Class: Html2rss::HtmlExtractor::Extractors::Pdf

Inherits:
Object
  • Object
show all
Defined in:
lib/html2rss/html_extractor/enclosure_extractor.rb

Overview

Extracts PDF enclosures from HTML tags.

Class Method Summary collapse

Class Method Details

.call(article_tag, base_url:) ⇒ Object 



54
55
56
57
58
59
60
61
62
63
64
65
 
# File 'lib/html2rss/html_extractor/enclosure_extractor.rb', line 54

def self.call(article_tag, base_url:)
  article_tag.css('a[href$=".pdf"]').filter_map do |link|
    href = link['href'].to_s
    next if href.empty?

    abs_url = Url.from_relative(href, base_url)
    {
      url: abs_url,
      type: RssBuilder::Enclosure.guess_content_type_from_url(abs_url)
    }
  end
end