A Python wrapper around OpenWeatherMap web APIs

PyOWM is a client Python wrapper library for OpenWeatherMap (OWM) web APIs. It allows quick and easy consumption of OWM data from Python applications via a simple object model and in a human-friendly fashion.

PyOWM runs on Python 3.7+

Former Dark Sky API users: you can can use PyOWM to get OpenWeatherMap's OneCall API data as an easy replacement to Dark Sky

With PyOWM you can integrate into your code any of the following OpenWeatherMap web APIs:

• Weather API v2.5 + OneCall API, providing current weather data, weather forecasts, weather history
• Agro API v1.0, providing soil data and satellite imagery search and download
• Air Pollution API v3.0, providing data about CO, O3, NO2 and SO2
• UV Index API v3.0, providing data about Ultraviolet exposition
• Stations API v3.0, allowing to create and manage meteostation and publish local weather measurements
• Weather Alerts API v3.0, allowing to set triggers on weather conditions and areas and poll for spawned alerts
• Image tiles for several map layers provided by OWM

As OpenWeatherMap APIs need a valid API key to allow responses, PyOWM won't work if you don't provide one. This stands for both free and paid (pro) subscription plans. You can signup for a free API key on the OWM website Please notice that the free API subscription plan is subject to requests throttling.

With a free OWM API Key:

from pyowm import OWM
from pyowm.utils import config
from pyowm.utils import timestamps

# ---------- FREE API KEY examples ---------------------

owm = OWM('your free OWM API key')
mgr = owm.weather_manager()


# Search for current weather in London (Great Britain) and get details
observation = mgr.weather_at_place('London,GB')
w = observation.weather

w.detailed_status         # 'clouds'
w.wind()                  # {'speed': 4.6, 'deg': 330}
w.humidity                # 87
w.temperature('celsius')  # {'temp_max': 10.5, 'temp': 9.7, 'temp_min': 9.0}
w.rain                    # {}
w.heat_index              # None
w.clouds                  # 75

# Will it be clear tomorrow at this time in Milan (Italy) ?
forecast = mgr.forecast_at_place('Milan,IT', 'daily')
answer = forecast.will_be_clear_at(timestamps.tomorrow())

# ---------- PAID API KEY example ---------------------

config_dict = config.get_default_config_for_subscription_type('professional')
owm = OWM('your paid OWM API key', config_dict)

# What's the current humidity in Berlin (Germany) ?
one_call_object = mgr.one_call(lat=52.5244, lon=13.4105)
one_call_object.current.humidity

Install with pip for your ease:

$ pip install pyowm

There are alternatives: setuptools, Windows installers and common Linux package managers such as Yaourt (Arch Linux) YaST/Zypper (OpenSuse) (please refer to the documentation for more detail)

Eager to fetch the very latest updates to PyOWM? Install the development trunk (which might be unstable). Eg on Linux:

$ git clone https://github.com/csparpa/pyowm.git
$ cd pyowm && git checkout develop
$ pip install -r requirements.txt && python setup.py install

The library software API documentation is available on Read the Docs.

The Code recipes section comes in handy!

Please read the FAQ section of the documentation before filing a new issue on GitHub!

There are many common issues, therefore a fix for your issue might come easier than you think

Here are some cool projects that use PyOWM

Join the PyOWM public Slack team by signing up here

Contributors (coding, testing, packaging, reporting issues) are welcome! See the the official documentation website for details or the CONTRIBUTING.md file for a quick primer.

MIT license