Merge pull request #751 from realpython/gemini-cli-vs-claude-code

Sample code for the article on Gemini CLI vs Claude Code

Author: lpozo

gemini-cli-vs-claude-code/README.md

@@ -0,0 +1,11 @@
+# Gemini CLI vs Claude Code: Which to Choose for Python Tasks
+
+This folder provides the code examples for the Real Python tutorial [Gemini CLI vs Claude Code: Which to Choose for Python Tasks](https://realpython.com/gemini-cli-vs-claude-code/).
+
+## Important Note
+
+The code in this folder has been modified from the original tutorial version to meet this repository's quality rules. The changes are as follows:
+
+- Used double quotes consistently in string literals
+- Removed unused imports
+- Reformatted lines longer than 79 characters

gemini-cli-vs-claude-code/claude-code/test_todo.py

@@ -0,0 +1,206 @@
+"""Unit tests for the to-do application."""
+
+import json
+import os
+
+import tempfile
+import unittest
+from unittest.mock import patch
+
+import todo
+
+# Point store at a temp file for every test
+import todo_store as store
+
+
+class BaseTest(unittest.TestCase):
+    """Set up a temporary tasks file for each test."""
+
+    def setUp(self):
+        self._tmp = tempfile.NamedTemporaryFile(
+            suffix=".json",
+            delete=False,
+            mode="w",
+        )
+        self._tmp.write("[]")
+        self._tmp.close()
+        self._orig = store.TASKS_FILE
+        store.TASKS_FILE = self._tmp.name
+
+    def tearDown(self):
+        store.TASKS_FILE = self._orig
+        os.unlink(self._tmp.name)
+
+
+# ── store tests ─────────────────────────────────────────────────────────────
+
+
+class TestAddTask(BaseTest):
+    def test_add_returns_task(self):
+        task = store.add_task("Buy milk")
+        self.assertEqual(task["description"], "Buy milk")
+        self.assertFalse(task["completed"])
+        self.assertEqual(task["id"], 1)
+
+    def test_ids_increment(self):
+        t1 = store.add_task("First")
+        t2 = store.add_task("Second")
+        self.assertEqual(t1["id"], 1)
+        self.assertEqual(t2["id"], 2)
+
+    def test_empty_description_raises(self):
+        with self.assertRaises(ValueError):
+            store.add_task("")
+
+    def test_whitespace_only_raises(self):
+        with self.assertRaises(ValueError):
+            store.add_task("   ")
+
+    def test_persists_to_disk(self):
+        store.add_task("Persisted")
+        with open(store.TASKS_FILE) as f:
+            data = json.load(f)
+        self.assertEqual(len(data), 1)
+        self.assertEqual(data[0]["description"], "Persisted")
+
+
+class TestCompleteTask(BaseTest):
+    def test_complete_task(self):
+        store.add_task("Write tests")
+        task = store.complete_task(1)
+        self.assertTrue(task["completed"])
+        self.assertIsNotNone(task["completed_at"])
+
+    def test_complete_nonexistent_raises(self):
+        with self.assertRaises(KeyError):
+            store.complete_task(999)
+
+    def test_complete_already_done_raises(self):
+        store.add_task("Already done")
+        store.complete_task(1)
+        with self.assertRaises(ValueError):
+            store.complete_task(1)
+
+
+class TestDeleteTask(BaseTest):
+    def test_delete_task(self):
+        store.add_task("To delete")
+        deleted = store.delete_task(1)
+        self.assertEqual(deleted["description"], "To delete")
+        self.assertEqual(store.load_tasks(), [])
+
+    def test_delete_nonexistent_raises(self):
+        with self.assertRaises(KeyError):
+            store.delete_task(42)
+
+    def test_remaining_tasks_intact(self):
+        store.add_task("Keep me")
+        store.add_task("Delete me")
+        store.delete_task(2)
+        tasks = store.load_tasks()
+        self.assertEqual(len(tasks), 1)
+        self.assertEqual(tasks[0]["description"], "Keep me")
+
+
+class TestFilterTasks(BaseTest):
+    def setUp(self):
+        super().setUp()
+        store.add_task("Pending task")
+        store.add_task("Completed task")
+        store.complete_task(2)
+        self.tasks = store.load_tasks()
+
+    def test_filter_all(self):
+        self.assertEqual(len(store.filter_tasks(self.tasks, "all")), 2)
+
+    def test_filter_pending(self):
+        result = store.filter_tasks(self.tasks, "pending")
+        self.assertEqual(len(result), 1)
+        self.assertFalse(result[0]["completed"])
+
+    def test_filter_completed(self):
+        result = store.filter_tasks(self.tasks, "completed")
+        self.assertEqual(len(result), 1)
+        self.assertTrue(result[0]["completed"])
+
+    def test_filter_unknown_raises(self):
+        with self.assertRaises(ValueError):
+            store.filter_tasks(self.tasks, "invalid")
+
+
+class TestCorruptedFile(BaseTest):
+    def test_corrupted_json_raises(self):
+        with open(store.TASKS_FILE, "w") as f:
+            f.write("not valid json{{{")
+        with self.assertRaises(ValueError):
+            store.load_tasks()
+
+    def test_non_array_json_raises(self):
+        with open(store.TASKS_FILE, "w") as f:
+            json.dump({"key": "value"}, f)
+        with self.assertRaises(ValueError):
+            store.load_tasks()
+
+    def test_missing_file_returns_empty(self):
+        os.unlink(store.TASKS_FILE)
+        self.assertEqual(store.load_tasks(), [])
+        # restore so tearDown doesn't crash
+        with open(store.TASKS_FILE, "w") as f:
+            f.write("[]")
+
+
+# ── CLI integration tests ───────────────────────────────────────────────────
+
+
+class TestCLI(BaseTest):
+    def _run(self, argv):
+        """Run CLI with given argv list, return exit code."""
+        with patch("sys.argv", ["todo"] + argv):
+            parser = todo.build_parser()
+            args = parser.parse_args()
+            return args.func(args)
+
+    def test_add_command(self):
+        code = self._run(["add", "CLI task"])
+        self.assertEqual(code, 0)
+        self.assertEqual(len(store.load_tasks()), 1)
+
+    def test_list_command(self):
+        store.add_task("Listed task")
+        code = self._run(["list"])
+        self.assertEqual(code, 0)
+
+    def test_list_pending_filter(self):
+        store.add_task("Pending")
+        store.add_task("Done")
+        store.complete_task(2)
+        code = self._run(["list", "--status", "pending"])
+        self.assertEqual(code, 0)
+
+    def test_done_command(self):
+        store.add_task("Mark done")
+        code = self._run(["done", "1"])
+        self.assertEqual(code, 0)
+        self.assertTrue(store.load_tasks()[0]["completed"])
+
+    def test_delete_command(self):
+        store.add_task("Remove me")
+        code = self._run(["delete", "1"])
+        self.assertEqual(code, 0)
+        self.assertEqual(store.load_tasks(), [])
+
+    def test_done_missing_id_returns_error(self):
+        code = self._run(["done", "99"])
+        self.assertEqual(code, 1)
+
+    def test_delete_missing_id_returns_error(self):
+        code = self._run(["delete", "99"])
+        self.assertEqual(code, 1)
+
+    def test_add_empty_returns_error(self):
+        code = self._run(["add", ""])
+        self.assertEqual(code, 1)
+
+
+if __name__ == "__main__":
+    unittest.main()

