0
@@ -1,4 +1,3 @@
0
-#!/usr/bin/ruby
0
 # 
0
 # Bluecloth is a Ruby implementation of Markdown, a text-to-HTML conversion
0
 # tool.
0
@@ -49,81 +48,18 @@
0
 #   regexps.
0
 #
0
 # * Put the StringScanner in the render state for thread-safety.
0
-#
0
-# == Version
0
-#
0
-#  $Id$
0
-# 
0
 
0
-require 'digest/md5'
0
 require 'logger'
0
-require 'strscan'
0
-
0
-
0
-### BlueCloth is a Ruby implementation of Markdown, a text-to-HTML conversion
0
-### tool.
0
-class BlueCloth < String
0
-
0
-	### Exception class for formatting errors.
0
-	class FormatError < RuntimeError
0
-
0
-		### Create a new FormatError with the given source +str+ and an optional
0
-		### message about the +specific+ error.
0
-		def initialize( str, specific=nil )
0
-			if specific
0
-				msg = "Bad markdown format near %p: %s" % [ str, specific ]
0
-			else
0
-				msg = "Bad markdown format near %p" % str
0
-			end
0
-
0
-			super( msg )
0
-		end
0
-	end
0
-
0
 
0
+# BlueCloth is a Ruby implementation of Markdown, a text-to-HTML conversion
0
+# tool.
0
+class BlueCloth
0
+  
0
   # Release Version
0
   Version = '0.0.3'
0
 
0
-	# SVN Revision
0
-	SvnRev = %q$Rev$
0
-
0
-	# SVN Id tag
0
-	SvnId = %q$Id$
0
-
0
-	# SVN URL
0
-	SvnUrl = %q$URL$
0
-
0
-
0
-	# Rendering state struct. Keeps track of URLs, titles, and HTML blocks
0
-	# midway through a render. I prefer this to the globals of the Perl version
0
-	# because globals make me break out in hives. Or something.
0
-	RenderState = Struct::new( "RenderState", :urls, :titles, :html_blocks, :log )
0
-
0
-	# Tab width for #detab! if none is specified
0
-	TabWidth = 4
0
-
0
-	# The tag-closing string -- set to '>' for HTML
0
-	EmptyElementSuffix = "/>";
0
-
0
-	# Table of MD5 sums for escaped characters
0
-	EscapeTable = {}
0
-	'\\`*_{}[]()#.!'.split(//).each {|char|
0
-		hash = Digest::MD5::hexdigest( char )
0
-
0
-		EscapeTable[ char ] = {
0
- 			:md5 => hash,
0
-			:md5re => Regexp::new( hash ),
0
-			:re  => Regexp::new( '\\\\' + Regexp::escape(char) ),
0
-		}
0
-	}
0
-
0
-
0
-	#################################################################
0
-	###	I N S T A N C E   M E T H O D S
0
-	#################################################################
0
-
0
-	### Create a new BlueCloth string.
0
-	def initialize( content="", *restrictions )
0
+	# Create a new BlueCloth string.
0
+	def initialize(content = "", *restrictions)
0
     @log = Logger::new( $deferr )
0
     @log.level = $DEBUG ?
0
       Logger::DEBUG :
0
@@ -137,16 +73,11 @@ class BlueCloth < String
0
     restrictions.flatten.each {|r| __send__("#{r}=", true) }
0
     @fold_lines = true
0
 
0
-		super( content )
0
+    @content = content
0
 
0
     @log.debug "String is: %p" % self
0
   end
0
 
0
-
0
-	######
0
-	public
0
-	######
0
-
0
   # Filters for controlling what gets output for untrusted input. (But really,
0
   # you're filtering bad stuff out of untrusted input at submission-time via
0
   # untainting, aren't you?)
0
@@ -156,989 +87,7 @@ class BlueCloth < String
0
   # so this isn't used by anything.
0
   attr_accessor :fold_lines
0
 
0
+end
0
 
0
-	### Render Markdown-formatted text in this string object as HTML and return
0
-	### it. The parameter is for compatibility with RedCloth, and is currently
0
-	### unused, though that may change in the future.
0
-	def to_html( lite=false )
0
-
0
-		# Create a StringScanner we can reuse for various lexing tasks
0
-		@scanner = StringScanner::new( '' )
0
-
0
-		# Make a structure to carry around stuff that gets placeholdered out of
0
-		# the source.
