Project

General

Profile

1
#                                vim:ts=4:sw=4:
2
# = RedCloth - Textile and Markdown Hybrid for Ruby
3
#
4
# Homepage::  http://whytheluckystiff.net/ruby/redcloth/
5
# Author::    why the lucky stiff (http://whytheluckystiff.net/)
6
# Copyright:: (cc) 2004 why the lucky stiff (and his puppet organizations.)
7
# License::   BSD
8
#
9
# (see http://hobix.com/textile/ for a Textile Reference.)
10
#
11
# Based on (and also inspired by) both:
12
#
13
# PyTextile: http://diveintomark.org/projects/textile/textile.py.txt
14
# Textism for PHP: http://www.textism.com/tools/textile/
15
#
16
#
17

    
18
# = RedCloth
19
#
20
# RedCloth is a Ruby library for converting Textile and/or Markdown
21
# into HTML.  You can use either format, intermingled or separately.
22
# You can also extend RedCloth to honor your own custom text stylings.
23
#
24
# RedCloth users are encouraged to use Textile if they are generating
25
# HTML and to use Markdown if others will be viewing the plain text.
26
#
27
# == What is Textile?
28
#
29
# Textile is a simple formatting style for text
30
# documents, loosely based on some HTML conventions.
31
#
32
# == Sample Textile Text
33
#
34
#  h2. This is a title
35
#
36
#  h3. This is a subhead
37
#
38
#  This is a bit of paragraph.
39
#
40
#  bq. This is a blockquote.
41
#
42
# = Writing Textile
43
#
44
# A Textile document consists of paragraphs.  Paragraphs
45
# can be specially formatted by adding a small instruction
46
# to the beginning of the paragraph.
47
#
48
#  h[n].   Header of size [n].
49
#  bq.     Blockquote.
50
#  #       Numeric list.
51
#  *       Bulleted list.
52
#
53
# == Quick Phrase Modifiers
54
#
55
# Quick phrase modifiers are also included, to allow formatting
56
# of small portions of text within a paragraph.
57
#
58
#  \_emphasis\_
59
#  \_\_italicized\_\_
60
#  \*strong\*
61
#  \*\*bold\*\*
62
#  ??citation??
63
#  -deleted text-
64
#  +inserted text+
65
#  ^superscript^
66
#  ~subscript~
67
#  @code@
68
#  %(classname)span%
69
#
70
#  ==notextile== (leave text alone)
71
#
72
# == Links
73
#
74
# To make a hypertext link, put the link text in "quotation 
75
# marks" followed immediately by a colon and the URL of the link.
76
# 
77
# Optional: text in (parentheses) following the link text, 
78
# but before the closing quotation mark, will become a Title 
79
# attribute for the link, visible as a tool tip when a cursor is above it.
80
# 
81
# Example:
82
#
83
#  "This is a link (This is a title) ":http://www.textism.com
84
# 
85
# Will become:
86
# 
87
#  <a href="http://www.textism.com" title="This is a title">This is a link</a>
88
#
89
# == Images
90
#
91
# To insert an image, put the URL for the image inside exclamation marks.
92
#
93
# Optional: text that immediately follows the URL in (parentheses) will 
94
# be used as the Alt text for the image. Images on the web should always 
95
# have descriptive Alt text for the benefit of readers using non-graphical 
96
# browsers.
97
#
98
# Optional: place a colon followed by a URL immediately after the 
99
# closing ! to make the image into a link.
100
# 
101
# Example:
102
#
103
#  !http://www.textism.com/common/textist.gif(Textist)!
104
#
105
# Will become:
106
#
107
#  <img src="http://www.textism.com/common/textist.gif" alt="Textist" />
108
#
109
# With a link:
110
#
111
#  !/common/textist.gif(Textist)!:http://textism.com
112
#
113
# Will become:
114
#
115
#  <a href="http://textism.com"><img src="/https/www.redmine.org/common/textist.gif" alt="Textist" /></a>
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
#  <acronym title="American Civil Liberties Union">ACLU</acronym>
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 <strong>fell</strong>!"
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 <b>bold</b> man", [:filter_html] )
251
    #   r.to_html
252
    #     #=>"<h1>A &lt;b&gt;bold&lt;/b&gt; man</h1>"
253
    #
254
    def initialize( string, restrictions = [] )
255
        restrictions.each { |r| method( "#{ r }=" ).call( true ) }
256
        super( string )
257
    end
258

    
259
    #
260
    # Generates HTML from the Textile contents.
261
    #
262
    #   r = RedCloth.new( "And then? She *fell*!" )
263
    #   r.to_html( true )
264
    #     #=>"And then? She <strong>fell</strong>!"
265
    #
266
    def to_html( *rules )
267
        rules = DEFAULT_RULES if rules.empty?
268
        # make our working copy
269
        text = self.dup
270
        
