Skip to content

Latest commit

 

History

History
123 lines (80 loc) · 2.89 KB

INSTALL.md

File metadata and controls

123 lines (80 loc) · 2.89 KB
Raw

Running Open Data Maker on your computer

If you just want to install and run, then you can just download a zip file.

You will still need the the dependencies below, but you don't need to clone the git repo for the source code.

Install Prerequisites

You can run our bootstrap script to make sure you have all the dependencies. It will also install and start up Elasticsearch:

script/bootstrap

To run Open Data Maker, you will need to have the following software installed on your computer:

**NOTE: Open Data Maker indexing currently is very slow on ES2.x, however, an index created on 1.x can be restored to 2.x.

Mac OS X

On a Mac, we recommend installing RVM.

If you are contributing to development, you will also need Git.

Get the Source Code

For development, fork the repo first, then clone your fork.

git clone https://github.com/<your GitHub username>/open-data-maker.git
cd open-data-maker

Run the App

Import the data

To get started, you can import sample data with:

rake import

Start the app

padrino start

Go to: http://127.0.0.1:3000/

and you should see the text Welcome to Open Data Maker with a link to the API created by the sample data.

You can verify that the import was successful by visiting http://127.0.0.1:3000/v1/cities?name=Cleveland. You should see something like:

{
  "state": "OH",
  "name": "Cleveland",
  "population": 396815,
  "land_area": 77.697,
  "location": {
    "lat": 41.478138,
    "lon": -81.679486
  }

Custom Datasets

While the app is running (or anytime) you can run rake import. For instance, if you had a presidents/data.yaml file, you would import it with:

export DATA_PATH=presidents
rake import
# or, more succintly:
DATA_PATH=presidents rake import

to clear the data, assuming the data set had an index named "president-data"

rake es:delete[president-data]

you may alternately delete all the indices (which could affect other apps if they are using your local Elasticsearch)

rake es:delete[_all]

The data directory can optionally include a file called data.yaml (see the sample one for its schema) that references one or more .csv files and specifies data types, field name mapping, and other support data.

Debugging

ES_DEBUG environment variable will turn on verbose tracer in the Elasticsearch client

optional performance profiling for rake import: rake import[profile=true]

Want to help?

See Contribution Guide