108 | #
109 | # With a link:
110 | #
111 | # !/common/textist.gif(Textist)!:http://textism.com
112 | #
113 | # Will become:
114 | #
115 | #
116 | #
117 | # == Defining Acronyms
118 | #
119 | # HTML allows authors to define acronyms via the tag. The definition appears as a
120 | # tool tip when a cursor hovers over the acronym. A crucial aid to clear writing,
121 | # this should be used at least once for each acronym in documents where they appear.
122 | #
123 | # To quickly define an acronym in Textile, place the full text in (parentheses)
124 | # immediately following the acronym.
125 | #
126 | # Example:
127 | #
128 | # ACLU(American Civil Liberties Union)
129 | #
130 | # Will become:
131 | #
132 | # ACLU
133 | #
134 | # == Adding Tables
135 | #
136 | # In Textile, simple tables can be added by seperating each column by
137 | # a pipe.
138 | #
139 | # |a|simple|table|row|
140 | # |And|Another|table|row|
141 | #
142 | # Attributes are defined by style definitions in parentheses.
143 | #
144 | # table(border:1px solid black).
145 | # (background:#ddd;color:red). |{}| | | |
146 | #
147 | # == Using RedCloth
148 | #
149 | # RedCloth is simply an extension of the String class, which can handle
150 | # Textile formatting. Use it like a String and output HTML with its
151 | # RedCloth#to_html method.
152 | #
153 | # doc = RedCloth.new "
154 | #
155 | # h2. Test document
156 | #
157 | # Just a simple test."
158 | #
159 | # puts doc.to_html
160 | #
161 | # By default, RedCloth uses both Textile and Markdown formatting, with
162 | # Textile formatting taking precedence. If you want to turn off Markdown
163 | # formatting, to boost speed and limit the processor:
164 | #
165 | # class RedCloth::Textile.new( str )
166 |
167 | class RedCloth < String
168 |
169 | VERSION = '3.0.4'
170 | DEFAULT_RULES = [:textile, :markdown]
171 |
172 | #
173 | # Two accessor for setting security restrictions.
174 | #
175 | # This is a nice thing if you're using RedCloth for
176 | # formatting in public places (e.g. Wikis) where you
177 | # don't want users to abuse HTML for bad things.
178 | #
179 | # If +:filter_html+ is set, HTML which wasn't
180 | # created by the Textile processor will be escaped.
181 | #
182 | # If +:filter_styles+ is set, it will also disable
183 | # the style markup specifier. ('{color: red}')
184 | #
185 | attr_accessor :filter_html, :filter_styles
186 |
187 | #
188 | # Accessor for toggling hard breaks.
189 | #
190 | # If +:hard_breaks+ is set, single newlines will
191 | # be converted to HTML break tags. This is the
192 | # default behavior for traditional RedCloth.
193 | #
194 | attr_accessor :hard_breaks
195 |
196 | # Accessor for toggling lite mode.
197 | #
198 | # In lite mode, block-level rules are ignored. This means
199 | # that tables, paragraphs, lists, and such aren't available.
200 | # Only the inline markup for bold, italics, entities and so on.
201 | #
202 | # r = RedCloth.new( "And then? She *fell*!", [:lite_mode] )
203 | # r.to_html
204 | # #=> "And then? She fell!"
205 | #
206 | attr_accessor :lite_mode
207 |
208 | #
209 | # Accessor for toggling span caps.
210 | #
211 | # Textile places `span' tags around capitalized
212 | # words by default, but this wreaks havoc on Wikis.
213 | # If +:no_span_caps+ is set, this will be
214 | # suppressed.
215 | #
216 | attr_accessor :no_span_caps
217 |
218 | #
219 | # Establishes the markup predence. Available rules include:
220 | #
221 | # == Textile Rules
222 | #
223 | # The following textile rules can be set individually. Or add the complete
224 | # set of rules with the single :textile rule, which supplies the rule set in
225 | # the following precedence:
226 | #
227 | # refs_textile:: Textile references (i.e. [hobix]http://hobix.com/)
228 | # block_textile_table:: Textile table block structures
229 | # block_textile_lists:: Textile list structures
230 | # block_textile_prefix:: Textile blocks with prefixes (i.e. bq., h2., etc.)
