Migrate references to one s3 bucket to another on Discourse

mozilla.rake

# frozen_string_literal: true

# paste this file into /var/www/discourse/lib/tasks/mozilla.rake
# then run it with: rake mozilla:rebake_posts_with_upload["example.s3.dualstack.us-west-2.amazonaws.com"]

require "db_helper"
require "file_store/s3_store"

task "mozilla:rebake_posts_with_upload", [:s3_host] => :environment do |t, args|
  s3_host = args[:s3_host]

  puts "Rewriting urls in Uploads"
  Upload.where("url LIKE '//%'")
    .update_all("url = REGEXP_REPLACE(url,'^\/\/[^\/]*amazonaws\.com\/','//#{s3_host}/')")

  puts "Rewriting urls in OptimizedImages"
  OptimizedImage.where("url LIKE '//%'")
    .update_all("url = REGEXP_REPLACE(url,'^\/\/[^\/]*amazonaws\.com\/','//#{s3_host}/')")

  Post
    .where("user_id > 0")
    .where("raw LIKE '%.s3%.amazonaws.com/%' OR raw LIKE '%(upload://%'")
    .find_each do |post|
    begin
      updated = false
      post.acting_user = Discourse.system_user

      post.raw.gsub!(/(\/\/[\w.-]+amazonaws\.com\/(original|optimized)\/([a-z0-9]+\/)+\h{40}([\w.-]+)?)/i) do |url|
        begin
          new_url = url.gsub(/\/\/[^\/]*amazonaws\.com\//,"//#{s3_host}/")
          if upload = Upload.find_by(url: new_url)
            updated = true
            url = new_url
          end

          url
        rescue StandardError => e
          puts ""
          puts "XXXXXXXXXXXXXXXX ERROR XXXXXXXXXXXXXXXX"
          puts "post id: #{post.id}"
          puts e
          url
        end
      end

      post.raw.gsub(/(upload:\/\/[0-9a-zA-Z]+\.\w+)/) do |url|
        begin
          post.rebake!
          putc "@"
          url
        rescue StandardError => e
          puts ""
          puts "XXXXXXXXXXXXXXXX ERROR XXXXXXXXXXXXXXXX"
          puts "post id: #{post.id}"
          puts e
          url
        end
      end

      if updated
        post.save!
        post.rebake!
        putc "#"
      else
        putc "."
      end

    rescue StandardError => e
      puts ""
      puts "XXXXXXXXXXXXXXXX ERROR XXXXXXXXXXXXXXXX"
      puts "post id: #{post.id}"
      puts e
    end
  end

  puts "Done!"
end