dg-axlsx 2.1.0

data/lib/axlsx/drawing/vml_shape.rb

@@ -0,0 +1,66 @@
+module Axlsx
+
+  # A VmlShape is used to position and render a comment.
+  class VmlShape
+
+    include Axlsx::OptionsParser
+    include Axlsx::Accessors
+
+    # Creates a new VmlShape
+    # @option options [Integer] row
+    # @option options [Integer] column
+    # @option options [Integer] left_column
+    # @option options [Integer] left_offset
+    # @option options [Integer] top_row
+    # @option options [Integer] top_offset
+    # @option options [Integer] right_column
+    # @option options [Integer] right_offset
+    # @option options [Integer] bottom_row
+    # @option options [Integer] bottom_offset
+    def initialize(options={})
+      @row = @column = @left_column = @top_row = @right_column = @bottom_row = 0
+      @left_offset = 15
+      @top_offset = 2
+      @right_offset = 50
+      @bottom_offset = 5
+      @visible = true
+      @id = (0...8).map{65.+(rand(25)).chr}.join
+      parse_options options
+      yield self if block_given?
+    end
+
+    unsigned_int_attr_accessor :row, :column, :left_column, :left_offset, :top_row, :top_offset,
+                               :right_column, :right_offset, :bottom_row, :bottom_offset
+
+    boolean_attr_accessor :visible
+
+    # serialize the shape to a string
+    # @param [String] str
+    # @return [String]
+    def to_xml_string(str ='')
+str << <<SHAME_ON_YOU
+
+<v:shape id="#{@id}" type="#_x0000_t202" fillcolor="#ffffa1 [80]" o:insetmode="auto"
+  style="visibility:#{@visible ? 'visible' : 'hidden'}">
+  <v:fill color2="#ffffa1 [80]"/>
+  <v:shadow on="t" obscured="t"/>
+  <v:path o:connecttype="none"/>
+  <v:textbox style='mso-fit-text-with-word-wrap:t'>
+   <div style='text-align:left'></div>
+  </v:textbox>
+
+  <x:ClientData ObjectType="Note">
+   <x:MoveWithCells/>
+   <x:SizeWithCells/>
+   <x:Anchor>#{left_column}, #{left_offset}, #{top_row}, #{top_offset}, #{right_column}, #{right_offset}, #{bottom_row}, #{bottom_offset}</x:Anchor>
+   <x:AutoFill>False</x:AutoFill>
+   <x:Row>#{row}</x:Row>
+   <x:Column>#{column}</x:Column>
+   #{@visible ? '<x:Visible/>' : ''}
+  </x:ClientData>
+ </v:shape>
+SHAME_ON_YOU
+
+    end
+  end
+end

data/lib/axlsx/package.rb

@@ -0,0 +1,363 @@
+# encoding: UTF-8
+module Axlsx
+  # Package is responsible for managing all the bits and peices that Open Office XML requires to make a valid
+  # xlsx document including valdation and serialization.
+  class Package
+    include Axlsx::OptionsParser
+
+    # provides access to the app doc properties for this package
+    # see App
+    attr_reader :app
+
+    # provides access to the core doc properties for the package
+    # see Core
+    attr_reader :core
+
+    # Initializes your package
+    #
+    # @param [Hash] options A hash that you can use to specify the author and workbook for this package.
+    # @option options [String] :author The author of the document
+    # @option options [Time] :created_at Timestamp in the document properties (defaults to current time).
+    # @option options [Boolean] :use_shared_strings This is passed to the workbook to specify that shared strings should be used when serializing the package.
+    # @example Package.new :author => 'you!', :workbook => Workbook.new
+    def initialize(options={})
+      @workbook = nil
+      @core, @app = Core.new, App.new
+      @core.creator = options[:author] || @core.creator
+      @core.created = options[:created_at]
+      parse_options options
+      yield self if block_given?
+    end
+
+    # Shortcut to specify that the workbook should use autowidth
+    # @see Workbook#use_autowidth
+    def use_autowidth=(v)
+      Axlsx::validate_boolean(v);
+      workbook.use_autowidth = v
+    end
+
+
+
+    # Shortcut to determine if the workbook is configured to use shared strings
+    # @see Workbook#use_shared_strings
+    def use_shared_strings
+      workbook.use_shared_strings
+    end
+
+    # Shortcut to specify that the workbook should use shared strings
+    # @see Workbook#use_shared_strings
+    def use_shared_strings=(v)
+      Axlsx::validate_boolean(v);
+      workbook.use_shared_strings = v
+    end
+    # The workbook this package will serialize or validate.
