Solved. It was actually the GPIO library for the ASUS tinker board, its RPi.GPIO implementation actually. It had a useless printf that sent that annoying message that messed up the json message. I removed that line and rebuilt from source and now it`s working flawlessly. Thanks.

@mykle1 Thanks, it`s getting closer to what I want. I will check the module you are mentioning. I am not that familiar with js so I did not recognize the things I need to change to accomplish what I was looking for. I will try again and see if I can make the image a bit bigger and the title white. Thanks a lot for your patience.

@oceank Yea that worked to some extent. When I say show me images it says that something is not supported.

Amazon Photos is not supported on this device. View your photos using the Amazon Photos app or website, or on devices like Fire TV or Echo Show.

When I say Alexa, start magic mirror, it promptly replies : “Hello my Queen, what can I do for you?” so that is working (Although I would like to change that message later)

Hello all,

I have tried to use the flick Large sensor together with my Raspberry Pi 2 to control the interface of the MagicMirror, as I have made some pages and wanted to scroll through them.

What I have managed to do is to control MMM-pages, MMM-page-indicator and the default news module via my MMM-flick module.

The controls are as follows:

Swipe Left - Decrement Page
Swipe Right - Increment Page
Swipe Up - Show news description
Swipe Up again - Open News page
Swipe Up as the news page is opened - Scroll down the news page
Swipe Down - Close News Page
Swipe Down again - Close News Description

I`m currently working on implementing the touch functionality. I want to also have mouse control.

The code is not perfect, I`m working it.

Here is the code (note that I have changed some libraries for it to work with the ASUS Tinker Board, as far as I know I`m the only one who has working flicklibs on the TinkerBoard. If someone is interested in getting the flick sensor to work with the TinkerBoard pm me ):

MMM-flick.py - used to read sensor data

 #!/usr/bin/env python
 
 import sys
 
 import json
 
 import time
 
 import signal
 
 import flicklib
 
 import RPi.GPIO as GPIO
 
 #import autopy
 
 
 
 GPIO.setmode(GPIO.BOARD)
 
 GPIO.setup(15, GPIO.OUT) 
 
 GPIO.setup(7, GPIO.OUT)
 
 
 
 #Turn on both LEDs for orange color in Stand By
 
 #GPIO.output(7, True) # Turn on RED LED
 
 #GPIO.output(15, True)# Turn on GREEN LED
 
 
 
 
 
 #Airwheel data
 
 
 
 some_value = 5000
 

 
 last_airwheel = 0
 
 delay = 5000
 
 
 
 
 
 #Get display size
 
 
 
 #width, height = autopy.screen.get_size()
 
 
 
 def to_node(type, message):
 
 	# convert to json and print (node helper will read from stdout)
 
 	try:
 
 		print(json.dumps({type: message}))
 
 	except Exception:
 
 		pass
 
 	# stdout has to be flushed manually to prevent delays in the node helper communication
 
 	sys.stdout.flush()
 
 
 
 to_node("status", 'Flick has started...')
 
 
 
 @flicklib.flick()
 
 def flick(start,finish):
 
   
 
 	#Slide down the newsfeed (DOWN GESTURE)
 
 	if(start == "north" and finish == "south"):
 
 		to_node("gesture", "ARTICLE_LESS_DETAILS")
 
 		GPIO.output(7, True)
 
 		time.sleep(0.5)
 
 		GPIO.output(7, False)
 
 		
 
 	#Slide up the newsfeed (UP GESTURE)
 
 	elif(start == "south" and finish == "north"):
 
 		to_node("gesture", "ARTICLE_MORE_DETAILS")
 
 		GPIO.output(15, True)
 
 		time.sleep(0.5)
 
 		GPIO.output(15, False)
 
 		
 
 	#Next page (RIGHT GESTURE)
 
 	elif(start == "west" and finish == "east"):
 
 		to_node("gesture", "PAGE_DECREMENT")
 
 		GPIO.output(7, True)
 
 		time.sleep(0.5)
 
