Git Product home page Git Product logo

discourse-algolia's Introduction

discourse-algolia [beta]

Power the search on your Discourse with Algolia ๐Ÿ”Ž๐ŸŽ‰

The discourse-algolia plugin indexes users, posts, topics and tags with Algolia and adds a search box built with autocomplete.js to the Discourse header. Try it live on Algolia's Community Forum:

discourse-algolia autocomplete

Features

The discourse-algolia plugin uses the Algolia API, which provides powerful indexing and search capabilities.

  • Speed: Results update immediately with each keystroke
  • Relevance: Metrics like views and likes are used to surface the best content first
  • User experience: All indicies are searched simultaneously and presented in a unified view
  • Analytics: See what users are searching for, see top queries that produce 0 results

Prerequisites

You'll need an Algolia account if you don't have one already.

  • Go here to sign up for a free account (10,000 records and 100,000 indexing operations monthly)
  • If you're using Discourse for an open source project and need higher limits, fill out the form on the Algolia for Open Source page.
  • If you need higher limits for your business, we're happy to help you find the right plan.

Beta testers wanted

Please see this thread on meta.discourse.org for more information: Add Algolia search to your Discourse.

Installation

Add this repository's git clone url to your container's app.yml file, at the bottom of the cmd section:

hooks:
  after_code:
    - exec:
        cd: $home/plugins
        cmd:
          - mkdir -p plugins
          - git clone https://github.com/discourse/discourse-algolia.git

Rebuild your container:

cd /var/discourse
git pull
./launcher rebuild app

Configuration

Once you've installed the plugin and restarted your Discourse, you will see a new plugin available in your admin configuration. Click the Settings button next to the discourse-algolia plugin.

Discourse admin plugins list

You will now see all of the plugin configuration options available.

discourse-algolia configuration options

Populate the settings as follows:

  • algolia enabled: Check this box to enable indexing new content with Algolia.
  • algolia autocomplete enabled: Check this box to replace the default Discourse autocomplete with the Algolia autocomplete. It is recommended to do this only once you have all content indexed (see below).
  • algolia application id: The ID of an Algolia application you have created.
  • algolia search api key: A search-only API key of the Algolia application. Do not use an admin API key, as this will be visible to the clients of your Discourse.
  • algolia admin api key: The admin API key of your Discourse application, or any API key that can write and configure indices.
  • algolia discourse username: If content is visible to this Discourse user, it will be indexed. If not, it will be skipped. This defaults to system, which is an admin account and can see content in all categories, including Staff. (Private messages, however, are always excluded from indexing.) It is recommended that you create a dummy separate user who can only see content you consider to be public, and change this value to their username.

You can find the algolia application id, algolia search api key, and algolia admin api key in the API Keys page of your Algolia dashboard.

Algolia dashboard API Keys page

Once all of the settings are turned on and populated, the plugin configuration should look like this:

discourse-algolia populated configuration options

Initial indexing

Once you have enabled the Algolia plugin and added an application ID and admin API key, you can now index all of your forum's content. Run the following rake task in your Discourse directory:

LOAD_PLUGINS=1 bundle exec rails algolia:initialize

This will create and configure three indices - discourse-users, discourse-posts, and discourse-tags - and then populate them by loading data from your database and sending it to Algolia. The data will be searchable as soon as the task is finished. You can now enable the algolia autocomplete enabled setting of the plugin, reload the page of your Discourse, and try the autocomplete search.

Rake tasks

To run rake tasks, first cd into your Discourse directory. If you're using the Docker install, just run this command:

./launcher enter app

Additional rake tasks are provided for configuring indices and syncing data.

rake algolia:configure                                                 # configure algolia index settings
rake algolia:initialize                                                # configure indices and upload data
rake algolia:reindex                                                   # reindex everything to algolia
rake algolia:reindex_posts                                             # reindex posts in algolia
rake algolia:reindex_tags                                              # reindex tags in algolia
rake algolia:reindex_users                                             # reindex users in algolia

Styling with CSS

To see how you can change the look and feel of the autocomplete, see the CSS styles in the plugin's assets/stylesheets directory.

 .algolia-holder .aa-input {
    color: #919191;
    font-family: Open Sans, Arial, sans-serif;
    border: solid 1px rgba(137,149,199,0.2);
    border-radius: 6px;
    font-size: 13px;
    line-height: 13px;
    padding: 8px 10px 8px 35px;
    height: auto;
    width: 250px;
    outline: 0;
    background-image: url("data:image/svg+xml;utf8,<svg width='14' height='14' viewBox='0 0 14 14' xmlns='http://www.w3.org/2000/svg'><path d='M12.5 11h-.79l-.28-.27C12.41 9.59 13 8.11 13 6.5 13 2.91 10.09 0 6.5 0S0 2.91 0 6.5 2.91 13 6.5 13c1.61 0 3.09-.59 4.23-1.57l.27.28v.79l5 4.99L17.49 16l-4.99-5zm-6 0C4.01 11 2 8.99 2 6.5S4.01 2 6.5 2 11 4.01 11 6.5 8.99 11 6.5 11z' fill='%237C8A94' fill-rule='evenodd'/></svg>");
    background-size: 14px 14px;
    background-repeat: no-repeat;
    background-position: 10px center;
  }

You'll see style definitions like this, which you can override by creating a fork of the plugin or more easily by using the CSS/HTML customization that Discourse makes available in the Admin tool.

Customizing ranking

Once you've indexed your data with Algolia, you can see it and search it from the Algolia dashboard. That will give you an idea of what attributes the plugin indexes. You can also make changes to relevance and other index configuration settings. A full treatment of ranking, relevance and the Algolia dashboard is out of scope here, but we encourage you to head over to the Algolia documentation and start digging in.

algolia dashboard indices tab

Support

Have a question? Ran into an issue? If it's code-related, please file an issue in this repository and include as much information as possible. If it's a general question, please post it on the Algolia Community Forum and add the tag #discourse-algolia. We look forward to hearing from you!

discourse-algolia's People

Contributors

joshed-io avatar dzello avatar jjaffeux avatar samsaffron avatar majakomel avatar nbianca avatar danielmai avatar jomaxro avatar justindirose avatar xrav3nz avatar xfalcox avatar

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    ๐Ÿ–– Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. ๐Ÿ“Š๐Ÿ“ˆ๐ŸŽ‰

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google โค๏ธ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.