271
        @urlrefs = {}
272
        @shelf = []
273
        textile_rules = [:refs_textile, :block_textile_table, :block_textile_lists,
274
                         :block_textile_prefix, :inline_textile_image, :inline_textile_link,
275
                         :inline_textile_code, :inline_textile_span]
276
        markdown_rules = [:refs_markdown, :block_markdown_setext, :block_markdown_atx, :block_markdown_rule,
277
                          :block_markdown_bq, :block_markdown_lists, 
278
                          :inline_markdown_reflink, :inline_markdown_link]
279
        @rules = rules.collect do |rule|
280
            case rule
281
            when :markdown
282
                markdown_rules
283
            when :textile
284
                textile_rules
285
            else
286
                rule
287
            end
288
        end.flatten
289

    
290
        # standard clean up
291
        incoming_entities text 
292
        clean_white_space text 
293

    
294
        # start processor
295
        @pre_list = []
296
        rip_offtags text
297
        no_textile text
298
        hard_break text 
299
        unless @lite_mode
300
            refs text
301
            blocks text
302
        end
303
        inline text
304
        smooth_offtags text
305

    
306
        retrieve text
307

    
308
        text.gsub!( /<\/?notextile>/, '' )
309
        text.gsub!( /x%x%/, '&#38;' )
310
        clean_html text if filter_html
311
        text.strip!
312
        text
313

    
314
    end
315

    
316
    #######
317
    private
318
    #######
319
    #
320
    # Mapping of 8-bit ASCII codes to HTML numerical entity equivalents.
321
    # (from PyTextile)
322
    #
323
    TEXTILE_TAGS =
324

    
325
        [[128, 8364], [129, 0], [130, 8218], [131, 402], [132, 8222], [133, 8230], 
326
         [134, 8224], [135, 8225], [136, 710], [137, 8240], [138, 352], [139, 8249], 
327
         [140, 338], [141, 0], [142, 0], [143, 0], [144, 0], [145, 8216], [146, 8217], 
328
         [147, 8220], [148, 8221], [149, 8226], [150, 8211], [151, 8212], [152, 732], 
329
         [153, 8482], [154, 353], [155, 8250], [156, 339], [157, 0], [158, 0], [159, 376]].
330

    
331
        collect! do |a, b|
332
            [a.chr, ( b.zero? and "" or "&#{ b };" )]
333
        end
334

    
335
    #
336
    # Regular expressions to convert to HTML.
337
    #
338
    A_HLGN = /(?:(?:<>|<|>|\=|[()]+)+)/
339
    A_VLGN = /[\-^~]/
340
    C_CLAS = '(?:\([^)]+\))'
341
    C_LNGE = '(?:\[[^\]]+\])'
342
    C_STYL = '(?:\{[^}]+\})'
343
    S_CSPN = '(?:\\\\\d+)'
344
    S_RSPN = '(?:/\d+)'
345
    A = "(?:#{A_HLGN}?#{A_VLGN}?|#{A_VLGN}?#{A_HLGN}?)"
346
    S = "(?:#{S_CSPN}?#{S_RSPN}|#{S_RSPN}?#{S_CSPN}?)"
347
    C = "(?:#{C_CLAS}?#{C_STYL}?#{C_LNGE}?|#{C_STYL}?#{C_LNGE}?#{C_CLAS}?|#{C_LNGE}?#{C_STYL}?#{C_CLAS}?)"
348
    # PUNCT = Regexp::quote( '!"#$%&\'()*+,-./:;<=>?@[\\]^_`{|}~' )
349
    PUNCT = Regexp::quote( '!"#$%&\'*+,-./:;=?@\\^_`|~' )
350
    PUNCT_NOQ = Regexp::quote( '!"#$&\',./:;=?@\\`|' )
351
    PUNCT_Q = Regexp::quote( '*-_+^~%' )
352
    HYPERLINK = '(\S+?)([^\w\s/;=\?]*?)(?=\s|<|$)'
353

    
354
    # Text markup tags, don't conflict with block tags
355
    SIMPLE_HTML_TAGS = [
356
        'tt', 'b', 'i', 'big', 'small', 'em', 'strong', 'dfn', 'code', 
357
        'samp', 'kbd', 'var', 'cite', 'abbr', 'acronym', 'a', 'img', 'br',
358
        'br', 'map', 'q', 'sub', 'sup', 'span', 'bdo'
359
    ]
360

    
361
    QTAGS = [
362
        ['**', 'b', :limit],
363
        ['*', 'strong', :limit],
364
        ['??', 'cite', :limit],
365
        ['-', 'del', :limit],
366
        ['__', 'i', :limit],
367
        ['_', 'em', :limit],
368
        ['%', 'span', :limit],
369
        ['+', 'ins', :limit],
370
        ['^', 'sup', :limit],
371
        ['~', 'sub', :limit]
372
    ] 
373
    QTAGS.collect! do |rc, ht, rtype|
374
        rcq = Regexp::quote rc
375
        re =
376
            case rtype
377
            when :limit
378
                /(\W)
379
                (#{rcq})
380
                (#{C})
381
                (?::(\S+?))?
382
                (\S.*?\S|\S)
383
                #{rcq}
384
                (?=\W)/x
385
            else
386
                /(#{rcq})
387
                (#{C})
388
                (?::(\S+))?
389
                (\S.*?\S|\S)
390
                #{rcq}/xm 
391
            end
392
        [rc, ht, re, rtype]
393
    end
394

