Class: Raakt::Test

Inherits:
Object
  • Object
show all
Defined in:
lib/raakt.rb,
lib/iso_language_codes.rb

Constant Summary collapse

ISO_CODES = 
[
"aa",
"ab",
"ae",
"af",
"ak",
"am",
"an",
"ar",
"as",
"av",
"ay",
"az",
"ba",
"be",
"bg",
"bh",
"bi",
"bm",
"bn",
"bo",
"br",
"bs",
"ca",
"ce",
"ch",
"co",
"cr",
"cs",
"cv",
"cy",
"da",
"de",
"dv",
"dz",
"ee",
"el",
"en",
"eo",
"es",
"et",
"eu",
"fa",
"ff",
"fi",
"fj",
"fo",
"fr",
"fy",
"ga",
"gd",
"gl",
"gn",
"gu",
"gv",
"ha",
"he",
"hi",
"ho",
"hr",
"ht",
"hu",
"hy",
"hz",
"ia",
"id",
"ie",
"ig",
"ii",
"ik",
"io",
"is",
"it",
"iu",
"ja",
"jv",
"ka",
"kg",
"ki",
"kj",
"kk",
"kl",
"km",
"kn",
"ko",
"kr",
"ks",
"ku",
"kv",
"kw",
"ky",
"la",
"lb",
"lg",
"li",
"ln",
"lo",
"lt",
"lv",
"mg",
"mh",
"mi",
"mk",
"ml",
"mn",
"mo",
"mr",
"ms",
"mt",
"my",
"na",
"nb",
"nd",
"ne",
"ng",
"nl",
"nn",
"no",
"nr",
"nv",
"ny",
"oc",
"oj",
"om",
"or",
"os",
"pa",
"pi",
"pl",
"ps",
"pt",
"qu",
"rm",
"rn",
"ro",
"ru",
"rw",
"sa",
"sc",
"sd",
"se",
"sg",
"sh",
"si",
"sk",
"sl",
"sm",
"sn",
"so",
"sq",
"sr",
"ss",
"st",
"su",
"sv",
"sw",
"ta",
"te",
"tg",
"th",
"ti",
"tk",
"tl",
"tn",
"to",
"tr",
"ts",
"tt",
"tw",
"ty",
"ug",
"uk",
"ur",
"uz",
"ve",
"vi",
"vo",
"wa",
"wo",
"xh",
"yi",
"yo",
"za",
"zh",
"zu"]

Instance Attribute Summary collapse

Instance Method Summary collapse

Constructor Details

#initialize(html = nil, headers = nil) ⇒ Test

Returns a new instance of Test.



83
84
85
86
87
88
89
 
# File 'lib/raakt.rb', line 83

def initialize(html=nil, headers=nil)
  @html = html
	  @headers = headers
  self.doc = @html if html
	  self.headers = @headers if headers
	  @ignore_bi = false 
end

Instance Attribute Details

#headers ⇒ Object

Returns the value of attribute headers.



81
82
83
 
# File 'lib/raakt.rb', line 81

def headers
  @headers
end

#html ⇒ Object

Returns the value of attribute html.



81
82
83
 
# File 'lib/raakt.rb', line 81

def html
  @html
end

#ignore_bi ⇒ Object

Returns the value of attribute ignore_bi.



81
82
83
 
# File 'lib/raakt.rb', line 81

def ignore_bi
  @ignore_bi
end

#user_agent ⇒ Object

Returns the value of attribute user_agent.



81
82
83
 
# File 'lib/raakt.rb', line 81

def user_agent
  @user_agent
end

Instance Method Details

#all ⇒ Object

Call all check methods.



