This application scrapes the metservice.com website every half an hour for the current temperature with and without windchill, before storing it in an SQL database.
To run this you will need python2.7, sqlite3 and scrapy:
http://doc.scrapy.org/en/latest/intro/install.html#intro-install
Run.py is executed, it uses multiprocessing to start a process which sleeps for half an hour at a time before executing the scrapy shell with a spider targeting the metservice website. This is to save on processing power but also allow user input to quit the application or see how much time is left until the next scrape.
Much more information can be gathered, just modify the wget_spider.py script and the weather_data.db to store the required data. If you want it to point to another location you will have to look for it using your web browsers developer tools while loading the correct metserivce page (it will be named similiarly to metservice.com/publicdata/localObs_porirua). If you're into web scraping at all this won't be hard for you.
The scrapy script recieves a json module in response to GET query and just parses it for the temperature data, and then enters it into the sql data base called weather_data.db.
It's very lightweight, designed to run on a small webserver so that the data from the sql database could be displayed dynamically via a webpage.