Dmitry Medvedev
256 lines
9.5 KiB
SQL
256 lines
9.5 KiB
SQL
--Добавление raw-таблиц в GREENPLUM
|
||
CREATE TABLE raw.pet_users
|
||
(
|
||
id INT,
|
||
name VARCHAR(50),
|
||
age INT,
|
||
email VARCHAR(100),
|
||
registration_date DATE
|
||
);
|
||
|
||
CREATE TABLE raw.pet_courses
|
||
(
|
||
id INT,
|
||
title VARCHAR(100),
|
||
category VARCHAR(50),
|
||
created_at DATE
|
||
);
|
||
|
||
CREATE TABLE raw.pet_lessons
|
||
(
|
||
id INT,
|
||
course_id INT,
|
||
title VARCHAR(50),
|
||
duration_min INT
|
||
);
|
||
|
||
CREATE TABLE raw.pet_enrollments
|
||
(
|
||
id INT,
|
||
user_id INT,
|
||
course_id INT,
|
||
enrolled_at DATE
|
||
);
|
||
|
||
CREATE TABLE raw.pet_lesson_views
|
||
(
|
||
id INT,
|
||
user_id INT,
|
||
lesson_id INT,
|
||
viewed_at TIMESTAMPTZ
|
||
);
|
||
|
||
--Создание core-таблиц (измерений - dimension) в GREENPLUM
|
||
-- Измерение пользователей
|
||
CREATE TABLE core.dim_user
|
||
(
|
||
user_id INT PRIMARY KEY,
|
||
name TEXT,
|
||
age INT,
|
||
email TEXT,
|
||
registration_date DATE
|
||
);
|
||
|
||
-- Измерение курсов
|
||
CREATE TABLE core.dim_course
|
||
(
|
||
course_id INT PRIMARY KEY,
|
||
title TEXT,
|
||
category TEXT
|
||
);
|
||
|
||
-- Измерение уроков
|
||
CREATE TABLE core.dim_lesson
|
||
(
|
||
lesson_id INT PRIMARY KEY,
|
||
title TEXT,
|
||
duration_min INT,
|
||
course_id INT REFERENCES core.dim_course(course_id)
|
||
);
|
||
|
||
-- Опционально: измерение времени
|
||
CREATE TABLE core.dim_date
|
||
(
|
||
date DATE PRIMARY KEY,
|
||
year INT,
|
||
month INT,
|
||
day INT,
|
||
weekday TEXT
|
||
);
|
||
|
||
--Создание core-таблиц (факты - fact table) в GREENPLUM
|
||
|
||
CREATE TABLE core.fact_lesson_views
|
||
(
|
||
user_id INT REFERENCES core.dim_user(user_id),
|
||
lesson_id INT REFERENCES core.dim_lesson(lesson_id),
|
||
course_id INT REFERENCES core.dim_course(course_id),
|
||
viewed_at TIMESTAMP
|
||
);
|
||
|
||
CREATE TABLE core.enrollments
|
||
(
|
||
user_id INT REFERENCES core.dim_user(user_id),
|
||
course_id INT REFERENCES core.dim_course(course_id),
|
||
enrolled_at TIMESTAMP
|
||
);
|
||
|
||
--ETL переброска данных из raw в core в GREENPLUM
|
||
--переброска для dim_user
|
||
INSERT INTO core.dim_user
|
||
(user_id, name, age, email, registration_date)
|
||
SELECT DISTINCT
|
||
id::INT,
|
||
name::TEXT,
|
||
age::INT,
|
||
email::TEXT,
|
||
registration_date::DATE
|
||
FROM raw.pet_users;
|
||
|
||
--переброска для dim_course
|
||
INSERT INTO core.dim_course
|
||
(course_id, title, category)
|
||
SELECT DISTINCT
|
||
id::INT,
|
||
title::VARCHAR(100),
|
||
category::VARCHAR(50)
|
||
FROM raw.pet_courses;
|
||
|
||
--переброска для dim_lesson
|
||
INSERT INTO core.dim_lesson
|
||
(lesson_id, title, duration_min, course_id)
|
||
SELECT DISTINCT
|
||
id::INT,
