#                                vim:ts=4:sw=4:
# = RedCloth - Textile and Markdown Hybrid for Ruby
#
# Homepage::  http://whytheluckystiff.net/ruby/redcloth/
# Author::    why the lucky stiff (http://whytheluckystiff.net/)
# Copyright:: (cc) 2004 why the lucky stiff (and his puppet organizations.)
# License::   BSD
#
# (see http://hobix.com/textile/ for a Textile Reference.)
#
# Based on (and also inspired by) both:
#
# PyTextile: http://diveintomark.org/projects/textile/textile.py.txt
# Textism for PHP: http://www.textism.com/tools/textile/
#
#
# = RedCloth
#
# RedCloth is a Ruby library for converting Textile and/or Markdown
# into HTML.  You can use either format, intermingled or separately.
# You can also extend RedCloth to honor your own custom text stylings.
#
# RedCloth users are encouraged to use Textile if they are generating
# HTML and to use Markdown if others will be viewing the plain text.
#
# == What is Textile?
#
# Textile is a simple formatting style for text
# documents, loosely based on some HTML conventions.
#
# == Sample Textile Text
#
#  h2. This is a title
#
#  h3. This is a subhead
#
#  This is a bit of paragraph.
#
#  bq. This is a blockquote.
#
# = Writing Textile
#
# A Textile document consists of paragraphs.  Paragraphs
# can be specially formatted by adding a small instruction
# to the beginning of the paragraph.
#
#  h[n].   Header of size [n].
#  bq.     Blockquote.
#  #       Numeric list.
#  *       Bulleted list.
#
# == Quick Phrase Modifiers
#
# Quick phrase modifiers are also included, to allow formatting
# of small portions of text within a paragraph.
#
#  \_emphasis\_
#  \_\_italicized\_\_
#  \*strong\*
#  \*\*bold\*\*
#  ??citation??
#  -deleted text-
#  +inserted text+
#  ^superscript^
#  ~subscript~
#  @code@
#  %(classname)span%
#
#  ==notextile== (leave text alone)
#
# == Links
#
# To make a hypertext link, put the link text in "quotation 
# marks" followed immediately by a colon and the URL of the link.
# 
# Optional: text in (parentheses) following the link text, 
# but before the closing quotation mark, will become a Title 
# attribute for the link, visible as a tool tip when a cursor is above it.
# 
# Example:
#
#  "This is a link (This is a title) ":http://www.textism.com
# 
# Will become:
# 
#  This is a link
#
# == Images
#
# To insert an image, put the URL for the image inside exclamation marks.
#
# Optional: text that immediately follows the URL in (parentheses) will 
# be used as the Alt text for the image. Images on the web should always 
# have descriptive Alt text for the benefit of readers using non-graphical 
# browsers.
#
# Optional: place a colon followed by a URL immediately after the 
# closing ! to make the image into a link.
# 
# Example:
#
#  !http://www.textism.com/common/textist.gif(Textist)!
#
# Will become:
#
#   #
# With a link:
#
#  !/common/textist.gif(Textist)!:http://textism.com
#
# Will become:
#
#
#
# With a link:
#
#  !/common/textist.gif(Textist)!:http://textism.com
#
# Will become:
#
#   #
# == Defining Acronyms
#
# HTML allows authors to define acronyms via the tag. The definition appears as a 
# tool tip when a cursor hovers over the acronym. A crucial aid to clear writing, 
# this should be used at least once for each acronym in documents where they appear.
#
# To quickly define an acronym in Textile, place the full text in (parentheses) 
# immediately following the acronym.
# 
# Example:
#
#  ACLU(American Civil Liberties Union)
#
# Will become:
#
#  ACLU
#
# == Adding Tables
#
# In Textile, simple tables can be added by seperating each column by
# a pipe.
#
#     |a|simple|table|row|
#     |And|Another|table|row|
#
# Attributes are defined by style definitions in parentheses.
#
#     table(border:1px solid black).
#     (background:#ddd;color:red). |{}| | | |
#
# == Using RedCloth
# 
# RedCloth is simply an extension of the String class, which can handle
# Textile formatting.  Use it like a String and output HTML with its
# RedCloth#to_html method.
#
#  doc = RedCloth.new "
#
#  h2. Test document
#
#  Just a simple test."
#
#  puts doc.to_html
#
# By default, RedCloth uses both Textile and Markdown formatting, with
# Textile formatting taking precedence.  If you want to turn off Markdown
# formatting, to boost speed and limit the processor:
#
#  class RedCloth::Textile.new( str )
class RedCloth3 < String
  VERSION = '3.0.4'
  DEFAULT_RULES = [:textile, :markdown]