108
109
110
111
112
113
114
115
116
117
118
 
# File 'lib/raakt.rb', line 108

def all
  messages = []
  
  self.methods.each do |method|
    if method[0..5] == "check_"
      messages += self.send(method)
    end
  end
  
  return messages
end

#alt_to_text(element) ⇒ Object 



477
478
479
480
481
482
483
 
# File 'lib/raakt.rb', line 477

def alt_to_text(element)
		if element.kind_of?(Hpricot::Elem) then
  		element.has_attribute?("alt") ? element['alt'] : ""
		else
			""
		end
end

#check_areas ⇒ Object

Verify that all area elements have a non-empty alt attribute. See UWEM 1.0 Test 1.1_HTML_01 (together with check_images)



200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
 
# File 'lib/raakt.rb', line 200

def check_areas
		messages = []
		area_elements = (@doc/"area")
		area_elements.map { |element| 
			unless element['alt']
messages << ErrorMessage.new(:missing_area_alt, element['name'] || element['id'] || "unknown") 
			else
if element['alt'].length == 0
	messages << ErrorMessage.new(:missing_area_alt_text, element['name'] || element['id'] || "unknown")
end
			end
		}

		messages
end

#check_character_set ⇒ Object

Verify that the charater set specified in HTTP headers match that specidied in the HTML meta element.



144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
 
# File 'lib/raakt.rb', line 144

def check_character_set
	messages = []
	header_charset = meta_charset = ""
	if @headers and @headers.length > 0 then
		if @headers.has_key?("content-type")
			header_charset = parse_charset(@headers["content-type"].to_s)
		end

		#get meta element charset
		meta_elements = @doc.search("//meta[@http-equiv]")
		for element in meta_elements do
			if element["http-equiv"].downcase == "content-type" then
				meta_charset = parse_charset(element["content"])
			end
		end

		if header_charset.length > 0 and meta_charset.length > 0
			unless meta_charset == header_charset
				messages << ErrorMessage.new(:charset_mismatch) 
			end
		end
	end

	return messages

end

#check_document_structure ⇒ Object

Verify that heading elements (h1-h6) appear in the correct order (no levels skipped). See UWEM 1.0 Test 3.5_HTML_03.



235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
 
# File 'lib/raakt.rb', line 235

def check_document_structure
  messages = []
  currentitem = 0
  docheadings = headings
  
  for heading in docheadings
    if currentitem == 0
      if level(heading.name) != 1
        messages << ErrorMessage.new(:first_h_not_h1, "h" + heading.name[1,1])
      end
    else
      if level(heading.name) - level(docheadings[currentitem - 1].name) > 1
        messages << ErrorMessage.new(:wrong_h_structure)
        break
      end  
    end
    
    currentitem += 1
    
  end
  
  messages
end

#check_embed ⇒ Object

Verify that the embed element isn’t used. See UWEM 1.0 Test 1.1_HTML_06.



137
138
139
140
 
# File 'lib/raakt.rb', line 137

def check_embed
	return [ErrorMessage.new(:embed_used)] unless (@doc/'embed').empty?
	[]
end

#check_fieldset_legend ⇒ Object

Verify that all fieldset elements have a legend child element. See UWEM 1.0 Test 12.3_HTML_01.



122
123
124
125
126
127
128
129
130
131
132
133
 
# File 'lib/raakt.rb', line 122

def check_fieldset_legend
	messages = []
	fieldsets = (@doc/"fieldset")
	fieldset_instance = 1
	for fieldset in fieldsets 
		if (fieldset/"legend").empty?
			messages << ErrorMessage.new(:fieldset_missing_legend, fieldset_instance.to_s)				
		end
		fieldset_instance += 1
	end
	messages
end

#check_for_formatting_elements ⇒ Object

Verify that no formatting elements have been used. See UWEM 1.0 Test 7.2_HTML_01 and Test 7.3_HTML_01.



