hexapdf 0.32.2 → 0.34.0

data/lib/hexapdf/document/layout.rb

@@ -71,21 +71,20 @@ module HexaPDF
     #
     # == Box Styles
     #
-    # All box creation methods accept HexaPDF::Layout::Style objects or names for style objects
-    # (defined via #style). This allows one to predefine certain styles (like first level heading,
-    # second level heading, paragraph, ...) and consistently use them throughout the document
-    # creation process.
+    # All box creation methods accept Layout::Style objects or names for style objects (defined via
+    # #style). This allows one to predefine certain styles (like first level heading, second level
+    # heading, paragraph, ...) and consistently use them throughout the document creation process.
     #
-    # One style property, HexaPDF::Layout::Style#font, is handled specially:
+    # One style property, Layout::Style#font, is handled specially:
     #
     # * If no font is set on a style, the font "Times" is automatically set because otherwise there
     #   would be problems with text drawing operations (font is the only style property that has no
     #   valid default value).
     #
-    # * Standard style objects only allow font wrapper objects to be set via the
-    #   HexaPDF::Layout::Style#font method. This class makes usage easier by allowing strings or an
-    #   array [name, options_hash] to be used, like with e.g Content::Canvas#font. So to use
-    #   Helvetica as font, one could just do:
+    # * Standard style objects only allow font wrapper objects to be set via the Layout::Style#font
+    #   method. This class makes usage easier by allowing strings or an array [name, options_hash]
+    #   to be used, like with e.g Content::Canvas#font. So to use Helvetica as font, one could just
+    #   do:
     #
     #     style.font = 'Helvetica'
     #
@@ -108,11 +107,14 @@ module HexaPDF
       # text, formatted_text, image, ...:: Any method accepted by the Layout class without the _box
       #                                    suffix.
       #
-      # list, column, ...:: Any name registered for the configuration option +layout.boxes.map+.
+      # list, column, ...:: Any name registered with the configuration option +layout.boxes.map+.
+      #
+      # The special method #multiple allows adding multiple boxes as a single array to the collected
+      # children.
       #
       # Example:
       #
-      #   document.layout.box(:list) do |list|
+      #   document.layout.box(:list) do |list|  # list is a ChildrenCollector
       #     list.text_box("Some text here")     # layout method
       #     list.image(image_path)              # layout method without _box suffix
       #     list.column(columns: 3) do |column| # registered box name
@@ -166,7 +168,7 @@ module HexaPDF
 
       end
 
-      # The mapping of style name (a Symbol) to HexaPDF::Layout::Style instance.
+      # The mapping of style name (a Symbol) to Layout::Style instance.
       attr_reader :styles
 
