prawn-format 0.1.0

data/manual/pdf.rb

@@ -0,0 +1,204 @@
+$LOAD_PATH.unshift "#{File.dirname(__FILE__)}/../lib"
+require 'prawn'
+require 'prawn/format'
+require 'prawn/format/version'
+require 'coderay'
+
+def process_style(document, style, content, line_number)
+  content = process_substitutions(content)
+
+  case style
+  when "h1"        then h1(document, content)
+  when "h2"        then h2(document, content)
+  when "p"         then paragraph(document, content)
+  when "fp"        then paragraph(document, content, false)
+  when "ul"        then start_list(document)
+  when "li"        then list_item(document, content)
+  when "/ul"       then end_list(document)
+  when "page"      then new_page(document)
+  when "highlight" then highlight(document, content)
+  when "hr"        then horiz_rule(document)
+  when "center"    then center(document, content)
+  else warn "unknown style #{style.inspect}"
+  end
+
+rescue Exception => err
+  puts "[error occurred while processing line ##{line_number}]"
+  raise
+end
+
+def process_substitutions(content)
+  content.
+    gsub(/%FORMAT:VERSION%/, Prawn::Format::Version::STRING).
+    gsub(/%NOW%/, Time.now.utc.strftime("%e %B %Y at %H:%M UTC")).
+    gsub(/%PDF\{(.*?)\}HTML\{(.*?)\}END%/, '\\1')
+end
+
+def center(document, content)
+  padding(document, :clear => true)
+  document.text(content, :plain => false, :align => :center)
+  padding(document)
+end
+
+def horiz_rule(document)
+  padding(document, :clear => true)
+  document.stroke_color "000000"
+  document.stroke_horizontal_rule
+  padding(document)
+end
+
+def h1(document, content)
+  clear_padding!
+  document.text "<h1>#{content}</h1>"
+  document.stroke_color "000080"
+  document.stroke_horizontal_rule
+  padding(document, :size => document.font_size * 2)
+end
+
+def h2(document, content)
+  clear_padding!
+  document.text "<h2>#{content}</h2>"
+  document.stroke_color "000080"
+  document.stroke_horizontal_rule
+  padding(document, :size => document.font_size * 2)
+end
+
+def paragraph(document, content, indent=true)
+  return unless content.strip.length > 0
+  clear_padding!
+  document.text "#{content}", :align => :justify
+  padding(document, :size => document.font_size / 2)
+end
+
+def start_list(document)
+  padding(document)
+end
+
+def list_item(document, content)
+  clear_padding!
+
+  indent_b = document.font_size * 3
+  indent   = document.font_size * 4
+
+  document.start_new_page if document.y < document.font_size
+  y = document.y - document.bounds.absolute_bottom
+  document.text "&bull;", :at => [indent_b, y - document.font.ascender]
+  document.layout(content, :align => :justify) do |helper|
+    while !helper.done?
+      y = helper.fill(indent, y, document.bounds.width-indent, :height => document.y)
+      if helper.done?
+        document.y = y + document.bounds.absolute_bottom
+      else
+        document.start_new_page
+        y = document.y - document.bounds.absolute_bottom
+      end
+    end
+  end
+
+  padding(document, :size => document.font_size / 4)
+end
+
+def end_list(document)
+  padding(document)
+end
+
+def new_page(document)
+  document.start_new_page
+  clear_padding!
+end
+
+def highlight(document, content)
+  file, syntax = content.split(/,/)
+  analyzed = CodeRay.scan(File.read(File.join(File.dirname(__FILE__), file)), syntax.to_sym)
+  html = "<pre>" + analyzed.html + "</pre>"
+
+  padding(document, :size => document.font_size * 2, :clear => true)
+  y = document.y - document.bounds.absolute_bottom
+  start_y = y + document.font_size
+
+  indent = document.font_size * 2
+
+  document.layout(html) do |helper|
+    while !helper.done?
+      y = helper.fill(indent, y, document.bounds.width-indent*2, :height => document.y)
+      if helper.done?
+        document.y = y + document.bounds.absolute_bottom
+      else
+        document.start_new_page
+        y = document.y - document.bounds.absolute_bottom
+      end
+    end
+  end
+
+  document.stroke_color "a0a0a0"
+  document.rectangle [document.font_size, start_y], bounds.width - document.font_size*2, (start_y - y)
+  document.stroke
+
+  padding(document, :size => document.font_size)
+end
+
+def clear_padding!
+  @last_padding = nil
+end
+
+# This padding stuff is mostly here just so that adjacent vertical spaces
+# will collapse.
+def padding(document, options={})
+  size = options[:size] || document.font_size
+
+  if @last_padding
+    full_size = [@last_padding, size].max
+    size = full_size - @last_padding
+    @last_padding = full_size
+  else
+    @last_padding = size
+  end
+
+  document.y -= size
+  clear_padding! if options[:clear]
+end
+
+SERIF_FONT  = "/Library/Fonts/Baskerville.dfont"
+
+Prawn::Document.generate("prawn-format.pdf", :compress => true) do
+  if File.exists?(SERIF_FONT)
+    font_families["Baskerville"] = {
+      :normal      => { :file => SERIF_FONT, :font => 1 },
+      :italic      => { :file => SERIF_FONT, :font => 2 },
+      :bold        => { :file => SERIF_FONT, :font => 4 }, # semi-bold, not bold
+      :bold_italic => { :file => SERIF_FONT, :font => 3 }
+    }
+    font "Baskerville", :size => 14
+  else
+    warn "Baskerville font is preferred for the manual, but could not be found. Using Times-Roman."
+    font "Times-Roman", :size => 14
+  end
+
+  tags :h1 => { :font_size => "2em", :font_weight => :bold, :color => "navy" },
+       :h2 => { :font_size => "1.5em", :font_weight => :bold, :color => "navy" },
+       :about => { :font_size => "80%", :color => "808080", :font_style => :italic }
+
+  styles :no  => { :color => "gray" },
+         :c   => { :color => "#666" },
+         :s   => { :color => "#d20" },
+         :dl  => { :color => "black" },
+         :co  => { :color => "#036", :font_weight => :bold },
+         :pc  => { :color => "#038", :font_weight => :bold },
+         :sy  => { :color => "#A60" },
+         :r   => { :color => "#080" },
+         :i   => { :color => "#00D", :font_weight => :bold },
+         :idl => { :color => "#888", :font_weight => :bold },
+         :dl  => { :color => "#840", :font_weight => :bold }
+
+  File.open("#{File.dirname(__FILE__)}/manual.txt") do |source|
+    number = 0
+    source.each_line do |line|
+      number += 1
+      line.chomp!
+      next if line.length == 0
+
+      style, content = line.match(/^(\S+)\.\s*(.*)/)[1,2]
+      process_style(self, style, content, number)
+    end
+  end
+end

