Change Thing type on Inrupt pod browser

As my profile file somehow got corrupted on I am unable to use it for the moment but I was thankfully still able to download my backup data from the UI and reformatted it for the Inrupt pod (replacing the simple <#id> with <>).

Upon uploading it to my pod using the PodBrowser it shows in the UI as Thing Type: application/octet-stream instead of as Thing Type: text/turtle;charset=UTF-8. The file extension is .ttl and I’m 99.9% sure it’s valid Turtle. Does anyone know how I can change its Thing type?

If I try to fetch it from my application using getSolidDataset it fails with a 406 error code.

Thank you.

Unfortunately PodBrowser can only get the MIME type of a file from the browser, so if the browser doesn’t know that it’s a text/turtle file, PodBrowser can’t know it either. (Although I guess one thing you might be able to do, if you’re comfortable with that, is open the “Network” tab of Firefox’s developer tools, look for the request that uploads your file, right-click and choose “Edit and resend”, and then change the Content-Type header yourself.)

(PS. The concept of a “Thing” in solid-client is part of a file. (If it were up to me, I’d have called SolidDatasets ThingSets to clarify that.))

1 Like

Thanks for the suggestion @Vincent . You definitely need Firefox for the edit and resent, good tip! Also in Brave it was hard to find the actual request as pod browser seemed to make many tens of repeated requests (I think this was might have been due to Brave interfering somehow) for other resources like:


I couldn’t manage to get it to work with Firefox, edit and resend. It seemed ok but when I tried to get the data it downloaded it all (you could tell from the time needed to complete the request) but the dataset was empty. I think it was because I didn’t set the header correctly. Pod browser displayed a type of text/turtle but not text/turtle;charset=UTF-8.

In the end I parsed the .ttl doc (thank VSCode for multi cursor and wrap/unwrap lines!), constructed the objects from the json and wrote a “small” client app (99.99% data :joy:) to upload them to