Merge branch 'main' into OS-233
@@ -0,0 +1,2 @@
|
|||||||
|
itikhomirov/.DS_Store
|
||||||
|
.DS_Store
|
||||||
@@ -0,0 +1,50 @@
|
|||||||
|
DROP SCHEMA IF EXISTS edu_platform CASCADE;
|
||||||
|
CREATE SCHEMA edu_platform;
|
||||||
|
|
||||||
|
drop table if exists edu_platform.users;
|
||||||
|
CREATE TABLE edu_platform.users (
|
||||||
|
id SERIAL PRIMARY KEY,
|
||||||
|
name VARCHAR(100) NOT NULL,
|
||||||
|
age INT NOT NULL,
|
||||||
|
email VARCHAR(100) NOT NULL,
|
||||||
|
registration_date DATE NOT NULL
|
||||||
|
);
|
||||||
|
|
||||||
|
drop table if exists edu_platform.courses;
|
||||||
|
CREATE TABLE edu_platform.courses (
|
||||||
|
id SERIAL PRIMARY KEY,
|
||||||
|
title VARCHAR(100) NOT NULL,
|
||||||
|
category VARCHAR(100) NOT NULL,
|
||||||
|
created_at DATE not NULL
|
||||||
|
);
|
||||||
|
|
||||||
|
drop table if exists edu_platform.lessons;
|
||||||
|
create TABLE edu_platform.lessons (
|
||||||
|
id SERIAL PRIMARY KEY,
|
||||||
|
course_id INT NOT NULL,
|
||||||
|
title VARCHAR(100) NOT NULL,
|
||||||
|
duration_min INT NOT null,
|
||||||
|
FOREIGN KEY (course_id) REFERENCES edu_platform.courses(id)
|
||||||
|
);
|
||||||
|
|
||||||
|
drop table if exists edu_platform.enrollments;
|
||||||
|
create TABLE edu_platform.enrollments (
|
||||||
|
id SERIAL PRIMARY KEY,
|
||||||
|
user_id INT NOT NULL,
|
||||||
|
course_id INT NOT NULL,
|
||||||
|
enrolled_at DATE not null,
|
||||||
|
foreign key (user_id) references edu_platform.users(id),
|
||||||
|
foreign key (course_id) references edu_platform.courses(id)
|
||||||
|
);
|
||||||
|
|
||||||
|
drop table if exists edu_platform.lesson_views;
|
||||||
|
create table edu_platform.lesson_views (
|
||||||
|
id SERIAL PRIMARY KEY,
|
||||||
|
user_id INT NOT NULL,
|
||||||
|
lesson_id INT NOT NULL,
|
||||||
|
viewed_at TIMESTAMP not null,
|
||||||
|
foreign key (user_id) references edu_platform.users(id),
|
||||||
|
foreign key (lesson_id) references edu_platform.lessons(id)
|
||||||
|
);
|
||||||
|
|
||||||
|
|
||||||
@@ -0,0 +1,38 @@
|
|||||||
|
INSERT INTO edu_platform.users (id, name, age, email, registration_date) VALUES
|
||||||
|
(1, 'Alice', 25, 'alice@mail.com', '2023-01-10'),
|
||||||
|
(2,'Bob', 30, 'bob@gmail.com', '2023-02-05'),
|
||||||
|
(3,'Charlie', 22, 'charlie@mail.com', '2023-02-20'),
|
||||||
|
(4,'Diana', 28, 'diana@mail.com', '2023-03-01'),
|
||||||
|
(5,'Ethan', 35, 'ethan@gmail.com', '2023-03-10');
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
|
INSERT INTO edu_platform.courses (id, title, category, created_at) VALUES
|
||||||
|
(1, 'SQL для начинающих', 'data', '2023-01-01'),
|
||||||
|
(2, 'Python для анализа данных', 'programming', '2023-01-15'),
|
||||||
|
(3, 'BI с нуля', 'business', '2023-03-01');
|
||||||
|
|
||||||
|
-- Таблица lessons
|
||||||
|
INSERT INTO edu_platform.lessons (id, course_id, title, duration_min) VALUES
|
||||||
|
(1, 1, 'SELECT и FROM', 10),
|
||||||
|
(2, 1, 'JOIN', 15),
|
||||||
|
(3, 2, 'Pandas', 20),
|
||||||
|
(4, 3, 'Основы BI', 12),
|
||||||
|
(5, 3, 'Метрики и дашборды', 18);
|
||||||
|
|
||||||
|
-- Таблица enrollments
|
||||||
|
INSERT INTO edu_platform.enrollments (id, user_id, course_id, enrolled_at) VALUES
|
||||||
|
(1, 1, 1, '2023-01-15'),
|
||||||
|
(2, 1, 2, '2023-02-01'),
|
||||||
|
(3, 2, 1, '2023-01-20'),
|
||||||
|
(4, 3, 2, '2023-03-05'),
|
||||||
|
(5, 4, 3, '2023-04-01');
|
||||||
|
|
||||||
|
-- Таблица lesson_views
|
||||||
|
INSERT INTO edu_platform.lesson_views (id, user_id, lesson_id, viewed_at) VALUES
|
||||||
|
(1, 1, 1, '2023-01-16 10:00:00'),
|
||||||
|
(2, 1, 2, '2023-01-16 10:15:00'),
|
||||||
|
(3, 2, 1, '2023-01-21 09:00:00'),
|
||||||
|
(4, 3, 3, '2023-03-06 12:00:00'),
|
||||||
|
(5, 4, 4, '2023-04-02 10:00:00'),
|
||||||
|
(6, 4, 5, '2023-04-03 10:00:00');
|
||||||
@@ -0,0 +1,74 @@
|
|||||||
|
-- Активность пользователей
|
||||||
|
|
||||||
|
-- До анализа EXPLAIN:
|
||||||
|
drop Materialized VIEW if exists edu_platform.user_activity_summary;
|
||||||
|
create Materialized VIEW edu_platform.user_activity_summary as
|
||||||
|
select
|
||||||
|
u.id as user_id ,
|
||||||
|
u.name as name,
|
||||||
|
COUNT(distinct e.course_id) as enrollment_count,
|
||||||
|
COUNT(distinct lv.lesson_id) as viewed_lessons_count,
|
||||||
|
MAX(lv.viewed_at) as last_view,
|
||||||
|
MIN(lv.viewed_at) as first_view,
|
||||||
|
ROUND(
|
||||||
|
COUNT(DISTINCT l.id) * 100.0 / NULLIF(COUNT(distinct l2.id), 0),
|
||||||
|
2) as engagement_rate
|
||||||
|
from
|
||||||
|
edu_platform.users u
|
||||||
|
left join edu_platform.enrollments e on u.id = e.user_id
|
||||||
|
inner join edu_platform.lesson_views lv on u.id = lv.user_id
|
||||||
|
left join edu_platform.lessons l on lv.lesson_id = l.id
|
||||||
|
left join edu_platform.lessons l2 on e.course_id = l2.course_id
|
||||||
|
group by
|
||||||
|
u.id, u.name;
|
||||||
|
|
||||||
|
-- После анализа EXPLAIN
|
||||||
|
-- Убрал диблирование строк, возникающее при Join-ах.
|
||||||
|
-- Минимизировал DISTINCT.
|
||||||
|
|
||||||
|
|
||||||
|
drop Materialized VIEW if exists edu_platform.user_activity_summary;
|
||||||
|
create Materialized VIEW edu_platform.user_activity_summary as
|
||||||
|
WITH
|
||||||
|
course_lessons AS (
|
||||||
|
SELECT course_id, COUNT(*) AS lesson_count
|
||||||
|
FROM edu_platform.lessons
|
||||||
|
GROUP BY course_id
|
||||||
|
),
|
||||||
|
users_enroll as (
|
||||||
|
select
|
||||||
|
u.id as user_id ,
|
||||||
|
u.name as name,
|
||||||
|
COUNT(DISTINCT e.course_id) as enrollment_count,
|
||||||
|
SUM(cl.lesson_count) as lesson_sum
|
||||||
|
from
|
||||||
|
edu_platform.users u
|
||||||
|
left join edu_platform.enrollments e on u.id = e.user_id
|
||||||
|
left join course_lessons cl on e.course_id = cl.course_id
|
||||||
|
group by u.id, u.name
|
||||||
|
),
|
||||||
|
users_lesson_views as (
|
||||||
|
select
|
||||||
|
lv.user_id,
|
||||||
|
COUNT(distinct lv.lesson_id) as viewed_lessons_count,
|
||||||
|
MAX(lv.viewed_at) as last_view,
|
||||||
|
MIN(lv.viewed_at) as first_view
|
||||||
|
from
|
||||||
|
edu_platform.lesson_views lv
|
||||||
|
group by lv.user_id
|
||||||
|
)
|
||||||
|
select
|
||||||
|
ue.user_id,
|
||||||
|
ue.name,
|
||||||
|
ue.enrollment_count,
|
||||||
|
ulv.viewed_lessons_count,
|
||||||
|
ulv.last_view,
|
||||||
|
ulv.first_view,
|
||||||
|
ROUND( (nullif(ulv.viewed_lessons_count, 0) * 100.0) / ue.lesson_sum,
|
||||||
|
2) as engagement_rate
|
||||||
|
from users_enroll ue
|
||||||
|
left join users_lesson_views ulv on ue.user_id = ulv.user_id;
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
@@ -0,0 +1,51 @@
|
|||||||
|
-- Эффективность курсов
|
||||||
|
|
||||||
|
-- После анализа EXPLAIN
|
||||||
|
-- Убрал диблирование строк, возникающее при Join-ах.
|
||||||
|
-- Минимизировал DISTINCT.
|
||||||
|
drop Materialized VIEW if exists edu_platform.course_summary;
|
||||||
|
create Materialized VIEW edu_platform.course_summary as
|
||||||
|
with
|
||||||
|
enroll as (
|
||||||
|
select
|
||||||
|
course_id,
|
||||||
|
COUNT(DISTINCT user_id) as unique_users -- Может ли один и тот же юзер записаться на курс больше одного раза? Если да, то нужен DISTINCT
|
||||||
|
from
|
||||||
|
edu_platform.enrollments
|
||||||
|
group by
|
||||||
|
course_id),
|
||||||
|
lessons as (
|
||||||
|
select
|
||||||
|
course_id,
|
||||||
|
COUNT(*) as lessons_count
|
||||||
|
from
|
||||||
|
edu_platform.lessons
|
||||||
|
group by
|
||||||
|
course_id),
|
||||||
|
views as (
|
||||||
|
select
|
||||||
|
l.course_id as course_id,
|
||||||
|
count(*) as total_views,
|
||||||
|
MIN(lv.viewed_at) as first_view,
|
||||||
|
MAX(lv.viewed_at) as last_view
|
||||||
|
from
|
||||||
|
edu_platform.lesson_views lv
|
||||||
|
left join edu_platform.lessons l on lv.lesson_id = l.id
|
||||||
|
group by
|
||||||
|
l.course_id)
|
||||||
|
select
|
||||||
|
c.id AS course_id,
|
||||||
|
c.title AS course_title,
|
||||||
|
e.unique_users,
|
||||||
|
l.lessons_count,
|
||||||
|
v.total_views,
|
||||||
|
ROUND( v.total_views * 1.0 / nullif(e.unique_users, 0)
|
||||||
|
,2
|
||||||
|
) as avg_views_per_user,
|
||||||
|
v.first_view,
|
||||||
|
v.last_view
|
||||||
|
from
|
||||||
|
edu_platform.courses c
|
||||||
|
LEFT JOIN enroll e ON c.id = e.course_id
|
||||||
|
LEFT JOIN lessons l ON c.id = l.course_id
|
||||||
|
LEFT JOIN views v ON c.id = v.course_id;
|
||||||
@@ -0,0 +1,21 @@
|
|||||||
|
-- Общая отчетность по взаимодействию
|
||||||
|
|
||||||
|
drop Materialized VIEW if exists edu_platform.platform_summary;
|
||||||
|
create Materialized VIEW edu_platform.platform_summary as
|
||||||
|
select
|
||||||
|
(select COUNT(*) from edu_platform.users) as total_users,
|
||||||
|
(select COUNT(*) from edu_platform.courses) as total_courses,
|
||||||
|
(select COUNT(*) from edu_platform.lessons) as total_lessons,
|
||||||
|
(select COUNT(distinct user_id) from edu_platform.lesson_views) as users_with_views,
|
||||||
|
ROUND(
|
||||||
|
(select AVG(t1.cnt) from
|
||||||
|
(select course_id, COUNT(id) cnt
|
||||||
|
from edu_platform.lessons
|
||||||
|
group by course_id) t1)
|
||||||
|
,2) as avg_lessons_per_course,
|
||||||
|
ROUND(
|
||||||
|
(select AVG(t2.cnt) from
|
||||||
|
(select lesson_id, COUNT(id) cnt
|
||||||
|
from edu_platform.lesson_views
|
||||||
|
group by lesson_id) t2)
|
||||||
|
,2) as avg_views_per_lesson;
|
||||||
@@ -0,0 +1,64 @@
|
|||||||
|
-- 0. Создаём БД
|
||||||
|
CREATE DATABASE dwh_db
|
||||||
|
WITH OWNER = tester
|
||||||
|
TEMPLATE = template0
|
||||||
|
ENCODING = 'UTF8';
|
||||||
|
|
||||||
|
-- 1. Создём схемы
|
||||||
|
CREATE SCHEMA raw; -- сырые данные
|
||||||
|
CREATE SCHEMA core; -- или core. Тут хранилище: факты и измерения
|
||||||
|
CREATE SCHEMA dm; -- BI-витрины
|
||||||
|
|
||||||
|
|
||||||
|
-- 2. Создаём raw-таблицы:
|
||||||
|
|
||||||
|
CREATE TABLE raw.users (
|
||||||
|
id serial4 NOT NULL,
|
||||||
|
"name" varchar(100) NOT NULL,
|
||||||
|
age int4 NOT NULL,
|
||||||
|
email varchar(100) NOT NULL,
|
||||||
|
registration_date date NOT NULL,
|
||||||
|
CONSTRAINT users_pkey PRIMARY KEY (id)
|
||||||
|
)
|
||||||
|
DISTRIBUTED BY (id);
|
||||||
|
|
||||||
|
CREATE TABLE raw.courses (
|
||||||
|
id serial4 NOT NULL,
|
||||||
|
title varchar(100) NOT NULL,
|
||||||
|
category varchar(100) NOT NULL,
|
||||||
|
created_at date NOT NULL,
|
||||||
|
CONSTRAINT courses_pkey PRIMARY KEY (id)
|
||||||
|
)
|
||||||
|
DISTRIBUTED BY (id);
|
||||||
|
|
||||||
|
CREATE TABLE raw.lessons (
|
||||||
|
id serial4 NOT NULL,
|
||||||
|
course_id int4 NOT NULL,
|
||||||
|
title varchar(100) NOT NULL,
|
||||||
|
duration_min int4 NOT NULL,
|
||||||
|
CONSTRAINT lessons_pkey PRIMARY KEY (id)
|
||||||
|
)
|
||||||
|
DISTRIBUTED BY (id);
|
||||||
|
ALTER TABLE raw.lessons ADD CONSTRAINT lessons_course_id_fkey FOREIGN KEY (course_id) REFERENCES raw.courses(id);
|
||||||
|
|
||||||
|
CREATE TABLE raw.enrollments (
|
||||||
|
id serial4 NOT NULL,
|
||||||
|
user_id int4 NOT NULL,
|
||||||
|
course_id int4 NOT NULL,
|
||||||
|
enrolled_at date NOT NULL,
|
||||||
|
CONSTRAINT enrollments_pkey PRIMARY KEY (id)
|
||||||
|
)
|
||||||
|
DISTRIBUTED BY (id);
|
||||||
|
ALTER TABLE raw.enrollments ADD CONSTRAINT enrollments_course_id_fkey FOREIGN KEY (course_id) REFERENCES raw.courses(id);
|
||||||
|
ALTER TABLE raw.enrollments ADD CONSTRAINT enrollments_user_id_fkey FOREIGN KEY (user_id) REFERENCES raw.users(id);
|
||||||
|
|
||||||
|
CREATE TABLE raw.lesson_views (
|
||||||
|
id serial4 NOT NULL,
|
||||||
|
user_id int4 NOT NULL,
|
||||||
|
lesson_id int4 NOT NULL,
|
||||||
|
viewed_at timestamp NOT NULL,
|
||||||
|
CONSTRAINT lesson_views_pkey PRIMARY KEY (id)
|
||||||
|
)
|
||||||
|
DISTRIBUTED BY (id);
|
||||||
|
ALTER TABLE raw.lesson_views ADD CONSTRAINT lesson_views_lesson_id_fkey FOREIGN KEY (lesson_id) REFERENCES raw.lessons(id);
|
||||||
|
ALTER TABLE raw.lesson_views ADD CONSTRAINT lesson_views_user_id_fkey FOREIGN KEY (user_id) REFERENCES raw.users(id);
|
||||||
@@ -0,0 +1,48 @@
|
|||||||
|
/* Копирую в файлы в контейнер:
|
||||||
|
docker cp D:/work/code/DE/GP/Pet-project/IMPORT_DATA/users_202512050730.csv gpdb1:/tmp/users_202512050730.csv
|
||||||
|
docker cp D:/work/code/DE/GP/Pet-project/IMPORT_DATA/courses_202512050730.csv gpdb1:/tmp/courses_202512050730.csv
|
||||||
|
docker cp D:/work/code/DE/GP/Pet-project/IMPORT_DATA/enrollments_202512050730.csv gpdb1:/tmp/enrollments_202512050730.csv
|
||||||
|
docker cp D:/work/code/DE/GP/Pet-project/IMPORT_DATA/lesson_views_202512050730.csv gpdb1:/tmp/lesson_views_202512050730.csv
|
||||||
|
docker cp D:/work/code/DE/GP/Pet-project/IMPORT_DATA/lessons_202512050730.csv gpdb1:/tmp/lessons_202512050730.csv
|
||||||
|
*/
|
||||||
|
|
||||||
|
|
||||||
|
COPY raw.users
|
||||||
|
FROM '/tmp/users_202512050730.csv'
|
||||||
|
WITH (
|
||||||
|
FORMAT csv,
|
||||||
|
HEADER true,
|
||||||
|
DELIMITER ','
|
||||||
|
);
|
||||||
|
|
||||||
|
COPY raw.courses
|
||||||
|
FROM '/tmp/courses_202512050730.csv'
|
||||||
|
WITH (
|
||||||
|
FORMAT csv,
|
||||||
|
HEADER true,
|
||||||
|
DELIMITER ','
|
||||||
|
);
|
||||||
|
|
||||||
|
COPY raw.lessons
|
||||||
|
FROM '/tmp/lessons_202512050730.csv'
|
||||||
|
WITH (
|
||||||
|
FORMAT csv,
|
||||||
|
HEADER true,
|
||||||
|
DELIMITER ','
|
||||||
|
);
|
||||||
|
|
||||||
|
COPY raw.enrollments
|
||||||
|
FROM '/tmp/enrollments_202512050730.csv'
|
||||||
|
WITH (
|
||||||
|
FORMAT csv,
|
||||||
|
HEADER true,
|
||||||
|
DELIMITER ','
|
||||||
|
);
|
||||||
|
|
||||||
|
COPY raw.lesson_views
|
||||||
|
FROM '/tmp/lesson_views_202512050730.csv'
|
||||||
|
WITH (
|
||||||
|
FORMAT csv,
|
||||||
|
HEADER true,
|
||||||
|
DELIMITER ','
|
||||||
|
);
|
||||||
@@ -0,0 +1,51 @@
|
|||||||
|
-- Измерение пользователей
|
||||||
|
CREATE TABLE core.dim_user (
|
||||||
|
user_id INT PRIMARY KEY,
|
||||||
|
name TEXT,
|
||||||
|
age INT,
|
||||||
|
email TEXT,
|
||||||
|
registration_date DATE
|
||||||
|
)
|
||||||
|
DISTRIBUTED BY (user_id);
|
||||||
|
|
||||||
|
-- Измерение курсов
|
||||||
|
CREATE TABLE core.dim_course (
|
||||||
|
course_id INT PRIMARY KEY,
|
||||||
|
title TEXT,
|
||||||
|
category TEXT
|
||||||
|
)
|
||||||
|
DISTRIBUTED BY (course_id);
|
||||||
|
|
||||||
|
-- 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)
|
||||||
|
)
|
||||||
|
DISTRIBUTED BY (lesson_id);
|
||||||
|
|
||||||
|
-- Опционально: измерение времени
|
||||||
|
CREATE TABLE core.dim_date (
|
||||||
|
date DATE PRIMARY KEY,
|
||||||
|
year INT,
|
||||||
|
month INT,
|
||||||
|
day INT,
|
||||||
|
weekday TEXT
|
||||||
|
)
|
||||||
|
DISTRIBUTED BY (date);
|
||||||
|
|
||||||
|
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
|
||||||
|
)
|
||||||
|
DISTRIBUTED BY (user_id);
|
||||||
|
|
||||||
|
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
|
||||||
|
)
|
||||||
|
DISTRIBUTED BY (user_id);
|
||||||
@@ -0,0 +1,49 @@
|
|||||||
|
BEGIN;
|
||||||
|
-- UPDATE с проверкой изменений
|
||||||
|
update
|
||||||
|
core.dim_user d
|
||||||
|
set
|
||||||
|
name = s.name,
|
||||||
|
age = s.age,
|
||||||
|
email = s.email,
|
||||||
|
registration_date = s.registration_date
|
||||||
|
from
|
||||||
|
(
|
||||||
|
select
|
||||||
|
distinct
|
||||||
|
id::INT,
|
||||||
|
name::TEXT,
|
||||||
|
age::INT,
|
||||||
|
email::TEXT,
|
||||||
|
registration_date::DATE
|
||||||
|
from
|
||||||
|
raw.users
|
||||||
|
) s
|
||||||
|
where
|
||||||
|
d.user_id = s.id
|
||||||
|
and (
|
||||||
|
d.name is distinct from s.name
|
||||||
|
or d.age is distinct from s.age
|
||||||
|
or d.email is distinct from s.email
|
||||||
|
or d.registration_date is distinct from s.registration_date
|
||||||
|
);
|
||||||
|
-- INSERT новых строк
|
||||||
|
INSERT INTO core.dim_user (user_id, name, age, email, registration_date)
|
||||||
|
SELECT
|
||||||
|
s.id,
|
||||||
|
s.name,
|
||||||
|
s.age,
|
||||||
|
s.email,
|
||||||
|
s.registration_date
|
||||||
|
FROM (
|
||||||
|
SELECT DISTINCT
|
||||||
|
id::INT,
|
||||||
|
name::TEXT,
|
||||||
|
age::INT,
|
||||||
|
email::TEXT,
|
||||||
|
registration_date::DATE
|
||||||
|
FROM raw.users
|
||||||
|
) s
|
||||||
|
LEFT JOIN core.dim_user d on d.user_id = s.id
|
||||||
|
WHERE d.user_id IS NULL;
|
||||||
|
COMMIT;
|
||||||
@@ -0,0 +1,39 @@
|
|||||||
|
BEGIN;
|
||||||
|
-- UPDATE с проверкой изменений
|
||||||
|
update
|
||||||
|
core.dim_course dc
|
||||||
|
set
|
||||||
|
title = s.title,
|
||||||
|
category = s.category
|
||||||
|
from
|
||||||
|
(
|
||||||
|
select
|
||||||
|
distinct
|
||||||
|
id::INT,
|
||||||
|
title::TEXT,
|
||||||
|
category::TEXT
|
||||||
|
from
|
||||||
|
raw.courses
|
||||||
|
) s
|
||||||
|
where
|
||||||
|
dc.course_id = s.id
|
||||||
|
and (
|
||||||
|
dc.title is distinct from s.title
|
||||||
|
or dc.category is distinct from s.category
|
||||||
|
);
|
||||||
|
-- INSERT новых строк
|
||||||
|
INSERT INTO core.dim_course (course_id, title, category)
|
||||||
|
SELECT
|
||||||
|
s.id,
|
||||||
|
s.title,
|
||||||
|
s.category
|
||||||
|
FROM (
|
||||||
|
SELECT DISTINCT
|
||||||
|
id::INT,
|
||||||
|
title::TEXT,
|
||||||
|
category::TEXT
|
||||||
|
FROM raw.courses
|
||||||
|
) s
|
||||||
|
LEFT JOIN core.dim_course cd on cd.course_id = s.id
|
||||||
|
WHERE cd.course_id IS NULL;
|
||||||
|
COMMIT;
|
||||||
@@ -0,0 +1,44 @@
|
|||||||
|
BEGIN;
|
||||||
|
-- UPDATE с проверкой изменений
|
||||||
|
update
|
||||||
|
core.dim_lesson dl
|
||||||
|
set
|
||||||
|
title = l.title,
|
||||||
|
duration_min = l.duration_min,
|
||||||
|
course_id = l.course_id
|
||||||
|
from
|
||||||
|
(
|
||||||
|
select distinct
|
||||||
|
id::INT,
|
||||||
|
course_id::INT,
|
||||||
|
title::TEXT,
|
||||||
|
duration_min::INT
|
||||||
|
from
|
||||||
|
raw.lessons
|
||||||
|
) l
|
||||||
|
where
|
||||||
|
dl.lesson_id = l.id
|
||||||
|
and (
|
||||||
|
dl.title is distinct from l.title
|
||||||
|
or dl.duration_min is distinct from l.duration_min
|
||||||
|
or dl.course_id is distinct from l.course_id
|
||||||
|
);
|
||||||
|
-- INSERT новых строк
|
||||||
|
INSERT INTO core.dim_lesson (lesson_id, title, duration_min, course_id)
|
||||||
|
SELECT
|
||||||
|
l.id,
|
||||||
|
l.title,
|
||||||
|
l.duration_min,
|
||||||
|
l.course_id
|
||||||
|
FROM (
|
||||||
|
select distinct
|
||||||
|
id::INT,
|
||||||
|
course_id::INT,
|
||||||
|
title::TEXT,
|
||||||
|
duration_min::INT
|
||||||
|
from
|
||||||
|
raw.lessons
|
||||||
|
) l
|
||||||
|
LEFT JOIN core.dim_lesson dl on dl.lesson_id = l.id
|
||||||
|
WHERE dl.lesson_id IS NULL;
|
||||||
|
COMMIT;
|
||||||
@@ -0,0 +1,17 @@
|
|||||||
|
insert into core.enrollments (user_id, course_id, enrolled_at)
|
||||||
|
select
|
||||||
|
re.user_id,
|
||||||
|
re.course_id,
|
||||||
|
re.enrolled_at
|
||||||
|
from
|
||||||
|
(select DISTINCT
|
||||||
|
user_id::INT,
|
||||||
|
course_id::INT,
|
||||||
|
enrolled_at
|
||||||
|
from
|
||||||
|
raw.enrollments) re
|
||||||
|
LEFT join core.enrollments ce on
|
||||||
|
ce.user_id = re.user_id
|
||||||
|
and ce.course_id = re.course_id
|
||||||
|
and ce.enrolled_at = re.enrolled_at
|
||||||
|
WHERE ce.user_id IS NULL;
|
||||||
@@ -0,0 +1,22 @@
|
|||||||
|
-- INSERT новых строк
|
||||||
|
INSERT INTO core.fact_lesson_views (user_id, lesson_id, course_id, viewed_at)
|
||||||
|
select
|
||||||
|
l.user_id,
|
||||||
|
l.lesson_id,
|
||||||
|
l.course_id,
|
||||||
|
l.viewed_at
|
||||||
|
from (select
|
||||||
|
lv.user_id,
|
||||||
|
lv.lesson_id,
|
||||||
|
l.course_id,
|
||||||
|
lv.viewed_at
|
||||||
|
from
|
||||||
|
raw.lesson_views lv
|
||||||
|
left join raw.lessons l on lv.lesson_id = l.id
|
||||||
|
) l
|
||||||
|
LEFT join core.fact_lesson_views flv on
|
||||||
|
flv.user_id = l.user_id
|
||||||
|
and flv.lesson_id = l.lesson_id
|
||||||
|
and flv.course_id = l.course_id
|
||||||
|
and flv.viewed_at = l.viewed_at
|
||||||
|
WHERE flv.user_id IS NULL;
|
||||||
@@ -0,0 +1,29 @@
|
|||||||
|
DROP MATERIALIZED VIEW IF EXISTS dm.lesson_popularity_summary;
|
||||||
|
CREATE MATERIALIZED VIEW dm.lesson_popularity_summary
|
||||||
|
DISTRIBUTED BY (lesson_id)
|
||||||
|
AS
|
||||||
|
select
|
||||||
|
t1.lesson_id as lesson_id,
|
||||||
|
dl.title as lesson_title,
|
||||||
|
t1.course_id as course_id,
|
||||||
|
dc.title as course_title,
|
||||||
|
t1.total_views as total_views,
|
||||||
|
t1.unique_users as unique_users,
|
||||||
|
t1.first_view as first_view,
|
||||||
|
t1.last_view as last_view
|
||||||
|
from
|
||||||
|
(select
|
||||||
|
flv.lesson_id,
|
||||||
|
flv.course_id,
|
||||||
|
COUNT(*) as total_views,
|
||||||
|
COUNT(distinct flv.user_id) as unique_users,
|
||||||
|
MIN(flv.viewed_at) as first_view,
|
||||||
|
MAX(flv.viewed_at) as last_view
|
||||||
|
from
|
||||||
|
core.fact_lesson_views flv
|
||||||
|
group by
|
||||||
|
flv.lesson_id,
|
||||||
|
flv.course_id
|
||||||
|
) as t1
|
||||||
|
right join core.dim_lesson dl on dl.lesson_id = t1.lesson_id
|
||||||
|
left join core.dim_course dc on t1.course_id = dc.course_id;
|
||||||
@@ -0,0 +1,25 @@
|
|||||||
|
DROP MATERIALIZED VIEW IF EXISTS dm.inactive_users_summary;
|
||||||
|
CREATE MATERIALIZED VIEW dm.inactive_users_summary
|
||||||
|
DISTRIBUTED BY (user_id)
|
||||||
|
AS
|
||||||
|
select
|
||||||
|
du.user_id as user_id,
|
||||||
|
du.name as user_name,
|
||||||
|
du.email as email,
|
||||||
|
du.age as age,
|
||||||
|
COALESCE(e.enrollments_count, 0) as enrollments_count,
|
||||||
|
du.registration_date as registration_date
|
||||||
|
from
|
||||||
|
core.dim_user du
|
||||||
|
left join
|
||||||
|
(select
|
||||||
|
user_id,
|
||||||
|
COUNT(*) as enrollments_count
|
||||||
|
from
|
||||||
|
core.enrollments
|
||||||
|
group by
|
||||||
|
user_id
|
||||||
|
) 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;
|
||||||
@@ -0,0 +1,39 @@
|
|||||||
|
DROP MATERIALIZED VIEW IF EXISTS dm.course_completion_rate;
|
||||||
|
CREATE MATERIALIZED VIEW dm.course_completion_rate
|
||||||
|
DISTRIBUTED BY (user_id)
|
||||||
|
AS
|
||||||
|
with activity_users as (
|
||||||
|
select
|
||||||
|
e.user_id as user_id,
|
||||||
|
e.course_id as course_id,
|
||||||
|
COUNT(DISTINCT flv.viewed_at) as lessons_viewed
|
||||||
|
from
|
||||||
|
core.enrollments e
|
||||||
|
left join core.dim_lesson dl on e.course_id = dl.course_id
|
||||||
|
left join core.fact_lesson_views flv on e.user_id = flv.user_id and e.course_id = flv.course_id and dl.lesson_id = flv.lesson_id
|
||||||
|
group by
|
||||||
|
e.user_id,
|
||||||
|
e.course_id),
|
||||||
|
course2 as (
|
||||||
|
select
|
||||||
|
course_id,
|
||||||
|
COUNT(*) as lessons_in_course
|
||||||
|
from core.dim_lesson
|
||||||
|
group by
|
||||||
|
course_id)
|
||||||
|
select
|
||||||
|
au.user_id as user_id,
|
||||||
|
du.name as user_name,
|
||||||
|
au.course_id as course_id,
|
||||||
|
dc.title as course_title,
|
||||||
|
c2.lessons_in_course as lessons_in_course,
|
||||||
|
au.lessons_viewed as lessons_viewed,
|
||||||
|
ROUND(au.lessons_viewed * 1.0 / c2.lessons_in_course , 2) as completion_rate
|
||||||
|
from
|
||||||
|
activity_users au
|
||||||
|
left join core.dim_user du on au.user_id = du.user_id
|
||||||
|
left join core.dim_course dc on au.course_id = dc.course_id
|
||||||
|
left join course2 c2 on au.course_id = c2.course_id
|
||||||
|
order by
|
||||||
|
au.user_id,
|
||||||
|
au.course_id;
|
||||||
@@ -0,0 +1,4 @@
|
|||||||
|
"id","title","category","created_at"
|
||||||
|
1,SQL для начинающих,data,2023-01-01
|
||||||
|
2,Python для анализа данных,programming,2023-01-15
|
||||||
|
3,BI с нуля,business,2023-03-01
|
||||||
|
@@ -0,0 +1,6 @@
|
|||||||
|
"id","user_id","course_id","enrolled_at"
|
||||||
|
1,1,1,2023-01-15
|
||||||
|
2,1,2,2023-02-01
|
||||||
|
3,2,1,2023-01-20
|
||||||
|
4,3,2,2023-03-05
|
||||||
|
5,4,3,2023-04-01
|
||||||
|
@@ -0,0 +1,7 @@
|
|||||||
|
"id","user_id","lesson_id","viewed_at"
|
||||||
|
1,1,1,2023-01-16 10:00:00.000
|
||||||
|
2,1,2,2023-01-16 10:15:00.000
|
||||||
|
3,2,1,2023-01-21 09:00:00.000
|
||||||
|
4,3,3,2023-03-06 12:00:00.000
|
||||||
|
5,4,4,2023-04-02 10:00:00.000
|
||||||
|
6,4,5,2023-04-03 10:00:00.000
|
||||||
|
@@ -0,0 +1,6 @@
|
|||||||
|
"id","course_id","title","duration_min"
|
||||||
|
1,1,SELECT и FROM,10
|
||||||
|
2,1,JOIN,15
|
||||||
|
3,2,Pandas,20
|
||||||
|
4,3,Основы BI,12
|
||||||
|
5,3,Метрики и дашборды,18
|
||||||
|
@@ -0,0 +1,6 @@
|
|||||||
|
"id","name","age","email","registration_date"
|
||||||
|
1,Alice,25,alice@mail.com,2023-01-10
|
||||||
|
2,Bob,30,bob@gmail.com,2023-02-05
|
||||||
|
3,Charlie,22,charlie@mail.com,2023-02-20
|
||||||
|
4,Diana,28,diana@mail.com,2023-03-01
|
||||||
|
5,Ethan,35,ethan@gmail.com,2023-03-10
|
||||||
|
@@ -0,0 +1,86 @@
|
|||||||
|
-- Этап 1: Создание таблиц и связей
|
||||||
|
-- Создай все таблицы в SQL
|
||||||
|
-- Заполни данными (INSERT)
|
||||||
|
-- Настрой внешние ключи
|
||||||
|
--Таблица 1:
|
||||||
|
CREATE TABLE agoncharov.courses (
|
||||||
|
id int4 NOT NULL,
|
||||||
|
title varchar(50) NOT NULL,
|
||||||
|
category varchar(50) NOT NULL,
|
||||||
|
created_at date NOT NULL,
|
||||||
|
CONSTRAINT courses_pkey PRIMARY KEY (id)
|
||||||
|
);
|
||||||
|
insert into (id,title,category,created_at)
|
||||||
|
values (1,'SQL для начинающих','data','2023-01-01'),
|
||||||
|
(2,'Python для анализа данных','programming','2023-01-15'),
|
||||||
|
(3,'BI с нуля','business','2023-03-01');
|
||||||
|
|
||||||
|
-- Таблица 2:
|
||||||
|
CREATE TABLE agoncharov.enrollmets (
|
||||||
|
id int4 NOT NULL,
|
||||||
|
user_id int4 NULL,
|
||||||
|
course_id int4 NULL,
|
||||||
|
enrolled_at date NULL,
|
||||||
|
CONSTRAINT enrollmets_pkey PRIMARY KEY (id)
|
||||||
|
);
|
||||||
|
ALTER TABLE agoncharov.enrollmets ADD CONSTRAINT enrollmets_course_id_fkey FOREIGN KEY (course_id) REFERENCES agoncharov.courses(id);
|
||||||
|
ALTER TABLE agoncharov.enrollmets ADD CONSTRAINT enrollmets_user_id_fkey FOREIGN KEY (user_id) REFERENCES agoncharov.users(id);
|
||||||
|
|
||||||
|
INSERT INTO (id,user_id,course_id,enrolled_at)
|
||||||
|
values (1,1,1,'2023-01-15'),(2,1,2,'2023-02-01'),(3,2,1,'2023-01-20'),
|
||||||
|
(4,3,2,'2023-03-05'),(5,4,3,'2023-04-01');
|
||||||
|
|
||||||
|
-- Таблица 3 :
|
||||||
|
CREATE TABLE agoncharov.lesson_views (
|
||||||
|
id int4 NOT NULL,
|
||||||
|
user_id int4 NULL,
|
||||||
|
lesson_id int4 NULL,
|
||||||
|
viewed_at timestamp NULL,
|
||||||
|
CONSTRAINT lesson_views_pkey PRIMARY KEY (id)
|
||||||
|
);
|
||||||
|
|
||||||
|
|
||||||
|
-- agoncharov.lesson_views внешние включи
|
||||||
|
|
||||||
|
ALTER TABLE agoncharov.lesson_views ADD CONSTRAINT lesson_views_lesson_id_fkey FOREIGN KEY (lesson_id) REFERENCES agoncharov.lessons(id);
|
||||||
|
ALTER TABLE agoncharov.lesson_views ADD CONSTRAINT lesson_views_user_id_fkey FOREIGN KEY (user_id) REFERENCES agoncharov.users(id);
|
||||||
|
|
||||||
|
insert into (id,user_id,lesson_id,viewed_at)
|
||||||
|
values (1,1,1,'2023-01-16 10:00:00'),(2,1,2,'2023-01-16 10:15:00'),
|
||||||
|
(3,2,1,'2023-01-21 09:00:00'), (4,3,3,'2023-04-06 12:00:00'),
|
||||||
|
(5,4,4,'2023-04-02 10:00:00'),(6,4,5,'2023-04-03 10:00:00');
|
||||||
|
|
||||||
|
-- Таблица 4
|
||||||
|
CREATE TABLE agoncharov.lessons (
|
||||||
|
id int4 NOT NULL,
|
||||||
|
course_id int4 NULL,
|
||||||
|
title varchar(50) NOT NULL,
|
||||||
|
duration_min int4 NOT NULL,
|
||||||
|
CONSTRAINT lessons_pkey PRIMARY KEY (id)
|
||||||
|
);
|
||||||
|
|
||||||
|
|
||||||
|
-- agoncharov.lessons внешние включи
|
||||||
|
|
||||||
|
ALTER TABLE agoncharov.lessons ADD CONSTRAINT lessons_course_id_fkey FOREIGN KEY (course_id) REFERENCES agoncharov.courses(id);
|
||||||
|
|
||||||
|
insert into (id,course_id,title,duration_min)
|
||||||
|
values (1,1,1,'SELECT и FROM',10),(2,2,1,'JOIN',15),
|
||||||
|
(3,3,2,'PANDAS',20),(4,4,3,'Основы BI',12),(5,5,3,'Метрики и дашборды',18);
|
||||||
|
|
||||||
|
-- Таблица 5:
|
||||||
|
CREATE TABLE agoncharov.users (
|
||||||
|
id int4 NOT NULL,
|
||||||
|
"name" varchar(50) NOT NULL,
|
||||||
|
age int4 NULL,
|
||||||
|
email varchar(100) NULL,
|
||||||
|
registration_date date NULL,
|
||||||
|
CONSTRAINT users_pkey PRIMARY KEY (id)
|
||||||
|
);
|
||||||
|
|
||||||
|
INSERT INTO (id,name,age,email,registration_date)
|
||||||
|
values (1,'Alice',25,'alice@mail.com','2023-01-10'),
|
||||||
|
(2,'Bob',30,'bob@gmail.com','2023-02-05'),
|
||||||
|
(3,'Charlie',22,'charlie@mail.com','2023-02-20'),
|
||||||
|
(4,'Diana',28,'diana@mail.com','2023-03-01'),
|
||||||
|
(5,'Ethan',35,'ethan@gmail.com','2023-03-10');
|
||||||
@@ -0,0 +1,24 @@
|
|||||||
|
CREATE MATERIALIZED VIEW course_summary as
|
||||||
|
with
|
||||||
|
lw as (
|
||||||
|
select
|
||||||
|
user_id,
|
||||||
|
lesson_id,
|
||||||
|
viewed_at
|
||||||
|
from lesson_views
|
||||||
|
)
|
||||||
|
select
|
||||||
|
id as course_id,
|
||||||
|
title as course_title,
|
||||||
|
count(distinct e.user_id) as unique_users,
|
||||||
|
count(distinct l.lesson_id) as lessons_count,
|
||||||
|
count(distinct lw.viewed_at) as total_views,
|
||||||
|
round(count(distinct lw.viewed_at)::decimal / count(distinct e.user_id), 2) as avg_views_per_user,
|
||||||
|
min(lw.viewed_at) as first_viewed_at,
|
||||||
|
max(lw.viewed_at) as last_viewed_at
|
||||||
|
from
|
||||||
|
courses as c
|
||||||
|
join (select course_id, user_id from enrollments) as e on c.id = e.course_id
|
||||||
|
join (select id as lesson_id, course_id from lessons) as l on c.id = l.course_id
|
||||||
|
join lw on l.lesson_id = lw.lesson_id
|
||||||
|
group by id;
|
||||||
@@ -0,0 +1,24 @@
|
|||||||
|
create materialized view platform_summary as
|
||||||
|
with
|
||||||
|
lv_stats as (
|
||||||
|
select
|
||||||
|
count(distinct user_id) as users_with_views,
|
||||||
|
count(*)::decimal / nullif(count(distinct lesson_id), as avg_views_per_lesson
|
||||||
|
from lesson_views
|
||||||
|
)
|
||||||
|
select
|
||||||
|
total_users,
|
||||||
|
total_courses,
|
||||||
|
total_lessons,
|
||||||
|
users_with_views,
|
||||||
|
round((total_lessons::decimal / nullif(total_courses, 2)), 2) AS avg_lessons_per_course,
|
||||||
|
round(avg_views_per_lesson, 2) as avg_views_per_lesson
|
||||||
|
from
|
||||||
|
(
|
||||||
|
select
|
||||||
|
(select count(*) from users) as total_users,
|
||||||
|
(select count(*) from courses) as total_courses,
|
||||||
|
count(*) as total_lessons
|
||||||
|
from lessons
|
||||||
|
) as s
|
||||||
|
cross join lv_stats as lv;
|
||||||
@@ -0,0 +1,35 @@
|
|||||||
|
-- Витрина активности пользователей
|
||||||
|
CREATE MATERIALIZED VIEW user_activity_summary as
|
||||||
|
with
|
||||||
|
lessons_cte as (
|
||||||
|
select
|
||||||
|
user_id,
|
||||||
|
count (*) as viewed_lessons_count,
|
||||||
|
max(viewed_at) as last_view,
|
||||||
|
min(viewed_at) as first_view
|
||||||
|
from lesson_views
|
||||||
|
group by user_id
|
||||||
|
),
|
||||||
|
enrollments_cte as (
|
||||||
|
select user_id, count(*) as enrollment_count
|
||||||
|
from enrollments as e
|
||||||
|
group by user_id
|
||||||
|
),
|
||||||
|
engagement_cte as (
|
||||||
|
select user_id, count(*) as possible_lessons
|
||||||
|
from lessons as l
|
||||||
|
join enrollments as en on l.course_id = en.course_id
|
||||||
|
group by user_id
|
||||||
|
)
|
||||||
|
select
|
||||||
|
id as user_id,
|
||||||
|
name,
|
||||||
|
e.enrollment_count,
|
||||||
|
l.viewed_lessons_count,
|
||||||
|
l.last_view,
|
||||||
|
l.first_view,
|
||||||
|
round(l.viewed_lessons_count::decimal * 100 / nullif(en.possible_lessons, 0), 2) as engagement_rate
|
||||||
|
from users as u
|
||||||
|
join lessons_cte as l on u.id = l.user_id
|
||||||
|
join engagement_cte as en on u.id = en.user_id
|
||||||
|
join enrollments_cte as e on u.id = e.user_id;
|
||||||
@@ -0,0 +1,69 @@
|
|||||||
|
create table users (
|
||||||
|
id serial primary key,
|
||||||
|
name varchar(50) not null,
|
||||||
|
age int not null,
|
||||||
|
email varchar(100) not null,
|
||||||
|
registration_date date not null
|
||||||
|
);
|
||||||
|
|
||||||
|
insert into users (name, age, email, registration_date) values
|
||||||
|
('Alice', 25, 'alice@mail.com', ; '2023-01-10'),
|
||||||
|
('Bob', 30, 'bob@gmail.com', '2023-02-05'),
|
||||||
|
('Charlie', 22, 'Charlie@mail.com', '2023-02-20'),
|
||||||
|
('Diana', 28, 'diana@mail.com', '2023-03-01'),
|
||||||
|
('Ethan', 35, 'ethan@gmail.com', '2023-03-10');
|
||||||
|
|
||||||
|
create table courses (
|
||||||
|
id serial primary key,
|
||||||
|
title varchar(100) not null,
|
||||||
|
category varchar(30) not null,
|
||||||
|
created_at date not_null
|
||||||
|
);
|
||||||
|
|
||||||
|
insert into courses (title, category, created_at) values
|
||||||
|
('SQL для начинающих', 'data', '2023-01-01'),
|
||||||
|
('Python для анализа данных', 'programming', '2023-01-15'),
|
||||||
