mill 0.1 → 0.3

data/lib/mill/resources/dir.rb

@@ -0,0 +1,31 @@
+module Mill
+
+  class Resource
+
+    class Index < Text
+
+      include HTMLHelpers
+
+      attr_accessor :pages
+
+      def initialize(pages:, **args)
+        @pages = pages
+        super(**args)
+      end
+
+      def load
+        super
+        @content = html_fragment do |html|
+          html.dl do
+            @pages.each do |page|
+              html.dt(page.title)
+            end
+          end
+        end
+      end
+
+    end
+
+  end
+
+end

data/lib/mill/resources/feed.rb

@@ -1,6 +1,6 @@
 # see http://www.sitemaps.org/protocol.php
 
-class Mill
+module Mill
 
   class Resource
 
@@ -8,25 +8,21 @@ class Mill
 
       include HTMLHelpers
 
-      def self.type
-        :feed
-      end
-
-      def load
-        resources = @mill.public_resources.sort_by(&:date)
+      def build
+        resources = @site.feed_resources
         builder = Nokogiri::XML::Builder.new do |xml|
           xml.feed(xmlns: 'http://www.w3.org/2005/Atom') do
-            xml.id(@mill.tag_uri)
-            xml.generator(*@mill.feed_generator)
-            xml.title(@mill.site_title)
-            xml.link(rel: 'alternate', type: 'text/html',             href: @mill.home_resource.uri)
+            xml.id(@site.tag_uri)
+            # xml.generator(*@site.feed_generator)
+            xml.title(@site.site_title) if @site.site_title
+            xml.link(rel: 'alternate', type: 'text/html',             href: @site.home_resource.uri) if @site.home_resource
             xml.link(rel: 'self',      type: 'application/atom+xml',  href: uri)
             xml.author do
-              xml.name(@mill.feed_author_name)
-              xml.uri(@mill.feed_author_uri)
-              xml.email(@mill.feed_author_email)
+              xml.name(@site.feed_author_name) if @site.feed_author_name
+              xml.uri(@site.feed_author_uri) if @site.feed_author_uri
+              xml.email(@site.feed_author_email) if @site.feed_author_email
             end
-            xml.updated(resources.last.date.iso8601)
+            xml.updated(resources.last.date.iso8601) unless resources.empty?
             resources.each do |resource|
               xml.entry do
                 xml.title(resource.title) if resource.title
@@ -34,13 +30,8 @@ class Mill
                 xml.id(resource.tag_uri)
                 xml.updated(resource.date.iso8601)
                 xml.published(resource.date.iso8601)
-                if (resource.respond_to?(:feed_summary))
-                  type, data = resource.feed_summary
-                  xml.summary(type: type) { xml.cdata(data) } if type && data
-                end
-                if (resource.respond_to?(:feed_content))
-                  type, data = resource.feed_content
-                  xml.content(type: type) { xml.cdata(data) }
+                if (html = resource.feed_content)
+                  xml.content(type: 'html') { xml.cdata(html.to_html) }
                 end
               end
             end

data/lib/mill/resources/google_site_verification.rb

@@ -0,0 +1,24 @@
+module Mill
+
+  class Resource
+
+    class GoogleSiteVerification < Resource
+
+      attr_accessor :key
+
+      def initialize(key:, **args)
+        @key = key
+        @public = false
+        super(**args)
+      end
+
+      def load
+        @content = "google-site-verification: #{@key}.html\n"
+        super
+      end
+
+    end
+
+  end
+
+end

data/lib/mill/resources/image.rb

@@ -1,4 +1,4 @@
-class Mill
+module Mill
 
   class Resource
 
@@ -6,13 +6,20 @@ class Mill
 
       include HTMLHelpers
 
+      FileTypes = %w{
+        image/gif
+        image/jpeg
+        image/png
+        image/tiff
+        image/vnd.microsoft.icon
+        image/x-icon
+        image/svg
+        image/svg+xml
+      }
+
       attr_accessor :width
       attr_accessor :height
 
