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
I have searched the existing issues, and I could not find an existing issue for this feature
I am requesting a straightforward extension of existing dbt-redshift functionality, rather than a Big Idea better suited to a discussion
Describe the feature
Wherever possible, fully rely on redshift driver for metadata and not SQL info_schema or any postgres system view like pg_*.
This will future-proof the adapter as well as unblock robust support for Datasharing (see #742)
Exhaustive list of related issues and PRs (so far)
Throughout the process, we may discover shortcomings in the driver. As @ryanwaldof says in #742:
If this isn't possible because our APIs don't support your use cases lets chat so we can figure out how to fix this. If we need to provide more info in our APIs, we need to provide more info in our APIs.
In these circumstances, migration to driver for metadata would pause while the driver adds the required information.
Additionally, it is crucially important that these changes are sufficiently tested and performance is considered.
One idea that @colin-rogers-dbt and I discussed was that these metadata-related changes might be merged into a feature branch for longer-term testing with customers before merging and releasing as a new minor version of dbt-redshift.
related issues
The content you are editing has changed. Please copy your edits and refresh the page.
This issue has been marked as Stale because it has been open for 180 days with no activity. If you would like the issue to remain open, please comment on the issue or else it will be closed in 7 days.
Although we are closing this issue as stale, it's not gone forever. Issues can be reopened if there is renewed community interest. Just add a comment to notify the maintainers.
Is this your first time submitting a feature request?
Describe the feature
Wherever possible, fully rely on redshift driver for metadata and not SQL info_schema or any postgres system view like pg_*.
This will future-proof the adapter as well as unblock robust support for Datasharing (see #742)
Exhaustive list of related issues and PRs (so far)
Throughout the process, we may discover shortcomings in the driver. As @ryanwaldof says in #742:
In these circumstances, migration to driver for metadata would pause while the driver adds the required information.
Additionally, it is crucially important that these changes are sufficiently tested and performance is considered.
One idea that @colin-rogers-dbt and I discussed was that these metadata-related changes might be merged into a feature branch for longer-term testing with customers before merging and releasing as a new minor version of dbt-redshift.
related issues
Tasks
adapter.get_relation
for relations in other databases [RA3] #179adapter.get_columns_in_relation
does not work cross database #736Describe alternatives you've considered
No response
Who will this benefit?
No response
Are you interested in contributing this feature?
No response
Anything else?
No response
The text was updated successfully, but these errors were encountered: