## SQLにおけるEXTRACT関数の理解

SQLの**EXTRACT**関数は、日付や時刻の値から特定の部分を抽出するためのものです。例えば、完全な日付から年、月、日だけを取得することができます。これは、年ごとのサインアップ数を分析するなど、日付の一部でデータをグループ化したりフィルタリングしたりする際に特に便利です。

**例：**

```sql
SELECT EXTRACT(YEAR FROM signup_date) AS signup_year FROM customers;
```

このステートメントは、各顧客の`signup_date`から年だけを返します。`EXTRACT`を使用し、関数内の引数を変更することで、月や日を取得することも可能です。

SELECT EXTRACT(YEAR FROM DATE '2024-06-01') AS year;

import unittest
import psycopg2
import pandas as pd
import re

class TestPostgresQuery(unittest.TestCase):
    @classmethod
    def setUpClass(cls):
        cls.connection = psycopg2.connect(
            host="localhost",
            user="agent",
            password="qwerty",
            dbname="testdb",
            sslmode="disable"
        )
        cls.cursor = cls.connection.cursor()

    def setUp(self):
        try:
            self.cursor.execute("ROLLBACK;")
        except Exception:
            pass
        self.connection.commit()
        # Only add signup_date column if it does not exist
        self.cursor.execute("""
            SELECT column_name FROM information_schema.columns
            WHERE table_name = 'customers' AND column_name = 'signup_date';
        """)
        exists = self.cursor.fetchone()
        if not exists:
            self.cursor.execute("ALTER TABLE customers ADD COLUMN signup_date DATE;")
            self.connection.commit()
        # Always update signup dates regardless
        self.cursor.execute("""
            UPDATE customers SET signup_date = '2021-03-15' WHERE customer_id = 1;
            UPDATE customers SET signup_date = '2022-07-22' WHERE customer_id = 2;
            UPDATE customers SET signup_date = '2022-12-05' WHERE customer_id = 3;
            UPDATE customers SET signup_date = '2023-01-18' WHERE customer_id = 4;
            UPDATE customers SET signup_date = '2023-06-09' WHERE customer_id = 5;
            UPDATE customers SET signup_date = '2023-09-27' WHERE customer_id = 6;
            UPDATE customers SET signup_date = '2024-02-14' WHERE customer_id = 7;
            UPDATE customers SET signup_date = '2024-04-02' WHERE customer_id = 8;
        """)
        self.connection.commit()

    def test_group_by_signup_year(self):
        """Check if the query groups customers by signup year and counts signups correctly."""
        with open("query.sql", "r") as file:
            query = file.read()
        # Remove any ALTER TABLE statement from the query
        query = re.sub(r"(?im)^.*alter table customers add column signup_date date;.*$", "", query)
        self.cursor.execute(query)
        rows = self.cursor.fetchall()
        columns = [desc[0] for desc in self.cursor.description]
        df = pd.DataFrame(rows, columns=columns)
        expected = pd.DataFrame({
            "signup_year": [2021.0, 2022.0, 2023.0, 2024.0],
            "signup_count": [1, 2, 3, 2]
        })
        # Check that all expected years and counts are present
        merged = pd.merge(df, expected, on=["signup_year", "signup_count"], how="outer", indicator=True)
        missing = merged[merged['_merge'] != 'both']
        condition = missing.empty
        _dynamic_test(
            self,
            condition,
            "Query groups and counts signups by year correctly.",
            f"Query result missing or incorrect for signup years/counts: {missing.to_dict(orient='records')}"
        )

    def test_one_row_per_year(self):
        """Check that only one row per unique signup year is returned."""
        with open("query.sql", "r") as file:
            query = file.read()
        # Remove any ALTER TABLE statement from the query
        query = re.sub(r"(?im)^.*alter table customers add column signup_date date;.*$", "", query)
        self.cursor.execute(query)
        rows = self.cursor.fetchall()
        columns = [desc[0] for desc in self.cursor.description]
        df = pd.DataFrame(rows, columns=columns)
        years = df['signup_year'].tolist()
        unique_years = set(years)
        counts = df['signup_year'].value_counts()
        condition = all(counts[year] == 1 for year in unique_years)
        _dynamic_test(
            self,
            condition,
            "Only one row per unique signup year.",
            "There are duplicate rows for some signup years."
        )

    def test_columns_and_order(self):
        """Check that columns are named correctly and ordered by signup_year ascending."""
        with open("query.sql", "r") as file:
            query = file.read()
        # Remove any ALTER TABLE statement from the query
        query = re.sub(r"(?im)^.*alter table customers add column signup_date date;.*$", "", query)
        self.cursor.execute(query)
        rows = self.cursor.fetchall()
        columns = [desc[0] for desc in self.cursor.description]
        df = pd.DataFrame(rows, columns=columns)
        # Check column names
        condition_cols = (normalize_text(columns[0]) == "signup_year" and normalize_text(columns[1]) == "signup_count")
        # Check order ascending
        condition_order = (df['signup_year'].tolist() == sorted(df['signup_year'].tolist()))
        _dynamic_test(
            self,
            condition_cols and condition_order,
            "Columns are named correctly and results are ordered by signup_year ascending.",
            "Columns are misnamed or results are not ordered by signup_year ascending."
        )

    def test_grouping_uses_extract(self):
        """Check that EXTRACT(YEAR FROM signup_date) is used in SELECT or GROUP BY."""
        with open("query.sql", "r") as file:
            query = file.read().lower()
        uses_extract = "extract(year from signup_date)" in query
        _dynamic_test(
            self,
            uses_extract,
            "Query uses EXTRACT(YEAR FROM signup_date) to group by year.",
            "Query does not use EXTRACT(YEAR FROM signup_date) to group by year."
        )

def _dynamic_test(test_case, condition, success_message, failure_message):
    if condition:
        test_case._testMethodName = success_message
        test_case.assertTrue(True, success_message)
    else:
        test_case._testMethodName = failure_message
        test_case.fail(failure_message)

def normalize_text(text):
    text = text.lower()
    text = re.sub(r"\\s{2,}", " ", text)
    text = re.sub(r"\\s*([,:?])\\s*", r"\\1 ", text)
    return text.strip()

if __name__ == "__main__":
    unittest.main()


test_solution.py

実践的なデータ分析スキルに焦点を当て、SQLを基礎から学べる包括的で初心者向けのコースです。分かりやすい解説、実際の事例、実践的な課題を通じて、SQLを用いたデータのクエリ、変換、分析を習得します。

リレーショナルデータベースとSQLの基礎概念を学習します。テーブルからデータを取得、フィルタリング、並べ替える方法を探求し、より高度なデータ分析の基礎を築きます。

SQLの強力な集計機能を深く学びます。データを要約し、グループ化し、より深い洞察のために分析する方法を習得します。

複数のテーブルからデータを組み合わせる力を引き出します。さまざまな種類の結合を実行し、複雑なビジネス上の質問に答える方法を学びます。

サブクエリ、ウィンドウ関数、データ変換などを含む高度なSQL技術を習得し、強力なデータ分析を実現します。

チャレンジ：カスタマーサインアップ年

SQLにおけるEXTRACT関数の理解

解答