datapoint-python¶
Installation¶
DataPoint for Python can be installed like any other Python module.
It is available on PyPI and the source is available on GitHub.
Pip¶
Pip makes Python package installation simple. For the latest stable version just fire up your terminal and run:
pip install datapoint
or for the very latest code from the repository’s master branch run:
pip install git+git://github.com/ejep/datapoint-python.git@master
and to upgrade it in the future:
pip install git+git://github.com/ejep/datapoint-python.git@master --upgrade
Easy Install¶
Or if you really feel the need then you can use easy_install.
easy_install datapoint
But you probably shouldn’t.
Source¶
You can also install from the source in GitHub.
First checkout the GitHub repository (or you can download the zip and extract it).
git clone https://github.com/ejep/datapoint-python.git datapoint-python
Navigate to that directory
cd datapoint-python
Then run the setup
python setup.py install
Windows¶
- Install python - you can see supported versions in the readme
- Install the appropriate setuptools python extension for your machine
- Install the appropriate pip python extension for your machine
- Add pip to your environment variables:
- Run Start > Edit the environment variables for your account
- Create a new variable:
- name pip
- value the path to pip.exe (this should be something like
C:\Python27\Scripts)
- From the command line run pip install datapoint
Getting started¶
Getting started with DataPoint for Python is simple and you can write a simple script which prints out data in just 6 lines of Python.
API Key¶
To access DataPoint you need to register with the Met Office and get yourself an API key. The process is simple and just ensures that you don’t abuse the service.
Connecting to DataPoint¶
Now that we have an API key we can import the module:
import datapoint
And create a connection to DataPoint:
conn = datapoint.connection(api_key="aaaaaaaa-bbbb-cccc-dddd-eeeeeeeeeeee")
This creates a Manager object which manages our connection and interacts with DataPoint for us, we’ll discuss Manager Objects in depth later but for now you just need to know that it looks after your API key and has a load of methods to return data from DataPoint.
Getting data from DataPoint¶
So now that we have our Manager Object with a connection to DataPoint we can request some data. Our goal is to request some forecast data but first we need to know the site ID for the location we want data for. Luckily the Manager Object has a method to return a Site object, which contains the ID among other things, from a specified latitude and longitude.
We can simply request a Site Object like so:
site = conn.get_nearest_forecast_site(51.500728, -0.124626)
For now we’re just going to use this object to get us our forecast but you’ll find more information about what the Site Object contains later.
Let’s call another of the Manager Object’s methods to give us a Forecast object for our site:
forecast = conn.get_forecast_for_site(site.id, "3hourly")
We’ve given this method two parameters, the site ID for the forecast we want and also a forecast type of “3hourly”. We’ll discuss the forecast types later on.
This Forecast Object which has been returned to us contains lots of information which we will cover in a later section, right now we’re just going to get the Timestep object which represents right this minute:
current_timestep = forecast.now()
This Timestep Object contains many different details about the weather but for now we’ll just print out the weather text.
print current_timestep.weather.text
And there you have it. If you followed all the steps you should have printed out the current weather for your chosen location.
Further Examples¶
For more code examples please have a look in the examples folder in the GitHub project.
Locations¶
There are around 6000 sites available for forecasts, and around 150 sites available for observations.
Forecast sites¶
The locations of the sites available for which forecasts are shown on the map below. Forecasts are restricted to locations within 30 km of the nearest site. This includes all of the United Kingdom. Most of the UK is within 10 km of the nearest site.
The locations of the sites for which forecasts are available. Produced using cartopy.
Objects¶
DataPoint for Python makes use of objects for almost everything. There are 6 different objects which will be returned by the module.
The diagram below shows the main classes used by the library, and how to move between them.
classes
Manager¶
The object which stores your API key and has methods to access the API.
Attributes¶
| attribute | type |
| api_key | string |
| call_response | dict |
| forecast_sites_last_update | float |
| forecast_sites_last_request | list of Site objects |
| forecast_sites_update_time | int |
| observation_sites_last_update | float |
| observation_sites_last_request | list of Site objects |
| observation_sites_update_time | int |
Methods¶
get_nearest_forecast_site(latitude=False, longitude=False)¶
Returns the nearest Site object to the specified coordinates which can provide a forecast.
- param latitude: int or float
- param longitude: int or float
- returns: Site
get_forecast_for_site(site_id, frequency=“daily”)¶
Get a forecast for the provided site. A frequency of “daily” will return two timesteps: “Day” and “Night”. A frequency of “3hourly” will return 8 timesteps: 0, 180, 360 … 1260 (minutes since midnight UTC)
- param site_id: string or unicode
- param frequency: string (“daily” or “3hourly”)
- returns: Forecast
get_observation_sites()¶
Returns a list of sites for which observations are available.
- returns: list of Site objects
get_nearest_observation_site(longitude=False, latitude=False)¶
Returns the nearest Site object to the specified coordinates.
- param longitude: int or float
- param latitude: int or float
- returns: Site
get_observations_for_site(site_id, frequency=’hourly’)¶
Get the observations for the provided site. Only hourly observations are available, and provide the last 24 hours of data.
- param site_id: string or unicode
- param frequency: string (“daily” or “3hourly”)
- returns: Observation
Site¶
An object containing details about a specific forecast site.
Attributes¶
| attribute | type |
| api_key | string |
| name | unicode |
| id | unicode |
| elevation | unicode |
| latitude | unicode |
| longitude | unicode |
| nationalPark | unicode |
| region | unicode |
| unitaryAuthArea | unicode |
Forecast¶
An object with properties of a single forecast and a list of Day objects.
Attributes¶
| attribute | type |
| api_key | string |
| data_date | datetime |
| continent | unicode |
| country | unicode |
| name | unicode |
| longitude | unicode |
| latitude | unicode |
| id | unicode |
| elevation | unicode |
| days | list of Day objects |
Observation¶
An object with the properties of a single observation and a list of Day objects.
Attributes¶
| attribute | type |
| api_key | string |
| data_date | datetime |
| continent | unicode |
| country | unicode |
| name | unicode |
| longitude | unicode |
| latitude | unicode |
| id | unicode |
| elevation | unicode |
| days | list of Day objects |
Day¶
An object with properties of a single day and a list of Timestep objects.
Attributes¶
| attribute | type |
| api_key | string |
| date | datetime |
| timesteps | list of Timestep objects |
Timestep¶
An object with each forecast property (wind, temp, etc) for a specific time, in the form of Element objects.
Attributes¶
| attribute | type |
| api_key | string |
| name | string |
| date | datetime |
| weather | Element |
| temperature | Element |
| feels_like_temperature | Element |
| wind_speed | Element |
| wind_direction | Element |
| wind_gust | Element |
| visibility | Element |
| uv | Element |
| precipitation | Element |
| humidity | Element |