crash with libnfc 1.7.0

[nmav]

Hello,
I'm trying to create a package of rfidiot in Fedora 20 which ships libnfc 1.7.0. I'm testing with omnikey CardMan 5321 and when I run:
/usr/bin/cardselect.py -N
I get a segmentation fault.

A valgrind backtrace (don't know how helpful it is), is as follows:
==11960== Invalid read of size 1
==11960== at 0x4A092F2: strlen (in /usr/lib64/valgrind/vgpreload_memcheck-amd64-linux.so)
==11960== by 0x3E0309821F: PyString_FromString (stringobject.c:121)
==11960== by 0xCFC8E58: _ctypes_callproc (callproc.c:927)
==11960== by 0xCFC2A84: PyCFuncPtr_call (_ctypes.c:3929)
==11960== by 0x3E0304A0C2: PyObject_Call (abstract.c:2529)
==11960== by 0x3E030DE38B: PyEval_EvalFrameEx (ceval.c:4316)
==11960== by 0x3E030E098F: PyEval_EvalFrameEx (ceval.c:4184)
==11960== by 0x3E030E21EC: PyEval_EvalCodeEx (ceval.c:3330)
==11960== by 0x3E0306F0F7: function_call (funcobject.c:526)
==11960== by 0x3E0304A0C2: PyObject_Call (abstract.c:2529)
==11960== by 0x3E030590C4: instancemethod_call (classobject.c:2602)
==11960== by 0x3E0304A0C2: PyObject_Call (abstract.c:2529)
==11960== Address 0x20 is not stack'd, malloc'd or (recently) free'd
==11960==
==11960==
==11960== Process terminating with default action of signal 11 (SIGSEGV)
==11960== Access not within mapped region at address 0x20
==11960== at 0x4A092F2: strlen (in /usr/lib64/valgrind/vgpreload_memcheck-amd64-linux.so)
==11960== by 0x3E0309821F: PyString_FromString (stringobject.c:121)
==11960== by 0xCFC8E58: _ctypes_callproc (callproc.c:927)
==11960== by 0xCFC2A84: PyCFuncPtr_call (_ctypes.c:3929)
==11960== by 0x3E0304A0C2: PyObject_Call (abstract.c:2529)
==11960== by 0x3E030DE38B: PyEval_EvalFrameEx (ceval.c:4316)
==11960== by 0x3E030E098F: PyEval_EvalFrameEx (ceval.c:4184)
==11960== by 0x3E030E21EC: PyEval_EvalCodeEx (ceval.c:3330)
==11960== by 0x3E0306F0F7: function_call (funcobject.c:526)
==11960== by 0x3E0304A0C2: PyObject_Call (abstract.c:2529)
==11960== by 0x3E030590C4: instancemethod_call (classobject.c:2602)
==11960== by 0x3E0304A0C2: PyObject_Call (abstract.c:2529)
==11960== If you believe this happened as a result of a stack
==11960== overflow in your program's main thread (unlikely but
==11960== possible), you can try to increase the size of the
==11960== main thread stack using the --main-stacksize= flag.
==11960== The main thread stack size used in this run was 8388608.

The equivalent for PCSC works pretty ok:
$ cardselect.py -L
PCSC devices:
No: 0 OMNIKEY CardMan 5321 (OKCM0072903120829265736952848153) 00 00

Any ideas?

[doegox]

Hello

I'm trying to create a package of rfidiot in Fedora 20 which ships
libnfc 1.7.0. I'm testing with omnikey CardMan 5321 and when I run:
/usr/bin/cardselect.py -N
I get a segmentation fault.

Not nice error handling I admit but omnikey is not a reader supported by
libnfc