PsychoPy backend causes OpenSesame Hang-up

imnotamember

Hi all,
I recently had to reformat my Kubuntu system and reinstalled OpenSesame. It worked previously without issue using PsychoPy backend. Now with a fresh system install and new install of OpenSesame I cannot run an experiment using the PsychoPy backend.

For clarification, when I hit any of the run buttons (windowed, fullscreen, quickrun) OpenSesame grays out fields as normal when an experiment begins. But that is as far as it gets. Here is the output from the run (including the output following me killing the experiment process to return control to the main OpenSesame window).

Python 2.7.12

* Type "help()", "copyright()", "credits()" or "license()" for more information.
* Use the "print([msg])" function in inline_script items to print to this debug window.
* Inspect inline_script variables when an experiment is finished.


In [1]: Starting experiment as ExperimentProcess-3
openexp._canvas.psycho.init_display(): waitblanking = True
openexp._canvas.psycho.init_display(): monitor = testMonitor
openexp._canvas.psycho.init_display(): screen = 0

Python seems to have crashed. This should not happen. If Python crashes often, please report it on the OpenSesame forum.

item-stack: 
time: Mon Mar 20 15:37:25 2017

No window/full-screen comes on showing the stimuli. I've reinstalled various times with different versions of PsychoPy installed seperately and nothing changes this result. This is also just the initial 'welcome' experiment that OpenSesame defaults to when starting.

Thought I'd bring this up in case it is a new bug (I didn't see references to this anywhere on the forums or the rest of the internet).

Here is my system info list from within OpenSesame:

System

System: Linux-4.4.0-66-generic-x86_64-with-Ubuntu-16.04-xenial
Architecture: 64bit
Modules and packages

OpenSesame 3.1.6
Python 2.7.12 (default, Nov 19 2016, 06:48:10) [GCC 5.4.0 20160609]
QProgEdit 4.0.9
datamatrix 0.4.4
qdatamatrix 0.1.13
pseudorandom 0.2.2
fileinspector 1.0.2
QNotifications 1.1.1
QOpenScienceFramework [not available]
opencv 2.4.9.1
expyriment [not available]
IPython 2.4.1
numpy 1.11.0
scipy 0.17.0
PIL/ PILLOW 1.1.7
psychopy 1.83.04
pygame 1.9.1release
pygaze 0.6.0a21
pyglet 1.1.4
PyQt 5.5.1
serial 3.0.1
markdown 2.6.6
bidi [not available]
yaml 3.11

sebastiaan

Hi,

I would first check if PsychoPy works by itself (not from within OpenSesame), for example by running this example script:

from psychopy import visual, core  # import some libraries from PsychoPy

#create a window
mywin = visual.Window([800,600], monitor="testMonitor", units="deg")

#create some stimuli
grating = visual.GratingStim(win=mywin, mask="circle", size=3, pos=[-4,0], sf=3)
fixation = visual.GratingStim(win=mywin, size=0.5, pos=[0,0], sf=0, rgb=-1)

#draw the stimuli and update the window
grating.draw()
fixation.draw()
mywin.update()

#pause, so you get a chance to see it!
core.wait(5.0)

This should show a simple grating and fixation square for 5 s. Does that work?

Cheers,
Sebastiaan

imnotamember

Hi Sebastiaan,

Thanks for the help! I ran your testing code both in the PsychoPy Coder and in a terminal within my python environment, and it works in both.

As a side note, when I run this:

---
API: 2
OpenSesame: 3.1.6
Platform: posix
---
set width 1024
set uniform_coordinates yes
set title "New experiment"
set subject_parity even
set subject_nr 0
set start experiment
set sound_sample_size -16
set sound_freq 48000
set sound_channels 2
set sound_buf_size 1024
set sampler_backend legacy
set round_decimals 2
set mouse_backend psycho
set keyboard_backend psycho
set height 768
set fullscreen no
set form_clicks no
set foreground white
set font_underline no
set font_size 18
set font_italic no
set font_family mono
set font_bold no
set disable_garbage_collection yes
set description "The main experiment item"
set coordinates uniform
set compensation 0
set color_backend psycho
set clock_backend psycho
set canvas_backend psycho
set bidi no
set background black

define sequence experiment
    set flush_keyboard yes
    set description "Runs a number of items in sequence"
    run new_inline_script always

define inline_script new_inline_script
    set description "Executes Python code"
    set _run "print 'testing in run phase'"
    set _prepare "print 'testing in prepare phase'"

My output is:

Python 2.7.12

* Type "help()", "copyright()", "credits()" or "license()" for more information.
* Use the "print([msg])" function in inline_script items to print to this debug window.
* Inspect inline_script variables when an experiment is finished.


