Closed GoogleCodeExporter closed 9 years ago
Sounds nice.
Another use-case that would be made possible and I think would be greate is the
possibility to set-up regression tests. I am planning some refactoring in the
libnfc and would be really less nervous if there were some knd of regression
tests suite that I could use to ensure I don't break any feature.
Are you just planning PCD emulation or willing to support say "modules" that
emulates PICCs on the PCD ? Has work already started (the issue is in the
accepted state… but Google trends to use this as default if you are a
developer and add an issue in a project) ? How have you planned to do the
implementation ?
Original comment by romain.t...@gmail.com
on 6 Aug 2010 at 9:42
Currently my idea is to simulate the NXP PN532 firmware with only a small set
of supported commands and only for passive targets (all tags supported by the
chip).
Tags could be simulated with dump files.
We are in the initial case-study phase so all suggestions are welcome!
Original comment by emanuele.bertoldi
on 6 Aug 2010 at 9:58
Okay
I would really be pleased if the code had enough abstraction to easily add
support for emulating more devices (PN531, PN533) and tags (e.g. DESFire
targets for which a dump will not be enough) in the future.
One way might be having a PCD emulation daemon for emulating the NFC device
(one daemon per device) that communicates with any number of PICC emulator
daemons (one daemon per target). The libnfc then would only require a simple
"pass-through" interface to communicate with the PCD emulator.
While it is probably not the easiest way to reach the goal, this modularity has
some benefits:
- It is easy to add new types of target (possibly custom ones);
- Targets emulators can be programmed in almost any language, that allows someone to use the same code for testing purpose and in an OS programmed in a contact-less Smart Card (e.g. FLOSS implementation of Mifare DESFire OS for a Java / BASIC card… I have never played with these card so maybe it is not feasible, but if it is, it may be a real advantage for developers to have such a framework);
But it might be really over-skill regarding what you are planning ;-)
Maybe the wiki could be a good place for the brainstorming ?
Romain
Original comment by romain.t...@gmail.com
on 6 Aug 2010 at 1:20
Nobody is working on this issue anymore.
Is someone interesting by this feature ?
Original comment by romu...@libnfc.org
on 27 Sep 2012 at 9:43
Original comment by romu...@libnfc.org
on 29 Jan 2013 at 2:36
Original issue reported on code.google.com by
emanuele.bertoldi
on 6 Aug 2010 at 9:04