0
-		rs = RenderState::new( {}, {}, {} )
0
-
0
-		# Make a copy of the string with normalized line endings, tabs turned to
0
-		# spaces, and a couple of guaranteed newlines at the end
0
-		text = self.gsub( /\r\n?/, "\n" ).detab
0
-		text += "\n\n"
0
-		@log.debug "Normalized line-endings: %p" % text
0
-
0
-		# Filter HTML if we're asked to do so
0
-		if self.filter_html
0
-			text.gsub!( "<", "&lt;" )
0
-			text.gsub!( ">", "&gt;" )
0
-			@log.debug "Filtered HTML: %p" % text
0
-		end
0
-
0
-		# Simplify blank lines
0
-		text.gsub!( /^ +$/, '' )
0
-		@log.debug "Tabs -> spaces/blank lines stripped: %p" % text
0
-
0
-		# Replace HTML blocks with placeholders
0
-		text = hide_html_blocks( text, rs )
0
-		@log.debug "Hid HTML blocks: %p" % text
0
-		@log.debug "Render state: %p" % rs
0
-
0
-		# Strip link definitions, store in render state
0
-		text = strip_link_definitions( text, rs )
0
-		@log.debug "Stripped link definitions: %p" % text
0
-		@log.debug "Render state: %p" % rs
0
-
0
-		# Escape meta-characters
0
-		text = escape_special_chars( text )
0
-		@log.debug "Escaped special characters: %p" % text
0
-
0
-		# Transform block-level constructs
0
-		text = apply_block_transforms( text, rs )
0
-		@log.debug "After block-level transforms: %p" % text
0
-
0
-		# Now swap back in all the escaped characters
0
-		text = unescape_special_chars( text )
0
-		@log.debug "After unescaping special characters: %p" % text
0
-
0
-		return text
0
-	end
0
-	
0
-
0
-	### Convert tabs in +str+ to spaces.
0
-	def detab( tabwidth=TabWidth )
0
-		copy = self.dup
0
-		copy.detab!( tabwidth )
0
-		return copy
0
-	end
0
-
0
-
0
-	### Convert tabs to spaces in place and return self if any were converted.
0
-	def detab!( tabwidth=TabWidth )
0
-		newstr = self.split( /\n/ ).collect {|line|
0
-			line.gsub( /(.*?)\t/ ) do
0
-				$1 + ' ' * (tabwidth - $1.length % tabwidth)
0
-			end
0
-		}.join("\n")
0
-		self.replace( newstr )
0
-	end
0
-
0
-
0
-	#######
0
-	#private
0
-	#######
0
-
0
-	### Do block-level transforms on a copy of +str+ using the specified render
0
-	### state +rs+ and return the results.
0
-	def apply_block_transforms( str, rs )
0
-		# Port: This was called '_runBlockGamut' in the original
0
-
0
-		@log.debug "Applying block transforms to:\n  %p" % str
0
-		text = transform_headers( str, rs )
0
-		text = transform_hrules( text, rs )
0
-		text = transform_lists( text, rs )
0
-		text = transform_code_blocks( text, rs )
0
-		text = transform_block_quotes( text, rs )
0
-		text = transform_auto_links( text, rs )
0
-		text = hide_html_blocks( text, rs )
0
-
0
-		text = form_paragraphs( text, rs )
0
-
0
-		@log.debug "Done with block transforms:\n  %p" % text
0
-		return text
0
-	end
0
-
0
-
0
-	### Apply Markdown span transforms to a copy of the specified +str+ with the
0
-	### given render state +rs+ and return it.
0
-	def apply_span_transforms( str, rs )
0
-		@log.debug "Applying span transforms to:\n  %p" % str
0
-
0
-		str = transform_code_spans( str, rs )
0
-		str = encode_html( str )
0
-		str = transform_images( str, rs )
0
-		str = transform_anchors( str, rs )
0
-		str = transform_italic_and_bold( str, rs )
0
-
0
-		# Hard breaks
0
-		str.gsub!( / {2,}\n/, "<br#{EmptyElementSuffix}\n" )
0
-
0
-		@log.debug "Done with span transforms:\n  %p" % str
0
-		return str
0
-	end
0
-
0
-
0
-	# The list of tags which are considered block-level constructs and an
0
-	# alternation pattern suitable for use in regexps made from the list
0
-	StrictBlockTags = %w[ p div h[1-6] blockquote pre table dl ol ul script noscript
0
-		form fieldset iframe math ins del ]
0
-	StrictTagPattern = StrictBlockTags.join('|')
0
-
0
-	LooseBlockTags = StrictBlockTags - %w[ins del]
0
-	LooseTagPattern = LooseBlockTags.join('|')
0
-
0
-	# Nested blocks:
0
-	# 	<div>
0
-	# 		<div>
0
-	# 		tags for inner block must be indented.
