Updated to latest Maruku.

config/spam_patterns.txt

@@ -49,7 +49,6 @@ lust cartoon
 mijneigenweblog
 Mortage
 myspace
-naked
 netfirms\.com
 nice site
 overflow:\s*auto

lib/maruku/defaults.rb

@@ -23,8 +23,10 @@ module MaRuKu
 	
 Globals = {
 	:unsafe_features => false,
+	:on_error => :warning,
 	
-	:debug_keep_ials => false,
+	
+	:use_numbered_headers => false,
 	
 	:maruku_signature => false,
 	:code_background_color => '#fef',
@@ -37,7 +39,12 @@ Globals = {
 	:html_png_resolution => 200,
 	
 	:html_use_syntax => false,
-	:on_error => :warning
+	
+	:latex_use_listings => false,
+	:latex_cjk => false,
+	
+	:debug_keep_ials => false,
+	
 }
 
 class MDElement

lib/maruku/input/html_helper.rb

@@ -29,12 +29,13 @@ module MaRuKu; module In; module Markdown; module SpanLevelParser
 		include MaRuKu::Strings
 		
 		Tag = %r{^<(/)?(\w+)\s*([^>]*)>}m
+		PartialTag = %r{^<.*}m
+
 		EverythingElse = %r{^[^<]+}m
 		CommentStart = %r{^<!--}x
 		CommentEnd = %r{^.*-->}
 		TO_SANITIZE = ['img','hr'] 
 		
-#		attr_accessor :inside_comment
 		attr_reader :rest
 		
 		def initialize 
@@ -42,30 +43,70 @@ module MaRuKu; module In; module Markdown; module SpanLevelParser
 			@tag_stack = []
 			@m = nil
 			@already = ""
-			@inside_comment = false
+			self.state = :inside_element
 		end
 
+		attr_accessor :state # :inside_element, :inside_tag, :inside_comment,
+		
 		def eat_this(line)
 			@rest = line  + @rest
 			things_read = 0
 			until @rest.empty?
-				if @inside_comment
+				case self.state
+					when :inside_comment
 						if @m = CommentEnd.match(@rest)
-						@inside_comment = false
 							@already += @m.pre_match + @m.to_s
 							@rest = @m.post_match
+							self.state = :inside_element
+						else 
+							@already += @rest 
+							@rest = ""
+							self.state = :inside_comment
+						end
+					when :inside_element 
+						if @m = CommentStart.match(@rest)
+							things_read += 1
+							@already += @m.pre_match + @m.to_s
+							@rest = @m.post_match
+							self.state = :inside_comment
+						elsif @m = Tag.match(@rest) then
+							things_read += 1
+							handle_tag
+							self.state = :inside_element
+						elsif @m = PartialTag.match(@rest) then
+							@already += @m.pre_match 
+							@rest = @m.post_match
+							@partial_tag = @m.to_s
+							self.state = :inside_tag
 						elsif @m = EverythingElse.match(@rest)
 							@already += @m.pre_match + @m.to_s
 							@rest = @m.post_match
+							self.state = :inside_element
+						else
+							error "Malformed HTML: not complete: #{@rest.inspect}"
+						end
+					when :inside_tag
+						if @m = /^[^>]*>/.match(@rest) then
+							@partial_tag += @m.to_s
+							@rest = @partial_tag + @m.post_match
+							@partial_tag = nil
+							self.state = :inside_element
+						else
+							@partial_tag += @rest
+							@rest = ""
+							self.state = :inside_tag
 						end
 					else
-					if @m = CommentStart.match(@rest)
+						raise "Bug bug: state = #{self.state.inspect}"
-						things_read += 1
+				end # not inside comment
-						@inside_comment = true
+				
-						@already += @m.pre_match + @m.to_s
+#				puts inspect
-						@rest = @m.post_match
+#				puts "Read: #{@tag_stack.inspect}"
-					elsif @m = Tag.match(@rest)
+				break if is_finished? and things_read>0	
-						things_read += 1
+			end
+		end
+
+		def handle_tag()
 			@already += @m.pre_match
 			@rest = @m.post_match
 
@@ -73,12 +114,15 @@ module MaRuKu; module In; module Markdown; module SpanLevelParser
 			tag = @m[2]
 			attributes = @m[3]
 
+		
 			is_single = false
 			if attributes =~ /\A(.*)\/\Z/
 				attributes = $1
 				is_single = true
 			end
 