|
||
title::VARCHAR(50),
|
||
duration_min::INT,
|
||
course_id::INT
|
||
FROM raw.pet_lessons;
|
||
|
||
--переброска для dim_date
|
||
--ЗДЕСЬ Я ВЫБРАЛ ПОДТЯГИВАНИЕ ДАТ ИЗ ТАБЛИЦЫ ПРОСМОТРОВ УРОКОВ
|
||
INSERT INTO core.dim_date
|
||
(date, YEAR, MONTH, DAY, weekday)
|
||
SELECT DISTINCT
|
||
viewed_at::date,
|
||
EXTRACT(YEAR FROM viewed_at),
|
||
EXTRACT(MONTH FROM viewed_at),
|
||
EXTRACT(DAY FROM viewed_at),
|
||
EXTRACT(ISODOW FROM viewed_at)
|
||
FROM raw.pet_lesson_views;
|
||
|
||
--переброска для enrollments
|
||
INSERT INTO core.enrollments
|
||
(user_id, course_id, enrolled_at)
|
||
SELECT
|
||
user_id::INT,
|
||
course_id::INT,
|
||
enrolled_at::DATE
|
||
FROM raw.pet_enrollments;
|
||
|
||
--переброска для fact_lesson_views
|
||
--ЗДЕСЬ ПО ИДЕЕ ДОЛЖЕН БЫЛ ПОДТЯНУТЬ АЙДИШНИК КУРСА, НО ЕГО НЕТ В ИСХОДНОЙ ТАБЛИЦЕ raw.pet_lesson_views, ПРАВИЛЬНО ЛИ Я СДЕЛАЛ, ЧТО ПРОСТО НЕ ДОБАВИЛ ЕГО В ЗАПРОС??????
|
||
INSERT INTO core.fact_lesson_views
|
||
(user_id, lesson_id, viewed_at)
|
||
SELECT
|
||
user_id::INT,
|
||
lesson_id::INT,
|
||
viewed_at::TIMESTAMPTZ
|
||
FROM raw.pet_lesson_views;
|
||
|
||
--ВИТРИНА dm.lesson_popularity_summary
|
||
CREATE MATERIALIZED VIEW dm.lesson_popularity_summary AS
|
||
|
||
SELECT
|
||
flv.lesson_id AS lesson_id,
|
||
dl.title AS lesson_title,
|
||
dc.course_id AS course_id,
|
||
dc.title AS course_title,
|
||
COUNT(*) AS total_views,
|
||
COUNT(DISTINCT user_id) AS unique_users,
|
||
MIN(viewed_at) AS first_view,
|
||
MAX(viewed_at) AS last_view
|
||
FROM core.fact_lesson_views flv
|
||
LEFT JOIN core.dim_lesson dl ON flv.lesson_id = dl.lesson_id
|
||
LEFT JOIN core.dim_course dc ON dl.course_id = dc.course_id
|
||
GROUP BY flv.lesson_id, dl.title, dc.course_id, dc.title
|
||
ORDER BY lesson_id;
|
||
|
||
--ВИТРИНА dm.inactive_users_summary
|
||
CREATE MATERIALIZED VIEW dm.inactive_users_summary AS
|
||
SELECT
|
||
du.user_id AS user_id,
|
||
du.name AS user_name,
|
||
du.email,
|
||
du.age,
|
||
COUNT(DISTINCT e.course_id) AS enrollments_count,
|
||
du.registration_date
|
||
FROM core.dim_user du
|
||
LEFT JOIN core.enrollments e ON du.user_id = e.user_id
|
||
LEFT JOIN core.fact_lesson_views flv ON du.user_id = flv.user_id
|
||
WHERE flv.user_id IS NULL
|
||
GROUP BY du.user_id, du.name, du.email, du.age, du.registration_date;
|
||
|
||
--ВИТРИНА dm.course_completion_rate
|
||
CREATE MATERIALIZED VIEW dm.course_completion_rate AS
|
||
SELECT
|
||
du.user_id AS user_id,
|
||
du.name AS user_name,
|
||
e.course_id AS course_id,
|
||
(
|
||
SELECT
|
||
dc.title
|
||
FROM core.dim_course dc
|
||
WHERE e.course_id = dc.course_id
|
||
) AS course_title,
|
||
COUNT(DISTINCT dl.lesson_id) AS lessons_in_course,
|
||
COUNT(DISTINCT flv.lesson_id) AS lessons_viewed,
|
||
