Make it easier to clean up results schema
Open
- Dominant language
- Jupyter Notebook
- Stars
- 201
- Forks
- 62
- PR merge metrics
- No merged PRs in 30d
Description
_From @thcrock on May 23, 2017 19:51_
The full models table (and child tables) can get cluttered with aborted/buggy runs.
We should make it easier to delete all the results schema data that we've decided is bad.
This is partially done in https://github.com/dssg/results-schema/blob/master/results_schema/schema.py#L79
There could be an interface that doesn't require a SQLAlchemy session, though. Maybe something that takes an engine/connectable, and an experiment hash.
But maybe it should be a soft delete? We can add an is_deleted column to each of the tables, and cascade a soft delete.
_Copied from original issue: dssg/results-schema#5_
Contributor guide
Assessment
This issue has not been assessed yet.