Creating a Travel Planner RESTful API with Flask

Introduction and Lesson Overview

Welcome to the first lesson of our course, Building a CrewAI-Powered Travel Planner App with Flask . In this lesson, we will embark on an exciting journey to create a RESTful API using Flask, a popular web framework for Python. This API will serve as the backbone of our travel planner application, allowing users to submit travel planning requests and receive structured itinerary responses. Our API will integrate seamlessly with the CrewAI travel planner crew, which you have previously learned to work with. By the end of this lesson, you will have a solid understanding of how to set up and run a Flask application that interacts with CrewAI to deliver personalized travel plans.

Setting Up Flask

Flask is a lightweight and flexible web framework that is perfect for building small to medium-sized web applications. It provides the tools needed to create web servers and handle HTTP requests. To get started with Flask on your local device, you would typically install it using pip with the command: Shell pip install Flask pip install Flask However, since you are working in the CodeSignal environment, Flask is already pre-installed, so you can focus on learning and building without worrying about setup.

Structuring the Project

Our project will be organized in a structured manner to support the design and functionality of our API. Here's how we'll organize our files: text app/ ├── main.py # Main application file └── travel_planner/ ├── travel_planner_crew.py # Main travel planner logic ├── models/ │ ├── daily_plan.py # Daily planning model │ ├── travel_itinerary.py # Itinerary model │ └── attraction.py # Attraction model ├── tools/ │ └── custom_search_tool.py # Custom search tool └── config/ ├── tasks.yaml # Task configurations └── agents.yaml # Agent configurations app/ ├── main.py # Main application file └── travel_planner/ ├── travel_planner_crew.py # Main travel planner logic ├── models/ │ ├── daily_plan.py # Daily planning model │ ├── travel_itinerary.py # Itinerary model │ └── attraction.py # Attraction model ├── tools/ │ └── custom_search_tool.py # Custom search tool └── config/ ├── tasks.yaml # Task configurations └── agents.yaml # Agent configurations The main.py file will house the core logic of our Flask application, including initializing the app, defining routes, and handling requests. The travel_planner directory contains all the CrewAI-related components: travel_planner_crew.py implements the main travel planning logic The models directory contains Pydantic models for structuring our data The tools directory includes tools like our custom search tool The config directory stores YAML configuration files for tasks and agents This modular structure ensures our code is organized, maintainable, and scalable as we build more complex features throughout the course.

Join the 1M+ learners on CodeSignal

Be a part of our community of 1M+ users who develop and demonstrate their skills on CodeSignal

Shell

pip install Flask

text

app/
├── main.py                        # Main application file
└── travel_planner/
    ├── travel_planner_crew.py     # Main travel planner logic
    ├── models/
    │   ├── daily_plan.py          # Daily planning model
    │   ├── travel_itinerary.py    # Itinerary model
    │   └── attraction.py          # Attraction model
    ├── tools/
    │   └── custom_search_tool.py  # Custom search tool
    └── config/
        ├── tasks.yaml             # Task configurations
        └── agents.yaml            # Agent configurations

Python

from flask import Flask, request, jsonify
from travel_planner.travel_planner_crew import TravelPlannerCrew

# Initialize Flask application
app = Flask(__name__)

from flask import Flask, request, jsonify
from travel_planner.travel_planner_crew import TravelPlannerCrew

# Initialize Flask application
app = Flask(__name__)

Python

@app.route('/api/plan', methods=['POST'])
def plan_trip():
    # This function handles POST requests to the /api/plan endpoint

@app.route('/api/plan', methods=['POST'])
def plan_trip():
    # This function handles POST requests to the /api/plan endpoint

Python

@app.route('/api/plan', methods=['POST'])
def plan_trip():
    try:
        # Get form data
        city = request.form.get('city')
        days = int(request.form.get('days'))
        attractions_per_day = int(request.form.get('attractions_per_day'))

        # Calculate total attractions needed
        total_attractions = days * attractions_per_day

        # Further processing will be done here...
        
    except Exception as e:
        # Return an error message if there is an exception in the request processing
        return jsonify({"error": str(e)}), 400