In [1]: Starting experiment as ExperimentProcess-2

Which makes me think the error is occurring very early in the set-up (usually it at least indicates some packages were loaded after the 'Starting experiment...' line).

I just tried the same code by running OpenSesame in terminal with the debug and warning flags on and got this message numerous times, but I assume it isn't necessarily related since it's just a warning(?).

***startwarning***
  File "/usr/bin/opensesame", line 25, in <module>
    __main__.opensesame()
  File "/usr/lib/python2.7/dist-packages/libqtopensesame/__main__.py", line 87, in opensesame
    sys.exit(app.exec_())
  File "/usr/lib/python2.7/dist-packages/libqtopensesame/qtopensesame.py", line 1007, in run_quick
    self.run_experiment(fullscreen=False, quick=True)
  File "/usr/lib/python2.7/dist-packages/libqtopensesame/qtopensesame.py", line 973, in run_experiment
    auto_response=self.experiment.auto_response)
  File "/usr/lib/python2.7/dist-packages/libqtopensesame/runners/base_runner.py", line 256, in run
    ret_val = self.execute()
  File "/usr/lib/python2.7/dist-packages/libqtopensesame/runners/multiprocess_runner.py", line 58, in execute
    self.exp_process.start()
  File "/usr/lib/python2.7/multiprocessing/process.py", line 130, in start
    self._popen = Popen(self)
  File "/usr/lib/python2.7/multiprocessing/forking.py", line 126, in __init__
    code = process_obj._bootstrap()
  File "/usr/lib/python2.7/multiprocessing/process.py", line 258, in _bootstrap
    self.run()
  File "/usr/lib/python2.7/dist-packages/libqtopensesame/misc/process.py", line 128, in run
    subject_nr=self.subject_nr, logfile=self.logfile)
  File "/usr/lib/python2.7/dist-packages/libopensesame/experiment.py", line 140, in __init__
    item.item.__init__(self, name, self, string)
  File "/usr/lib/python2.7/dist-packages/libopensesame/item.py", line 67, in __init__
    self.from_string(string)
  File "/usr/lib/python2.7/dist-packages/libopensesame/experiment.py", line 345, in from_string
    self.items.new(item_type, item_name, def_str)
  File "/usr/lib/python2.7/dist-packages/libopensesame/item_store.py", line 156, in new
    name = self.valid_name(_type, suggestion=name)
  File "/usr/lib/python2.7/dist-packages/libopensesame/item_store.py", line 215, in valid_name
    while _name in self:
  File "/usr/lib/python2.7/dist-packages/libopensesame/item_store.py", line 290, in __contains__
    name = cistr(name)
  File "/usr/lib/python2.7/dist-packages/libopensesame/cistr.py", line 39, in __init__
    str.__init__(self, s)
  File "/usr/lib/python2.7/dist-packages/libqtopensesame/qtopensesame.py", line 343, in warn_with_traceback
    traceback.print_stack()

/usr/lib/python2.7/dist-packages/libopensesame/cistr.py:39: DeprecationWarning: object.__init__() takes no parameters
  str.__init__(self, s)

***endwarning***

Any help you can offer is greatly appreciated. I usually try to troubleshoot these things on my own but this is getting deeper into the OpenSesame core than I know at the moment. I thought you might have seen this in the past.

Thanks,
Josh

sebastiaan

Hi Josh,

It took me a while, but I've seen this problem too now. On Ubuntu 16.10. On Ubuntu 16.04 or 17.04 things work fine for me. Are you running 16.10 as well?

I don't think you'll be able to properly find the cause of this crash. It's a low-level crash somewhere, and it can be anything, really. I think you're best bet is to upgrade to 17.04, if you haven't done so already.

Cheers!
Sebastiaan

imnotamember

Hi Sebastiaan,
Great news, thanks for the update! I had a feeling it was something very close to the kernel. I'll update and repost if things are resolved then.

Thanks again,
Josh