data/prawn-format.gemspec

@@ -0,0 +1,45 @@
+Gem::Specification.new do |s|
+  s.name = %q{prawn-format}
+  s.version = "0.1.0"
+
+  s.required_rubygems_version = Gem::Requirement.new(">= 1.2") if s.respond_to? :required_rubygems_version=
+  s.authors = ["Jamis Buck"]
+  s.date = %q{2009-01-26}
+  s.description = %q{an extension of Prawn that allows inline formatting}
+  s.email = %q{jamis@jamisbuck.org}
+  s.extra_rdoc_files = ["lib/prawn/format/effects/link.rb", "lib/prawn/format/effects/underline.rb", "lib/prawn/format/instructions/base.rb", "lib/prawn/format/instructions/tag_close.rb", "lib/prawn/format/instructions/tag_open.rb", "lib/prawn/format/instructions/text.rb", "lib/prawn/format/layout_builder.rb", "lib/prawn/format/lexer.rb", "lib/prawn/format/line.rb", "lib/prawn/format/parser.rb", "lib/prawn/format/state.rb", "lib/prawn/format/text_object.rb", "lib/prawn/format/version.rb", "lib/prawn/format.rb"]
+  s.files = ["examples/basic-formatting.rb", "examples/christmas-carol.txt", "examples/document.rb", "examples/flowing.rb", "examples/style-classes.rb", "examples/syntax-highlighting.rb", "examples/tags.rb", "lib/prawn/format/effects/link.rb", "lib/prawn/format/effects/underline.rb", "lib/prawn/format/instructions/base.rb", "lib/prawn/format/instructions/tag_close.rb", "lib/prawn/format/instructions/tag_open.rb", "lib/prawn/format/instructions/text.rb", "lib/prawn/format/layout_builder.rb", "lib/prawn/format/lexer.rb", "lib/prawn/format/line.rb", "lib/prawn/format/parser.rb", "lib/prawn/format/state.rb", "lib/prawn/format/text_object.rb", "lib/prawn/format/version.rb", "lib/prawn/format.rb", "manual/html.rb", "manual/include/basics.rb", "manual/include/breaks.rb", "manual/include/custom-tags.rb", "manual/include/custom-tags2.rb", "manual/include/indent.rb", "manual/include/options.rb", "manual/include/style-classes.rb", "manual/manual.txt", "manual/pdf.rb", "Rakefile", "spec/layout_builder_spec.rb", "spec/lexer_spec.rb", "spec/parser_spec.rb", "spec/spec_helper.rb", "Manifest", "prawn-format.gemspec"]
+  s.has_rdoc = true
+  s.homepage = %q{http://rubyforge.org/projects/prawn}
+  s.rdoc_options = ["--line-numbers", "--inline-source", "--title", "Prawn-format"]
+  s.require_paths = ["lib"]
+  s.rubyforge_project = %q{prawn}
+  s.rubygems_version = %q{1.2.0}
+  s.summary = %q{an extension of Prawn that allows inline formatting}
+  s.test_files = ["spec/layout_builder_spec.rb", "spec/lexer_spec.rb", "spec/parser_spec.rb"]
+
+  if s.respond_to? :specification_version then
+    current_version = Gem::Specification::CURRENT_SPECIFICATION_VERSION
+    s.specification_version = 2
+
+    if current_version >= 3 then
+      s.add_runtime_dependency(%q<prawn>, [">= 0.4"])
+      s.add_runtime_dependency(%q<prawn>, ["< 0.5"])
+      s.add_development_dependency(%q<echoe>, [">= 0"])
+      s.add_development_dependency(%q<rake>, [">= 0"])
+      s.add_development_dependency(%q<coderay>, [">= 0"])
+    else
+      s.add_dependency(%q<prawn>, [">= 0.4"])
+      s.add_dependency(%q<prawn>, ["< 0.5"])
+      s.add_dependency(%q<echoe>, [">= 0"])
+      s.add_dependency(%q<rake>, [">= 0"])
+      s.add_dependency(%q<coderay>, [">= 0"])
+    end
+  else
+    s.add_dependency(%q<prawn>, [">= 0.4"])
+    s.add_dependency(%q<prawn>, ["< 0.5"])
+    s.add_dependency(%q<echoe>, [">= 0"])
+    s.add_dependency(%q<rake>, [">= 0"])
+    s.add_dependency(%q<coderay>, [">= 0"])
+  end
+end

