about summary refs log tree commit diff
path: root/app/lib/link_details_extractor.rb
blob: f8a0be636ee36e456c5c67a70ce77e02d00c703e (plain) (blame)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
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
71
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
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
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
161
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
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
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
262
263
264
265
266
267
268
269
270
271
272
273
274
275
# frozen_string_literal: true

class LinkDetailsExtractor
  include ActionView::Helpers::TagHelper
  include LanguagesHelper

  # Some publications wrap their JSON-LD data in their <script> tags
  # in commented-out CDATA blocks, they need to be removed before
  # attempting to parse JSON
  CDATA_JUNK_PATTERN = %r{^[\s]*(
    (/\*[\s]*<!\[CDATA\[[\s]*\*/) # Block comment style opening
    |
    (//[\s]*<!\[CDATA\[) # Single-line comment style opening
    |
    (/\*[\s]*\]\]>[\s]*\*/) # Block comment style closing
    |
    (//[\s]*\]\]>) # Single-line comment style closing
  )[\s]*$}x

  class StructuredData
    SUPPORTED_TYPES = %w(
      NewsArticle
      WebPage
    ).freeze

    def initialize(data)
      @data = data
    end

    def headline
      json['headline']
    end

    def description
      json['description']
    end

    def language
      json['inLanguage']
    end

    def type
      json['@type']
    end

    def image
      obj = first_of_value(json['image'])

      return obj['url'] if obj.is_a?(Hash)

      obj
    end

    def date_published
      json['datePublished']
    end

    def date_modified
      json['dateModified']
    end

    def author_name
      author['name']
    end

    def author_url
      author['url']
    end

    def publisher_name
      publisher['name']
    end

    def publisher_logo
      publisher.dig('logo', 'url')
    end

    def valid?
      json.present?
    end

    private

    def author
      first_of_value(json['author']) || {}
    end

    def publisher
      first_of_value(json['publisher']) || {}
    end

    def first_of_value(arr)
      arr.is_a?(Array) ? arr.first : arr
    end

    def root_array(root)
      root.is_a?(Array) ? root : [root]
    end

    def json
      @json ||= root_array(Oj.load(@data)).find { |obj| SUPPORTED_TYPES.include?(obj['@type']) } || {}
    end
  end

  def initialize(original_url, html, html_charset)
    @original_url = Addressable::URI.parse(original_url)
    @html         = html
    @html_charset = html_charset
  end

  def to_preview_card_attributes
    {
      title: title || '',
      description: description || '',
      image_remote_url: image,
      type: type,
      link_type: link_type,
      width: width || 0,
      height: height || 0,
      html: html || '',
      provider_name: provider_name || '',
      provider_url: provider_url || '',
      author_name: author_name || '',
      author_url: author_url || '',
      embed_url: embed_url || '',
      language: language,
    }
  end

  def type
    player_url.present? ? :video : :link
  end

  def link_type
    if structured_data&.type == 'NewsArticle' || opengraph_tag('og:type') == 'article'
      :article
    else
      :unknown
    end
  end

  def html
    player_url.present? ? content_tag(:iframe, nil, src: player_url, width: width, height: height, allowtransparency: 'true', scrolling: 'no', frameborder: '0') : nil
  end

  def width
    opengraph_tag('twitter:player:width')
  end

  def height
    opengraph_tag('twitter:player:height')
  end

  def title
    html_entities.decode(structured_data&.headline || opengraph_tag('og:title') || document.xpath('//title').map(&:content).first)
  end

  def description
    html_entities.decode(structured_data&.description || opengraph_tag('og:description') || meta_tag('description'))
  end

  def image
    valid_url_or_nil(opengraph_tag('og:image'))
  end

  def canonical_url
    valid_url_or_nil(link_tag('canonical') || opengraph_tag('og:url'), same_origin_only: true) || @original_url.to_s
  end

  def provider_name
    html_entities.decode(structured_data&.publisher_name || opengraph_tag('og:site_name'))
  end

  def provider_url
    valid_url_or_nil(host_to_url(opengraph_tag('og:site')))
  end

  def author_name
    html_entities.decode(structured_data&.author_name || opengraph_tag('og:author') || opengraph_tag('og:author:username'))
  end

  def author_url
    structured_data&.author_url
  end

  def embed_url
    valid_url_or_nil(opengraph_tag('twitter:player:stream'))
  end

  def language
    valid_locale_or_nil(structured_data&.language || opengraph_tag('og:locale') || document.xpath('//html').pick('lang'))
  end

  def icon
    valid_url_or_nil(structured_data&.publisher_icon || link_tag('apple-touch-icon') || link_tag('shortcut icon'))
  end

  private

  def player_url
    valid_url_or_nil(opengraph_tag('twitter:player'))
  end

  def host_to_url(str)
    return if str.blank?

    str.start_with?(/https?:\/\//) ? str : "http://#{str}"
  end

  def valid_url_or_nil(str, same_origin_only: false)
    return if str.blank? || str == 'null'

    url = @original_url + Addressable::URI.parse(str)

    return if url.host.blank? || !%w(http https).include?(url.scheme) || (same_origin_only && url.host != @original_url.host)

    url.to_s
  rescue Addressable::URI::InvalidURIError
    nil
  end

  def link_tag(name)
    document.xpath("//link[@rel=\"#{name}\"]").pick('href')
  end

  def opengraph_tag(name)
    document.xpath("//meta[@property=\"#{name}\" or @name=\"#{name}\"]").pick('content')
  end

  def meta_tag(name)
    document.xpath("//meta[@name=\"#{name}\"]").pick('content')
  end

  def structured_data
    # Some publications have more than one JSON-LD definition on the page,
    # and some of those definitions aren't valid JSON either, so we have
    # to loop through here until we find something that is the right type
    # and doesn't break
    @structured_data ||= document.xpath('//script[@type="application/ld+json"]').filter_map do |element|
      json_ld = element.content&.gsub(CDATA_JUNK_PATTERN, '')

      next if json_ld.blank?

      structured_data = StructuredData.new(html_entities.decode(json_ld))

      next unless structured_data.valid?

      structured_data
    rescue Oj::ParseError, EncodingError
      Rails.logger.debug { "Invalid JSON-LD in #{@original_url}" }
      next
    end.first
  end

  def document
    @document ||= Nokogiri::HTML(@html, nil, encoding)
  end

  def encoding
    @encoding ||= begin
      guess = detector.detect(@html, @html_charset)
      guess&.fetch(:confidence, 0).to_i > 60 ? guess&.fetch(:encoding, nil) : nil
    end
  end

  def detector
    @detector ||= CharlockHolmes::EncodingDetector.new.tap do |detector|
      detector.strip_tags = true
    end
  end

  def html_entities
    @html_entities ||= HTMLEntities.new
  end
end