sablon 0.0.18 → 0.0.19.beta1

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: a32d344154e7d3212cf75d5360c2f593122bd082
-  data.tar.gz: 2224e7a3f29c836876bae9aa12cb812c91feb2fd
+  metadata.gz: 4a08176b5a9b4646da6f24b4aff58ac09dcbc7cd
+  data.tar.gz: f2549ddf6142a9ffb02dde86b6a13d9eb38cfbbd
 SHA512:
-  metadata.gz: 711140b3b7ea9723188d5937d79d8fc8be86a76ccad7b9d3ae1c60c268da0ed072b129c9eef4447183c150bcdec7c98c08d1735836a67b1c9109dd500b3b7d0a
-  data.tar.gz: 04000b91fb2cd8e4941a5b1bb073bf22f73a8d1de5214f5b9b6eff8a25942bc60946126235fffa457dd14feac8fb9b13a1472878848ec89d33606ad803f6a0e8
+  metadata.gz: 1f54003471153af06f163732d84eeab5768e7430f744702062d6868a9ca9f1be6847875b141f7e69ed6e5d4c62957d2f8853dc058696d2bf6234beee9b74a6fd
+  data.tar.gz: 25731c1257863f69c9b68586041a35e0c2ca85b1d85e01bc6c8c14a2ec569920aa57658a40fc0d7f002681181eba59c1b71e4c0e37f17c3b83157b42f8f5512d

data/.travis.yml

@@ -1,5 +1,5 @@
 language: ruby
 rvm:
-  - 1.9.3
-  - 2.0.0
-  - 2.1.2
+  - 2.1.8
+  - 2.2.4
+  - 2.3.0

data/Gemfile.lock

@@ -1,7 +1,7 @@
 PATH
   remote: .
   specs:
-    sablon (0.0.18)
+    sablon (0.0.19.beta1)
       nokogiri (>= 1.6.0)
       redcarpet (>= 3.2)
       rubyzip (>= 1.1)
@@ -9,12 +9,12 @@ PATH
 GEM
   remote: https://rubygems.org/
   specs:
-    mini_portile (0.6.2)
+    mini_portile2 (2.0.0)
     minitest (5.8.0)
-    nokogiri (1.6.6.2)
-      mini_portile (~> 0.6.0)
+    nokogiri (1.6.7.1)
+      mini_portile2 (~> 2.0.0.rc2)
     rake (10.4.2)
-    redcarpet (3.3.2)
+    redcarpet (3.3.3)
     rubyzip (1.1.7)
     xml-simple (1.1.5)
 

data/README.md

@@ -100,7 +100,7 @@ IMPORTANT: This feature is very much *experimental*. Currently, the insertion
     will replace the containing paragraph. This means that other content in the same
     paragraph is discarded.
 
-##### Markdown
+##### Markdown [experimental]
 
 Similar to WordProcessingML it's possible to use markdown while processing the
 tempalte. You don't need to modify your templates, a simple insertion operation
@@ -138,10 +138,42 @@ etc. according to the header level. Ordered lists will use the style
 supported. Normal text paragraphs use the style `Paragraph`. It's not necessary
 to have that style in the template. Word will fall back to using the `Normal` style.
 
+IMPORTANT: This feature is very much *experimental*. Currently, the insertion
+    will replace the containing paragraph. This means that other content in the same
+    paragraph is discarded. In the near future this feature will most likely be
+    implemented on top of HTML insertion.
+
+##### HTML [experimental]
+
+Similar to WordProcessingML it's possible to use html as input while processing the
+tempalte. You don't need to modify your templates, a simple insertion operation
+is sufficient:
+
+```
+«=article.body»
+```
+
+To use HTML insertion prepare the context like so:
+
+```ruby
+html_body = <<-HTML
+<div>This text can contain <em>additional formatting</em>
+according to the <strong>Markdown</strongstrong> specification.</div>
+HTML
+context = {
+  article: { html_body: Sablon.content(:html, html_body) }
+}
+template.render_to_file File.expand_path("~/Desktop/output.docx"), context
+```
+
+Currently HTML insertion is very limited and strongly focused on the HTML
+generated by [Trix editor](https://github.com/basecamp/trix).
+
 IMPORTANT: This feature is very much *experimental*. Currently, the insertion
     will replace the containing paragraph. This means that other content in the same
     paragraph is discarded.
 