+    # @return [Workbook] If no workbook instance has been assigned with this package a new Workbook instance is returned.
+    # @raise ArgumentError if workbook parameter is not a Workbook instance.
+    # @note As there are multiple ways to instantiate a workbook for the package,
+    #   here are a few examples:
+    #     # assign directly during package instanciation
+    #     wb = Package.new(:workbook => Workbook.new).workbook
+    #
+    #     # get a fresh workbook automatically from the package
+    #     wb = Pacakge.new().workbook
+    #     #     # set the workbook after creating the package
+    #     wb = Package.new().workbook = Workbook.new
+    def workbook
+      @workbook || @workbook = Workbook.new
+      yield @workbook if block_given?
+      @workbook
+    end
+
+    #def self.parse(input, confirm_valid = false)
+    #  p = Package.new
+    #  z = Zip::File.open(input)
+    #  p.workbook = Workbook.parse z.get_entry(WORKBOOK_PN)
+    #  p
+    #end
+
+    # @see workbook
+    def workbook=(workbook) DataTypeValidator.validate :Package_workbook, Workbook, workbook; @workbook = workbook; end
+
+    # Serialize your workbook to disk as an xlsx document.
+    #
+    # @param [String] output The name of the file you want to serialize your package to
+    # @param [Boolean] confirm_valid Validate the package prior to serialization.
+    # @return [Boolean] False if confirm_valid and validation errors exist. True if the package was serialized
+    # @note A tremendous amount of effort has gone into ensuring that you cannot create invalid xlsx documents.
+    #   confirm_valid should be used in the rare case that you cannot open the serialized file.
+    # @see Package#validate
+    # @example
+    #   # This is how easy it is to create a valid xlsx file. Of course you might want to add a sheet or two, and maybe some data, styles and charts.
+    #   # Take a look at the README for an example of how to do it!
+    #
+    #   #serialize to a file
+    #   p = Axlsx::Package.new
+    #   # ......add cool stuff to your workbook......
+    #   p.serialize("example.xlsx")
+    #
+    #   # Serialize to a stream
+    #   s = p.to_stream()
+    #   File.open('example_streamed.xlsx', 'w') { |f| f.write(s.read) }
+    def serialize(output, confirm_valid=false)
+      return false unless !confirm_valid || self.validate.empty?
+      Relationship.clear_cached_instances
+      Zip::OutputStream.open(output) do |zip|
+        write_parts(zip)
+      end
+      true
+    end
+
+
+    # Serialize your workbook to a StringIO instance
+    # @param [Boolean] confirm_valid Validate the package prior to serialization.
+    # @return [StringIO|Boolean] False if confirm_valid and validation errors exist. rewound string IO if not.
+    def to_stream(confirm_valid=false)
+      return false unless !confirm_valid || self.validate.empty?
+      Relationship.clear_cached_instances
+      zip = write_parts(Zip::OutputStream.new(StringIO.new, true))
+      stream = zip.close_buffer
+      stream.rewind
+      stream
+    end
+
+    # Encrypt the package into a CFB using the password provided
+    # This is not ready yet
+    def encrypt(file_name, password)
+      return false
+      # moc = MsOffCrypto.new(file_name, password)
+      # moc.save
+    end
+
+    # Validate all parts of the package against xsd schema.
+    # @return [Array] An array of all validation errors found.
+    # @note This gem includes all schema from OfficeOpenXML-XMLSchema-Transitional.zip and OpenPackagingConventions-XMLSchema.zip
+    #   as per ECMA-376, Third edition. opc schema require an internet connection to import remote schema from dublin core for dc,
+    #   dcterms and xml namespaces. Those remote schema are included in this gem, and the original files have been altered to
+    #   refer to the local versions.
+    #
+    #   If by chance you are able to creat a package that does not validate it indicates that the internal
+    #   validation is not robust enough and needs to be improved. Please report your errors to the gem author.
+    # @see http://www.ecma-international.org/publications/standards/Ecma-376.htm
+    # @example
+    #  # The following will output any error messages found in serialization.
+    #  p = Axlsx::Package.new
+    #  # ... code to create sheets, charts, styles etc.
+    #  p.validate.each { |error| puts error.message }
+    def validate
+      errors = []
+      parts.each do |part|
+        unless part[:schema].nil?
+          if part[:doc].is_a? String
+            errors.concat validate_single_doc(part[:schema], part[:doc])
+          else
+            errors.concat validate_single_doc(part[:schema], part[:doc].to_xml_string)
+          end
+        end
+      end
+      errors
+    end
+
+    private
+
+    # Writes the package parts to a zip archive.