       # Creates a new Layout object for the given PDF document.
@@ -179,20 +181,20 @@ module HexaPDF
       #    layout.style(name)                              -> style
       #    layout.style(name, base: :base, **properties)   -> style
       #
-      # Creates or updates the HexaPDF::Layout::Style object called +name+ with the given property
-      # values and returns it.
+      # Creates or updates the Layout::Style object called +name+ with the given property values and
+      # returns it.
+      #
+      # If neither +base+ nor any style properties are specified, the style +name+ is just returned.
       #
       # This method allows convenient access to the stored styles and to update them. Such styles
       # can then be used by name in the various box creation methods, e.g. #text_box or #image_box.
       #
-      # If neither +base+ nor any style properties are specified, the style +name+ is just returned.
-      #
       # If the style +name+ does not exist yet and the argument +base+ specifies the name of another
       # style, that style is duplicated and used as basis for the style. This also means that the
       # referenced +base+ style needs be defined first!
       #
-      # The special name :base should be used for setting the base style which is used when no
-      # specific style is set.
+      # The special name :base should be used for setting the base style which is used for the
+      # +base+ argument when no specific style is specified.
       #
       # Note that the style property 'font' is handled specially, see the class documentation for
       # details.
@@ -269,26 +271,26 @@ module HexaPDF
       #
       # +style+, +style_properties+::
       #     The box and the text are styled using the given +style+. This can either be a style name
-      #     set via #style or anything HexaPDF::Layout::Style::create accepts. If any additional
+      #     set via #style or anything Layout::Style::create accepts. If any additional
       #     +style_properties+ are specified, the style is duplicated and the additional styles are
       #     applied.
       #
       # +properties+::
-      #     This can be used to set custom properties on the created text box. See Box#properties
-      #     for details and usage.
+      #     This can be used to set custom properties on the created text box. See
+      #     Layout::Box#properties for details and usage.
       #
       # +box_style+::
       #     Sometimes it is necessary for the box to have a different style than the text, e.g. when
       #     using overlays. In such a case use +box_style+ for specifiying the style of the box (a
-      #     style name set via #style or anything HexaPDF::Layout::Style::create accepts).
+      #     style name set via #style or anything Layout::Style::create accepts).
       #
       #     The +style+ together with the +style_properties+ will be used for the text style.
       #
       # Examples:
       #
-      #   layout.text_box("Test " * 15)
+      #   layout.text_box("Test is on " * 15)
       #   layout.text_box("Now " * 7, width: 100)
-      #   layout.text_box("Another test", font_size: 15, fill_color: "green")
+      #   layout.text_box("Another test", font_size: 15, fill_color: "hp-blue")
       #   layout.text_box("Different box style", fill_color: 'white', box_style: {
       #     underlays: [->(c, b) { c.rectangle(0, 0, b.content_width, b.content_height).fill }]
       #   })
@@ -307,7 +309,8 @@ module HexaPDF
       # formatted differently.
       #
       # The argument +data+ needs to be an array of String, HexaPDF::Layout::InlineBox and/or Hash
-      # objects and is transformed so that it is suitable as argument for the text box:
+      # objects and is transformed so that it is suitable as argument for the text box
+      # initialization method.
       #
       # * A String object is treated like {text: data}.
       #
@@ -322,9 +325,9 @@ module HexaPDF
       #          for the text. If this is set and :box is not, the hash will be transformed into a
       #          text fragment with an appropriate link overlay.
       #
-      #   style:: The style to be use as base style instead of the style created from the +style+
-      #           and +style_properties+ arguments. See HexaPDF::Layout::Style::create for allowed
-      #           values.
+      #   style:: The style to use as base style instead of the style created from the +style+ and
+      #           +style_properties+ arguments. This can either be a style name set via #style or
+      #           anything HexaPDF::Layout::Style::create allows.
       #
       #           If any style properties are set, the used style is duplicated and the additional
       #           properties applied.
@@ -336,23 +339,42 @@ module HexaPDF
       #
       #         The value must be one or more (as an array) positional arguments to be used with the
       #         #inline_box method. The rest of the hash keys are passed as keyword arguments to
-      #         #inline_box except for :block that value of which would be passed as the block.
+      #         #inline_box except for :block which would be passed as the block.
       #
       # See #text_box for details on +width+, +height+, +style+, +style_properties+, +properties+
       # and +box_style+.
       #
       # Examples:
       #
+      #   # Text without special styling
       #   layout.formatted_text_box(["Some string"])
-      #   layout.formatted_text_box(["Some ", {text: "string", fill_color: 128}])
-      #   layout.formatted_text_box(["Some ", {link: "https://example.com",
-      #                                        fill_color: 'blue', text: "Example"}])
-      #   layout.formatted_text_box(["Some ", {text: "string", style: {font_size: 20}}])
-      #   layout.formatted_text_box(["Some ", {box: [:text, "string"], valign: :top}])
+      #
+      #   # A predefined inline box
+      #   ibox = layout.inline_box(:text, 'Hello')
+      #   layout.formatted_text_box([ibox])
+      #
+      #   # Text with styling properties
+      #   layout.formatted_text_box([{text: "string", fill_color: 128}])
+      #
+      #   # Text referencing a base style
+      #   layout.formatted_text_box([{text: "string", style: :bold}])
+      #
+      #   # Text with a link
+      #   layout.formatted_text_box([{link: "https://example.com",
+      #                               fill_color: 'blue', text: "Example"}])
+      #
+      #   # Inline boxes created from the given data
+      #   layout.formatted_text_box([{box: [:text, "string"], valign: :top}])
       #   block = lambda {|list| list.text("First item"); list.text("Second item") }
       #   layout.formatted_text_box(["Some ", {box: :list, item_spacing: 10, block: block}])
       #
-      # See: #text_box, HexaPDF::Layout::TextBox, HexaPDF::Layout::TextFragment
+      #   # Combining the above variants
+      #   layout.formatted_text_box(["Hello", {box: [:text, 'World!']}, "Here comes a ",
+      #                             {link: 'https://example.com', text: 'link'}, '!',
+      #                             {text: 'And more!', style: :bold, font_size: 20}])
+      #
+      # See: #text_box, #inline_box, HexaPDF::Layout::TextBox, HexaPDF::Layout::TextFragment,
+      # HexaPDF::Layout::InlineBox
       def formatted_text_box(data, width: 0, height: 0, style: nil, properties: nil, box_style: nil,
                              **style_properties)
         style = retrieve_style(style, style_properties)
@@ -406,25 +428,131 @@ module HexaPDF
                                        properties: properties, style: style)
       end
 