    
395
    # Elements to handle
396
    GLYPHS = [
397
    #   [ /([^\s\[{(>])?\'([dmst]\b|ll\b|ve\b|\s|:|$)/, '\1&#8217;\2' ], # single closing
398
    #   [ /([^\s\[{(>#{PUNCT_Q}][#{PUNCT_Q}]*)\'/, '\1&#8217;' ], # single closing
399
    #   [ /\'(?=[#{PUNCT_Q}]*(s\b|[\s#{PUNCT_NOQ}]))/, '&#8217;' ], # single closing
400
    #   [ /\'/, '&#8216;' ], # single opening
401
        [ /</, '&lt;' ], # less-than
402
        [ />/, '&gt;' ], # greater-than
403
    #   [ /([^\s\[{(])?"(\s|:|$)/, '\1&#8221;\2' ], # double closing
404
    #   [ /([^\s\[{(>#{PUNCT_Q}][#{PUNCT_Q}]*)"/, '\1&#8221;' ], # double closing
405
    #   [ /"(?=[#{PUNCT_Q}]*[\s#{PUNCT_NOQ}])/, '&#8221;' ], # double closing
406
    #   [ /"/, '&#8220;' ], # double opening
407
        [ /\b( )?\.{3}/, '\1&#8230;' ], # ellipsis
408
        [ /\b([A-Z][A-Z0-9]{2,})\b(?:[(]([^)]*)[)])/, '<acronym title="\2">\1</acronym>' ], # 3+ uppercase acronym
409
        [ /(^|[^"][>\s])([A-Z][A-Z0-9 ]+[A-Z0-9])([^<A-Za-z0-9]|$)/, '\1<span class="caps">\2</span>\3', :no_span_caps ], # 3+ uppercase caps
410
        [ /(\.\s)?\s?--\s?/, '\1&#8212;' ], # em dash
411
        [ /\s->\s/, ' &rarr; ' ], # right arrow
412
        [ /\s-\s/, ' &#8211; ' ], # en dash
413
        [ /(\d+) ?x ?(\d+)/, '\1&#215;\2' ], # dimension sign
414
        [ /\b ?[(\[]TM[\])]/i, '&#8482;' ], # trademark
415
        [ /\b ?[(\[]R[\])]/i, '&#174;' ], # registered
416
        [ /\b ?[(\[]C[\])]/i, '&#169;' ] # copyright
417
    ]
418

    
419
    H_ALGN_VALS = {
420
        '<' => 'left',
421
        '=' => 'center',
422
        '>' => 'right',
423
        '<>' => 'justify'
424
    }
425

    
426
    V_ALGN_VALS = {
427
        '^' => 'top',
428
        '-' => 'middle',
429
        '~' => 'bottom'
430
    }
431

    
432
    #
433
    # Flexible HTML escaping
434
    #
435
    def htmlesc( str, mode )
436
        str.gsub!( '&', '&amp;' )
437
        str.gsub!( '"', '&quot;' ) if mode != :NoQuotes
438
        str.gsub!( "'", '&#039;' ) if mode == :Quotes
439
        str.gsub!( '<', '&lt;')
440
        str.gsub!( '>', '&gt;')
441
    end
442

    
443
    # Search and replace for Textile glyphs (quotes, dashes, other symbols)
444
    def pgl( text )
445
        GLYPHS.each do |re, resub, tog|
446
            next if tog and method( tog ).call
447
            text.gsub! re, resub
448
        end
449
    end
450

    
451
    # Parses Textile attribute lists and builds an HTML attribute string
452
    def pba( text_in, element = "" )
453
        
454
        return '' unless text_in
455

    
456
        style = []
457
        text = text_in.dup
458
        if element == 'td'
459
            colspan = $1 if text =~ /\\(\d+)/
460
            rowspan = $1 if text =~ /\/(\d+)/
461
            style << "vertical-align:#{ v_align( $& ) };" if text =~ A_VLGN
462
        end
463

    
464
        style << "#{ $1 };" if not filter_styles and
465
            text.sub!( /\{([^}]*)\}/, '' )
466

    
467
        lang = $1 if
468
            text.sub!( /\[([^)]+?)\]/, '' )
469

    
470
        cls = $1 if
471
            text.sub!( /\(([^()]+?)\)/, '' )
472
                        
473
        style << "padding-left:#{ $1.length }em;" if
474
            text.sub!( /([(]+)/, '' )
475

    
476
        style << "padding-right:#{ $1.length }em;" if text.sub!( /([)]+)/, '' )
477

