##// END OF EJS Templates
Fixed: lines in wiki content containing just a single wiki link are not displayed....
Jean-Philippe Lang -
r724:780d5fa070ce
parent child
Show More
@@ -1,1130 +1,1130
1 1 # vim:ts=4:sw=4:
2 2 # = RedCloth - Textile and Markdown Hybrid for Ruby
3 3 #
4 4 # Homepage:: http://whytheluckystiff.net/ruby/redcloth/
5 5 # Author:: why the lucky stiff (http://whytheluckystiff.net/)
6 6 # Copyright:: (cc) 2004 why the lucky stiff (and his puppet organizations.)
7 7 # License:: BSD
8 8 #
9 9 # (see http://hobix.com/textile/ for a Textile Reference.)
10 10 #
11 11 # Based on (and also inspired by) both:
12 12 #
13 13 # PyTextile: http://diveintomark.org/projects/textile/textile.py.txt
14 14 # Textism for PHP: http://www.textism.com/tools/textile/
15 15 #
16 16 #
17 17
18 18 # = RedCloth
19 19 #
20 20 # RedCloth is a Ruby library for converting Textile and/or Markdown
21 21 # into HTML. You can use either format, intermingled or separately.
22 22 # You can also extend RedCloth to honor your own custom text stylings.
23 23 #
24 24 # RedCloth users are encouraged to use Textile if they are generating
25 25 # HTML and to use Markdown if others will be viewing the plain text.
26 26 #
27 27 # == What is Textile?
28 28 #
29 29 # Textile is a simple formatting style for text
30 30 # documents, loosely based on some HTML conventions.
31 31 #
32 32 # == Sample Textile Text
33 33 #
34 34 # h2. This is a title
35 35 #
36 36 # h3. This is a subhead
37 37 #
38 38 # This is a bit of paragraph.
39 39 #
40 40 # bq. This is a blockquote.
41 41 #
42 42 # = Writing Textile
43 43 #
44 44 # A Textile document consists of paragraphs. Paragraphs
45 45 # can be specially formatted by adding a small instruction
46 46 # to the beginning of the paragraph.
47 47 #
48 48 # h[n]. Header of size [n].
49 49 # bq. Blockquote.
50 50 # # Numeric list.
51 51 # * Bulleted list.
52 52 #
53 53 # == Quick Phrase Modifiers
54 54 #
55 55 # Quick phrase modifiers are also included, to allow formatting
56 56 # of small portions of text within a paragraph.
57 57 #
58 58 # \_emphasis\_
59 59 # \_\_italicized\_\_
60 60 # \*strong\*
61 61 # \*\*bold\*\*
62 62 # ??citation??
63 63 # -deleted text-
64 64 # +inserted text+
65 65 # ^superscript^
66 66 # ~subscript~
67 67 # @code@
68 68 # %(classname)span%
69 69 #
70 70 # ==notextile== (leave text alone)
71 71 #
72 72 # == Links
73 73 #
74 74 # To make a hypertext link, put the link text in "quotation
75 75 # marks" followed immediately by a colon and the URL of the link.
76 76 #
77 77 # Optional: text in (parentheses) following the link text,
78 78 # but before the closing quotation mark, will become a Title
79 79 # attribute for the link, visible as a tool tip when a cursor is above it.
80 80 #
81 81 # Example:
82 82 #
83 83 # "This is a link (This is a title) ":http://www.textism.com
84 84 #
85 85 # Will become:
86 86 #
87 87 # <a href="http://www.textism.com" title="This is a title">This is a link</a>
88 88 #
89 89 # == Images
90 90 #
91 91 # To insert an image, put the URL for the image inside exclamation marks.
92 92 #
93 93 # Optional: text that immediately follows the URL in (parentheses) will
94 94 # be used as the Alt text for the image. Images on the web should always
95 95 # have descriptive Alt text for the benefit of readers using non-graphical
96 96 # browsers.
97 97 #
98 98 # Optional: place a colon followed by a URL immediately after the
99 99 # closing ! to make the image into a link.
100 100 #
101 101 # Example:
102 102 #
103 103 # !http://www.textism.com/common/textist.gif(Textist)!
104 104 #
105 105 # Will become:
106 106 #
107 107 # <img src="http://www.textism.com/common/textist.gif" alt="Textist" />
108 108 #
109 109 # With a link:
110 110 #
111 111 # !/common/textist.gif(Textist)!:http://textism.com
112 112 #
113 113 # Will become:
114 114 #
115 115 # <a href="http://textism.com"><img src="/common/textist.gif" alt="Textist" /></a>
116 116 #
117 117 # == Defining Acronyms
118 118 #
119 119 # HTML allows authors to define acronyms via the tag. The definition appears as a
120 120 # tool tip when a cursor hovers over the acronym. A crucial aid to clear writing,
121 121 # this should be used at least once for each acronym in documents where they appear.
122 122 #
123 123 # To quickly define an acronym in Textile, place the full text in (parentheses)
124 124 # immediately following the acronym.
125 125 #
126 126 # Example:
127 127 #
128 128 # ACLU(American Civil Liberties Union)
129 129 #
130 130 # Will become:
131 131 #
132 132 # <acronym title="American Civil Liberties Union">ACLU</acronym>
133 133 #
134 134 # == Adding Tables
135 135 #
136 136 # In Textile, simple tables can be added by seperating each column by
137 137 # a pipe.
138 138 #
139 139 # |a|simple|table|row|
140 140 # |And|Another|table|row|
141 141 #
142 142 # Attributes are defined by style definitions in parentheses.
143 143 #
144 144 # table(border:1px solid black).
145 145 # (background:#ddd;color:red). |{}| | | |
146 146 #
147 147 # == Using RedCloth
148 148 #
149 149 # RedCloth is simply an extension of the String class, which can handle
150 150 # Textile formatting. Use it like a String and output HTML with its
151 151 # RedCloth#to_html method.
152 152 #
153 153 # doc = RedCloth.new "
154 154 #
155 155 # h2. Test document
156 156 #
157 157 # Just a simple test."
