Description of the problem: Greetings, I am trying to run a study from Psychopy3 builder to Pavlovia but I get the following message after Sync is completed: “We could not find a record for experiment designer: undefined”. Any suggestions about what to do?
Sharing the URL of the experiment won’t make it public. The data and the experiment will remain private but others will be able to see the address for where to run it (or not, as the case may be). But I don’t think I or @apitiot can help unless you’re prepared to share more than your current info
Apologies for the late reply…@AlexV Unfortunately I have not found a way to fix this. I tried it again numerous times and now it is stuck at the synchronising phase… @jon Although I followed all the steps meticulously when I click on the project I have created on pavlovia, there is no URL link (that space is empty).
If this was in a browser then there must be a URL. It’s the address at the top of your browser. Please send that. Or send your pavlovia username. I need something more to go on than the information here
The gitlab repository for your experiment seems to be empty. Have you perhaps deleted the experiment on your local machine? Could you resurect it so I can check what is going on on our end?
Cheers,
I went through the whole process again and managed to upload the experiment. The problem now is that it will not read the sentences from the excel file. It just shows ‘NaN’
Username: Marina Gibson
Name of project: SPR_Greek
Here’s the link to my thread on it. If you think there is a quick fix to it, then do let me know. I actually made another pavlovia account and I’ve managed to get all my studies running from there.
OK, thanks for sharing Marina. Now I can dig into it properly.
I very much doubt that the issue is the greek letters. I think it’s the code where you have $eval("S%s" % (example)) for text_4 and similar for your text_8. It’s to do with the variable not being updated correctly during the trials and I’m not sure why, but I also wonder if you really need to use variables like this. Maybe this can be handled by conditions files instead. I’m thinking that one file could contain all the words for one ‘trial’ and then you switch files for the different trials. But I’m not sure I really understand the study (given the greek instructions). @dvbridges can probably jump in and help here too
By the way, I think the issue with not synchronising before might have been due to you trying to sync the experiment directly from your desktop. When you create a project it contains all the files in the current folder so your project here contains all your pdfs and files from your desktop. Some computers won’t even allow you to do that, e.g. in computer labs where the desktop is “roaming” because of security issues. I think we might add in future a warning that studies should be in a folder and not on desktop.
OK, here’s a fixed version of your study that does run online and locally. The issue was with the variables that were being set up and altered in your JS environment. There werew several issues that needed fixing:
you had variables that were being set to operate locally in your JS (by including the var statement inside a function they only exist inside that function)
you then specified them as belonging to window to try and get around that (which might have worked but…)
your code to access the values in the text used the values that weren’t assigned to the window (they were just created as var and never updated)
I’ve attached below a working version where the code snippets look almost exactly the same between Py/JS and hopefully you can see how it works now.
This will hopefully be handled soon by our upcoming Py->JS autoconversion