Angular VizGrimoire dashboard for GitHub data

ng-vizgrimoire-github project is a rewritten from scratch version of VizGrimoireJS dashboard to be used in biterg.io website (not available anymore).

Since our visualization toolkit has move to Kibana, this project is considered deprecated.

Dependencies

Currently the dashboard depends mainly on:

• Angular JS (1.4.x), including angular-route and angular-bootstrap modules
• Angular nvD3
• bootstrap-css-only for baseline styling

Check bower.json file for detailed info.

How to install it

If you want use it as standalone dashboard, clone the repository and:

$ npm install

$ bower install

Dashboard renders data produced by current standard MetricsGrimoire toolkit applied to one GitHub repository. Data (set of json files) must be copied under data folder in root directory.

Note:

Currently, there is a set of data included for testing purposes.

Once there, just:

$ grunt serve

This will let you test and develop over your own dashboard version.

If you want to create a distributable dashboard:

$ grunt build

FIXME: data folder (by now) needs to be copied to dist folder to ensure data is shown

dist folder content could be uploaded to any web server to be available for browsing through internet

#How it works

VizGrimoire Widgets (AKA vgwidets)

Main component for the dashboard are widgets used to represent data. They are defined as Angular directives in the html code:

<vgwidet-{type} datasource={datasource} metrics={metric1,metric2,etc}></vgwidet>

Basically, you define:

• vgwidget-{type}: the type of visualization to be used to represent the data:
  • vgwidget-timeseries: used for timeseries representation
  • vgwidget-trends: used for data trends representation
  • vgwidget-onion: used for onion analysis representation
  • vgwidget-tops: used for top contributors representation
  • vgwidget-demography: used for demographic representation
  • vgwidget-stackedarea: used for aggregated set of timeseries visualization, like activity by domains. Activity by companies, or by repositories, etc. are still under development or pending features.
• widget-datasource: to define where the data to show come from:
• scm: Source code management (data related with commits and committers)
• its: Issue tracking system (data related with issues and issue openers and closers)
• scr: Source code review (data related with pull requests and submitters, reviewers and mergers)
• widget-metrics: a list of metrics to be shown (like commits, domains, authors, etc.).

VGWidgets are placed as blocks in the html structure, and the will fit the available width. Usually vgwidet height is hardcoded fixed.

Some examples

To show a chart of source code authors evolution, current implementation is:

<vgwidet-timeseries widget-datasource="scm" widget-metrics="authors"></vgwidet>

If you would like to see both authors and new authors, that would be:

<vgwidet-timeseries widget-datasource="scm" widget-metrics="authors,newauthors"></vgwidet>

TODO

There are a lot of things to do and that could be improved, so, test it, paly with it and submit pull requests if you find it interesting...