This provided constants that translated from the physical header pin numbers to GPIO lines. Instead, we (and the majority of other Pi related documentation out there) now just use GPIO numbers exclusively.
The class hasn't been used in examples since a few versions, and also wasn't ever properly documented in the reference, so unlikely that many people will notice.
This function is used like this:
GPIO.waitFor(pin number, rising/falling/change, timeout or -1)
While it's using an interrupt internally, in the way it is used it doesn't relate to the interrupt functions (attachInterrupt, etc) at all. To prevent confusion, rename it to waitFor. (Another possible name would be waitForPin.)
Some measurements with a logic analyzer and the Raspberry Pi 2
sleep: 0.001500, measured avg: 0.0015357, measured 0.95 perc.: 0.0015573
sleep: 0.0185, measured avg: 0.0186177, measured 0.95 perc.: 0.0186345
servo_pulse_oversleep was set to account for the (expected) overhead of waking up and toggling the pin with help from the numbers above.
Instead of throwing an exception early on, this makes it possible to test all other parts of sketches that make light use of the IO library. A warning is printed once after startup for those environments.
This was tested on OS X against all examples that come with the IO library. Note: x86 and x64 are technically supported platforms. If a user on those prefers to use simulation instead, it is necessary to call NativeInterface.alwaysSimulate() before any other IO library function.
It appears as if using the I2C pins for GPIO requires a restart of the Raspberry Pi before the I2C interface can be used again. Play safe and move the examples to pins that aren't used for anything else.
This makes everyone's sketches a bit lighter, since it is not longer necessary to explicitly cast to byte or call byte(). Instead, we throw an exception if the value does not fit into the byte.
