ballmer 0.0.1

data/lib/ballmer/presentation/slide.rb

@@ -0,0 +1,19 @@
+module Ballmer
+  class Presentation
+    # Load a slide up in thar. 
+    class Slide < Document::XMLPart
+      # Key used to look up slides from [Content-Types].xml.
+      CONTENT_TYPE = "application/vnd.openxmlformats-officedocument.presentationml.slide+xml".freeze
+
+      # Key used to look up slides from .xml.rel documents
+      REL_TYPE = "http://schemas.openxmlformats.org/officeDocument/2006/relationships/slide".freeze
+
+      def notes
+        # TODO - Move a type caster into rels based on content type like 
+        # rels[Notes::REL_TYPE].first
+        notes_path = rels.targets(Notes::REL_TYPE).first.expand_path(@path.dirname)
+        Notes.new(@doc, notes_path)
+      end
+    end
+  end
+end

data/lib/ballmer/presentation/slides.rb

@@ -0,0 +1,172 @@
+module Ballmer
+  class Presentation
+    # Manages concerns around keeping slide and notesSlides files in
+    # sync with an array of slides. These basically needs to trasnact
+    # the slide\d+ and slideNote\d+ numbers to be in sync with an array.
+    # Its a big, ugly ass complicated beast. Send your thank you cards to Bill Gates.
+    class Slides
+      include Enumerable
+
+      def initialize(doc)
+        @doc = doc
+      end
+
+      def each(&block)
+        # TODO - Do NOT read content-types, but read Rels instead (and move this type casting in there.)
+        @doc.content_types[Slide::CONTENT_TYPE].each { |path| block.call slide path }
+      end
+
+      # This method is crazy because it has to manipulate a ton of files within the PPTX. Most of
+      # what happens in here I figured out by diff-ing PPTX files that had copies of identical slides, but
+      # a different number of slides.
+      def push(slide)
+        n = to_a.size + 1
+        # Paths within the zip file of new files we have to write.
+        slide_path = Pathname.new("/ppt/slides/slide#{n}.xml")
+        slide_rels_path = Pathname.new("/ppt/slides/_rels/slide#{n}.xml.rels")
+        slide_notes_path = Pathname.new("/ppt/notesSlides/notesSlide#{n}.xml")
+        slide_notes_rels_path = Pathname.new("/ppt/notesSlides/_rels/notesSlide#{n}.xml.rels")
+
+        # Update ./ppt
+        #   !!! CREATE !!!
+        #   ./slides
+        #     Create new files
+        #       ./slide(\d+).xml file
+        @doc.copy slide_path, slide.path
+        #       ./_rels/slide(\d+).xml.rels
+        @doc.copy slide_rels_path, slide.rels.path
+        #   ./notesSlides
+        #     Create new files
+        #       ./notesSlide(\d+).xml file
+        @doc.copy slide_notes_path, slide.notes.path
+        #       ./_rels/notesSlide(\d+).xml.rels
+        @doc.copy slide_notes_rels_path, slide.notes.rels.path
+        
+        #   !!! UPDATES !!!
+        # Update the notes in the new slide to point at the new notes
+        @doc.edit_xml slide_rels_path do |xml|
+          # TODO - Move this rel logic into the parts so that we don't have to repeat ourselves when calculating this stuff out.
+          xml.at_xpath("//xmlns:Relationship[@Type='#{Notes::REL_TYPE}']")['Target'] = slide_notes_path.relative_path_from(slide_path.dirname)
+        end
+
+        # Update teh slideNotes reference to point at the new slide
+        @doc.edit_xml slide_notes_rels_path do |xml|
+          xml.at_xpath("//xmlns:Relationship[@Type='#{Slide::REL_TYPE}']")['Target'] = slide_path.relative_path_from(slide_notes_path.dirname)
+        end
+
+        #   ./_rels/presentation.xml.rels
+        #     Update Relationship ids
+        #     Insert a new one slideRef
+        @doc.edit_xml @doc.presentation.rels.path do |xml|
+          # Calucate the next id
+          next_id = xml.xpath('//xmlns:Relationship[@Id]').map{ |n| n['Id'] }.sort.last.succ
+          # TODO - Figure out how to make this more MS idiomatic up 9->10 instead of incrementing
+          # the character....
+          # Insert that into the slide and crakc open the presentation.xml file
+          types = xml.at_xpath('/xmlns:Relationships')
+          types << Nokogiri::XML::Node.new("Relationship", xml).tap do |n|
+            n['Id'] = next_id
+            n['Type'] = Slide::REL_TYPE