    
478
        style << "text-align:#{ h_align( $& ) };" if text =~ A_HLGN
479

    
480
        cls, id = $1, $2 if cls =~ /^(.*?)#(.*)$/
481
        
482
        atts = ''
483
        atts << " style=\"#{ style.join }\"" unless style.empty?
484
        atts << " class=\"#{ cls }\"" unless cls.to_s.empty?
485
        atts << " lang=\"#{ lang }\"" if lang
486
        atts << " id=\"#{ id }\"" if id
487
        atts << " colspan=\"#{ colspan }\"" if colspan
488
        atts << " rowspan=\"#{ rowspan }\"" if rowspan
489
        
490
        atts
491
    end
492

    
493
    TABLE_RE = /^(?:table(_?#{S}#{A}#{C})\. ?\n)?^(#{A}#{C}\.? ?\|.*?\|)(\n\n|\Z)/m
494
    
495
    # Parses a Textile table block, building HTML from the result.
496
    def block_textile_table( text ) 
497
        text.gsub!( TABLE_RE ) do |matches|
498

    
499
            tatts, fullrow = $~[1..2]
500
            tatts = pba( tatts, 'table' )
501
            tatts = shelve( tatts ) if tatts
502
            rows = []
503

    
504
            fullrow.
505
            split( /\|$/m ).
506
            delete_if { |x| x.empty? }.
507
            each do |row|
508

    
509
                ratts, row = pba( $1, 'tr' ), $2 if row =~ /^(#{A}#{C}\. )(.*)/m
510
                
511
                cells = []
512
                row.split( '|' ).each do |cell|
513
                    ctyp = 'd'
514
                    ctyp = 'h' if cell =~ /^_/
515

    
516
                    catts = ''
517
                    catts, cell = pba( $1, 'td' ), $2 if cell =~ /^(_?#{S}#{A}#{C}\. ?)(.*)/
518

    
519
                    unless cell.strip.empty?
520
                        catts = shelve( catts ) if catts
521
                        cells << "\t\t\t<t#{ ctyp }#{ catts }>#{ cell }</t#{ ctyp }>" 
522
                    end
523
                end
524
                ratts = shelve( ratts ) if ratts
525
                rows << "\t\t<tr#{ ratts }>\n#{ cells.join( "\n" ) }\n\t\t</tr>"
526
            end
527
            "\t<table#{ tatts }>\n#{ rows.join( "\n" ) }\n\t</table>\n\n"
528
        end
529
    end
530

    
531
    LISTS_RE = /^([#*]+?#{C} .*?)$(?![^#*])/m
532
    LISTS_CONTENT_RE = /^([#*]+)(#{A}#{C}) (.*)$/m
533

    
534
    # Parses Textile lists and generates HTML
535
    def block_textile_lists( text ) 
536
        text.gsub!( LISTS_RE ) do |match|
537
            lines = match.split( /\n/ )
538
            last_line = -1
539
            depth = []
540
            lines.each_with_index do |line, line_id|
541
                if line =~ LISTS_CONTENT_RE 
542
                    tl,atts,content = $~[1..3]
543
                    if depth.last
544
                        if depth.last.length > tl.length
545
                            (depth.length - 1).downto(0) do |i|
546
                                break if depth[i].length == tl.length
547
                                lines[line_id - 1] << "</li>\n\t</#{ lT( depth[i] ) }l>\n\t"
548
                                depth.pop
549
                            end
550
                        end
551
                        if depth.last and depth.last.length == tl.length
552
                            lines[line_id - 1] << '</li>'
553
                        end
554
                    end
555
                    unless depth.last == tl
556
                        depth << tl
557
                        atts = pba( atts )
558
                        atts = shelve( atts ) if atts
559
                        lines[line_id] = "\t<#{ lT(tl) }l#{ atts }>\n\t<li>#{ content }"
560
                    else
561
                        lines[line_id] = "\t\t<li>#{ content }"
562
                    end
563
                    last_line = line_id
564

    
565
                else
566
                    last_line = line_id
567
                end
568
                if line_id - last_line > 1 or line_id == lines.length - 1
569
                    depth.delete_if do |v|
570
                        lines[last_line] << "</li>\n\t</#{ lT( v ) }l>"
571
                    end
572
                end
573
            end
574
            lines.join( "\n" )
575
        end
576
    end
577

    
578
    CODE_RE = /(\W)
579
        @
580
        (?:\|(\w+?)\|)?
581
        (.+?)
582
        @
583
        (?=\W)/x
584

    
585
    def inline_textile_code( text ) 
586
        text.gsub!( CODE_RE ) do |m|
587
            before,lang,code,after = $~[1..4]
588
            lang = " lang=\"#{ lang }\"" if lang
589
            rip_offtags( "#{ before }<code#{ lang }>#{ code }</code>#{ after }" )
590
        end
591
    end
592

    
593
    def lT( text ) 
594
        text =~ /\#$/ ? 'o' : 'u'
595
    end
596

    
597
    def hard_break( text )
598
        text.gsub!( /(.)\n(?!\Z| *([#*=]+(\s|$)|[{|]))/, "\\1<br />" ) if hard_breaks
599
    end
600

