+
+ def get_title_from_html(pagedata)
+ return unless TITLE_RE.match(pagedata)
+ $1.ircify_html
+ end
+
+ def get_title_for_url(uri_str, nick = nil, channel = nil)
+
+ url = uri_str.kind_of?(URI) ? uri_str : URI.parse(uri_str)
+ return if url.scheme !~ /https?/
+
+ logopts = Hash.new
+ logopts[:nick] = nick if nick
+ logopts[:channel] = channel if channel
+
+ title = nil
+ extra = String.new
+
+ begin
+ debug "+ getting #{url.request_uri}"
+ @bot.httputil.get_response(url) { |resp|
+ case resp
+ when Net::HTTPSuccess
+
+ debug resp.to_hash
+
+ if resp['content-type'] =~ /^text\/|(?:x|ht)ml/
+ # The page is text or HTML, so we can try finding a title and, if
+ # requested, the first par.
+ #
+ # We act differently depending on whether we want the first par or
+ # not: in the first case we download the initial part and the parse
+ # it; in the second case we only download as much as we need to find
+ # the title
+ #
+ if @bot.config['url.first_par']
+ partial = resp.partial_body(@bot.config['http.info_bytes'])
+ logopts[:title] = title = get_title_from_html(partial)
+ first_par = Utils.ircify_first_html_par(partial, :strip => title)
+ unless first_par.empty?
+ logopts[:extra] = first_par
+ extra << ", #{Bold}text#{Bold}: #{first_par}"
+ end
+ call_event(:url_added, url.to_s, logopts)
+ return "#{Bold}title#{Bold}: #{title}#{extra}" if title
+ else
+ resp.partial_body(@bot.config['http.info_bytes']) { |part|
+ logopts[:title] = title = get_title_from_html(part)
+ call_event(:url_added, url.to_s, logopts)
+ return "#{Bold}title#{Bold}: #{title}" if title
+ }
+ end
+ # if nothing was found, provide more basic info, as for non-html pages
+ else
+ resp.no_cache = true
+ end
+
+ enc = resp['content-encoding']
+ logopts[:extra] = String.new
+ logopts[:extra] << "Content Type: #{resp['content-type']}"
+ if enc
+ logopts[:extra] << ", encoding: #{enc}"
+ extra << ", #{Bold}encoding#{Bold}: #{enc}"
+ end
+
+ unless @bot.config['url.titles_only']
+ # content doesn't have title, just display info.
+ size = resp['content-length'].gsub(/(\d)(?=\d{3}+(?:\.|$))(\d{3}\..*)?/,'\1,\2') rescue nil
+ if size
+ logopts[:extra] << ", size: #{size} bytes"
+ size = ", #{Bold}size#{Bold}: #{size} bytes"
+ end
+ call_event(:url_added, url.to_s, logopts)
+ return "#{Bold}type#{Bold}: #{resp['content-type']}#{size}#{extra}"
+ end
+ call_event(:url_added, url.to_s, logopts)
+ else
+ raise UrlLinkError, "getting link (#{resp.code} - #{resp.message})"
+ end
+ }
+ return nil
+ rescue Exception => e
+ case e
+ when UrlLinkError
+ raise e
+ else
+ error e
+ raise "connecting to site/processing information (#{e.message})"
+ end
+ end
+ end
+