+            n['Target'] = slide_path.relative_path_from(@doc.presentation.path.dirname)
+          end
+          #   ./presentation.xml
+          #     Update attr
+          #       p:notesMasterId
+          #     Insert attr
+          #       p:sldId, increment, etc.
+          @doc.edit_xml '/ppt/presentation.xml' do |xml|
+            slides = xml.at_xpath('/p:presentation/p:sldIdLst')
+            next_slide_id = slides.xpath('//p:sldId[@id]').map{ |n| n['id'] }.sort.last.succ
+            slides << Nokogiri::XML::Node.new("p:sldId", xml).tap do |n|
+              # TODO - Fix the ID that's jacked up.
+              n['id'] = next_slide_id
+              n['r:id'] = next_id
+            end
+          end
+        end
+
+        # Update ./[Content-Types].xml with new slide link and slideNotes link
+        @doc.edit_xml @doc.content_types.path do |xml|
+          types = xml.at_xpath('/xmlns:Types')
+          types << Nokogiri::XML::Node.new("Override", xml).tap do |n|
+            n['PartName'] = slide_path
+            n['ContentType'] = Slide::CONTENT_TYPE
+          end
+          types << Nokogiri::XML::Node.new("Override", xml).tap do |n|
+            n['PartName'] = slide_notes_path
+            n['ContentType'] = Notes::CONTENT_TYPE
+          end
+        end
+
+        # Great, that's all done, so lets return the slide eh?
+        slide slide_path
+      end
+
+      # Removes a slide from the slides collection
+      def delete(slide)
+        #   ./_rels/presentation.xml.rels
+        #     Update Relationship ids
+        #     Insert a new one slideRef
+        @doc.edit_xml @doc.presentation.rels.path do |xml|
+          # Calucate the next id
+            # next_id = xml.xpath('//xmlns:Relationship[@Id]').map{ |n| n['Id'] }.sort.last.succ
+          # TODO - Figure out how to make this more MS idiomatic up 9->10 instead of incrementing
+          # the character....
+          # Insert that into the slide and crakc open the presentation.xml file
+
+          target = slide.path.relative_path_from(@doc.presentation.path.dirname)
+          relationship = xml.at_xpath("/xmlns:Relationships/xmlns:Relationship[@Type='#{Slide::REL_TYPE}' and @Target='#{target}']")
+          #   ./presentation.xml
+          #     Update attr
+          #       p:notesMasterId
+          #     Insert attr
+          #       p:sldId, increment, etc.
+          @doc.edit_xml '/ppt/presentation.xml' do |xml|
+            xml.at_xpath("/p:presentation/p:sldIdLst/p:sldId[@r:id='#{relationship['Id']}']").remove
+          end
+          relationship.remove
+        end
+
+        # Delete slide link and slideNotes link from ./[Content-Types].xml 
+        @doc.edit_xml @doc.content_types.path do |xml|
+          xml.at_xpath("/xmlns:Types/xmlns:Override[@ContentType='#{Slide::CONTENT_TYPE}' and @PartName='#{slide.path}']").remove
+          xml.at_xpath("/xmlns:Types/xmlns:Override[@ContentType='#{Notes::CONTENT_TYPE}' and @PartName='#{slide.notes.path}']").remove
+        end
+
+        # Update ./ppt
+        #   !!! DESTROY !!!
+        #   ./slides
+        #     Delete files
+        #       ./_rels/notesSlide(\d+).xml.rels
+        @doc.delete slide.notes.rels.path
+        #       ./notesSlide(\d+).xml file
+        @doc.delete slide.notes.path
+        #       ./_rels/slide(\d+).xml.rels
+        @doc.delete slide.rels.path
+        #       ./slide(\d+).xml file
+        @doc.delete slide.path
+        #   ./notesSlides
+        #     Delete files
+
+        # Hooray! We're done! Ummm, what should we return though? can't be the slide since
+        # its destroyed and there's no practical way to keep it around in memory.
+      end
+
+      private
+      def slide(path)
+        Slide.new(@doc, path) 
+      end
+      # Reads from the [Content_Types].xml file the paths for the slide
+      #   <Override PartName="/ppt/slides/slide1.xml" ContentType="application/vnd.openxmlformats-officedocument.presentationml.slide+xml"/>
+      def parts
+        @doc.content_types.parts Slide::CONTENT_TYPE
+      end
+
+      # Microsoft decided it would be cool to start at 1 instead of 0 
+      # for the part indices, so this deals with that seperatly
+      def next_number
+        self.to_a.size + 1
+      end
+    end
+  end
+end

