API Exception Handling

This repository provides a guide on how to handle exceptions in a basic API that communicates with a SQLite database. It includes four versions, each with their own set of enhancements.

api_exception_handling
│
├── notebooks/
│   ├── logging_decorator.ipynb
│   └── retry_decorator.ipynb
│
├── logs/
│   ├── README.md
│   └── application.log
│
├── utils/
│   ├── __init__.py
│   ├── decorator1.py
│   ├── decorator2.py
│   └── ...
│
├── versions/
│   ├── legacy/
│   │   └── __init__.py
│   ├── v1/
│   │   ├── __init__.py
│   │   └── TO-DO.md
│   ├── v2/
│   │   ├── __init__.py
│   │   └── TO-DO.md
│   ├── v3/
│   │   ├── __init__.py
│   │   └── TO-DO.md
│   └── v4/
│       ├── __init__.py
│       └── TO-DO.md
│
├── data/
│   └── application.db
│
├── .gitignore
├── README.md
└── __init__.py

Version 1

The first version of the API implements the following steps:

1. Identify potential error scenarios
2. Separate library-specific errors
3. Handle low-level errors
4. Raise errors suitable to the level above
5. Add custom exception types

Version 2

The second version of the API implements the following steps:

• Automate the database creation and run it as the main program
• Integrate Context Manager

class SQLite:
    def __init__(self, file="application.db"):
        self.file = file
    def __enter__(self):
        self.conn = sqlite3.connect(self.file)
        return self.conn.cursor()
    def __exit__(self, type, value, traceback):
        self.conn.close()

Version 3

The third version of the API implements the following steps:

• Create a retry() decorator to retry database connection in case of a failure
• Create a log_exceptions() decorator to log errors and exceptions during data fetching.

@retry(sqlite3.OperationalError, tries=3, delay=1, backoff=2)
def create_database() -> None:
    with sqlite3.connect('application.db') as conn:
        cur = conn.cursor()

        cur.execute('''CREATE TABLE blogs
                    (id text NOT NULL PRIMARY KEY,
                    date TEXT,
                    title TEXT,
                    content TEXT,
                    public INTEGER)''')

        cur.execute("INSERT INTO blogs VALUES ('first-blog', '2021-03-07', 'My first blog' ,'Some content', 1)")

        cur.execute("INSERT INTO blogs VALUES ('private-blog', '2021-03-07', 'Secret blog' ,'This is a secret', 0)")


        conn.commit()

@app.route('/blogs/<id>')
@log_exceptions()
def get_blog(blog_id):
    try:
        return jsonify(fetch_blog(blog_id))
    except NotFoundError:
        abort(404, description="Resource not found.")
    except NotAuthorizedError:
        abort(403, description="Access denied.")

Version 4

The fourth version refactors the code to demonstrate monadic error handling concepts with the Result type from the returns library. The Result type is a monadic data type that represents the outcome of a computation that may succeed or fail. The Success subclass holds the result of the computation, while the Failure subclass holds the exception.

The updated version of the @log_exceptions decorator now logs exceptions to version4_logs.log file.

Each version is contained within its own directory and is documented in a separate markdown file.