Welcome to this brand new Apisearch newsletter about new features, improvements in the application and next steps. In this newsletter you will find all you need to improve your search layer in your website by using the last updates, some advices and tons of information about the project.
You will find the Apisearch server in this Github repository.
- HEALTH_CHECK_TOKEN environment variable - Pull Request
- Database encryption strategy for sensible fields - Pull Request
- Campaigns plugin - Pull Request
- JWT plugin - Pull Request
- Added new interactions (buy, fav) - Pull Request
You will find the Apisearch UI library in this Github repository.
- Infinite Scroll- Docs - Demo
- Autocomplete - Docs - Demo
- Range filters - Docs - Demo
- Checkbox filter - Docs
You can start using Apisearch as a service in our hosted application. This feature is still only available for partners and resellers, so feel free to contact us - [email protected] - if you want a easy access user.
- Experimental UI access. Make you index use the more experimental UI features. Available for all indices, but only recommended for development ones.
- Autocomplete and Highlights options under the Search Layer configuration
- New Iframe endpoint. You can embed your search engine my using an iframe. You will find all the information inside the Search layer configuration.
- New User usage endpoint. Check your whole usage in the platform.
The Apisearch team is working so hard in a Deep Learning model in order to be able to provide a smart recommender for your website. By using our interactions API, you will send us your user interactions with your items, and by processing all this data layer, we will provide you an amazing API with recommendations for your users.
This layer will also work as a more-like-this processor, so we will return you N items similar to some other items provided by you.
Let's see some possible usages for this feature.
- You can suggest similar items when in the cart page
- You can suggest similar items when in a product page
- You can recommend items to your users similar to their last 10 viewed items
- You can recommend items to your users by having historical interactions and global behavior
- You can add an "Other users bought" slider in your home page
...and many more. This plugin will be the start of this second project phase, mainly focused on how to return better results and how to learn from your users behavior.
This service will be exposed as a private service, automatically integrated with the Apisearch SAAS version.
We're working in a new partnership with a Catalan company that work in Magento projects. As soon as we close this partnership, we will start creating the open sourced Magento plugin for magento, including v1 and v2. In this plugin you will easily add Apisearch in your ecommerce, having access to the amazing admin application and entering a whole new world of search power.
We're working as well in some admin improvements
- Token permissions - Right now, you can define if a Token is an admin one, or a query one. Admin tokens can't be used as query ones, so that would introduce a serious security issue. In this new feature, you will be able to configure permissions per endpoint or group of endpoints.
- Admin translations - You will find the whole Admin platform in english. Soon in more languages.
- Multiple UI layers - Right now, you can attach one single template and UI configuration per index. What if you could have several UI layers for the same index? What if you could build several search engines from the same data?
- Parsing service - Do you want us to parse your data once per day? Well, this feature will be soon available