refactor: rename direction to address, inline the last SQL files

README.md

@@ -185,7 +185,6 @@ sql-performance/
 │   ├── index_usage.py       # B-tree index impact
 │   ├── join_vs_subquery.py  # JOIN vs IN vs EXISTS
 │   └── pagination.py        # OFFSET vs keyset
-├── queries/                 # SQL files (loaded by benchmarks)
 ├── sql/
 │   ├── init.sql             # Schema only
 │   └── seed.py              # Parametrized seeder (small/medium/large)

app/config.py

@@ -2,10 +2,6 @@
 from pathlib import Path
 
 BASE_DIR = Path(__file__).resolve().parent.parent
-QUERIES_DIR = BASE_DIR / 'queries'
-
-QUERY_1_NAME = 'SELECT *'
-QUERY_2_NAME = 'SELECT id, name, email'
 
 
 class Config:

benchmarks/select_star.py

@@ -7,7 +7,6 @@
 from matplotlib.backends.backend_agg import FigureCanvasAgg
 from matplotlib.figure import Figure
 
-from app.config import QUERIES_DIR, QUERY_1_NAME, QUERY_2_NAME
 from benchmarks.base import (
     BenchmarkBase,
     BenchmarkNotApplicable,
@@ -20,6 +19,12 @@
 
 DATA_POINTS = 10
 
+QUERY_ALL_COLUMNS = "SELECT * FROM users LIMIT %s"
+QUERY_THREE_COLUMNS = "SELECT id, name, email FROM users LIMIT %s"
+
+LABEL_ALL_COLUMNS = "SELECT *"
+LABEL_THREE_COLUMNS = "SELECT id, name, email"
+
 
 @register
 class SelectStarBenchmark(BenchmarkBase):
@@ -28,10 +33,6 @@ class SelectStarBenchmark(BenchmarkBase):
     description = "Demonstrates why SELECT * is slow and selecting specific columns is fast"
     required_tables = ["users"]
 
-    def _get_query(self, filename: str) -> str:
-        with open(QUERIES_DIR / filename) as f:
-            return f.read()
-
     def _limits(self, total: int) -> list[int]:
         """LIMITs must stay inside the table: a LIMIT above the row count
         returns the whole table every time and flattens the curve into noise."""
@@ -47,27 +48,25 @@ def setup(self, conn) -> None:
 
     def run(self, conn) -> BenchmarkResult:
         limits = self._limits(table_row_count(conn))
-        query_1 = self._get_query("query_1.sql")
-        query_2 = self._get_query("query_2.sql")
 
         q1_times, q2_times = [], []
         q1_rows, q2_rows = [], []
 
         with conn.cursor() as cursor:
             for limit in limits:
-                rows_1, t1 = measure(cursor, query_1, (limit,))
-                rows_2, t2 = measure(cursor, query_2, (limit,))
+                rows_1, t1 = measure(cursor, QUERY_ALL_COLUMNS, (limit,))
+                rows_2, t2 = measure(cursor, QUERY_THREE_COLUMNS, (limit,))
                 q1_times.append(t1)
                 q2_times.append(t2)
                 q1_rows.append(rows_1)
                 q2_rows.append(rows_2)
 
         q1 = QueryResult(
-            name=QUERY_1_NAME, query=query_1,
+            name=LABEL_ALL_COLUMNS, query=QUERY_ALL_COLUMNS,
             times=q1_times, limits=limits, rows_fetched=q1_rows,
         )
         q2 = QueryResult(
-            name=QUERY_2_NAME, query=query_2,
+            name=LABEL_THREE_COLUMNS, query=QUERY_THREE_COLUMNS,
             times=q2_times, limits=limits, rows_fetched=q2_rows,
         )
 

sql/init.sql

@@ -8,7 +8,7 @@ CREATE TABLE IF NOT EXISTS users (
     name VARCHAR(100),
     surname VARCHAR(100),
     email VARCHAR(100),
-    direction VARCHAR(200),
+    address VARCHAR(200),
     city VARCHAR(100),
     country VARCHAR(100),
     postal_code VARCHAR(20),

sql/seed.py

@@ -40,7 +40,7 @@
            %(cities)s::TEXT[]      AS cities,
            %(countries)s::TEXT[]   AS countries
 )
-INSERT INTO users (name, surname, email, direction, city, country,
+INSERT INTO users (name, surname, email, address, city, country,
                    postal_code, phone, age, bio)
 SELECT
     a.first_names[1 + floor(random() * array_length(a.first_names, 1))::INT],

tests/test_config.py

@@ -1,4 +1,4 @@
-from app.config import BASE_DIR, QUERIES_DIR, Config
+from app.config import BASE_DIR, Config
 
 
 def test_config_defaults():
@@ -35,9 +35,3 @@ def test_config_port_cast(monkeypatch):
 def test_base_dir():
     assert BASE_DIR.exists()
     assert (BASE_DIR / 'app').exists()
-
-
-def test_queries_dir():
-    assert QUERIES_DIR.exists()
-    assert (QUERIES_DIR / 'query_1.sql').exists()
-    assert (QUERIES_DIR / 'query_2.sql').exists()

tests/test_integration.py

@@ -45,7 +45,7 @@ def test_seed_leaves_no_null_values_in_generated_columns(conn):
     with conn.cursor() as cur:
         cur.execute(
             "SELECT count(*) FROM users "
-            "WHERE name IS NULL OR surname IS NULL "
+            "WHERE name IS NULL OR surname IS NULL OR address IS NULL "
             "OR city IS NULL OR country IS NULL"
         )
         assert cur.fetchone()[0] == 0