인프런 커뮤니티 질문&답변

ARRAY(배열) / STRUCT(구조체)

# 1. array_exercises 테이블에서 각 영화(title)별로 장르(genres)를 UNNEST해서 보여주세요
SELECT
  title, 
  genre # UNNEST에서 사용한 새로운 컬럼으로 사용 / 기존 ARRAY_COULMN은 사용 X
FROM advanced.array_exercises 
CROSS JOIN UNNEST(genres) AS genre;

# 2. array_exercises 테이블에서 각 영화(title)별로 배우(actor)와 배역(character)을 보여주세요. 배우와 배역은 별도의 컬럼으로 나와야 합니다
SELECT
  title, 
  actor.actor, 
  actor.character 
FROM advanced.array_exercises
CROSS JOIN UNNEST(actors) AS actor;

# 3. array_exercises 테이블에서 각 영화(title)별로 배우(actor), 배역(character), 장르(genre)를 출력하세요. 한 Row에 배우, 배역, 장르가 모두 표시되어야 합니다
SELECT
  title, 
  actor.actor, 
  actor.character, 
  genre
FROM advanced.array_exercises
CROSS JOIN UNNEST(actors) AS actor
CROSS JOIN UNNEST(genres) AS genre;

# 4. 앱 로그 데이터(app_logs)의 배열을 풀어주세요
SELECT
  user_id, 
  event_date, 
  event_name, 
  user_pseudo_id, 
  event_param.key, 
  event_param.value.string_value, 
  event_param.value.int_value
FROM advanced.app_logs
CROSS JOIN UNNEST(event_params) AS event_param
WHERE 
  event_date = "2022-08-01";

PIVOT(피봇)

: 축을 중심으로 회전한다

# 1. orders테이블에서 유저(user_id)별로 주문 금액(amount)의 합계를 PIVOT해주세요. 날짜(order_date)를 행(Row)으로, user_id를 열(Column)으로 만들어야 합니다
SELECT
  order_date,
  SUM(IF(user_id = 1, amount, 0)) AS user_1,
  SUM(IF(user_id = 2, amount, 0)) AS user_2,
  SUM(IF(user_id = 3, amount, 0)) AS user_3,
FROM advanced.orders
GROUP BY ALL
ORDER BY
  order_date;

# 2. orders 테이블에서 날짜(order_date)별로 유저들의 주문 금액(amount)의 합계를 PIVOT 해주세요. user_id를 행(Row)으로, order_date를 열(Column)으로 만들어야 합니다
SELECT
  user_id,
  SUM(IF(order_date = "2023-05-01", amount, 0)) AS `2023-05-01`,
  SUM(IF(order_date = "2023-05-02", amount, 0)) AS `2023-05-02`,
  SUM(IF(order_date = "2023-05-03", amount, 0)) AS `2023-05-03`,
  SUM(IF(order_date = "2023-05-04", amount, 0)) AS `2023-05-04`,
  SUM(IF(order_date = "2023-05-05", amount, 0)) AS `2023-05-05`
FROM advanced.orders
GROUP BY ALL
ORDER BY
  user_id;

# 3. orders 테이블에서 사용자(user_id)별, 날짜(order_date)별로 주문이 있다면 1, 없다면 0으로 PIVOT 해주세요. user_id를 행(Row)으로, order_date를 열(Column)로 만들고 주문을 많이 해도 1로 처리합니다
SELECT
  user_id, 
  MAX(IF(order_date = "2023-05-01", 1, 0)) AS `2023-05-01`,
  MAX(IF(order_date = "2023-05-02", 1, 0)) AS `2023-05-02`,
  MAX(IF(order_date = "2023-05-03", 1, 0)) AS `2023-05-03`,
  MAX(IF(order_date = "2023-05-04", 1, 0)) AS `2023-05-04`,
  MAX(IF(order_date = "2023-05-05", 1, 0)) AS `2023-05-05`
FROM advanced.orders
GROUP BY ALL
ORDER BY
  user_id;