  #
  # Two accessor for setting security restrictions.
  #
  # This is a nice thing if you're using RedCloth for
  # formatting in public places (e.g. Wikis) where you
  # don't want users to abuse HTML for bad things.
  #
  # If +:filter_html+ is set, HTML which wasn't
  # created by the Textile processor will be escaped.
  #
  # If +:filter_styles+ is set, it will also disable
  # the style markup specifier. ('{color: red}')
  #
  attr_accessor :filter_html, :filter_styles
  #
  # Accessor for toggling hard breaks.
  #
  # If +:hard_breaks+ is set, single newlines will
  # be converted to HTML break tags.  This is the
  # default behavior for traditional RedCloth.
  #
  attr_accessor :hard_breaks
  # Accessor for toggling lite mode.
  #
  # In lite mode, block-level rules are ignored.  This means
  # that tables, paragraphs, lists, and such aren't available.
  # Only the inline markup for bold, italics, entities and so on.
  #
  #   r = RedCloth.new( "And then? She *fell*!", [:lite_mode] )
  #   r.to_html
  #   #=> "And then? She fell!"
  #
  attr_accessor :lite_mode
  #
  # Accessor for toggling span caps.
  #
  # Textile places `span' tags around capitalized
  # words by default, but this wreaks havoc on Wikis.
  # If +:no_span_caps+ is set, this will be
  # suppressed.
  #
  attr_accessor :no_span_caps
  #
  # Establishes the markup predence.  Available rules include:
  #
  # == Textile Rules
  #
  # The following textile rules can be set individually.  Or add the complete
  # set of rules with the single :textile rule, which supplies the rule set in
  # the following precedence:
  #
  # refs_textile::          Textile references (i.e. [hobix]http://hobix.com/)
  # block_textile_table::   Textile table block structures
  # block_textile_lists::   Textile list structures
  # block_textile_prefix::  Textile blocks with prefixes (i.e. bq., h2., etc.)
  # inline_textile_image::  Textile inline images
  # inline_textile_link::   Textile inline links
  # inline_textile_span::   Textile inline spans
  # glyphs_textile:: Textile entities (such as em-dashes and smart quotes)
  #
  # == Markdown
  #
  # refs_markdown::         Markdown references (for example: [hobix]: http://hobix.com/)
  # block_markdown_setext:: Markdown setext headers
  # block_markdown_atx::    Markdown atx headers
  # block_markdown_rule::   Markdown horizontal rules
  # block_markdown_bq::     Markdown blockquotes
  # block_markdown_lists::  Markdown lists
  # inline_markdown_link::  Markdown links
  attr_accessor :rules
  # Returns a new RedCloth object, based on _string_ and
  # enforcing all the included _restrictions_.
  #
  #   r = RedCloth.new( "h1. A bold man", [:filter_html] )
  #   r.to_html
  #     #=>"
#
# == Defining Acronyms
#
# HTML allows authors to define acronyms via the tag. The definition appears as a 
# tool tip when a cursor hovers over the acronym. A crucial aid to clear writing, 
# this should be used at least once for each acronym in documents where they appear.
#
# To quickly define an acronym in Textile, place the full text in (parentheses) 
# immediately following the acronym.
# 
# Example:
#
#  ACLU(American Civil Liberties Union)
#
# Will become:
#
#  ACLU
#
# == Adding Tables
#
# In Textile, simple tables can be added by seperating each column by
# a pipe.
#
#     |a|simple|table|row|
#     |And|Another|table|row|
#
# Attributes are defined by style definitions in parentheses.
#
#     table(border:1px solid black).
#     (background:#ddd;color:red). |{}| | | |
#
# == Using RedCloth
# 
# RedCloth is simply an extension of the String class, which can handle
# Textile formatting.  Use it like a String and output HTML with its
# RedCloth#to_html method.
#
#  doc = RedCloth.new "
#
#  h2. Test document
#
#  Just a simple test."
#
#  puts doc.to_html
#
# By default, RedCloth uses both Textile and Markdown formatting, with
# Textile formatting taking precedence.  If you want to turn off Markdown
# formatting, to boost speed and limit the processor:
#
#  class RedCloth::Textile.new( str )
class RedCloth3 < String
  VERSION = '3.0.4'
  DEFAULT_RULES = [:textile, :markdown]
