CREATE OR REPLACE VIEW optimized_stats AS
WITH
  artist_stats AS (
    SELECT
      TO_CHAR(COUNT(DISTINCT artist_name), 'FM999,999,999') AS artist_count
    FROM
      optimized_listens
    WHERE
      artist_name IS NOT NULL
  ),
  track_stats AS (
    SELECT
      TO_CHAR(COUNT(*), 'FM999,999,999') AS listen_count
    FROM
      optimized_listens
  ),
  concert_stats AS (
    SELECT
      TO_CHAR(COUNT(*), 'FM999,999,999') AS concert_count
    FROM
      concerts
  ),
  venue_stats AS (
    SELECT
      TO_CHAR(COUNT(DISTINCT venue), 'FM999,999,999') AS venue_count
    FROM
      concerts
  ),
  yearly_data AS (
    SELECT
      EXTRACT(
        YEAR
        FROM
          e.last_watched_at
      ) AS year,
      0 AS artist_count,
      0 AS listen_count,
      0 AS genre_count,
      COUNT(DISTINCT e.show) AS show_count,
      COUNT(*) AS episode_count,
      0 AS post_count,
      0 AS link_count,
      0 AS book_count,
      0 AS movie_count,
      0 AS concert_count,
      0 AS venue_count
    FROM
      episodes e
    GROUP BY
      EXTRACT(
        YEAR
        FROM
          e.last_watched_at
      )
    HAVING
      EXTRACT(
        YEAR
        FROM
          e.last_watched_at
      ) >= 2023
    UNION ALL
    SELECT
      EXTRACT(
        YEAR
        FROM
          p.date
      ) AS year,
      0 AS artist_count,
      0 AS listen_count,
      0 AS genre_count,
      0 AS show_count,
      0 AS episode_count,
      COUNT(*) AS post_count,
      0 AS link_count,
      0 AS book_count,
      0 AS movie_count,
      0 AS concert_count,
      0 AS venue_count
    FROM
      optimized_posts p
    GROUP BY
      EXTRACT(
        YEAR
        FROM
          p.date
      )
    HAVING
      EXTRACT(
        YEAR
        FROM
          p.date
      ) >= 2023
    UNION ALL
    SELECT
      EXTRACT(
        YEAR
        FROM
          o.date
      ) AS year,
      0 AS artist_count,
      0 AS listen_count,
      0 AS genre_count,
      0 AS show_count,
      0 AS episode_count,
      0 AS post_count,
      COUNT(*) AS link_count,
      0 AS book_count,
      0 AS movie_count,
      0 AS concert_count,
      0 AS venue_count
    FROM
      optimized_links o
    GROUP BY
      EXTRACT(
        YEAR
        FROM
          o.date
      )
    HAVING
      EXTRACT(
        YEAR
        FROM
          o.date
      ) >= 2023
    UNION ALL
    SELECT
      EXTRACT(
        YEAR
        FROM
          b.date_finished
      ) AS year,
      0 AS artist_count,
      0 AS listen_count,
      0 AS genre_count,
      0 AS show_count,
      0 AS episode_count,
      0 AS post_count,
      0 AS link_count,
      COUNT(*) AS book_count,
      0 AS movie_count,
      0 AS concert_count,
      0 AS venue_count
    FROM
      optimized_books b
    WHERE
      LOWER(b.status) = 'finished'
    GROUP BY
      EXTRACT(
        YEAR
        FROM
          b.date_finished
      )
    HAVING
      EXTRACT(
        YEAR
        FROM
          b.date_finished
      ) >= 2023
    UNION ALL
    SELECT
      EXTRACT(
        YEAR
        FROM
          m.last_watched
      ) AS year,
      0 AS artist_count,
      0 AS listen_count,
      0 AS genre_count,
      0 AS show_count,
      0 AS episode_count,
      0 AS post_count,
      0 AS link_count,
      0 AS book_count,
      COUNT(*) AS movie_count,
      0 AS concert_count,
      0 AS venue_count
    FROM
      optimized_movies m
    GROUP BY
      EXTRACT(
        YEAR
        FROM
          m.last_watched
      )
    HAVING
      EXTRACT(
        YEAR
        FROM
          m.last_watched
      ) >= 2023
    UNION ALL
    SELECT
      EXTRACT(
        YEAR
        FROM
          TO_TIMESTAMP(l.listened_at)
      ) AS year,
      COUNT(DISTINCT l.artist_name) AS artist_count,
      COUNT(l.id) AS listen_count,
      COUNT(DISTINCT l.genre_name) AS genre_count,
      0 AS show_count,
      0 AS episode_count,
      0 AS post_count,
      0 AS link_count,
      0 AS book_count,
      0 AS movie_count,
      0 AS concert_count,
      0 AS venue_count
    FROM
      optimized_listens l
    GROUP BY
      EXTRACT(
        YEAR
        FROM
          TO_TIMESTAMP(l.listened_at)
