🐛 gbd-risk: Add missing percentages

[Marigold]

Percentages from gbd_risk dataset such as this one got removed by this commit. Because of that, ETL tries to rebuild gbd_risk every time it runs and then throws a warning.

This PR adds run_percent=True argument to regional aggregates to put them back. It also raises an error on staging servers when there are variables that cannot be deleted automatically.

[owidbot]

Quick links (staging server):

Site	Admin	Wizard

Login: ssh owid@staging-site-fix-gbd-risk

chart-diff: ✅

No charts for review.

data-diff: ✅ No differences found

Legend: +New  ~Modified  -Removed  =Identical  Details
Hint: Run this locally with etl diff REMOTE data/ --include yourdataset --verbose --snippet

Automatically updated datasets matching weekly_wildfires|excess_mortality|covid|fluid|flunet|country_profile|garden/ihme_gbd/2019/gbd_risk are not included

Edited: 2024-09-10 09:04:27 UTC
Execution time: 12.92 seconds

[spoonerf]

Ah thanks for finding this! So we don't want regional aggregates for percentages for gbd_risk as there's some weird stuff going on their (some 'risks' reduce your chance of dying), so run_percent has to stay False.

But I think there was a broader issue in shared.py, meaning that all aggregate percentages were getting dropped unless run_percent = True, which isn't what we wanted.

Have changed it now, I think that also fixes the original issue?

[Marigold]

Have changed it now, I think that also fixes the original issue?

Yep, much better!