+#			puts "READ TAG #{@m.to_s.inspect} tag = #{tag} closing? #{is_closing} single = #{is_single}"
+	
 			if TO_SANITIZE.include? tag 
 				attributes.strip!
 		#		puts "Attributes: #{attributes.inspect}"
@@ -98,30 +142,17 @@ module MaRuKu; module In; module Markdown; module SpanLevelParser
 					      "closes <#{@tag_stack.last}>"
 				end
 				@tag_stack.pop
-						elsif not is_single
-							@tag_stack.push tag
-							@already += @m.to_s
-						end
-					elsif @m = EverythingElse.match(@rest)
-						@already += @m.pre_match + @m.to_s
-						@rest = @m.post_match
 			else 
-						error "Malformed HTML: not complete: #{@rest.inspect}"
+				@already += @m.to_s
-					end
-				end # not inside comment
 				
-#				puts inspect
+				@tag_stack.push(tag) unless is_single
-#				puts "Read: #{@tag_stack.inspect}"
-				break if is_finished? and things_read>0	
 			end
 		end
-
-
 		def error(s)
 			raise Exception, "Error: #{s} \n"+ inspect, caller
 		end
 
-		def inspect; "HTML READER\n comment=#{@inside_comment} "+
+		def inspect; "HTML READER\n state=#{self.state} "+
 			"match=#{@m.to_s.inspect}\n"+
 			"Tag stack = #{@tag_stack.inspect} \n"+
 			"Before:\n"+
@@ -137,7 +168,7 @@ module MaRuKu; module In; module Markdown; module SpanLevelParser
 		end
 		
 		def is_finished?
-			not @inside_comment and @tag_stack.empty?
+			(self.state == :inside_element)  and @tag_stack.empty?
 		end
 	end # html helper 
 

lib/maruku/input/parse_span_better.rb

@@ -557,6 +557,11 @@ module MaRuKu; module In; module Markdown; module SpanLevelParser
 		when ?[ # link ref
 			ref_id = read_ref_id(src,con)
 			if ref_id
+				if ref_id.size == 0
+					ref_id =  children.to_s.downcase.gsub(' ','_')
+				else
+					ref_id = ref_id.downcase
+				end	
 				con.push_element md_link(children, ref_id)
 			else 
 				maruku_error "Could not read ref_id", src, con
@@ -566,7 +571,8 @@ module MaRuKu; module In; module Markdown; module SpanLevelParser
 				return
 			end
 		else # empty [link]
-			con.push_element md_link(children, "")
+			id = children.to_s.downcase.gsub(' ','_')
+			con.push_element md_link(children, id)
 		end
 	end # read link
 

lib/maruku/input/type_detection.rb

@@ -44,11 +44,12 @@ module MaRuKu; module Strings
 		return :definition       if l =~ Definition
 		# I had a bug with emails and urls at the beginning of the 
 		# line that were mistaken for raw_html
-		return :text     if l=~EMailAddress or l=~ URL
+		return :text if l=~ /^#{EMailAddress}/
+		return :text if l=~ /^<http:/
 		# raw html is like PHP Markdown Extra: at most three spaces before
 		return :xml_instr if l =~ %r{^\s*<\?}
 		return :raw_html if l =~ %r{^[ ]?[ ]?[ ]?</?\s*\w+}
-		return :raw_html if l =~ %r{[ ]{0,3}<\!\-\-}
+		return :raw_html if l =~ %r{^[ ]?[ ]?[ ]?<\!\-\-}
 		return :ulist    if l =~ /^\s?([\*\-\+])\s+.*\w+/
 		return :olist    if l =~ /^\s?\d+\..*\w+/
 		return :header1  if l =~ /^(=)+/ 
@@ -137,5 +138,4 @@ module MaRuKu; module Strings
 
 
 	EMailAddress = /<([^:]+@[^:]+)>/
-	URL = /^<http:/
 end end

lib/maruku/output/to_html.rb

@@ -21,8 +21,6 @@
 
 require 'rexml/document'
 