 		GPIO.output(7, False)
 
 		
 
   
 
 	#Previous page (LEFT GESTURE)
 
 	elif(start == "east" and finish == "west"):
 
 		to_node("gesture", "PAGE_INCREMENT")
 
 		GPIO.output(15, True)
 
 		time.sleep(0.5)
 
 		GPIO.output(15, False)
 
    
 
 @flicklib.airwheel()
 
 def spinny(delta):
 
 	global some_value
 
 	global last_airwheel
 
 	global delay
 
 	some_value += delta
 
 	if some_value < 0:
 
 		some_value = 0
 
 	if some_value > 10000:
 
 		some_value = 10000
 
 	now = int(round(time.time() * 1000))
 
 	if(now - last_airwheel > delay):
 
 		#to_node()
 
 		last_airwheel = now
 
 
 
 #Mouse control via flick board
 
 
 
 #@flicklib.move()
 
 #def move(x,y,z):
 
 #	x = (x) * width
 
 #	y = (y) * height
 
 
 
 #	x = int(x)
 
 #	y = height - int (y)
 
 	
 
 #	if( y > 799 ):
 
 #		y = 799
 
 
 
 	#autopy.mouse.move(x, y)
 
 
 
 #Double tap gesture 
 
 
 
 #@flicklib.double_tap()
 
 #def doubletap(position):
 
 	#
 
 
 
 #Tap gesture
 
 
 
 @flicklib.tap()
 
 def tap(position):
 
 	if position == 'center':
 
 		GPIO.output(15, True)# Turn on GREEN LED
 
 		time.sleep(0.5)
 
 		GPIO.output(15, False)# Turn off GREEN LED
 
 
 
 #Touch gesture
 
 
 
 #@flicklib.touch()
 
 #def touch(position):
 
 	#
 
 
 
 
 
 
 
 signal.pause() 

node_helper.js

 'use strict';
 
 
 
 const NodeHelper = require('node_helper');
 
 const {PythonShell} = require('python-shell');
 
 var pythonStarted = false
 
 
 
 module.exports = NodeHelper.create({
 
   
 
 	python_start: function () {
 
 		const self = this;
 
 		const pyshell = new PythonShell('modules/' + this.name + '/MMM-flick.py', { mode: 'json', args: [JSON.stringify(this.config)]});
 
 		
 
 		pyshell.on('message', function (message) {
 
       
 
 			if (message.hasOwnProperty('status')){
 
 			console.log("node_helper_[" + self.name + "]" + message.status);
 
 			}
 
 			
 
 			if (message.hasOwnProperty('gesture')){
 
 			console.log("node_helper_[" + self.name + "] " + message.gesture);
 
 			self.sendSocketNotification("gesture_observed", message.gesture);
 
 			}
 
 		});
 
 		
 
 		 pyshell.end(function (err) {
 
 			if (err) throw err;
 
 			console.log("node_helper_[" + self.name + "] " + 'finished running...');
 
 		});
 
 	},
 
 
 
 	// Subclass socketNotificationReceived received.
 
 	socketNotificationReceived: function(notification, payload) {
 
 		if(notification === 'CONFIG') {
 
 		  this.config = payload
 
 		  if(!pythonStarted) {
 
 			pythonStarted = true;
 
 			this.python_start();
 
 		  };
 
 		};
 
 	}
 
 	
 
 });

MMM-flick.js

Module.register("MMM-flick",{	

		gesture_up: 0,

		gesture_right: 0,	

	// Override socket notification handler.

	socketNotificationReceived: function(notification, payload) {

		if (notification === "gesture_observed"){

			const self = this;

			self.sendNotification(payload);

			if (payload === "up"){

					MM.getModules().withClass(this.config.defaultClass).exceptWithClass(this.config.everyoneClass).enumerate(function(module) {

						module.hide(1000, function() {

							Log.log(module.name + ' is hidden.');

						});

					});

			

					MM.getModules().withClass("class_up_1_show").enumerate(function(module) {

						module.show(1000, function() {

							Log.log(module.name + ' is shown.');

						});

					});

			}

			else if (payload === "down") {

				MM.getModules().withClass("class_up_1_show").enumerate(function(module) {

						module.hide(1000, function() {

							Log.log(module.name + ' is hidden by gesture.');

						});

					});

			}

		}

	},

	

	start: function() {

		this.current_user = null;

		this.sendSocketNotification('CONFIG', this.config);

		Log.info('Starting module: ' + this.name);

	}

	

});

@mattsharp Thanks for trying, I will try soon enough, I have tried the google assistant module, but I`ll probably return to alexa soon.