|
('BI с нуля', 'business', '2023-03-01');
|
||||||
|
|
||||||
|
create table lessons (
|
||||||
|
id serial primary key,
|
||||||
|
course_id int references courses(id),
|
||||||
|
title varchar(100) not null,
|
||||||
|
duration_min int not null
|
||||||
|
);
|
||||||
|
|
||||||
|
insert into lessons (course_id, title, duration_min) values
|
||||||
|
(1, 'SELECT и FROM', 10),
|
||||||
|
(1, 'JOIN', 15),
|
||||||
|
(2, 'Pandas', 20),
|
||||||
|
(3, 'Основы BI', 12),
|
||||||
|
(3, 'Метрики и дашборды', 18);
|
||||||
|
|
||||||
|
create table enrollments (
|
||||||
|
id serial primary key,
|
||||||
|
user_id int references users(id),
|
||||||
|
course_id int references courses(id),
|
||||||
|
enrolled_at date not null
|
||||||
|
);
|
||||||
|
|
||||||
|
insert into enrollments (user_id, course_id, enrolled_at) values
|
||||||
|
(1, 1, '2023-01-15'),
|
||||||
|
(1, 2, '2023-02-01'),
|
||||||
|
(2, 1, '2023-01-20'),
|
||||||
|
(3, 2, '2023-03-05'),
|
||||||
|
(4, 3, '2023-04-01');
|
||||||
|
|
||||||
|
create table lesson_views (
|
||||||
|
id serial primary key,
|
||||||
|
user_id int references users(id),
|
||||||
|
lesson_id int references lessons(id),
|
||||||
|
viewed_at timestamp(0) not null
|
||||||
|
);
|
||||||
|
|
||||||
|
insert into lesson_views (user_id, lesson_id, viewed_at) values
|
||||||
|
(1, 1, '2023-01-16 10:00:00'),
|
||||||
|
(1, 2, '2023-01-16 10:15:00'),
|
||||||
|
(2, 1, '2023-01-21 09:00:00'),
|
||||||
|
(3, 3, '2023-03-06 12:00:00'),
|
||||||
|
(4, 4, '2023-04-02 10:00:00'),
|
||||||
|
(4, 5, '2023-04-03 10:00:00');
|
||||||
@@ -0,0 +1,418 @@
|
|||||||
|
-- Этап 1
|
||||||
|
-- Задание 1
|
||||||
|
-- Перенести данные как в пет-проекте с Greenplum
|
||||||
|
|
||||||
|
CREATE TABLE itikhomirov.core_user
|
||||||
|
ON CLUSTER clickhouse_cluster
|
||||||
|
(
|
||||||
|
user_id Int32,
|
||||||
|
name String,
|
||||||
|
age Int32,
|
||||||
|
email String,
|
||||||
|
registration_date DateTime64(3, 'UTC')
|
||||||
|
)
|
||||||
|
ENGINE = ReplicatedMergeTree(
|
||||||
|
'/clickhouse/tables/{shard}/itikhomirov/core_user',
|
||||||
|
'{replica}'
|
||||||
|
)
|
||||||
|
ORDER BY user_id;
|
||||||
|
|
||||||
|
ALTER TABLE itikhomirov.core_user
|
||||||
|
MODIFY COLUMN registration_date DateTime64(3);
|
||||||
|
|
||||||
|
CREATE TABLE itikhomirov.core_lesson
|
||||||
|
ON CLUSTER clickhouse_cluster
|
||||||
|
(
|
||||||
|
lesson_id Int32,
|
||||||
|
title String,
|
||||||
|
duration_min Int32,
|
||||||
|
course_id Int32
|
||||||
|
)
|
||||||
|
ENGINE = ReplicatedMergeTree(
|
||||||
|
'/clickhouse/tables/{shard}/itikhomirov/core_lesson',
|
||||||
|
'{replica}'
|
||||||
|
)
|
||||||
|
ORDER BY lesson_id;
|
||||||
|
|
||||||
|
CREATE TABLE itikhomirov.core_fact_lesson_views
|
||||||
|
ON CLUSTER clickhouse_cluster
|
||||||
|
(
|
||||||
|
user_id Int32,
|
||||||
|
lesson_id Int32,
|
||||||
|
course_id Int32,
|
||||||
|
viewed_at DateTime64(3)
|
||||||
|
)
|
||||||
|
ENGINE = ReplicatedMergeTree(
|
||||||
|
'/clickhouse/tables/{shard}/itikhomirov/core_fact_lesson_views',
|
||||||
|
'{replica}'
|
||||||
|
)
|
||||||
|
ORDER BY user_id;
|
||||||
|
|
||||||
|
ALTER TABLE itikhomirov.core_fact_lesson_views
|
||||||
|
MODIFY COLUMN viewed_at DateTime64(3);
|
||||||
|
|
||||||
|
DROP TABLE IF EXISTS itikhomirov.core_fact_lesson_views
|
||||||
|
ON CLUSTER clickhouse_cluster;
|
||||||
|
|
||||||
|
CREATE TABLE itikhomirov.core_enrollments
|
||||||
|
ON CLUSTER clickhouse_cluster
|
||||||
|
(
|
||||||
|
user_id Int32,
|
||||||
|
course_id Int32,
|
||||||
|
enrolled_at DateTime64(3)
|
||||||
|
)
|
||||||
|
ENGINE = ReplicatedMergeTree(
|
||||||
|
'/clickhouse/tables/{shard}/itikhomirov/core_enrollmentsы',
|
||||||
|
'{replica}'
|
||||||
|
)
|
||||||
|
ORDER BY user_id;
|
||||||
|
|
||||||
|
CREATE TABLE itikhomirov.core_course
|
||||||
|
ON CLUSTER clickhouse_cluster
|
||||||
|
(
|
||||||
|
course_id Int32,
|
||||||
|
title String,
|
||||||
|
category String
|
||||||
|
)
|
||||||
|
ENGINE = ReplicatedMergeTree(
|
||||||
|
'/clickhouse/tables/{shard}/itikhomirov/core_course',
|
||||||
|
'{replica}'
|
||||||
|
)
|
||||||
|
ORDER BY course_id;
|
||||||
|
|
||||||
|
-- Этап 2. Создание витринных таблиц
|
||||||
|
-- Задание 2: Подготовь структуру витрины lesson_popularity_summary
|
||||||
|
|
||||||
|
-- Создаем пустую таблицу lesson_popularity_summary(структуру витрины)
|
||||||
|
|
||||||
|
CREATE TABLE itikhomirov.lesson_popularity_summary
|
||||||
|
ON CLUSTER clickhouse_cluster
|
||||||
|
(
|
||||||
|
lesson_id UInt32,
|
||||||
|
lesson_title String,
|
||||||
|
course_id UInt32,
|
||||||
|
course_title String,
|
||||||
|
total_views UInt32,
|
||||||
|
unique_users UInt32,
|
||||||
|
first_view DateTime64(3),
|
||||||
|
last_view DateTime64(3)
|
||||||
|
)
|
||||||
|
ENGINE = ReplicatedMergeTree(
|
||||||
|
'/clickhouse/tables/{shard}/itikhomirov/lesson_popularity_summary',
|
||||||
|
'{replica}'
|
||||||
|
)
|
||||||
|
ORDER BY (lesson_id, course_id);
|
||||||
|
|
||||||
|
-- Создаем Materialized View
|
||||||
|
|
||||||
|
DROP TABLE IF EXISTS itikhomirov.mv_lesson_popularity_summary
|
||||||
|
ON CLUSTER clickhouse_cluster;
|
||||||
|
|
||||||
|
CREATE MATERIALIZED VIEW itikhomirov.mv_lesson_popularity_summary
|
||||||
|
ON CLUSTER clickhouse_cluster
|
||||||
|
TO itikhomirov.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 itikhomirov.core_fact_lesson_views AS flv
|
||||||
|
JOIN itikhomirov.core_lesson AS l ON flv.lesson_id = l.lesson_id
|
||||||
|
JOIN itikhomirov.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 lesson_popularity_summary
|
||||||
|
|
||||||
|
INSERT INTO itikhomirov.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 itikhomirov.core_fact_lesson_views AS flv
|
||||||
|
JOIN itikhomirov.core_lesson AS l ON flv.lesson_id = l.lesson_id
|
||||||
|
JOIN itikhomirov.core_course AS c ON flv.course_id = c.course_id
|
||||||
|
GROUP BY
|
||||||
|
flv.lesson_id,
|
||||||
|
l.title,
|
||||||
|
flv.course_id,
|
||||||
|
c.title;
|
||||||
|
|
||||||
|
-- Задание 3: Создай витрину inactive_users_summary
|
||||||
|
|
||||||
|
-- Создаем пустую таблицу inactive_users_summary(структуру витрины)
|
||||||
|
|
||||||
|
DROP TABLE IF EXISTS itikhomirov.inactive_users_summary
|
||||||
|
ON CLUSTER clickhouse_cluster;
|
||||||
|
|
||||||
|
DROP TABLE IF EXISTS itikhomirov.inactive_users_summary
|
||||||
|
ON CLUSTER clickhouse_cluster;
|
||||||
|
|
||||||
|
CREATE TABLE itikhomirov.inactive_users_summary
|
||||||
|
ON CLUSTER clickhouse_cluster
|
||||||
|
(
|
||||||
|
user_id UInt32,
|
||||||
|
name String,
|
||||||
|
email String,
|
||||||
|
age UInt8,
|
||||||
|
registered_courses_count UInt32,
|
||||||
|
registration_date DateTime64(3)
|
||||||
|
)
|
||||||
|
ENGINE = ReplicatedMergeTree(
|
||||||
|
'/clickhouse/tables/{shard}/itikhomirov/inactive_users_summary',
|
||||||
|
'{replica}'
|
||||||
|
)
|
||||||
|
ORDER BY (user_id);
|
||||||
|
|
||||||
|
-- Создаем Materialized View
|
||||||
|
|
||||||
|
DROP TABLE IF EXISTS itikhomirov.mv_inactive_users_summary
|
||||||
|
ON CLUSTER clickhouse_cluster;
|
||||||
|
|
||||||
|
CREATE MATERIALIZED VIEW itikhomirov.mv_inactive_users_summary
|
||||||
|
ON CLUSTER clickhouse_cluster
|
||||||
|
TO itikhomirov.inactive_users_summary
|
||||||
|
AS
|
||||||
|
WITH user_courses AS (
|
||||||
|
SELECT
|
||||||
|
e.user_id,
|
||||||
|
countDistinct(e.course_id) AS registered_courses_count
|
||||||
|
FROM itikhomirov.core_enrollments AS e
|
||||||
|
GROUP BY
|
||||||
|
e.user_id
|
||||||
|
)
|
||||||
|
SELECT
|
||||||
|
u.user_id AS user_id,
|
||||||
|
u.name AS name,
|
||||||
|
u.email AS email,
|
||||||
|
u.age AS age,
|
||||||
|
coalesce(uc.registered_courses_count, 0) AS registered_courses_count,
|
||||||
|
u.registration_date AS registration_date
|
||||||
|
FROM itikhomirov.core_user AS u
|
||||||
|
LEFT JOIN user_courses AS uc
|
||||||
|
ON u.user_id = uc.user_id
|
||||||
|
WHERE u.user_id NOT IN (
|
||||||
|
SELECT DISTINCT user_id
|
||||||
|
FROM itikhomirov.core_fact_lesson_views
|
||||||
|
);
|
||||||
|
|
||||||
|
|
||||||
|
INSERT INTO itikhomirov.inactive_users_summary
|
||||||
|
WITH user_courses AS (
|
||||||
|
SELECT
|
||||||
|
e.user_id,
|
||||||
|
countDistinct(e.course_id) AS registered_courses_count
|
||||||
|
FROM itikhomirov.core_enrollments AS e
|
||||||
|
GROUP BY
|
||||||
|
e.user_id
|
||||||
|
)
|
||||||
|
-- В CTE user_courses считается число курсов по каждому пользователю независимо от просмотров
|
||||||
|
-- Снаружи берутся пользователи без просмотров (NOT IN по core_fact_lesson_views),
|
||||||
|
-- и к ним подмешивается уже готовый registered_courses_count.