+      # This helper class is used by Layout#table_box to allow specifying the keyword arguments used
+      # when converting cell data to box instances.
+      class CellArgumentCollector
+
+        # Stores a single keyword argument definition for a number of rows/columns.
+        ArgumentInfo = Struct.new(:rows, :cols, :args)
+
+        # Returns all stored ArgumentInfo instances.
+        attr_reader :argument_infos
+
+        # Creates a new instance, providing the number of rows and columns of the table.
+        def initialize(number_of_rows, number_of_columns)
+          @argument_infos = []
+          @number_of_rows = number_of_rows
+          @number_of_columns = number_of_columns
+        end
+
+        # Stores the keyword arguments in +args+ for the given 0-based rows and columns which can
+        # either be a single number or a range of numbers.
+        def []=(rows = 0..-1, cols = 0..-1, args)
+          rows = adjust_range(rows.kind_of?(Integer) ? rows..rows : rows, @number_of_rows)
+          cols = adjust_range(cols.kind_of?(Integer) ? cols..cols : cols, @number_of_columns)
+          @argument_infos << ArgumentInfo.new(rows, cols, args)
+        end
+
+        # Retrieves the merged keyword arguments for the cell in +row+ and +col+.
+        #
+        # Earlier defined arguments are overridden by later ones.
+        def retrieve_arguments_for(row, col)
+          @argument_infos.each_with_object({}) do |arg_info, result|
+            next unless arg_info.rows.cover?(row) && arg_info.cols.cover?(col)
+            result.update(arg_info.args)
+          end
+        end
+
+        private
+
+        # Adjusts the +range+ so that both the begin and the end of the range are zero or positive
+        # integers smaller than +max+.
+        def adjust_range(range, max)
+          (range.begin % max)..(range.end % max)
+        end
+
+      end
+
+      # Creates a HexaPDF::Layout::TableBox for the given table data.
+      #
+      # This method is a small wrapper around the actual class and mainly facilitates transforming
+      # the contents of the +data+ into the box instances needed by the table box implementation.
+      #
+      # In addition to everything the table box implementation allows for +data+, it is also
+      # possible to specify strings as cell contents. Those strings will be converted to text boxes
+      # by using the #text_box method. *Note* that this functionality is *not* available for the
+      # header and footer!
+      #
+      # Additional arguments for the #text_box invocations can be specified using the optional block
+      # that yields a CellArgumentCollector instance. This allows customization of the text boxes.
+      # By specifying the special key +:cell+ it is also possible to assign style properties to the
+      # cells themselves.
+      #
+      # See HexaPDF::Layout::TableBox::new for details on +column_widths+, +header+, +footer+, and
+      # +cell_style+.
+      #
+      # See #text_box for details on +width+, +height+, +style+, +style_properties+ and
+      # +properties+.
+      #
+      # Examples:
+      #
+      #   layout.table_box([[layout.text('A'), layout.text('B')],
+      #                     [layout.image(image_path), layout.text('D')]]
+      #   layout.table_box([['A', 'B'], [layout.image(image_path), 'D]])     # same as above
+      #
+      #   layout.table_box([['A', 'B'], ['C', 'D]]) do |args|
+      #     # assign the predefined style :cell_text to all texts
+      #     args[] = {style: :cell_text}
+      #     # row 0 has a grey background and bold text
+      #     args[0] = {font: ['Helvetica', variant: :bold], cell: {background_color: 'eee'}}
+      #     # text in last column is right aligned
+      #     args[0..-1, -1] = {align: :right}
+      #   end
+      #
+      # See: HexaPDF::Layout::TableBox
+      def table_box(data, column_widths: nil, header: nil, footer: nil, cell_style: nil,
+                    width: 0, height: 0, style: nil, properties: nil, **style_properties)
+        style = retrieve_style(style, style_properties)
+        cells = HexaPDF::Layout::TableBox::Cells.new(data, cell_style: cell_style)
+        collector = CellArgumentCollector.new(cells.number_of_rows, cells.number_of_columns)
+        yield(collector) if block_given?
+        cells.style do |cell|
+          args = collector.retrieve_arguments_for(cell.row, cell.column)
+          cstyle = args.delete(:cell)
+          result = case cell.children
+                   when Array, HexaPDF::Layout::Box
+                     cell.children
+                   else
+                     text_box(cell.children.to_s, **args)
+                   end
+          cell.children = result
+          cell.style.update(**cstyle) if cstyle
+        end
+        box_class_for_name(:table).new(cells: cells, column_widths: column_widths, header: header,
+                                       footer: footer, cell_style: cell_style, width: width,
+                                       height: height, properties: properties, style: style)
+      end
+
       LOREM_IPSUM = [ # :nodoc:
         "Lorem ipsum dolor sit amet, con\u{00AD}sectetur adipis\u{00AD}cing elit, sed " \
           "do eiusmod tempor incididunt ut labore et dolore magna aliqua.",
         "Ut enim ad minim veniam, quis nostrud exer\u{00AD}citation ullamco laboris nisi ut " \
-          "aliquip ex ea commodo consequat. ",
+          "aliquip ex ea commodo consequat.",
         "Duis aute irure dolor in reprehen\u{00AD}derit in voluptate velit esse cillum dolore " \
-          "eu fugiat nulla pariatur. ",
+          "eu fugiat nulla pariatur.",
         "Excepteur sint occaecat cupidatat non proident, sunt in culpa qui officia deserunt " \
           "mollit anim id est laborum.",
       ]
 
