utilities/snippets_cli_go#1: Project Overview

Background

snippets.bentasker.co.uk is currently created with my home-grown PHP static site generator.

However, it's now years old and I'd like to replace it with something more modern.

My intention is to migrate to Nikola just like I did with my recipes site.

But, most of my snippets usage happens via a custom cli:

sbt_cli foo

That CLI relies on fetching the JSON sitemap that my SSG generates. No other SSG generates one of these. Whilst I could create a Nikola plugin to generate a compatible file, it would seem to make more sense to create a new utility which consumes a RSS feed.

The old CLI was in Python, but I've elected to write this in Go - there's no other reason for that than that I fancied doing it in Go.

Featureset

The Python CLI has a number of features, not all of which are intended for the initial Go implementation

• [x] search: My primary mode of interaction is searching
• [x] [number]: Output snippet with id (see note below)
• [x] autodisplay: if a search results in a single snippet, output that snippet
• [ ] list: list all snippets in the system (not implemented because I've never used it)
• [ ] local caching: probably will be implemented in future, but not a high priority
• [ ] analytics ping: not needed. I never actually updated the embedded URL when I stopped using piwik

Snippet IDs

The old implementation leans quite heavily on IDs - each snippet is given an ID at generation time.

Whilst we could (and probably will, come import time) expand the front-matter fed into Nikola to include these they'll still not be available via the RSS feed.

So, if we want to use IDs as a means of specifying a snippet they'll need to be assigned deterministically by the CLI

Snippet Output

The old CLI retrieves a JSON object describing the snippet and then constructs output section by section.

We won't be able to do that - we'll have to process HTML and reformat into something CLI friendly.

We can get that HTML from one of two sources

• RSS feed: requires the feed to be set to contain full content, increasing its size
• Fetching the page itself: means an extra request but keeps the RSS feed small (and reduces it's utility to sites that steal content by rendering RSS feeds)