+
 #### Conditionals
 
 Sablon can render parts of the template conditonally based on the value of a

data/lib/sablon.rb

@@ -1,14 +1,19 @@
+require 'singleton'
+require 'zip'
+require 'nokogiri'
+
 require "sablon/version"
+require "sablon/numbering"
 require "sablon/context"
 require "sablon/template"
-require "sablon/processor"
+require "sablon/processor/document"
 require "sablon/processor/section_properties"
+require "sablon/processor/numbering"
 require "sablon/parser/mail_merge"
 require "sablon/operations"
+require "sablon/html/converter"
 require "sablon/content"
 
-require 'zip'
-require 'nokogiri'
 require 'redcarpet'
 require "sablon/redcarpet/render/word_ml"
 

data/lib/sablon/content.rb

@@ -93,8 +93,25 @@ module Sablon
       end
     end
 
+    class HTML < Struct.new(:word_ml)
+      include Sablon::Content
+      def self.id; :html end
+      def self.wraps?(value) false end
+
+      def initialize(html)
+        converter = HTMLConverter.new
+        word_ml = Sablon.content(:word_ml, converter.process(html))
+        super word_ml
+      end
+
+      def append_to(*args)
+        word_ml.append_to(*args)
+      end
+    end
+
     register Sablon::Content::String
     register Sablon::Content::WordML
     register Sablon::Content::Markdown
+    register Sablon::Content::HTML
   end
 end

data/lib/sablon/html/ast.rb

@@ -0,0 +1,130 @@
+module Sablon
+  class HTMLConverter
+    class Node
+      def accept(visitor)
+        visitor.visit(self)
+      end
+
+      def self.node_name
+        @node_name ||= name.split('::').last
+      end
+    end
+
+    class Collection < Node
+      attr_reader :nodes
+      def initialize(nodes)
+        @nodes = nodes
+      end
+
+      def accept(visitor)
+        super
+        @nodes.each do |node|
+          node.accept(visitor)
+        end
+      end
+
+      def to_docx
+        nodes.map(&:to_docx).join
+      end
+    end
+
+    class Root < Collection
+      def to_a
+        nodes
+      end
+
+      def grep(pattern)
+        visitor = GrepVisitor.new(pattern)
+        accept(visitor)
+        visitor.result
+      end
+    end
+
+    class Paragraph < Node
+      attr_accessor :style, :runs
+      def initialize(style, runs)
+        @style, @runs = style, runs
+      end
+
+      PATTERN = <<-XML.gsub("\n", "")
+<w:p>
+<w:pPr>
+<w:pStyle w:val="%s" />
+%s
+</w:pPr>
+%s
+</w:p>
+XML
+
+      def to_docx
+        PATTERN % [style, ppr_docx, runs.to_docx]
+      end
+
+      def accept(visitor)
+        super
+        runs.accept(visitor)
+      end
+
+      private
+      def ppr_docx
+      end
+    end
+
+    class ListParagraph < Paragraph
+      LIST_STYLE = <<-XML.gsub("\n", "")
+<w:numPr>
+<w:ilvl w:val="%s" />
+<w:numId w:val="%s" />
+</w:numPr>
+XML
+      attr_accessor :numid, :ilvl
+      def initialize(style, runs, numid, ilvl)
+        super style, runs
+        @numid = numid
+        @ilvl = ilvl
+      end
+
+      private
+      def ppr_docx
+        LIST_STYLE % [@ilvl, numid]
+      end
+    end
+
+    class Text < Node
+      attr_reader :string
+      def initialize(string)
+        @string = string
+      end
+
+      def to_docx
+        "<w:r>#{style_docx}<w:t xml:space=\"preserve\">#{normalized_string}</w:t></w:r>"
+      end
+
+      private
+      def style_docx
+      end
+
+      def normalized_string
+        string.tr("\u00A0", ' ')
+      end
+    end
+
+    class Bold < Text
+      def style_docx
+        '<w:rPr><w:b /></w:rPr>'
+      end
+    end
+
+    class Italic < Text
+      def style_docx
+        '<w:rPr><w:i /></w:rPr>'
+      end
+    end
+
+    class Newline < Node
+      def to_docx
+        "<w:r><w:br/></w:r>"
+      end
+    end
+  end
+end