-      # Uses #text_box to create +count+ paragraphs of lorem ipsum text.
+      # Uses #text_box to create +count+ paragraphs with +sentences+ number of sentences (1 to 4) of
+      # lorem ipsum text.
       #
       # The +text_box_properties+ arguments are passed as is to #text_box.
       def lorem_ipsum_box(sentences: 4, count: 1, **text_box_properties)
         text_box(([LOREM_IPSUM[0, sentences].join(" ")] * count).join("\n\n"), **text_box_properties)
       end
 
-      BOX_METHOD_NAMES = [:text, :formatted_text, :image, :lorem_ipsum] #:nodoc:
+      BOX_METHOD_NAMES = [:text, :formatted_text, :image, :table, :lorem_ipsum] #:nodoc:
 
       # Allows creating boxes using more convenient method names:
       #
@@ -479,6 +607,9 @@ module HexaPDF
       # Finally, a default font is set if necessary to ensure that the style object works in all
       # cases.
       def retrieve_style(style, properties = nil)
+        if style.kind_of?(Symbol) && !@styles.key?(style)
+          raise HexaPDF::Error, "Style #{style} not defined"
+        end
         style = HexaPDF::Layout::Style.create(@styles[style] || style || @styles[:base])
         style = style.dup.update(**properties) unless properties.nil? || properties.empty?
         style.font('Times') unless style.font?

data/lib/hexapdf/document/pages.rb

@@ -39,7 +39,7 @@ require 'hexapdf/error'
 module HexaPDF
   class Document
 
-    # This class provides methods for managing the pages of a PDF file.
+    # This class provides methods for managing the pages and page labels of a PDF file.
     #
     # For page manipulation it uses the methods of HexaPDF::Type::PageTreeNode underneath but
     # provides a more convenient interface.
@@ -178,7 +178,8 @@ module HexaPDF
         @document.catalog.pages.each_page(&block)
       end
 
-      # Returns the number of pages in the PDF document. May be zero if the document has no pages.
+      # Returns the number of pages in the PDF document. May be zero if the document has no pages
+      # yet.
       def count
         @document.catalog.pages.page_count
       end
@@ -245,7 +246,7 @@ module HexaPDF
 
         labels = @document.catalog.page_labels(create: true)
         labels.add_entry(start_index, page_label)
-        labels.add_entry(0, {S: :d}) unless labels.find_entry(0)
+        labels.add_entry(0, {S: :D}) unless labels.find_entry(0)
 
         page_label
       end