+    # @param [Zip::OutputStream] zip
+    # @return [Zip::OutputStream]
+    def write_parts(zip)
+      p = parts
+      p.each do |part|
+        #next unless part[:entry] == CORE_PN
+        unless part[:doc].nil?
+          zip.put_next_entry(zip_entry_for_part(part))
+          if part[:doc].is_a? String
+            entry = ['1.9.2', '1.9.3'].include?(RUBY_VERSION) ? part[:doc].force_encoding('BINARY') : part[:doc]
+            zip.puts(entry)
+          else
+            part[:doc].to_xml_string(zip)
+          end
+        end
+        unless part[:path].nil?
+          zip.put_next_entry(zip_entry_for_part(part))
+          # binread for 1.9.3
+          zip.write IO.respond_to?(:binread) ? IO.binread(part[:path]) : IO.read(part[:path])
+        end
+      end
+      zip
+    end
+
+    # Generate a Entry for the given package part.
+    # The important part here is to explicitly set the timestamp for the zip entry: Serializing axlsx packages
+    # with identical contents should result in identical zip files – however, the timestamp of a zip entry
+    # defaults to the time of serialization and therefore the zip file contents would be different every time
+    # the package is serialized.
+    #
+    # Note: {Core#created} also defaults to the current time – so to generate identical axlsx packages you have
+    # to set this explicitly, too (eg. with `Package.new(created_at: Time.local(2013, 1, 1))`).
+    #
+    # @param part A hash describing a part of this pacakge (see {#parts})
+    # @return [Zip::Entry]
+    def zip_entry_for_part(part)
+      timestamp = Zip::DOSTime.at(@core.created.to_i)
+      Zip::Entry.new("", part[:entry], "", "", 0, 0, Zip::Entry::DEFLATED, 0, timestamp)
+    end
+
+    # The parts of a package
+    # @return [Array] An array of hashes that define the entry, document and schema for each part of the package.
+    # @private
+    def parts
+      parts = [
+       {:entry => RELS_PN, :doc => relationships, :schema => RELS_XSD},
+       {:entry => "xl/#{STYLES_PN}", :doc => workbook.styles, :schema => SML_XSD},
+       {:entry => CORE_PN, :doc => @core, :schema => CORE_XSD},
+       {:entry => APP_PN, :doc => @app, :schema => APP_XSD},
+       {:entry => WORKBOOK_RELS_PN, :doc => workbook.relationships, :schema => RELS_XSD},
+       {:entry => CONTENT_TYPES_PN, :doc => content_types, :schema => CONTENT_TYPES_XSD},
+       {:entry => WORKBOOK_PN, :doc => workbook, :schema => SML_XSD}
+      ]
+
+      workbook.drawings.each do |drawing|
+        parts << {:entry => "xl/#{drawing.rels_pn}", :doc => drawing.relationships, :schema => RELS_XSD}
+        parts << {:entry => "xl/#{drawing.pn}", :doc => drawing, :schema => DRAWING_XSD}
+      end
+
+
+      workbook.tables.each do |table|
+        parts << {:entry => "xl/#{table.pn}", :doc => table, :schema => SML_XSD}
+      end
+      workbook.pivot_tables.each do |pivot_table|
+        cache_definition = pivot_table.cache_definition
+        parts << {:entry => "xl/#{pivot_table.rels_pn}", :doc => pivot_table.relationships, :schema => RELS_XSD}
+        parts << {:entry => "xl/#{pivot_table.pn}", :doc => pivot_table} #, :schema => SML_XSD}
+        parts << {:entry => "xl/#{cache_definition.pn}", :doc => cache_definition} #, :schema => SML_XSD}
+      end
+
+      workbook.comments.each do|comment|
+        if comment.size > 0
+          parts << { :entry => "xl/#{comment.pn}", :doc => comment, :schema => SML_XSD }
+          parts << { :entry => "xl/#{comment.vml_drawing.pn}", :doc => comment.vml_drawing, :schema => nil }
+        end
+      end
+
+      workbook.charts.each do |chart|
+        parts << {:entry => "xl/#{chart.pn}", :doc => chart, :schema => DRAWING_XSD}
+      end
+
+      workbook.images.each do |image|
+        parts << {:entry => "xl/#{image.pn}", :path => image.image_src}
+      end
+
+      if use_shared_strings
+        parts << {:entry => "xl/#{SHARED_STRINGS_PN}", :doc => workbook.shared_strings, :schema => SML_XSD}
+      end
+
+      workbook.worksheets.each do |sheet|
+        parts << {:entry => "xl/#{sheet.rels_pn}", :doc => sheet.relationships, :schema => RELS_XSD}
+        parts << {:entry => "xl/#{sheet.pn}", :doc => sheet, :schema => SML_XSD}
+      end
+      parts
+    end
+
+    # Performs xsd validation for a signle document
+    #
+    # @param [String] schema path to the xsd schema to be used in validation.
