Tembo Website

Goodbye Database Sprawl, Hello Postgres

https://tembo.io

The Tembo website is built using Astro + React and Docusaurus 2, both are modern static website and docs generators.

Installation

Run this from the root of repo:

$ npm install

Local Development

Landing page:

1. Install dependencies first:

cd landing
npm install

2. Start the local dev server from the project root:

npm run landing

Docs:

$ npm run docs

These command(s) start local dev servers and open up browsers on localhost:3000 and localhost:4321 by default. All changes are reflected live without having to restart the server.

Build

Landing:

$ npm run build-landing

Docs:

$ npm run build-docs

This commands will generate static content into the associated build directory and can be served using any static contents hosting service.

Deployment

• main is auto-deployed to https://tembo.io via AWS Amplify
• Open PRs (draft or regular) have preview environments deployed, comment including link will be posted in the PR

Recommended Workflow

• Checkout new branch
• npm run docs or npm run landing
  • commits locally... looking at site in browser
  • push to remote each commit (opened draft PR)
• Ready to go?
  • Review preview environment
  • Squash and merge and get approval from Darren/Samay

Writing a blog post ✍️

Refer to this example post as needed

1. Create a new folder inside /landing/src/content/blog directory

mkdir -p landing/src/content/blog/2024-09-20-example-post

2. Make a new markdown file inside the same directory (this is where you will write your blogpost)

touch landing/src/content/blog/2024-09-20-example-post/index.md

3. Frontmatter

Note: the Tembo blog works with both markdown (.md) and MDX (.mdx) files for posts.

Each blog post supports frontmatter, which is a block of YAML at the top of the file that contains metadata about the post. Here's an example of what frontmatter looks like:

---
slug: sentence-transformers
title: 'Automate vector search in Postgres with any Hugging Face transformer'
authors: [adam]
tags: [postgres, extensions, stacks, vector-search]
image: './tembo_vector.png'
date: 2024-02-02T09:00
description: Walk through using pg_vectorize to automate the vector search workflow in Postgres. Use pg_vectorize transform text to embeddings, and host Sentence Transformers in a container next to Postgres.
---

• slug is the URL of the blog post. It should be a unique identifier for the post and should (ideally) not change after the post is published. Each post will be live at a url like https://tembo.io/blog/your-post-slug.

• title is the title of the blog post. This will be displayed at the top of each post, on the main card list page, and in the title meta tag.

• authors is an array of the authors of the post. Each author should be a string that matches the key of an author in the AUTHORS object inside of here.

• tags is an array of tags that the post is associated with. Each tag will be displayed inside of the left sidebar in each post and will also be used inside of the keywords meta tag. The main tags that we use are postgres. extensions, stacks, engineering, and data.

• image is the path to the image that will be displayed at the top of the post and in the og:image meta tags. This MUST be duplicated in the local blog folder and inside of the public folder for it to work properly. Please contact the Tembo design team for the optimal size + branding for these images.

• date is the date that the post was published. This will be displayed at the top of the post and used inside of the pubDate field in the RSS feed.

• description (optional) is a short description of the post. This will be used inside of the description meta tag.

4. Code blocks

Code blocks can be done as you usually would in typical markdown and will be automatically syntax highlighted + styled

// # Your code goes here :)

5. Adding images

Images can be added to the post by using relative paths to any image files (.jpg, .png, .svg, etc are all supported) that are placed inside of the content/blog/2024-09-20-example-post directory (identical to how our old blog worked). An example post with images can be found here.

6. Admonitions (callouts)

The Tembo blog supports admonitions or callouts.

These are special styled blocks of text that are used to highlight information. To use a callout you must first create an index.mdx file for your post (instead of just plain markdown). There are four different types of callouts: info, tip, warning, and danger. Each callout can have a body with content along with a optional title. Below is an example of what a callout looks like:

Code:

import Callout from '../../../components/Callout.astro'; // be sure to import the callout component first :)

<Callout title='Your title goes here' variant='info'>
  Your body text goes here! (this gets rendered + styled as markdown)
</Callout>

Rendered in the blog post:

7. Embedding tweets and videos

• You can embed a tweet by using the Tweet component and specifying the tweet id that you can find inside it's URL:

import Tweet from '../../../components/Tweet' // the `Tweet` component must be imported first

<Tweet id="1752060288814420168" client:load />

• Videos can be embedded within any blog post using the iframe embed provided by youtube (please make sure that you use the 900x400 dimensions):

 <iframe
    style={{ position: 'absolute', top:'10px', width: '100%', height: '100%' }}
    width="900"
    height="400"
    src="https://www.youtube.com/embed/4vK0JqCNuok?si=60cxZCsfiYEiIiNm"
    title="YouTube video player"
    frameBorder="0"
    allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture"
    allowFullScreen>
  </iframe>

8. RSS feed

The RSS feed for the Tembo Blog gets generated on ever new merge to main and can be found at https://tembo.io/feed.xml.