158 158 #
159 159 # puts doc.to_html
160 160 #
161 161 # By default, RedCloth uses both Textile and Markdown formatting, with
162 162 # Textile formatting taking precedence. If you want to turn off Markdown
163 163 # formatting, to boost speed and limit the processor:
164 164 #
165 165 # class RedCloth::Textile.new( str )
166 166
167 167 class RedCloth < String
168 168
169 169 VERSION = '3.0.4'
170 170 DEFAULT_RULES = [:textile, :markdown]
171 171
172 172 #
173 173 # Two accessor for setting security restrictions.
174 174 #
175 175 # This is a nice thing if you're using RedCloth for
176 176 # formatting in public places (e.g. Wikis) where you
177 177 # don't want users to abuse HTML for bad things.
178 178 #
179 179 # If +:filter_html+ is set, HTML which wasn't
180 180 # created by the Textile processor will be escaped.
181 181 #
182 182 # If +:filter_styles+ is set, it will also disable
183 183 # the style markup specifier. ('{color: red}')
184 184 #
185 185 attr_accessor :filter_html, :filter_styles
186 186
187 187 #
188 188 # Accessor for toggling hard breaks.
189 189 #
190 190 # If +:hard_breaks+ is set, single newlines will
191 191 # be converted to HTML break tags. This is the
192 192 # default behavior for traditional RedCloth.
193 193 #
194 194 attr_accessor :hard_breaks
195 195
196 196 # Accessor for toggling lite mode.
197 197 #
198 198 # In lite mode, block-level rules are ignored. This means
199 199 # that tables, paragraphs, lists, and such aren't available.
200 200 # Only the inline markup for bold, italics, entities and so on.
201 201 #
202 202 # r = RedCloth.new( "And then? She *fell*!", [:lite_mode] )
203 203 # r.to_html
204 204 # #=> "And then? She <strong>fell</strong>!"
205 205 #
206 206 attr_accessor :lite_mode
207 207
208 208 #
209 209 # Accessor for toggling span caps.
210 210 #
211 211 # Textile places `span' tags around capitalized
212 212 # words by default, but this wreaks havoc on Wikis.
213 213 # If +:no_span_caps+ is set, this will be
214 214 # suppressed.
215 215 #
216 216 attr_accessor :no_span_caps
217 217
218 218 #
219 219 # Establishes the markup predence. Available rules include:
220 220 #
221 221 # == Textile Rules
222 222 #
223 223 # The following textile rules can be set individually. Or add the complete
224 224 # set of rules with the single :textile rule, which supplies the rule set in
225 225 # the following precedence:
226 226 #
227 227 # refs_textile:: Textile references (i.e. [hobix]http://hobix.com/)
228 228 # block_textile_table:: Textile table block structures
229 229 # block_textile_lists:: Textile list structures
230 230 # block_textile_prefix:: Textile blocks with prefixes (i.e. bq., h2., etc.)
231 231 # inline_textile_image:: Textile inline images
232 232 # inline_textile_link:: Textile inline links
233 233 # inline_textile_span:: Textile inline spans
234 234 # glyphs_textile:: Textile entities (such as em-dashes and smart quotes)
235 235 #
236 236 # == Markdown
237 237 #
238 238 # refs_markdown:: Markdown references (for example: [hobix]: http://hobix.com/)
239 239 # block_markdown_setext:: Markdown setext headers
240 240 # block_markdown_atx:: Markdown atx headers
241 241 # block_markdown_rule:: Markdown horizontal rules
242 242 # block_markdown_bq:: Markdown blockquotes
243 243 # block_markdown_lists:: Markdown lists
244 244 # inline_markdown_link:: Markdown links
245 245 attr_accessor :rules
246 246
247 247 # Returns a new RedCloth object, based on _string_ and
248 248 # enforcing all the included _restrictions_.
249 249 #
250 250 # r = RedCloth.new( "h1. A <b>bold</b> man", [:filter_html] )
251 251 # r.to_html
252 252 # #=>"<h1>A &lt;b&gt;bold&lt;/b&gt; man</h1>"
253 253 #
254 254 def initialize( string, restrictions = [] )
255 255 restrictions.each { |r| method( "#{ r }=" ).call( true ) }
256 256 super( string )
257 257 end
258 258
259 259 #
260 260 # Generates HTML from the Textile contents.
261 261 #
262 262 # r = RedCloth.new( "And then? She *fell*!" )
263 263 # r.to_html( true )
264 264 # #=>"And then? She <strong>fell</strong>!"
265 265 #
266 266 def to_html( *rules )
267 267 rules = DEFAULT_RULES if rules.empty?
268 268 # make our working copy
269 269 text = self.dup
270 270
271 271 @urlrefs = {}
272 272 @shelf = []
273 273 textile_rules = [:refs_textile, :block_textile_table, :block_textile_lists,
274 274 :block_textile_prefix, :inline_textile_image, :inline_textile_link,
275 275 :inline_textile_code, :inline_textile_span, :glyphs_textile]
276 276 markdown_rules = [:refs_markdown, :block_markdown_setext, :block_markdown_atx, :block_markdown_rule,
277 277 :block_markdown_bq, :block_markdown_lists,
278 278 :inline_markdown_reflink, :inline_markdown_link]
279 279 @rules = rules.collect do |rule|
280 280 case rule
281 281 when :markdown
282 282 markdown_rules
283 283 when :textile
284 284 textile_rules
285 285 else
286 286 rule
287 287 end
288 288 end.flatten
289 289
290 290 # standard clean up
291 291 incoming_entities text
292 292 clean_white_space text
293 293
294 294 # start processor
295 295 @pre_list = []
296 296 rip_offtags text
297 297 no_textile text
298 298 hard_break text
299 299 unless @lite_mode
300 300 refs text
301 301 blocks text
302 302 end
303 303 inline text
304 304 smooth_offtags text
305 305
306 306 retrieve text
307 307
308 308 text.gsub!( /<\/?notextile>/, '' )
309 309 text.gsub!( /x%x%/, '&#38;' )
310 310 clean_html text if filter_html
311 311 text.strip!