298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
 
# File 'lib/raakt.rb', line 298

def check_for_formatting_elements
  
  	messages = []

formatting_elements = %w(font b i u tt small big strike s)
		formatting_elements = %w(font u tt small big strike s) if @ignore_bi
	  
 formatting_items = (@doc/formatting_elements.join('|'))
  
  	unless formatting_items.empty?
    	messages << ErrorMessage.new(:missing_semantics, "#{formatting_items.join(', ')}")  
 end
	  
 flicker_elements = %w(blink marquee)
 flicker_items = (@doc/flicker_elements.join('|'))
 
  	unless flicker_items.empty?
    	messages << ErrorMessage.new(:has_flicker)  
  	end

  	messages   
end

#check_for_language_info ⇒ Object

Verify that the root documet html element as a lang attribute.



323
324
325
326
327
328
329
330
331
332
333
334
 
# File 'lib/raakt.rb', line 323

def check_for_language_info
  messages = []  
	  unless (@doc/'html[@lang]').empty?
lang_code = (@doc/"html").first["lang"].to_s
if lang_code.length < 2
  		messages << ErrorMessage.new(:missing_lang_info) 
end
	  else
  	messages << ErrorMessage.new(:missing_lang_info) 
	  end
	  messages
end

#check_for_nested_tables ⇒ Object

Verify that the document does not have any nested tabled. This is indicative of a table-based layout.



261
262
263
264
265
266
267
268
269
270
271
272
273
 
# File 'lib/raakt.rb', line 261

def check_for_nested_tables
  
  messages = []  
  tables = (@doc/"table")
  
  for table in tables
    unless (table/"table").empty?
      return messages << ErrorMessage.new(:has_nested_tables)
    end
  end
  
  messages
end

#check_form ⇒ Object

Verify that all form fields have a corresponding label element. See UWEM 1.0 Test 12.4_HTML_02.



369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
 
# File 'lib/raakt.rb', line 369

def check_form
  messages = []
  labels = get_labels
  fields = get_editable_fields
  
  #make sure all fields have associated labels
  label_for_ids = []
  for label in labels
    if label["for"]
      label_for_ids << label["for"]
    end
  end
  
  field_id = nil
  
  for field in fields
    field_id = (field["id"] || "")
    field_identifier = (field["id"] || field["name"] || "unknown")
    if not label_for_ids.include?(field_id)
      messages << ErrorMessage.new(:field_missing_label, field_identifier)
    end
  end   
  
  messages
end

#check_frames ⇒ Object

Verify that all frame elements have a title atribute.



397
398
399
400
401
402
403
404
405
 
# File 'lib/raakt.rb', line 397

def check_frames
	  # Covers UWEM Test 12.1_HTML_01
  return [] unless is_frameset
  
  (@doc/"frame").find_all do |frame|
    frame_title = frame['title'] || ''
    normalize_text(frame_title).empty?
  end.map { |frame| ErrorMessage.new(:missing_frame_title, frame['src']) }            
end

#check_has_heading ⇒ Object

Verify that the document has at least one h1 element.



228
229
230
231
 
# File 'lib/raakt.rb', line 228

def check_has_heading
  return [ErrorMessage.new(:missing_heading)] if (@doc/'h1').empty?
  []
end

#check_images ⇒ Object

Verify that all img elements have an alt attribute.



193
194
195
196
 
# File 'lib/raakt.rb', line 193

def check_images
  no_alt_images = (@doc/"img:not([@alt])")
  no_alt_images.map { |img| ErrorMessage.new(:missing_alt, img['src']) }
end

#check_input_type_img ⇒ Object

Verify that all input type=image elements have an alt attribute.



173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
 
# File 'lib/raakt.rb', line 173

def check_input_type_img
	#Covers UWEM 1.0 Test 1.1_HTML_01

	messages = []
	image_input_buttons = @doc.search("input").select { |element| element['type'] =~ /image/i }
	image_input_buttons.map { |element| 
		unless element['alt']
			messages << ErrorMessage.new(:missing_input_alt, element['name'] || element['id'] || "") 
		else
			if element['alt'].length == 0
				messages << ErrorMessage.new(:missing_input_alt_text, element['name'] || element['id'] || "")
			end
		end
	}

	messages
end

#check_link_text ⇒ Object

Verify that no link texts are ambiguous. A typical example is the presence of multiple “Read more” links.