0
-	# 		</div>
0
-	# 	</div>
0
-	StrictBlockRegex = %r{
0
-		^						# Start of line
0
-		<(#{StrictTagPattern})	# Start tag: \2
0
-		\b						# word break
0
-		(.*\n)*?				# Any number of lines, minimal match
0
-		</\1>					# Matching end tag
0
-		[ ]*					# trailing spaces
0
-		$						# End of line or document
0
-	  }ix
0
-
0
-	# More-liberal block-matching
0
-	LooseBlockRegex = %r{
0
-		^						# Start of line
0
-		<(#{LooseTagPattern})	# start tag: \2
0
-		\b						# word break
0
-		(.*\n)*?				# Any number of lines, minimal match
0
-		.*</\1>					# Anything + Matching end tag
0
-		[ ]*					# trailing spaces
0
-		$						# End of line or document
0
-	  }ix
0
-
0
-	# Special case for <hr />.
0
-	HruleBlockRegex = %r{
0
-		(						# $1
0
-			\A\n?				# Start of doc + optional \n
0
-			|					# or
0
-			.*\n\n				# anything + blank line
0
-		)
0
-		(						# save in $2
0
-			[ ]*				# Any spaces
0
-			<hr					# Tag open
0
-			\b					# Word break
0
-			([^<>])*?			# Attributes
0
-			/?>					# Tag close
0
-			$					# followed by a blank line or end of document
0
-		)
0
-	  }ix
0
-
0
-	### Replace all blocks of HTML in +str+ that start in the left margin with
0
-	### tokens.
0
-	def hide_html_blocks( str, rs )
0
-		@log.debug "Hiding HTML blocks in %p" % str
0
-		
0
-		# Tokenizer proc to pass to gsub
0
-		tokenize = lambda {|match|
0
-			key = Digest::MD5::hexdigest( match )
0
-			rs.html_blocks[ key ] = match
0
-			@log.debug "Replacing %p with %p" % [ match, key ]
0
-			"\n\n#{key}\n\n"
0
-		}
0
-
0
-		rval = str.dup
0
-
0
-		@log.debug "Finding blocks with the strict regex..."
0
-		rval.gsub!( StrictBlockRegex, &tokenize )
0
-
0
-		@log.debug "Finding blocks with the loose regex..."
0
-		rval.gsub!( LooseBlockRegex, &tokenize )
0
-
0
-		@log.debug "Finding hrules..."
0
-		rval.gsub!( HruleBlockRegex ) {|match| $1 + tokenize[$2] }
0
-
0
-		return rval
0
-	end
0
-
0
-
0
-	# Link defs are in the form: ^[id]: url "optional title"
0
-	LinkRegex = %r{
0
-		^[ ]*\[(.+)\]:		# id = $1
0
-		  [ ]*
0
-		  \n?				# maybe *one* newline
0
-		  [ ]*
0
-		<?(\S+?)>?				# url = $2
0
-		  [ ]*
0
-		  \n?				# maybe one newline
0
-		  [ ]*
0
-		(?:
0
-			# Titles are delimited by "quotes" or (parens).
0
-			["(]
0
-			(.+?)			# title = $3
0
-			[")]			# Matching ) or "
0
-			[ ]*
0
-		)?	# title is optional
0
-		(?:\n+|\Z)
0
-	  }x
0
-
0
-	### Strip link definitions from +str+, storing them in the given RenderState
0
-	### +rs+.
0
-	def strip_link_definitions( str, rs )
0
-		str.gsub( LinkRegex ) {|match|
0
-			id, url, title = $1, $2, $3
0
-
0
-			rs.urls[ id.downcase ] = encode_html( url )
0
-			unless title.nil?
0
-				rs.titles[ id.downcase ] = title.gsub( /"/, "&quot;" )
0
-			end
0
-			""
0
-		}
0
-	end
0
-
0
-
0
-	### Escape special characters in the given +str+
0
-	def escape_special_chars( str )
0
-		@log.debug "  Escaping special characters"
0
-		text = ''
0
-
0
-		# The original Markdown source has something called '$tags_to_skip'
0
-		# declared here, but it's never used, so I don't define it.
