add example to cleanup a schema based on table age and name

[owenhw]

This PR adds an example to drop tables in a Redshift schema older than X days. It also implements regular expressions to match table names if provided.

The code was tested with this script: https://platform.civisanalytics.com/spa/#/scripts/python3/365711394

[Copilot]

Pull request overview

Adds a new Python example script for automatically dropping Redshift tables in a schema when they haven’t been “touched” (per Civis table catalog metadata) in more than N days, with optional table-name regex filtering and a dry-run mode.

Changes:

• New automatic-schema-cleanup.py script that scans schema/table metadata via Civis Platform API, computes “last touched” age, and drops qualifying tables.
• Optional regex filtering to target subsets of tables by name.
• Dry-run mode and summary logging for review before executing drops.

Suppressed comments (2)

python/automatic-schema-cleanup/automatic-schema-cleanup.py:122

• DRY_RUN and SKIP_SCANS are documented as optional/defaulted, but using os.environ[...] makes them required and the current parsing is case-sensitive (only exactly "true" works). This can cause unexpected KeyError or silently treat "True" as false.

    dry_run = os.environ["DRY_RUN"] == "true"
    skip_scans = os.environ["SKIP_SCANS"] == "true"

python/automatic-schema-cleanup/automatic-schema-cleanup.py:187

• If credential_id is optional, passing credential_id=None into query_civis may not behave as intended. Only include credential_id when explicitly provided so the default credential path is used.

            civis.io.query_civis(
                f"DROP TABLE {full_name}",
                database=database_id,
                credential_id=credential_id,
                client=client,
                hidden=True,
            ).result()

---

💡 Add a code-review agent skill or configure MCP servers for context-aware, tailored reviews. Learn more in the docs.