356
357
358
359
360
361
362
363
364
365
 
# File 'lib/raakt.rb', line 356

def check_link_text
  links = get_links
  
  link = links.find do |link|
    links.find { |cmp_link| is_ambiguous_link(link, cmp_link) }
  end
  
  return [] unless link
  [ErrorMessage.new(:ambiguous_link_text, get_link_text(link))]
end

#check_refresh ⇒ Object

Verify that the document does not use meta-refresh to redirect the user away after a period of time.



409
410
411
412
413
414
415
 
# File 'lib/raakt.rb', line 409

def check_refresh
  meta_elements = (@doc/'meta')
  
  meta_elements.find_all do |element|
    element["http-equiv"] == "refresh"
  end.map { ErrorMessage.new(:has_meta_refresh) }
end

#check_tables ⇒ Object

Verify that all tables have at least on table header (th) element.



277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
 
# File 'lib/raakt.rb', line 277

def check_tables
  messages = []  
  tables = (@doc/"table")   
  currenttable = 1
  
  for table in tables     
  	hasth = false
    hasth = true unless (table/">tr>th").empty?
    hasth = true unless (table/">thead>tr>th").empty?
    
    messages << ErrorMessage.new(:missing_th, currenttable.to_s) unless hasth
            
    currenttable += 1
  end
  
  messages
end

#check_title ⇒ Object

Verify that the document has a non-empty title element.



219
220
221
222
223
224
 
# File 'lib/raakt.rb', line 219

def check_title
  title = @doc.at('title')
  return [ErrorMessage.new(:missing_title)] unless title
  return [ErrorMessage.new(:empty_title)] if normalize_text(title.inner_html).empty?
  []			
end

#check_valid_language_code ⇒ Object

Verify that the html element has a valid lang code.



338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
 
# File 'lib/raakt.rb', line 338

def check_valid_language_code
  messages = []
  unless (@doc/"html[@lang]").empty?
	#load list of valid language codes
	#iso_lang_codes = []
	#IO.foreach(File.dirname(__FILE__) + "/iso_language_codes.txt") { |code| iso_lang_codes << code.chomp }

	doc_main_lang_code = (@doc/"html").first["lang"].to_s.downcase
	unless ISO_CODES.include?(doc_main_lang_code[0..1])
		messages << ErrorMessage.new(:wrong_lang_code, doc_main_lang_code)
	end
  end

  messages
end

#doc=(html) ⇒ Object

Set the HTML used in the test.



92
93
94
95
 
# File 'lib/raakt.rb', line 92

def doc=(html)
	  Hpricot.buffer_size = 262144 #Allow for asp.net bastard-sized viewstate attributes...
  @doc = Hpricot(html)
end

#downcase_hash_keys(a_hash) ⇒ Object 



434
435
436
437
438
 
# File 'lib/raakt.rb', line 434

def downcase_hash_keys(a_hash)
	downcased_hash = {}
	a_hash.collect {|key,value| downcased_hash[key.downcase] = value}
	return downcased_hash
end

#elements_to_text(element) ⇒ Object 



485
486
487
488
489
490
491
492
 
# File 'lib/raakt.rb', line 485

def elements_to_text(element)
  str = ''
  element.traverse_all_element do |elem|
    elem.kind_of?(Hpricot::Text) ? str += "#{elem}" : str += alt_to_text(elem)
  end
  
  str
end

#get_editable_fields ⇒ Object 



517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
 
# File 'lib/raakt.rb', line 517

def get_editable_fields
  allfields = (@doc/"textarea|select|input")
  fields = []
  field_type = ""
  
  for field in allfields do
    field_type = field["type"] || ""
    unless ["button", "submit", "hidden", "image"].include?(field_type)
      fields << field
    end
    
  end
  
  return fields
end

#get_labels ⇒ Object 



512
513
514
 
# File 'lib/raakt.rb', line 512

def get_labels
  @doc/'label'
end

#get_link_text(link) ⇒ Object 



552
553
554
555
 
# File 'lib/raakt.rb', line 552

def get_link_text(link)
  text = (elements_to_text(link) || '').strip
  normalize_text(text)
end