      )
    HAVING
      EXTRACT(
        YEAR
        FROM
          TO_TIMESTAMP(l.listened_at)
      ) >= 2023
    UNION ALL
    SELECT
      EXTRACT(
        YEAR
        FROM
          c.date
      ) AS year,
      0 AS artist_count,
      0 AS listen_count,
      0 AS genre_count,
      0 AS show_count,
      0 AS episode_count,
      0 AS post_count,
      0 AS link_count,
      0 AS book_count,
      0 AS movie_count,
      COUNT(*) AS concert_count,
      COUNT(DISTINCT c.venue) AS venue_count
    FROM
      concerts c
    GROUP BY
      EXTRACT(
        YEAR
        FROM
          c.date
      )
    HAVING
      EXTRACT(
        YEAR
        FROM
          c.date
      ) >= 2023
  ),
  aggregated_yearly_stats AS (
    SELECT
      year,
      SUM(artist_count) AS artist_count,
      SUM(listen_count) AS listen_count,
      SUM(genre_count) AS genre_count,
      SUM(show_count) AS show_count,
      SUM(episode_count) AS episode_count,
      SUM(post_count) AS post_count,
      SUM(link_count) AS link_count,
      SUM(book_count) AS book_count,
      SUM(movie_count) AS movie_count,
      SUM(concert_count) AS concert_count,
      SUM(venue_count) AS venue_count
    FROM
      yearly_data
    GROUP BY
      year
    ORDER BY
      year DESC
  )
SELECT
  (
    SELECT
      artist_count
    FROM
      artist_stats
  ) AS artist_count,
  (
    SELECT
      listen_count
    FROM
      track_stats
  ) AS listen_count,
  (
    SELECT
      concert_count
    FROM
      concert_stats
  ) AS concert_count,
  (
    SELECT
      venue_count
    FROM
      venue_stats
  ) AS venue_count,
  (
    SELECT
      TO_CHAR(COUNT(DISTINCT e.show), 'FM999,999,999')
    FROM
      episodes e
  ) AS show_count,
  (
    SELECT
      TO_CHAR(COUNT(*), 'FM999,999,999')
    FROM
      episodes e
  ) AS episode_count,
  (
    SELECT
      TO_CHAR(COUNT(*), 'FM999,999,999')
    FROM
      optimized_posts
  ) AS post_count,
  (
    SELECT
      TO_CHAR(COUNT(*), 'FM999,999,999')
    FROM
      optimized_links
  ) AS link_count,
  (
    SELECT
      TO_CHAR(COUNT(*), 'FM999,999,999')
    FROM
      optimized_books
    WHERE
      LOWER(status) = 'finished'
  ) AS book_count,
  (
    SELECT
      TO_CHAR(COUNT(*), 'FM999,999,999')
    FROM
      optimized_movies
    WHERE
      last_watched IS NOT NULL
  ) AS movie_count,
  (
    SELECT
      TO_CHAR(COUNT(DISTINCT genre_name), 'FM999,999,999')
    FROM
      optimized_listens
    WHERE
      genre_name IS NOT NULL
  ) AS genre_count,
  JSON_AGG(
    JSON_BUILD_OBJECT(
      'year',
      ys.year,
      'artist_count',
      CASE
        WHEN ys.artist_count > 0 THEN TO_CHAR(ys.artist_count, 'FM999,999,999')
        ELSE NULL
      END,
      'listen_count',
      CASE
        WHEN ys.listen_count > 0 THEN TO_CHAR(ys.listen_count, 'FM999,999,999')
        ELSE NULL
      END,
      'genre_count',
      CASE
        WHEN ys.genre_count > 0 THEN TO_CHAR(ys.genre_count, 'FM999,999,999')
        ELSE NULL
      END,
      'show_count',
      CASE
        WHEN ys.show_count > 0 THEN TO_CHAR(ys.show_count, 'FM999,999,999')
        ELSE NULL
      END,
      'episode_count',
      CASE
        WHEN ys.episode_count > 0 THEN TO_CHAR(ys.episode_count, 'FM999,999,999')
        ELSE NULL
      END,
      'post_count',
      CASE
        WHEN ys.post_count > 0 THEN TO_CHAR(ys.post_count, 'FM999,999,999')
        ELSE NULL
      END,
      'link_count',
      CASE
        WHEN ys.link_count > 0 THEN TO_CHAR(ys.link_count, 'FM999,999,999')
        ELSE NULL
      END,
      'book_count',
      CASE
        WHEN ys.book_count > 0 THEN TO_CHAR(ys.book_count, 'FM999,999,999')
        ELSE NULL
      END,
      'movie_count',
      CASE
        WHEN ys.movie_count > 0 THEN TO_CHAR(ys.movie_count, 'FM999,999,999')
        ELSE NULL
      END,
      'concert_count',
      CASE
        WHEN ys.concert_count > 0 THEN TO_CHAR(ys.concert_count, 'FM999,999,999')
        ELSE NULL
      END,
      'venue_count',
      CASE
        WHEN ys.venue_count > 0 THEN TO_CHAR(ys.venue_count, 'FM999,999,999')
        ELSE NULL
      END
    )
  ) AS yearly_breakdown
FROM
  aggregated_yearly_stats ys;