+    # @param [String] doc The xml text to be validated
+    # @return [Array] An array of all validation errors encountered.
+    # @private
+    def validate_single_doc(schema, doc)
+      schema = Nokogiri::XML::Schema(File.open(schema))
+      doc = Nokogiri::XML(doc)
+      errors = []
+      schema.validate(doc).each do |error|
+        errors << error
+      end
+      errors
+    end
+
+    # Appends override objects for drawings, charts, and sheets as they exist in your workbook to the default content types.
+    # @return [ContentType]
+    # @private
+    def content_types
+      c_types = base_content_types
+      workbook.drawings.each do |drawing|
+        c_types << Axlsx::Override.new(:PartName => "/xl/#{drawing.pn}",
+                                       :ContentType => DRAWING_CT)
+      end
+
+      workbook.charts.each do |chart|
+        c_types << Axlsx::Override.new(:PartName => "/xl/#{chart.pn}",
+                                       :ContentType => CHART_CT)
+      end
+
+      workbook.tables.each do |table|
+        c_types << Axlsx::Override.new(:PartName => "/xl/#{table.pn}",
+                                       :ContentType => TABLE_CT)
+      end
+
+      workbook.pivot_tables.each do |pivot_table|
+        c_types << Axlsx::Override.new(:PartName => "/xl/#{pivot_table.pn}",
+                                       :ContentType => PIVOT_TABLE_CT)
+        c_types << Axlsx::Override.new(:PartName => "/xl/#{pivot_table.cache_definition.pn}",
+                                       :ContentType => PIVOT_TABLE_CACHE_DEFINITION_CT)
+      end
+
+      workbook.comments.each do |comment|
+        if comment.size > 0
+        c_types << Axlsx::Override.new(:PartName => "/xl/#{comment.pn}",
+                                       :ContentType => COMMENT_CT)
+        end
+      end
+
+      if workbook.comments.size > 0
+        c_types << Axlsx::Default.new(:Extension => "vml", :ContentType => VML_DRAWING_CT)
+      end
+
+      workbook.worksheets.each do |sheet|
+        c_types << Axlsx::Override.new(:PartName => "/xl/#{sheet.pn}",
+                                         :ContentType => WORKSHEET_CT)
+      end
+      exts = workbook.images.map { |image| image.extname.downcase }
+      exts.uniq.each do |ext|
+        ct = if  ['jpeg', 'jpg'].include?(ext)
+               JPEG_CT
+             elsif ext == 'gif'
+               GIF_CT
+             elsif ext == 'png'
+               PNG_CT
+             end
+        c_types << Axlsx::Default.new(:ContentType => ct, :Extension => ext )
+      end
+      if use_shared_strings
+        c_types << Axlsx::Override.new(:PartName => "/xl/#{SHARED_STRINGS_PN}",
+                                       :ContentType => SHARED_STRINGS_CT)
+      end
+      c_types
+    end
+
+    # Creates the minimum content types for generating a valid xlsx document.
+    # @return [ContentType]
+    # @private
+    def base_content_types
+      c_types = ContentType.new()
+      c_types << Default.new(:ContentType => RELS_CT, :Extension => RELS_EX)
+      c_types << Default.new(:Extension => XML_EX, :ContentType => XML_CT)
+      c_types << Override.new(:PartName => "/#{APP_PN}", :ContentType => APP_CT)
+      c_types << Override.new(:PartName => "/#{CORE_PN}", :ContentType => CORE_CT)
+      c_types << Override.new(:PartName => "/xl/#{STYLES_PN}", :ContentType => STYLES_CT)
+      c_types << Axlsx::Override.new(:PartName => "/#{WORKBOOK_PN}", :ContentType => WORKBOOK_CT)
+      c_types.lock
+      c_types
+    end
+
+    # Creates the relationships required for a valid xlsx document
+    # @return [Relationships]
+    # @private
+    def relationships
+      rels = Axlsx::Relationships.new
+      rels << Relationship.new(self, WORKBOOK_R, WORKBOOK_PN)
+      rels << Relationship.new(self, CORE_R, CORE_PN)
+      rels << Relationship.new(self, APP_R, APP_PN)
+      rels.lock
+      rels
+    end
+  end
+end
+

data/lib/axlsx/rels/relationship.rb

@@ -0,0 +1,129 @@
+# encoding: UTF-8
+module Axlsx
+  # A relationship defines a reference between package parts.