    
601
    BLOCKS_GROUP_RE = /\n{2,}(?! )/m
602

    
603
    def blocks( text, deep_code = false )
604
        text.replace( text.split( BLOCKS_GROUP_RE ).collect do |blk|
605
            plain = blk !~ /\A[#*> ]/
606

    
607
            # skip blocks that are complex HTML
608
            if blk =~ /^<\/?(\w+).*>/ and not SIMPLE_HTML_TAGS.include? $1
609
                blk
610
            else
611
                # search for indentation levels
612
                blk.strip!
613
                if blk.empty?
614
                    blk
615
                else
616
                    code_blk = nil
617
                    blk.gsub!( /((?:\n(?:\n^ +[^\n]*)+)+)/m ) do |iblk|
618
                        flush_left iblk
619
                        blocks iblk, plain
620
                        iblk.gsub( /^(\S)/, "\t\\1" )
621
                        if plain
622
                            code_blk = iblk; ""
623
                        else
624
                            iblk
625
                        end
626
                    end
627

    
628
                    block_applied = 0 
629
                    @rules.each do |rule_name|
630
                        block_applied += 1 if ( rule_name.to_s.match /^block_/ and method( rule_name ).call( blk ) )
631
                    end
632
                    if block_applied.zero?
633
                        if deep_code
634
                            blk = "\t<pre><code>#{ blk }</code></pre>"
635
                        else
636
                            blk = "\t<p>#{ blk }</p>"
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<blockquote#{ cite }>\n\t\t<p#{ atts }>#{ content }</p>\n\t</blockquote>"
652
    end
653

    
654
    def textile_p( tag, atts, cite, content )
655
        atts = shelve( atts ) if atts
656
        "\t<#{ tag }#{ atts }>#{ content }</#{ tag }>"
657
    end
658

    
659
    alias textile_h1 textile_p
660
    alias textile_h2 textile_p
661
    alias textile_h3 textile_p
662
    alias textile_h4 textile_p
663
    alias textile_h5 textile_p
664
    alias textile_h6 textile_p
665

    
666
    def textile_fn_( tag, num, atts, cite, content )
667
        atts << " id=\"fn#{ num }\""
668
        content = "<sup>#{ num }</sup> #{ content }"
669
        atts = shelve( atts ) if atts
670
        "\t<p#{ atts }>#{ content }</p>"
671
    end
672

    
673
    BLOCK_RE = /^(([a-z]+)(\d*))(#{A}#{C})\.(?::(\S+))? (.*)$/m
674

    
675
    def block_textile_prefix( text ) 
676
        if text =~ BLOCK_RE
677
            tag,tagpre,num,atts,cite,content = $~[1..6]
678
            atts = pba( atts )
679

    
680
            # pass to prefix handler
681
            if respond_to? "textile_#{ tag }", true
682
                text.gsub!( $&, method( "textile_#{ tag }" ).call( tag, atts, cite, content ) )
683
            elsif respond_to? "textile_#{ tagpre }_", true
684
                text.gsub!( $&, method( "textile_#{ tagpre }_" ).call( tagpre, num, atts, cite, content ) )
685
            end
686
        end
687
    end
688
    
689
    SETEXT_RE = /\A(.+?)\n([=-])[=-]* *$/m
690
    def block_markdown_setext( text )
691
        if text =~ SETEXT_RE
692
            tag = if $2 == "="; "h1"; else; "h2"; end
693
            blk, cont = "<#{ tag }>#{ $1 }</#{ tag }>", $'
694
            blocks cont
695
            text.replace( blk + cont )
696
        end
697
    end
698

    
699
    ATX_RE = /\A(\#{1,6})  # $1 = string of #'s
700
              [ ]*
701
              (.+?)       # $2 = Header text
702
              [ ]*
703
              \#*         # optional closing #'s (not counted)
704
              $/x
705
    def block_markdown_atx( text )
706
        if text =~ ATX_RE
707
            tag = "h#{ $1.length }"
708
            blk, cont = "<#{ tag }>#{ $2 }</#{ tag }>\n\n", $'
709
            blocks cont
710
            text.replace( blk + cont )
711
        end
712
    end
713

    
714
    MARKDOWN_BQ_RE = /\A(^ *> ?.+$(.+\n)*\n*)+/m
715

    
716
    def block_markdown_bq( text )
717
        text.gsub!( MARKDOWN_BQ_RE ) do |blk|
718
            blk.gsub!( /^ *> ?/, '' )
719
            flush_left blk
720
            blocks blk
721
            blk.gsub!( /^(\S)/, "\t\\1" )
722
            "<blockquote>\n#{ blk }\n</blockquote>\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
            "<hr />"
733
        end
734
    end
735

    
736
    # XXX TODO XXX
737
    def block_markdown_lists( text )
738
    end
739

    
740
    def inline_textile_span( text ) 
741
        QTAGS.each do |qtag_rc, ht, qtag_re, rtype|
742
            text.gsub!( qtag_re ) do |m|
743
             