0
-
0
-		tokenize_html( str ) {|token, str|
0
-			@log.debug "   Adding %p token %p" % [ token, str ]
0
-			case token
0
-
0
-			# Within tags, encode * and _
0
-			when :tag
0
-				text += str.
0
-					gsub( /\*/, EscapeTable['*'][:md5] ).
0
-					gsub( /_/, EscapeTable['_'][:md5] )
0
-
0
-			# Encode backslashed stuff in regular text
0
-			when :text
0
-				text += encode_backslash_escapes( str )
0
-			else
0
-				raise TypeError, "Unknown token type %p" % token
0
-			end
0
-		}
0
-
0
-		@log.debug "  Text with escapes is now: %p" % text
0
-		return text
0
-	end
0
-
0
-
0
-	### Swap escaped special characters in a copy of the given +str+ and return
0
-	### it.
0
-	def unescape_special_chars( str )
0
-		EscapeTable.each {|char, hash|
0
-			@log.debug "Unescaping escaped %p with %p" % [ char, hash[:md5re] ]
0
-			str.gsub!( hash[:md5re], char )
0
-		}
0
-
0
-		return str
0
-	end
0
-
0
-
0
-	### Return a copy of the given +str+ with any backslashed special character
0
-	### in it replaced with MD5 placeholders.
0
-	def encode_backslash_escapes( str )
0
-		# Make a copy with any double-escaped backslashes encoded
0
-		text = str.gsub( /\\\\/, EscapeTable['\\'][:md5] )
0
-		
0
-		EscapeTable.each_pair {|char, esc|
0
-			next if char == '\\'
0
-			text.gsub!( esc[:re], esc[:md5] )
0
-		}
0
-
0
-		return text
0
-	end
0
-
0
-
0
-	### Transform any Markdown-style horizontal rules in a copy of the specified
0
-	### +str+ and return it.
0
-	def transform_hrules( str, rs )
0
-		@log.debug " Transforming horizontal rules"
0
-		str.gsub( /^( ?[\-\*_] ?){3,}$/, "\n<hr#{EmptyElementSuffix}\n" )
0
-	end
0
-
0
-
0
-
0
-	# Patterns to match and transform lists
0
-	ListMarkerOl = %r{\d+\.}
0
-	ListMarkerUl = %r{[*+-]}
0
-	ListMarkerAny = Regexp::union( ListMarkerOl, ListMarkerUl )
0
-
0
-	ListRegexp = %r{
0
-		  (?:
0
-			^[ ]{0,#{TabWidth - 1}}		# Indent < tab width
0
-			(#{ListMarkerAny})			# unordered or ordered ($1)
0
-			[ ]+						# At least one space
0
-		  )
0
-		  (?m:.+?)						# item content (include newlines)
0
-		  (?:
0
-			  \z						# Either EOF
0
-			|							#  or
0
-			  \n{2,}					# Blank line...
0
-			  (?=\S)					# ...followed by non-space
0
-			  (?![ ]*					# ...but not another item
0
-				(#{ListMarkerAny})
0
-			   [ ]+)
0
-		  )
0
-	  }x
0
-
0
-	### Transform Markdown-style lists in a copy of the specified +str+ and
0
-	### return it.
0
-	def transform_lists( str, rs )
0
-		@log.debug " Transforming lists at %p" % (str[0,100] + '...')
0
-
0
-		str.gsub( ListRegexp ) {|list|
0
-			@log.debug "  Found list %p" % list
0
-			bullet = $1
0
-			list_type = (ListMarkerUl.match(bullet) ? "ul" : "ol")
0
-			list.gsub!( /\n{2,}/, "\n\n\n" )
0
-
0
-			%{<%s>\n%s</%s>\n} % [
0
-				list_type,
0
-				transform_list_items( list, rs ),
0
-				list_type,
0
-			]
0
-		}
0
-	end
0
-
0
-
0
-	# Pattern for transforming list items
0
-	ListItemRegexp = %r{
0
-		(\n)?							# leading line = $1
0
-		(^[ ]*)							# leading whitespace = $2
0
-		(#{ListMarkerAny}) [ ]+			# list marker = $3
0
-		((?m:.+?)						# list item text   = $4
0
-		(\n{1,2}))
0
-		(?= \n* (\z | \2 (#{ListMarkerAny}) [ ]+))
0
-	  }x
0
-
0
-	### Transform list items in a copy of the given +str+ and return it.
