Class: HRefPreview::Preview

Inherits:
Object
  • Object
show all
Defined in:
lib/href_preview/preview.rb

Constant Summary collapse

SANITIZE_OPTIONS = 
options

Instance Attribute Summary collapse

Instance Method Summary collapse

Constructor Details

#initialize(response, connection = DEFAULT_CONNECTION) ⇒ Preview

Initializes a ‘Preview` from an HTTP response.



28
29
30
31
 
# File 'lib/href_preview/preview.rb', line 28

def initialize(response, connection=DEFAULT_CONNECTION)
  @response = response
  @connection = connection
end

Instance Attribute Details

#response ⇒ Object (readonly)

Returns the value of attribute response.



33
34
35
 
# File 'lib/href_preview/preview.rb', line 33

def response
  @response
end

Instance Method Details

#article_html ⇒ Object 



342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
 
# File 'lib/href_preview/preview.rb', line 342

def article_html
  @article_html ||= (if is_html?
    begin
      html = nil
      if article_node
        html = article_node.children.reject do |child|
          next unless child.attribute('class')
          [
            'related_links_inline',
            'inline-share-btn-label',
            'inline-share-btn'
          ].include?(child.attribute('class').value)
        end.map(&:to_s).join('')
      end
      if html
        html = Sanitize.clean(html, SANITIZE_OPTIONS)
        html.gsub!("\r\n", "\n")
        html.gsub!("\t", "  ")
        html.gsub!(/ *\n */, "\n")
        html.gsub!(/\n\n+/, "\n\n")
        html.gsub!(/<p>\n+/, "<p>\n")
        html.gsub!(/\n+<\/p>/, "\n</p>")
        html.gsub!(/<\/p>\n+/, "</p>\n")
        html.strip!

        # Excise empty elements
        reparsed = Nokogiri::HTML.fragment(html)
        excise_empty = lambda do |node|
          if node.respond_to?(:name) && node.name == "script"
            node.unlink
          else
            node.children.each do |node|
              excise_empty.call(node) if node.element?
            end
            if node.respond_to?(:attribute_nodes) && node.respond_to?(:text)
              if node.attribute_nodes.size == 0 && node.text.to_s.strip =~ /^\s*$/ &&
                  node.children.all? { |child| child.text? }
                node.unlink
              end
            end
          end
        end
        excise_empty.call(reparsed)
        html = reparsed.to_s
      end
      html
    end
  end)
end

#article_node ⇒ Object 



297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
 
# File 'lib/href_preview/preview.rb', line 297

def article_node
  @article_node ||= (if is_html?
    begin
      nodes = dom.xpath("/html[@itemtype='http://schema.org/NewsArticle']//article[@id='story']")
      nodes.first if nodes.size == 1
    end or
    begin
      nodes = dom.xpath("//*/*[@itemtype='http://schema.org/NewsArticle']")
      nodes.first if nodes.size == 1
    end or
    begin
      nodes = dom.xpath("//*/*[@itemprop='articleBody']")
      nodes.first if nodes.size == 1
    end or
    begin
      nodes = dom.css("article div.article-entry")
      nodes.first if nodes.size == 1
    end or
    begin
      nodes = dom.css("article.post div.entry-content")
      nodes.first if nodes.size == 1
    end or
    begin
      nodes = dom.css("div.post div.postBody")
      nodes.first if nodes.size == 1
    end or
    begin
      nodes = dom.css(".pg_story div#leftcolumn div.body")
      nodes.first if nodes.size == 1
    end
  end)
end

#article_text ⇒ Object 



392
393
394
 
# File 'lib/href_preview/preview.rb', line 392

def article_text
  @article_text ||= is_html? ? Sanitize.clean(article_html) : nil
end

#canonical_uri ⇒ Object 



203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
 
# File 'lib/href_preview/preview.rb', line 203

def canonical_uri
  @canonical_uri ||= (if is_html?
    begin
      node = dom.xpath("//*/link[@rel='canonical']/@href").first
      Addressable::URI.parse(node.value) if node && node.value && node.value != ''
    end or
    begin
      node = dom.xpath("//*/meta[@property='og:url']/@content").first
      Addressable::URI.parse(node.value) if node && node.value && node.value != ''
    end or
    Addressable::URI.parse(response.env.url.to_s)
  else
    Addressable::URI.parse(response.env.url.to_s)
  end)
end

#charset ⇒ String

Returns the charset declared in the HTTP headers or HTML meta tags.



