⚠️ Deprecated: this project moved to a monorepo and is now maintained at https://github.com/datatlas-erasme/datatlas

🌍 DatAtlas

DatAtlas allows you to create a map of the territory on the fly by embedding data from different sources: open sourced data but also crowdsourced data

All infos about the front and the back are stored here

React Front using [kepler.gl]()

Front Setup

more infos here WIP

Using docker

The easiest way to launch datAtlas can be by using docker-compose

1. Clone the repo

   https://github.com/datatlas-erasme/front.git

2. Copy env.example and rename it into .env

3. Put your mapbox token in the .env file

4. Launch DatAtlas

   docker-compose up

Now you can access the server trough http://localhost:5007

Manually

Requirements

• nodejs
• a mapbox api access token

1. Clone the repo

   https://github.com/datatlas-erasme/front.git

2. Copy env.example and rename it into .env

3. Put your mapbox token in the .env file

[WIP TODO]

• Change the backend var URl

4. Install and start npm

   npm install npm start

Back Setup

more infos here WIP

How does datatlas work [WIP]

Datatlas is made of a front and a back

The front is a keplergl surcharged with a styling and theme management [WIP]

The back is here to give the front the styling,dataset configuration and urls He is also able to handle notion tables he converts into a readable keplergl json format (and soon other unconventionnal data sources / unconventionnal => no geojson )

The front needs the back server to know :

• wich data layers he needs to load
• wich name he has (ex: Datatlas-Industries)
• wich optionnal button he has (ex: crowdsourcing buttons to add a new point on the map)
• how are the layers and filters configured by default

two API routes are used for it :

GET api/

more infos on routes here

Production

DatAtlas is not ready for production yet

Configuration

See InstanceConfigurationInterface.ts.

Search

A search field shows up when you configure a list of searchfilters :

{
  "searchFilters": ["s66961zse"]
}

Where s66961zse is a filter id.

The search query look for matches in the specified filter domains via a fuzzy search and sort them using a scoring algorithm.

Warning : Specifying more than on filter id may lead to unexpected results.