data/spec/layout_builder_spec.rb

@@ -0,0 +1,27 @@
+# encoding: utf-8
+
+require File.join(File.expand_path(File.dirname(__FILE__)), "spec_helper")      
+
+describe "when building a text layout" do     
+
+  before(:each) { create_pdf }
+
+  it "should understand common tags by default" do
+    layout = new_layout("<b>hi</b> <i>there</i>")
+    assert_nothing_raised { lines(layout) }
+  end
+
+  private
+
+    def new_layout(text, opts={})
+      Prawn::Format::LayoutBuilder.new(@pdf, text, opts)
+    end
+
+    def lines(layout)
+      lines = []
+      while (line = layout.next)
+        lines << line
+      end
+      return lines
+    end
+end

data/spec/lexer_spec.rb

@@ -0,0 +1,91 @@
+# encoding: utf-8
+
+require File.join(File.expand_path(File.dirname(__FILE__)), "spec_helper")      
+
+def lexer_for(text)
+  Prawn::Format::Lexer.new(text)
+end
+
+describe "when scanning text" do     
+
+  it "should scan a single text word as a chunk" do
+    lexer = lexer_for("christmas")
+    assert_equal({ :type => :text, :text => ["christmas"] }, lexer.next)
+  end
+
+  it "should delimit multiple words by spaces" do
+    lexer = lexer_for("a christmas carol by charles dickens")
+    assert_equal({ :type => :text, :text => ["a", " ", "christmas", " ", "carol", " ", "by", " ", "charles", " ", "dickens"] }, lexer.next)
+  end
+
+  it "should delimit multiple words by hyphens" do
+    lexer = lexer_for("christmas-carol-thingy")
+    assert_equal({ :type => :text, :text => ["christmas", "-", "carol", "-", "thingy"] }, lexer.next)
+  end
+
+  it "should delimit multiple words by em-dashes" do
+    lexer = lexer_for("christmas—carol—thingy")
+    assert_equal({ :type => :text, :text => ["christmas", "—", "carol", "—", "thingy"] }, lexer.next)
+  end
+
+  it "should report nil as end-of-stream after scanning" do
+    lexer = lexer_for("a christmas carol by charles dickens")
+    assert_not_nil lexer.next
+    assert_nil lexer.next
+    assert_nil lexer.next
+  end
+
+end
+
+describe "when scanning XML entities" do
+
+  Prawn::Format::Lexer::ENTITY_MAP.each do |key, value|
+    it "should map #{key} to #{value}" do
+      lexer = lexer_for("&#{key};")
+      assert_equal({ :type => :text, :text => [value] }, lexer.next)
+    end
+  end
+
+  it "should convert decimal entities to utf-8" do
+    lexer = lexer_for('—')
+    assert_equal({ :type => :text, :text => ["\xe2\x80\x94"] }, lexer.next)
+  end
+
+  it "should convert hexadecimal entities to utf-8" do
+    lexer = lexer_for('—')
+    assert_equal({ :type => :text, :text => ["\xe2\x80\x94"] }, lexer.next)
+  end
+
+  it "should raise InvalidFormat on unrecognized entities" do
+    lexer = lexer_for('&bogus;')
+    assert_raises(Prawn::Format::Lexer::InvalidFormat) do
+      lexer.next
+    end
+  end
+
+end
+
+describe "when scanning tags" do
+
+  it "should return tag with empty options for a tag with no options" do
+    lexer = lexer_for("<test>")
+    assert_equal({ :type => :open, :tag => :test, :options => {} }, lexer.next)
+  end
+
+  it "should scan tags with different delimiters consistently" do
+    lexer = lexer_for("<test first=january second=\"february\" third='march'>")
+    assert_equal({ :type => :open, :tag => :test, :options => { :first => "january", :second => "february", :third => "march" } }, lexer.next)
+  end
+
+  it "should scan closing tag" do
+    lexer = lexer_for("</test>")
+    assert_equal({ :type => :close, :tag => :test }, lexer.next)
+  end
+
+  it "should scan self-closing tag as two tokens" do
+    lexer = lexer_for("<test/>")
+    assert_equal({ :type => :open, :tag => :test, :options => {} }, lexer.next)
+    assert_equal({ :type => :close, :tag => :test }, lexer.next)
+  end
+
+end

