setWatch issue

I tried the 'cutting edge build' and have verified that it is working with my 8uS pulse using

{repeat:true,edge:"falling",hispeed:true­}

Thanks @Gordon and @allObjects

@Gordon, this new option is absolutely great news. I was looking for something about detectable minimum pulse width, but could not find it. Now we know another tech spec detail and dependency on power (saving) mode.

I do not really care about the reported pin state, because when push comes to shove, I control it anyway myself. Important is that the pulse comes thu. This even helps for complicated debouncing.

@billsalt 's addition of the flip flop that cuts the frequency in half and watching both edges is really a great solution pattern.

The most though I liked the test driver. When I suggested it a while ago I was not sure if it is possible to create such a short pulse... now it's proven (I got the idea from another test driver implementation when exciting an LC circuit for figuring the resonance: Espruino controlling LC resonance experiment in HAM Radio class. In this example I was not sure how high the frequency could go. Since for finding the resonance frequencies, excitement works also just 'provoking' along the harmonics...).

Great! Also worth noting that while we do create events for rising and falling edges, to make this work I extended the code to allow some extra flags to be specified when creating a watch. It'd allow us to make some changes to (finally) being able to specify whether to record only rising or falling edges.

So far that hasn't been a big deal, but at least it can be done relatively easily now.

One thing to add: The setWatch code isn't smart enough to know whether you have one hispeed watch and one not - so to make this work reliably you can only really have one watch on each pin (if you have >1 watch on a pin it'll use the hispeed setting from the first call to setWatch with that pin).