|
||||||
|
SELECT
|
||||||
|
u.user_id AS user_id,
|
||||||
|
u.name AS name,
|
||||||
|
u.email AS email,
|
||||||
|
u.age AS age,
|
||||||
|
coalesce(uc.registered_courses_count, 0) AS registered_courses_count,
|
||||||
|
u.registration_date AS registration_date
|
||||||
|
FROM itikhomirov.core_user AS u
|
||||||
|
LEFT JOIN user_courses AS uc
|
||||||
|
ON u.user_id = uc.user_id
|
||||||
|
WHERE u.user_id NOT IN (
|
||||||
|
SELECT DISTINCT user_id
|
||||||
|
FROM itikhomirov.core_fact_lesson_views
|
||||||
|
)
|
||||||
|
ORDER BY u.user_id;
|
||||||
|
|
||||||
|
-- Задание 4: Создай витрину course_completion_rate
|
||||||
|
|
||||||
|
CREATE TABLE itikhomirov.course_completion_rate
|
||||||
|
ON CLUSTER clickhouse_cluster
|
||||||
|
(
|
||||||
|
user_id UInt32,
|
||||||
|
user_name String,
|
||||||
|
course_id UInt32,
|
||||||
|
course_title String,
|
||||||
|
lessons_in_course UInt32,
|
||||||
|
lessons_viewed UInt32,
|
||||||
|
completion_rate Float32
|
||||||
|
)
|
||||||
|
ENGINE = ReplicatedMergeTree(
|
||||||
|
'/clickhouse/tables/{shard}/itikhomirov/course_completion_rate',
|
||||||
|
'{replica}'
|
||||||
|
)
|
||||||
|
ORDER BY (user_id, course_id);
|
||||||
|
|
||||||
|
|
||||||
|
-- Создаем Materialized View
|
||||||
|
|
||||||
|
CREATE MATERIALIZED VIEW itikhomirov.mv_course_completion_rate
|
||||||
|
ON CLUSTER clickhouse_cluster
|
||||||
|
TO itikhomirov.course_completion_rate
|
||||||
|
AS
|
||||||
|
WITH
|
||||||
|
lessons_per_course AS (
|
||||||
|
SELECT
|
||||||
|
course_id,
|
||||||
|
countDistinct(lesson_id) AS lessons_in_course
|
||||||
|
FROM itikhomirov.core_lesson
|
||||||
|
GROUP BY course_id
|
||||||
|
),
|
||||||
|
views_per_user_course AS (
|
||||||
|
SELECT
|
||||||
|
user_id,
|
||||||
|
course_id,
|
||||||
|
countDistinct(lesson_id) AS lessons_viewed
|
||||||
|
FROM itikhomirov.core_fact_lesson_views
|
||||||
|
GROUP BY user_id, course_id
|
||||||
|
)
|
||||||
|
SELECT
|
||||||
|
u.user_id AS user_id,
|
||||||
|
u.name AS user_name,
|
||||||
|
c.course_id AS course_id,
|
||||||
|
c.title AS course_title,
|
||||||
|
lpc.lessons_in_course AS lessons_in_course,
|
||||||
|
coalesce(vpc.lessons_viewed, 0) AS lessons_viewed,
|
||||||
|
ROUND(
|
||||||
|
COALESCE(
|
||||||
|
NULLIF(coalesce(vpc.lessons_viewed, 0) * 1.0, 0)
|
||||||
|
/ NULLIF(lpc.lessons_in_course * 1.0, 0),
|
||||||
|
0
|
||||||
|
),
|
||||||
|
2
|
||||||
|
) AS completion_rate
|
||||||
|
FROM itikhomirov.core_user AS u
|
||||||
|
JOIN itikhomirov.core_enrollments AS e ON u.user_id = e.user_id
|
||||||
|
JOIN itikhomirov.core_course AS c ON e.course_id = c.course_id
|
||||||
|
JOIN lessons_per_course AS lpc ON c.course_id = lpc.course_id
|
||||||
|
LEFT JOIN views_per_user_course AS vpc ON vpc.user_id = u.user_id AND vpc.course_id = c.course_id
|
||||||
|
ORDER BY
|
||||||
|
u.user_id,
|
||||||
|
c.course_id;
|
||||||
|
|
||||||
|
-- Вставляем данные в таблицу pet.course_completion_rate
|
||||||
|
|
||||||
|
INSERT INTO itikhomirov.course_completion_rate
|
||||||
|
-- Создаем две временные таблицы:
|
||||||
|
-- 1) Сколько уроков в каждом курсе
|
||||||
|
WITH
|
||||||
|
lessons_per_course AS (
|
||||||
|
SELECT
|
||||||
|
course_id,
|
||||||
|
countDistinct(lesson_id) AS lessons_in_course
|
||||||
|
FROM itikhomirov.core_lesson
|
||||||
|
GROUP BY course_id
|
||||||
|
),
|
||||||
|
-- 2) Сколько уроков каждого курса посмотрел пользователь
|
||||||
|
views_per_user_course AS (
|
||||||
|
SELECT
|
||||||
|
user_id,
|
||||||
|
course_id,
|
||||||
|
countDistinct(lesson_id) AS lessons_viewed
|
||||||
|
FROM itikhomirov.core_fact_lesson_views
|
||||||
|
GROUP BY user_id, course_id
|
||||||
|
)
|
||||||
|
-- Основной запрос
|
||||||
|
SELECT
|
||||||
|
u.user_id AS user_id,
|
||||||
|
u.name AS user_name,
|
||||||
|
c.course_id AS course_id,
|
||||||
|
c.title AS course_title,
|
||||||
|
lpc.lessons_in_course AS lessons_in_course,
|
||||||
|
coalesce(vpc.lessons_viewed, 0) AS lessons_viewed,
|
||||||
|
ROUND(
|
||||||
|
COALESCE(
|
||||||
|
NULLIF(coalesce(vpc.lessons_viewed, 0) * 1.0, 0)
|
||||||
|
/ NULLIF(lpc.lessons_in_course * 1.0, 0),
|
||||||
|
0
|
||||||
|
),
|
||||||
|
2
|
||||||
|
) AS completion_rate
|
||||||
|
FROM itikhomirov.core_user AS u
|
||||||
|
JOIN itikhomirov.core_enrollments AS e ON u.user_id = e.user_id
|
||||||
|
JOIN itikhomirov.core_course AS c ON e.course_id = c.course_id
|
||||||
|
JOIN lessons_per_course AS lpc ON c.course_id = lpc.course_id
|
||||||
|
LEFT JOIN views_per_user_course AS vpc ON vpc.user_id = u.user_id AND vpc.course_id = c.course_id
|
||||||
|
ORDER BY
|
||||||
|
u.user_id,
|
||||||
|
c.course_id;
|
||||||
|
|
||||||
|
-- Этап 3. Анализ и запросы
|
||||||
|
|
||||||
|
-- Задание 5: Напиши SQL-запросы к витринам
|
||||||
|
|
||||||
|
-- Уроки с хотя бы одним просмотром
|
||||||
|
SELECT
|
||||||
|
lesson_title,
|
||||||
|
course_title,
|
||||||
|
total_views
|
||||||
|
FROM itikhomirov.lesson_popularity_summary
|
||||||
|
WHERE total_views > 0
|
||||||
|
ORDER BY lesson_title ASC;
|
||||||
|
|
||||||
|
-- Топ‑3 урока внутри одного курса
|
||||||
|
SELECT
|
||||||
|
lesson_title,
|
||||||
|
course_title,
|
||||||
|
total_views
|
||||||
|
FROM itikhomirov.lesson_popularity_summary
|
||||||
|
WHERE course_title = 'SQL для начинающих'
|
||||||
|
ORDER BY total_views DESC
|
||||||
|
LIMIT 3;
|
||||||
|
|
||||||
|
-- Пользователи, которые завершили курсы на 100%
|
||||||
|
SELECT
|
||||||
|
user_id,
|
||||||
|
user_name,
|
||||||
|
course_title,
|
||||||
|
completion_rate
|
||||||
|
FROM itikhomirov.course_completion_rate
|
||||||
|
WHERE completion_rate = 1
|
||||||
|
ORDER BY user_id;
|
||||||
|
|
||||||
|
-- Этап 4. Тестирование и валидация
|
||||||
|
|
||||||
|
-- Задание 6: Сравни данные с Greenplum**
|
||||||
|
|
||||||
|
-- 1. Выбери 2–3 значения из таблиц `_summary_source` в Greenplum
|
||||||
|
|
||||||
|
|
||||||
|
-- 2. Сравни с ClickHouse-результатами:
|
||||||
|
-- - Совпадает ли `SUM`, `AVG`, `MIN`, `MAX`
|
||||||
|
-- - Проверяй `NULL`, `0`, округления
|
||||||
|
|
||||||
|
|
||||||
|
-- 3. Напиши мини-отчёт: **"где совпадает, где нет, почему"**
|
||||||
|
|
||||||
|
SELECT
|
||||||
|
sum(total_views) AS sum_total_views,
|
||||||
|
avg(total_views) AS avg_total_views,
|
||||||
|
min(total_views) AS min_total_views,
|
||||||
|
max(total_views) AS max_total_views
|
||||||
|
FROM itikhomirov.lesson_popularity_summary
|
||||||
|
WHERE lesson_id IN (1, 3, 5);
|
||||||
|
|
||||||
|
SELECT
|
||||||
|
sum(total_views) AS sum_total_views,
|
||||||
|
avg(total_views) AS avg_total_views,
|
||||||
|
min(total_views) AS min_total_views,
|
||||||
|
max(total_views) AS max_total_views
|
||||||
|
FROM itikhomirov.lesson_popularity_summary
|
||||||
|
WHERE lesson_id = 999;
|
||||||
|
|
||||||
|
-- Отчёт в .doc файле
|
||||||
|
|
||||||
|
After Width: | Height: | Size: 292 KiB |
|
After Width: | Height: | Size: 359 KiB |
|
After Width: | Height: | Size: 215 KiB |
|
After Width: | Height: | Size: 129 KiB |
|
After Width: | Height: | Size: 234 KiB |
|
After Width: | Height: | Size: 122 KiB |
