sutty-migration 0.2.0 → 0.2.1

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 046cf945de1c0736e329224151a4b331c87e44cab6e6c2bc1e69f22b7639fe68
-  data.tar.gz: 58effbee202ab51c7ff1ed4e8e98498cd5a65f618c6ae6fc4f924fe6aed2e0e1
+  metadata.gz: bbc62b40240f7bdea15ea0994e1af5db0e7590a500459ca82e4727f86bb791a5
+  data.tar.gz: '087218a58be167e9348d4314922a8985a56339b9c6dd340edbfa21be1aabc9ed'
 SHA512:
-  metadata.gz: a21cb549bddd9bc55218c0633932300811e547d2d4cfde2525fbcdcaf8a2ff3bb89f0542813b46fdb9b90c9d739166ce37e303fb4de76b918b52bee9402fcb6d
-  data.tar.gz: 544f18359b4e9996c07f6828643bb8d4856d55457b4a6134b2ef2cedf01d4471d294effc6024598ef66f9b1d2258345f44ac370e1678166a2b9d19b5c4a55e74
+  metadata.gz: 5a6933d3c03a86aa0b469f75afa42c969bc22301672054cba60eb5256b87a4381c325e7d49fd9af1259cef0d44094c5fe78df2aec51f859446b34625b24921a5
+  data.tar.gz: 3ce853bddcf43a8c993ef8c8f1e9601fb4da59bdf7eb58a96611baef8e6934375b4bd20e1cb67a008025e03b0c94d62772ab18f4d23550d063d6d381a1190bdd

data/lib/sutty_migration/wordpress.rb

@@ -38,15 +38,19 @@ module SuttyMigration
     #
     # @return [Hash] { "ID" => SuttyMigration::Wordpress }
     def blogs
-      @blogs ||= wp["select blog_id as id, domain, path from #{prefix}blogs"].to_a.map do |blog|
+      @blogs ||= wp["select * from #{prefix}blogs"].to_a.map do |blog|
         url   = "https://#{blog[:domain]}#{blog[:path]}"
-        pfx   = "#{prefix}#{blog[:id]}_" if blog[:id] > 1
+        pfx   = "#{prefix}#{blog[:blog_id]}_" if blog[:blog_id] > 1
         pfx ||= prefix
 
-        [ blog[:id], self.class.new(site: site, url: url, prefix: pfx, database: database, limit: limit, multisite: self) ]
+        [ blog[:blog_id], blog.merge(db: self.class.new(site: site, url: url, prefix: pfx, database: database, limit: limit, multisite: self)) ]
       end.to_h
     end
 
+    def options
+      @options ||= wp["select option_name, option_value from #{prefix}options"].to_a.map(&:values).to_h.transform_keys(&:to_sym)
+    end
+
     # Open the database.
     #
     # @return [Sequel::SQLite::Database]

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: sutty-migration
 version: !ruby/object:Gem::Version
-  version: 0.2.0
+  version: 0.2.1
 platform: ruby
 authors:
 - f
@@ -138,7 +138,6 @@ files:
 - lib/sutty_migration/data.rb
 - lib/sutty_migration/jekyll/document_creator.rb
 - lib/sutty_migration/wordpress.rb
-- lib/wordpress.rb
 homepage: https://0xacab.org/sutty/jekyll/sutty-migration
 licenses:
 - GPL-3.0

