This project is basically for getting information about diseases from Health Organizations - WHO, CDC et al. It is written in Ruby, to contribute, check out the Contribution Guide.
Clone the project repository by running the command below if you use SSH
git clone [email protected]:devcenter-square/disease-info.git
If you use https, use this instead
git clone https://github.com/devcenter-square/disease-info.git
This project uses SQLite3 for development and testing and PostgreSQL for production. To contribute, you only need to install SQLite3. If you already have it installed on your machine, you can skip to the next step.
If you don't have it installed, check the following links for guides on how to install:
- SQLite - Installation by Tutorials Point
- How to install SQLite3 by Mislav Marohnić
cd
into the project directory.- Run
bundle install
- then
rake db:create
- then
rake db:migrate
Populate Database with disease info by running
rake mine_data:who
When this is done, visit the endpoint for diseases on http://localhost:3000/diseases.json to get a list of all diseases on your database.
The present scrapers are:
- WHO, collects data from WHO's Infectious diseases page
Feel free to add another scraper by first adding your rake task to the present list on lib/tasks/mine_data.rake
.
Then create your scraper in app/lib/scrapers
. You can follow the example of the WHO scraper.
If you have to add new columns to the existing Disease structure based on the data you get, feel free to do that.
More importantly, add the source of your data in the more
column present on the Disease object.
Make sure to name space your Disease name with the source. For example Malaria from WHO is saved with name Malaria - WHO.
End Point | Method | Expected response |
---|---|---|
/diseases.json | GET | Gets a list of all diseases |
/diseases.json?data_source=source | GET | Gets a list of diseases from a particular source. Available sources are; WHO |
/diseases/disease_name.json | GET | Gets a particular disease with the supplied name |
This will be the first release and we are tagging it v1.0. To see it live, go to https://disease-info-api.herokuapp.com/diseases. For specific a specific disease, use https://disease-info-api.herokuapp.com/diseases/tuberculosis. Also Disease-info prettifies JSON on production, however if you want JSON prettified on develoment environment and you are using Chrome to access the API you could install JSON Formatter to get a prettier display.
First see the Contribution Guide for how to contribute.
- Add more Health Organizations' disease information sites
- Include a way to verify the scrapped data
- Put the symptoms of each disease in an array