-
-
 class String
 	# A string is rendered into HTML by creating
 	# a REXML::Text node. REXML takes care of all the encoding.
@@ -101,6 +99,52 @@ xhtml11_mathml2_svg11 =
 	
 	def xml_newline() Text.new("\n") end
 		
+
+=begin maruku_doc
+Attribute: title
+Scope: document
+
+Sets the title of the document.
+If a title is not specified, the first header will be used.
+
+These should be equivalent:
+
+	Title: my document
+
+	Content
+
+and
+
+	my document
+	===========
+
+	Content
+
+In both cases, the title is set to "my document".
+=end
+
+=begin maruku_doc
+Attribute: subject
+Scope: document
+
+Synonim for `title`.
+=end
+
+
+=begin maruku_doc
+Attribute: css
+Scope: document
+Output: HTML
+Summary: Activates CSS stylesheets for HTML.
+
+`css` should be a space-separated list of urls.
+
+Example:
+	
+	CSS: style.css math.css
+
+=end
+
 	# Render to a complete HTML document (returns a REXML document tree)
 	def to_html_document_tree
 		doc = Document.new(nil,{:respect_whitespace =>:all})
@@ -121,6 +165,7 @@ xhtml11_mathml2_svg11 =
 #			me.attributes['content'] = 'text/html;charset=utf-8'	
 			me.attributes['content'] = 'application/xhtml+xml;charset=utf-8'	
 		
+
 			# Create title element
 			doc_title = self.attributes[:title] || self.attributes[:subject] || ""
 			title = Element.new 'title', head
@@ -243,6 +288,34 @@ xhtml11_mathml2_svg11 =
 		m
 	end
 	
+=begin maruku_doc
+Attribute: id
+Scope: element
+Output: LaTeX, HTML
+
+It is copied as a standard HTML attribute.
+
+Moreover, it used as a label name for hyperlinks in both HTML and
+in PDF.
+
+=end
+
+=begin maruku_doc
+Attribute: class
+Scope: element
+Output: HTML
+
+It is copied as a standard HTML attribute.
+=end
+
+=begin maruku_doc
+Attribute: style
+Scope: element
+Output: HTML
+
+It is copied as a standard HTML attribute.
+=end
+	
 	StandardAttributes = [:id, :style, :class]
 	def create_html_element(name, attributes_to_copy=[])
 		m = Element.new name
@@ -272,9 +345,20 @@ xhtml11_mathml2_svg11 =
 	def to_html_strong;    wrap_as_element('strong')           end
 	def to_html_emphasis;  wrap_as_element('em')               end
 
+=begin maruku_doc
+Attribute: use_numbered_headers
+Scope: document
+Summary: Activates the numbering of headers.
+
+If `true`, section headers will be numbered.
+
+In LaTeX export, the numbering of headers is managed
+by Maruku, to have the same results in both HTML and LaTeX.
+=end
+
 	# nil if not applicable, else string
 	def section_number
-		return nil if not @doc.attributes[:use_numbered_headers]
+		return nil if not get_setting(:use_numbered_headers)
 		
 		n = @attributes[:section_number]
 		if n && (not n.empty?)
@@ -315,14 +399,35 @@ xhtml11_mathml2_svg11 =
 		
 =begin maruku_doc
 Attribute: html_use_syntax
-Scope: document
+Scope: global, document, element
-Output: html
+Output: HTML
 Summary: Enables the use of the `syntax` package.
 Related: lang, code_lang
-Default: <?mrk Globals[:html_use_syntax].to_s ?>
+Default: <?mrk md_code(Globals[:html_use_syntax].to_s) ?>
+
+If true, the `syntax` package is used. It supports the `ruby` and `xml`
+languages. Remember to set the `lang` attribute of the code block.
+
+Examples:
+
+		require 'maruku'
+	{:lang=ruby html_use_syntax=true}
+
+and 
+	
+		<div style="text-align:center">Div</div>
+	{:lang=html html_use_syntax=true}
+
+produces:
+
+	require 'maruku'
+{:lang=ruby html_use_syntax=true}
+
+and
+
+	<div style="text-align:center">Div</div>
+{:lang=html html_use_syntax=true}
 
-If false, Maruku does not append a signature to the
-generated file.
 =end
 
 	def to_html_code; 
@@ -378,6 +483,26 @@ generated file.
 		element
 	end
 	
+=begin maruku_doc
+Attribute: code_background_color
+Scope: global, document, element
+Summary: Background color for code blocks.
+
+The format is either a named color (`green`, `red`) or a CSS color
+of the form `#ff00ff`. 
+
+* for **HTML output**, the value is put straight in the `background-color` CSS 
+  property of the block.
+
+* for **LaTeX output**, if it is a named color, it must be a color accepted
+  by the LaTeX `color` packages. If it is of the form `#ff00ff`, Maruku
+  defines a color using the `\color[rgb]{r,g,b}` macro. 
+
+  For example, for `#0000ff`, the macro is called as: `\color[rgb]{0,0,1}`.
+
+=end
+	
+	
 	def to_html_code_using_pre(source)
 		pre = create_html_element  'pre'
 		code = Element.new 'code', pre
@@ -427,11 +552,6 @@ generated file.
 	def to_html_link
 		a =  wrap_as_element 'a'
 		id = self.ref_id
-		# if empty, use text
-		if id.size == 0
-			id = children.to_s.downcase.gsub(' ','_')
-			
-		end
 		
 		if ref = @doc.refs[id]
 			url = ref[:url]

lib/maruku/output/to_latex.rb

@@ -88,7 +88,7 @@ while the default is to add this:
 <?mrk md_codeblock(Maruku::MDDocument::Latex_preamble_enc_utf8) ?>
 
