can ;-).
But all this wouldn't help if the 8bit text came across a mail server capable of 7bit text only and therefore deleted all set 8th bits in the message. That was the cause I recommended the use of Base64 encoding.
But those servers become rare, so it might not be nessessary (in fact, we all got the message from Faraz Siddiqi with correct
content, but declared with wrong headers).
-- Dominikus Scherkl
This archive was generated by hypermail 2.1.5 : Tue May 10 2005 - 10:37:11 CDT