flask-selfheal 🔗

Automatically fix broken URLs in your Flask app using intelligent resolvers.

Features

• Multiple Matching Strategies: Choose from 1:1 exact matching, fuzzy matching, and even hybrid database-backed approaches to find the best matching URL.
• Configurable: Fine-tune matching strategies, thresholds, and normalization rules to suit your needs.
• Chainable Resolvers: Combine multiple resolvers to create a robust URL healing strategy.
• Easy Integration: Plug into any Flask app with minimal setup.
• Performant: Optimized to try fast methods first and only use expensive ones as a last resort.

Installation

uv pip install flask-selfheal

Usage

Fuzzy Flask Routes Resolver

This example demonstrates how to use the FlaskRoutesResolver to automatically correct common typos in your Flask routes using fuzzy matching (using difflib under the hood).

from flask import Flask
from flask_selfheal import SelfHeal
from flask_selfheal.resolvers import FlaskRoutesResolver

app = Flask(__name__)

@app.route("/home")
def home():
    return "Welcome Home!"

@app.route("/about")
def about():
    return "About Us"

@app.route("/contact")
def contact():
    return "Contact Us"

# Configure SelfHeal with fuzzy matching
SelfHeal(app, resolvers=[FlaskRoutesResolver()])

if __name__ == "__main__":
    app.run(debug=True)

https://example.com/hme    --> Redirects to /home
https://example.com/abot   --> Redirects to /about
https://example.com/contat --> Redirects to /contact

Chaining Multiple Resolvers

You can combine multiple resolvers to create a more robust URL healing strategy. In this example, we use both AliasMappingResolver and FuzzyMappingResolver to handle obsolete URLs and common typos.

from flask import Flask
from flask_selfheal import SelfHeal
from flask_selfheal.resolvers import AliasMappingResolver, FuzzyMappingResolver

app = Flask(__name__)

@app.route("/home")
def home():
    return "Welcome Home!"

@app.route("/new-path")
def new_path():
    return "This is the new path!"

# Define resolvers - will be tried in order defined
resolvers = [
    AliasMappingResolver(
        {"old-path": "new-path"}  # Handle obsolete URLs
    ),
    FuzzyMappingResolver(
        ["home", "new-path"]  # Handle typos
    )
]

SelfHeal(app, resolvers=resolvers)

if __name__ == "__main__":
    app.run(debug=True)

https://example.com/old-path  --> Redirects to /new-path
https://example.com/hme       --> Redirects to /home
https://example.com/new-pth   --> Redirects to /new-path

Database-Backed Resolver

The DatabaseResolver allows you to resolve URLs based on the slug in your database (using SQLAlchemy).

Ensure that you have a model with a slug field in your database (example below).

class Articles(db.Model):
    id = db.Column(db.Integer, primary_key=True)
    slug = db.Column(db.String(100), unique=True, nullable=False) # <--
    title = db.Column(db.String(100), nullable=False)
    content = db.Column(db.Text, nullable=False)

from flask import Flask
from app.models import Articles  # Your SQLAlchemy model
from flask_selfheal import SelfHeal
from flask_selfheal.resolvers import DatabaseResolver

app = Flask(__name__)
app.config["SQLALCHEMY_DATABASE_URI"] = "sqlite:///yourdatabase.db"
db.init_app(app)

@app.route("/articles/<slug>")
def article_detail(slug):
    article = Articles.query.filter_by(slug=slug).first_or_404()
    return f"Article: {article.title}"

# Configure the DatabaseResolver
db_resolver = [
    DatabaseResolver(
        Articles,
        slug_field='slug',
    )
]

SelfHeal(app, resolvers=db_resolver, redirect_pattern="/articles/{slug}")

if __name__ == "__main__":
    app.run(debug=True)

https://example.com/articles/1234         --> Redirects to /articles/hello-world-1234567
https://example.com/articles/hello-world  --> Redirects to /articles/hello-world-1234567
https://example.com/articles/hell-wrl     --> Redirects to /articles/hello-world-1234567
https://example.com/articles/world        --> Redirects to /articles/hello-world-1234567

The DatabaseResolver follows these strategies in order when trying to find a matching slug:

1. Exact match: Fastest, direct database lookup
2. SQL LIKE matching: Handle simple variations using SQL LIKE wildcards
3. Normalized matching: Handle common typos (0 -> o, 1 -> l, etc.)
4. Word-based matching: Match individual significant words
5. Partial matching: Match meaningful substrings
6. Fuzzy matching: Handle more complex typos by fuzzy matching

Fine-tuning the Database Resolver

The DatabaseResolver can be customized with various parameters to adjust its behavior:

resolver = DatabaseResolver(
    model=Product,
    slug_field='slug',
    use_fuzzy=True,                # Enable/disable last resort fuzzy matching
    fuzzy_cutoff=0.7,              # Similarity threshold (0-1)
    enable_word_matching=True,     # Match individual words
    enable_partial_matching=True,  # Match partial strings
    min_word_length=3,             # Minimum word length to consider
    custom_normalizers={           # Custom normalizer mappings (0 -> o, ph -> f, etc.)
        '0': 'o', 'ph': 'f'
    }
)

You can take a look at more examples in the examples/ directory

License

This project is licensed under the MIT License. See the LICENSE file for details.

Contributing

Contributions are welcome! Please open an issue or submit a pull request on GitHub.