Cross-posting From Letterboxd to Jekyll
I want to meet the movie-lovers where they are, so I’m going to try writing my movie journal on Letterboxd. I have friends there already. Unfortunately, Letterboxd doesn’t have an API to post from my site into it, so I have to post there and pull them in here.
So, similar to what I built for Mastodon, I’m now pulling my Letterboxd reviews into the watching blog category (PESOS). I still want them here for archival/linking/discovery/etc. And this way the other systems of syndicating them to Mastodon and Bluesky can also “just work” once imported.
Workflow
I tacked this on to the GitHub workflow that I run regularly to syndicate my posts into and out of this site. The steps are pretty simple:
- Execute the
utilities/pesos_letterboxdscript - If there are any content changes, commit them and push them up to the repository
The script, written in Ruby, pulls in the public RSS (yes! Bless the public standard of RSS!) feed of my Letterboxd account’s reviews. It then parses the XML to find reviews and imports them as Jekyll posts. For every other post/review, it looks to see if there’s already a Jekyll post on the site that corresponds (has the same Jekyll post frontmatter of letterboxd_review_url). Any that are already present get skipped. Any that doesn’t gets imported as a blog post.
The script is using the little Post class I wrote for syndicating earlier/elsewhere.
Code
require 'jekyll'
require 'rexml/document'
require 'net/http'
require 'time'
require_relative 'models/post'
module PESOS
class Letterboxd
class Error < StandardError; end
attr_reader :site
def initialize
@site = Jekyll::Site.new(Jekyll.configuration({}))
@username = ENV.fetch('LETTERBOXD_USERNAME', 'joshbeckman')
@site.read
end
def import_posts
posts = outbox.filter do |item|
next false if post_exists?(item['link'])
true
end
posts.each { |post| import_post(post) }
end
def post_exists?(url)
site.posts.docs.any? do |post|
post.data['letterboxd_review_url'] == url
end
end
def import_post(post)
title = post['filmTitle']
image = find_image(post['description'])
letterboxd_id = post['link'].split('/').last
slug = "#{post['guid']}-#{letterboxd_id}"
post = Post.new(
body: post['description'],
category: 'blog/watching',
date: Time.parse(post['watchedDate']),
letterboxd_review_url: post['link'],
tmdb_id: post['movieId'],
letterboxd_id: letterboxd_id,
bluesky_status_url: false,
mastodon_social_status_url: false,
image: image,
slug: slug,
tags: %w[letterboxd movies],
title: title,
rating: post['memberRating'].to_i
)
post.create_file
end
def find_image(description)
return nil if description.nil?
image = description.scan(/<img src="([^"]+)"/)
return nil if image.empty?
image.first.first
end
def outbox
uri = URI.parse("https://letterboxd.com/#{@username}/rss/")
req = Net::HTTP::Get.new(uri)
req['Accept'] = 'application/rss+xml'
res = Net::HTTP.start(uri.hostname, uri.port, use_ssl: true) do |http|
http.request(req)
end
raise Error, 'Outbox request failed' unless res.is_a?(Net::HTTPSuccess)
REXML::Document.new(res.body).elements['rss/channel'].elements.map do |item|
next unless item.name == 'item'
item.elements.each_with_object({}) do |element, hash|
hash[element.name] = element.text
end
end.compact
end
end
end
letterbox = PESOS::Letterboxd.new
letterbox.import_posts
Future Work
Things I’d like to add in the future:
- Detect any tags in the review and set them correctly on the Jekyll post
- Figure out a way to determine the IMDB ID for the movie being reviewed (I have a hack in mind for this, but it’s ugly).
Reference
Widgets
Insight Agent
This widget generates “insights” about the post using an agentic loop and MCP server for this site. A response may take up to a minute to generate.