Module: Artext

Defined in:
lib/artext.rb,
lib/artext/version.rb

Constant Summary collapse

USER_AGENT = 
"facebookexternalhit/1.1 (+http://www.facebook.com/externalhit_uatext.php)"
VERSION = 
"0.0.4"

Class Method Summary collapse

Class Method Details

.extract(url) ⇒ Object 



13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
 
# File 'lib/artext.rb', line 13

def self.extract(url)
  url = (url =~ /^(http|https):\/\/(.)*/i) ? url : "http://#{url}"
  return {:url => url, :data => [], :article => []} if ((url =~ /^(http|https):\/\/[a-z0-9]+([\-\.]{1}[a-z0-9]+)*\.[a-z]{2,5}(:[0-9]{1,5})?(\/.*)?$/ix).nil?)
  begin
    res = HTTParty.get(url, headers: {"User-Agent" => USER_AGENT})
    raise Exception.new("Unable to crawl URL") if res.code != 200
    doc = Nokogiri::HTML(res)
  rescue Exception => e
    doc = Nokogiri::HTML(open(url, "User-Agent" => USER_AGENT))
  end
  data = get_data_from_url(doc, url)
  article = get_article_from_url(doc, url, data[:type])
  data[:type] = "image" if (article[:score] == 1)
  response = {:url => url, :data => [data], :article => [article]}
end

.extractp(element, score) ⇒ Object 



286
287
288
289
290
291
292
293
294
295
 
# File 'lib/artext.rb', line 286

def self.extractp(element, score)
  p_elem = nil
  imgs = []
  as = element.search("a")
  if (!is_blank?(as) && element.text == as.text && score < 0.8)
    return nil
  end
  p_elem, imgs = phandle(element, "", []) if (!is_blank?(element))
  return p_elem, imgs
end

.figurehandle(element, html, images) ⇒ Object 



327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
 
# File 'lib/artext.rb', line 327

def self.figurehandle(element, html, images)
  if (element.name == "img" || (!is_blank?(element.attribute("class")) && element.attribute("class").value.include?("js-delayed-image-load")))
    img = get_valid_image(element)
    if (!is_blank?(img))
      html = html + "<img src=\"#{img}\">"
      images << img
    end
  elsif (is_blank?(element.children))
    return html, images
  else
    element.children.each do |elem|
      html, images = figurehandle(elem, html, images)
    end
  end
  return html, images
end

.find_article(doc) ⇒ Object 



361
362
363
364
365
366
 
# File 'lib/artext.rb', line 361

def self.find_article(doc)
  article = doc.xpath('//*[@*="articleBody"]')
  article = doc.xpath('//*[contains(@class, "article")]') if (is_blank?(article) || article.text.split.join(" ").length < 450)
  article = doc.xpath('//*[contains(@class, "body")]') if (is_blank?(article) || article.text.split.join(" ").length < 450)
  return article
end

.find_relevant(article, score) ⇒ Object 



152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
 
# File 'lib/artext.rb', line 152

def self.find_relevant(article, score)
  p_elems = article.search("p")
  total_p = p_elems.count
  rel = article
  max_p = 0
  last_p = 0
  if (!is_blank?(p_elems))
    while (p_elems[0].text != article.text || p_elems[0].attribute("class") != article.attribute("class"))
      parent_elems = p_elems.map {|p| p.parent()}
      parent_elems.each do |ps|
        if (ps.search("p").count > max_p)
          max_p = ps.search("p").count
          rel = ps
        end
      end
      if ((last_p > 5 && last_p == max_p) || total_p - max_p < 2)
        score = 0.95 if (score < 1)
        break
      end
      last_p = max_p
      p_elems = parent_elems
    end
  end
  return rel, score
end

.get_article_from_url(doc, url, type) ⇒ Object 



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
 
# File 'lib/artext.rb', line 117

def self.get_article_from_url(doc, url, type)
  dates = get_author_and_date(doc)
  article = doc.search("//article")
  score = 0.9
  article = [] if (article.count == 1 && article.text.split.join(" ").length < 500)
  if (article.count > 1)
    article = get_correct_article(article)
    score = 0.8
  end
  if (is_blank?(article))
    article = find_article(doc)
    score = 0.6
  end
  if (is_blank?(article))
    # image url
    begin
      html = doc.to_html
      raise Exception.new("Image URL") if is_blank?(html)
      return {:body => "", :text => "", :images => [], :score => 0}
    rescue Exception => e
      return {:body => "<figure><img src=\"#{url}\"></figure>", :text => "", :images => [url], :score => 1}
    end
  else
    article = remove_unwanted_items_from(article)
    article, score = find_relevant(article, score)
    score = score - 0.5 if (type != "article")
    if (score > 0.9)
      html, imgs = iteratively_clean(article, "", [], score)
    else
      html, imgs = recursively_clean(article, "", [], score)
    end
    response = {:body => html, :text => article.text.split.join(" "), :images => imgs, :score => score}.merge(dates)
  end
end