data/spec/parser_spec.rb

@@ -0,0 +1,103 @@
+# encoding: utf-8
+
+require File.join(File.expand_path(File.dirname(__FILE__)), "spec_helper")      
+
+describe "when parsing formatted text" do     
+
+  before(:each) { create_pdf }
+
+  it "should raise TagError when it does not recognize a tag" do
+    parser = parser_for("<hello>")
+    assert_raises(Prawn::Format::Parser::TagError) { parser.next }
+  end
+
+  it "should raise TagError when it encounters an unmatched closing tag" do
+    parser = parser_for("</b>")
+    assert_raises(Prawn::Format::Parser::TagError) { parser.next }
+  end
+
+  it "should raise TagError when it a tag is closed with the wrong type" do
+    parser = parser_for("<b></i>")
+    assert_raises(Prawn::Format::Parser::TagError) { parse(parser) }
+  end
+
+  it "should apply styles defined for tag" do
+    parser = parser_for("a<b>c</b>d")
+    weights = [:normal, :bold, :bold, :bold, :normal]
+    assert_equal weights, parse(parser).map { |i| i.state.font_weight }
+  end
+
+  it "should honor custom tag styles" do
+    parser = parser_for("a<k>c</k>d", :tags => { :k => { :text_decoration => :underline } })
+    decorations = [:none, :underline, :underline, :underline, :none]
+    assert_equal decorations, parse(parser).map { |i| i.state.text_decoration }
+  end
+
+  it "should honor custom style classes" do
+    parser = parser_for("a<j class='test'>c</j>d", :tags => { :j => {} }, :styles => { :test => { :text_decoration => :underline } })
+    decorations = [:none, :underline, :underline, :underline, :none]
+    assert_equal decorations, parse(parser).map { |i| i.state.text_decoration }
+  end
+
+  it "should parse delimited text as separate instructions" do
+    parser = parser_for("a b-cd")
+    bits = ["a", " ", "b", "-", "cd"]
+    assert_equal bits, parse(parser).map { |i| i.text }
+  end
+
+  it "should return nil after last token is parsed" do
+    parser = parser_for("a")
+    assert_not_nil parser.next
+    assert_nil parser.next
+    assert_nil parser.next
+  end
+
+  it "should report eos? at end of stream" do
+    parser = parser_for("a")
+    assert !parser.eos?
+    parser.next
+    assert parser.eos?
+  end
+
+  it "should return next instruction without consuming it when peek is called" do
+    parser = parser_for("a")
+    assert_equal "a", parser.peek.text
+    assert !parser.eos?
+    assert_equal "a", parser.next.text
+    assert parser.eos?
+  end
+
+  it "should save instruction for next call when push is called" do
+    parser = parser_for("a")
+    k = parser.next
+    assert parser.eos?
+    parser.push(k)
+    assert !parser.eos?
+    assert_equal k, parser.next
+    assert parser.eos?
+  end
+
+  it "should map meta styles to styles on the tag" do
+    parser = parser_for("<k name='bob'>a</k>", :tags => { :k => { :meta => { :name => :__name__ } } })
+    i = parser.next
+    assert_equal "bob", i.tag[:style][:__name__]
+  end
+
+  private
+
+    def parser_for(text, opts={})
+      tags = @pdf.tags.merge(opts[:tags] || {})
+      styles = @pdf.styles.merge(opts[:styles] || {})
+      @parser = Prawn::Format::Parser.new(@pdf, text, opts.merge(:styles => styles, :tags => tags))
+    end
+
+    def parse(parser)
+      instructions = []
+
+      while instr = parser.next
+        instructions << instr
+      end
+
+      return instructions
+    end
+end