Re: non-ASCII stuff in manuf

[Guy Harris <guy () alum mit edu>]

On Oct 25, 2010, at 3:13 PM, Jeff Morriss wrote:

> The manuf entry for that is (escaped):
>
> 00:50:C2:0A:20:00/36    J\xC3\xAF\xC2\xBF\xC2\xBDgerCom
>
> And, actually, Wireshark displays it pretty much like Unidecode() does:
>
> Ji?1/2gerC

And so does Safari - if I do

        http://www.google.com/search?client=safari&rls=en&q=00:50:C2:0A:20:00&ie=UTF-8&oe=UTF-8

it finds

        http://hwaddress.com/?q=J%20&%20F%20Labs

which speaks of

        J�ger Computergesteuerte Messtechnik GmbH

and, not to any great surprise of mine, that's supposed to be

        http://www.adwin.de/index-us.html

"Jäger Computergesteuerte Messtechnik".

Now, "a with diaresis" shouldn't take 6 octets to encode it in UTF-8 - that's 0x00E4 in Unicode, which turns into 0xC3 
0xA4 in UTF-8.  Even if you use "a" plus a combining diaresis, it's not 0xC3 0xAF 0xC2 0xBF 0xC2 0xBD in UTF-8 - that 
UTF-8 sequence is 0x00EF 0x00BF 0x00BD in Unicode, which is, not surprisingly, ï ¿ ½.  (And it's not ISO 8859/1, either 
- that'd just be 0xE4 for "a with diaresis".  I've no idea what character encoding would turn "a with diaresis", or "a" 
plus a combining diaresis, into that octet sequence.

I suspect that's just something corrupted in the entry in the IEEE database; perhaps we need to override that entry 
when we generate our manuf file.
___________________________________________________________________________
Sent via:    Wireshark-dev mailing list <wireshark-dev () wireshark org>
Archives:    http://www.wireshark.org/lists/wireshark-dev
Unsubscribe: https://wireshark.org/mailman/options/wireshark-dev
             mailto:wireshark-dev-request () wireshark org?subject=unsubscribe