.get_author_and_date(doc) ⇒ Object 



72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
 
# File 'lib/artext.rb', line 72

def self.get_author_and_date(doc)
  authors = []
  possible_authors = doc.xpath('//meta[@property="author" or @name="author"]')
  possible_authors.each {|a| authors << a["content"] if (!is_blank?(a["content"]))}

  date = nil
  possible_dates = doc.xpath('//meta[contains(@property, "date") or contains(@name, "date")]')
  if (!is_blank?(possible_dates))
    possible_dates.each do |po|
      if (po["content"][0..3].to_i > 2000)
        date = po["content"] if (!is_blank?(po["content"]))
        break
      end
    end
  end
  if (date.nil?)
    possible_dates = doc.xpath('//*[contains(@datetime, "2015")]')
    date = possible_dates.first.attribute("datetime").value if (!is_blank?(possible_dates))
  end
  if (!date.nil?)
    datetime = date.split.join(" ")[0..9].split("-")
    t = Time.new(datetime[0], datetime[1], datetime[2])
    date = "#{Date::MONTHNAMES[t.month]} #{t.day}, #{t.year}"
  end
  return {:date => date, :authors => authors}
end

.get_best_image(images) ⇒ Object 



99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
 
# File 'lib/artext.rb', line 99

def self.get_best_image(images)
  return nil if (is_blank?(images))
  return images[0] if (images.size == 1)
  # reject logo or similar images
  refined_images = images.reject{|i| i =~ /logo|fallback/i}
  return refined_images[0] if (refined_images.size == 1)
  refined_images = images if is_blank?(refined_images)
  dimensions = []
  refined_images.each do |i|
    type = FastImage.type(Addressable::URI.escape(i))
    size = FastImage.size(Addressable::URI.escape(i))
    return i if((type == :gif) && (size && size[0] > 299 && size[1] > 199))
    dimensions << {:x => size[0], :y => size[1], :image => i} if !size.nil?
  end
  image = is_blank?(dimensions) ? nil : dimensions.max_by{|d| d[:x]}[:image]
  return image
end

.get_correct_article(articles) ⇒ Object 



368
369
370
371
372
373
374
375
 
# File 'lib/artext.rb', line 368

def self.get_correct_article(articles)
  articles.each do |article|
    if (article.text.split.join(" ").length > 200)
      return article
    end
  end
  return nil
end

.get_data_from_url(doc, url) ⇒ Object 



29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
 
# File 'lib/artext.rb', line 29

def self.get_data_from_url(doc, url)
  og_image = doc.search("//meta[@property='og:image' or @name='og:image']")
  og_images = []
  if !is_blank?(og_image)
    og_image.each do |ogi|
      if !is_blank?(ogi["content"])
        image = ogi["content"]
        if (image =~ /^\/\/(.)*/)
          uri = URI.parse(url)
          image = "#{uri.scheme}:#{image}"
        elsif (image =~ /^\/(.)*/)
          uri = URI.parse(url)
          image = File.join("#{uri.scheme}://#{uri.host}", image)
        end
        og_images << image
      end
    end
  end
  # Try to get the best image based on heuristics
  image = get_best_image(og_images)

  og_title = doc.search("//meta[@property='og:title' or @name='og:title']")
  if (!is_blank?(og_title) && !is_blank?(og_title[0]["content"]))
    clip_title = og_title[0]["content"]
  else
    page_title = doc.search("//title")[0]
    clip_title = page_title.text if !is_blank?(page_title)
  end

  tags = []
  possible_tags = doc.xpath('//meta[contains(@name, "tag") or contains(@name, "keyword") or contains(@property, "tag") or contains(@property, "keyword")]')
  possible_tags.each{|e| tags << e["content"].split(',') if (!e["content"].nil?)}
  tags = tags.flatten.map(&:strip).uniq!