data/lib/sablon/html/converter.rb

@@ -0,0 +1,133 @@
+require "sablon/html/ast"
+require "sablon/html/visitor"
+
+module Sablon
+  class HTMLConverter
+    class ASTBuilder
+      Layer = Struct.new(:items, :ilvl)
+
+      def initialize(nodes)
+        @layers = [Layer.new(nodes, false)]
+        @root = Root.new([])
+      end
+
+      def to_ast
+        @root
+      end
+
+      def new_layer(ilvl: false)
+        @layers.push Layer.new([], ilvl)
+      end
+
+      def next
+        current_layer.items.shift
+      end
+
+      def push(node)
+        @layers.last.items.push node
+      end
+
+      def push_all(nodes)
+        nodes.each(&method(:push))
+      end
+
+      def done?
+        !current_layer.items.any?
+      end
+
+      def nested?
+        ilvl > 0
+      end
+
+      def ilvl
+        @layers.select { |layer| layer.ilvl }.size - 1
+      end
+
+      def emit(node)
+        @root.nodes << node
+      end
+
+      private
+      def current_layer
+        if @layers.any?
+          last_layer = @layers.last
+          if last_layer.items.any?
+            last_layer
+          else
+            @layers.pop
+            current_layer
+          end
+        else
+          Layer.new([], false)
+        end
+      end
+    end
+
+    def process(input)
+      processed_ast(input).to_docx
+    end
+
+    def processed_ast(input)
+      ast = build_ast(input)
+      ast
+    end
+
+    def build_ast(input)
+      doc = Nokogiri::HTML.fragment(input)
+      @builder = ASTBuilder.new(doc.children)
+
+      while !@builder.done?
+        ast_next_paragraph
+      end
+      @builder.to_ast
+    end
+
+    private
+    def ast_next_paragraph
+      node = @builder.next
+      if node.name == 'div' || node.name == 'p'
+        @builder.new_layer
+        @builder.emit Paragraph.new('Paragraph', text(node.children))
+      elsif node.name == 'ul'
+        @builder.new_layer ilvl: true
+        unless @builder.nested?
+          @definition = Sablon::Numbering.instance.register('ListBullet')
+        end
+        @builder.push_all(node.children)
+      elsif node.name == 'ol'
+        @builder.new_layer ilvl: true
+        unless @builder.nested?
+          @definition = Sablon::Numbering.instance.register('ListNumber')
+        end
+        @builder.push_all(node.children)
+      elsif node.name == 'li'
+        @builder.new_layer
+        @builder.emit ListParagraph.new(@definition.style, text(node.children), @definition.numid, @builder.ilvl)
+      elsif node.text?
+        # SKIP?
+      else
+        raise ArgumentError, "Don't know how to handle node: #{node.inspect}"
+      end
+    end
+
+    def text(nodes)
+      runs = nodes.map do |node|
+        if node.text?
+          Text.new(node.text)
+        elsif node.name == 'br'
+          Newline.new
+        elsif node.name == 'strong'
+          Bold.new(node.text)
+        elsif node.name == 'em'
+          Italic.new(node.text)
+        elsif ['ul', 'ol', 'p', 'div'].include?(node.name)
+          @builder.push(node)
+          nil
+        else
+          raise ArgumentError, "Don't know how to handle node: #{node.inspect}"
+        end
+      end
+      Collection.new(runs.compact)
+    end
+  end
+end

data/lib/sablon/html/visitor.rb

@@ -0,0 +1,26 @@
+module Sablon
+  class HTMLConverter
+    class Visitor
+      def visit(node)
+        method_name = "visit_#{node.class.node_name}"
+        if respond_to? method_name
+          public_send method_name, node
+        end
+      end
+    end
+
+    class GrepVisitor
+      attr_reader :result
+      def initialize(pattern)
+        @pattern = pattern
+        @result = []
+      end
+
+      def visit(node)
+        if @pattern === node
+          @result << node
+        end
+      end
+    end
+  end
+end

data/lib/sablon/numbering.rb

@@ -0,0 +1,28 @@
+module Sablon
+  class Numbering
+    include Singleton
+    attr_reader :definitions
+
+    Definition = Struct.new(:numid, :style) do
+      def inspect
+        "#<Numbering #{numid}:#{style}"
+      end
+    end
+
+    def initialize
+      reset!
+    end
+
+    def reset!
+      @numid = 1000
+      @definitions = []
+    end
+
+    def register(style)
+      @numid += 1
+      definition = Definition.new(@numid, style)
+      @definitions << definition
+      definition
+    end
+  end
+end