data/lib/ballmer/presentation/tags.rb

@@ -0,0 +1,50 @@
+module Ballmer
+  class Presentation
+    class Tags < Document::XMLPart
+      # TODO, there are three types of notes. We need to figure out 
+      # how to resolve the slide number, notes, and whatever the hell else
+      # the first note type is.
+
+      # Key used to look up notes from [Content-Types].xml.
+      CONTENT_TYPE = "application/vnd.openxmlformats-officedocument.presentationml.tags+xml".freeze
+
+      # Key used to look up notes from .xml.rel documents
+      REL_TYPE = 'http://schemas.openxmlformats.org/officeDocument/2006/relationships/tags'.freeze
+
+      # Read tag
+      def [](key)
+        if tag = tag(key)
+          tag['val']
+        end
+      end
+
+      # If the tag exists, create it -- otherwise update.
+      #   <p:tag name="" val=""/>
+      def []=(key, value)
+        unless tag key # Don't add the tag if it already exists.
+          tag_list << Nokogiri::XML::Node.new("p:tag", @xml) do |tag|
+            tag['name'] = key
+            tag['val'] = value
+          end
+        end
+      end
+
+      private
+
+      def tag(key)
+        tag_list.at_xpath("./p:tag[@name='#{key}']")
+      end
+
+      def tag_list
+        xml.at_xpath('p:tagLst')
+      end
+
+      def self.build(doc, path)
+        # Write the template
+        doc.write path, File.read(File.join(File.dirname(__FILE__), 'tags.xml'))
+        # K, now lets read it and decorate with the part.
+        new doc, path
+      end
+    end
+  end
+end

data/lib/ballmer/presentation/tags.xml

@@ -0,0 +1,3 @@
+<?xml version="1.0" encoding="UTF-8" standalone="yes"?>
+<p:tagLst xmlns:a="http://schemas.openxmlformats.org/drawingml/2006/main" xmlns:r="http://schemas.openxmlformats.org/officeDocument/2006/relationships" xmlns:p="http://schemas.openxmlformats.org/presentationml/2006/main">
+</p:tagLst>

data/lib/ballmer/presentation.rb

@@ -0,0 +1,20 @@
+module Ballmer
+  # Represents a presentation that has many slides.
+  class Presentation < Document
+    autoload :Slides,       'ballmer/presentation/slides'
+    autoload :Slide,        'ballmer/presentation/slide'
+    autoload :Notes,        'ballmer/presentation/notes'
+    autoload :NotesParser,  'ballmer/presentation/notes_parser'
+    autoload :Tags,         'ballmer/presentation/tags'
+    
+    # Return an array of slides.
+    def slides
+      @slides ||= Slides.new(self)
+    end
+
+    # Presentation XML file.
+    def presentation
+      XMLPart.new(self, '/ppt/presentation.xml')
+    end
+  end
+end

data/lib/ballmer/version.rb

@@ -0,0 +1,3 @@
+module Ballmer
+  VERSION = "0.0.1"
+end

data/lib/ballmer.rb

@@ -0,0 +1,7 @@
+require "ballmer/version"
+require "nokogiri"
+
+module Ballmer
+  autoload :Presentation, 'ballmer/presentation'
+  autoload :Document,     'ballmer/document'
+end