  type = doc.search("//meta[@property='og:type' or @name='og:type']")
  type = is_blank?(type) ? nil : type[0]["content"]

  favicon = "http://www.google.com/s2/favicons?domain_url=#{url}"
  theme = get_dominant_color(favicon)

  res = {:image => image, :title => clip_title, :tags => tags, :type => type, :favicon => favicon, :theme => theme}
end

.get_dominant_color(url) ⇒ Object 



389
390
391
392
393
 
# File 'lib/artext.rb', line 389

def self.get_dominant_color(url)
  image = MiniMagick::Image.open(url)
  color = image.run_command("convert", image.path, "-format", "%c\n",  "-colors", 1, "-depth", 8, "histogram:info:").split(' ')
  return color[Hash[color.map{|h| h =~ /^#/}.map.with_index.to_a][0]][0..6]
end

.get_element_html(element, images, score) ⇒ Object 



217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
 
# File 'lib/artext.rb', line 217

def self.get_element_html(element, images, score)
  if (element.name == "a")
    # Eat it
  elsif (element.name == "img")
    img = get_valid_image(element)
    if (!is_blank?(img))
      tv = "<figure><img src=\"#{img}\"></figure>"
      images = images + [img]
    end
  elsif (element.name == "h1" || element.name == "h2" || element.name == "h3" || element.name == "h4")
    tv = "<h2>#{element.text.split.join(" ")}</h2>" if (!is_blank?(element.text.split.join(" ")))
  elsif (element.name == "p")
    p_elem, ti = extractp(element, score)
    tv = "<p>#{p_elem}</p>" if (!is_blank?(p_elem))
    tv = tv.gsub("<p></p>", "")
    images = images + ti if (!is_blank?(ti))
  elsif (element.name == "figure")
    cap = element.search("figcaption").inner_html
    cap = is_blank?(cap) ? "" : "<figcaption>#{cap}</figcaption>"
    tv, ti = figurehandle(element, "", [])
    tv = "<figure>#{tv}#{cap}</figure>" if (!is_blank?(tv))
    images = images + ti
  elsif (element.name == "text")
    tv = element.text.split.join(" ")
    tv = nil if tv == "advertisement"
    tv = "<p class\"inline\">#{tv}</p>" if (!is_blank?(tv))
  elsif (element.name == "i")
    tv = element.text.split.join(" ")
    tv = "<i>#{tv}</i>" if (!is_blank?(tv))
  elsif (element.name == "ol" || element.name == "ul")
    tv, ti = listhandle(element)
    images = images + ti
  elsif (element.name == "div" || element.name == "span")
    html = ""
    imgs = []
    element.children.each do |elem|
      tv, ti = get_element_html(elem, [], score)
      html = html + tv if (!is_blank?(tv))
      imgs = imgs + ti if (!is_blank?(ti))
    end
    tv = html
    images = imgs
  end
  return tv, images
end

.get_valid_image(element) ⇒ Object 



263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
 
# File 'lib/artext.rb', line 263

def self.get_valid_image(element)
  if (!is_blank?(element))
    tsrc1 = nil
    search_in = ["data-image", "data-original", "srcset", "data-src", "datasrc", "rel:bf_image_src", "src"]
    search_in.each do |search|
      tsrc = element.attribute(search)
      if (!is_blank?(tsrc))
        tsrc1 = tsrc.value
        tsrc1 = tsrc1.split(",").first.split(" ")[0] if (search == "srcset")
        break
      end
    end
    if (!is_blank?(tsrc1))
      tsrc1 = "http:" + tsrc1 if (tsrc1[0..1] == "//")
      tv_size = FastImage.size(Addressable::URI.escape(tsrc1))
      if (!tv_size.nil? && (tv_size[0] > 100 || tv_size[1] > 100))
        return tsrc1
      end
    end
  end
  return ""
end

.is_blank?(value) ⇒ Boolean

Returns:

  • (Boolean) 


395
396
397
398
399
400
401
 
# File 'lib/artext.rb', line 395

def self.is_blank?(value)
  if (value.class == Nokogiri::XML::Element || value.class == Nokogiri::XML::Attr)
    return (value.nil? || value.blank?)
  else
    return (value.nil? || value.empty?)
  end
end