|
After Width: | Height: | Size: 458 KiB |
|
After Width: | Height: | Size: 136 KiB |
|
After Width: | Height: | Size: 523 KiB |
|
After Width: | Height: | Size: 106 KiB |
|
After Width: | Height: | Size: 194 KiB |
|
After Width: | Height: | Size: 363 KiB |
|
After Width: | Height: | Size: 272 KiB |
|
After Width: | Height: | Size: 333 KiB |
|
After Width: | Height: | Size: 293 KiB |
@@ -0,0 +1,241 @@
|
|||||||
|
-- Этап 1. Инициализация проекта
|
||||||
|
-- Задание 1: Создай схемы
|
||||||
|
|
||||||
|
CREATE SCHEMA raw;
|
||||||
|
CREATE SCHEMA core;
|
||||||
|
CREATE SCHEMA dm;
|
||||||
|
|
||||||
|
-- Задание 2: Создай raw-таблицы
|
||||||
|
|
||||||
|
CREATE TABLE raw.users (
|
||||||
|
id int,
|
||||||
|
name varchar(15),
|
||||||
|
age int,
|
||||||
|
email varchar(30),
|
||||||
|
registration_date DATE
|
||||||
|
)
|
||||||
|
DISTRIBUTED BY (id);
|
||||||
|
|
||||||
|
CREATE TABLE raw.courses (
|
||||||
|
id int,
|
||||||
|
title text,
|
||||||
|
category varchar(30),
|
||||||
|
created_at DATE
|
||||||
|
)
|
||||||
|
DISTRIBUTED BY (id);
|
||||||
|
|
||||||
|
CREATE TABLE raw.lessons (
|
||||||
|
id int,
|
||||||
|
course_id int,
|
||||||
|
title text,
|
||||||
|
duration_min int
|
||||||
|
)
|
||||||
|
DISTRIBUTED BY (id);
|
||||||
|
|
||||||
|
CREATE TABLE raw.enrollments (
|
||||||
|
id int,
|
||||||
|
user_id int,
|
||||||
|
course_id int,
|
||||||
|
enrolled_at DATE
|
||||||
|
)
|
||||||
|
DISTRIBUTED BY (id);
|
||||||
|
|
||||||
|
CREATE TABLE raw.lesson_views (
|
||||||
|
id int,
|
||||||
|
user_id int,
|
||||||
|
lesson_id int,
|
||||||
|
viewed_at timestamp
|
||||||
|
)
|
||||||
|
DISTRIBUTED BY (id);
|
||||||
|
|
||||||
|
-- Этап 2. Загрузка данных (эмуляция Airflow)
|
||||||
|
-- Задание 3: Выгрузить данные из PostgreSQL и Загрузить CSV в Greenplum
|
||||||
|
|
||||||
|
|
||||||
|
-- Экспортировал данные из PostgreSQL в CSV-файлы и загрузил их в Greenplum
|
||||||
|
|
||||||
|
-- Этап 3. Слой Core — обогащённые данные
|
||||||
|
-- Задание 4: Измерения (dimension tables)
|
||||||
|
|
||||||
|
-- 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
|
||||||
|
);
|
||||||
|
|
||||||
|
-- Задание 5: Факт (fact table)
|
||||||
|
|
||||||
|
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
|
||||||
|
);
|
||||||
|
|
||||||
|
-- Задание 6: ETL переброса данных из raw в core
|
||||||
|
|
||||||
|
INSERT INTO core.dim_user (user_id, name, age, email, registration_date) -- ok
|
||||||
|
SELECT DISTINCT
|
||||||
|
id::INT,
|
||||||
|
name::TEXT,
|
||||||
|
age::INT,
|
||||||
|
email::TEXT,
|
||||||
|
registration_date::DATE
|
||||||
|
FROM raw.users;
|
||||||
|
|
||||||
|
INSERT INTO core.dim_lesson (lesson_id, title, duration_min, course_id) -- ok
|
||||||
|
SELECT DISTINCT
|
||||||
|
id::INT,
|
||||||
|
title::TEXT,
|
||||||
|
duration_min::INT,
|
||||||
|
course_id::INT
|
||||||
|
FROM raw.lessons;
|
||||||
|
|
||||||
|
INSERT INTO core.fact_lesson_views (user_id, lesson_id, course_id, viewed_at) --заполнить из нескольких таблиц с помощью join
|
||||||
|
SELECT DISTINCT
|
||||||
|
lv.user_id::INT,
|
||||||
|
lv.lesson_id::INT,
|
||||||
|
le.course_id::INT,
|
||||||
|
lv.viewed_at::TIMESTAMP
|
||||||
|
FROM raw.lesson_views AS lv
|
||||||
|
JOIN raw.lessons AS le ON lv.lesson_id = le.id;
|
||||||
|
|
||||||
|
INSERT INTO core.enrollments (user_id, course_id, enrolled_at) -- ok
|
||||||
|
SELECT DISTINCT
|
||||||
|
user_id::INT,
|
||||||
|
course_id::INT,
|
||||||
|
enrolled_at::DATE
|
||||||
|
FROM raw.enrollments;
|
||||||
|
|
||||||
|
INSERT INTO core.dim_course (course_id, title, category) -- ok
|
||||||
|
SELECT DISTINCT
|
||||||
|
id::INT,
|
||||||
|
title::TEXT,
|
||||||
|
category::VARCHAR
|
||||||
|
FROM raw.courses;
|
||||||
|
|
||||||
|
INSERT INTO core.dim_date (date, year, month, day, weekday)
|
||||||
|
SELECT DISTINCT
|
||||||
|
d::DATE,
|
||||||
|
EXTRACT(YEAR FROM d)::INT,
|
||||||
|
EXTRACT(MONTH FROM d)::INT,
|
||||||
|
EXTRACT(DAY FROM d)::INT,
|
||||||
|
TO_CHAR(d, 'Day')::TEXT
|
||||||
|
FROM (
|
||||||
|
SELECT registration_date AS d FROM raw.users
|
||||||
|
UNION
|
||||||
|
SELECT enrolled_at FROM raw.enrollments
|
||||||
|
UNION
|
||||||
|
SELECT viewed_at FROM raw.lesson_views
|
||||||
|
) all_dates
|
||||||
|
WHERE d IS NOT NULL;
|
||||||
|
|
||||||
|
-- Надо сделать, чтобы вместо номера месяца выводилось название месяца
|
||||||
|
|
||||||
|
ALTER TABLE core.dim_date
|
||||||
|
ALTER COLUMN month TYPE text;
|
||||||
|
|
||||||
|
UPDATE core.dim_date
|
||||||
|
SET month = TO_CHAR(date, 'Month');
|
||||||
|
|
||||||
|
-- Этап 4: Подготовка витрин в Greenplum
|
||||||
|
-- Задание 7: Витрина dm.lesson_popularity_summary
|
||||||
|
|
||||||
|
CREATE MATERIALIZED VIEW dm.lesson_popularity_summary AS
|
||||||
|
SELECT
|
||||||
|
dl.lesson_id,
|
||||||
|
dl.title AS lesson_title,
|
||||||
|
dl.course_id,
|
||||||
|
dc.title AS course_title,
|
||||||
|
COUNT(flv.user_id) AS total_views,
|
||||||
|
COUNT(DISTINCT flv.user_id) AS unique_users,
|
||||||
|
MIN(flv.viewed_at) AS first_view,
|
||||||
|
MAX(flv.viewed_at) AS last_view
|
||||||
|
FROM core.fact_lesson_views AS flv
|
||||||
|
JOIN core.dim_lesson AS dl ON flv.lesson_id = dl.lesson_id
|
||||||
|
JOIN core.dim_course AS dc ON dl.course_id = dc.course_id
|
||||||
|
GROUP BY dl.lesson_id, dl.title, dl.course_id, dc.title
|
||||||
|
ORDER BY dl.lesson_id ASC;
|
||||||
|
|
||||||
|
-- Задание 8: Витрина dm.inactive_users_summary
|
||||||
|
|
||||||
|
CREATE MATERIALIZED VIEW dm.inactive_users_summary AS
|
||||||
|
SELECT
|
||||||
|
du.user_id,
|
||||||
|
du.name AS user_name,
|
||||||
|
du.email,
|
||||||
|
du.age,
|
||||||
|
COUNT(e.course_id) AS enrollments_count,
|
||||||
|
du.registration_date
|
||||||
|
FROM core.dim_user AS du
|
||||||
|
LEFT JOIN core.enrollments AS e ON du.user_id = e.user_id
|
||||||
|
LEFT JOIN core.fact_lesson_views AS 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;
|
||||||
|
|
||||||
|
-- Задание 9: Витрина dm.course_completion_rate
|
||||||
|
|
||||||
|
CREATE MATERIALIZED VIEW dm.course_completion_rate AS
|
||||||
|
SELECT
|
||||||
|
du.user_id,
|
||||||
|
du.name AS user_name,
|
||||||
|
dc.course_id,
|
||||||
|
dc.title AS course_title,
|
||||||
|
COUNT(DISTINCT dl.lesson_id) AS lesson_in_course,
|
||||||
|
COUNT(DISTINCT flv.lesson_id) AS lessons_viewed,
|
||||||
|
ROUND(
|
||||||
|
COALESCE(
|
||||||
|
NULLIF(COUNT(DISTINCT flv.lesson_id)::NUMERIC,0) / NULLIF(COUNT(DISTINCT dl.lesson_id), 0),
|
||||||
|
0
|
||||||
|
) ,2
|
||||||
|
) AS completion_rate
|
||||||
|
FROM core.dim_user AS du
|
||||||
|
JOIN core.enrollments AS e ON du.user_id = e.user_id
|
||||||
|
JOIN core.dim_course AS dc ON e.course_id= dc.course_id
|
||||||
|
JOIN core.dim_lesson AS dl ON dc.course_id = dl.course_id
|
||||||
|
LEFT JOIN core.fact_lesson_views AS flv ON du.user_id = flv.user_id AND dc.course_id = flv.course_id
|
||||||
|
GROUP BY du.user_id, user_name, dc.course_id, course_title
|
||||||
|
ORDER BY du.user_id ASC;
|
||||||
|
|
||||||
|
-- 🧪 Этап 5. Проверка
|
||||||
|
-- Задание 10: Проверь витрины
