Hi Ludovic,

>Do you know how the IFD-Handler differentiate an normal APDU from a CT-BCS
>command?
>The IFD-Handler detect the VERIFY command defined in ISO 7816-4 and
>perform a special treatment instead?

Maybe Jens Miltner could go into this in more detail? I would assume that only command 
class 0x20 for CT-BCS and instruction byte (0x18 in case of verify) are used.

Ahh... There would of course be two completely different ways to handle this... What 
we're doing is setting up the CT-BCS commands in MacGiro. The other way would of 
course be to capture the ISO 7816 verify command in the driver. There's several 
reasons against doing so:

In a non secure PIN entry environment, our application will display appropriate prompt 
strings when a PIN entry is required. This is especially important if a user has to 
juggle multiple cards in sequence when working with more than one bank account.

With secure PIN entry, these prompt strings may be displayed by the reader itself if 
it has a display, through parameters of the CT-BCS verify command.

Also, the user needs to be able do decide if he wants to operate any given card with 
or without secure PIN entry: There might be cards with an alphanumerical PIN wich can 
not be input on the numerical keypad of the reader.

>
>> This is certainly an arguable design decision. But, as there have been drivers 
>> deployed before using this scheme, we have decided to do the same thing in MacGiro. 
>> Currently, the cyberJack pinpad by Reiner SCT has a pcsc-lite driver for MacOS X 
>> that works together with MacGiro in this way.
>
>I don't know if I will follow this direction. It seems to be a serious
>hack :-)

I will not dispute this. It's just that there's IFD-Handler for both Linux and MacOS 
existing now that work this way.

>My proposal is not incompatible with the present situation. A driver
>using the "old" scheme (CT-BCS command in SCardTransmit) could _also_
>support the new one (SCardControl) without breaking any thing.

Yes, but a client application would have to know if the driver supported the old   or 
the new scheme.


-- 
Best Regards,

Eric Walter

med-i-bit GmbH

_____________________________________________________________________
mailto:[EMAIL PROTECTED]               http://www.med-i-bit.de
_______________________________________________
Muscle mailing list
[EMAIL PROTECTED]
http://lists.drizzle.com/mailman/listinfo/muscle

Reply via email to