README

Xillion

Xillion is an Attribute-based Resource Framework.

It allows you to work with Resources and their collection of attributes.

Use your entities in XACML, ABAC, OPA and other external policy-based frameworks
Standardise the way remote objects from different sources are represented for integration purposes

Resource: representation of 1 real-world entity. Has a unique ID (in the form of a URL) and an array of Attributes
Attribute: Defines a single attribute. Has a unique ID (in the form of a URL), description, an AttributeType and other metadata
DataType: Data type supported by an Attribute (i.e. string, int, etc)
ResourceRepository: Container for a set of Resources that allows you to search, get and set Resource instances in bulk
ResourceContext: Container for a set of Resources Repositories. The root of the structure.
Profile: A "Contract" or "Interface", specifying which attributes Resources of this type need to "implement'. A single Resource can implement multiple profiles simultaniously.

Specify a Attributes, DataTypes, Profiles and Resources to use in your projects in YAML configuration files
Predefined instances of xml schema datatypes are provided that are commonly used in XACML environments
Resource Repositories: currently ArrayRepository, with plans for PDO and Elasticsearch based Repositories.
Resource Loader
Resource Resolver that can resolve pre-populated Resources with their attributes and types from objects through various means (currently direct and provider-based)
Resolve Resources directly from your entities: Quick and simple method if you "own" those entities.
Resolve Resources through a ResourcesProvider: Useful to extract attributes for entities from external libraries.
Collect your resources in a single Resource Context for bulk operations
Validate one Attribute, a Resource or an entire Resource Context

All identifiers are prefixed with a FQDN (Fully Qualified Domain Name). I.e. core.xillion.cloud/display
The identifier FQDN is owned by the entity that defines the identifier (i.e. do not randomly defined resource or attribute identifiers with domain names that you do not own).
Attribute identifiers do not specify a scheme (i.e. http://, https://, file:// etc)
Attribute identifiers do not contain a path prefix (i.e. do not prefix paths with /xillion etc)
Identifiers are lower-case only, and allow dashes for word-spacing (i.e. /some-example)
Any level of sub-paths is allowed in identifiers. The recommendation is to keep the levels to a minimum (i.e `x.example.web/a/b/c/d/e/f is allowed but discouraged)
Profile identifiers are in a sub-path prefixed with /profiles.
Data-type identifiers are in a sub-path prefixed with /data-types.
Attribute identifiers do not use a prefix. (i.e. do not use prefix /attributes)
Resource identifiers do not use a prefix. (i.e. do not use prefix /resources)
Packages (FQDN) define either a set of attributes and profiles (library) -or- resources (content), not both.

MIT. Please refer to the license file for details.

Check out our other projects at linkorb.com/engineering.

Btw, we're hiring!