312 312 text
313 313
314 314 end
315 315
316 316 #######
317 317 private
318 318 #######
319 319 #
320 320 # Mapping of 8-bit ASCII codes to HTML numerical entity equivalents.
321 321 # (from PyTextile)
322 322 #
323 323 TEXTILE_TAGS =
324 324
325 325 [[128, 8364], [129, 0], [130, 8218], [131, 402], [132, 8222], [133, 8230],
326 326 [134, 8224], [135, 8225], [136, 710], [137, 8240], [138, 352], [139, 8249],
327 327 [140, 338], [141, 0], [142, 0], [143, 0], [144, 0], [145, 8216], [146, 8217],
328 328 [147, 8220], [148, 8221], [149, 8226], [150, 8211], [151, 8212], [152, 732],
329 329 [153, 8482], [154, 353], [155, 8250], [156, 339], [157, 0], [158, 0], [159, 376]].
330 330
331 331 collect! do |a, b|
332 332 [a.chr, ( b.zero? and "" or "&#{ b };" )]
333 333 end
334 334
335 335 #
336 336 # Regular expressions to convert to HTML.
337 337 #
338 338 A_HLGN = /(?:(?:<>|<|>|\=|[()]+)+)/
339 339 A_VLGN = /[\-^~]/
340 340 C_CLAS = '(?:\([^)]+\))'
341 341 C_LNGE = '(?:\[[^\]]+\])'
342 342 C_STYL = '(?:\{[^}]+\})'
343 343 S_CSPN = '(?:\\\\\d+)'
344 344 S_RSPN = '(?:/\d+)'
345 345 A = "(?:#{A_HLGN}?#{A_VLGN}?|#{A_VLGN}?#{A_HLGN}?)"
346 346 S = "(?:#{S_CSPN}?#{S_RSPN}|#{S_RSPN}?#{S_CSPN}?)"
347 347 C = "(?:#{C_CLAS}?#{C_STYL}?#{C_LNGE}?|#{C_STYL}?#{C_LNGE}?#{C_CLAS}?|#{C_LNGE}?#{C_STYL}?#{C_CLAS}?)"
348 348 # PUNCT = Regexp::quote( '!"#$%&\'()*+,-./:;<=>?@[\\]^_`{|}~' )
349 349 PUNCT = Regexp::quote( '!"#$%&\'*+,-./:;=?@\\^_`|~' )
350 350 PUNCT_NOQ = Regexp::quote( '!"#$&\',./:;=?@\\`|' )
351 351 PUNCT_Q = Regexp::quote( '*-_+^~%' )
352 352 HYPERLINK = '(\S+?)([^\w\s/;=\?]*?)(?=\s|<|$)'
353 353
354 354 # Text markup tags, don't conflict with block tags
355 355 SIMPLE_HTML_TAGS = [
356 356 'tt', 'b', 'i', 'big', 'small', 'em', 'strong', 'dfn', 'code',
357 357 'samp', 'kbd', 'var', 'cite', 'abbr', 'acronym', 'a', 'img', 'br',
358 358 'br', 'map', 'q', 'sub', 'sup', 'span', 'bdo'
359 359 ]
360 360
361 361 QTAGS = [
362 362 ['**', 'b'],
363 363 ['*', 'strong'],
364 364 ['??', 'cite', :limit],
365 365 ['-', 'del', :limit],
366 366 ['__', 'i'],
367 367 ['_', 'em', :limit],
368 368 ['%', 'span', :limit],
369 369 ['+', 'ins', :limit],
370 370 ['^', 'sup'],
371 371 ['~', 'sub']
372 372 ]
373 373 QTAGS.collect! do |rc, ht, rtype|
374 374 rcq = Regexp::quote rc
375 375 re =
376 376 case rtype
377 377 when :limit
378 378 /(\W)
379 379 (#{rcq})
380 380 (#{C})
381 381 (?::(\S+?))?
382 382 (\S.*?\S|\S)
383 383 #{rcq}
384 384 (?=\W)/x
385 385 else
386 386 /(#{rcq})
387 387 (#{C})
388 388 (?::(\S+))?
