require 'rev-api/version' require 'rev-api/http_client' require 'rev-api/exceptions' require 'json' # automatically include business logic objects Dir[File.dirname(__FILE__) + '/models/*.rb'].each do |file| require file end # Rev API Ruby SDK module Rev # Main point of interaction with API. # Wraps common REST operations, returning plain objects. # Internally utilizes JSON resource representation. class Api # Production host. Used by default for new Rev::Api client PRODUCTION_HOST = 'www.rev.com' # Sandbox domain - pass 'Rev::Api::SANDBOX_HOST' as third param # into Rev::Api ctor SANDBOX_HOST = 'api-sandbox.rev.com' # @note http://www.rev.com/api/security # @param client_api_key [String] secret key specific to each partner that wishes to use the Rev API # @param user_api_key [String] secret key specific to a Rev user, which identifies the user account under whose privileges the requested operation executes # @param host [String] use {Rev::Api::PRODUCTION_HOST} or {Rev::Api::SANDBOX_HOST}. Production is default value # @return [HttpClient] client obj def initialize(client_api_key, user_api_key, host = PRODUCTION_HOST) @client = HttpClient.new(client_api_key, user_api_key, host) end # Loads single page of existing orders for current client # # @note http://www.rev.com/api/ordersget # @param page [Int, nil] 0-based page number, defaults to 0 # @return [OrdersListPage] paged result cointaining 'orders' def get_orders_page(page = 0) response = @client.get("/orders?page=#{page.to_i}") Api.verify_get_response(response) OrdersListPage.new(Api.parse(response)) end # Loads all orders for current client. Works by calling get_orders_page multiple times. # Use with caution if your order list might be large. # # @note http://www.rev.com/api/ordersget # @return [Array of Order] list of orders def get_all_orders orders = [] page = 0 loop do orders_page = self.get_orders_page page page += 1 orders.push *orders_page.orders break if (page * orders_page.results_per_page >= orders_page.total_count) end orders end # Returns Order given an order number. # # @note http://www.rev.com/api/ordersgetone # @param number [String] order number, like 'TCXXXXXXXX' # @return [Order] order obj def get_order(number) response = @client.get("/orders/#{number}") Api.verify_get_response(response) Order.new(Api.parse(response)) end # Cancel an order by number. If cancellation is not allowed, Rev::Api::BadRequestError is raised. # # @note http://www.rev.com/api/orderscancel # @param number [String] order number # @return [Boolean] true on success, raised Exception from Rev::Api namespace otherwise def cancel_order(number) data = { :order_num => number } response = @client.post("/orders/#{number}/cancel", data) Api.verify_post_response(response) end # Get metadata about an order attachment. # Use this method to retrieve information about an order attachment (either transcript, # translation, or source file). # # @note http://www.rev.com/api/attachmentsget # @param id [String] attachment id, as returned in info about an order # @return [Attachment] attachment object def get_attachment_metadata(id) response = @client.get("/attachments/#{id}") Api.verify_get_response(response) Attachment.new(Api.parse(response)) end # Get the raw data for the attachment with given id. # Download the contents of an attachment. Use this method to download either a finished transcript, # finished translation or a source file for an order. # For transcript and translation attachments, you may request to get the contents in a specific # representation, specified via a mime-type. # # See {Rev::Order::Attachment::REPRESENTATIONS} hash, which contains symbols for currently supported mime types. # The authoritative list is in the API documentation at http://www.rev.com/api/attachmentsgetcontent # # If a block is given, the response is passed to the block directly, to allow progressive reading of the data. # In this case, the block must itself check for error responses, using Api.verify_get_response. # If no block is given, the full response is returned. In that case, if the response is an error, an appropriate # error is raised. # # @param id [String] attachment id # @param mime_type [String, nil] mime-type for the desired format in which the content should be retrieved. # @yieldparam resp [Net::HTTP::Response] the response, ready to be read # @return [Net::HTTP::Response] the response containing raw data def get_attachment_content(id, mime_type = nil, &block) headers = {} unless mime_type.nil? headers['Accept'] = mime_type headers['Accept-Charset'] = 'utf-8' if mime_type.start_with? 'text/' end if block_given? @client.get_binary("/attachments/#{id}/content", headers, &block) else response = @client.get_binary("/attachments/#{id}/content", headers) Api.verify_get_response(response) response end end # Get the raw data for the attachment with given id. # Download the contents of an attachment and save it into a file. Use this method to download either a finished transcript, # finished translation or a source file for an order. # For transcript and translation attachments, you may request to get the contents in a specific # representation, specified via a mime-type. # # See {Rev::Order::Attachment::REPRESENTATIONS} hash, which contains symbols for currently supported mime types. # The authoritative list is in the API documentation at http://www.rev.com/api/attachmentsgetcontent # # @param id [String] attachment id # @param path [String, nil] path to file into which the content is to be saved. # @param mime_type [String, nil] mime-type for the desired format in which the content should be retrieved. # @return [String] filepath content has been saved to. Might raise standard IO exception if file creation files def save_attachment_content(id, path, mime_type = nil) headers = {} unless mime_type.nil? headers['Accept'] = mime_type headers['Accept-Charset'] = 'utf-8' if mime_type.start_with? 