231 | # inline_textile_image:: Textile inline images
232 | # inline_textile_link:: Textile inline links
233 | # inline_textile_span:: Textile inline spans
234 | # glyphs_textile:: Textile entities (such as em-dashes and smart quotes)
235 | #
236 | # == Markdown
237 | #
238 | # refs_markdown:: Markdown references (for example: [hobix]: http://hobix.com/)
239 | # block_markdown_setext:: Markdown setext headers
240 | # block_markdown_atx:: Markdown atx headers
241 | # block_markdown_rule:: Markdown horizontal rules
242 | # block_markdown_bq:: Markdown blockquotes
243 | # block_markdown_lists:: Markdown lists
244 | # inline_markdown_link:: Markdown links
245 | attr_accessor :rules
246 |
247 | # Returns a new RedCloth object, based on _string_ and
248 | # enforcing all the included _restrictions_.
249 | #
250 | # r = RedCloth.new( "h1. A bold man", [:filter_html] )
251 | # r.to_html
252 | # #=>"#{ blk }#{ blk }
" 637 | end 638 | end 639 | # hard_break blk 640 | blk + "\n#{ code_blk }" 641 | end 642 | end 643 | 644 | end.join( "\n\n" ) ) 645 | end 646 | 647 | def textile_bq( tag, atts, cite, content ) 648 | cite, cite_title = check_refs( cite ) 649 | cite = " cite=\"#{ cite }\"" if cite 650 | atts = shelve( atts ) if atts 651 | "\t\n#{ blk }\n\n\n" 723 | end 724 | end 725 | 726 | MARKDOWN_RULE_RE = /^(#{ 727 | ['*', '-', '_'].collect { |ch| '( ?' + Regexp::quote( ch ) + ' ?){3,}' }.join( '|' ) 728 | })$/ 729 | 730 | def block_markdown_rule( text ) 731 | text.gsub!( MARKDOWN_RULE_RE ) do |blk| 732 | "
|.|^) # start of line?
875 | \! # opening
876 | (\<|\=|\>)? # optional alignment atts
877 | (#{C}) # optional style,class atts
878 | (?:\. )? # optional dot-space
879 | ([^\s(!]+?) # presume this is the src
880 | \s? # optional space
881 | (?:\(((?:[^\(\)]|\([^\)]+\))+?)\))? # optional title
882 | \! # closing
883 | (?::#{ HYPERLINK })? # optional href
884 | /x
885 |
886 | def inline_textile_image( text )
887 | text.gsub!( IMAGE_RE ) do |m|
888 | stln,algn,atts,url,title,href,href_a1,href_a2 = $~[1..8]
889 | atts = pba( atts )
890 | atts = " src=\"#{ url }\"#{ atts }"
891 | atts << " title=\"#{ title }\"" if title
892 | atts << " alt=\"#{ title }\""
893 | # size = @getimagesize($url);
894 | # if($size) $atts.= " $size[3]";
895 |
896 | href, alt_title = check_refs( href ) if href
897 | url, url_title = check_refs( url )
898 |
899 | out = ''
900 | out << " "
907 | out = " #{ out }"
908 | else
909 | out = "#{ stln }#{ out }"
910 | end
911 | else
912 | out = stln + out
913 | end
914 |
915 | out
916 | end
917 | end
918 |
919 | def shelve( val )
920 | @shelf << val
921 | " :redsh##{ @shelf.length }:"
922 | end
923 |
924 | def retrieve( text )
925 | @shelf.each_with_index do |r, i|
926 | text.gsub!( " :redsh##{ i + 1 }:", r )
927 | end
928 | end
929 |
930 | def incoming_entities( text )
931 | ## turn any incoming ampersands into a dummy character for now.