|
||||||
|
|
||||||
|
SELECT * FROM dm.lesson_popularity_summary LIMIT 10;
|
||||||
|
SELECT * FROM dm.inactive_users_summary LIMIT 10;
|
||||||
|
SELECT * FROM dm.course_completion_rate ORDER BY user_id ASC LIMIT 10;
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
@@ -0,0 +1,45 @@
|
|||||||
|
BEGIN;
|
||||||
|
|
||||||
|
CREATE TABLE SKONEV.users(
|
||||||
|
id INT PRIMARY KEY,
|
||||||
|
name VARCHAR(100),
|
||||||
|
age INT,
|
||||||
|
email VARCHAR(255) UNIQUE,
|
||||||
|
registration_date DATE NOT NULL
|
||||||
|
);
|
||||||
|
|
||||||
|
CREATE TABLE SKONEV.courses(
|
||||||
|
id INT PRIMARY KEY,
|
||||||
|
title TEXT NOT NULL,
|
||||||
|
category VARCHAR(255) CHECK (category IN ('data', 'programming', 'business')),
|
||||||
|
created_at DATE NOT NULL
|
||||||
|
);
|
||||||
|
|
||||||
|
CREATE TABLE SKONEV.lessons(
|
||||||
|
id INT PRIMARY KEY,
|
||||||
|
course_id INT NOT NULL,
|
||||||
|
FOREIGN KEY (course_id) REFERENCES courses(id),
|
||||||
|
title TEXT NOT NULL,
|
||||||
|
duration_min INT NOT NULL
|
||||||
|
);
|
||||||
|
|
||||||
|
|
||||||
|
CREATE TABLE SKONEV.enrollments(
|
||||||
|
id INT PRIMARY KEY,
|
||||||
|
user_id INT NOT NULL,
|
||||||
|
FOREIGN KEY (user_id) REFERENCES users(id),
|
||||||
|
course_id INT NOT NULL,
|
||||||
|
FOREIGN KEY (course_id) REFERENCES courses(id),
|
||||||
|
enrolled_at DATE NOT NULL
|
||||||
|
);
|
||||||
|
|
||||||
|
CREATE TABLE SKONEV.lesson_views(
|
||||||
|
id INT PRIMARY KEY,
|
||||||
|
user_id INT NOT NULL,
|
||||||
|
FOREIGN KEY (user_id) REFERENCES users(id),
|
||||||
|
lesson_id INT NOT NULL,
|
||||||
|
FOREIGN KEY (lesson_id) REFERENCES lessons(id),
|
||||||
|
viewed_at TIMESTAMP NOT NULL
|
||||||
|
);
|
||||||
|
|
||||||
|
COMMIT;
|
||||||
@@ -0,0 +1,47 @@
|
|||||||
|
BEGIN;
|
||||||
|
|
||||||
|
CREATE MATERIALIZED VIEW course_summary AS
|
||||||
|
WITH courses_users AS (
|
||||||
|
SELECT
|
||||||
|
c.id AS course_id,
|
||||||
|
c.title AS course_name,
|
||||||
|
COUNT(DISTINCT user_id) AS unique_users
|
||||||
|
FROM courses c
|
||||||
|
LEFT join enrollments e ON c.id = e.course_id
|
||||||
|
GROUP BY c.id, c.title
|
||||||
|
),
|
||||||
|
courses_views AS (
|
||||||
|
SELECT
|
||||||
|
l.course_id,
|
||||||
|
COUNT(lv.id) AS total_views,
|
||||||
|
MAX(lv.viewed_at) AS last_viewed_at,
|
||||||
|
MIN(lv.viewed_at) AS first_viewed_at
|
||||||
|
FROM lessons l
|
||||||
|
JOIN lesson_views lv ON l.id = lv.lesson_id
|
||||||
|
GROUP BY l.course_id
|
||||||
|
),
|
||||||
|
courses_lessons AS (
|
||||||
|
SELECT
|
||||||
|
course_id,
|
||||||
|
COUNT(DISTINCT id) AS lessons_count
|
||||||
|
FROM lessons
|
||||||
|
GROUP BY course_id
|
||||||
|
)
|
||||||
|
|
||||||
|
SELECT
|
||||||
|
cu.course_id,
|
||||||
|
cu.course_name,
|
||||||
|
cu.unique_users,
|
||||||
|
cl.lessons_count,
|
||||||
|
cv.total_views,
|
||||||
|
ROUND(cv.total_views::DECIMAL / NULLIF(cu.unique_users, 0), 2) AS avg_views_per_user,
|
||||||
|
cv.first_viewed_at,
|
||||||
|
cv.last_viewed_at
|
||||||
|
FROM courses_users cu
|
||||||
|
LEFT JOIN courses_views cv USING(course_id)
|
||||||
|
LEFT JOIN courses_lessons cl USING(course_id)
|
||||||
|
ORDER BY cu.course_id;
|
||||||
|
|
||||||
|
COMMIT;
|
||||||
|
|
||||||
|
|
||||||
@@ -0,0 +1,20 @@
|
|||||||
|
BEGIN;
|
||||||
|
|
||||||
|
CREATE MATERIALIZED VIEW platform_summary AS
|
||||||
|
SELECT
|
||||||
|
(SELECT COUNT(*) FROM users) AS total_users,
|
||||||
|
(SELECT COUNT(*) FROM courses) AS total_courses,
|
||||||
|
(SELECT COUNT(*) FROM lessons) AS total_lessons,
|
||||||
|
(SELECT COUNT(DISTINCT user_id) FROM lesson_views) AS users_with_views,
|
||||||
|
ROUND(
|
||||||
|
(SELECT COUNT(*)::DECIMAL FROM lessons) /
|
||||||
|
NULLIF((SELECT COUNT(DISTINCT course_id) FROM lessons), 0),
|
||||||
|
2
|
||||||
|
) AS avg_lessons_per_course,
|
||||||
|
ROUND(
|
||||||
|
(SELECT COUNT(*)::DECIMAL FROM lesson_views) /
|
||||||
|
NULLIF((SELECT COUNT(DISTINCT lesson_id) FROM lesson_views), 0),
|
||||||
|
2
|
||||||
|
) AS avg_views_per_lesson;
|
||||||
|
|
||||||
|
COMMIT;
|
||||||
@@ -0,0 +1,38 @@
|
|||||||
|
BEGIN;
|
||||||
|
|
||||||
|
CREATE MATERIALIZED VIEW user_activity_summary AS
|
||||||
|
WITH users_views AS (
|
||||||
|
SELECT
|
||||||
|
u.id AS user_id,
|
||||||
|
u.name,
|
||||||
|
COUNT(DISTINCT lv.lesson_id) AS viewed_lessons_count,
|
||||||
|
MAX(lv.viewed_at) AS last_view,
|
||||||
|
MIN(lv.viewed_at) AS first_view
|
||||||
|
FROM users AS u
|
||||||
|
JOIN lesson_views AS lv on u.id = lv.user_id
|
||||||
|
GROUP BY u.id
|
||||||
|
|
||||||
|
),
|
||||||
|
users_engagement AS (
|
||||||
|
SELECT
|
||||||
|
e.user_id AS user_id,
|
||||||
|
COUNT(DISTINCT l.id) AS lessons_cnt,
|
||||||
|
COUNT(DISTINCT course_id) AS enrollment_count
|
||||||
|
FROM enrollments AS e
|
||||||
|
JOIN lessons AS l using(course_id)
|
||||||
|
GROUP BY e.user_id
|
||||||
|
)
|
||||||
|
|
||||||
|
SELECT
|
||||||
|
uv.user_id,
|
||||||
|
uv.name,
|
||||||
|
ue.enrollment_count,
|
||||||
|
uv.viewed_lessons_count,
|
||||||
|
uv.last_view,
|
||||||
|
uv.first_view,
|
||||||
|
ROUND(uv.viewed_lessons_count::DECIMAL / NULLIF(ue.lessons_cnt, 0) * 100, 2) AS engagement_rate
|
||||||
|
FROM users_views AS uv
|
||||||
|
LEFT JOIN users_engagement AS ue using (user_id)
|
||||||
|
ORDER BY uv.user_id
|
||||||
|
|
||||||
|
COMMIT;
|
||||||
@@ -0,0 +1,42 @@
|
|||||||
|
BEGIN;
|
||||||
|
|
||||||
|
INSERT INTO SKONEV.users (id, name, age, email, registration_date)
|
||||||
|
VALUES
|
||||||
|
(1, 'Alice', 25, 'alice@mail.com', '2023-01-10'),
|
||||||
|
(2, 'Bob', 30, 'bob@gmail.com', '2023-02-05'),
|
||||||
|
(3, 'Charlie', 22, 'charlie@mail.com', '2023-02-20'),
|
||||||
|
(4, 'Diana', 28, 'diana@mail.com', '2023-03-01'),
|
||||||
|
(5, 'Ethan', 35, 'ethan@gmail.com', '2023-03-10');
|
||||||
|
|
||||||
|
INSERT INTO SKONEV.courses (id, title, category, created_at)
|
||||||
|
VALUES
|
||||||
|
(1, 'SQL для начинающих', 'data', '2023-01-01'),
|
||||||
|
(2, 'Python для анализа данных', 'programming', '2023-01-15'),
|
||||||
|
(3, 'BI с нуля', 'business', '2023-03-01');
|
||||||
|
|
||||||
|
INSERT INTO SKONEV.lessons (id, course_id, title, duration_min)
|
||||||
|
VALUES
|
||||||
|
(1, 1, 'SELECT и FROM', 10),
|
||||||
|
(2, 1, 'JOIN', 15),
|
||||||
|
(3, 2, 'Pandas', 20),
|
||||||
|
(4, 3, 'Основы BI', 12),
|
||||||
|
(5, 3, 'Метрики и дашборды', 18);
|
||||||
|
|
||||||
|
INSERT INTO SKONEV.enrollments (id, user_id, course_id, enrolled_at)
|
||||||
|
VALUES
|
||||||
|
(1, 1, 1, '2023-01-15'),
|
||||||
|
(2, 1, 2, '2023-02-01'),
|
||||||
|
(3, 2, 1, '2023-01-20'),
|
||||||
|
(4, 3, 2, '2023-03-05'),
|
||||||
|
(5, 4, 3, '2023-04-01');
|
||||||
|
|
||||||
|
INSERT INTO SKONEV.lesson_views (id, user_id, lesson_id, viewed_at)
|
||||||
|
VALUES
|
||||||
|
(1, 1, 1, '2023-01-16 10:00:00'),
|
||||||
|
(2, 1, 2, '2023-01-16 10:15:00'),
|
||||||
|
(3, 2, 1, '2023-01-21 09:00:00'),
|
||||||
|
(4, 3, 3, '2023-03-06 12:00:00'),
|
||||||
|
(5, 4, 4, '2023-04-02 10:00:00'),
|
||||||
|
(6, 4, 5, '2023-04-03 10:00:00');
|
||||||
|
|
||||||
|
COMMIT;
|
||||||