Add functional tests

.github/workflows/tests.yml

@@ -12,6 +12,13 @@ env:
 jobs:
   the_job:
     runs-on: ubuntu-latest
+    services:
+      postgres:
+        image: postgres:15
+        env:
+          POSTGRES_PASSWORD: password
+        ports:
+          - 5432:5432
     steps:
       - name: Checkout Code
         uses: actions/checkout@v3
@@ -40,7 +47,11 @@ jobs:
         if: steps.poetry-cache.outputs.cache-hit != 'true'
         run: |
           python -m poetry install
+      - name: Create src database
+        shell: bash
+        run: |
+          PGPASSWORD=password psql --host=localhost --username=postgres --file=tests/examples/src.dump
       - name: Run Unit Tests
         shell: bash
         run: |
-          poetry run python -m unittest discover --verbose tests
+          FUNCTIONAL_TESTS=1 poetry run python -m unittest discover --verbose tests

README.md

@@ -1,3 +1,32 @@
 # sqlsynthgen
 
 Synthetic data for SQL databases
+
+## Development
+
+### Setup
+
+### Poetry
+
+1. Install [Poetry](https://python-poetry.org/docs/#installation).
+1. Create a Poetry environment and install dependencies with `poetry install`.
+1. Activate a Poetry shell with `poetry shell`.
+   We will assume that this shell is active for the remainder of the instructions.
+
+### Pre-Commit
+
+1. Install [Pre-commit](https://pre-commit.com/#install).
+1. Install Pre-commit hooks with `pre-commit install --install-hooks`.
+   The hooks will run whenever you perform a Git commit.
+
+### Testing
+
+You can execute most unit tests by running `python -m unittest discover --verbose tests/` from the root directory.
+However, the functional tests require:
+
+1. A local PostgreSQL server to be running with:
+    1. A `postgres` user whose password is `password`.
+    1. Databases named `src` and `dst`.
+    1. See [tests.yml](.github/workflows/tests.yml) and [test_functional.py](tests/test_functional.py) for more.
+1. The `FUNCTIONAL_TESTS` environment variable to be set to `1`.
+   For example, you could run `FUNCTIONAL_TESTS=1 python -m unittest discover --verbose tests/`.

tests/examples/dst.dump

@@ -0,0 +1,27 @@
+--
+-- PostgreSQL database dump
+--
+
+-- Dumped from database version 14.2 (Debian 14.2-1.pgdg110+1)
+-- Dumped by pg_dump version 14.6 (Homebrew)
+
+SET statement_timeout = 0;
+SET lock_timeout = 0;
+SET idle_in_transaction_session_timeout = 0;
+SET client_encoding = 'UTF8';
+SET standard_conforming_strings = on;
+SELECT pg_catalog.set_config('search_path', '', false);
+SET check_function_bodies = false;
+SET xmloption = content;
+SET client_min_messages = warning;
+SET row_security = off;
+
+DROP DATABASE IF EXISTS dst;
+--
+-- Name: dst; Type: DATABASE; Schema: -; Owner: postgres
+--
+
+CREATE DATABASE dst WITH TEMPLATE = template0 ENCODING = 'UTF8' LOCALE = 'en_US.utf8';
+
+
+ALTER DATABASE dst OWNER TO postgres;

