On Mon, 14 Mar 2005 15:51:08 +0100, Ludovic Rousseau <[EMAIL PROTECTED]> wrote: > - You don't need to use i2dw_be()/dw2i_be() since the value of > IOCTL_FEATURE_VERIFY_PIN_DIRECT is "opaque" for the application. It is > just 4 bytes and the byte order is not important. I confirm that i'm lazy and thus stupid too. I'm now somewhat ignorant as i'm too lazy to think but might it cause problems if the application and actual reader ar eon different endian machines ?
> - Maybe you could define a structure containing the TVL triplet to > ease the parse of the TLV structure? Nod. > > Due to the fact that CCID spec at > > the same time is somewhat lacking proper information about such > > things, the translation between teletrust and ccid blocks is somewhat > > ugly > I confirm the ugliness :-) > > Maybe we should wait until PCSC v2 becomes public & final before I include > anything related to secure PIN? (I hope it is compatible with teletrust class > 2) > When is PCSC v2 supposed to be ready? I've been told that some time this summer. I've also been told that teletrust class 2 spec will be included as part10 to pcsc v2 spec and the interface will undergo only cosmetics. So the interface should not change. Wither way, what is needed in pcsc for this is the introduction of SCARD_CTL_CODE and the ability to send no data with SCardControl. Everything else can be kept out of the scope of pcsclite. Needed tag codes can be kept separately at ccid and application side if needed. But it would be nice to be able to use the capabilities of the reader before somebody decides that it is now nice time to publish some kind of a spec. -- Martin Paljak [EMAIL PROTECTED] http://martin.paljak.pri.ee/ +372.5156495 - phone _______________________________________________ Muscle mailing list [email protected] http://lists.drizzle.com/mailman/listinfo/muscle