#get_link_title(link) ⇒ Object 



561
562
563
564
 
# File 'lib/raakt.rb', line 561

def get_link_title(link)
  text = (link['title'] || '').strip
  normalize_text(text)
end

#get_link_url(link) ⇒ Object 



557
558
559
 
# File 'lib/raakt.rb', line 557

def get_link_url(link)
  link['href']
end

#get_links ⇒ Object 



458
459
460
 
# File 'lib/raakt.rb', line 458

def get_links      
  (@doc/'a')
end

#headings ⇒ Object

Utility methods



420
421
422
423
424
425
426
 
# File 'lib/raakt.rb', line 420

def headings
  headings = []
  1.upto(6) do |i| 
    headings.push((@doc/"h#{i}")) if (@doc/"h#{i}").length > 0
  end
  headings.flatten
end

#is_ambiguous_link(link_a, link_b) ⇒ Object 



450
451
452
453
454
455
456
 
# File 'lib/raakt.rb', line 450

def is_ambiguous_link(link_a, link_b)
  return false if links_point_to_same_resource?(link_a, link_b)
  return true if link_text_identical?(link_a, link_b) &&
                 link_title_identical?(link_a, link_b)
  
  false
end

#is_frameset ⇒ Object 



534
535
536
 
# File 'lib/raakt.rb', line 534

def is_frameset
  (@doc/"frameset").length > 0
end

#langinfo(element) ⇒ Object 



462
463
464
465
466
467
468
469
470
471
472
473
474
 
# File 'lib/raakt.rb', line 462

def langinfo(element)
  langval = ""
  
  if element.class.to_s == 'Tag'      
    if element['lang']
      langval = element['lang']
    end      
  else
    return nil
  end
  
  return langval
end

#level(heading) ⇒ Object 



429
430
431
 
# File 'lib/raakt.rb', line 429

def level(heading)
  Integer(heading[1].chr)
end

#link_text_identical?(link_a, link_b) ⇒ Boolean

Returns:

  • (Boolean) 


539
540
541
 
# File 'lib/raakt.rb', line 539

def link_text_identical?(link_a, link_b)
  get_link_text(link_a) == get_link_text(link_b)
end

#link_title_identical?(link_a, link_b) ⇒ Boolean

Returns:

  • (Boolean) 


543
544
545
 
# File 'lib/raakt.rb', line 543

def link_title_identical?(link_a, link_b)
  get_link_title(link_a) == get_link_title(link_b)
end

#links_point_to_same_resource?(link_a, link_b) ⇒ Boolean

Returns:

  • (Boolean) 


547
548
549
550
 
# File 'lib/raakt.rb', line 547

def links_point_to_same_resource?(link_a, link_b)
  (link_a == link_b) ||
  (get_link_url(link_a) == get_link_url(link_b))
end

#normalize_text(text) ⇒ Object 



495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
 
# File 'lib/raakt.rb', line 495

def normalize_text(text)
  text ||= ''
  retval = text.gsub(/&nbsp;/, ' ')
  retval = retval.gsub(/&#160;/, ' ')
  retval = retval.gsub(/\n/, '')
  retval = retval.gsub(/\r/, '')
  retval = retval.gsub(/\t/, '')
  while /  /.match(retval) do
    retval = retval.gsub(/  /, ' ')
  end
  
  retval = retval.strip
  
  return retval
end

#parse_charset(contenttype) ⇒ Object 



440
441
442
443
444
445
446
447
 
# File 'lib/raakt.rb', line 440

def parse_charset(contenttype)
	# get charset identifier from content type string
	if contenttype=~/charset=(.*)\w?/ then
		return $1.downcase.strip
	end

	return ""
end