@app.route('/api/plan', methods=['POST'])
def plan_trip():
    try:
        # Get form data
        city = request.form.get('city')
        days = int(request.form.get('days'))
        attractions_per_day = int(request.form.get('attractions_per_day'))

        # Calculate total attractions needed
        total_attractions = days * attractions_per_day

        # Further processing will be done here...
        
    except Exception as e:
        # Return an error message if there is an exception in the request processing
        return jsonify({"error": str(e)}), 400

Python

@app.route('/api/plan', methods=['POST'])
def plan_trip():
    try:
        # Get form data and calculate total attractions needed...

        try:
            # Initialize the TravelPlannerCrew instance
            travel_planner = TravelPlannerCrew()
            
            # Run the crew
            result = travel_planner.travel_crew().kickoff(inputs={
                "city": city,
                "days": days,
                "attractions_per_day": attractions_per_day,
                "total_attractions": total_attractions
            })
        except Exception as e:
            # Return an error message if there is an exception during crew execution
            return jsonify({"error": f"Error generating travel plan: {str(e)}"}), 500

        # Further processing will be done here...

    except Exception as e:
        return jsonify({"error": str(e)}), 400

@app.route('/api/plan', methods=['POST'])
def plan_trip():
    try:
        # Get form data and calculate total attractions needed...

        try:
            # Initialize the TravelPlannerCrew instance
            travel_planner = TravelPlannerCrew()
            
            # Run the crew
            result = travel_planner.travel_crew().kickoff(inputs={
                "city": city,
                "days": days,
                "attractions_per_day": attractions_per_day,
                "total_attractions": total_attractions
            })
        except Exception as e:
            # Return an error message if there is an exception during crew execution
            return jsonify({"error": f"Error generating travel plan: {str(e)}"}), 500

        # Further processing will be done here...

    except Exception as e:
        return jsonify({"error": str(e)}), 400

Python

@app.route('/api/plan', methods=['POST'])
def plan_trip():
    try:
        # Get form data and calculate total attractions needed...
        
        # Create the crew and run it...

        # Check if the result has a pydantic model
        if result.pydantic:
            # Return the pydantic output in JSON format
            return jsonify(result.pydantic.model_dump())
        
        # Return an error message if no itinerary is generated
        return jsonify({"error": "No travel itinerary generated"}), 404

    except Exception as e:
        return jsonify({"error": str(e)}), 400

@app.route('/api/plan', methods=['POST'])
def plan_trip():
    try:
        # Get form data and calculate total attractions needed...
        
        # Create the crew and run it...

        # Check if the result has a pydantic model
        if result.pydantic:
            # Return the pydantic output in JSON format
            return jsonify(result.pydantic.model_dump())
        
        # Return an error message if no itinerary is generated
        return jsonify({"error": "No travel itinerary generated"}), 404

    except Exception as e:
        return jsonify({"error": str(e)}), 400

Python

if __name__ == '__main__':
    # Only run the app when this file is executed directly
    app.run(
        host='0.0.0.0',  # Listen on all network interfaces
        port=3000,       # Run on port 3000
        debug=True       # Enable debug mode for development
    )

if __name__ == '__main__':
    # Only run the app when this file is executed directly
    app.run(
        host='0.0.0.0',  # Listen on all network interfaces
        port=3000,       # Run on port 3000
        debug=True       # Enable debug mode for development
    )

text

POST http://localhost:3000/api/plan
Content-Type: application/x-www-form-urlencoded

city=New York&days=2&attractions_per_day=3

POST http://localhost:3000/api/plan
Content-Type: application/x-www-form-urlencoded

city=New York&days=2&attractions_per_day=3

JSON

{
  "city": "New York",
  "daily_plans": [
    {
      "attractions": [
        {
          "address": "Liberty Island, New York, NY 10004",
          "category": "Historical",
          "description": "An iconic symbol of freedom and democracy, the Statue of Liberty is located on Liberty Island.",
          "estimated_duration": "3 hours",
          "name": "Statue of Liberty"
        },
        ...
      ],
      "day_number": 1,
      "meal_suggestions": [
        "Lunch at a local deli near Battery Park (try bagels or a pastrami sandwich)",
        ...
      ]
    },
    ...
  ],
  "days": 2,
  "overall_tips": "Consider using the subway for quick transportation between attractions"
}

