dejaVu fits the unmet need of being a hackable data browser for Elasticsearch. Existing browsers were either built with a legacy UI and had a lacking user experience or used server side rendering (I am looking at you, Kibana).
So we decided to build our own, with a goal of making it 100% client side and using modern UI elements (no page reloads, infinite scrolling, creating filter views). It's available as a github hosted page, chrome extension and as an Elasticsearch plugin.
bin/plugin install appbaseio/dejaVu
Note: To make sure you enable CORS settings for your ElasticSearch instance, add the following lines in the ES configuration file.
http.port: 9200
http.cors.allow-origin: "http://127.0.0.1:1358"
http.cors.enabled: true
http.cors.allow-headers : X-Requested-With,X-Auth-Token,Content-Type, Content-Length, Authorization
http.cors.allow-credentials: trueAfter installing the plugin, start elasticsearch service
elasticsearchand visit the following URL to access it.
http://127.0.0.1:9200/_plugin/dejaVu Note: If you use Elasticsearch from a different port, the URL would change accordingly.
Sort through the data, find things visually, hide irrelevant data and make sense of all the numbers and dates. Filters work by identifying data mappings from the Elasticsearch index. If dejavu sees a string field, it will provide filters for search, has and has not and is also mindful if the data is analyzed. Similarly a numeric field allows filtering on ranges and a date field allows filtering data by dates.
dejavu also supports local filters like column sorting and showing a subset of columns.
To-do: dejavu doesn't support filtering on advanced data types (like geopoint, parent-child); but it's possible to add those at some point.
It's not uncommon to have thousands of records in a type. dejavu supports an infinite scroll based UI, pagination is so old school.
dejavu also supports browsing data from multiple types and bulk deletions. It also let's you add new records and update existing records.
dejavu uses a websockets based API and subscribes for data changes for the current filtered view. For this to work, the Elasticsearch server needs to support a websockets based publish API. Currently, you can take advantage of this feature by hosting your data with appbase.io.
| Features | dejaVu | ES-head | ES-kopf | ES-browser | Kibana |
|---|---|---|---|---|---|
| Installation | Chrome app, Elasticsearch plugin, static page | Elasticsearch plugin, static page | Elasticsearch plugin, static page | Elasticsearch plugin (doesn't work with v2.0 and above) | Elasticsearch plugin |
| Modern UI | Built with React v0.14.0, uses a live-reload interface. | Built with jQuery v1.6.1, slightly stodgy | Built with Angular 1.x | Built with ExtJs, but a bit stodgy | Built with Node.JS, Hapi, Jade |
| Browser features | CRUD with support for data filters. | Read data with support for full-text search. | No data view | Data view support for a single type | Read view with support for visualizations / charting |
| Open Source | MIT license | Apache v2.0 | MIT license | Apache v2.0 | Apache v2.0 |
Here's a rough roadmap of things to come in the v1.0 release.
- Create a test coverage suite
- Battle-testing with different datasets
- Feature support for advanced filtering
- Offline detection and reconnection for realtime updates
- Performance improvements while scrolling