Getting Started#

yr-weather provides support for multiple products from the MET.no API. These have been created by the Norwegian Meteorological Institute (MET).

Before proceeding, make sure you agree with the current API Terms of Service. Some products, for example Locationforecast have some additional requirements, such as providing a custom User-Agent.

Select the product you want to explore:

Locationforecast#

This product requires a custom User-Agent.

The Locationforecast API is used for getting weather data, such as air temperature, wind speed, and more. You can read more about the API on it’s own documentation.

To get started with Locationforecast using yr-weather, instantiate a Locationforecast object like so:

import yr_weather

# Replace with your own User-Agent. See MET API Terms of Service for correct User-Agents.
headers = {
    "User-Agent": "Your User-Agent"
}

my_client = yr_weather.Locationforecast(headers=headers)

# Get air temperature in Oslo, Norway
oslo_temp = my_client.get_air_temperature(59.91, 10.75)

# Get full weather data for Oslo, Norway
weather_data = my_client.get_forecast(59.91, 10.75)

print(oslo_temp)
# Example output: 8.0

# Get data from a typed dict
print(weather_data["properties"]["timeseries"][0]["data"]["instant"]["details"])
# Example output:
# {
#     "air_pressure_at_sea_level": 1034.7,
#     "air_temperature": -1.5,
#     "air_temperature_percentile_10": -2.5,
#     "air_temperature_percentile_90": -0.9,
#     "cloud_area_fraction": 99.9,
#     "cloud_area_fraction_high": 90.9,
#     "cloud_area_fraction_low": 62.9,
#     "cloud_area_fraction_medium": 95.7,
#     "dew_point_temperature": -4.4,
#     "fog_area_fraction": 0.0,
#     "relative_humidity": 82.1,
#     "ultraviolet_index_clear_sky": 0.0,
#     "wind_from_direction": 11.3,
#     "wind_speed": 0.2,
#     "wind_speed_of_gust": 0.7,
#     "wind_speed_percentile_10": 0.3,
#     "wind_speed_percentile_90": 0.7
# }

Like shown above, you can now use my_client to get full weather data, air temperature, and more.

Next steps:

Textforecast#

Textforecast is used to get text-based forecasts for selected areas in Norway. The forecasts are provided by MET and you can read more about the API on their documentation.

In the example below, a Textforecast class is instantiated and the text forecast for the first location in the list (from the API) is given.

Note

The "landoverview" forecast is only available in Norwegian. For more information, please see the links under the example.

import yr_weather

text_client = yr_weather.Textforecast()

land_overview = text_client.get_forecasts("landoverview")

newest_forecast = land_overview["time"][0]["forecasttype"]

first_location = newest_forecast["location"][0]

print(first_location)
# Example output:
# {
#     'name': 'Østlandet',
#     'id': '0503_0608',
#     'text': 'Skiftende eller sørvest bris, sørvest frisk bris på kysten. Oppholdsvær og noe sol.'
# }

Next steps:

Radar#

The API returns images and GIFs of radar images. yr-weather gives you the freedom to decide what you want to do with the data you receive. For more information about the API itself, check MET’s documentation.

What you see below is an example of getting an animated radar image and saving it in the current working directory.

import yr_weather

radar_client = yr_weather.Radar()

result = radar_client.get_radar("central_norway", "5level_reflectivity", "image")

with open("image.png", "wb") as f:
    for chunk in result:
        f.write(chunk)

Next steps:

Sunrise#

The Sunrise submodule allows you to calculate various events, such as sunrises, sunsets, low moons, high moons and so on. For more information about the API itself, check MET’s documentation.

In the example below, we use the Sunrise.get_detail() function to get data about the sunset.

import yr_weather

sunrise_client = yr_weather.Sunrise()

# Get sunset for Oslo, Norway
sunset = sunrise_client.get_detail("sunset", "2023-01-20", 59.91, 10.75, "+01:00")

print(sunset)
# Example output:
# {
#     'desc': 'LOCAL DIURNAL SUN SET',
#     'time': '2023-01-20T16:00:36+01:00'}
# }

Next steps: