lib/onebox/helpers.rb in onebox-2.2.16 vs lib/onebox/helpers.rb in onebox-2.2.17
- old
+ new
@@ -24,26 +24,10 @@
html.gsub(/<[^>]+>/, ' ').gsub(/\n/, '')
end
def self.fetch_html_doc(url, headers = nil, body_cacher = nil)
response = (fetch_response(url, headers: headers, body_cacher: body_cacher) rescue nil)
- doc = Nokogiri::HTML(response)
- uri = Addressable::URI.parse(url)
-
- ignore_canonical_tag = doc.at('meta[property="og:ignore_canonical"]')
- should_ignore_canonical = IGNORE_CANONICAL_DOMAINS.map { |hostname| uri.hostname.match?(hostname) }.any?
-
- unless (ignore_canonical_tag && ignore_canonical_tag['content'].to_s == 'true') || should_ignore_canonical
- # prefer canonical link
- canonical_link = doc.at('//link[@rel="canonical"]/@href')
- canonical_uri = Addressable::URI.parse(canonical_link)
- if canonical_link && "#{canonical_uri.host}#{canonical_uri.path}" != "#{uri.host}#{uri.path}"
- response = (fetch_response(canonical_uri.to_s, headers: headers, body_cacher: body_cacher) rescue nil)
- doc = Nokogiri::HTML(response) if response
- end
- end
-
- doc
+ Nokogiri::HTML(response)
end
def self.fetch_response(location, redirect_limit: 5, domain: nil, headers: nil, body_cacher: nil)
redirect_limit = Onebox.options.redirect_limit if redirect_limit > Onebox.options.redirect_limit