data/spec/lib/ballmer/document/content_types_spec.rb

@@ -0,0 +1,35 @@
+require 'spec_helper'
+
+describe Ballmer::Document::ContentTypes do
+  subject     { read_presentation('presentation3.pptx') }
+  let(:tags)  { Ballmer::Presentation::Tags.new(subject, 'tags/tag1.xml') }
+
+  describe "#exists?" do
+    it "should detect if part is in XML" do
+      subject.content_types.should exist(subject.slides.first)
+    end
+
+    it "should detect if part is not in XML" do
+      subject.content_types.should_not exist(tags)
+    end
+  end
+
+  describe "#append" do
+    before(:each) do
+      subject.content_types.append tags
+    end
+
+    let(:nodes) do
+      subject.content_types.xml.xpath("//xmlns:Override[@ContentType='#{tags.class::CONTENT_TYPE}' and @PartName='#{tags.path}']")
+    end
+
+    it "should write part to XML" do
+      nodes.should have(1).item
+    end
+
+    it "should not write duplicate parts to XML" do
+      subject.content_types.append tags
+      nodes.should have(1).item
+    end
+  end
+end

data/spec/lib/ballmer/document/rels_spec.rb

@@ -0,0 +1,43 @@
+require 'spec_helper'
+
+describe Ballmer::Document::Rels do
+  subject     { read_presentation('presentation3.pptx') }
+  let(:tags)  { Ballmer::Presentation::Tags.new(subject, '/ppt/tags/tag1.xml') }
+  let(:slide) { subject.slides.first }
+
+  describe "#exists?" do
+    it "should detect if part is in XML" do
+      subject.presentation.rels.should exist(subject.slides.first)
+    end
+
+    it "should detect if part is not in XML" do
+      subject.presentation.rels.should_not exist(tags)
+    end
+  end
+
+  describe "#append" do
+    before(:each) do
+      slide.rels.append tags
+    end
+
+    let(:nodes) do
+      slide.rels.xml.xpath("//xmlns:Relationship[@Type='#{tags.class::REL_TYPE}' and @Target='#{slide.relative_path_from(tags)}']")
+    end
+
+    it "should write part to XML" do
+      nodes.should have(1).item
+    end
+
+    it "should not write duplicate parts to XML" do
+      slide.rels.append tags
+      nodes.should have(1).item
+    end
+  end
+
+  describe "#id" do
+    it "should get id" do
+      # I just pulled this value out of the template.
+      subject.presentation.rels.id(subject.slides.first).should == 'rId2'
+    end
+  end
+end

data/spec/lib/ballmer/document_spec.rb

@@ -0,0 +1,12 @@
+require 'spec_helper'
+
+describe Ballmer::Document do
+  subject { read_presentation('presentation3.pptx') }
+
+  context "rels" do
+    it "should resolve"
+  end
+
+  context "content_types"
+  context "zip"
+end

data/spec/lib/ballmer/presentation/notes_parser_spec.rb

@@ -0,0 +1,19 @@
+# encoding: UTF-8
+
+require 'spec_helper'
+
+describe Ballmer::Presentation::NotesParser do
+  # Fixture contains a complex note with several different formats.
+  subject           { read_presentation('notes.pptx').slides.first.notes }
+
+  it "should have notes" do
+    subject.body.should == %[Some crazy notes are here. Some words are in bold, others are italic.
+
+Funny thing about notes. When there’s a mispullin’ it gets underlined and tokenized.]
+  end
+
+  it "should edit notes" do
+    subject.body = "Hi\bpig"
+    subject.body.should == "Hi\bpig"
+  end
+end

data/spec/lib/ballmer/presentation/tags_spec.rb

@@ -0,0 +1,22 @@
+require 'spec_helper'
+
+describe Ballmer::Presentation::Tags do
+  let(:presentation) { read_presentation('presentation1.pptx') }
+  subject { Ballmer::Presentation::Tags.build(presentation, 'ppt/tags/tags1.xml') }
+
+  it "should set key" do
+    subject['foo'] = 'bar'
+    subject.xml.at_xpath("//p:tag[@name='foo']")['val'].should == 'bar'
+  end
+
+  it "should read key" do
+    subject['foo'] = 'bar'
+    subject['foo'].should == 'bar'
+  end
+
+  it "should not write duplicate keys" do
+    subject['foo'] = 'bar'
+    subject['foo'] = 'bar'
+    subject.xml.xpath("//p:tag[@name='foo']").should have(1).item
+  end
+end

