- title = http.start do |http|
- url.path = '/' if url.path == ''
- head = http.request_head(url.path)
- case head
- when Net::HTTPRedirection then
- # call self recursively if this is a redirect
- redirect_to = head['location']
- puts "+ redirect location: #{redirect_to}"
- absolute_uris = URI.extract redirect_to
- raise "wtf! redirect = #{redirect_to}" if absolute_uris.size > 1
- if absolute_uris.size == 1
- url = URI.parse absolute_uris[0]
- else
- url.path = redirect_to
- end
- puts "+ whee, redirect to #{url.to_s}!"
- title = get_title_for_url(url.to_s)
- when Net::HTTPSuccess then
- if head['content-type'] =~ /^text\//
- # content is 'text/*'
- # retrieve the title from the page
- puts "+ getting #{url.path}"
- response = http.request_get(url.path)
- return get_title_from_html(response.body)
- else
- # content isn't 'text/*'... display info about the file.
- size = head['content-length'].gsub(/(\d)(?=\d{3}+(?:\.|$))(\d{3}\..*)?/,'\1,\2')
- #lastmod = head['last-modified']
- return "[Link Info] type: #{head['content-type']}#{size ? ", size: #{size} bytes" : ""}"
- end
- when Net::HTTPClientError then
- return "[Title] Error getting link (#{response.code} - #{response.message})"
- when Net::HTTPServerError then
- return "[Title] Error getting link (#{response.code} - #{response.message})"
- end
- end
+ title = http.start { |http|
+ url.path = '/' if url.path == ''\r
+\r
+ http.request_get(url.path, "User-Agent" => "rbot-url_plugin/666.666") { |response|\r
+
+ case response
+ when Net::HTTPRedirection then
+ # call self recursively if this is a redirect
+ redirect_to = response['location'] || './'
+ puts "+ redirect location: #{redirect_to.inspect}"
+ url = URI.join url.to_s, redirect_to
+ puts "+ whee, redirecting to #{url.to_s}!"
+ title = get_title_for_url(url.to_s, depth-1)
+ when Net::HTTPSuccess then
+ if response['content-type'] =~ /^text\//
+ # since the content is 'text/*' and is small enough to
+ # be a webpage, retrieve the title from the page
+ puts "+ getting #{url.request_uri}"\r
+ data = read_data_from_response(response, 50000)\r
+ return get_title_from_html(data)
+ else
+ # content doesn't have title, just display info.
+ size = response['content-length'].gsub(/(\d)(?=\d{3}+(?:\.|$))(\d{3}\..*)?/,'\1,\2')
+ return "[Link Info] type: #{response['content-type']}#{size ? ", size: #{size} bytes" : ""}"
+ end
+ when Net::HTTPClientError then
+ return "[Link Info] Error getting link (#{response.code} - #{response.message})"
+ when Net::HTTPServerError then
+ return "[Link Info] Error getting link (#{response.code} - #{response.message})"
+ end # end of "case response"\r
+ \r
+ } # end of request block
+ } # end of http start block\r
+