semantictext 0.2.1 → 0.2.2

data/README.rdoc

@@ -38,7 +38,7 @@ Semantic text supports:
 * inline hyperlinks for http: mailto: and ftp:
 * markup tags within bullet points
 * custom markup tags e.g. postal code, youtube video embed, ... whatever you define in a subclass of SemanticText::DefaultTagFactory
-
+* verbatim code blocks delimited by lines marked "{{{" and "}}}" respectively
 
 === Compatibility
 This project is being developed on OS X. Automated testing for Linux will be included in future releases.

data/TODO.rdoc

@@ -1,17 +1,18 @@
 ==to do
-* improve testing by mocking out tag factory used in tests - consider how/whether to do this
 * tighten up error test cases on ftptag, httptag and mailtotag
-* add verbatim code sections between {{{ and }}} brackets at start of line
-* support wikinames as a special custom tag
+* improve testing by mocking out tag factory used in tests - consider how/whether to do this
 * fix path to testfiles so test can run on gem - tests only run on source at the moment
 
 ==maybe
+* support wikinames as a special custom tag
 * consider including section objects that are delimited by headings and that contain all objects in a section
 * think about how to support twitter with special structure tags e.g. #keyword and @user
 * refactor parser into header parser and text parser
 * pull out parsers for different parts and use the state pattern
 
 ==done
+* add verbatim code sections between {{{ and }}} brackets at start of line
+* added multi-level headings (in release 0.2.1)
 * added basic rcov test coverage task to Rakefile
 * find a way to generate demo.txt file from rake
 * make demo.txt be up-to-date as dependency of rdoc

data/doc/demo.script

@@ -7,6 +7,10 @@ p.parse "!I'm a title"
 p.parse "I'm a paragraph..."
 p.parse "... yep, I'm still that paragraph."
 p.parse ''
+p.parse '{{{'
+p.parse '10 PRINT "HELLO"'
+p.parse '20 GOTO 10'
+p.parse '}}}'
 p.parse "* I'm a bullet point"
 p.parse "* I'm another bullet point"
 

data/lib/semantictext/document.rb

@@ -10,6 +10,7 @@ require 'semantictext/bulletedlist'
 require 'semantictext/bullet'
 require 'semantictext/bulleted_list_parser'
 require 'semantictext/rich_text_parser'
+require 'semantictext/verbatim'
 
 module SemanticText
  
@@ -18,109 +19,136 @@ module SemanticText
   #* I initiate and co-ordinate document-wide operations.
   class Document
     # title of the document
-  	attr_reader :title
-  	
-  	# date the document was created
-  	attr_reader :createdAt
-  	
-  	# keyword list for the current document
-  	attr_reader :keywords 
-  	
-  	# pathname of the file currently being parsed (if it exists, nil otherwise)
-  	attr_reader :pathname
-  	
-  	# the object model of the parsed document
-  	attr_reader :content
+    attr_reader :title
 
-  	def initialize(tag_factory=DefaultTagFactory.new)
-  	  @pathname=nil
-  		@headers_completed = false
-  		@content = []
-  		@current_paragraph = nil
-      @bulleted_list_parser = nil
-      @rich_text_parser = RichTextParser.new(tag_factory)
-  	end
+    # date the document was created
+    attr_reader :createdAt
 
-    # export as html
-    def export_html
-      out = ""
-      content.each {|element| out=out+element.export_html}
-      out = out + "\n"
-    end
+    # keyword list for the current document
+    attr_reader :keywords 
 