 =end
-		encoding = @doc.attributes[:latex_cjk] ? 
+		encoding = get_setting(:latex_cjk) ? 
 			Latex_preamble_enc_cjk : Latex_preamble_enc_utf8
 
 =begin maruku_doc
@@ -190,9 +190,31 @@ Admissible formats:
 		end
 	end
 	
+=begin maruku_doc
+Attribute: code_show_spaces
+Scope: global, document, element
+
+If `true`, shows spaces and tabs in code blocks.
+
+Example:
+
+		 One space
+		  Two spaces
+			 	Tab, space, tab
+					Tab, tab, tab and all is green!
+	{:code_show_spaces code_background_color=#ffeedd}
+{:markdown}
+	
+That will produce:
+
+	 One space
+	  Two spaces
+		 	Tab, space, tab
+				Tab, tab, tab and all is green!
+{:code_show_spaces code_background_color=#ffeedd}
+
+=end
 	
-	def to_latex_code;
-		raw_code = self.raw_code
 =begin maruku_doc
 Attribute: latex_use_listings
 Scope: document
@@ -227,7 +249,10 @@ Otherwise, a standard `verbatim` environment is used.
 
 =end
 
-		if @doc.attributes[:latex_use_listings]
+	def to_latex_code;
+		raw_code = self.raw_code
+		
+		if get_setting(:latex_use_listings)
 			@doc.latex_require_package('listings')
 				
 			s = "\\lstset{columns=fixed,frame=shadowbox}"
@@ -314,7 +339,7 @@ Otherwise, a standard `verbatim` environment is used.
 \\end{#{name}}\n"	
 	end
 	
-	SAFE_CHARS = Set.new([?\ ] + (?a..?z).to_a + (?A..?Z).to_a)
+	SAFE_CHARS = Set.new((?a..?z).to_a + (?A..?Z).to_a)
 	# the ultimate escaping
 	# (is much better than using \verb)
 	def latex_escape(source)
@@ -345,7 +370,6 @@ Otherwise, a standard `verbatim` environment is used.
 	end
 
 	def to_latex_immediate_link
-		a =  create_html_element 'a'
 		url = self.url
 		text = url.gsub(/^mailto:/,'') # don't show mailto
 #			gsub('~','$\sim$')
@@ -372,11 +396,6 @@ Otherwise, a standard `verbatim` environment is used.
 	
 	def to_latex_link
 		id = self.ref_id
-		# if empty, use text
-		if id.size == 0
-			id = children.to_s.downcase
-		end
-		
 		ref = @doc.refs[id]
 		if not ref
 			$stderr.puts "Could not find id = '#{id}'"

lib/maruku/tests/new_parser.rb

@@ -152,10 +152,10 @@ module MaRuKu; module Tests
 		["\\[a\\]", ["[a]"], 'Escaping 2'],
 # This is valid in the new Markdown version
 #		["[a]",   ["a"],   'Not a link'],
-		["[a]",   [ md_link(["a"],'')], 'Empty link'],
+		["[a]",   [ md_link(["a"],'a')], 'Empty link'],
 		["[a][]", ],
-		["[a][]b",   [ md_link(["a"],''),'b'], 'Empty link'],
+		["[a][]b",   [ md_link(["a"],'a'),'b'], 'Empty link'],
-		["[a\\]][]", [ md_link(["a]"],'')], 'Escape inside link'],
+		["[a\\]][]", [ md_link(["a]"],'a]')], 'Escape inside link'],
 		
 		["[a",  :throw,   'Link not closed'],
 		["[a][",  :throw,   'Ref not closed'],

lib/maruku/version.rb

@@ -19,7 +19,7 @@
 #++
 
 module MaRuKu
-	Version = '0.4.2.1'
+	Version = '0.5.0'
 	
 	MarukuURL = 'http://maruku.rubyforge.org/'