932 | ## This uses a negative lookahead for alphanumerics followed by a semicolon,
933 | ## implying an incoming html entity, to be skipped
934 |
935 | text.gsub!( /&(?![#a-z0-9]+;)/i, "x%x%" )
936 | end
937 |
938 | def no_textile( text )
939 | text.gsub!( /(^|\s)==([^=]+.*?)==(\s|$)?/,
940 | '\1, etc.
991 | if $1
992 | if line =~ OFFTAG_OPEN
993 | codepre += 1
994 | elsif line =~ OFFTAG_CLOSE
995 | codepre -= 1
996 | codepre = 0 if codepre < 0
997 | end
998 | elsif codepre.zero?
999 | glyphs_textile( line, level + 1 )
1000 | else
1001 | htmlesc( line, :NoQuotes )
1002 | end
1003 | # p [level, codepre, line]
1004 |
1005 | line
1006 | end
1007 | end
1008 | end
1009 |
1010 | def rip_offtags( text )
1011 | if text =~ /<.*>/
1012 | ## strip and encode
content
1013 | codepre, used_offtags = 0, {}
1014 | text.gsub!( OFFTAG_MATCH ) do |line|
1015 | if $3
1016 | offtag, aftertag = $4, $5
1017 | codepre += 1
1018 | used_offtags[offtag] = true
1019 | if codepre - used_offtags.length > 0
1020 | htmlesc( line, :NoQuotes ) unless used_offtags['notextile']
1021 | @pre_list.last << line
1022 | line = ""
1023 | else
1024 | htmlesc( aftertag, :NoQuotes ) if aftertag and not used_offtags['notextile']
1025 | line = " content
1046 | text.gsub!( /
'
1068 | end
1069 |
1070 | # HTML cleansing stuff
1071 | BASIC_TAGS = {
1072 | 'a' => ['href', 'title'],
1073 | 'img' => ['src', 'alt', 'title'],
1074 | 'br' => [],
1075 | 'i' => nil,
1076 | 'u' => nil,
1077 | 'b' => nil,
1078 | 'pre' => nil,
1079 | 'kbd' => nil,
1080 | 'code' => ['lang'],
1081 | 'cite' => nil,
1082 | 'strong' => nil,
1083 | 'em' => nil,
1084 | 'ins' => nil,
1085 | 'sup' => nil,
1086 | 'sub' => nil,
1087 | 'del' => nil,
1088 | 'table' => nil,
1089 | 'tr' => nil,
1090 | 'td' => ['colspan', 'rowspan'],
1091 | 'th' => nil,
1092 | 'ol' => nil,
1093 | 'ul' => nil,
1094 | 'li' => nil,
1095 | 'p' => nil,
1096 | 'h1' => nil,
1097 | 'h2' => nil,
1098 | 'h3' => nil,
1099 | 'h4' => nil,
1100 | 'h5' => nil,
1101 | 'h6' => nil,
1102 | 'blockquote' => ['cite']
1103 | }
1104 |
1105 | def clean_html( text, tags = BASIC_TAGS )
1106 | text.gsub!( /]*)>/ ) do
1108 | raw = $~
1109 | tag = raw[2].downcase
1110 | if tags.has_key? tag
1111 | pcs = [tag]
1112 | tags[tag].each do |prop|
1113 | ['"', "'", ''].each do |q|
1114 | q2 = ( q != '' ? q : '\s' )
1115 | if raw[3] =~ /#{prop}\s*=\s*#{q}([^#{q2}]+)#{q}/i
1116 | attrv = $1
1117 | next if prop == 'src' and attrv =~ %r{^(?!http)\w+:}
1118 | pcs << "#{prop}=\"#{$1.gsub('"', '\\"')}\""
1119 | break
1120 | end
1121 | end
1122 | end if tags[tag]
1123 | "<#{raw[1]}#{pcs.join " "}>"
1124 | else
1125 | " "
1126 | end
1127 | end
1128 | end
1129 | end
1130 |
1131 |
--------------------------------------------------------------------------------