-    # true iff I have seen the end of the headers section at the top of the document
-    def parameters_complete?
-      @headers_completed
-    end
+    # pathname of the file currently being parsed (if it exists, nil otherwise)
+    attr_reader :pathname
+
+    # the object model of the parsed document
+    attr_reader :content
+
+  def initialize(tag_factory=DefaultTagFactory.new)
+    @pathname=nil
+    @headers_completed = false
+    @content = []
+    @current_paragraph = nil
+    @verbatim = nil
+    @bulleted_list_parser = nil
+    @rich_text_parser = RichTextParser.new(tag_factory)
+  end
+
+  # export as html
+  def export_html
+    out = ""
+    content.each {|element| out=out+element.export_html}
+    out = out + "\n"
+  end
+
+  # true iff I have seen the end of the headers section at the top of the document
+  def parameters_complete?
+    @headers_completed
+  end
   
-    # parse a document into this object from pathname specified by file
-  	def parse_from(file)
-  	  @pathname=file
-  		f = File.new(file)
-  		f.each_line do |line|
-  		  parse(line)
-  		end
-  		f.close
-  	end
+  # parse a document into this object from pathname specified by file
+  def parse_from(file)
+    @pathname=file
+    f = File.new(file)
+    f.each_line do |line|
+      parse(line)
+    end
+    f.close
+  end
 
-    # parse an individual <i>line</i> of String appending content
-    # into the current document held by this object
-  	def parse(line)
-  	  line.chomp!
-      begin
-        if (!@headers_completed)
-        	process_header_line(line) 
-        else
-        	parse_line(line)
-        end
+  # parse an individual <i>line</i> of String appending content
+  # into the current document held by this object
+  def parse(line)
+    line.chomp!
+    begin
+      if (!@headers_completed)
+        process_header_line(line) 
+      else
+        parse_line(line)
+      end
       rescue  NotHeaderLine
         @headers_completed = true
         parse_line(line)
-  		end	
-  	end
+    end	
+  end
 	
-  	private
+    private
 	
-  	def process_header_line(headerLine)
-  		splitLine = headerLine.split(':',2)
-  		(attributeName, value) = splitLine
-  		raise NotHeaderLine.new() if splitLine.size <2
-  		attributeName.strip!
-  		@title = value if attributeName=='title'
-  		@createdAt = DateExtractor.new.extract_from(value) if attributeName=='createdAt'
-  		@keywords = KeywordExtractor.new.extract_from(value) if attributeName=='keywords'
-  	end
+  def process_header_line(headerLine)
+    splitLine = headerLine.split(':',2)
+    (attributeName, value) = splitLine
+    raise NotHeaderLine.new() if splitLine.size <2
+    attributeName.strip!
+    @title = value if attributeName=='title'
+    @createdAt = DateExtractor.new.extract_from(value) if attributeName=='createdAt'
+    @keywords = KeywordExtractor.new.extract_from(value) if attributeName=='keywords'
+  end
 
-    def parse_paragraph_line(line)
-  	  if @current_paragraph.nil?
-        @current_paragraph = Paragraph.new
-        @content << @current_paragraph
-      end
-      @rich_text_parser.parse(line, @current_paragraph)    
+  def parse_paragraph_line(line)
+    if @current_paragraph.nil?
+      @current_paragraph = Paragraph.new
+      @content << @current_paragraph
     end
+    @rich_text_parser.parse(line, @current_paragraph)    
+  end
     
-  	def parse_line(line)
-  	  @bulleted_list_parser = nil if !line.begins_with '*'
-  	  if (line =='')
-        @current_paragraph = nil
-        @bulleted_list = nil
-  	    return
-      end
-  		if (line.begins_with('!'))
-  		  @content << Heading.parse(line)
-  		else
-  		  if (line.begins_with('*'))
-  		    if @bulleted_list_parser.nil?
-  		      @bulleted_list_parser = BulletedListParser.new(@rich_text_parser)
-  		      @content << @bulleted_list_parser.bulleted_list
-  	      end
-  	      @bulleted_list_parser.parse_line(line)
-  	    else
-  		    parse_paragraph_line(line)
-  	    end
-  		end
-  	end	
-
+  # When a bullet point is seen:
+  #* ensure we have a @bulleted_list_parser,
+  #* create one if necessary,
+  #* register its output in the @content.
+  def ensure_bulleted_list_active
+    if @bulleted_list_parser.nil?
+      @bulleted_list_parser = BulletedListParser.new(@rich_text_parser)
+      @content << @bulleted_list_parser.bulleted_list
+    end
   end
+    
+  def ensure_verbatim_active
+    @content << @verbatim = Verbatim.new { @verbatim = nil}
+  end
+  
+  def parse_line(line)
+    if @verbatim
+      @verbatim.parse(line)
+      return
+    end
+    
+    @bulleted_list_parser = nil unless line.begins_with '*'
+
+    if (line =='')
+      @current_paragraph = nil
+      @bulleted_list = nil
+      return
+    end
 