data/spec/lib/ballmer/presentation_spec.rb

@@ -0,0 +1,118 @@
+require 'spec_helper'
+
+describe Ballmer::Presentation do
+  subject { read_presentation('presentation3.pptx') }
+
+  context "presentation" do
+    it "should have slides" do
+      subject.slides.should have(3).items
+    end
+  end
+
+  context "slides" do
+    let(:original)  { subject.slides.to_a.first }
+    let(:copy)      { subject.slides.push original }
+
+    # Explicitly call copy before each test in case
+    # we don't call "copy" from a spec below since
+    # we don't make assertions on "copy" per test.
+    before(:each){ copy }
+
+    context "#push" do
+      it "should copy slide" do
+        original.notes.body.should == copy.notes.body
+      end
+
+      context "slide.xml" do
+        let(:slide) { copy }
+
+        it "should append" do
+          slide.path.to_s.should == '/ppt/slides/slide4.xml'
+        end
+
+        context "rels" do
+          it "should append ./_rels/slide.xml" do
+            slide.rels.path.to_s.should == '/ppt/slides/_rels/slide4.xml.rels'
+          end
+
+          it "should reference notesSlides.xml" do
+            rels = slide.rels.targets(Ballmer::Presentation::Notes::REL_TYPE)
+            rels.should have(1).item
+            rels.first.to_s.should == '../notesSlides/notesSlide4.xml'
+          end
+        end
+      end
+
+      context "notesSlides.xml" do
+        let(:notes) { copy.notes }
+
+        it "should append" do
+          notes.path.to_s.should == '/ppt/notesSlides/notesSlide4.xml'
+        end
+
+        context "rels" do
+          it "should append ./_rels/notesSlides.xml" do
+            notes.rels.path.to_s.should == '/ppt/notesSlides/_rels/notesSlide4.xml.rels'
+          end
+
+          it "should reference slides.xml" do
+            rels = notes.rels.targets(Ballmer::Presentation::Slide::REL_TYPE)
+            rels.should have(1).item
+            rels.first.to_s.should == '../slides/slide4.xml'
+          end
+        end
+      end
+
+      context "[Content-Type].xml" do
+        it "should add slide Override" do
+          subject.content_types[Ballmer::Presentation::Slide::CONTENT_TYPE].should have(4).items
+        end
+      end
+
+      context "presentation.xml" do
+        it "should add slide"
+        it "should add slideId"
+
+        context "_rels" do
+          it "should add slide"
+          it "should add "
+        end
+      end
+    end
+
+    context "#delete" do
+      before(:each) do
+        subject.slides.delete subject.slides.first
+      end
+
+      it "should remove slide" do
+        subject.should have(3).slides
+      end
+
+      context "[Content-Type].xml" do
+        it "should remove slide"
+      end
+
+      context "presentation.xml" do
+        it "should remove slide"
+        context "_rels" do
+          it "should remove slide"
+        end
+      end
+
+      context "slide.xml" do
+        it "should delete"
+        context "rels" do
+          it "should delete"
+        end
+      end
+
+      context "noteSlide.xml" do
+        it "should delete"
+        context "rels" do
+          it "should delete"
+        end
+      end
+    end
+  end
+end

data/spec/spec_helper.rb

@@ -0,0 +1,38 @@
+require "ballmer"
+
+# Helpers
+module Ballmer
+  module Helpers
+    # Open a presentation, k?
+    def read_presentation(file)
+      Presentation.read(read(file))
+    end
+
+    private
+    # Read the bytes of a document.
+    def read(file)
+      File.read(File.join(File.dirname(__FILE__), 'fixtures', file))
+    end
+  end
+end
+
+# This file was generated by the `rspec --init` command. Conventionally, all
+# specs live under a `spec` directory, which RSpec adds to the `$LOAD_PATH`.
+# Require this file using `require "spec_helper"` to ensure that it is only
+# loaded once.
+#
+# See http://rubydoc.info/gems/rspec-core/RSpec/Core/Configuration
+RSpec.configure do |config|
+  config.treat_symbols_as_metadata_keys_with_true_values = true
+  config.run_all_when_everything_filtered = true
+  config.filter_run :focus
+
+  # Run specs in random order to surface order dependencies. If you find an
+  # order dependency and want to debug it, you can fix the order by providing
+  # the seed, which is printed after each run.
+  #     --seed 1234
+  config.order = 'random'
+
+  # Include the helpers module
+  config.include Ballmer::Helpers
+end