  #
  # Two accessor for setting security restrictions.
  #
  # This is a nice thing if you're using RedCloth for
  # formatting in public places (e.g. Wikis) where you
  # don't want users to abuse HTML for bad things.
  #
  # If +:filter_html+ is set, HTML which wasn't
  # created by the Textile processor will be escaped.
  #
  # If +:filter_styles+ is set, it will also disable
  # the style markup specifier. ('{color: red}')
  #
  attr_accessor :filter_html, :filter_styles
  #
  # Accessor for toggling hard breaks.
  #
  # If +:hard_breaks+ is set, single newlines will
  # be converted to HTML break tags.  This is the
  # default behavior for traditional RedCloth.
  #
  attr_accessor :hard_breaks
  # Accessor for toggling lite mode.
  #
  # In lite mode, block-level rules are ignored.  This means
  # that tables, paragraphs, lists, and such aren't available.
  # Only the inline markup for bold, italics, entities and so on.
  #
  #   r = RedCloth.new( "And then? She *fell*!", [:lite_mode] )
  #   r.to_html
  #   #=> "And then? She fell!"
  #
  attr_accessor :lite_mode
  #
  # Accessor for toggling span caps.
  #
  # Textile places `span' tags around capitalized
  # words by default, but this wreaks havoc on Wikis.
  # If +:no_span_caps+ is set, this will be
  # suppressed.
  #
  attr_accessor :no_span_caps
  #
  # Establishes the markup predence.  Available rules include:
  #
  # == Textile Rules
  #
  # The following textile rules can be set individually.  Or add the complete
  # set of rules with the single :textile rule, which supplies the rule set in
  # the following precedence:
  #
  # refs_textile::          Textile references (i.e. [hobix]http://hobix.com/)
  # block_textile_table::   Textile table block structures
  # block_textile_lists::   Textile list structures
  # block_textile_prefix::  Textile blocks with prefixes (i.e. bq., h2., etc.)
  # inline_textile_image::  Textile inline images
  # inline_textile_link::   Textile inline links
  # inline_textile_span::   Textile inline spans
  # glyphs_textile:: Textile entities (such as em-dashes and smart quotes)
  #
  # == Markdown
  #
  # refs_markdown::         Markdown references (for example: [hobix]: http://hobix.com/)
  # block_markdown_setext:: Markdown setext headers
  # block_markdown_atx::    Markdown atx headers
  # block_markdown_rule::   Markdown horizontal rules
  # block_markdown_bq::     Markdown blockquotes
  # block_markdown_lists::  Markdown lists
  # inline_markdown_link::  Markdown links
  attr_accessor :rules
  # Returns a new RedCloth object, based on _string_ and
  # enforcing all the included _restrictions_.
  #
  #   r = RedCloth.new( "h1. A bold man", [:filter_html] )
  #   r.to_html
  #     #=>"
' * (l-indent) : '' * (indent-l)) + "\n\n") indent = l end quotes << (content + "\n") end quotes << ("\n" + '' * indent + "\n\n") quotes end end CODE_RE = /(\W) @ (?:\|(\w+?)\|)? (.+?) @ (?=\W)/x def inline_textile_code( text ) text.gsub!( CODE_RE ) do |m| before,lang,code,after = $~[1..4] lang = " lang=\"#{ lang }\"" if lang rip_offtags( "#{ before }