389 389 (\S.*?\S|\S)
390 390 #{rcq}/xm
391 391 end
392 392 [rc, ht, re, rtype]
393 393 end
394 394
395 395 # Elements to handle
396 396 GLYPHS = [
397 397 # [ /([^\s\[{(>])?\'([dmst]\b|ll\b|ve\b|\s|:|$)/, '\1&#8217;\2' ], # single closing
398 398 [ /([^\s\[{(>#{PUNCT_Q}][#{PUNCT_Q}]*)\'/, '\1&#8217;' ], # single closing
399 399 [ /\'(?=[#{PUNCT_Q}]*(s\b|[\s#{PUNCT_NOQ}]))/, '&#8217;' ], # single closing
400 400 [ /\'/, '&#8216;' ], # single opening
401 401 [ /</, '&lt;' ], # less-than
402 402 [ />/, '&gt;' ], # greater-than
403 403 # [ /([^\s\[{(])?"(\s|:|$)/, '\1&#8221;\2' ], # double closing
404 404 [ /([^\s\[{(>#{PUNCT_Q}][#{PUNCT_Q}]*)"/, '\1&#8221;' ], # double closing
405 405 [ /"(?=[#{PUNCT_Q}]*[\s#{PUNCT_NOQ}])/, '&#8221;' ], # double closing
406 406 [ /"/, '&#8220;' ], # double opening
407 407 [ /\b( )?\.{3}/, '\1&#8230;' ], # ellipsis
408 408 [ /\b([A-Z][A-Z0-9]{2,})\b(?:[(]([^)]*)[)])/, '<acronym title="\2">\1</acronym>' ], # 3+ uppercase acronym
409 409 [ /(^|[^"][>\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
410 410 [ /(\.\s)?\s?--\s?/, '\1&#8212;' ], # em dash
411 411 [ /\s->\s/, ' &rarr; ' ], # right arrow
412 412 [ /\s-\s/, ' &#8211; ' ], # en dash
413 413 [ /(\d+) ?x ?(\d+)/, '\1&#215;\2' ], # dimension sign
414 414 [ /\b ?[(\[]TM[\])]/i, '&#8482;' ], # trademark
415 415 [ /\b ?[(\[]R[\])]/i, '&#174;' ], # registered
416 416 [ /\b ?[(\[]C[\])]/i, '&#169;' ] # copyright
417 417 ]
418 418
419 419 H_ALGN_VALS = {
420 420 '<' => 'left',
421 421 '=' => 'center',
422 422 '>' => 'right',
423 423 '<>' => 'justify'
424 424 }
425 425
426 426 V_ALGN_VALS = {
427 427 '^' => 'top',
428 428 '-' => 'middle',
429 429 '~' => 'bottom'
430 430 }
431 431
432 432 #
433 433 # Flexible HTML escaping
434 434 #
435 435 def htmlesc( str, mode )
436 436 str.gsub!( '&', '&amp;' )
437 437 str.gsub!( '"', '&quot;' ) if mode != :NoQuotes
438 438 str.gsub!( "'", '&#039;' ) if mode == :Quotes
439 439 str.gsub!( '<', '&lt;')
440 440 str.gsub!( '>', '&gt;')
441 441 end
442 442
443 443 # Search and replace for Textile glyphs (quotes, dashes, other symbols)
444 444 def pgl( text )
445 445 GLYPHS.each do |re, resub, tog|
446 446 next if tog and method( tog ).call
447 447 text.gsub! re, resub
448 448 end
449 449 end
450 450
451 451 # Parses Textile attribute lists and builds an HTML attribute string
452 452 def pba( text_in, element = "" )
453 453
454 454 return '' unless text_in
455 455
456 456 style = []
457 457 text = text_in.dup
458 458 if element == 'td'
459 459 colspan = $1 if text =~ /\\(\d+)/
460 460 rowspan = $1 if text =~ /\/(\d+)/
461 461 style << "vertical-align:#{ v_align( $& ) };" if text =~ A_VLGN
462 462 end
463 463
464 464 style << "#{ $1 };" if not filter_styles and
465 465 text.sub!( /\{([^}]*)\}/, '' )
466 466
467 467 lang = $1 if
468 468 text.sub!( /\[([^)]+?)\]/, '' )
469 469
470 470 cls = $1 if
471 471 text.sub!( /\(([^()]+?)\)/, '' )
472 472
473 473 style << "padding-left:#{ $1.length }em;" if
474 474 text.sub!( /([(]+)/, '' )
475 475
476 476 style << "padding-right:#{ $1.length }em;" if text.sub!( /([)]+)/, '' )
477 477
478 478 style << "text-align:#{ h_align( $& ) };" if text =~ A_HLGN
479 479
480 480 cls, id = $1, $2 if cls =~ /^(.*?)#(.*)$/
481 481
482 482 atts = ''
483 483 atts << " style=\"#{ style.join }\"" unless style.empty?
484 484 atts << " class=\"#{ cls }\"" unless cls.to_s.empty?
485 485 atts << " lang=\"#{ lang }\"" if lang
486 486 atts << " id=\"#{ id }\"" if id
487 487 atts << " colspan=\"#{ colspan }\"" if colspan
488 488 atts << " rowspan=\"#{ rowspan }\"" if rowspan
489 489
490 490 atts
491 491 end
492 492
493 493 TABLE_RE = /^(?:table(_?#{S}#{A}#{C})\. ?\n)?^(#{A}#{C}\.? ?\|.*?\|)(\n\n|\Z)/m
494 494
495 495 # Parses a Textile table block, building HTML from the result.
496 496 def block_textile_table( text )
497 497 text.gsub!( TABLE_RE ) do |matches|
498 498
499 499 tatts, fullrow = $~[1..2]
500 500 tatts = pba( tatts, 'table' )
501 501 tatts = shelve( tatts ) if tatts
502 502 rows = []
503 503
504 504 fullrow.
505 505 split( /\|$/m ).
506 506 delete_if { |x| x.empty? }.
507 507 each do |row|
508 508
509 509 ratts, row = pba( $1, 'tr' ), $2 if row =~ /^(#{A}#{C}\. )(.*)/m
510 510
511 511 cells = []
512 512 row.split( '|' ).each do |cell|
513 513 ctyp = 'd'
514 514 ctyp = 'h' if cell =~ /^_/
515 515
516 516 catts = ''
517 517 catts, cell = pba( $1, 'td' ), $2 if cell =~ /^(_?#{S}#{A}#{C}\. ?)(.*)/
518 518
519 519 unless cell.strip.empty?
