motion.get_magnetic_field()

Omega0

With 1.6 in beta I wanted to bring up an unresolved issue from the past: motion.get_magnetic_field(). I wanted to start doing some tests with magnetism but I keep getting (0.0, 0.0, 0.0, 0) as the return value. Is there any chance this will be fixed in the final release of 1.6?

dgelessus

What kind of device are you using, and does it actually have a magnetic field sensor? (Don't ask me which models have one. I didn't even know that there was such a thing in iDevices or that Pythonista supported it.)

Do any of the other motion functions return meaningful values? Did you remember to call motion.start_updates()?

Omega0

I'm using an iPhone 5s which does have a magnetic field sensor (and does work in other apps) and I did remeber to call motion.start_updates(), so those aren't the issues. I remeber seeing another thread about this a while ago and don't remember that being resolved so I think this is just a glitch in Pythonista.

wradcliffe

I just checked it on my ipad4 using 1.6

>>> import motion
>>> motion.get_magnetic_field()
(0.0, 0.0, 0.0, 0)
>>> motion.start_updates()
>>> motion.get_magnetic_field()
(-22.983476638793945, -37.92658233642578, -45.98847198486328, 1)
>>> motion.stop_updates()

zencuke

I just tried on my brand new iPhone 6. Failed with:

'''
>>> import motion
>>> motion.get_magnetic_field()
(0.0, 0.0, 0.0, 0)
>>> motion.start_updates()
>>> motion.get_magnetic_field()
(0.0,0.0,0.0,-1)
'''

Shouldn't magnetic field require location services. It used to. I can't figure out how to turn on location services for Pythonista.

dgelessus

Settings app > Privacy > Location Services > turn on the switch for Pythonista

Omega0

Enabled location services for Pythonista, still not working.

polymerchm

Some more data points. I get (0.0, 0.0,0.0,0) before enabling updates as above, then (0.0,0.0,0.0,-1) using beta 1.6 on a Ipad Air 2. ioS 8.1.1.

Other motion.get_* calls work just fine.

The motion request preferences does not include Pythonista as an application requesting access.
OMZ needs to work on this. Not fixable by mere mortals.

omz

For 1.6 beta users: Could you try moving your device after enabling updates to calibrate the compass?

import motion
import time

try:
    motion.start_updates()
    while True:
        print motion.get_magnetic_field()
        # if you get (0, 0, 0,...) here, try moving your device in circles... like you would in the Compass app
        time.sleep(0.5)
finally:
    motion.stop_updates()

polymerchm

After ~10 (0,0,0,-1) returns, it starts to give real values for x,y,z and 1 for the last parameter.

omz

Okay, I don't think there's more I can do. If you want to use the magnetometer data, you might want to check if the compass is calibrated first, and wait for calibration before using the data otherwise, something like this:

import motion
import time

motion.start_updates()
while True:
    m = motion.get_magnetic_field()
    if m[3] != -1:
        break
    print 'Please calibrate the compass by moving the device...'
    time.sleep(0.5)

print 'Magnetometer successfully calibrated'
print 'Data:', motion.get_magnetic_field()

motion.stop_updates()

zencuke

Just tried again. I get (0,0, 0.0, 0.0, -1) forever, i.e hundreds of iterations regardless of motion. Just as I did when I tried 4 days ago. I have calibrated the compass with the compass app. I can switch between Pythonista and the compass app seeing the compass app working but Pythonista still giving the same empty results.

I'm using iPhone 6 with iOS 8.1.1. As far as I remember I haven't been able to make get_magnetic_field to work since I upgraded from iPhone 5 to iPhone 6 several days ago. I'm pretty sure it worked on the iPhone 5 but I don't have it anymore to test.

Note: I just updated the infinite loop in my test script to print all 4 motion.get_<> function results. Everything prints normal (at least non zero and changing as the iPhone is rotated and tilted) except get_magnetic_field(); all this while jumping back and forth between Pythionista and the compass app.

I don't think it is a calibration problem. Also since the compass app works it is hard to imagine that it is a problem with my new iPhone 6 either. Note: I just downloaded 4 of the free compass apps. They all work; even while switching between them and the failing Pythonista script. So far 5 compass apps work every time and my Pythonista script fails every time.

I don't know if this helps but I did notice a repeatable state change. When I run the script from a cold start get_magnetic_fileld() reports 0 accuracy before the call to start_updates() and for the first call after the call to start_updates. Then it switches to -1. Rerunning the script always sees accuracy of -1 even after a call to stop_updates(). The only way I can get an accuracy of 0 again is to kill Pythonista and start over.

brumm

hi, I tried the same (iphone 6 + pythonista 1.5). Compass app works, but in pythonista I only got (0.0, 0.0, 0.0, -1).


import motion
import time

t1 = time.time()
run = True 
motion.get_magnetic_field()
motion.start_updates()
while run:
  m = motion.get_magnetic_field()
  t2 = time.time()
  if (t2 - t1) > 5.0:
    run = False
  else:
    print m
motion.stop_updates()

polymerchm

@omz, @Brumm, @zencuke:

omz's test ran fine on an Ipad Air 2 w/ 8.1.1. Sounds like an iPhone 6 issue.

JonB

Ipad 2 has the same issue, w/ 1.5.

zencuke

@dgelessus - Sorry I wasn't clear. I know how to turn on Location services in general. I just don't see Pythonista in the list when I go there.

Edit: When a script imports location a dialog pops up. After that Pythonista is in the location services list. Hmm. I wonder if this the normal behavior for all apps. I've never noticed it before but this may be the first time I checked location settings before attempting to use the app location functionality.

ashook

Does anyone know of a workaround or fix for this problem? motion.get_magnetic_field() fails on both of my iOS devices: iPad mini 9.02, iPhone 5 9.02.