# 4. app_log를 pivot하기
WITH base AS(
SELECT 
  user_id, 
  event_date,
  event_name,
  event_timestamp,
  user_pseudo_id, 
  MAX(if(event_param.key = "firebase_screen", event_param.value.string_value, NULL)) AS firebase_screen,
  MAX(if(event_param.key ="food_id", event_param.value.int_value, NULL)) AS food_id, 
  MAX(if(event_param.key ="session_id", event_param.value.string_value, NULL)) AS session_id
FROM avdanced.app_logs
CROSS JOIN UNNEST (event_params) AS event_param
WHERE
   event_date ='2022-08-01'
GROUP BY ALL
)

SELECT
  event_date, 
  COUNT(user_id) AS cnt
FROM base
WHERE
  event_name = 'click_cart' 
GROUP BY ALL;

퍼널(Funnel)

: 깔때기 의미 → 넓은 시작점에서 점점 좁하지는 흐름

WITH base AS (SELECT
    event_date,
    event_timestamp,
    event_name,
    user_id,
    user_pseudo_id,
    platform,
    MAX(IF(event_param.key = "firebase_screen", event_param.value.string_value, NULL)) AS fire_screen,
    -- MAX(IF(event_param.key = "food_id", event_param.value.int_value, NULL)) AS food_id,
    MAX(IF(event_param.key = "session_id", event_param.value.string_value, NULL)) AS session_id
  FROM advanced.app_logs
  CROSS JOIN UNNEST(event_params) AS event_param
  WHERE
    event_date BETWEEN "2022-08-01" AND "2022-08-18"
  GROUP BY ALL
), fiter_event_and_concat_event_and_screen AS (
  -- event_name + screen (필요한 이벤트만 WHERE 조건에 걸어서 사용)
  SELECT
    * EXCEPT(event_name, fire_screen, event_timestamp),
    CONCAT(event_name, "-", fire_screen) AS event_name_with_screen,
    DATETIME(TIMESTAMP_MICROS(event_timestamp), 'Asia/Seoul') AS event_datetime
  FROM base
  WHERE
    event_name IN ("screen_view", "click_payment")
)

-- step_number + COUNT
-- step_number : CASE WHEN을 사용해 숫자 지정
SELECT
  event_name_with_screen,
  CASE
    WHEN event_name_with_screen = "screen_view-welcome" THEN 1
    WHEN event_name_with_screen = "screen_view-home" THEN 2
    WHEN event_name_with_screen = "screen_view-food_category" THEN 3
    WHEN event_name_with_screen = "screen_view-restaurant" THEN 4
    WHEN event_name_with_screen = "screen_view-cart" THEN 5
    WHEN event_name_with_screen = "click_payment-cart" THEN 6
  ELSE NULL
  END AS step_number,
  COUNT(DISTINCT user_pseudo_id) AS cnt
FROM fiter_event_and_concat_event_and_screen
GROUP BY ALL
HAVING step_number IS NOT NULL
WHERE
  user_pseudo_id = "1350836585.3421064109"

-- 일자별
SELECT
  event_date,
  event_name_with_screen,
  CASE
    WHEN event_name_with_screen = "screen_view-welcome" THEN 1
    WHEN event_name_with_screen = "screen_view-home" THEN 2
    WHEN event_name_with_screen = "screen_view-food_category" THEN 3
    WHEN event_name_with_screen = "screen_view-restaurant" THEN 4
    WHEN event_name_with_screen = "screen_view-cart" THEN 5
    WHEN event_name_with_screen = "click_payment-cart" THEN 6
  ELSE NULL
  END AS step_number,
  COUNT(DISTINCT user_pseudo_id) AS cnt
FROM fiter_event_and_concat_event_and_screen
GROUP BY ALL
HAVING step_number IS NOT NULL
ORDER BY event_date;