tests/examples/src.dump

@@ -0,0 +1,109 @@
+--
+-- PostgreSQL database dump
+--
+
+-- Dumped from database version 14.2 (Debian 14.2-1.pgdg110+1)
+-- Dumped by pg_dump version 14.6 (Homebrew)
+
+SET statement_timeout = 0;
+SET lock_timeout = 0;
+SET idle_in_transaction_session_timeout = 0;
+SET client_encoding = 'UTF8';
+SET standard_conforming_strings = on;
+SELECT pg_catalog.set_config('search_path', '', false);
+SET check_function_bodies = false;
+SET xmloption = content;
+SET client_min_messages = warning;
+SET row_security = off;
+
+DROP DATABASE IF EXISTS src;
+--
+-- Name: src; Type: DATABASE; Schema: -; Owner: postgres
+--
+
+CREATE DATABASE src WITH TEMPLATE = template0 ENCODING = 'UTF8' LOCALE = 'en_US.utf8';
+
+
+ALTER DATABASE src OWNER TO postgres;
+
+\connect src
+
+SET statement_timeout = 0;
+SET lock_timeout = 0;
+SET idle_in_transaction_session_timeout = 0;
+SET client_encoding = 'UTF8';
+SET standard_conforming_strings = on;
+SELECT pg_catalog.set_config('search_path', '', false);
+SET check_function_bodies = false;
+SET xmloption = content;
+SET client_min_messages = warning;
+SET row_security = off;
+
+SET default_tablespace = '';
+
+SET default_table_access_method = heap;
+
+--
+-- Name: hospital_visit; Type: TABLE; Schema: public; Owner: postgres
+--
+
+CREATE TABLE public.hospital_visit (
+    hospital_visit_id bigint NOT NULL,
+    person_id integer NOT NULL,
+    visit_start date NOT NULL,
+    visit_duration_seconds real NOT NULL,
+    visit_image bytea NOT NULL
+);
+
+
+ALTER TABLE public.hospital_visit OWNER TO postgres;
+
+--
+-- Name: person; Type: TABLE; Schema: public; Owner: postgres
+--
+
+CREATE TABLE public.person (
+    person_id integer NOT NULL,
+    name text NOT NULL,
+    research_opt_out boolean NOT NULL,
+    stored_from timestamp with time zone NOT NULL
+);
+
+
+ALTER TABLE public.person OWNER TO postgres;
+
+--
+-- Data for Name: hospital_visit; Type: TABLE DATA; Schema: public; Owner: postgres
+--
+
+COPY public.hospital_visit (hospital_visit_id, person_id, visit_start, visit_duration_seconds, visit_image) FROM stdin;
+\.
+
+
+--
+-- Data for Name: person; Type: TABLE DATA; Schema: public; Owner: postgres
+--
+
+COPY public.person (person_id, name, research_opt_out, stored_from) FROM stdin;
+\.
+
+
+--
+-- Name: hospital_visit hospital_visit_pkey; Type: CONSTRAINT; Schema: public; Owner: postgres
+--
+
+ALTER TABLE ONLY public.hospital_visit
+    ADD CONSTRAINT hospital_visit_pkey PRIMARY KEY (hospital_visit_id);
+
+
+--
+-- Name: person person_pkey; Type: CONSTRAINT; Schema: public; Owner: postgres
+--
+
+ALTER TABLE ONLY public.person
+    ADD CONSTRAINT person_pkey PRIMARY KEY (person_id);
+
+
+--
+-- PostgreSQL database dump complete
+--

tests/test_functional.py

@@ -0,0 +1,77 @@
+"""Tests for the main module."""
+import os
+from pathlib import Path
+from subprocess import run
+from unittest import TestCase, skipUnless
+
+
+@skipUnless(
+    os.environ.get("FUNCTIONAL_TESTS") == "1", "Set 'FUNCTIONAL_TESTS=1' to enable."
+)
+class FunctionalTests(TestCase):
+    """End-to-end tests."""
+
+    orm_file_path = Path("tests/tmp/orm.py")
+    ssg_file_path = Path("tests/tmp/ssg.py")
+
+    def setUp(self) -> None:
+        """Pre-test setup."""
+        self.orm_file_path.unlink(missing_ok=True)
+        self.ssg_file_path.unlink(missing_ok=True)
+
+        # If you need to update src.dump or dst.dump, use
+        # pg_dump -d src|dst -h localhost -U postgres -C -c > tests/examples/src|dst.dump
+
+        env = os.environ.copy()
+        env = {**env, "PGPASSWORD": "password"}
+
+        # Clear and re-create the destination database
+        completed_process = run(
+            [
+                "psql",
+                "--host=localhost",
+                "--username=postgres",
+                "--file=" + str(Path("tests/examples/dst.dump")),
+            ],
+            capture_output=True,
+            env=env,
+            check=True,
+        )
+
+        # psql doesn't always return != 0 if it fails
+        assert completed_process.stderr == b"", completed_process.stderr
+
+    def test_workflow(self) -> None:
+        """Test the recommended CLI workflow runs without errors."""
+
+        env = os.environ.copy()
+        env = {
+            **env,
+            "src_host_name": "localhost",
+            "src_user_name": "postgres",
+            "src_password": "password",
+            "src_db_name": "src",
+            "src_schema": "",
+            "dst_host_name": "localhost",
+            "dst_user_name": "postgres",
+            "dst_password": "password",
+            "dst_db_name": "dst",
+        }
+
+        with open(self.orm_file_path, "wb") as file:
+            run(["sqlsynthgen", "make-tables"], stdout=file, env=env, check=True)
+
+        with open(self.ssg_file_path, "wb") as file:
+            run(
+                ["sqlsynthgen", "make-generators", self.orm_file_path],
+                stdout=file,
+                env=env,
+                check=True,
+            )
+
+        run(["sqlsynthgen", "create-tables", self.orm_file_path], env=env, check=True)
+        run(
+            ["sqlsynthgen", "create-data", self.orm_file_path, self.ssg_file_path],
+            env=env,
+            check=True,
+        )

tests/tmp/.gitignore

@@ -0,0 +1 @@
+*