Closed kevinelliott closed 3 months ago
We need to ensure the indices are created, as well. I think there are some table joins in the views that require indices or they just explode. Also, there's at least one sequence in there, but it might be autogenerated at table create time. We also need to make sure table ownership is correct. I guess that can be handled by the create script when the DB is instantiated.
We need to ensure the indices are created, as well. I think there are some table joins in the views that require indices or they just explode. Also, there's at least one sequence in there, but it might be autogenerated at table create time. We also need to make sure table ownership is correct. I guess that can be handled by the create script when the DB is instantiated.
Indeed, indexes. Feel free to take a pass here and push up a commit if you have a moment. Otherwise, I can tackle. I was mostly waiting until we solidified the structure.
@ab0oo Of course this PR is based on a logical "merging" of each of our original schemas, with the intention to provide similar capabilities while also preserving history data. I would think we'll need a few more tables and perhaps a couple more views.
This looks good to me, i'm lacking a lot of context of the data structures though, so take my input with a grain of salt.
Merging as initial SQL but we can adjust and add to in follow ups.
This is a draft PR intentionally so that we can discuss and shape this initial SQL pass.