HTML5lib Sanitizer

Replaced native Sanitizer with HTML5lib version.
Synced with latest Maruku.

vendor/plugins/maruku/lib/maruku/defaults.rb

@@ -31,6 +31,9 @@ Globals = {
 	:maruku_signature => false,
 	:code_background_color => '#fef',
 	:code_show_spaces => false,
+	
+	:filter_html => false,
+	
 	:html_math_output_mathml => true, # also set :html_math_engine
 	:html_math_engine => 'itex2mml', #ritex, itex2mml, none
 	

vendor/plugins/maruku/lib/maruku/input/parse_block.rb

@@ -477,7 +477,7 @@ module MaRuKu; module In; module Markdown; module BlockLevelParser
 		end
 		
 		id = match[1]; url = match[2]; title = match[3]; 
-		id = id.strip.downcase.gsub(' ','_')
+		id = sanitize_ref_id(id)
 		
 		hash = self.refs[id] = {:url=>url,:title=>title}
 		

vendor/plugins/maruku/lib/maruku/input/parse_span_better.rb

@@ -287,7 +287,7 @@ module MaRuKu; module In; module Markdown; module SpanLevelParser
 	end
 
 	def extension_meta(src, con, break_on_chars)
-		if m = src.read_regexp(/([^\s\:]+):/)
+		if m = src.read_regexp(/([^\s\:\"\']+):/)
 			name = m[1]
 			al = read_attribute_list(src, con, break_on_chars)
 #			puts "#{name}=#{al.inspect}"
@@ -581,9 +581,9 @@ module MaRuKu; module In; module Markdown; module SpanLevelParser
 			ref_id = read_ref_id(src,con)
 			if ref_id
 				if ref_id.size == 0
-					ref_id =  children.to_s.downcase.gsub(' ','_')
+					ref_id = sanitize_ref_id(children.to_s)
 				else
-					ref_id = ref_id.downcase
+					ref_id = sanitize_ref_id(ref_id)
 				end	
 				con.push_element md_link(children, ref_id)
 			else 

vendor/plugins/maruku/lib/maruku/input_textile2/t2_parser.rb

@@ -108,6 +108,7 @@ module MaRuKu
 		# Input is a LineSource
 		def t2_parse_blocks(src, output)
 			while src.cur_line
+				l = src.shift_line
 				
 				# ignore empty line
 				if l.t2_empty? then 
@@ -115,7 +116,6 @@ module MaRuKu
 					next 
 				end
 				
-				l = src.shift_line
 				# TODO: lists
 				# TODO: xml
 				# TODO: `==`

vendor/plugins/maruku/lib/maruku/output/to_html.rb

@@ -741,7 +741,17 @@ of the form `#ff00ff`.
 		return a
 	end
 
+=begin maruku_doc
+Attribute: filter_html
+Scope: document
+
+If true, raw HTML is discarded from the output.
+
+=end
+
 	def to_html_raw_html
+		return [] if get_setting(:filter_html)
+		
 		raw_html = self.raw_html
 		if rexml_doc = @parsed_html
 			root = rexml_doc.root

vendor/plugins/maruku/lib/maruku/output/to_latex.rb

@@ -152,7 +152,7 @@ end end
 module MaRuKu; module Out; module Latex
 	
 	def to_latex_hrule; "\n\\vspace{.5em} \\hrule \\vspace{.5em}\n" end
-	def to_latex_linebreak; "\\linebreak " end
+	def to_latex_linebreak; "\\newline " end
 	
 	def to_latex_paragraph 
 		children_to_latex+"\n\n"

vendor/plugins/maruku/lib/maruku/string_utils.rb

@@ -146,6 +146,10 @@ module MaRuKu; module Strings
 		s[0, i+1].strip
 	end
 	
+	def sanitize_ref_id(x)
+		x.downcase.gsub(' ','_').gsub(/[^\w]/,'')
+	end
+
 
 	# removes initial quote
 	def unquote(s)

vendor/plugins/maruku/lib/maruku/tests/new_parser.rb

@@ -155,7 +155,7 @@ module MaRuKu; module Tests
 		["[a]",   [ md_link(["a"],'a')], 'Empty link'],
 		["[a][]", ],
 		["[a][]b",   [ md_link(["a"],'a'),'b'], 'Empty link'],
-		["[a\\]][]", [ md_link(["a]"],'a]')], 'Escape inside link'],
+		["[a\\]][]", [ md_link(["a]"],'a')], 'Escape inside link (throw ?] away)'],
 		
 		["[a",  :throw,   'Link not closed'],
 		["[a][",  :throw,   'Ref not closed'],

vendor/plugins/maruku/lib/maruku/version.rb

@@ -19,7 +19,7 @@
 #++
 
 module MaRuKu
-	Version = '0.5.5'
+	Version = '0.5.6'
 	
 	MarukuURL = 'http://maruku.rubyforge.org/'