520 520 catts = shelve( catts ) if catts
521 521 cells << "\t\t\t<t#{ ctyp }#{ catts }>#{ cell }</t#{ ctyp }>"
522 522 end
523 523 end
524 524 ratts = shelve( ratts ) if ratts
525 525 rows << "\t\t<tr#{ ratts }>\n#{ cells.join( "\n" ) }\n\t\t</tr>"
526 526 end
527 527 "\t<table#{ tatts }>\n#{ rows.join( "\n" ) }\n\t</table>\n\n"
528 528 end
529 529 end
530 530
531 531 LISTS_RE = /^([#*]+?#{C} .*?)$(?![^#*])/m
532 532 LISTS_CONTENT_RE = /^([#*]+)(#{A}#{C}) (.*)$/m
533 533
534 534 # Parses Textile lists and generates HTML
535 535 def block_textile_lists( text )
536 536 text.gsub!( LISTS_RE ) do |match|
537 537 lines = match.split( /\n/ )
538 538 last_line = -1
539 539 depth = []
540 540 lines.each_with_index do |line, line_id|
541 541 if line =~ LISTS_CONTENT_RE
542 542 tl,atts,content = $~[1..3]
543 543 if depth.last
544 544 if depth.last.length > tl.length
545 545 (depth.length - 1).downto(0) do |i|
546 546 break if depth[i].length == tl.length
547 547 lines[line_id - 1] << "</li>\n\t</#{ lT( depth[i] ) }l>\n\t"
548 548 depth.pop
549 549 end
550 550 end
551 551 if depth.last and depth.last.length == tl.length
552 552 lines[line_id - 1] << '</li>'
553 553 end
554 554 end
555 555 unless depth.last == tl
556 556 depth << tl
557 557 atts = pba( atts )
558 558 atts = shelve( atts ) if atts
559 559 lines[line_id] = "\t<#{ lT(tl) }l#{ atts }>\n\t<li>#{ content }"
560 560 else
561 561 lines[line_id] = "\t\t<li>#{ content }"
562 562 end
563 563 last_line = line_id
564 564
565 565 else
566 566 last_line = line_id
567 567 end
568 568 if line_id - last_line > 1 or line_id == lines.length - 1
569 569 depth.delete_if do |v|
570 570 lines[last_line] << "</li>\n\t</#{ lT( v ) }l>"
571 571 end
572 572 end
573 573 end
574 574 lines.join( "\n" )
575 575 end
576 576 end
577 577
578 578 CODE_RE = /(\W)
579 579 @
580 580 (?:\|(\w+?)\|)?
581 581 (.+?)
582 582 @
583 583 (?=\W)/x
584 584
585 585 def inline_textile_code( text )
586 586 text.gsub!( CODE_RE ) do |m|
587 587 before,lang,code,after = $~[1..4]
588 588 lang = " lang=\"#{ lang }\"" if lang
589 589 rip_offtags( "#{ before }<code#{ lang }>#{ code }</code>#{ after }" )
590 590 end
591 591 end
592 592
593 593 def lT( text )
594 594 text =~ /\#$/ ? 'o' : 'u'
595 595 end
596 596
597 597 def hard_break( text )
598 598 text.gsub!( /(.)\n(?!\Z| *([#*=]+(\s|$)|[{|]))/, "\\1<br />" ) if hard_breaks
599 599 end
600 600
601 601 BLOCKS_GROUP_RE = /\n{2,}(?! )/m
602 602
603 603 def blocks( text, deep_code = false )
604 604 text.replace( text.split( BLOCKS_GROUP_RE ).collect do |blk|
605 605 plain = blk !~ /\A[#*> ]/
606 606
607 607 # skip blocks that are complex HTML
608 608 if blk =~ /^<\/?(\w+).*>/ and not SIMPLE_HTML_TAGS.include? $1
609 609 blk
610 610 else
611 611 # search for indentation levels
612 612 blk.strip!
613 613 if blk.empty?
614 614 blk
615 615 else
616 616 code_blk = nil
617 617 blk.gsub!( /((?:\n(?:\n^ +[^\n]*)+)+)/m ) do |iblk|
618 618 flush_left iblk
619 619 blocks iblk, plain
620 620 iblk.gsub( /^(\S)/, "\t\\1" )
621 621 if plain
622 622 code_blk = iblk; ""
623 623 else
624 624 iblk
625 625 end
626 626 end
627 627
628 628 block_applied = 0
629 629 @rules.each do |rule_name|
630 630 block_applied += 1 if ( rule_name.to_s.match /^block_/ and method( rule_name ).call( blk ) )
631 631 end
632 632 if block_applied.zero?
633 633 if deep_code
634 634 blk = "\t<pre><code>#{ blk }</code></pre>"
635 635 else
636 636 blk = "\t<p>#{ blk }</p>"
637 637 end
638 638 end
639 639 # hard_break blk
640 640 blk + "\n#{ code_blk }"
641 641 end
642 642 end
643 643
644 644 end.join( "\n\n" ) )
645 645 end
646 646
647 647 def textile_bq( tag, atts, cite, content )
648 648 cite, cite_title = check_refs( cite )
649 649 cite = " cite=\"#{ cite }\"" if cite
650 650 atts = shelve( atts ) if atts
651 651 "\t<blockquote#{ cite }>\n\t\t<p#{ atts }>#{ content }</p>\n\t</blockquote>"
652 652 end
653 653
654 654 def textile_p( tag, atts, cite, content )
655 655 atts = shelve( atts ) if atts
656 656 "\t<#{ tag }#{ atts }>#{ content }</#{ tag }>"
657 657 end
658 658
659 659 alias textile_h1 textile_p
660 660 alias textile_h2 textile_p
661 661 alias textile_h3 textile_p
662 662 alias textile_h4 textile_p
663 663 alias textile_h5 textile_p
664 664 alias textile_h6 textile_p
665 665
666 666 def textile_fn_( tag, num, atts, cite, content )
667 667 atts << " id=\"fn#{ num }\""
668 668 content = "<sup>#{ num }</sup> #{ content }"
669 669 atts = shelve( atts ) if atts
670 670 "\t<p#{ atts }>#{ content }</p>"
671 671 end
672 672
673 673 BLOCK_RE = /^(([a-z]+)(\d*))(#{A}#{C})\.(?::(\S+))? (.*)$/m
674 674
675 675 def block_textile_prefix( text )
676 676 if text =~ BLOCK_RE
677 677 tag,tagpre,num,atts,cite,content = $~[1..6]
678 678 atts = pba( atts )
679 679
680 680 # pass to prefix handler
681 681 if respond_to? "textile_#{ tag }", true
682 682 text.gsub!( $&, method( "textile_#{ tag }" ).call( tag, atts, cite, content ) )
683 683 elsif respond_to? "textile_#{ tagpre }_", true
684 684 text.gsub!( $&, method( "textile_#{ tagpre }_" ).call( tagpre, num, atts, cite, content ) )
685 685 end
686 686 end