'text/' end # same simple approach as box-api does for now: return response.body as-is if path for saving is nil File.open(path, 'wb') do |file| response = @client.get_binary("/attachments/#{id}/content", headers) do |resp| resp.read_body do |segment| file.write(segment) end end Api.verify_get_response(response) end # we don't handle IO-related exceptions path end # Get the content of the attachment with given id as a string. Use this method to grab the contents of a finished transcript # or translation as a string. This method should generally not be used for source attachments, as those are typically # binary files like MP3s, which cannot be converted to a string. # # May raise Rev::Api::NotAcceptableError if the attachment cannot be converted into a text representation. # # @param id [String] attachment id # @return [String] the content of the attachment as a string def get_attachment_content_as_string(id) response = self.get_attachment_content(id, Attachment::REPRESENTATIONS[:txt]) response.body end # Submit a new order using {Rev::OrderRequest}. # @note http://www.rev.com/api/ordersposttranscription - for full information # # @param order_request [OrderRequest] object specifying payment, inputs, options and notification info. # inputs must previously be uploaded using upload_input or create_input_from_link # @return [String] order number for the new order # Raises {Rev::BadRequestError} on failure (.code attr exposes API error code - # see {Rev::OrderRequestError}). def submit_order(order_request) response = @client.post("/orders", order_request.to_json, { 'Content-Type' => 'application/json' }) Api.verify_post_response(response) new_order_uri = response.headers['Location'] return new_order_uri.split('/')[-1] end # Upload given local file directly as source input for order. # @note http://www.rev.com/api/inputspost # # @param path [String] mandatory, path to local file (relative or absolute) to upload # @param content_type [String] mandatory, content-type of the file you're uploading # @return [String] URI identifying newly uploaded media. This URI can be used to identify the input # when constructing a OrderRequest object to submit an order. # {Rev::BadRequestError} is raised on failure (.code attr exposes API error code - # see {Rev::InputRequestError}). def upload_input(path, content_type) filename = Pathname.new(path).basename headers = { 'Content-Disposition' => "attachment; filename=#{filename}", 'Content-Type' => content_type } File.open(path) do |data| response = @client.post_binary("/inputs", data, headers) Api.verify_post_response(response) headers = HTTParty::Response::Headers.new(response.to_hash) return headers['Location'] end end # Request creation of a source input based on an external URL which the server will attempt to download. # @note http://www.rev.com/api/inputspost # # @param url [String] mandatory, URL where the media can be retrieved. Must be publicly accessible. # HTTPS urls are ok as long as the site in question has a valid certificate # @param filename [String, nil] optional, the filename for the media. If not specified, we will # determine it from the URL # @param content_type [String, nil] optional, the content type of the media to be retrieved. # If not specified, we will try to determine it from the server response # @return [String] URI identifying newly uploaded media. This URI can be used to identify the input # when constructing a OrderRequest object to submit an order. # {Rev::BadRequestError} is raised on failure (.code attr exposes API error code - # see {Rev::InputRequestError}). def create_input_from_link(url, filename = nil, content_type = nil) request = { :url => url } request[:filename] = filename unless filename.nil? request[:content_type] = content_type unless content_type.nil? response = @client.post("/inputs", request.to_json, { 'Content-Type' => 'application/json' }) Api.verify_post_response(response) response.headers['Location'] end private # Below are utility helper methods for handling response class << self # Parse given response's body JSON into Hash, so that it might be # easily mapped onto business logic object. # # @param response [Response] HTTParty response obj # @return [Hash] hash of values parsed from JSON def parse(response) JSON.load response.body.to_s end # Raises exception if response is not considered as success # # @param response [HTTPParty::Response] HTTParty response obj. Net::HTTPResponse represented by .response # @return [Boolean] true if response is considered as successful def verify_get_response(response) # HTTP response codes are handled here and propagated up to the caller, since caller should be able # to handle all types of errors the same - using exceptions unless response.response.instance_of? Net::HTTPOK Api.handle_error(response) end true end # (see #verify_get_response) def verify_post_response(response) # see http://www.rev.com/api/errorhandling unless response.response.instance_of?(Net::HTTPCreated) || response.response.instance_of?(Net::HTTPNoContent) Api.handle_error(response) end true end # Given a response, raises a corresponding Exception. # Full response is given for the sake of BadRequest reporting, # which usually contains validation errors. # # @param response [Response] containing failing status to look for def handle_error(response) case response.response when Net::HTTPBadRequest # Bad request - response contains error code and details. Usually means failed validation body = JSON.load response.body.to_s msg = "API responded with code #{body['code']}: #{body['message']}" msg += " Details: #{body['detail'].to_s}" if body['detail'] raise BadRequestError.new msg, body['code'] when Net::HTTPUnauthorized raise NotAuthorizedError when Net::HTTPForbidden raise ForbiddenError when Net::HTTPNotFound raise NotFoundError when Net::HTTPNotAcceptable raise NotAcceptableError when Net::HTTPServerError raise ServerError, "Status code: #{response.code}" else raise UnknownError end end end end end