Factmint Charts allows you to create interactive data visualizations, which can be rendered from an HTML table or directly from JSON. For more information and live examples, see http://factmint.com/charts-overview/. This repo includes the charts as well as the API that you can use to create your own (see the /api
directory).
-
Download:
bower install factmint-charts
or manually download raw CSS and JS from the dist folder -
Create a table with a classname of
fm-[visualisation name]
and include the CSS and JS in your document.
e.g.
<table class="fm-pie">
<thead>
<tr>
<th>Product</th>
<th>Units sold</th>
</tr>
</thead>
<tbody>
<tr>
<td>Ice cream</td>
<td>1500</td>
</tr>
<tr>
<td>Umbrellas</td>
<td>6500</td>
</tr>
<tr>
<td>Horse shoes</td>
<td>4200</td>
</tr>
</tbody>
</table>
<link rel="stylesheet" href="pie.min.css">
<script async src="pie.min.js"></script>
For the two choropleths and the pie chart, you also have the option of using a Javascript object/JSON rather than an HTML table:
index.html
<html>
<head>
<title>Pie chart</title>
<link rel="stylesheet" href="../../../../dist/choropleth-uk-constituencies.min.css">
<script defer src="../../../../dist/choropleth-uk-constituencies.min.js"></script>
<script defer src="main.js"></script>
</head>
main.js
var data = [
{
"title": "Rowing",
"value": 27
},
{
"title": "Kayaking",
"value": 18
},
{
"title": "Cycling",
"details": "Lorem ipsum",
"value": 50
}
];
window.factmint.pie(data);
First, clone this repository with git clone https://github.com/factmint/Charts.git
. Then run npm install
from the project root.
To see running examples, navigate to the directory of a given chart in the terminal and (if it's the first time) npm install
and grunt install
. Finally, run grunt serve
from the project root (not the chart directory).
e.g.
cd choropleth-uk-constituencies
npm install && grunt install
cd ../../
grunt serve
This will start a web server. You should see a message telling you which port it is running on (e.g. "Started connect web server on http://0.0.0.0:15002"). You can now navigate to http://locahost:15002 (the port number will be different if 15002 was not open when you ran grunt serve
, so be sure to check), where you will see a directory listing. If you look in the examples/unbuilt/
directory, you will find example use cases.
If you run grunt build
from the chart's root directory, it will create or update three files in the /dist/
directory: a minified standalone script, a minified CSS file, and a text file containing all of the available options for the given chart. You can see examples using the built script under examples/built/
in the chart's directory.
To see unit tests for the API, navigate to api/test
in your browser.
(See http://factmint.com/documentation/bubble-chart/)
(See http://factmint.com/documentation/candlestick-chart/)
(See http://factmint.com/documentation/choropleth/)
(See http://factmint.com/documentation/world-countries-choropleth/)
(See http://factmint.com/documentation/column-bar-chart/)
(See http://factmint.com/documentation/doughnut-chart/)
(See http://factmint.com/documentation/line-chart/)
(See http://factmint.com/documentation/line-over-column-chart/)
(See http://factmint.com/documentation/pictorial-bar-chart/)
(See http://factmint.com/documentation/pie-chart/)
(See http://factmint.com/documentation/scatter-graph/)
(See http://factmint.com/documentation/stacked-area-chart/)
(See http://factmint.com/documentation/stacked-column-bar-chart/)
The /api
directory contains the API, which can be used to build your own chart. See https://github.com/factmint/Charts/blob/master/api/README.md for more information.