You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
Repository navigation
Auto-Generated Count field unstable during TSC datasource publish #1625
Describe the bug
Tableau's autogenerated count field is not recognized when refreshing a hyper and publishing it to Cloud.
Publishing with TSC causes that field to be regenerated when opening a web edit workbook or a local workbook that is connected to the published datasource. The original count field becomes an invalid field and any calc field referencing it breaks.
I think I've narrowed it down to the publish portion of the script.
Versions
Tableau Cloud
Python version: 3.11.8
TSC library version: 0.37
To Reproduce
Create a new simple hyper and publish it. I published it from Desktop to Cloud.
Set the workbook to use as datasource during publish.
Connect another workbook to the local hyper file.
Run refresh and publish
Only the workbooks that are connected to the published version of the datasource see the error. This includes a web edit version and the local desktop version of the workbook.
On recommendation, I used TSC.update_hyper_dataaction:replace and this corrected the error in Cloud and local workbooks.
running the original script on the original broken then fixed datasource again changes the outcome a little bit. Calculated fields that reference the Extract(count) field break, the invalid version of the field is not present, but a working version is and the Number of Records field appears.
Results
The original count field becomes an invalid field and any calc field referencing it breaks.
NOTE: Be careful not to post user names, passwords, auth tokens or any other private or sensitive information.
This issue has been automatically marked as stale because it has not had recent activity. It will be closed in 14 days if no further activity occurs. If this is still relevant, please comment to keep it open.
Closing as not-planned. The auto-generated Extract(count) field being invalidated after a refreshed .hyper publish is a behavior inside the extract file itself — that surface is owned by the Hyper API (https://tableau.github.io/hyper-db/docs/hyper-api/), not TSC. TSC just uploads the bytes; it doesn't touch the internal structure of the extract. If you can identify a case where TSC is corrupting the file during upload (rather than the file already being in this state on disk), please open a fresh issue with a reproducer.
Activity
github-actions commentedon Aug 18, 2026
This issue has been automatically marked as stale because it has not had recent activity. It will be closed in 14 days if no further activity occurs. If this is still relevant, please comment to keep it open.
jacalata commentedon Aug 25, 2026
Closing as not-planned. The auto-generated
Extract(count)field being invalidated after a refreshed.hyperpublish is a behavior inside the extract file itself — that surface is owned by the Hyper API (https://tableau.github.io/hyper-db/docs/hyper-api/), not TSC. TSC just uploads the bytes; it doesn't touch the internal structure of the extract. If you can identify a case where TSC is corrupting the file during upload (rather than the file already being in this state on disk), please open a fresh issue with a reproducer.