744
                case rtype
745
                when :limit
746
                    sta,qtag,atts,cite,content = $~[1..5]
747
                else
748
                    qtag,atts,cite,content = $~[1..4]
749
                    sta = ''
750
                end
751
                atts = pba( atts )
752
                atts << " cite=\"#{ cite }\"" if cite
753
                atts = shelve( atts ) if atts
754

    
755
                "#{ sta }<#{ ht }#{ atts }>#{ content }</#{ ht }>"
756

    
757
            end
758
        end
759
    end
760

    
761
    LINK_RE = /
762
            ([\s\[{(]|[#{PUNCT}])?     # $pre
763
            "                          # start
764
            (#{C})                     # $atts
765
            ([^"]+?)                   # $text
766
            \s?
767
            (?:\(([^)]+?)\)(?="))?     # $title
768
            ":
769
            (\S+?)                     # $url
770
            (\/)?                      # $slash
771
            ([^\w\/;]*?)               # $post
772
            (?=<|\s|$)
773
        /x 
774

    
775
    def inline_textile_link( text ) 
776
        text.gsub!( LINK_RE ) do |m|
777
            pre,atts,text,title,url,slash,post = $~[1..7]
778

    
779
            url, url_title = check_refs( url )
780
            title ||= url_title
781
            
782
            atts = pba( atts )
783
            atts = " href=\"#{ url }#{ slash }\"#{ atts }"
784
            atts << " title=\"#{ title }\"" if title
785
            atts = shelve( atts ) if atts
786
            
787
            external = (url =~ /^http:\/\//) ? ' class="external"' : ''
788
            
789
            "#{ pre }<a#{ atts }#{ external }>#{ text }</a>#{ post }"
790
        end
791
    end
792

    
793
    MARKDOWN_REFLINK_RE = /
794
            \[([^\[\]]+)\]      # $text
795
            [ ]?                # opt. space
796
            (?:\n[ ]*)?         # one optional newline followed by spaces
797
            \[(.*?)\]           # $id
798
        /x 
799

    
800
    def inline_markdown_reflink( text ) 
801
        text.gsub!( MARKDOWN_REFLINK_RE ) do |m|
802
            text, id = $~[1..2]
803

    
804
            if id.empty?
805
                url, title = check_refs( text )
806
            else
807
                url, title = check_refs( id )
808
            end
809
            
810
            atts = " href=\"#{ url }\""
811
            atts << " title=\"#{ title }\"" if title
812
            atts = shelve( atts )
813
            
814
            "<a#{ atts }>#{ text }</a>"
815
        end
816
    end
817

    
818
    MARKDOWN_LINK_RE = /
819
            \[([^\[\]]+)\]      # $text
820
            \(                  # open paren
821
            [ \t]*              # opt space
822
            <?(.+?)>?           # $href
823
            [ \t]*              # opt space
824
            (?:                 # whole title
825
            (['"])              # $quote
826
            (.*?)               # $title
827
            \3                  # matching quote
828
            )?                  # title is optional
829
            \)
830
        /x 
831

    
832
    def inline_markdown_link( text ) 
833
        text.gsub!( MARKDOWN_LINK_RE ) do |m|
834
            text, url, quote, title = $~[1..4]
835

    
836
            atts = " href=\"#{ url }\""
837
            atts << " title=\"#{ title }\"" if title
838
            atts = shelve( atts )
839
            
840
            "<a#{ atts }>#{ text }</a>"
841
        end
842
    end
843

    
844
    TEXTILE_REFS_RE =  /(^ *)\[([^\[\n]+?)\](#{HYPERLINK})(?=\s|$)/
845
    MARKDOWN_REFS_RE = /(^ *)\[([^\n]+?)\]:\s+<?(#{HYPERLINK})>?(?:\s+"((?:[^"]|\\")+)")?(?=\s|$)/m
846

    
847
    def refs( text )
848
        @rules.each do |rule_name|
849
            method( rule_name ).call( text ) if rule_name.to_s.match /^refs_/
850
        end
851
    end
852

    
853
    def refs_textile( text ) 
854
        text.gsub!( TEXTILE_REFS_RE ) do |m|
855
            flag, url = $~[2..3]
856
            @urlrefs[flag.downcase] = [url, nil]
857
            nil
858
        end
859
    end
860
    
861
    def refs_markdown( text )
862
        text.gsub!( MARKDOWN_REFS_RE ) do |m|
863
            flag, url = $~[2..3]
864
            title = $~[6]
865
            @urlrefs[flag.downcase] = [url, title]
866
            nil
867
        end
868
    end
869

    
870
    def check_refs( text ) 
871
        ret = @urlrefs[text.downcase] if text
872
        ret || [text, nil]
873
    end
874

    
875
    IMAGE_RE = /
876
            (<p>|.|^)            # start of line?
877
            \!                   # opening
878
            (\<|\=|\>)?          # optional alignment atts
879
            (#{C})               # optional style,class atts
880
            (?:\. )?             # optional dot-space
881
            ([^\s(!]+?)          # presume this is the src
882
            \s?                  # optional space
883
            (?:\(((?:[^\(\)]|\([^\)]+\))+?)\))?   # optional title
884
            \!                   # closing
885
            (?::#{ HYPERLINK })? # optional href
886
        /x 
887