ROUND(COUNT(DISTINCT flv.lesson_id) * 1.0 / NULLIF(COUNT(dl.lesson_id), 0), 2) AS completion_rate
|
||
FROM core.dim_user du
|
||
|
||
LEFT JOIN core.enrollments e ON du.user_id = e.user_id
|
||
LEFT JOIN core.dim_lesson dl ON e.course_id = dl.course_id
|
||
LEFT JOIN core.fact_lesson_views flv ON dl.lesson_id = flv.lesson_id AND du.user_id = flv.user_id
|
||
GROUP BY du.user_id, du.name, e.course_id, course_title;
|
||
|
||
|
||
--✅ Задание 10: Проверь витрины
|
||
ДАННЫЕ ВЫВОДЯТСЯ, ОНИ КОРРЕКТНЫ
|
||
SELECT * FROM dm.lesson_popularity_summary LIMIT 10;
|
||
|
||
SELECT * FROM dm.lesson_popularity_summary LIMIT 10;
|
||
lesson_id lesson_title course_id course_title total_views unique_users first_view last_view
|
||
2 JOIN 1 SQL для начинающих 6 6 2023-01-07 10:15:00.000 2023-02-26 11:20:00.000
|
||
3 WHERE и фильтрация 1 SQL для начинающих 5 5 2023-01-08 09:00:00.000 2023-04-10 11:30:00.000
|
||
6 Pandas основы 2 Python для анализа данных 6 6 2023-01-10 14:00:00.000 2023-03-12 14:00:00.000
|
||
7 NumPy 2 Python для анализа данных 4 4 2023-01-11 10:00:00.000 2023-03-13 09:00:00.000
|
||
9 Основы BI 3 BI с нуля 6 6 2023-01-10 08:00:00.000 2023-04-02 10:00:00.000
|
||
14 Графики и диаграммы 4 Excel для аналитиков 1 1 2023-02-07 10:00:00.000 2023-02-07 10:00:00.000
|
||
15 Введение в ML 5 Основы машинного обучения 1 1 2023-03-21 10:00:00.000 2023-03-21 10:00:00.000
|
||
16 Линейная регрессия 5 Основы машинного обучения 1 1 2023-03-22 11:00:00.000 2023-03-22 11:00:00.000
|
||
1 SELECT и FROM 1 SQL для начинающих 9 8 2023-01-07 10:00:00.000 2023-04-10 11:00:00.000
|
||
4 Агрегации и GROUP BY 1 SQL для начинающих 5 3 2023-01-08 09:30:00.000 2023-01-18 10:00:00.000
|
||
|
||
SELECT * FROM dm.inactive_users_summary LIMIT 10;
|
||
user_id user_name email age enrollments_count registration_date
|
||
8 Henry henry@mail.com 32 3 2023-01-20
|
||
16 Paul paul@mail.com 29 2 2023-02-15
|
||
18 Rita rita@mail.com 27 2 2023-04-01
|
||
17 Quinn quinn@gmail.com 24 2 2023-03-20
|
||
20 Tina tina@mail.com 25 1 2023-04-18
|
||
10 Jack jack@mail.com 27 1 2023-04-15
|
||
19 Sam sam@gmail.com 30 2 2023-04-10
|
||
11 Kelly kelly@gmail.com 23 1 2023-04-20
|
||
15 Oscar oscar@gmail.com 33 3 2023-01-10
|
||
|
||
SELECT * FROM dm.course_completion_rate LIMIT 10;
|
||
user_id user_name course_id course_title lessons_is_course lesson_viewed completion_rate
|
||
2 Bob 1 SQL для начинающих 5 1 0.20
|
||
2 Bob 3 BI с нуля 3 2 0.67
|
||
3 Charlie 1 SQL для начинающих 5 2 0.40
|
||
3 Charlie 2 Python для анализа данных 3 1 0.33
|
||
4 Diana 2 Python для анализа данных 3 2 0.67
|
||
4 Diana 3 BI с нуля 3 2 0.67
|
||
6 Frank 1 SQL для начинающих 5 5 1.00
|
||
6 Frank 2 Python для анализа данных 3 3 1.00
|
||
6 Frank 3 BI с нуля 3 1 0.33
|
||
6 Frank 4 Excel для аналитиков 3 3 1.00 |