require 'json' require 'uri' module ArchiveAPI def get_raw_list_from_api(url, page_index, http) # Automatically append /* if the URL doesn't contain a path after the domain # This is a workaround for an issue with the API and *some* domains. # See https://github.com/StrawberryMaster/wayback-machine-downloader/issues/6 if url && !url.match(/^https?:\/\/.*\//i) url = "#{url}/*" end request_url = URI("https://web.archive.org/cdx/search/cdx") params = [["output", "json"], ["url", url]] + parameters_for_api(page_index) request_url.query = URI.encode_www_form(params) begin response = http.get(request_url) body = response.body.to_s.strip return [] if body.empty? json = JSON.parse(body) # Check if the response contains the header ["timestamp", "original"] json.shift if json.first == ["timestamp", "original"] json rescue JSON::ParserError, StandardError => e warn "Failed to fetch data from API: #{e.message}" [] end end def parameters_for_api(page_index) parameters = [["fl", "timestamp,original"], ["collapse", "digest"], ["gzip", "false"]] parameters.push(["filter", "statuscode:200"]) unless @all parameters.push(["from", @from_timestamp.to_s]) if @from_timestamp && @from_timestamp != 0 parameters.push(["to", @to_timestamp.to_s]) if @to_timestamp && @to_timestamp != 0 parameters.push(["page", page_index]) if page_index parameters end end