Files
homework/itikhomirov/clickhouse/Pet_project.sql
T
2025-12-07 23:03:34 +03:00

125 lines
3.0 KiB
SQL

CREATE database pet;
-- Этап 1
-- Задание 1
-- Перенести данные как в пет-проекте с Greenplum
-- Создаем таблицы, чтобы потом загрузить в них данные из таблиц GreenPlum
-- Таблица user из GP
CREATE TABLE pet.core_user (
user_id Int32,
name String,
age Int32,
email String,
registration_date Date
)
ENGINE = MergeTree
ORDER BY user_id;
-- Таблица course из GP
CREATE TABLE pet.core_course (
course_id Int32,
title String,
category String
)
ENGINE = MergeTree
ORDER BY course_id;
-- Таблица lesson из GP
CREATE TABLE pet.core_lesson (
lesson_id Int32,
title String,
duration_min Int32,
course_id Int32
)
ENGINE = MergeTree
ORDER BY lesson_id;
-- Таблица fact_lesson_views из GP
CREATE TABLE pet.core_fact_lesson_views (
user_id Int32,
lesson_id Int32,
course_id Int32,
viewed_at DateTime64(3)
)
ENGINE = MergeTree
ORDER BY user_id;
-- Таблица enrollments из GP
CREATE TABLE pet.core_enrollments (
id Int32,
user_id Int32,
course_id Int32,
enrolled_at Date
)
ENGINE = MergeTree
ORDER BY user_id;
-- Этап 2. Создание витринных таблиц
-- Задание 2: Подготовь структуру витрины lesson_popularity_summary
-- Создаем пустую таблицу lesson_popularity_summary(структуру витрины)
CREATE TABLE pet.lesson_popularity_summary
(
lesson_id UInt32,
lesson_title String,
course_id UInt32,
course_title String,
total_views UInt32,
unique_users UInt32,
first_view DateTime64(3, 'UTC'),
last_view DateTime64(3, 'UTC')
)
ENGINE = MergeTree
ORDER BY (lesson_id, course_id);
-- Создаем Materialized View
CREATE MATERIALIZED VIEW pet.mv_lesson_popularity_summary
TO pet.lesson_popularity_summary
AS
SELECT
flv.lesson_id AS lesson_id,
l.title AS lesson_title,
flv.course_id AS course_id,
c.title AS course_title,
count() AS total_views,
uniqExact(flv.user_id) AS unique_users,
min(flv.viewed_at) AS first_view,
max(flv.viewed_at) AS last_view
FROM pet.core_fact_lesson_views AS flv
JOIN pet.core_lesson AS l ON flv.lesson_id = l.lesson_id
JOIN pet.core_course AS c ON flv.course_id = c.course_id
GROUP BY
flv.lesson_id,
l.title,
flv.course_id,
c.title;
-- Вставляем данные в таблицу pet.lesson_popularity_summary
INSERT INTO pet.lesson_popularity_summary
SELECT
flv.lesson_id,
l.title AS lesson_title,
flv.course_id,
c.title AS course_title,
count() AS total_views,
uniqExact(flv.user_id) AS unique_users,
min(flv.viewed_at) AS first_view,
max(flv.viewed_at) AS last_view
FROM pet.core_fact_lesson_views AS flv
JOIN pet.core_lesson AS l ON flv.lesson_id = l.lesson_id
JOIN pet.core_course AS c ON flv.course_id = c.course_id
GROUP BY
flv.lesson_id,
l.title,
flv.course_id,
c.title;