data/lib/wordpress.rb

@@ -1,192 +0,0 @@
-# frozen_string_literal: true
-
-# Generar UUIDs
-require 'securerandom'
-# Traer resultados de la base de datos
-require 'sequel'
-require 'sqlite3'
-require 'json'
-# Descargar archivos
-require 'faraday'
-require 'progressbar'
-
-class Wordpress
-  attr_reader :site, :prefix, :limit, :url
-
-  def initialize(site:, url:, prefix: 'wp_', limit: 10)
-    @site = site
-    @prefix = prefix.freeze
-    @limit = limit.freeze
-    @url = url.freeze
-
-    # Conectarse a la base de datos
-    @wp = Sequel.sqlite(File.join(site.source, '_data', 'wordpress', 'post.sqlite3'))
-    # Las funciones de JSON usan mucha CPU, vamos a traer de a pocos
-    # registros.
-    @wp.extension :pagination
-  end
-
-  def download(file)
-    dest = "wp-content/uploads/#{file}"
-    full = File.join(site.source, dest)
-
-    return dest if File.exist? full
-
-    Jekyll.logger.info "Downloading #{dest}"
-
-    FileUtils.mkdir_p File.dirname(full)
-
-    File.open(full, 'w') do |f|
-      url = "#{url}/#{dest}"
-      head = Faraday.head(url)
-      content_length = head.headers['content-length']
-      progress_bar = ProgressBar.new
-
-      Faraday.get(url) do |req|
-        req.options.on_data = Proc.new do |chunk, downloaded_bytes|
-          f.write chunk
-        end
-      end
-    end
-
-    dest
-  end
-
-  # Obtiene todos los tipos de artículos disponibles
-  #
-  # @return [Array]
-  def layouts
-    @layouts ||= @wp["select distinct post_type from #{prefix}posts"].to_a.map(&:values).flatten
-  end
-
-  # Obtiene todos los posts opcionalmente filtrando por tipo de post.
-  # No es la forma oficial de Sequel pero no tenemos tiempo de
-  # aprenderla específicamente y además tenemos las opciones en formato
-  # JSON que no estarían soportadas.
-  #
-  # @param :layout [String] Layout name, one of #layouts
-  # @param :with_meta [Boolean] 
-  # @return [Enumerator]
-  def posts(**options)
-    if options[:layout] && !layouts.include?(options[:layout])
-      raise ArgumentError, "#{layout} must be one of #{layouts.join(', ')}"
-    end
-
-    @posts ||= {}
-    @posts[options[:layout] || 'all'] ||= @wp[post_query(**options)].each_page(limit).to_a.map(&:to_a).flatten.tap do |p|
-      next unless options[:with_meta]
-
-      p.map do |post|
-        post[:front_matter] = JSON.parse(post[:front_matter]) unless post[:front_matter].nil?
-        post[:terms] = JSON.parse(post[:terms]) unless post[:terms].nil?
-      end
-    end
-  end
-
-  private
-
-  # Consulta para los posts, incluyendo metadatos en JSON.  Los
-  # metadatos vienen en dos partes porque tienen dos 
-  #
-  # @return [String]
-  def post_query(layout: nil, with_meta: true)
-    @post_query ||= <<~EOQ
-      select
-        p.ID as id,
-        p.post_title as title,
-        p.post_name as slug,
-        p.post_type as layout,
-        p.strftime('%Y-%m-%d', post_date) as date,
-        p.post_status as status,
-        p.post_content as content
-        #{", json_group_object(f.meta_key, f.meta_value) as front_matter" if with_meta}
-        #{", t.meta as meta" if with_meta}
-      from #{prefix}posts as p
-      left join #{prefix}postmeta as f on p.ID = f.post_id
-      #{"left join (#{meta_query(layout: layout)}) as as t on t.id = p.ID" if with_meta}
-      #{"where p.post_type = :layout" if layout}
-      group by p.ID
-    EOQ
-  end
-
-  # 
-  def meta_query(layout: nil)
-    @meta_query ||= <<~EOQ
-      select
-        p.ID as id,
-        json_group_object(tt.taxonomy, t.name) as meta
-      from #{prefix}posts as p
-      left join #{prefix}term_relationships as r on r.object_id = p.ID
-      left join #{prefix}term_taxonomy as tt on tt.term_taxonomy_id = r.term_taxonomy_id
-      left join #{prefix}terms as t on t.term_id = tt.term_id
-      #{"where p.post_type = :layout" if layout}
-      group by p.ID
-    EOQ
-  end
-end
-
-# Antes de generar el sitio vamos a leer todos los artículos desde la
-# base de datos y generarlos localmente.
-Jekyll::Hooks.register :site, :post_read do |site|
-  wp = Wordpress.new(site: site,
-                     url: site.config.dig('wordpress', 'url'),
-                     prefix: site.config.dig('wordpress', 'prefix'))
-
-  collection = site.collections['posts']
-  ascii_re = Regexp.new("\P{ASCII}").freeze
-  sanitizer = Rails::Html::SafeListSanitizer.new
-
-  # Traer todas las imágenes cargadas y descargarlas
-  attachments = wp.posts(layout: 'attachment').map do |page|
-    page.map do |attachment|
-      attachment[:data] = JSON.parse(attachment[:data]) unless attachment[:data].nil?
-      file = attachment.dig(:data, '_wp_attached_file')
-
-      next unless file
-
-      dest = wp.download(file)
-
-      # Tener un mapa de IDs y archivos destino
-      [ attachment[:id], dest ]
-    end
-  end.compact.flatten(1).to_h
-
-  %w[post page].each do |type|
-    wp.posts(layout: type).each do |page|
-      page.each do |post|
-        # Convertir los datos extra en un Hash
-        post[:data] = JSON.parse(post[:data]) unless post[:data].nil?
-        post[:slug] = Jekyll::Utils.slugify(post[:title], mode: 'latin') if post[:slug].empty?
-        post[:meta] = wp.meta id: post[:id]
-
-        path = File.join(site.source, '_posts', post.slice(:date, :slug).values.join('-') + '.markdown')
-
-        if File.exist? path
-          Jekyll.logger.info "#{path} ya fue migrado, actualizando"
-
-          doc = site.documents.find do |d|
-            d['id'] == post[:id]
-          end
-        else
-          # Crear un post nuevo y agregarlo a la colección
-          collection.docs << doc = Jekyll::Document.new(path, site: site, collection: collection)
-          doc.data['uuid'] = SecureRandom.uuid
-        end
-
-        thumbnail = post.dig(:data, '_thumbnail_id')&.to_i
-
-        doc.data['layout'] = type
-        doc.data['title'] = post[:title]
-        doc.data['draft'] = post[:status] != 'publish'
-        doc.data['id'] = post[:id]
-        doc.data['date'] = Jekyll::Utils.parse_date(post[:date])
-        doc.data['tags'] = post[:meta].select { |k| k[:type] == 'post_tag' }.map { |k| k[:name] }
-        doc.data['categories'] = post[:meta].select { |k| k[:type] == 'category' }.map { |k| k[:name] }
-        doc.data['image'] = attachments[thumbnail] if thumbnail
-
-        doc.content = ReverseMarkdown.convert(sanitizer.sanitize(post[:content]))
-        doc.save
-      end
-    end
-  end
-end