Error for problem with corrupt/missing .js file?

I have a counterbalanced experiment where participant must simply respond using a or l key if they are in group a, l or a if in group b to have responses mapped.
This runs perfectly on my PC but upon upload, once I select the group (a or b) I get the error below. I’m not very experienced with psychopy builder yet and I’m doing my best but this one I can’t unravel as it suggests something is wrong with the pixi-4.8.1.min.js file or it cannot find it. I know an update to 4.8.5 has been released, could that be the problem?

I’m using Psychopy3 Psychopy Builder, on Windows 7 (64bit), exporting html and uploading to Pavlovia.

[Exception… “Failure” nsresult: “0x80004005 (NS_ERROR_FAILURE)” location: "JS frame :: https://pavlovia.org/run/penning/smilesvalidityandreliability/html/lib/vendors/pixi-4.8.1.min.js :: [108]

Try to run the experiment again. If the error persists, contact the experimenter.

In short I think this is the sort of thing where we’re failing to convert python to JavaScript fully. Interesting challenge for us to make it work more smoothly :slight_smile:

I can see your experiment and I’ll try and get it working!

Brilliant, I was all pleased with myself for repairing one error, only to get another I realised I had no clue what to do.

Thanks

I’ve made some progress.
https://pavlovia.org/run/jon/smilesvalidityandreliability/html/

The instructions screen still isn’t working right, which I don’t understand, but that is our problem not yours. When I’ve got it working I’ll push the changes back to your version (then you can sync and be up and running again! :slight_smile:

OK, I’ve submitted a “pull request” to your experiment from my fixed version. Hopefully you got an email and it’s reasonably obvious from that how to “pull in” my changes :slight_smile:

If you get stuck let me know. You’re finding things that we do need to improve at our end so keep it coming!

thanks
Jon

Things I’ve changed:

  • moved stims into a folder to make life easier (I’d recommend that usually)

Then these all need some fixing at our end basically:

  • I had to manually copy this folder into html/resources because PsychoPy hadn’t found the stimuli automatically
  • you set your stimuli to inherit the units (pix) from the window but it turns out psychoJS can’t do that yet so I had to go through and manually set them all to be pix

Thank you. I’m learning so much! I’ve pulled the changes and it runs and collects data and I’m incredibly delighted by this. I’ll take on board your changes, I actually wasn’t sure how to have a stim folder with the way the conditions files were set up but I will explore. Not sure re the units/pix aspect, but I will hopefully get to grips with that once I have looked at the updated version properly (without having consumed Friday night wine).
Thank you again, all very helpful.
glen

Hi all,
I’m new to the PsychoPy/Pavlovia world and after a few bug fixing myself, I also ran into the same problem as Glen here (see also Screenshot)!
My experiment is straight forward (and works on my local laptop fine - Latest PsychoPy version). After a first instruction text, participants see a fixation cross, then a horizontally moving circle, and then the mouse is shown with which participants have to indicate the final location of the circle. I build it completely with builder and only uses the code component to manipulate the visibility and starting location of the mouse (the mouse should only be visible after the moving circle disappeared and should always appear in the center of the screen). The Experiment ends with a final text.

In contrast to Glen, I don’t use any external stimuli which have to be loaded (if i understand it correctly), yet I also define the size information (of the text as well as the fixation cross and moving circle) in pixel.

“you set your stimuli to inherit the units (pix) from the window but it turns out psychoJS can’t do that yet so I had to go through and manually set them all to be pix” <

So probably I also have to go though the js script and manuallly set the unit to pix? If so, does somebody have a blueprint how the unit can be defined in psychojs - I’m not familiar with js-programming…? Or am I on the wrong path?

Error Message Pavlovia|646x439

I found a work around, the text components were causing this error message and in fact the problem was that I used pix to define the letter height → I changed the unit to ‘norm’, adapted the values so that the text is readable (the exact size is not relevant to me as I only used the text component for the participants introduction) and it works now.

EDIT: I was imprecise with this:

In contrast to Glen, I don’t use any external stimuli which have to be loaded (if i understand it correctly), yet I also define the size information (of the text as well as the fixation cross and moving circle) in pixel.

I defined the unit information with ‘from Exp settings’ (which were set to pix). When defined directly as pix, it works.