gemini-cli-vs-claude-code/claude-code/todo.py

@@ -0,0 +1,132 @@
+#!/usr/bin/env python3
+"""CLI to-do application.
+
+Usage:
+    todo.py add "Buy groceries"
+    todo.py list
+    todo.py list --status pending
+    todo.py list --status completed
+    todo.py done <id>
+    todo.py delete <id>
+"""
+
+import argparse
+import sys
+
+import todo_store as store
+
+# ── Formatting helpers ──────────────────────────────────────────────────────
+
+CHECK = "[x]"
+EMPTY = "[ ]"
+
+
+def _fmt_task(task: dict) -> str:
+    status = CHECK if task["completed"] else EMPTY
+    suffix = (
+        f"  (done {task['completed_at'][:10]})" if task["completed_at"] else ""
+    )
+    return f"  {task['id']:>3}  {status}  {task['description']}{suffix}"
+
+
+# ── Command handlers ────────────────────────────────────────────────────────
+
+
+def cmd_add(args: argparse.Namespace) -> int:
+    try:
+        task = store.add_task(args.description)
+        print(f"Added task #{task['id']}: {task['description']}")
+        return 0
+    except ValueError as e:
+        print(f"Error: {e}", file=sys.stderr)
+        return 1
+
+
+def cmd_list(args: argparse.Namespace) -> int:
+    try:
+        tasks = store.load_tasks()
+        filtered = store.filter_tasks(tasks, args.status)
+    except ValueError as e:
+        print(f"Error: {e}", file=sys.stderr)
+        return 1
+
+    if not filtered:
+        label = "" if args.status == "all" else f"{args.status} "
+        print(f"No {label}tasks found.")
+        return 0
+
+    label = "" if args.status == "all" else f"{args.status} "
+    print(f"\n--- {label}tasks ({len(filtered)}) ---")
+    for task in filtered:
+        print(_fmt_task(task))
+    print()
+    return 0
+
+
+def cmd_done(args: argparse.Namespace) -> int:
+    try:
+        task = store.complete_task(args.id)
+        print(f"Completed task #{task['id']}: {task['description']}")
+        return 0
+    except (KeyError, ValueError) as e:
+        print(f"Error: {e}", file=sys.stderr)
+        return 1
+
+
+def cmd_delete(args: argparse.Namespace) -> int:
+    try:
+        task = store.delete_task(args.id)
+        print(f"Deleted task #{task['id']}: {task['description']}")
+        return 0
+    except KeyError as e:
+        print(f"Error: {e}", file=sys.stderr)
+        return 1
+
+
+# ── Argument parsing ────────────────────────────────────────────────────────
+
+
+def build_parser() -> argparse.ArgumentParser:
+    parser = argparse.ArgumentParser(
+        prog="todo",
+        description="A simple CLI to-do application.",
+    )
+    sub = parser.add_subparsers(dest="command", metavar="<command>")
+    sub.required = True
+
+    # add
+    p_add = sub.add_parser("add", help="Add a new task")
+    p_add.add_argument("description", help="Task description")
+    p_add.set_defaults(func=cmd_add)
+
+    # list
+    p_list = sub.add_parser("list", help="List tasks")
+    p_list.add_argument(
+        "--status",
+        choices=["all", "pending", "completed"],
+        default="all",
+        help="Filter by status (default: all)",
+    )
+    p_list.set_defaults(func=cmd_list)
+
+    # done
+    p_done = sub.add_parser("done", help="Mark a task as completed")
+    p_done.add_argument("id", type=int, help="Task ID")
+    p_done.set_defaults(func=cmd_done)
+
+    # delete
+    p_del = sub.add_parser("delete", help="Delete a task")
+    p_del.add_argument("id", type=int, help="Task ID")
+    p_del.set_defaults(func=cmd_delete)
+
+    return parser
+
+
+def main() -> int:
+    parser = build_parser()
+    args = parser.parse_args()
+    return args.func(args)
+
+
+if __name__ == "__main__":
+    sys.exit(main())