Hello,
We are using Dataverse version 6.6
In some of our recently ingested datasets, we have CSV files. In the file list, some show up as CSV files and some show up as the converted TAB file. This made me think the ones showing up as CSV weren't converted, but it looks like they were, as TAB is an option for download and I get an error that they are already converted, if I try to reingest them.
Should all files show up in the list as the TAB version of them selves?
Why would some show up as TAB and some as CSV?
Thanks!
Here is an example where this is happening (you can search on csv in the file section to see the files) https://dataverse.yale.edu/dataset.xhtml?persistentId=doi:10.60600/YU/THFJQA
Actually, I think this might have been something we did through an automated process - where the file's information got changed because of it, after ingest.
It does look a little odd:
![]()
By that I mean that only the W1_attrition_public-1.tab file looks right because the filename is ".tab" and it shows the number of variables.
The csv files all show the number of variables but the filename should show ".tab".
So yeah, maybe your automated process changed something?
Thanks for looking, Phil. I suspect that an "edit" of an already existing file, via the API, somehow caused the file name to get overwritten. (not a dataverse bug, but probably on our end)
gotcha
Last updated: Apr 03 2026 at 06:08 UTC