#{ code }#{ after }", false )
    end
  end
  def lT( text )
    text =~ /\#$/ ? 'o' : 'u'
  end
  def hard_break( text )
    text.gsub!( /(.)\n(?!\Z| *([#*=]+(\s|$)|[{|]))/, "\\1#{ blk }#{ blk }
" end end # hard_break blk blk + "\n#{ code_blk }" end end end.join( "\n\n" ) ) end def textile_bq( tag, atts, cite, content ) cite, cite_title = check_refs( cite ) cite = " cite=\"#{ cite }\"" if cite atts = shelve( atts ) if atts "\t\n\t\t" end def textile_p( tag, atts, cite, content ) atts = shelve( atts ) if atts "\t<#{ tag }#{ atts }>#{ content }#{ tag }>" end alias textile_h1 textile_p alias textile_h2 textile_p alias textile_h3 textile_p alias textile_h4 textile_p alias textile_h5 textile_p alias textile_h6 textile_p def textile_fn_( tag, num, atts, cite, content ) atts << " id=\"fn#{ num }\" class=\"footnote\"" content = "#{ num } #{ content }" atts = shelve( atts ) if atts "\t#{ content }
\n\t
#{ content }
" end BLOCK_RE = /^(([a-z]+)(\d*))(#{A}#{C})\.(?::(\S+))? (.*)$/m def block_textile_prefix( text ) if text =~ BLOCK_RE tag,tagpre,num,atts,cite,content = $~[1..6] atts = pba( atts ) # pass to prefix handler replacement = nil if respond_to? "textile_#{ tag }", true replacement = method( "textile_#{ tag }" ).call( tag, atts, cite, content ) elsif respond_to? "textile_#{ tagpre }_", true replacement = method( "textile_#{ tagpre }_" ).call( tagpre, num, atts, cite, content ) end text.gsub!( $& ) { replacement } if replacement end end SETEXT_RE = /\A(.+?)\n([=-])[=-]* *$/m def block_markdown_setext( text ) if text =~ SETEXT_RE tag = if $2 == "="; "h1"; else; "h2"; end blk, cont = "<#{ tag }>#{ $1 }#{ tag }>", $' blocks cont text.replace( blk + cont ) end end ATX_RE = /\A(\#{1,6}) # $1 = string of #'s [ ]* (.+?) # $2 = Header text [ ]* \#* # optional closing #'s (not counted) $/x def block_markdown_atx( text ) if text =~ ATX_RE tag = "h#{ $1.length }" blk, cont = "<#{ tag }>#{ $2 }#{ tag }>\n\n", $' blocks cont text.replace( blk + cont ) end end MARKDOWN_BQ_RE = /\A(^ *> ?.+$(.+\n)*\n*)+/m def block_markdown_bq( text ) text.gsub!( MARKDOWN_BQ_RE ) do |blk| blk.gsub!( /^ *> ?/, '' ) flush_left blk blocks blk blk.gsub!( /^(\S)/, "\t\\1" ) "\n#{ blk }\n\n\n" end end MARKDOWN_RULE_RE = /^(#{ ['*', '-', '_'].collect { |ch| ' ?(' + Regexp::quote( ch ) + ' ?){3,}' }.join( '|' ) })$/ def block_markdown_rule( text ) text.gsub!( MARKDOWN_RULE_RE ) do |blk| "
" out = "
#{ out }" else out = "#{ stln }
,  etc.
        if $1
          if line =~ OFFTAG_OPEN
            codepre += 1
          elsif line =~ OFFTAG_CLOSE
            codepre -= 1
            codepre = 0 if codepre < 0
          end
        elsif codepre.zero?
