Comparing Weather Data API vs Web Scraping

Introduction

When building apps, there are usually two ways to get data from a service:
1. Use the official API provided by the service.
2. Scrape the data directly from the website.

Both approaches work, but they come with different trade‑offs. Let’s look at a simple example: pulling weather data for Jakarta using OpenWeatherMap. We’ll compare what happens when we use the official API versus scraping the HTML page with Selenium.

Getting Data from the API

The API endpoint looks like this:

https://api.openweathermap.org/data/2.5/weather?q=Jakarta&appid=API_KEY&units=metric

The response is structured JSON. Here’s a simplified version:

{
  "name": "Jakarta",
  "weather": [{"description": "overcast clouds"}],
  "main": {"temp": 26.37, "temp_min": 24.94, "temp_max": 26.72, "pressure": 1009, "humidity": 70},
  "wind": {"speed": 5.14}
}

API summary:
- City: Jakarta
- Weather: overcast clouds
- Temperature: 26.37 °C
- Range: 24.94 – 26.72 °C
- Pressure: 1009 hPa
- Humidity: 70%
- Wind: 5.14 m/s

Working with APIs? You may also enjoy Backend API Testing with Laravel for deeper insights into validating API reliability.

Getting Data by Scraping HTML

The website page is:

https://openweathermap.org/find?q=jakarta

With Selenium, we can grab the table row that shows the weather. The output looks like this:

{'source': 'Scraping (HTML via Selenium)', 'city': 'Jakarta, ID', 'weather': 'overcast clouds', 'temp': '26.4°С'}

Scraping summary:
- City: Jakarta, ID
- Weather: overcast clouds
- Temperature: 26.4 °C

Side‑by‑Side Comparison

Conclusion

• The API is the best choice for real applications: it’s stable, fast, and gives complete data.
• Scraping HTML is useful for learning, experimenting, or showing off portfolio skills.
• In Jakarta’s case, both methods give similar results, but the API provides more detail behind the scenes.