.iteratively_clean(element, html, images, score) ⇒ Object 



178
179
180
181
182
183
184
185
186
187
188
 
# File 'lib/artext.rb', line 178

def self.iteratively_clean(element, html, images, score)
  html = ""
  imgs = []
  element.children.each do |elem|
    tv, ti = get_element_html(elem, [], score)
    html = html + tv if (!is_blank?(tv))
    imgs = imgs + ti if (!is_blank?(ti))
  end
  imgs.uniq!
  return html, imgs
end

.listhandle(element) ⇒ Object 



344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
 
# File 'lib/artext.rb', line 344

def self.listhandle(element)
  html = ""
  imgs = []
  li_elems = element.search("li")
  li_elems.each do |elem|
    tv, ti = recursively_clean(elem, "", [], 0.95)
    html = html + "<li>#{tv}</li>" if (!is_blank?(tv))
    imgs = imgs + ti if (!is_blank?(ti))
  end
  if (element.name == "ul")
    html = "<ul>#{html}</ul>" if (!is_blank?(html))
  elsif (element.name == "ol")
    html = "<ol>#{html}</ol>" if (!is_blank?(html))
  end
  return html, imgs
end

.phandle(element, html, images) ⇒ 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
 
# File 'lib/artext.rb', line 297

def self.phandle(element, html, images)
  if (!is_blank?(element.children) && !((element.name == "a" || element.name == "figure") && is_blank?(element.search("img"))))
    element.children.each do |elem|
      html, images = phandle(elem, html, images)
    end
  end
  if (element.name == "img")
    img = get_valid_image(element)
    if (!is_blank?(img))
      html = "</p><figure><img src=\"#{img}\"></figure><p>"
      images << img
    end
  elsif (element.name == "figure")
    cap = element.search("figcaption").inner_html
    cap = is_blank?(cap) ? "" : "<figcaption>#{cap}</figcaption>"
    tv, ti = figurehandle(element, "", [])
    html = "</p><figure>#{tv}#{cap}</figure><p>" if (!is_blank?(tv))
    images << ti if (!is_blank?(ti))
  elsif (element.name == "a")
    html = html + " <a href=\"#{element.attribute("href").value if (!is_blank?(element.attribute("href")))}\">#{element.text.split.join(" ")}</a> "
  elsif (element.name == "text")
    html = html + element.text.split.join(" ")
  elsif (element.name == "br")
    html = html + "<br>"
  elsif (element.name == "p" && is_blank?(html))
    html = element.text.split.join(" ")
  end
  return html, images
end

.recursively_clean(element, html, images, score) ⇒ Object 



190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
 
# File 'lib/artext.rb', line 190

def self.recursively_clean(element, html, images, score)
  allowable = ["p", "figure", "a", "h1", "h2", "h3", "h4", "text"]
  if (is_blank?(element.children) || (element.class != Nokogiri::XML::NodeSet && (allowable.include?(element.name))))
    tv, ti = get_element_html(element, [], score)
    html = html + tv if (!is_blank?(tv))
    images = images + ti
    images.uniq!
  else
    if (element.class != Nokogiri::XML::NodeSet && (element.name == "header" || element.name == "footer"))
      # Eat it
    elsif (element.class != Nokogiri::XML::NodeSet && element.name == "div" && !is_blank?(element.attribute("class")) && element.attribute("class").value.downcase.include?("meta") && element.text.split.join(" ").length < 300)
      #Eat it
    elsif (element.class != Nokogiri::XML::NodeSet && element.name == "section" && score > 0.9)
      #Eat it
    elsif (element.class != Nokogiri::XML::NodeSet && element.name == "ul" && score > 0.9)
      #Eat it
    elsif (element.class != Nokogiri::XML::NodeSet && element.name == "ol" && score > 0.9)
      #Eat it
    else
      element.children.each do |art|
        html, images = recursively_clean(art, html, images, score)
      end
    end
  end
  return html, images
end

.remove_unwanted_items_from(article) ⇒ Object 



377
378
379
380
381
382
383
384
385
386
387
 
# File 'lib/artext.rb', line 377

def self.remove_unwanted_items_from(article)
  unwanted_elements = ["//script", "//comment()", "//aside", ".aside", "iframe", "//noscript", "//form", "//header", "//footer"]
  unwanted_elements.each do |elem|
    article.search("#{elem}").remove
  end
  removable_elements = ["comment", "social", "advertisement", "share"]
  removable_elements.each do |rem|
    article.xpath("//*[contains(@*, '#{rem}')]").remove
  end
  return article
end