+    if (line.begins_with('!'))
+      @content << Heading.parse(line)
+      return
+    end
+
+    if (line.begins_with('*'))
+      ensure_bulleted_list_active
+      @bulleted_list_parser.parse_line(line)
+      return
+    end
+
+    if (line == '{{{')
+      ensure_verbatim_active
+      return
+    end
+    
+    parse_paragraph_line(line)
+    return
+  end	
+
+  end
 end

data/lib/semantictext/verbatim.rb

@@ -0,0 +1,31 @@
+require 'cgi'
+module SemanticText
+  
+  class Verbatim
+    attr_reader :lines
+
+    @end_of_section_block = {}
+
+    def initialize(&end_of_section_handler)
+      @lines= []
+      @end_of_section_block = end_of_section_handler
+    end
+
+    #Parse line of text into a verbatim block. 
+    #Runs end of verbatim handler when end of verbatim found.
+    def parse(line)
+      if line=='}}}'
+        @end_of_section_block.call
+      else
+        @lines << line
+      end
+    end
+
+    def export_html
+      out = "\n<pre>\n"
+      lines.each  {|line | out += CGI.escapeHTML(line)+"\n" }
+      out += "</pre>\n"
+    end
+
+  end
+end

data/test/test_export.rb

@@ -11,7 +11,7 @@ class TestExport < Test::Unit::TestCase
     
     expected_file=File.new(ENV['SANDBOX']+'/semantictext/testfiles/regression-exportsample.txt')
     expected = expected_file.readlines
-    
+
     (0..(expected.size-1)).each {|index| assert_equal expected[index],actual[index]+"\n"}
     
     assert_equal expected.size, actual.size

data/testfiles/complex.art

@@ -9,6 +9,13 @@ paragraph.
 
 This paragraph tests escaping < > &
 
+{{{
+verbatim block line 1 < > &
+verbatim block line 2
+
+verbatim block line 4
+}}}
+
 Theis is the third paragraph.
 
 !!second-level heading

data/testfiles/regression-exportsample.txt

@@ -2,6 +2,13 @@
 <h1>First Big Heading</h1>
 <p> This is another paragraph.</p>
 <p> This paragraph tests escaping &lt; &gt; &amp;</p>
+<pre>
+verbatim block line 1 &lt; &gt; &amp;
+verbatim block line 2
+
+verbatim block line 4
+</pre>
+
 <p> Theis is the third paragraph.</p>
 <h2>second-level heading</h2>
 <h3>third-level heading</h3>

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification 
 name: semantictext
 version: !ruby/object:Gem::Version 
-  version: 0.2.1
+  version: 0.2.2
 platform: ruby
 authors: 
 - Dafydd Rees
@@ -9,7 +9,7 @@ autorequire:
 bindir: bin
 cert_chain: []
 
-date: 2009-12-05 00:00:00 +00:00
+date: 2009-12-06 00:00:00 +00:00
 default_executable: 
 dependencies: []
 
@@ -45,6 +45,7 @@ files:
 - lib/semantictext/tag.rb
 - lib/semantictext/tag_parsing_failed.rb
 - lib/semantictext/unknown_tag.rb
+- lib/semantictext/verbatim.rb
 - lib/semantictext.rb
 - lib/string.rb
 - test/test_bullet.rb