Class: Html2rss::AutoSource::Scraper::SemanticHtml::Image

Inherits:
Object
  • Object
show all
Defined in:
lib/html2rss/auto_source/scraper/semantic_html/image.rb

Overview

Image is responsible for extracting image URLs the article_tag.

Class Method Summary collapse

Class Method Details

.call(article_tag, url:) ⇒ Object 



10
11
12
13
14
15
16
 
# File 'lib/html2rss/auto_source/scraper/semantic_html/image.rb', line 10

def self.call(article_tag, url:)
  img_src = from_source(article_tag) ||
            from_img(article_tag) ||
            from_style(article_tag)

  Utils.build_absolute_url_from_relative(img_src, url) if img_src
end

.from_img(article_tag) ⇒ Object 



18
19
20
 
# File 'lib/html2rss/auto_source/scraper/semantic_html/image.rb', line 18

def self.from_img(article_tag)
  article_tag.at_css('img[src]:not([src^="data"])')&.[]('src')
end

.from_source(article_tag) ⇒ Object

Extracts the largest image source from the srcset attribute of an img tag or a source tag inside a picture tag.

See Also:



29
30
31
32
33
34
35
36
37
38
39
40
41
42
 
# File 'lib/html2rss/auto_source/scraper/semantic_html/image.rb', line 29

def self.from_source(article_tag) # rubocop:disable Metrics/AbcSize
  hash = article_tag.css('img[srcset], picture > source[srcset]')
                    .flat_map { |source| source['srcset'].to_s.split(',') }
                    .filter_map do |line|
    width, url = line.split.reverse
    next if url.nil? || url.start_with?('data:')

    width_value = width.to_i.zero? ? 0 : width.scan(/\d+/).first.to_i

    [width_value, url.strip]
  end.to_h

  hash[hash.keys.max]
end

.from_style(article_tag) ⇒ Object 



44
45
46
47
48
49
 
# File 'lib/html2rss/auto_source/scraper/semantic_html/image.rb', line 44

def self.from_style(article_tag)
  article_tag.css('[style*="url"]')
             .map { |tag| tag['style'][/url\(['"]?(.*?)['"]?\)/, 1] }
             .reject { |src| !src || src.start_with?('data:') }
             .max_by(&:size)
end