-- Step 1: Delete duplicate entries, keeping the last one based on timestamp
-- Use a CTE to mark duplicates and delete them efficiently
WITH ranked_snapshots AS (
    SELECT 
        id,
        url,
        ROW_NUMBER() OVER(PARTITION BY url ORDER BY timestamp DESC) AS row_num
    FROM 
        snapshots
)
DELETE FROM snapshots
USING ranked_snapshots
WHERE snapshots.id = ranked_snapshots.id
AND ranked_snapshots.row_num > 1;

-- Step 2: Add a unique constraint on the url column to prevent future duplicates
ALTER TABLE snapshots
ADD CONSTRAINT unique_url UNIQUE (url);