59
60
61
62
63
64
65
66
67
68
69
70
71
72
 
# File 'lib/href_preview/preview.rb', line 59

def charset
  @charset ||= (begin
    charset = response.headers['Content-Type'].to_s[/;\s*charset=([^;,]*)/, 1] or
    begin
      node = dom.xpath("//*/meta[@http-equiv='Content-Type']/@content").first
      node.value.to_s[/;\s*charset=([^;,]*)/, 1] if node
    end or
    begin
      node = dom.xpath("//*/meta/@charset").first
      node.value if node
    end
    charset.strip if charset
  end)
end

#description ⇒ Object 



162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
 
# File 'lib/href_preview/preview.rb', line 162

def description
  @description ||= (begin
    if is_html?
      description = begin
        node = dom.xpath("//*/meta[@property='og:description']/@content").first
        node.value if node
      end or
      begin
        node = dom.xpath("//*/meta[@name='dc.description']/@content").first
        node.value if node
      end or
      begin
        node = dom.xpath("//*/meta[@itemprop='description']/@content").first
        node.value if node
      end or
      begin
        node = dom.xpath("//*/meta[@name='description']/@content").first
        node.value if node
      end or
      begin
        node = dom.xpath("//*/meta[@name='dcterms.abstract']/@content").first
        node.value if node
      end or
      begin
        # Unlikely to ever happen
        node = dom.xpath("//*/meta[@name='twitter:description']/@content").first
        node.value if node
      end or
      begin
        # Unlikely to ever happen
        node = dom.xpath("//*/meta[@name='sailthru.description']/@content").first
        node.value if node
      end
      if description
        description.gsub!(/&nbsp;/, ' ')
        description.strip
      end
    end
  end)
end

#dom ⇒ Nokogiri::HTML::Document

The DOM for the response body.



111
112
113
 
# File 'lib/href_preview/preview.rb', line 111

def dom
  @dom ||= Nokogiri::HTML(response.body)
end

#image_uri ⇒ Object 



232
233
234
 
# File 'lib/href_preview/preview.rb', line 232

def image_uri
  @image_uri ||= (images.first ? Addressable::URI.parse(images.first.uri) : nil)
end

#images ⇒ Object 



236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
 
# File 'lib/href_preview/preview.rb', line 236

def images
  @images ||= (begin
    image_uris = []
    if is_html?
      nodes = dom.xpath("//*/meta[@property='og:image']/@content")
      nodes.each do |node|
        if node && node.value && node.value != ''
          image_uris << Addressable::URI.parse(node.value)
        end
      end
      if article_node
        nodes = article_node.xpath("meta[@itemprop='thumbnailurl']/@content")
        nodes.each do |node|
          if node && node.value && node.value != ''
            image_uris << Addressable::URI.parse(node.value)
          end
        end
      end
    elsif mime_type && mime_type.media_type == 'image'
      image_uris << canonical_uri
    end
    image_uris.uniq.map { |uri| FastImage.new(uri, :timeout => 0.5) }
  end)
end

#inspect ⇒ Object 



446
447
448
449
 
# File 'lib/href_preview/preview.rb', line 446

def inspect
  addr = '0x' + ('%x' % (object_id << 1)).rjust(14, '0')
  "#<HRefPreview::Preview:#{addr} TITLE=#{title.inspect}>"
end

#is_html? ⇒ true, false

Returns true if the response had a 2xx HTTP code and the mime type is either HTML or XHTML.



99
100
101
102
103
104
 
# File 'lib/href_preview/preview.rb', line 99

def is_html?
  return (
    response.status >= 200 && response.status < 300 &&
    mime_type && mime_type.sub_type =~ /^x?html/
  )
end

#item_type ⇒ Object 



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

def item_type
  @item_type ||= (if is_html?
    begin
      node = dom.xpath("//*/meta[@property='og:type']/@content").first
      node.value if node
    end or
    if dom.xpath("//*[@itemtype='http://schema.org/NewsArticle']").first != nil
      'article'
    end
  end)
end

#language ⇒ Object 



76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
 
# File 'lib/href_preview/preview.rb', line 76

def language
  @language ||= (begin
    language = response.headers['Content-Language'] or
    begin
      node = dom.xpath("//*/meta[@http-equiv='Content-Language']/@content").first
      node.value if node
    end or
    begin
      node = dom.xpath("//*/meta[@name='dc.language']/@content").first
      node.value if node
    end
    if language
      # Strip the irrelevant '-US' from 'en-US' if it appears.
      language[/^([a-z]{2})/, 1].to_s.downcase
    end
  end)
end

#mime_type ⇒ MIME::Type

Returns the MIME type declared in the HTTP headers or HTML meta tags.