    
888
    def inline_textile_image( text ) 
889
        text.gsub!( IMAGE_RE )  do |m|
890
            stln,algn,atts,url,title,href,href_a1,href_a2 = $~[1..8]
891
            atts = pba( atts )
892
            atts = " src=\"#{ url }\"#{ atts }"
893
            atts << " title=\"#{ title }\"" if title
894
            atts << " alt=\"#{ title }\"" 
895
            # size = @getimagesize($url);
896
            # if($size) $atts.= " $size[3]";
897

    
898
            href, alt_title = check_refs( href ) if href
899
            url, url_title = check_refs( url )
900

    
901
            out = ''
902
            out << "<a#{ shelve( " href=\"#{ href }\"" ) }>" if href
903
            out << "<img#{ shelve( atts ) } />"
904
            out << "</a>#{ href_a1 }#{ href_a2 }" if href
905
            
906
            if algn 
907
                algn = h_align( algn )
908
                if stln == "<p>"
909
                    out = "<p style=\"float:#{ algn }\">#{ out }"
910
                else
911
                    out = "#{ stln }<div style=\"float:#{ algn }\">#{ out }</div>"
912
                end
913
            else
914
                out = stln + out
915
            end
916

    
917
            out
918
        end
919
    end
920

    
921
    def shelve( val ) 
922
        @shelf << val
923
        " :redsh##{ @shelf.length }:"
924
    end
925
    
926
    def retrieve( text ) 
927
        @shelf.each_with_index do |r, i|
928
            text.gsub!( " :redsh##{ i + 1 }:", r )
929
        end
930
    end
931

    
932
    def incoming_entities( text ) 
933
        ## turn any incoming ampersands into a dummy character for now.
934
        ## This uses a negative lookahead for alphanumerics followed by a semicolon,
935
        ## implying an incoming html entity, to be skipped
936

    
937
        text.gsub!( /&(?![#a-z0-9]+;)/i, "x%x%" )
938
    end
939

    
940
    def no_textile( text ) 
941
        text.gsub!( /(^|\s)==([^=]+.*?)==(\s|$)?/,
942
            '\1<notextile>\2</notextile>\3' )
943
        text.gsub!( /^ *==([^=]+.*?)==/m,
944
            '\1<notextile>\2</notextile>\3' )
945
    end
946

    
947
    def clean_white_space( text ) 
948
        # normalize line breaks
949
        text.gsub!( /\r\n/, "\n" )
950
        text.gsub!( /\r/, "\n" )
951
        text.gsub!( /\t/, '    ' )
952
        text.gsub!( /^ +$/, '' )
953
        text.gsub!( /\n{3,}/, "\n\n" )
954
        text.gsub!( /"$/, "\" " )
955

    
956
        # if entire document is indented, flush
957
        # to the left side
958
        flush_left text
959
    end
960

    
961
    def flush_left( text )
962
        indt = 0
963
        if text =~ /^ /
964
            while text !~ /^ {#{indt}}\S/
965
                indt += 1
966
            end unless text.empty?
967
            if indt.nonzero?
968
                text.gsub!( /^ {#{indt}}/, '' )
969
            end
970
        end
971
    end
972

    
973
    def footnote_ref( text ) 
974
        text.gsub!( /\b\[([0-9]+?)\](\s)?/,
975
            '<sup><a href="#fn\1">\1</a></sup>\2' )
976
    end
977
    
978
    OFFTAGS = /(code|pre|kbd|notextile)/
979
    OFFTAG_MATCH = /(?:(<\/#{ OFFTAGS }>)|(<#{ OFFTAGS }[^>]*>))(.*?)(?=<\/?#{ OFFTAGS }|\Z)/mi
980
    OFFTAG_OPEN = /<#{ OFFTAGS }/
981
    OFFTAG_CLOSE = /<\/?#{ OFFTAGS }/
982
    HASTAG_MATCH = /(<\/?\w[^\n]*?>)/m
983
    ALLTAG_MATCH = /(<\/?\w[^\n]*?>)|.*?(?=<\/?\w[^\n]*?>|$)/m
984

    
985
    def glyphs_textile( text, level = 0 )
986
        if text !~ HASTAG_MATCH
987
            pgl text
988
            footnote_ref text
989
        else
990
            codepre = 0
991
            text.gsub!( ALLTAG_MATCH ) do |line|
992
                ## matches are off if we're between <code>, <pre> etc.
993
                if $1
994
                    if line =~ OFFTAG_OPEN
995
                        codepre += 1
996
                    elsif line =~ OFFTAG_CLOSE
997
                        codepre -= 1
998
                        codepre = 0 if codepre < 0
999
                    end 
1000
                elsif codepre.zero?
1001
                    glyphs_textile( line, level + 1 )
1002
                else
1003
                    htmlesc( line, :NoQuotes )
1004
                end
1005
                # p [level, codepre, line]
1006