          glyphs_textile( line, level + 1 )
        else
          htmlesc( line, :NoQuotes )
        end
        # p [level, codepre, line]
        line
      end
    end
  end
  def rip_offtags( text, escape_aftertag=true, escape_line=true )
    if text =~ /<.*>/
      ## strip and encode  content
      codepre, used_offtags = 0, {}
      text.gsub!( OFFTAG_MATCH ) do |line|
        if $3
          first, offtag, aftertag = $3, $4, $5
          codepre += 1
          used_offtags[offtag] = true
          if codepre - used_offtags.length > 0
            htmlesc( line, :NoQuotes ) if escape_line
            @pre_list.last << line
            line = ""
          else
            ### htmlesc is disabled between CODE tags which will be parsed with highlighter
            ### Regexp in formatter.rb is : /\s?(.+)/m
            ### NB: some changes were made not to use $N variables, because we use "match"
            ###   and it breaks following lines
            htmlesc( aftertag, :NoQuotes ) if aftertag && escape_aftertag && !first.match(//)
            line = ""
            first.match(/<#{ OFFTAGS }([^>]*)>/)
            tag = $1
            $2.to_s.match(/(class\=("[^"]+"|'[^']+'))/i)
            tag << " #{$1}" if $1
            @pre_list << "<#{ tag }>#{ aftertag }"
          end
        elsif $1 and codepre > 0
          if codepre - used_offtags.length > 0
            htmlesc( line, :NoQuotes ) if escape_line
            @pre_list.last << line
            line = ""
          end
          codepre -= 1 unless codepre.zero?
          used_offtags = {} if codepre.zero?
        end
        line
      end
    end
    text
  end
  def smooth_offtags( text )
    unless @pre_list.empty?
      ## replace  content
      text.gsub!( // ) { @pre_list[$1.to_i] }
    end
  end
  def inline( text )
    [/^inline_/, /^glyphs_/].each do |meth_re|
      @rules.each do |rule_name|
        method( rule_name ).call( text ) if rule_name.to_s.match( meth_re )
      end
    end
  end
  def h_align( text )
    H_ALGN_VALS[text]
  end
  def v_align( text )
    V_ALGN_VALS[text]
  end
  def textile_popup_help( name, windowW, windowH )
    ' ' + name + '
'
  end
  # HTML cleansing stuff
  BASIC_TAGS = {
      'a' => ['href', 'title'],
      'img' => ['src', 'alt', 'title'],
      'br' => [],
      'i' => nil,
      'u' => nil,
      'b' => nil,
      'pre' => nil,
      'kbd' => nil,
      'code' => ['lang'],
      'cite' => nil,
      'strong' => nil,
      'em' => nil,
      'ins' => nil,
      'sup' => nil,
      'sub' => nil,
      'del' => nil,
      'table' => nil,
      'tr' => nil,
      'td' => ['colspan', 'rowspan'],
      'th' => nil,
      'ol' => nil,
      'ul' => nil,
      'li' => nil,
      'p' => nil,
      'h1' => nil,
      'h2' => nil,
      'h3' => nil,
      'h4' => nil,
      'h5' => nil,
      'h6' => nil,
      'blockquote' => ['cite']
  }
  def clean_html( text, tags = BASIC_TAGS )
    text.gsub!( /]*)>/ ) do
      raw = $~
      tag = raw[2].downcase
      if tags.has_key? tag
        pcs = [tag]
        tags[tag].each do |prop|
          ['"', "'", ''].each do |q|
            q2 = ( q != '' ? q : '\s' )
            if raw[3] =~ /#{prop}\s*=\s*#{q}([^#{q2}]+)#{q}/i
              attrv = $1
              next if prop == 'src' and attrv =~ %r{^(?!http)\w+:}
              pcs << "#{prop}=\"#{$1.gsub('"', '\\"')}\""
              break
            end
          end
        end if tags[tag]
        "<#{raw[1]}#{pcs.join " "}>"
      else
        " "
      end
    end
  end
  ALLOWED_TAGS = %w(redpre pre code notextile)
  def escape_html_tags(text)
    text.gsub!(%r{<(\/?([!\w]+)[^<>\n]*)(>?)}) {|m| ALLOWED_TAGS.include?($2) ? "<#{$1}#{$3}" : "<#{$1}#{'>' unless $3.blank?}" }
  end
end