0
-	def transform_list_items( str, rs )
0
-		@log.debug " Transforming list items"
0
-
0
-		# Trim trailing blank lines
0
-		str = str.sub( /\n{2,}\z/, "\n" )
0
-
0
-		str.gsub( ListItemRegexp ) {|line|
0
-			@log.debug "  Found item line %p" % line
0
-			leading_line, item = $1, $4
0
-
0
-			if leading_line or /\n{2,}/.match( item )
0
-				@log.debug "   Found leading line or item has a blank"
0
-				item = apply_block_transforms( outdent(item), rs )
0
-			else
0
-				# Recursion for sub-lists
0
-				@log.debug "   Recursing for sublist"
0
-				item = transform_lists( outdent(item), rs ).chomp
0
-				item = apply_span_transforms( item, rs )
0
-			end
0
-
0
-			%{<li>%s</li>\n} % item
0
-		}
0
-	end
0
-
0
-
0
-	# Pattern for matching codeblocks
0
-	CodeBlockRegexp = %r{
0
-		(?:\n\n|\A)
0
-		(									# $1 = the code block
0
-		  (?:
0
-			(?:[ ]{#{TabWidth}} | \t)		# a tab or tab-width of spaces
0
-			.*\n+
0
-		  )+
0
-		)
0
-		(^[ ]{0,#{TabWidth - 1}}\S|\Z)		# Lookahead for non-space at
0
-											# line-start, or end of doc
0
-	  }x
0
-
0
-	### Transform Markdown-style codeblocks in a copy of the specified +str+ and
0
-	### return it.
0
-	def transform_code_blocks( str, rs )
0
-		@log.debug " Transforming code blocks"
0
-
0
-		str.gsub( CodeBlockRegexp ) {|block|
0
-			codeblock = $1
0
-			remainder = $2
0
-
0
-			# Generate the codeblock
0
-			%{\n\n<pre><code>%s\n</code></pre>\n\n%s} %
0
-				[ encode_code( outdent(codeblock), rs ).rstrip, remainder ]
0
-		}
0
-	end
0
-
0
-
0
-	# Pattern for matching Markdown blockquote blocks
0
-	BlockQuoteRegexp = %r{
0
-		  (?:
0
-			^[ ]*>[ ]?		# '>' at the start of a line
0
-			  .+\n			# rest of the first line
0
-			(?:.+\n)*		# subsequent consecutive lines
0
-			\n*				# blanks
0
-		  )+
0
-	  }x
0
-	PreChunk = %r{ ( ^ \s* <pre> .+? </pre> ) }xm
0
-
0
-	### Transform Markdown-style blockquotes in a copy of the specified +str+
0
-	### and return it.
0
-	def transform_block_quotes( str, rs )
0
-		@log.debug " Transforming block quotes"
0
-
0
-		str.gsub( BlockQuoteRegexp ) {|quote|
0
-			@log.debug "Making blockquote from %p" % quote
0
-
0
-			quote.gsub!( /^ *> ?/, '' ) # Trim one level of quoting 
0
-			quote.gsub!( /^ +$/, '' )	# Trim whitespace-only lines
0
-
0
-			indent = " " * TabWidth
0
-			quoted = %{<blockquote>\n%s\n</blockquote>\n\n} %
0
-				apply_block_transforms( quote, rs ).
0
-				gsub( /^/, indent ).
0
-				gsub( PreChunk ) {|m| m.gsub(/^#{indent}/o, '') }
0
-			@log.debug "Blockquoted chunk is: %p" % quoted
0
-			quoted
0
-		}
0
-	end
0
-
0
-
0
-	AutoAnchorURLRegexp = /<((https?|ftp):[^'">\s]+)>/
0
-	AutoAnchorEmailRegexp = %r{
0
-		<
0
-		(
0
-			[-.\w]+
0
-			\@
0
-			[-a-z0-9]+(\.[-a-z0-9]+)*\.[a-z]+
0
-		)
0
-		>
0
-	  }xi
0
-
0
-	### Transform URLs in a copy of the specified +str+ into links and return
0
-	### it.
0
-	def transform_auto_links( str, rs )
0
-