-      def self.type
-        :image
-      end
-
       def load
         info = ImageSize.path(@input_file.to_s)
         @width, @height = *info.size

data/lib/mill/resources/other.rb

@@ -0,0 +1,29 @@
+module Mill
+
+  class Resource
+
+    class Other < Resource
+
+      FileTypes = %w{
+        application/pdf
+
+        application/x-shockwave-flash
+
+        application/font-sfnt
+        application/x-font-opentype
+        application/x-font-otf
+        application/x-font-truetype
+        application/x-font-ttf
+
+        application/rtf
+
+        application/javascript
+        application/x-javascript
+        text/javascript
+      }
+
+    end
+
+  end
+
+end

data/lib/mill/resources/redirect.rb

@@ -1,4 +1,4 @@
-class Mill
+module Mill
 
   class Resource
 
@@ -7,16 +7,10 @@ class Mill
       attr_accessor :redirect_uri
       attr_accessor :redirect_code
 
-      def self.type
-        :redirect
-      end
-
-      def initialize(params={})
-        super(
-          {
-            redirect_code: 303,
-          }.merge(params)
-        )
+      def initialize(redirect_uri:, redirect_code: 303, **args)
+        @redirect_uri = redirect_uri
+        @redirect_code = redirect_code
+        super(**args)
       end
 
       def load
@@ -24,7 +18,7 @@ class Mill
         super
       end
 
-      def build
+      def save
         @output_file = @output_file.add_extension('.redirect')
         super
       end
@@ -33,4 +27,4 @@ class Mill
 
   end
 
-end
+end

data/lib/mill/resources/robots.rb

@@ -1,19 +1,16 @@
 # see http://www.robotstxt.org/robotstxt.html
 
-class Mill
+module Mill
 
   class Resource
 
     class Robots < Resource
 
-      def self.type
-        :robots
-      end
-
-      def load
-        info = {
-          'Sitemap' => @mill.sitemap_resource.absolute_uri,
-        }
+      def build
+        info = {}
+        info['User-Agent'] = '*'
+        info['Disallow'] = '/' unless @site.allow_robots
+        info['Sitemap'] = @site.sitemap_resource.absolute_uri if @site.make_sitemap
         @content = info.map { |key, value| "#{key}: #{value}" }.join("\n")
         super
       end

data/lib/mill/resources/sitemap.rb

@@ -1,21 +1,17 @@
 # see http://www.sitemaps.org/protocol.php
 
-class Mill
+module Mill
 
   class Resource
 
     class Sitemap < Resource
 
-      def self.type
-        :sitemap
-      end
-
-      def load
+      def build
         builder = Nokogiri::XML::Builder.new do |xml|
           xml.urlset('xmlns' => 'http://www.sitemaps.org/schemas/sitemap/0.9',
                      'xmlns:xsi' => 'http://www.w3.org/2001/XMLSchema-instance',
                      'xsi:schemaLocation' => 'http://www.sitemaps.org/schemas/sitemap/0.9 http://www.sitemaps.org/schemas/sitemap/0.9/sitemap.xsd') do
-            @mill.public_resources.each do |resource|
+            @site.public_resources.each do |resource|
               xml.url do
                 xml.loc(resource.absolute_uri)
                 xml.lastmod(resource.date.iso8601)

data/lib/mill/resources/stylesheet.rb

@@ -0,0 +1,27 @@
+module Mill
+
+  class Resource
+
+    class Stylesheet < Resource
+
+      FileTypes = %w{
+        text/css
+      }
+
+      def load
+        super
+        unless @input_file.basename.to_s.end_with?('.min.css')
+          engine = Sass::Engine.for_file(@input_file.to_s, syntax: :scss, style: :compressed)
+          begin
+            @content = engine.render
+          rescue Sass::SyntaxError => e
+            raise "#{input_file}: error parsing CSS: #{e}"
+          end
+        end
+      end
+
+    end
+
+  end
+
+end

data/lib/mill/resources/text.rb

@@ -1,4 +1,4 @@
-class Mill
+module Mill
 
   class Resource
 
@@ -6,54 +6,79 @@ class Mill
 
       include HTMLHelpers
 
-      attr_accessor :title
+      FileTypes = %w{
+        text/plain
+        text/html
+      }
 
