WEBVTT 00:00.000 --> 00:05.880 You already know the difference between a button being held, pressed, and released. 00:05.880 --> 00:13.160 Let's trace the same button across five robot loops so we can see exactly when each signal becomes true. 00:13.160 --> 00:16.240 Each column represents one pass through the robot loop. 00:16.240 --> 00:20.280 The button values are false, false, true, true, and false. 00:20.280 --> 00:25.080 The button remains down for loops three and four, so the loop reads true twice. 00:25.080 --> 00:34.600 That does not mean the driver pressed it twice. The program sampled the same held button on two different loops. 00:34.600 --> 00:41.880 If we draw these values as a graph, the line starts low because the button is false during loops one and two. 00:41.880 --> 00:49.260 Before loop three, the line steps straight up. That rising edge is the instant the button becomes pressed. 00:49.260 --> 00:56.160 The line stays high through loops three and four. That flat high section means the button is being held. 00:56.160 --> 00:58.800 It does not create another press. 00:58.800 --> 01:05.320 Before loop five, the line steps straight back down. That falling edge is the release. 01:05.320 --> 01:08.520 The line remains low because the button is false again. 01:08.520 --> 01:14.760 This looks like one square pulse from a PWM-style graph, but it is not a PWM output. 01:14.760 --> 01:20.400 We are only using the shape to show a boolean button state over time. 01:20.400 --> 01:25.240 Now we can compare the current sample with the previous sample. 01:25.240 --> 01:32.680 Held is the current button value, so it is true on both loops where the button is down. 01:32.680 --> 01:38.240 A pressed signal, also called a rising edge, is true only on loop three. 01:38.240 --> 01:42.440 That is the moment current changes from false to true. 01:42.440 --> 01:51.200 A released signal, or falling edge, is true only on loop five, where current changes from true back to false. 01:51.200 --> 01:59.480 At the top of the loop, `currentA` reads the button now. 01:59.480 --> 02:04.360 `previousA` still contains the value from the last loop. 02:04.360 --> 02:13.480 `aPressed` is true when current is true and previous is false, so it is the rising edge. 02:13.480 --> 02:18.840 `aReleased` is true when current is false and previous is true, so it is the falling edge. 02:18.840 --> 02:24.860 The order matters here, and we calculate both edges first. 02:24.860 --> 02:33.440 Update `previousA` near the bottom of the loop, after the results have already been used. 02:33.440 --> 02:41.560 If you overwrite previous before then, both values become identical and the edge disappears. 02:41.560 --> 02:51.200 The slow-mode line reads `left_bumper` directly because slow mode should remain active while the bumper is held. 02:51.200 --> 03:00.400 The claw toggle uses `aPressed` because the claw should change once per button press, not repeatedly while held. 03:00.400 --> 03:08.440 Without a rising-edge detector, the toggle would only work if the driver pressed the button for exactly one sample. 03:08.440 --> 03:14.320 That is not reliable or realistic. 03:14.320 --> 03:23.240 While held uses the current value. 03:23.240 --> 03:28.200 When pressed uses the rising edge, and when released uses the falling edge. 03:28.200 --> 03:38.840 Edge detection cannot fix a confusing control layout, so talk with your drivers before writing the condition. 03:38.840 --> 03:46.840 Before running the code, complete the held, pressed, and released rows for these five samples. 03:46.840 --> 03:57.560 Choose one season-robot action that should repeat while held and one action that should happen when pressed. 03:57.560 --> 04:11.240 Predict first, compare with telemetry, and document the controls so new drivers know how they work after the season ends.