Exploring OramaSearch with Michele Riva - Indexing and searching blog posts
In this session, Roberto and Luciano are joined by Michele Riva (co-founder of OramaSearch) to discuss how to leverage OramaSearch for indexing and searching through a vast collection of 70+ blog posts. In this down-to-earth discussion, they explore the technical capabilities of OramaSearch, focusing on its efficiency, implementation, and optimization for edge networks.
We then tackle the challenge of scripting data ingestion from a series of blog posts in markdown format to the OramaSearch API for indexing. We discuss what the API looks like, how to obtain an API key, structuring JSON payloads, and optimizing HTTP requests, all while troubleshooting issues like ensuring unique document IDs and streamlining the process for maximum efficiency.
We also showcase how to use the OramaSearch client library to perform a search over the existing indices deployed on the edge. In this phase, Michele explains various technical details and nuances on how query parameters work, indexing thresholds, and search result sorting algorithms.
Finally, we discuss the pricing of OramaSearch Cloud, reviewing the free tier and the other pricing plans.
📖 Chapters
00:00:00 Intro
00:00:34 Intro to Michele Riva and OramaSearch
00:01:45 The plan for today: indexing blog posts from a static website
00:02:32 Differences between OramaSearch open source and OramaSearch Cloud
00:05:36 Signing up for an OramaSearch account
00:06:20 Getting started with OramaSearch Cloud
00:09:30 Creating a new index and using the webhook integration
00:10:03 Defining the schema for our index
00:17:01 Understanding the OramaSearch Cloud Webhook API
00:20:12 Getting an API Key and storing it in an environment variable
00:21:42 Creating a script to parse all the markdown files and extract the relevant metadata
00:29:01 Making the first request to the Webhook API
00:37:25 Ideas on how to productionise this script and call to the index deployment endpoint
00:43:02 Searching over our newly deployed index
00:57:50 How to highlight matches from searches
01:00:32 OramaSearch Cloud plans and pricing
01:03:37 Wrapping up
Check out OramaSearch at https://oramasearch.com/
If you want to connect with Michele Riva, you can find all his contacts at https://www.micheleriva.it/
The code examples are available here: https://github.com/lmammino/loige.co/tree/main/scripts
To follow the next streams live and interact with us, you can check us out on Twitch at https://twitch.tv/loige