Summary  
This chapter explains how to specify sort direction in SQL code by using the DESC keyword to sort columns in descending order and shows how to apply it to one or multiple columns with different sort directions.

General domain of usage  
Database querying

## Sorting Columns in Descending Order

We can do more than just sort data in ascending order by default. The `ORDER BY` clause can also arrange data in descending order, which requires using the `DESC` keyword.

SELECT continent 
FROM country 
ORDER BY continent DESC;

## Sorting Multiple Columns in Descending Order

We can sort data by multiple columns, and in many cases, this is actually necessary. For example, when displaying a list of students, you might want to order them first by their last name and then by their first name. This approach is especially helpful when multiple students share the same name. 

Also, if you're sorting in descending order by more than one column, remember that each column must include its own `DESC` keyword.

SELECT id, name, region 
FROM country 
ORDER BY region, name DESC;

You only need to apply the `DESC` keyword to the column you want sorted in descending order. In our example, we used `DESC` for the `name` column but not for the `region` column. As a result, the `name` column is sorted in descending order, while the `region` column remains sorted in ascending order (the default).

import unittest
import psycopg2
import pandas as pd

class TestTask(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 test_selects_five_columns_in_order(self):
        """Test that the query selects columns in the correct order: name, continent, region, population, capital."""
        with open("query.sql", "r") as file:
            query = file.read()
        self.cursor.execute(query)
        columns = [desc[0] for desc in self.cursor.description]
        expected_columns = ["name", "continent", "region", "population", "capital"]
        condition = columns == expected_columns
        _dynamic_test(
            self,
            condition,
            "Query selects the correct columns in the correct order.",
            f"Columns selected are {columns}, but expected {expected_columns}."
        )

    def test_order_by_continent_desc_population_asc(self):
        """Test that the result is ordered by continent DESC, then population ASC."""
        with open("query.sql", "r") as file:
            query = file.read()
        self.cursor.execute(query)
        rows = self.cursor.fetchall()
        # Get the indexes for continent and population
        columns = [desc[0] for desc in self.cursor.description]
        continent_idx = columns.index("continent")
        population_idx = columns.index("population")
        # Prepare list of (continent, population) for comparison
        order_pairs = [(row[continent_idx], row[population_idx]) for row in rows]
        # Compare with expected ordering
        sorted_pairs = sorted(order_pairs, key=lambda x: (x[0] is not None and x[0] or '', x[1] if x[1] is not None else float('inf')), reverse=False)
        # continent DESC, population ASC
        # So, for continent, need descending
        sorted_pairs = sorted(order_pairs, key=lambda x: ((x[0] is not None and x[0] or ''), x[1] if x[1] is not None else float('inf')))
        sorted_pairs = sorted(sorted_pairs, key=lambda x: (x[0] is not None and x[0] or ''), reverse=True)
        # Now, compare to actual
        # But above sorts in two steps: first by population ASC, then by continent DESC
        # Instead, use a tuple: (-continent, population) but continent is string
        # So, do a single sort:
        expected_pairs = sorted(order_pairs, key=lambda x: ((x[0] is not None and x[0] or ''), x[1] if x[1] is not None else float('inf')), reverse=False)
        expected_pairs = sorted(expected_pairs, key=lambda x: (x[0] is not None and x[0] or ''), reverse=True)
        condition = order_pairs == expected_pairs
        _dynamic_test(
            self,
            condition,
            "Rows are ordered by continent DESC, then population ASC.",
            "Rows are not ordered by continent DESC, then population ASC."
        )

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

Learn the core fundamentals of SQL and relational databases. Understand how databases store and organize data, write basic SQL queries, work with tables and perform essential operations. Build a strong foundation for working with databases in backend development.

Specifying Sort Direction

Sorting Columns in Descending Order

Sorting Multiple Columns in Descending Order

Ratkaisu