{
  "city": "New York",
  "daily_plans": [
    {
      "attractions": [
        {
          "address": "Liberty Island, New York, NY 10004",
          "category": "Historical",
          "description": "An iconic symbol of freedom and democracy, the Statue of Liberty is located on Liberty Island.",
          "estimated_duration": "3 hours",
          "name": "Statue of Liberty"
        },
        ...
      ],
      "day_number": 1,
      "meal_suggestions": [
        "Lunch at a local deli near Battery Park (try bagels or a pastrami sandwich)",
        ...
      ]
    },
    ...
  ],
  "days": 2,
  "overall_tips": "Consider using the subway for quick transportation between attractions"
}

Python

from flask import Flask, request, jsonify
from travel_planner.travel_planner_crew import TravelPlannerCrew

# Initialize Flask application
app = Flask(__name__)

Python

@app.route('/api/plan', methods=['POST'])
def plan_trip():
    # This function handles POST requests to the /api/plan endpoint

Python

@app.route('/api/plan', methods=['POST'])
def plan_trip():
    try:
        # Get form data
        city = request.form.get('city')
        days = int(request.form.get('days'))
        attractions_per_day = int(request.form.get('attractions_per_day'))

        # Calculate total attractions needed
        total_attractions = days * attractions_per_day

        # Further processing will be done here...
        
    except Exception as e:
        # Return an error message if there is an exception in the request processing
        return jsonify({"error": str(e)}), 400

Python

@app.route('/api/plan', methods=['POST'])
def plan_trip():
    try:
        # Get form data and calculate total attractions needed...

        try:
            # Initialize the TravelPlannerCrew instance
            travel_planner = TravelPlannerCrew()
            
            # Run the crew
            result = travel_planner.travel_crew().kickoff(inputs={
                "city": city,
                "days": days,
                "attractions_per_day": attractions_per_day,
                "total_attractions": total_attractions
            })
        except Exception as e:
            # Return an error message if there is an exception during crew execution
            return jsonify({"error": f"Error generating travel plan: {str(e)}"}), 500

        # Further processing will be done here...

    except Exception as e:
        return jsonify({"error": str(e)}), 400

Python

@app.route('/api/plan', methods=['POST'])
def plan_trip():
    try:
        # Get form data and calculate total attractions needed...
        
        # Create the crew and run it...

        # Check if the result has a pydantic model
        if result.pydantic:
            # Return the pydantic output in JSON format
            return jsonify(result.pydantic.model_dump())
        
        # Return an error message if no itinerary is generated
        return jsonify({"error": "No travel itinerary generated"}), 404

    except Exception as e:
        return jsonify({"error": str(e)}), 400

Python

if __name__ == '__main__':
    # Only run the app when this file is executed directly
    app.run(
        host='0.0.0.0',  # Listen on all network interfaces
        port=3000,       # Run on port 3000
        debug=True       # Enable debug mode for development
    )

text

POST http://localhost:3000/api/plan
Content-Type: application/x-www-form-urlencoded

city=New York&days=2&attractions_per_day=3

JSON

{
  "city": "New York",
  "daily_plans": [
    {
      "attractions": [
        {
          "address": "Liberty Island, New York, NY 10004",
          "category": "Historical",
          "description": "An iconic symbol of freedom and democracy, the Statue of Liberty is located on Liberty Island.",
          "estimated_duration": "3 hours",
          "name": "Statue of Liberty"
        },
        ...
      ],
      "day_number": 1,
      "meal_suggestions": [
        "Lunch at a local deli near Battery Park (try bagels or a pastrami sandwich)",
        ...
      ]
    },
    ...
  ],
  "days": 2,
  "overall_tips": "Consider using the subway for quick transportation between attractions"
}

Introduction and Lesson Overview

Setting Up Flask

Structuring the Project

Join the 1M+ learners on CodeSignal

Initializing the Flask Application

Creating the Travel Planning Endpoint

Processing Form Data and Error Handling

Integrating with the CrewAI Travel Planner

Returning the Travel Itinerary as JSON

Running the Flask API

Accessing the Plan Endpoint

Lesson Summary and Next Steps