40
41
42
43
44
45
46
47
48
49
50
51
52
 
# File 'lib/href_preview/preview.rb', line 40

def mime_type
  @mime_type ||= (begin
    MIME::Types[response.headers['Content-Type']].first or
    begin
      node = dom.xpath("//*/meta[@http-equiv='Content-Type']/@content").first
      MIME::Types[node.value].first if node && node.value
    end or
    begin
      node = dom.xpath("//*/meta[@name='dc.format']/@content").first
      MIME::Types[node.value].first if node && node.value
    end
  end)
end

#published ⇒ Object 



396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
 
# File 'lib/href_preview/preview.rb', line 396

def published
  @published ||= (begin
    # Check under the article node first, otherwise search all
    begin
      node = dom.xpath("//*/meta[@property='article:published_time']/@content").first
      Time.parse(node.value) if node && node.value && node.value != ''
    end or
    if article_node
      node = article_node.xpath("meta[@itemprop='datepublished']/@content").first
      Time.parse(node.value) if node && node.value && node.value != ''
    end or
    begin
      node = dom.xpath("//*/meta[@itemprop='datepublished']/@content").first
      Time.parse(node.value) if node && node.value && node.value != ''
    end or
    begin
      node = dom.xpath("//*/meta[@name='dcterms.created']/@content").first
      Time.parse(node.value) if node && node.value && node.value != ''
    end or
    begin
      # Only a date, not a time, and not particularly specific,
      # so this is a fallback at best.
      node = dom.xpath("//*/meta[@name='dc.date']/@content").first
      Time.parse(node.value) if node && node.value && node.value != ''
    end
  end)
end

#shortlink_uri ⇒ Object 



219
220
221
222
223
224
225
226
227
228
229
230
 
# File 'lib/href_preview/preview.rb', line 219

def shortlink_uri
  @shortlink_uri ||= (if is_html?
    begin
      node = dom.xpath("//*/link[@rel='shortlink']/@href").first
      Addressable::URI.parse(node.value) if node && node.value && node.value != ''
    end or
    begin
      node = dom.xpath("//*[@class='story-short-url']/a/@href").first
      Addressable::URI.parse(node.value) if node && node.value && node.value != ''
    end
  end)
end

#site_name ⇒ Object 



273
274
275
276
277
278
279
280
281
282
283
284
 
# File 'lib/href_preview/preview.rb', line 273

def site_name
  @site_name ||= (if is_html?
    begin
      node = dom.xpath("//*/meta[@property='og:site_name']/@content").first
      node.value if node
    end or
    begin
      node = dom.xpath("//*/meta[@name='dc.publisher']/@content").first
      node.value if node
    end
  end)
end

#title ⇒ String 



117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
 
# File 'lib/href_preview/preview.rb', line 117

def title
  @title ||= (begin
    if is_html?
      title = begin
        node = dom.xpath("//*/meta[@property='og:title']/@content").first
        node.value if node
      end or
      begin
        node = dom.xpath("//*/meta[@name='dc.title']/@content").first
        node.value if node
      end or
      begin
        if article_node
          node = article_node.xpath("*[@itemprop='headline']").first
          node.text if node
        end
      end or
      begin
        node = dom.xpath("//*/*[(self::h1 or self::h2) and @itemprop='headline']").first
        node.text if node
      end or
      begin
        node = dom.xpath("//*/head/title").first
        node.text if node
      end or
      begin
        # Unlikely to ever happen
        node = dom.xpath("//*/meta[@name='twitter:title']/@content").first
        node.value if node
      end or
      begin
        # Unlikely to ever happen
        node = dom.xpath("//*/meta[@name='sailthru.title']/@content").first
        node.value if node
      end
      if title
        title.gsub!(/&nbsp;/, ' ')
        title.gsub!(/^#{site_name}[\s\|\-\:]*/, '')
        title.gsub!(/[\s\|\-\:]*#{site_name}$/, '')
        title.strip
      end
    end
  end)
end

#twitter ⇒ String 



288
289
290
291
292
293
294
295
 
# File 'lib/href_preview/preview.rb', line 288

def twitter
  @twitter ||= (if is_html?
    begin
      node = dom.xpath("//*/meta[@name='twitter:site']/@content").first
      node.value if node && node.value && node.value =~ /^@/
    end
  end)
end

#updated ⇒ Object 



424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
 
# File 'lib/href_preview/preview.rb', line 424

def updated
  @updated ||= (begin
    # Check under the article node first, otherwise search all
    begin
      node = dom.xpath("//*/meta[@property='article:modified_time']/@content").first
      Time.parse(node.value) if node && node.value && node.value != ''
    end or
    if article_node
      node = article_node.xpath("meta[@itemprop='datemodified']/@content").first
      Time.parse(node.value) if node && node.value && node.value != ''
    end or
    begin
      node = dom.xpath("meta[@itemprop='datemodified']/@content").first
      Time.parse(node.value) if node && node.value && node.value != ''
    end or
    begin
      node = dom.xpath("//*/meta[@name='dcterms.modified']/@content").first
      Time.parse(node.value) if node && node.value && node.value != ''
    end
  end)
end