+  # @note Packages automatically manage relationships.
+  class Relationship
+    
+    class << self
+      # Keeps track of all instances of this class.
+      # @return [Array]
+      def instances
+        @instances ||= []
+      end
+      
+      # Clear cached instances.
+      # 
+      # This should be called before serializing a package (see {Package#serialize} and
+      # {Package#to_stream}) to make sure that serialization is idempotent (i.e. 
+      # Relationship instances are generated with the same IDs everytime the package
+      # is serialized).
+      # 
+      # Also, calling this avoids memory leaks (cached instances lingering around 
+      # forever). 
+      def clear_cached_instances
+        @instances = []
+      end
+      
+      # Generate and return a unique id (eg. `rId123`) Used for setting {#Id}. 
+      #
+      # The generated id depends on the number of cached instances, so using
+      # {clear_cached_instances} will automatically reset the generated ids, too.
+      # @return [String]
+      def next_free_id
+        "rId#{@instances.size + 1}"
+      end
+    end
+
+    # The id of the relationship (eg. "rId123"). Most instances get their own unique id. 
+    # However, some instances need to share the same id – see {#should_use_same_id_as?}
+    # for details.
+    # @return [String]
+    attr_reader :Id
+    
+    # The location of the relationship target
+    # @return [String]
+    attr_reader :Target
+
+    # The type of relationship
+    # @note Supported types are defined as constants in Axlsx:
+    # @see XML_NS_R
+    # @see TABLE_R
+    # @see PIVOT_TABLE_R
+    # @see WORKBOOK_R
+    # @see WORKSHEET_R
+    # @see APP_R
+    # @see RELS_R
+    # @see CORE_R
+    # @see STYLES_R
+    # @see CHART_R
+    # @see DRAWING_R
+    # @return [String]
+    attr_reader :Type
+
+    # The target mode of the relationship
+    # used for hyperlink type relationships to mark the relationship to an external resource
+    # TargetMode can be specified during initialization by passing in a :target_mode option
+    # Target mode must be :external for now.
+    attr_reader :TargetMode
+
+    # The source object the relations belongs to (e.g. a hyperlink, drawing, ...). Needed when
+    # looking up the relationship for a specific object (see {Relationships#for}).
+    attr_reader :source_obj
+    
+    # Initializes a new relationship. 
+    # @param [Object] source_obj see {#source_obj}
+    # @param [String] type The type of the relationship
+    # @param [String] target The target for the relationship
+    # @option [Symbol] :target_mode only accepts :external.
+    def initialize(source_obj, type, target, options={})
+      @source_obj = source_obj
+      self.Target=target
+      self.Type=type
+      self.TargetMode = options[:target_mode] if options[:target_mode]
+      @Id = if (existing = self.class.instances.find{ |i| should_use_same_id_as?(i) })
+        existing.Id
+      else
+        self.class.next_free_id
+      end
+      self.class.instances << self
+    end
+
+    # @see Target
+    def Target=(v) Axlsx::validate_string v; @Target = v end
+    # @see Type
+    def Type=(v) Axlsx::validate_relationship_type v; @Type = v end
+
+    # @see TargetMode
+    def TargetMode=(v) RestrictionValidator.validate 'Relationship.TargetMode', [:External, :Internal], v; @TargetMode = v; end
+
+    # serialize relationship
+    # @param [String] str
+    # @return [String]
+    def to_xml_string(str = '')
+      h = self.instance_values.reject{|k, _| k == "source_obj"}
+      str << '<Relationship '
+      str << (h.map { |key, value| '' << key.to_s << '="' << Axlsx::coder.encode(value.to_s) << '"'}.join(' '))
+      str << '/>'
+    end
+    
+    # Whether this relationship should use the same id as `other`.
+    #
+    # Instances designating the same relationship need to use the same id. We can not simply
+    # compare the {#Target} attribute, though: `foo/bar.xml`, `../foo/bar.xml`, 
+    # `../../foo/bar.xml` etc. are all different but probably mean the same file (this 
+    # is especially an issue for relationships in the context of pivot tables). So lets
+    # just ignore this attribute for now (except when {#TargetMode} is set to `:External` –
+    # then {#Target} will be an absolute URL and thus can safely be compared).
+    #
+    # @todo Implement comparison of {#Target} based on normalized path names.
+    # @param other [Relationship]
+    def should_use_same_id_as?(other)
+      result = self.source_obj == other.source_obj && self.Type == other.Type && self.TargetMode == other.TargetMode
+      if self.TargetMode == :External
+        result &&= self.Target == other.Target
+      end
+      result
+    end
+    
+  end
+end