-      def self.type
-        :text
+      attr_accessor :title
+      attr_accessor :summary
+      attr_accessor :author
+
+      def initialize(title: nil, summary: nil, author: nil, public: true, **args)
+        @title = title
+        @summary = summary
+        @author = author
+        super(public: public, **args)
       end
 
-      def initialize(params={})
-        super(
-          {
-            public: true,
-          }.merge(params)
-        )
+      def inspect
+        super + ", title: %p, summary: %p, author: %p" % [
+          @title,
+          @summary,
+          @author,
+        ]
       end
 
       def load
+        super
         if @input_file
           @content = @input_file.read
-          markup_class = case @input_file.extname
+          mode = case @input_file.extname.downcase
           when '.md', '.mdown', '.markdown'
-            Kramdown::Document
+            :markdown
           when '.textile'
-            RedCloth
+            :textile
           when '.txt'
-            PreText
+            :pre
+          when '.htm', '.html'
+            :html
           else
-            nil
+            raise "Unknown text type: #{@input_file}"
           end
-          if markup_class
+          if mode != :html
             parse_text_header
-            raise "#{uri}: Content is empty" unless @content
-            @content = markup_class.new(@content).to_html
+            @content = (@content || '').to_html(mode: mode, multiline: true)
             @output_file = @output_file.replace_extension('.html')
           end
           begin
             @content = parse_html(@content)
-          rescue HTMLError => e
-            raise "failed to parse #{@input_file}: #{e}"
+          rescue Error => e
+            raise e, "#{@input_file}: #{e}"
           end
           parse_html_header
         end
-        add_image_sizes
-        convert_relative_links
+      end
+
+      def build
+        replace_elements
         super
       end
 
+      def replace_elements
+        replace_pages_element
+        remove_comments
+        add_image_sizes
+        # shorten_links
+      end
+
       def parse_html_header
-        if (title_elem = @content.at_xpath('/html/head/title'))
-          @title = title_elem.text
+        unless @title
+          if (title_elem = @content.at_xpath('/html/head/title'))
+            @title = title_elem.text
+          else
+            @title = uri.to_s
+          end
         end
         @content.xpath('/html/head/meta[@name]').each do |meta|
           send("#{meta['name']}=", meta['content'])
@@ -61,7 +86,7 @@ class Mill
       end
 
       def parse_text_header
-        if @content =~ /^\w+:\s+/
+        if @content.split(/\n/, 2).first =~ /^\w+:\s+/
           header, @content = @content.split(/\n\n/, 2)
           header.split(/\n/).map do |line|
             key, value = line.strip.split(/:\s+/, 2)
@@ -72,82 +97,137 @@ class Mill
       end
 
       def final_content
-        html_document do |doc|
+        html_document(@site.html_version) do |doc|
           doc.html(lang: 'en') do |html|
-            html.head do
-              html << head.to_html
+            html << head.to_html
+            html << body.to_html
+          end
+        end.to_html
+      end
+
+      def head(&block)
+        html_fragment do |html|
+          html.head do
+            head = content_head
+            if (title = @title || (head && head.at_xpath('title')))
+              html.title { html << title.to_html }
             end
-            html.body do
-              html << body.to_html
+            yield(html) if block_given?
+            if head
+              head.children.reject { |e| e.text? || e.name == 'title' }.each do |e|
+                html << e.to_html
+              end
             end
           end
         end
       end
 
-      def head
-        @content.at_xpath('/html/head').children
+      def body(&block)
+        html_fragment do |html|
+          html.body do
+            if (elem = content_body)
+              html << elem.children.to_html
+            end
+            yield(html) if block_given?
+          end
+        end
       end
 
-      def body
-        @content.at_xpath('/html/body').children
+      def content_head
+        @content && @content.at_xpath('/html/head')
       end
 
-      def verify
-        tidy_html(@output_file.read) do |error_str|
-          warn "#{uri}: #{error_str}"
-        end
+      def content_body
+        @content && @content.at_xpath('/html/body')
       end
 
       def add_external_link_targets
         @content.xpath('//a').each do |a|
           if a['href'] && a['href'] =~ /^\w+:/
             a['target'] = '_blank'
+            a['rel'] = 'noopener noreferrer'
           end
         end
       end
 