Updated the code, and also support for the Asus Tinker Board.

Nvm, I got it working perfectly! xD

That is exactly what I’m planning to do. In theory it works behind the mirror, but you will lose the touch gestures more than likely. Unfortunately I do not have the mirror yet (it has not been delivered yet), but I’ll post a picture/results when I get it.

So far I wrote the code this way:

Left/right swipes go through MMM-pages & MMM-pages_indicator

Up/down swipes open/close newsfeed details

Each gesture also triggers the LED on the sensor board.

I’m still thinking what other things I can do with the remaining gestures.

I’ll update the correct code after I clean it up a bit.

@steffenschmidt, I was successful to some extent. I have managed to use the gestures from the board to control the interface. Up, down, left and right are working and more will follow when I know what I want them to do. The only problem is that the flick sensor is NOT WORKING behind the mirror, probably because of the aluminum/silver layer that makes a mirror a mirror.

I will work on the project in the winter holidays and post some more details.

@xne0n, I have made some progress, it runs perfectly smooth on the Asus Tinker board, I have managed to get mouse input working decently. It works by putting your finger close to the flick, the orange led turns on, and then the mouse follows your finger. It is not mouse pad like precision, but it gets the job done. I will update the source code, or maybe add it as a module on github after I finish writing my thesis.

@xne0n, I have made some progress, it runs perfectly smooth on the Asus Tinker board, I have managed to get mouse input working decently. It works by putting your finger close to the flick, the orange led turns on, and then the mouse follows your finger. It is not mouse pad like precision, but it gets the job done. I will update the source code, or maybe add it as a module on github after I finish writing my thesis.

Hi,

Sorry for my really late reply, I have been working on the module for a while, getting the mouse control to work.

As for your request, I did not know of the MMM-GroveGestures module. I had a look at it and the basic commands are basically the same although the code is written a bit differently. The gestures sequence though is a nice feature and I will look into that and try to do it the following days, but I am not really sure what actions should I bind to those new gestures.

Best regards,

Radu

Would it be possible to make LEAP run on a laptop board with magic mirror (i5, 8GB of ram). A laptop board could be easily integrated in a frame.

So you did you get it to work? I have an updated version if you are interested.

@mattsharp Thanks mate. I moved to MirrorMirroronthewall and almost got it working flawlessy. So don`t worry too much.

@oceank I start with Alexa, start magic mirror, which works just fine. Then Search image of apples, find image of apple or any other possible way of saying I want an image of an apple results in:

Amazon Photos is not supported on this device. View your photos using the Amazon Photos app or website, or on devices like Fire TV or Echo Show.

for videos I get:

I couldn’t find any enabled video skills that can find that. Go to the Alexa app to manage your video skills and devices.

#Solved. Forgot to modify the app ID line pointed out a few pages ago. Waiting for my echo dot to arrive to play around.

@oceank I can`t seem to find anything relevant in the logs… I keep getting

Amazon Photos is not supported on this device. View your photos using the Amazon Photos app or website, or on devices like Fire TV or Echo Show.

and similar for videos.

@oceank Okay, I`ll try that now. Nothing changed unfortunately…

@oceank I can confirm search images is on. I have also added youtube javascript v3 to the api. Are you sure 2 different projects are required?

@oceank I redid the steps. Just to be clear, I need to do only the api steps right? not installing whatever they are saying there?

I keep getting the same message after I re-uploaded to lamba with new api keys.