SQL-like Data Functions On Demand

Given a database schema and documentation, implement functions on the spot using the documentation.

Asked in the learning stage. VO2; repeated in the HM technical portion.

Implement the documented PostgreSQL function get_pipeline_health(workspace_id, month_start). Return one row for every active pipeline in the workspace. Include only runs from month_start through the final day of that month. A pipeline with no runs must still appear. failure_rate_pct is failed runs divided by all runs, rounded to two decimal places. Health is no runs for zero runs, healthy for a failure rate at most 10%, and attention otherwise.

Output

1. Columns: pipeline_id, pipeline_name, total_runs, successful_runs, failure_rate_pct, health_status
2. One row per active pipeline in workspace 1 for February 2025
3. Order by pipeline_name ascending

Use the documented function contract and schema below.