In December 2015 we released DocSearch, an easy way to make your software or API documentation searchable.
With just a little configuration, DocSearch can automatically crawl most documentation websites and then provide a nice autocomplete search box to add to documentation pages.
Today we have some exciting news for the DocSearch project:
- An improved design is here with the release of DocSearch.js v2
- We are open sourcing the full DocSearch crawling code!
A better default design
With several iterations, we upgraded the DocSearch default design incorporating all the feedback we received from the community. The new design is available today with the release of DocSearch v2.
Here’s what it looks like on the Middeman website:
From 0 to 75k searches a day
DocSearch is now powering 150 documentation searches. Most of them are in English, but we also have sites in Turkish, Chinese, French, Japanese, etc.
In total we’ve indexed around 1.5M records and are performing 75k searches per day and growing. An interesting trend we’ve noticed is that most of these searches are made during weekdays. Who ever said developers worked on the weekend?
The last bar, starts on the 26th of June, that’s why it’s smaller.
The DocSearch open source stack
DocSearch is composed of three different projects. as of today they will all be open source :
- The front end of DocSearch (which was already open source): https://github.com/algolia/docsearch
- The scraper which browses & indexes web pages: https://github.com/algolia/docsearch-scraper
- The configurations for the scraper: https://github.com/algolia/docsearch-configs
Want to try DocSearch on your documentation? Just follow these steps: https://github.com/algolia/docsearch-scraper
The scraper is a collection of submodules, each in its own directory:
- cli: A command line tool to manage DocSearch. Run `./docsearch` and follow the steps
- deployer: Tool used by Algolia to deploy the configuration in our Mesos infrastructure
- doctor: A monitoring & repairing tool to check if the indices built by the scraper are in good shape
- playground: An HTML page to easily test DocSearch indices
- scraper: The core of the scraper. It reads the configuration file, fetches the web pages and indexes them with Algolia.
Future of DocSearch
We want DocSearch to be easy to integrate and customize for everyone. To do that we are building a visual configuration tool that will help you auto-generate the DocSearch configuration and the styling of the autocomplete menu.
Request an awesome documentation search
We’re always looking for websites that are missing a great documentation search and are happy to help create it. We’ve seen DocSearch improve the search experience of many types of documentation and we think it could improve yours too.