mirror of
https://github.com/discourse/discourse.git
synced 2024-11-23 02:19:27 +08:00
30990006a9
This reduces chances of errors where consumers of strings mutate inputs and reduces memory usage of the app. Test suite passes now, but there may be some stuff left, so we will run a few sites on a branch prior to merging
75 lines
1.7 KiB
Ruby
75 lines
1.7 KiB
Ruby
# frozen_string_literal: true
|
|
|
|
require_dependency 'final_destination'
|
|
|
|
module RetrieveTitle
|
|
CRAWL_TIMEOUT = 1
|
|
|
|
def self.crawl(url)
|
|
fetch_title(url)
|
|
rescue Exception
|
|
# If there was a connection error, do nothing
|
|
end
|
|
|
|
def self.extract_title(html)
|
|
title = nil
|
|
if doc = Nokogiri::HTML(html)
|
|
|
|
title = doc.at('title')&.inner_text
|
|
|
|
# A horrible hack - YouTube uses `document.title` to populate the title
|
|
# for some reason. For any other site than YouTube this wouldn't be worth it.
|
|
if title == "YouTube" && html =~ /document\.title *= *"(.*)";/
|
|
title = Regexp.last_match[1].sub(/ - YouTube$/, '')
|
|
end
|
|
|
|
if !title && node = doc.at('meta[property="og:title"]')
|
|
title = node['content']
|
|
end
|
|
end
|
|
|
|
if title.present?
|
|
title.gsub!(/\n/, ' ')
|
|
title.gsub!(/ +/, ' ')
|
|
title.strip!
|
|
return title
|
|
end
|
|
nil
|
|
end
|
|
|
|
private
|
|
|
|
def self.max_chunk_size(uri)
|
|
|
|
# Amazon and YouTube leave the title until very late. Exceptions are bad
|
|
# but these are large sites.
|
|
return 500 if uri.host =~ /amazon\.(com|ca|co\.uk|es|fr|de|it|com\.au|com\.br|cn|in|co\.jp|com\.mx)$/
|
|
return 300 if uri.host =~ /youtube\.com$/ || uri.host =~ /youtu.be/
|
|
|
|
# default is 10k
|
|
10
|
|
end
|
|
|
|
# Fetch the beginning of a HTML document at a url
|
|
def self.fetch_title(url)
|
|
fd = FinalDestination.new(url, timeout: CRAWL_TIMEOUT)
|
|
|
|
current = nil
|
|
title = nil
|
|
|
|
fd.get do |_response, chunk, uri|
|
|
|
|
if current
|
|
current << chunk
|
|
else
|
|
current = chunk
|
|
end
|
|
|
|
max_size = max_chunk_size(uri) * 1024
|
|
title = extract_title(current)
|
|
throw :done if title || max_size < current.length
|
|
end
|
|
return title
|
|
end
|
|
end
|