Microsoft Technology Licensing, LLC

Jeffrey Anderson

John Healy

Ryan Munson

Josh W Honeyman

Ahmad Mahdi El Husseini

Michael Gall

William Leslie Frandsen

David J Lannoye

Lars Dragheim Blaaberg

Ganapathi Subramanian Sadasivam

Sanket M Vasa

Steven Degele

Tristan H Cartony

John W Honeyman

Some database tables include a relational field, where a record of the table stores a foreign-key reference value that identifies a related record of a second table. While fulfilling a request comprising a query over the database, the system may identify a context of the request, wherein the context is supplemental to the query, and wherein the context relates to the relational field. For example, a security policy may permit a requester to query the first table, but restrict the requester from accessing the second table that the relational field of the first table references, where dereferencing the relational field may cause unexpected results. Instead, the system may evaluate the context of the request, and may fulfill the request based on the context by providing a response that substitutes, for the value that references the second table, a substitute value that does not reference the second table.

Reconciling foreign key references and table security policies

An incoming string is received and an initial segmentation is generated based upon separators in the incoming string. The segment values in the initial segmentation are validated by determining whether they are found in a data store. Where a segment in the initial segmentation is not found, subsequent segments, and corresponding separators, are iteratively concatenated to find segment values. Where a segment is still not found, after iterative concatenation, the initial segment in the concatenation is output as an unfound segment and the subsequent segments are processed further.

Segmenting data with included separators

A consuming subsystem calculates information based on setup information from one or more other subsystems. Each of the one or more other subsystems generates a base version value that changes every time any of the setup information changes. The consuming subsystem caches information, including the base version values at the time the information was calculated by the consuming subsystem.

Caching using base setup version

A Chart of accounts (COA) is maintained as a logical set of constraints, criteria, and relationships and does not persist any of the valid combinations until a first time a particular combination is used. When any business logic or properties need to be set or referenced on an individual combination, that combination is physically persisted 'just in time' and then becomes a physical combination that may be used going forward. A hash value is associated with each combination such that lookups are simplified.

Logical chart of accounts with hashing

Combination based LRU caching employs a mapping mechanism in an LRU cache separate from a set of LRU caches for storing the values used in the combinations. The mapping mechanism is used to track the valid combinations of the values in the LRU caches storing the values resulting in any given value being stored at most once. Through the addition of a byte pointer significantly more combinations may be tracked in the same amount of cache memory with full LRU semantics on both the values and combinations.

Combination based LRU caching

A computer-implemented method and system are provided for retrieving a metadata instance defining at least part of an application. In the method, a first metadata instance is loaded. Then, one or more templates specified in the first metadata instance are identified. The identified templates are recursively loaded, and metadata from the first metadata instance is applied to the recursively loaded templates to produce a new metadata instance. The new metadata instance is returned.

Recursive metadata templating

A system for providing automatic event publication relative to data owned by a service is disclosed. A service which owns data publishes internal events when data is created, updated, or deleted. These internal events are only visible by consumers on the same physical computer and running within the same process or executable. Consumers who are located remotely or in a separate process cannot subscribe to these events. Embodiments provide the ability to expose local entity create, update, and delete events as 'public' events (or out-only messages) on a service contract. This allows external consumers who may be remote and even on a different platform, to subscribe and receive notifications of when data changes in the owning service.

Automatic create, update and delete event publishing

The present invention provides an entity projection in a client. The entity projection is an encapsulation of data, from an owning service, and residing on a consumer of that service. The entity projection provides an abstraction that conforms to an entity programming model on the consumer. The entity projection may illustratively include a subset of the properties of an entity in the owning service, the subset being only those properties required by the entity and those properties desired by the consumer.

Entity projection

An abstraction for rich data access for consumers of a service in a service oriented system. The abstraction is created by the author of a service which owns the data, and includes a number of service-public properties and associations. Further, the abstraction can contain metadata and logic that provides for a rich, interactive experience for the consumer interacting with the data encapsulated by the entity agent. These can include verifying that the data conforms to the contract published by the owning service, validating and defaulting properties, and validating associated data. Also the abstraction can contains metadata and logic for transmitting of the data back to the owning service and optionally saving to the local data store of the consumer.

Growing community of inventors

West Fargo, ND, United States of America

Jeffrey Anderson