Ludovic Rousseau wrote:
Hello,
On 13/06/07, Paul Klissner <[EMAIL PROTECTED]> wrote:
Does anyone know if and where it is documented under which
circumstances SCardReconnect() returns a sharing violation?
I have a case with the code I've been working with where I see one
application
thread issues SCardBeginTransaction() meantime the card is removed,
and before SCardEndTransaction() is called, another thread receives
an SCARD_W_RESET from SCardGetStatus() and issues an
SCardReconnect() [w/SCARD_S_SHARED set] as a result,
and SCardReconnect() returns the error SCARD_E_SHARING_VIOLATION.
Is that the proper behavior, or perhaps a bug in my code, or PCSClite,
or is it undocumented?
What version of pcsc-lite do you use?
Currently based on pcsc-lite-1.3.2, haven't synced up with later
builds yet.
What do you think would be the correct behavior?
Premature for me to speculate about the ideal. I'm trying to understand
what the expected behavior of pcsc-lite-1.3.2 is, and determine if it is
based on
the PC/SC Working Group standard or if it is implementation-specific and
curious if anyone here knows offhand.
The reason: We have a multi-threaded vendor app written to accommodate
another implementation of PC/SC ('PCBypass'). The vendor app is failing
with my implementation based on PCSClite-1.3.2 during Smart Card
removal/re-insertion, but doesn't seem to exhibit the behavior when run
with PCBypass.
I'm investigating sharing violations based on a casual comment made
by one of the vendor's employees who studied a log file and thought the
sharing violation errors 'didn't look right'. I want to rule it out.
I'm just
about finished writing even more focused test that I will use to compare
the sharing violation error behavior between my implementation,
PCSClite-1.3.2
and PCSCbypass to see if they all behave the same or not.
This could turn out to be an unrelated bug in my code, or a bug in the
vendor
app. Or it could be due to disagreements among the implementations. I want
to know if the behavior is defined in the standard and if said
implementations
adhere to it if so. I am just trying to sort it all out.
Thanks,
Paul
_______________________________________________
Muscle mailing list
[email protected]
http://lists.drizzle.com/mailman/listinfo/muscle
