Redetect files types during upgrading to dataverse 4.15

[Michel Bamouni]

Hi everyone

I'm currently plan to upgrade my dataverse from version 4.10.1 to 4.15 and I see in the release notes of 4.15 the needs to "redetect file types".

Here is the documentation of this release : https://github.com/IQSS/dataverse/releases/tag/v4.15

But when I go to the dataverse native api, the redetect file type api seems to use a file ID of PID. 

So I want to know if there is a way to do this in one command call or if I must call the api for each file ?

Best regards,

Michel

[Philip Durbin]

Hi Michel,

I just looked at the current code and the pull request* where we added that "redetect" API endpoint and I'm sorry to say that there is no "redetect all" or "redetect all in dataset" option. As you've observed, you must make an API call for each file.

Good luck with your upgrade.

Thanks,

Phil

* https://github.com/IQSS/dataverse/pull/5853

--
You received this message because you are subscribed to the Google Groups "Dataverse Users Community" group.
To unsubscribe from this group and stop receiving emails from it, send an email to dataverse-commu...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/dataverse-community/28c04936-3842-4e7a-8e33-87775170e482%40googlegroups.com.

--

Philip Durbin
Software Developer for http://dataverse.org
http://www.iq.harvard.edu/people/philip-durbin

[Michel Bamouni]

Hi Philip,

Thanks for your return.

Best regards

To view this discussion on the web visit https://groups.google.com/d/msgid/dataverse-community/CABbxx8FTdPRDFT8kdfahM1eV_Wxk_AEm2hFiWZMoZvejypkAwg%40mail.gmail.com.

[Philip Durbin]

Sure. Don Sizemore opened an issue about wanting "redetect all" API endpoint a while back: https://github.com/IQSS/dataverse/issues/6355

I was reminded of this by a dvcli issue he just opened about iterating through files, datasets, or PIDs: https://github.com/poikilotherm/dvcli/issues/3

To view this discussion on the web visit https://groups.google.com/d/msgid/dataverse-community/CAFovp2rV7248PcMK4uQgiDA3Dcr60COTiY2jB6joF07j7HXp8A%40mail.gmail.com.

[Stefan Kasberger]

Hi,

i wrote some small scripts in Python with pyDataverse to do this. First, I collected all Datafile IDs and wrote them into a JSON file, then I iterated over the IDs and called a curl command for each File. If you still need to solve the problem, I could share the scripts and workflow with you. Just write me at stefan.k...@univie.ac.at.

Greetz, Stefan