Re: Issue 150 in libnfc: nfc_initiator_poll_targets() could have more human understandable arguments

[lib...@googlecode.com]

Comment #1 on issue 150 by rco...@il4p.fr: nfc_initiator_poll_targets()
could have more human understandable arguments
http://code.google.com/p/libnfc/issues/detail?id=150

r1161 simplifies the polling we don't need a nfc_target_t array but a
simple pointer.

[lib...@googlecode.com]

Updates:
Cc: yob...@gmail.com r...@libnfc.org emanuele...@gmail.com

Comment #2 on issue 150 by rco...@il4p.fr: nfc_initiator_poll_targets()

could have more human understandable arguments
http://code.google.com/p/libnfc/issues/detail?id=150

In my opinion, we could provide a function that only need a period (in ms
or timeval struct) during the polling function will act.
We can compute the number of polling with the period.

We can also provide a second function that allow to set more precisely
theses parameters (nfc_poll_target_ext() ) or create some options to set
the way we compute the thing :)

What do you think about that ?

[lib...@googlecode.com]

Comment #3 on issue 150 by yob...@gmail.com: nfc_initiator_poll_targets()

could have more human understandable arguments
http://code.google.com/p/libnfc/issues/detail?id=150

Good idea IMHO.
Tuning the parameters of the polling itself is not that important in 99% of
cases.
At least this would provide a common denominator between native polling of
PN532 and sw polling of the other chips.