687 687 end
688 688
689 689 SETEXT_RE = /\A(.+?)\n([=-])[=-]* *$/m
690 690 def block_markdown_setext( text )
691 691 if text =~ SETEXT_RE
692 692 tag = if $2 == "="; "h1"; else; "h2"; end
693 693 blk, cont = "<#{ tag }>#{ $1 }</#{ tag }>", $'
694 694 blocks cont
695 695 text.replace( blk + cont )
696 696 end
697 697 end
698 698
699 699 ATX_RE = /\A(\#{1,6}) # $1 = string of #'s
700 700 [ ]*
701 701 (.+?) # $2 = Header text
702 702 [ ]*
703 703 \#* # optional closing #'s (not counted)
704 704 $/x
705 705 def block_markdown_atx( text )
706 706 if text =~ ATX_RE
707 707 tag = "h#{ $1.length }"
708 708 blk, cont = "<#{ tag }>#{ $2 }</#{ tag }>\n\n", $'
709 709 blocks cont
710 710 text.replace( blk + cont )
711 711 end
712 712 end
713 713
714 714 MARKDOWN_BQ_RE = /\A(^ *> ?.+$(.+\n)*\n*)+/m
715 715
716 716 def block_markdown_bq( text )
717 717 text.gsub!( MARKDOWN_BQ_RE ) do |blk|
718 718 blk.gsub!( /^ *> ?/, '' )
719 719 flush_left blk
720 720 blocks blk
721 721 blk.gsub!( /^(\S)/, "\t\\1" )
722 722 "<blockquote>\n#{ blk }\n</blockquote>\n\n"
723 723 end
724 724 end
725 725
726 726 MARKDOWN_RULE_RE = /^(#{
727 727 ['*', '-', '_'].collect { |ch| '( ?' + Regexp::quote( ch ) + ' ?){3,}' }.join( '|' )
728 728 })$/
729 729
730 730 def block_markdown_rule( text )
731 731 text.gsub!( MARKDOWN_RULE_RE ) do |blk|
732 732 "<hr />"
733 733 end
734 734 end
735 735
736 736 # XXX TODO XXX
737 737 def block_markdown_lists( text )
738 738 end
739 739
740 740 def inline_textile_span( text )
741 741 QTAGS.each do |qtag_rc, ht, qtag_re, rtype|
742 742 text.gsub!( qtag_re ) do |m|
743 743
744 744 case rtype
745 745 when :limit
746 746 sta,qtag,atts,cite,content = $~[1..5]
747 747 else
748 748 qtag,atts,cite,content = $~[1..4]
749 749 sta = ''
750 750 end
751 751 atts = pba( atts )
752 752 atts << " cite=\"#{ cite }\"" if cite
753 753 atts = shelve( atts ) if atts
754 754
755 755 "#{ sta }<#{ ht }#{ atts }>#{ content }</#{ ht }>"
756 756
757 757 end
758 758 end
759 759 end
760 760
761 761 LINK_RE = /
762 762 ([\s\[{(]|[#{PUNCT}])? # $pre
763 763 " # start
764 764 (#{C}) # $atts
765 765 ([^"]+?) # $text
766 766 \s?
767 767 (?:\(([^)]+?)\)(?="))? # $title
768 768 ":
769 769 (\S+?) # $url
770 770 (\/)? # $slash
771 771 ([^\w\/;]*?) # $post
772 772 (?=<|\s|$)
773 773 /x
774 774
775 775 def inline_textile_link( text )
776 776 text.gsub!( LINK_RE ) do |m|
777 777 pre,atts,text,title,url,slash,post = $~[1..7]
778 778
779 779 url, url_title = check_refs( url )
780 780 title ||= url_title
781 781
782 782 atts = pba( atts )
783 783 atts = " href=\"#{ url }#{ slash }\"#{ atts }"
784 784 atts << " title=\"#{ title }\"" if title
785 785 atts = shelve( atts ) if atts
786 786
787 787 "#{ pre }<a#{ atts }>#{ text }</a>#{ post }"
788 788 end
789 789 end
790 790
791 791 MARKDOWN_REFLINK_RE = /
792 792 \[([^\[\]]+)\] # $text
793 793 [ ]? # opt. space
794 794 (?:\n[ ]*)? # one optional newline followed by spaces
795 795 \[(.*?)\] # $id
796 796 /x
797 797
798 798 def inline_markdown_reflink( text )
799 799 text.gsub!( MARKDOWN_REFLINK_RE ) do |m|
800 800 text, id = $~[1..2]
801 801
802 802 if id.empty?
803 803 url, title = check_refs( text )
804 804 else
805 805 url, title = check_refs( id )
806 806 end
807 807
808 808 atts = " href=\"#{ url }\""
809 809 atts << " title=\"#{ title }\"" if title
810 810 atts = shelve( atts )
811 811
812 812 "<a#{ atts }>#{ text }</a>"
813 813 end
814 814 end
815 815
816 816 MARKDOWN_LINK_RE = /
817 817 \[([^\[\]]+)\] # $text
818 818 \( # open paren
819 819 [ \t]* # opt space
820 820 <?(.+?)>? # $href
821 821 [ \t]* # opt space
822 822 (?: # whole title
823 823 (['"]) # $quote
824 824 (.*?) # $title
825 825 \3 # matching quote
826 826 )? # title is optional
827 827 \)
828 828 /x
829 829
830 830 def inline_markdown_link( text )
831 831 text.gsub!( MARKDOWN_LINK_RE ) do |m|
832 832 text, url, quote, title = $~[1..4]
833 833
834 834 atts = " href=\"#{ url }\""
835 835 atts << " title=\"#{ title }\"" if title
836 836 atts = shelve( atts )
837 837
838 838 "<a#{ atts }>#{ text }</a>"
839 839 end
840 840 end
841 841
842 TEXTILE_REFS_RE = /(^ *)\[([^\n]+?)\](#{HYPERLINK})(?=\s|$)/
842 TEXTILE_REFS_RE = /(^ *)\[([^\[\n]+?)\](#{HYPERLINK})(?=\s|$)/
843 843 MARKDOWN_REFS_RE = /(^ *)\[([^\n]+?)\]:\s+<?(#{HYPERLINK})>?(?:\s+"((?:[^"]|\\")+)")?(?=\s|$)/m
844 844
845 845 def refs( text )
846 846 @rules.each do |rule_name|
847 847 method( rule_name ).call( text ) if rule_name.to_s.match /^refs_/
848 848 end
849 849 end
850 850
851 851 def refs_textile( text )
852 852 text.gsub!( TEXTILE_REFS_RE ) do |m|
853 853 flag, url = $~[2..3]
854 854 @urlrefs[flag.downcase] = [url, nil]
855 855 nil
856 856 end
857 857 end
858 858
859 859 def refs_markdown( text )
860 860 text.gsub!( MARKDOWN_REFS_RE ) do |m|
861 861 flag, url = $~[2..3]
862 862 title = $~[6]
863 863 @urlrefs[flag.downcase] = [url, title]
864 864 nil
865 865 end
866 866 end
867 867
868 868 def check_refs( text )
869 869 ret = @urlrefs[text.downcase] if text
870 870 ret || [text, nil]
871 871 end
872 872
873 873 IMAGE_RE = /
874 874 (<p>|.|^) # start of line?
