Abstract: A service testing system is disclosed to enable consistent replay of stateful requests on a service whose output depends on the service's execution state prior to the requests. In embodiments, the service implements a compute engine that executes service requests and a storage subsystem that maintains execution states during the execution of stateful requests. When a stateful request is received during testing, the storage subsystem creates an in-memory test copy of the execution state to support execution of the request, and provides the test copy to the compute engine. In embodiments, the storage subsystem will create a separate instance of execution state for each individual test run. The disclosed techniques enable mock execution states to be easily created for testing of stateful requests, in a manner that is transparent to the compute engine and does not impact production execution data maintained by the service.

Abstract: A service testing system is disclosed to enable consistent replay of stateful requests on a service whose output depends on the service's execution state prior to the requests. In embodiments, the service implements a compute engine that executes service requests and a storage subsystem that maintains execution states during the execution of stateful requests. When a stateful request is received during testing, the storage subsystem creates an in-memory test copy of the execution state to support execution of the request, and provides the test copy to the compute engine. In embodiments, the storage subsystem will create a separate instance of execution state for each individual test run. The disclosed techniques enable mock execution states to be easily created for testing of stateful requests, in a manner that is transparent to the compute engine and does not impact production execution data maintained by the service.

Abstract: Techniques for implementing a self-serve content classification platform are provided. In one technique, through a user interface (UI) of the content classification platform, a user selects an option corresponding to a particular processor type. A processor configuration is also received through the same or different UI. Based on the user selection and the processor configuration, processor data that associates the particular processor type and the processor configuration is stored in association with a client context. A content classification request is received from a client associated with the client context. In response to receiving the content classification request, content that is referenced by the content classification request is identified, the processor data is accessed based on the content classification request, a processor that is defined based on the processor data is caused to be executed based on the content, and a result is generated based on output from the processor.

Abstract: Techniques for implementing a self-serve content classification platform are provided. In one technique, through a user interface (UI) of the content classification platform, a user selects an option corresponding to a particular processor type. A processor configuration is also received through the same or different UI. Based on the user selection and the processor configuration, processor data that associates the particular processor type and the processor configuration is stored in association with a client context. A content classification request is received from a client associated with the client context. In response to receiving the content classification request, content that is referenced by the content classification request is identified, the processor data is accessed based on the content classification request, a processor that is defined based on the processor data is caused to be executed based on the content, and a result is generated based on output from the processor.

Abstract: An item review system is disclosed that detects that an item reviewed or being reviewed by a user is similar to one or more items previously reviewed by the user. In some embodiments, the system provides an option for the user to modify the prior review of each such similar item, as may be desirable to capture the user's changed perception of the previously reviewed item. The system may additionally or alternatively display the prior review(s) as the user generates the new review, such that the user can take these prior reviews into consideration.

Abstract: Selection refinement tool techniques are described in which a selection within an image is modified using a tool that automatically resizes according to the shape of selection boundaries. In one or more implementations, selection boundaries are displayed in an image that is exposed in a user interface. The selection boundaries may represent an automatic selection via an editing tool, or a selection that is based on manual manipulation of an editing tool by a user. The selection refinement tool is operable to modify the selection boundaries by positioning of a cursor of the refinement tool proximate to the selection boundaries. In response to positioning of the cursor relative to the selection boundaries, the cursor is dynamically resized in dependence upon a positional relationship between the cursor and the selection boundaries, such that boundaries of the cursor are kept from intersecting with the selection boundaries.

Abstract: Systems and methods are provided for generating multi-image content for online services using a single image. For example, a processor identifies publishing constraints specified for image categories in a profile of an online service. The processor generates a first image by applying a first publishing constraint to a copy of an input image and generates a second image by applying a second publishing constraint to a selected portion of the first image. The processor presents the first and second images in different preview interfaces. The processor applies manipulation effects to the second image that modify the second image independently of modifications to the first image. The manipulation effect is applied to the second in accordance with the second publishing constraint for the second image. The processor uploads the images with the applied manipulation effects to the online service for presentation in the different image categories.

Abstract: Selection refinement tool techniques are described in which a selection within an image is modified using a tool that automatically resizes according to the shape of selection boundaries. In one or more implementations, selection boundaries are displayed in an image that is exposed in a user interface. The selection boundaries may represent an automatic selection via an editing tool, or a selection that is based on manual manipulation of an editing tool by a user. The selection refinement tool is operable to modify the selection boundaries by positioning of a cursor of the refinement tool proximate to the selection boundaries. In response to positioning of the cursor relative to the selection boundaries, the cursor is dynamically resized in dependence upon a positional relationship between the cursor and the selection boundaries, such that boundaries of the cursor are kept from intersecting with the selection boundaries.

Abstract: Systems and methods are provided for generating multi-image content for online services using a single image. For example, a processor identifies publishing constraints specified for image categories in a profile of an online service. The processor generates a first image by applying a first publishing constraint to a copy of an input image and generates a second image by applying a second publishing constraint to a selected portion of the first image. The processor presents the first and second images in different preview interfaces. The processor applies manipulation effects to the second image that modify the second image independently of modifications to the first image. The manipulation effect is applied to the second in accordance with the second publishing constraint for the second image. The processor uploads the images with the applied manipulation effects to the online service for presentation in the different image categories.

Abstract: A system and method for automatically assessing the process quality capability of a shipping process is disclosed. The system monitors data on a database that is being constantly updated with information about shipment requests and shipment dates for various products. From that data, a statistical calculation is performed and the results are indicated on an Internet or intranet electronic page. The statistical calculation is designed to indicate the capabilities of the current process to deliver products to customers on time.

Abstract: The present invention discloses a method and apparatus for displaying real-time status of product availability, orders, and sales revenue. The present invention includes querying a database at regular recurring time intervals for information regarding the date a product will be ready for shipment. The number of days between the current date and the date when each product will be ready for shipment is calculated and is displayed along with the date each product will be ready for shipment. The present invention further includes one or more computer programs that when executed by one or more processors instructs the one or more processors to display real-time status of product availability. The program instructions further include populating the database with data to include a date when each product will be available for shipment.

Abstract: A system and method are disclosed for electronically reporting in real-time the status of work in progress by a computerized system. The invention provides alert notification when a customer requested shipment date falls before the date that a shipment was promised or is scheduled to be fulfilled. The invention further includes notification when the request date for shipment is near, thereby providing a highlighted alert notification to managers. The system is designed to update its data when a request is made, or automatically in real-time or on a regular basis. A proactive alert is provided to alert managers of a correctable problem for that order and a reactive alert is provided to track late shipments to prevent future late shipments.

Abstract: The present invention discloses a method and apparatus for displaying real-time status of product availability, orders, and sales revenue. The present invention includes querying a database at regular recurring time intervals for information regarding the date a product will be ready for shipment. The number of days between the current date and the date when each product will be ready for shipment is calculated and is displayed along with the date each product will be ready for shipment. The present invention further includes one or more computer programs that when executed by one or more processors instructs the one or more processors to display real-time status of product availability. The program instructions further include populating the database with data to include a date when each product will be available for shipment.

Abstract: A system and method for automatically assessing the process quality capability of a shipping process is disclosed. The system monitors data on a database that is being constantly updated with information about shipment requests and shipment dates for various products. From that data, a statistical calculation is performed and the results are indicated on an Internet or intranet electronic page. The statistical calculation is designed to indicate the capabilities of the current process to deliver products to customers on time.