redcloth.rb

来自「用ruby on rails写的一个博客程序,还不错..ruby on rail」· RB 代码 · 共 1,131 行 · 第 1/3 页

#                                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:
# 
#  <a href="http://www.textism.com" title="This is a title">This is a link</a>
#
# == 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:
#
#  <img src="http://www.textism.com/common/textist.gif" alt="Textist" />
#
# With a link:
#
#  !/common/textist.gif(Textist)!:http://textism.com
#
# Will become:
#
#  <a href="http://textism.com"><img src="/common/textist.gif" alt="Textist" /></a>
#
# == 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:
#
#  <acronym title="American Civil Liberties Union">ACLU</acronym>
#
# == 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 RedCloth < 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 <strong>fell</strong>!"
    #
    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 <b>bold</b> man", [:filter_html] )
    #   r.to_html
    #     #=>"<h1>A &lt;b&gt;bold&lt;/b&gt; man</h1>"
    #
    def initialize( string, restrictions = [] )
        restrictions.each { |r| method( "#{ r }=" ).call( true ) }
        super( string )
    end

    #
    # Generates HTML from the Textile contents.
    #
    #   r = RedCloth.new( "And then? She *fell*!" )
    #   r.to_html( true )
    #     #=>"And then? She <strong>fell</strong>!"
    #
    def to_html( *rules )
        rules = DEFAULT_RULES if rules.empty?
        # make our working copy
        text = self.dup
        
        @urlrefs = {}
        @shelf = []
        textile_rules = [:refs_textile, :block_textile_table, :block_textile_lists,
                         :block_textile_prefix, :inline_textile_image, :inline_textile_link,
                         :inline_textile_code, :inline_textile_span, :glyphs_textile]
        markdown_rules = [:refs_markdown, :block_markdown_setext, :block_markdown_atx, :block_markdown_rule,
                          :block_markdown_bq, :block_markdown_lists, 
                          :inline_markdown_reflink, :inline_markdown_link]
        @rules = rules.collect do |rule|
            case rule
            when :markdown
                markdown_rules
            when :textile
                textile_rules
            else
                rule
            end
        end.flatten

        # standard clean up
        incoming_entities text 
        clean_white_space text 

        # start processor
        @pre_list = []
        rip_offtags text
        no_textile text
        hard_break text 
        unless @lite_mode
            refs text
            blocks text
        end
        inline text
        smooth_offtags text

        retrieve text

        text.gsub!( /<\/?notextile>/, '' )
        text.gsub!( /x%x%/, '&#38;' )
        clean_html text if filter_html
        text.strip!
        text

    end

    #######
    private
    #######
    #
    # Mapping of 8-bit ASCII codes to HTML numerical entity equivalents.
    # (from PyTextile)
    #
    TEXTILE_TAGS =

        [[128, 8364], [129, 0], [130, 8218], [131, 402], [132, 8222], [133, 8230], 
         [134, 8224], [135, 8225], [136, 710], [137, 8240], [138, 352], [139, 8249], 
         [140, 338], [141, 0], [142, 0], [143, 0], [144, 0], [145, 8216], [146, 8217], 
         [147, 8220], [148, 8221], [149, 8226], [150, 8211], [151, 8212], [152, 732], 
         [153, 8482], [154, 353], [155, 8250], [156, 339], [157, 0], [158, 0], [159, 376]].

        collect! do |a, b|
            [a.chr, ( b.zero? and "" or "&#{ b };" )]
        end

    #
    # Regular expressions to convert to HTML.
    #
    A_HLGN = /(?:(?:<>|<|>|\=|[()]+)+)/
    A_VLGN = /[\-^~]/
    C_CLAS = '(?:\([^)]+\))'
    C_LNGE = '(?:\[[^\]]+\])'
    C_STYL = '(?:\{[^}]+\})'
    S_CSPN = '(?:\\\\\d+)'
    S_RSPN = '(?:/\d+)'
    A = "(?:#{A_HLGN}?#{A_VLGN}?|#{A_VLGN}?#{A_HLGN}?)"
    S = "(?:#{S_CSPN}?#{S_RSPN}|#{S_RSPN}?#{S_CSPN}?)"
    C = "(?:#{C_CLAS}?#{C_STYL}?#{C_LNGE}?|#{C_STYL}?#{C_LNGE}?#{C_CLAS}?|#{C_LNGE}?#{C_STYL}?#{C_CLAS}?)"
    # PUNCT = Regexp::quote( '!"#$%&\'()*+,-./:;<=>?@[\\]^_`{|}~' )
    PUNCT = Regexp::quote( '!"#$%&\'*+,-./:;=?@\\^_`|~' )
    PUNCT_NOQ = Regexp::quote( '!"#$&\',./:;=?@\\`|' )
    PUNCT_Q = Regexp::quote( '*-_+^~%' )
    HYPERLINK = '(\S+?)([^\w\s/;=\?]*?)(?=\s|<|$)'

    # Text markup tags, don't conflict with block tags
    SIMPLE_HTML_TAGS = [
        'tt', 'b', 'i', 'big', 'small', 'em', 'strong', 'dfn', 'code', 
        'samp', 'kbd', 'var', 'cite', 'abbr', 'acronym', 'a', 'img', 'br',
        'br', 'map', 'q', 'sub', 'sup', 'span', 'bdo'
    ]

    QTAGS = [
        ['**', 'b'],
        ['*', 'strong'],
        ['??', 'cite', :limit],
        ['-', 'del', :limit],
        ['__', 'i'],
        ['_', 'em', :limit],
        ['%', 'span', :limit],
        ['+', 'ins', :limit],
        ['^', 'sup'],
        ['~', 'sub']
    ] 
    QTAGS.collect! do |rc, ht, rtype|
        rcq = Regexp::quote rc
        re =
            case rtype
            when :limit