875 875 \! # opening
876 876 (\<|\=|\>)? # optional alignment atts
877 877 (#{C}) # optional style,class atts
878 878 (?:\. )? # optional dot-space
879 879 ([^\s(!]+?) # presume this is the src
880 880 \s? # optional space
881 881 (?:\(((?:[^\(\)]|\([^\)]+\))+?)\))? # optional title
882 882 \! # closing
883 883 (?::#{ HYPERLINK })? # optional href
884 884 /x
885 885
886 886 def inline_textile_image( text )
887 887 text.gsub!( IMAGE_RE ) do |m|
888 888 stln,algn,atts,url,title,href,href_a1,href_a2 = $~[1..8]
889 889 atts = pba( atts )
890 890 atts = " src=\"#{ url }\"#{ atts }"
891 891 atts << " title=\"#{ title }\"" if title
892 892 atts << " alt=\"#{ title }\""
893 893 # size = @getimagesize($url);
894 894 # if($size) $atts.= " $size[3]";
895 895
896 896 href, alt_title = check_refs( href ) if href
897 897 url, url_title = check_refs( url )
898 898
899 899 out = ''
900 900 out << "<a#{ shelve( " href=\"#{ href }\"" ) }>" if href
901 901 out << "<img#{ shelve( atts ) } />"
902 902 out << "</a>#{ href_a1 }#{ href_a2 }" if href
903 903
904 904 if algn
905 905 algn = h_align( algn )
906 906 if stln == "<p>"
907 907 out = "<p style=\"float:#{ algn }\">#{ out }"
908 908 else
909 909 out = "#{ stln }<div style=\"float:#{ algn }\">#{ out }</div>"
910 910 end
911 911 else
912 912 out = stln + out
913 913 end
914 914
915 915 out
916 916 end
917 917 end
918 918
919 919 def shelve( val )
920 920 @shelf << val
921 921 " :redsh##{ @shelf.length }:"
922 922 end
923 923
924 924 def retrieve( text )
925 925 @shelf.each_with_index do |r, i|
926 926 text.gsub!( " :redsh##{ i + 1 }:", r )
927 927 end
928 928 end
929 929
930 930 def incoming_entities( text )
931 931 ## turn any incoming ampersands into a dummy character for now.
932 932 ## This uses a negative lookahead for alphanumerics followed by a semicolon,
933 933 ## implying an incoming html entity, to be skipped
934 934
935 935 text.gsub!( /&(?![#a-z0-9]+;)/i, "x%x%" )
936 936 end
937 937
938 938 def no_textile( text )
939 939 text.gsub!( /(^|\s)==([^=]+.*?)==(\s|$)?/,
940 940 '\1<notextile>\2</notextile>\3' )
941 941 text.gsub!( /^ *==([^=]+.*?)==/m,
942 942 '\1<notextile>\2</notextile>\3' )
943 943 end
944 944
945 945 def clean_white_space( text )
946 946 # normalize line breaks
947 947 text.gsub!( /\r\n/, "\n" )
948 948 text.gsub!( /\r/, "\n" )
949 949 text.gsub!( /\t/, ' ' )
950 950 text.gsub!( /^ +$/, '' )
951 951 text.gsub!( /\n{3,}/, "\n\n" )
952 952 text.gsub!( /"$/, "\" " )
953 953
954 954 # if entire document is indented, flush
955 955 # to the left side
956 956 flush_left text
957 957 end
958 958
959 959 def flush_left( text )
960 960 indt = 0
961 961 if text =~ /^ /
962 962 while text !~ /^ {#{indt}}\S/
963 963 indt += 1
964 964 end unless text.empty?
965 965 if indt.nonzero?
966 966 text.gsub!( /^ {#{indt}}/, '' )
967 967 end
968 968 end
969 969 end
970 970
971 971 def footnote_ref( text )
972 972 text.gsub!( /\b\[([0-9]+?)\](\s)?/,
973 973 '<sup><a href="#fn\1">\1</a></sup>\2' )
974 974 end
975 975
976 976 OFFTAGS = /(code|pre|kbd|notextile)/
977 977 OFFTAG_MATCH = /(?:(<\/#{ OFFTAGS }>)|(<#{ OFFTAGS }[^>]*>))(.*?)(?=<\/?#{ OFFTAGS }|\Z)/mi
978 978 OFFTAG_OPEN = /<#{ OFFTAGS }/
979 979 OFFTAG_CLOSE = /<\/?#{ OFFTAGS }/
980 980 HASTAG_MATCH = /(<\/?\w[^\n]*?>)/m
981 981 ALLTAG_MATCH = /(<\/?\w[^\n]*?>)|.*?(?=<\/?\w[^\n]*?>|$)/m
982 982
983 983 def glyphs_textile( text, level = 0 )
984 984 if text !~ HASTAG_MATCH
985 985 pgl text
986 986 footnote_ref text
987 987 else
988 988 codepre = 0
989 989 text.gsub!( ALLTAG_MATCH ) do |line|
990 990 ## matches are off if we're between <code>, <pre> etc.