+      def remove_comments
+        @content.xpath('//comment()').each do |comment|
+          comment.remove
+        end
+      end
+
       def add_image_sizes
         @content.xpath('//img').each do |img|
           # skip elements that already have width/height defined
           next if img[:width] || img[:height]
           img_link = Addressable::URI.parse(img['src'])
-          raise "no link in <img> element: #{img.to_s}" if img_link.nil? || img_link.empty?
+          raise Error, "No link in <img> element: #{img.to_s}" if img_link.nil? || img_link.empty?
           next if img_link.host
           img_uri = uri + img_link
-          img_resource = @mill.find_resource(img_uri) or raise "Can't find image for #{img_uri}"
+          img_resource = @site.find_resource(img_uri) or raise Error, "Can't find image for #{img_uri}"
           img[:width], img[:height] = img_resource.width, img_resource.height
         end
       end
 
-      def convert_relative_links
-        @mill.link_elem_attrs.each do |xpath|
-          convert_relative_link(xpath)
+      def shorten_links
+        find_link_elements(@content).each do |attribute|
+          elem = attribute.parent
+          link_uri = Addressable::URI.parse(attribute.value) or raise Error, "Can't parse #{attribute.value.inspect} from #{xpath.inspect}"
+          link_uri = uri + link_uri
+          if link_uri.relative?
+            self_uri = uri.normalize
+            self_uri.scheme = 'http'
+            link_uri.scheme = 'http'
+            attribute.value = self_uri.route_to(link_uri)
+            # ;;warn "[#{uri}] shortened link #{elem.name}/@#{attribute.name}: #{link_uri} => #{attribute.value}"
+          end
         end
       end
 
-      def convert_relative_link(elem_attr)
-        @content.xpath("//#{elem_attr}").each do |attribute|
-          elem = attribute.parent
-          link_uri = Addressable::URI.parse(attribute.value) or raise "Can't parse #{attribute.value.inspect} from #{xpath.inspect}"
-          if !link_uri.path.empty? && link_uri.path[0] != '/'
-            attribute.value = uri + link_uri
-            # ;;warn "[#{uri}] absolutized #{elem.name}/@#{attribute.name}: #{link_uri} => #{attribute.value}"
+      def replace_pages_element
+        replace_element(@content, '//pages') do |elem|
+          html_fragment do |html|
+            html.dl do
+              find_sibling_resources(Resource::Text).each do |page|
+                html.dt do
+                  html.a(href: page.uri) { html << page.title.to_html }
+                end
+                html.dd do
+                  if (summary = page.summary)
+                    html << summary.to_html
+                  end
+                end
+              end
+            end
           end
         end
       end
 
-      def feed_summary
-        ;;raise "#{uri} has no content" unless @content
-        if (p = @content.at_xpath('/html/body/p[1]'))
-          ['html', p.to_html]
-        else
-          nil
-        end
+      def summary
+        @summary || ((p = feed_content.at_xpath('//p')) && p.children)
       end
 
       def feed_content
-        body = @content.at_xpath('/html/body') or raise "#{uri} has no content"
-        ['html', body.children.to_html]
+        if (body = content_body)
+          # If we have a main element (<div class="main"> or <main>), use that.
+          # Otherwise, use the body, but delete header/footer/nav divs or elements.
+          if (main = body.at_xpath('//div[@id="main"]')) || (main = body.at_xpath('//main'))
+            main.children
+          elsif (article = body.at_xpath('//article'))
+            article.children
+          else
+            body_elem = body.clone
+            %w{header nav masthead footer}.each do |name|
+              if (elem = body_elem.at_xpath("//div[@id=\"#{name}\"]")) || (elem = body_elem.at_xpath("//#{name}"))
+                elem.remove
+              end
+            end
+            body_elem.children
+          end
+        else
+          warn "Warning: Resource #{uri} (#{self.class}) has no content"
+          nil
+        end
       end
 
     end