End Redirect URL not working properly
Hello developers,
I'm running an online experiment on Prolific, which requires the experiment to redirect to a Prolific link once finished. I pasted the link in the 'end redirect url' box in the properties of my imported study, but it doesn't seem to be working as it should.
All studies that made it to the end got an error when they were supposed to be redirected to Prolific. See the image attached; under 'study code' should be an actual completion code, which is not there for any of the participants.
Cheers,
Jasper
Comments
The experiment was also set to full-screen mode, in case that matters.
Hi Jasper,
I just tried it on cortex.jatos.org and it seemed to have worked with me. Attached is a screenshot of the URL I put in JATOS in the study's properties and underneath is the screenshot of the Prolific page after the study was done. But I just tried with the Preview in Prolific - but I assume this works the same as the real thing.
Can you give me advice in how to replicate your issue?
Best,
Kristian
The only things I can think of that may have been different:
I'm really no expert, but I can email you whatever information/files you need if that helps.
Hi Jasper,
>Fullscreen mode, though it seems farfetched
Can't imagine. It's a redirect to the Prolific page. But to be sure you can try to run your study without fullscreen mode via Prolific.
>Running through a General Multiple Worker link (my bet)
A General Multiple is what I was using.
>Our server
I can't imagine since the redirect happens in the browser (in jatos.js).
The exeriment
Maybe. I don't know how. But if you send me the study, I'll try it out on my server.
Best,
Kristian
I've sent you the study.
And I just checked, without fullscreen mode the problem persists.
Sorry Jasper, I didn't get the notification about your answer (usually I get an email). I just got back here coincidentally because I was looking at your other issue (or someone with the same name?). And I also didn't get the study. I'm not sure what's going on. If your problem is still there, can you just sent me again your study. I'll monitor this thread here to see whether you answered.
Best,
Kristian
I've just sent you the files again. If you don't have them, could you share your email address with me?
The problem is sort of 'solved' now because @Yavor manually changed the server files to redirect to prolific. This is a work-around that he's been using for a while, but if the redirect link in JATOS would work it would save him a lot of time.
I've got the files. I also figured out the mystery of why I didn't get notification emails anymore: some of them got into my spam folder. Weirdly only some got into spam while others were delivered properly. I'll try it out at come back here.
Okay, I checked the URL redirect with your study and it worked here too. I used the cortex.jatos.org server. I did nothing special: 1) I added Prolific redirect URL in the properties of your study. 2) Run your study with the Preview button from Prolific. 3) Got redirected to Prolific.
I'm a bit lost - don't know what's happening with your JATOS/OSWeb experiment. What does your JATOS' log say, any errors in there?
Kristian
That is weird. I tested again with the shorter version that I also sent to you, and got the same errors.
I will also send the data file your way. It looks fine to me. Could it be something to do with our server?
Best, Jasper
Weird, indeed. What irritates me is this error msg you keep getting: "Errors occurred during the experiment". Although it is displayed by JATOS, it originates in OSWeb. Apart from being extremely non-specific, it still hints to some errors during your OSWeb study (that do not happen when I run it my browser).
Would it be possible for me to test your study on your server? Could you give me General Multiple link that I can then test in Prolific?
Best
Kristian
Of course, I'll send it to you!
Edit: We are also running a bunch of other experiments on this server, though, so I trust that you won't be doing anything that'll overload the server.
Just tried it out on your server. In short: The Prolific redirect seem to have worked with me but I've got OSWeb errors.
The good news: I've got redirected to Prolific in the end.
And the bad news: There were OSWeb errors in the browser console: OSWeb has stopped running due to a fatal error. and osweb.1.3.3.bundle.js:1 Script parsing error: Unexpected character ' '.
This would explain your JATOS end page: A problem occurred: ... - I'm not sure though why I didn't get this end page.
With OSWeb errors I'm not the expert. Sorry. Can't help there ;(
Hope you find a solution,
Kristian
Hmm, okay. I'm not sure I understand this correctly. What you're saying is the redirect is not working due to some errors in the experiment?
But how come you get correctly redirected and I don't? We're running the same experiment on the same server, just on a different pc.
This is a mystery to me too. Maybe it's the browser? I'm using Chrome. I have no clue.
I'm using Chrome too. (Edit: on a mac)
I'll see if we can try the redirect link thing with a different experiment, that hopefully doesn't have any errors, and see if it works.
Hi Jasper, I'm actually curious. Where you able to pin point the error?
Kristian
I've already finished collecting data for this experiment, so I haven't really looked for the error. I figured I'd try again with a new experiment, some time later on.
Cheers,
Jasper
I still don't know what caused the problem, but at least it seems to be solved now :)
At least it is solved now :D
Hi @JasperdeWaard, you mentioned earlier in the thread that @Yavor manually changed the server files to redirect to prolific. I was wondering if one of you two can tell me how you did that, since I believe I need to do that as well. I'm using my university's JATOS server, and unfortunately it is an old version in which it's not possible to simply paste in the Prolific end redirect URL into JATOS.
Hi @phidget, maybe this page offers some help with this: https://osdoc.cogsci.nl/3.3/manual/osweb/prolific/. I do think you will need the jatos version that support the custom redirect url though, or you indeed need to make some changes to the osweb sourcefiles to make sure that you get redirected back to prolific at the end of the experiment.