    
1007
                line
1008
            end
1009
        end
1010
    end
1011

    
1012
    def rip_offtags( text )
1013
        if text =~ /<.*>/
1014
            ## strip and encode <pre> content
1015
            codepre, used_offtags = 0, {}
1016
            text.gsub!( OFFTAG_MATCH ) do |line|
1017
                if $3
1018
                    offtag, aftertag = $4, $5
1019
                    codepre += 1
1020
                    used_offtags[offtag] = true
1021
                    if codepre - used_offtags.length > 0
1022
                        htmlesc( line, :NoQuotes ) unless used_offtags['notextile']
1023
                        @pre_list.last << line
1024
                        line = ""
1025
                    else
1026
                        htmlesc( aftertag, :NoQuotes ) if aftertag and not used_offtags['notextile']
1027
                        line = "<redpre##{ @pre_list.length }>"
1028
                        @pre_list << "#{ $3 }#{ aftertag }"
1029
                    end
1030
                elsif $1 and codepre > 0
1031
                    if codepre - used_offtags.length > 0
1032
                        htmlesc( line, :NoQuotes ) unless used_offtags['notextile']
1033
                        @pre_list.last << line
1034
                        line = ""
1035
                    end
1036
                    codepre -= 1 unless codepre.zero?
1037
                    used_offtags = {} if codepre.zero?
1038
                end 
1039
                line
1040
            end
1041
        end
1042
        text
1043
    end
1044

    
1045
    def smooth_offtags( text )
1046
        unless @pre_list.empty?
1047
            ## replace <pre> content
1048
            text.gsub!( /<redpre#(\d+)>/ ) { @pre_list[$1.to_i] }
1049
        end
1050
    end
1051

    
1052
    def inline( text ) 
1053
        [/^inline_/, /^glyphs_/].each do |meth_re|
1054
            @rules.each do |rule_name|
1055
                method( rule_name ).call( text ) if rule_name.to_s.match( meth_re )
1056
            end
1057
        end
1058
    end
1059

    
1060
    def h_align( text ) 
1061
        H_ALGN_VALS[text]
1062
    end
1063

    
1064
    def v_align( text ) 
1065
        V_ALGN_VALS[text]
1066
    end
1067

    
1068
    def textile_popup_help( name, windowW, windowH )
1069
        ' <a target="_blank" href="http://hobix.com/textile/#' + helpvar + '" onclick="window.open(this.href, \'popupwindow\', \'width=' + windowW + ',height=' + windowH + ',scrollbars,resizable\'); return false;">' + name + '</a><br />'
1070
    end
1071

    
1072
    # HTML cleansing stuff
1073
    BASIC_TAGS = {
1074
        'a' => ['href', 'title'],
1075
        'img' => ['src', 'alt', 'title'],
1076
        'br' => [],
1077
        'i' => nil,
1078
        'u' => nil, 
1079
        'b' => nil,
1080
        'pre' => nil,
1081
        'kbd' => nil,
1082
        'code' => ['lang'],
1083
        'cite' => nil,
1084
        'strong' => nil,
1085
        'em' => nil,
1086
        'ins' => nil,
1087
        'sup' => nil,
1088
        'sub' => nil,
1089
        'del' => nil,
1090
        'table' => nil,
1091
        'tr' => nil,
1092
        'td' => ['colspan', 'rowspan'],
1093
        'th' => nil,
1094
        'ol' => nil,
1095
        'ul' => nil,
1096
        'li' => nil,
1097
        'p' => nil,
1098
        'h1' => nil,
1099
        'h2' => nil,
1100
        'h3' => nil,
1101
        'h4' => nil,
1102
        'h5' => nil,
1103
        'h6' => nil, 
1104
        'blockquote' => ['cite']
1105
    }
1106

    
1107
    def clean_html( text, tags = BASIC_TAGS )
1108
        text.gsub!( /<!\[CDATA\[/, '' )
1109
        text.gsub!( /<(\/*)(\w+)([^>]*)>/ ) do
1110
            raw = $~
1111
            tag = raw[2].downcase
1112
            if tags.has_key? tag
1113
                pcs = [tag]
1114
                tags[tag].each do |prop|
1115
                    ['"', "'", ''].each do |q|
1116
                        q2 = ( q != '' ? q : '\s' )
1117
                        if raw[3] =~ /#{prop}\s*=\s*#{q}([^#{q2}]+)#{q}/i
1118
                            attrv = $1
1119
                            next if prop == 'src' and attrv =~ %r{^(?!http)\w+:}
1120
                            pcs << "#{prop}=\"#{$1.gsub('"', '\\"')}\""
1121
                            break
1122
                        end
1123
                    end
1124
                end if tags[tag]
1125
                "<#{raw[1]}#{pcs.join " "}>"
1126
            else
1127
                " "
1128
            end
1129
        end
1130
    end
1131
end
1132

    
(2-2/4)