991 991 if $1
992 992 if line =~ OFFTAG_OPEN
993 993 codepre += 1
994 994 elsif line =~ OFFTAG_CLOSE
995 995 codepre -= 1
996 996 codepre = 0 if codepre < 0
997 997 end
998 998 elsif codepre.zero?
999 999 glyphs_textile( line, level + 1 )
1000 1000 else
1001 1001 htmlesc( line, :NoQuotes )
1002 1002 end
1003 1003 # p [level, codepre, line]
1004 1004
1005 1005 line
1006 1006 end
1007 1007 end
1008 1008 end
1009 1009
1010 1010 def rip_offtags( text )
1011 1011 if text =~ /<.*>/
1012 1012 ## strip and encode <pre> content
1013 1013 codepre, used_offtags = 0, {}
1014 1014 text.gsub!( OFFTAG_MATCH ) do |line|
1015 1015 if $3
1016 1016 offtag, aftertag = $4, $5
1017 1017 codepre += 1
1018 1018 used_offtags[offtag] = true
1019 1019 if codepre - used_offtags.length > 0
1020 1020 htmlesc( line, :NoQuotes ) unless used_offtags['notextile']
1021 1021 @pre_list.last << line
1022 1022 line = ""
1023 1023 else
1024 1024 htmlesc( aftertag, :NoQuotes ) if aftertag and not used_offtags['notextile']
1025 1025 line = "<redpre##{ @pre_list.length }>"
1026 1026 @pre_list << "#{ $3 }#{ aftertag }"
1027 1027 end
1028 1028 elsif $1 and codepre > 0
1029 1029 if codepre - used_offtags.length > 0
1030 1030 htmlesc( line, :NoQuotes ) unless used_offtags['notextile']
1031 1031 @pre_list.last << line
1032 1032 line = ""
1033 1033 end
1034 1034 codepre -= 1 unless codepre.zero?
1035 1035 used_offtags = {} if codepre.zero?
1036 1036 end
1037 1037 line
1038 1038 end
1039 1039 end
1040 1040 text
1041 1041 end
1042 1042
1043 1043 def smooth_offtags( text )
1044 1044 unless @pre_list.empty?
1045 1045 ## replace <pre> content
1046 1046 text.gsub!( /<redpre#(\d+)>/ ) { @pre_list[$1.to_i] }
1047 1047 end
1048 1048 end
1049 1049
1050 1050 def inline( text )
1051 1051 [/^inline_/, /^glyphs_/].each do |meth_re|
1052 1052 @rules.each do |rule_name|
1053 1053 method( rule_name ).call( text ) if rule_name.to_s.match( meth_re )
1054 1054 end
1055 1055 end
1056 1056 end
1057 1057
1058 1058 def h_align( text )
1059 1059 H_ALGN_VALS[text]
1060 1060 end
1061 1061
1062 1062 def v_align( text )
1063 1063 V_ALGN_VALS[text]
1064 1064 end
1065 1065
1066 1066 def textile_popup_help( name, windowW, windowH )
1067 1067 ' <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 />'
1068 1068 end
1069 1069
1070 1070 # HTML cleansing stuff
1071 1071 BASIC_TAGS = {
1072 1072 'a' => ['href', 'title'],
1073 1073 'img' => ['src', 'alt', 'title'],
1074 1074 'br' => [],
1075 1075 'i' => nil,
1076 1076 'u' => nil,
1077 1077 'b' => nil,
1078 1078 'pre' => nil,
1079 1079 'kbd' => nil,
1080 1080 'code' => ['lang'],
1081 1081 'cite' => nil,
1082 1082 'strong' => nil,
1083 1083 'em' => nil,
1084 1084 'ins' => nil,
1085 1085 'sup' => nil,
1086 1086 'sub' => nil,
1087 1087 'del' => nil,
1088 1088 'table' => nil,
1089 1089 'tr' => nil,
1090 1090 'td' => ['colspan', 'rowspan'],
1091 1091 'th' => nil,
1092 1092 'ol' => nil,
1093 1093 'ul' => nil,
1094 1094 'li' => nil,
1095 1095 'p' => nil,
1096 1096 'h1' => nil,
1097 1097 'h2' => nil,
1098 1098 'h3' => nil,
1099 1099 'h4' => nil,
1100 1100 'h5' => nil,
1101 1101 'h6' => nil,
1102 1102 'blockquote' => ['cite']
1103 1103 }
1104 1104
1105 1105 def clean_html( text, tags = BASIC_TAGS )
1106 1106 text.gsub!( /<!\[CDATA\[/, '' )
1107 1107 text.gsub!( /<(\/*)(\w+)([^>]*)>/ ) do
1108 1108 raw = $~
1109 1109 tag = raw[2].downcase
1110 1110 if tags.has_key? tag
1111 1111 pcs = [tag]
1112 1112 tags[tag].each do |prop|
1113 1113 ['"', "'", ''].each do |q|
1114 1114 q2 = ( q != '' ? q : '\s' )
1115 1115 if raw[3] =~ /#{prop}\s*=\s*#{q}([^#{q2}]+)#{q}/i
1116 1116 attrv = $1
1117 1117 next if prop == 'src' and attrv =~ %r{^(?!http)\w+:}
1118 1118 pcs << "#{prop}=\"#{$1.gsub('"', '\\"')}\""
1119 1119 break
1120 1120 end
1121 1121 end
1122 1122 end if tags[tag]
1123 1123 "<#{raw[1]}#{pcs.join " "}>"
1124 1124 else
1125 1125 " "
1126 1126 end
1127 1127 end
1128 1128 end
1129 1129 end
1130 1130
General Comments 0
You need to be logged in to leave comments. Login now